2012 COURSE DATES: AUGUST 4 – 17, 2012 - Sirenian International
2012 COURSE DATES: AUGUST 4 – 17, 2012 - Sirenian International
2012 COURSE DATES: AUGUST 4 – 17, 2012 - Sirenian International
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
ECOLOGY, BEHAVIOR & CONSERVATION OF MANATEES & DOLPHINS AND THE<br />
COMPLEX MANGROVE-SEAGRASS-CORAL HABITATS WE SHARE WITH MARINE<br />
MEGAFAUNA WITHIN THE BELIZE BARRIER REEF LAGOON SYSTEM<br />
<strong>2012</strong> <strong>COURSE</strong> <strong>DATES</strong>: <strong>AUGUST</strong> 4 <strong>–</strong> <strong>17</strong>, <strong>2012</strong><br />
OVERVIEW & BACKGROUND<br />
Antillean manatees (Trichechus manatus manatus) are found throughout Central America and the Caribbean, but<br />
are Red Listed by the IUCN as endangered, in continuing decline, with severely fragmented populations. The UN<br />
Caribbean Environmental Programme considers them an endangered and protected species of regional concern,<br />
threatened by poaching, boat strikes, entanglement in fishing gear, and habitat degradation. Belize may be the last<br />
stronghold for Antillean manatees in the Caribbean; and the Drowned Cayes area is one of the most important<br />
activity centers in Belize. However, the existing body of knowledge is inadequate to develop and implement sitespecific<br />
management and recovery plans. Because manatees are elusive, endangered, and have slow reproductive<br />
rates, long-term studies in this area are necessary to evaluate and monitor the population status and develop practical<br />
conservation plans to ensure survival of the population, and ultimately, the sub-species. Our comprehensive and<br />
collaborative project began in October 1998 and hopefully, will continue indefinitely into the future!
Bottlenose dolphins (Tursiops truncatus) are not<br />
endangered, but their stocks are considered depleted<br />
by the U.S. Marine Mammal Protection Act. A<br />
study of the local population was started in the<br />
Drowned Cayes by Oceanic Society Expeditions<br />
(OSE) in 1997, however it was abandoned in 2001<br />
when OSE moved their base to Turneffe, an atoll<br />
approximately 10 miles east of the Belize Barrier<br />
Reef. We re-started a photo-id project in 2005 with<br />
the hope of additional collaboration with OSE and<br />
other dolphin researchers in the region.<br />
Although this project does not specifically<br />
investigate sea turtles in Belize, loggerhead,<br />
hawskbill, and green sea turtles are occasionally<br />
sighted during our expeditions. And, in 2011, our<br />
visiting scientist is a turtle expert!<br />
Students and volunteers have been an integral component of the project since its conception. We rely on you to<br />
assist us with a wide range of data collection and processing during each survey. During this course, half-day (3-4<br />
hours) manatee & dolphin surveys will be conducted from a small boat on 10 of your 14-day expedition.<br />
Additionally, each student will undertake an individual research project to investigate another component of our<br />
field site, exposing students the broader Conservation Biology issues (see syllabus for suggested research topics.<br />
WELCOME LETTER<br />
While in Belize, be warned, “If ya drink de watter, ya mus com bok.” This old Belizean Creole proverb is true! On<br />
my first trip to Belize in 1998, I drank the water and fell in love with both the people and the place…and the<br />
manatees, dolphins, and other wildlife of<br />
course! Since that first trip, Belize has been<br />
my 2 nd home. Hopefully, you will soon be<br />
equally enthusiastic about the opportunities and<br />
challenges that await you in this tropical<br />
paradise.<br />
This is no ordinary field course! During the<br />
course, you will be collecting data on<br />
manatees, dolphins, and their habitat that my<br />
research partners and I actually use for our<br />
long-term studies of Antillean manatees and<br />
bottlenose dolphins in the Drowned Cayes<br />
(pronounced "keys") area of Belize. Although<br />
our project focuses primarily on manatees and<br />
dolphins, you will also experience the<br />
mangrove-seagrass-coral reef system in which<br />
we live and work. Perhaps most importantly,<br />
the data are also used by Belizeans, local<br />
NGOs, agencies, and decision-makers, to<br />
better develop marine conservation strategies.<br />
The Drowned Cayes is a complex maze of<br />
mangrove islands, just east of Belize City,<br />
surrounded by the Caribbean Sea. Since we<br />
are just two miles inside the Belize Barrier<br />
Reef, we are quite well protected from severe<br />
wave action, but if a tropical storm or hurricane threatens the western Caribbean, we will implement our Hurricane<br />
Plan, which includes evacuation. The Caribbean Sea and sunshine are breathtaking, but they can cause great<br />
Belize Field Course Briefing Page 1 4/4/<strong>2012</strong>
discomfort without proper protection, so you must pay close attention to the Packing Checklist in this briefing.<br />
Photo above: (c) C. Self-Sullivan: “Mario” a friendly, unmarked manatee videoed at North Gallows in 2001.<br />
You should be prepared to spend long periods of time outdoors, on the island and on the water in a small boat,<br />
searching for and observing manatees and dolphins. If you wish, you will have an opportunity to snorkel in the<br />
seagrass beds, mangrove bogues, or on coral patches. But please don’t expect to swim with manatees and dolphins <strong>–</strong><br />
our wild animals are not acclimated to humans and intentional swimming with manatees & dolphins is not permitted<br />
in Belize. After a few days in the field, you will understand why we refer to manatees as elusive. We can pretty<br />
much guarantee that you will see both manatees & dolphins in the wild, but we can’t predict how many or how close<br />
they will come to our research boat. However, the more you read about manatee & dolphin ecology and behavior<br />
prior to your arrival, the more rewarding your observations will be.<br />
You should also be prepared for total immersion into private island living. As field researchers, we spend much of<br />
our time innovating and problem solving. On this<br />
expedition, you will experience the unique lifestyle<br />
of wildlife researchers under moderate conditions.<br />
While we live comfortably compared to most cayedwellers,<br />
it’s quite different from life in the northern<br />
hemisphere. Spanish Lookout Caye, a 184-acre<br />
mangrove island, is shared by Spanish Bay<br />
Conservation and Research Center, the Hugh Parkey<br />
Foundation for Marine Awareness and Education,<br />
and Hugh Parkey’s Belize Dive Connection and<br />
Adventure Lodge. During your two recreational<br />
days, we will be using Belize Dive Connections to<br />
experience an inland adventure to a Maya site, and<br />
a sea adventure on kayaks, snorkel or SCUBA (you<br />
may SCUBA only if you are a certified diver...bring<br />
your sea card & dive log). Photo (c) C. Self-<br />
Sullivan: “Dubya” a dolphin we’ve observed in our<br />
study area since 2004.<br />
Evening entertainment includes good conversation,<br />
star gazing, and a few old fashioned board and card<br />
games. So bring your favorite stories to share with us as we create some new stories to take back to your friends!<br />
Feel free to bring non-electronic musical instruments and games, too. Hopefully, participating on this expedition<br />
will leave you with a new perspective on sustainable living <strong>–</strong> are you up to the challenge?<br />
Cheers,<br />
Caryn Self-Sullivan, Ph.D.<br />
Adjunct Faculty, Nova Southeastern University<br />
President & Co-founder, <strong>Sirenian</strong> <strong>International</strong><br />
200 Stonewall Drive, Fredericksburg, VA 22401-2110<br />
Email: cselfsullivan@sirenian.org<br />
Voice: 540.287.8207 | Fax: 888.371.4998<br />
Belize Field Course Briefing Page 2 4/4/<strong>2012</strong>
TABLE OF CONTENTS PAGE<br />
BACKGROUND & PROJECT OVERVIEW Cover - 1<br />
WELCOME LETTER 1-2<br />
REGISTRATION FORM APPENDIX<br />
<strong>COURSE</strong> SYLLABUS (REQUIRES SIGNATURE) APPENDIX<br />
<strong>COURSE</strong> POLICY AND LIABILIT RELEASE FORM (REQUIRES SIGNATURE) APPENDIX<br />
REGISTRATION CHECK LIST 3 (BELOW)<br />
ABOUT BELIZE 4<br />
PRINCIPAL INVESTIGATOR, CO-PI, VISITING SCIENTISTS 4-5<br />
FIELD TRAINING AND ASSIGNMENTS 6-7<br />
ACCOMMOATIONS & FOOD 7-8<br />
ENVIRONMENTAL CONDITIONS 8<br />
POTENTIAL HAZARDS 9<br />
MEDICAL CONDITIONS OF SPECIAL CONCERN 9<br />
HEALTH INFORMATION 10<br />
PACKING CONSIDERATIONS 10<br />
EMERGENCIES IN THE FIELD 10-12<br />
OTHER USEFUL INFORMATION 12<br />
PACKING CHECK LIST 13-14<br />
APPENDIX (Forms, Data Sheets, Background Readings, Course Flyer, CVs) 15<br />
REGISTRATION & Pre-FIELDING CHECK LIST Session I<br />
Sessions II & III 1 st Group of items is due 90 days prior to fielding<br />
2 nd group of items is due 60 days prior to fielding Sent Date<br />
1 May <strong>2012</strong> Completed Registration Form _________<br />
1 May <strong>2012</strong> Deposit Invoice Paid _________<br />
1 May <strong>2012</strong> Signed Course Policy & Liability Release Form _________<br />
1 May <strong>2012</strong> Signed Syllabus & Unofficial Copy of Transcripts _________<br />
1 May <strong>2012</strong> Low-income Country Scholarship Essay _________<br />
1 May <strong>2012</strong> 1 st, 2 nd, and 3 rd Choice of Belize: Sea to Stars Topic _________<br />
1 May <strong>2012</strong> 1 st, 2 nd, and 3 rd Choice of Independent Research Topic _________<br />
1 June <strong>2012</strong> Balance Due Invoice Paid _________<br />
1 June <strong>2012</strong> Copy of Passport & Current Photo (Head Shot) _________<br />
1 June <strong>2012</strong> Copy of DAN Preferred Membership Card _________<br />
1 June <strong>2012</strong> Copy of SCUBA Certification Card (if diving) _________<br />
1 June <strong>2012</strong> Copy of Airline Travel Itinerary & Receipt _________<br />
1 June <strong>2012</strong> Proposed Presentation Format for Belize: Sea to Stars _________<br />
1 June <strong>2012</strong> Literature Review & Citations for Independent Research Topic _________<br />
Belize Field Course Briefing Page 3 4/4/<strong>2012</strong>
ABOUT BELIZE<br />
Research Site: The project is based on Spanish Lookout Caye, a 184-acre twin-mangrove island located in the<br />
Drowned Cayes, about 10 miles east-southeast of Belize City. The Drowned Cayes are a maze of mangrove islands<br />
inside the Belize Barrier Reef. Approximately 10 acres of Spanish Lookout Caye have been filled and developed,<br />
leaving over <strong>17</strong>0-acres of pristine mangroves and<br />
mangrove swamps divided by a fast flowing channel<br />
of water commonly known as Gilroy’s Creek.<br />
Mangrove islands are famous for their wildlife and<br />
you will share this island home with birds, crabs,<br />
mosquitoes, and sandflies; you might even be lucky<br />
enough to sight an American crocodile or boa<br />
constrictor. Thirty to forty hours will be spent on our<br />
research boat, and marine hazards in the area include<br />
jellyfish and fire coral. The tropical sun is strong and<br />
the humidity is very high. Among the best reasons for<br />
visiting Belize are its long-standing and rapidly<br />
expanding conservation ethic and the incredible<br />
diversity of natural habitats. The spectacular Belize<br />
Barrier Reef, which is the second largest barrier reef in the world and runs along the entire 280 km coastline<br />
(Beletsky 1999), is perhaps its most celebrated natural treasure.<br />
Cultural, Social and Political Environment: Belize is a small country located between Mexico and Guatemala on<br />
the Yucatan Peninsula and was formally known as British Honduras. Because Belize was a British protectorate,<br />
English is taught and spoken at all school levels and is the official language. The more common language spoken<br />
between Belizeans is Creole <strong>–</strong> an unwritten language that combines English with African, Maya, and Spanish words.<br />
Most Belizeans speak at least two languages: English and at least one other such as Creole, Garifuna, Spanish, Maya<br />
Mopan, Kechi Maya, German (Menonites), etc.<br />
Belize is a politically stable democracy, with a Parliamentary system of government and elections every five years.<br />
Elections were held in spring of 2008, and the United Democratic Party (UDP) currently holds the majority of<br />
governmental seats. The People’s United Party (PUP) is the dominate opposition, having just lost their first election<br />
in 10 years. There are a few independents in office. The current Prime Minister is the Honorable Dean Barrow (also<br />
father of Rapper Jamal “Shyne” Barrow). Belizeans are quite vocal about their political opinions; almost everyone<br />
turns out to vote, and local taxi drivers are a good source of information about the current issues. Christianity is the<br />
predominant religion with many churches serving both Catholic and Protestant congregations.<br />
Like any large urban area, Belize City has its share of crime. The relatively small population of less than 300,000<br />
people consists of predominantly five unique cultures: Maya, Mistiso, Creole, Garifuna, and European. More<br />
recently Mennonites, Chinese, and Taiwanese populations are growing in Belize.<br />
PRINCIPAL INVESTIGATOR<br />
Caryn Self-Sullivan, Ph.D., is currently adjunct faculty at Nova<br />
Southeastern University, President of <strong>Sirenian</strong> <strong>International</strong>,<br />
Scientific Advisor to the NCRC West African manatee project in<br />
Volta Lake, Ghana, and Marine Science Advisor to the Hugh<br />
Parkey Foundation for Marine Awareness and Education in Belize.<br />
She graduated from Coastal Carolina University in 1997 with a<br />
B.S. in Marine Science and minors in Mathematics and Biology.<br />
She was an NSF Graduate Fellow from 1999-2001, and received<br />
her Ph.D. in Wildlife & Fisheries Sciences in 2008. Additionally,<br />
Dr. C serves on the IUCN Species Survival Commission Sirenia<br />
Specialist Group, the Belize National Manatee Working Group, and<br />
the Belize Marine Mammal Stranding Network. She has taught<br />
university level courses in Environmental Education, General<br />
Biology and Conservation Biology. She has taught professional<br />
level workshops focused on the order Sirenia in the Dominican<br />
Dr. Self-Sullivan<br />
Belize Field Course Briefing Page 4 4/4/<strong>2012</strong>
Republic, Belize, Ghana, and the USA. Her research interests include marine biology, animal behavior, endangered<br />
marine species, and conservation biology, with a focus on marine mammals. In the field, Dr. C is responsible for<br />
supervising the students and staff, field training, coordinating the lecture/learning/discussion series, overseeing<br />
experimental design and data collection methods, capturing manatees with underwater video camera equipment, and<br />
capturing dolphins with above water digital camera equipment. During this course, Dr. C will be your primary<br />
expert on the local research project, manatees, and dolphins.<br />
CO-PI AND VISITING SCIENTISTS (CVS ARE IN THE APPENDIX SECTION)<br />
Katherine S. LaCommare, Ph.D., Co-PI will receive her<br />
PhD from the Environmental Biology Program, Department of<br />
Biology, University of Massachusetts, Boston, in May 2011!<br />
She is adjunct faculty at Lansing Community College in<br />
Michigan. Her previous degrees include a M.S. in Forestry,<br />
Conservation Biology Program, Department of Forestry and<br />
Natural Resources, Purdue University; and a B.S. in<br />
Anthropology/Zoology, University of Michigan. Katie cofounded<br />
<strong>Sirenian</strong> <strong>International</strong> with Dr. C and has been the co-<br />
PI on our long-term manatee research project since its<br />
inception in 1998. Unfortunately, Katie may not be able to<br />
join us during the 2001 field course.<br />
Heather J. Kalb, Ph.D., Visiting Scientist is an Assistant<br />
Dr. LaCommare<br />
Profession of Biology at West Liberty University in Wheeling,<br />
WV. She received her Ph.D. in Zoology from Texas A&M<br />
University in 1999. Dr. Kalb specializes in the reproductive<br />
physiology of turtles. She has taught university level courses in Zoology, Animal Diversity, General<br />
Biology, Scientific Communication, Comparative Vertebrate Anatomy & Physiology, Animal Behavior,<br />
Vertebrate Zoology, Conservation Biology, and Vertebrae Ecology. During this course, Dr. Kalb will be<br />
your primary expert on sea turtles and the fields of zoology, vertebrate anatomy, physiology.<br />
Bruce A. Schulte, Ph.D. Visiting Scientist, is the Biology Department Head at Western Kentucky<br />
University. He received his Ph.D. from the State University of New York, College of Environmental<br />
Science and Forestry in 1993; his M.S. in Biology from University of Southern California, and his B.S.<br />
from the College of William and Mary. He has regularly taught courses in Animal Behavior, Behavioral<br />
Ecology, Chemical Ecology, Conservation, and Environmental Biology. His research interests include<br />
communication and social behavior of herbivorous mammals, such as elephants, manatees, beavers and<br />
horses. His research group also examines how an understanding of behavior can facilitate positive<br />
human-animal interactions, such as reducing human-wildlife conflict. When in the field, Dr. Schulte will<br />
be your primary expert in the fields of animal behavior, chemical ecology, and conservation behavior.<br />
Jessica R. Young, Ph.D., Visiting Scientist is Associate Professor of Biology and Associate Vice<br />
President for Academic Affairs at Western State College of Colorado. She received her Ph.D. in<br />
Population Biology and<br />
Behavioral Ecology from Purdue University in 1994; and her B.A. in Ecology, Behavior, and Evolution<br />
from UC San Diego in 1988. Her general research interests integrate evolutionary theories of behavioral<br />
ecology and animal communication with applied aspects of conservation biology and wildlife<br />
management. For the past two decades, she has been working with a unique species of grouse, the<br />
Gunnison Sage-grouse, which was recognized by the AOU in 2000 as a distinct species based on<br />
physical, behavioral, and genetic traits and the first new bird species described in over 100 years. In the<br />
field, Dr. Young will be your primary expert in the fields of behavioral ecology and ornithology.<br />
Belize Field Course Briefing Page 5 4/4/<strong>2012</strong>
FIELD TRAINING AND ASSIGNMENTS<br />
During each field course, students are encouraged to<br />
become proficient in at least one aspect of the long-term<br />
research project and will have input on the assignments<br />
in which they would like to participate. For example,<br />
one participant might become proficient recording<br />
behavioral data, while another might be better at<br />
operating the Global Positioning System (GPS) or<br />
recording environmental data. Most days, students will<br />
spend 1/2 day (3-4 hours) on a small boat conducting<br />
surveys and observing manatee & dolphin behavior.<br />
Our data collection is generally done from 25-foot<br />
fiberglass research boats, equipped with a single 115 HP<br />
Yamaha four-stroke engine. We carry a mobile phone, a<br />
magnetic compass, GPS unit, fire extinguisher, First Aid<br />
kit, and life jackets for 13 passengers, including the<br />
captain. Other standard safety equipment includes an<br />
anchor with anchor line, swim ladder, pole, bailing<br />
bucket, and bimini top for shade. There is no head<br />
(toilet facility) on board. We relieve ourselves in the<br />
water or in a bucket.<br />
Skills and talents extremely helpful to the project<br />
include patience, flexibility, attentiveness, a love for<br />
watching animals, a passion for learning, a passion for<br />
living sustainably in an outdoor environment,<br />
swimming, snorkeling, good handwriting, positive<br />
problem solving, positive attitude, strong team spirit, respect for low-tech data collection methods, and the ability to<br />
design and create something from nothing.<br />
Students will be expected to participate in most of the following tasks:<br />
• Continuous scanning surveys: These include both boat surveys and point scans designed to search for manatees<br />
& dolphins within the study area. During boat surveys and point scans students are needed to actively watch for<br />
manatees, dolphins and signs of their presence, and to record survey and sighting data. During sightings,<br />
students will record additional data, including behavioral states, breath cycles, and movements, as well as data<br />
about other boats in the scan area.<br />
• Environmental variable recording: Air and water temperature, salinity, wind direction and speed, sea state,<br />
turbidity, water depth, and bottom type will all be recorded. During boat surveys, point scans, and/or whenever<br />
a manatee or dolphin is sighted, locations will be marked with a GPS unit and general location will be plotted<br />
on the field map. At the end of each point scan students will use field equipment to take these measurements.<br />
• Focal follows: During focal follows, which are 40-minute increments of time during which we focus on a single<br />
manatee or group of dolphins, students record manatee or dolphin behavioral states, breath cycles, movements,<br />
and any other activity in the area.<br />
• Underwater ID captures: These are attempted by the Principal Investigator (PI) only, using a digital video<br />
camera and underwater housing. During an underwater attempt, students record both the manatee and the PI<br />
movements on a focal follow data sheet.<br />
• Surface photo-identification: During dolphin focal follows, the PI photographs dolphins from the bow of the<br />
research vessel; students document details on data sheets. Students will have an opportunity to photo dolphins<br />
when conditions are appropriate.<br />
Note: Students should NOT expect to swim with the manatees and dolphins, as commercial swimming with these<br />
endangered animals is prohibited in Belize. The PI has special permission to get into the water to photograph and<br />
video-tape animals, but this permission does not extend to students.<br />
Belize Field Course Briefing Page 6 4/4/<strong>2012</strong>
ACCOMMODATIONS<br />
At Spanish Bay, students and staff will share dormitory style housing. Each room will consist of a combination of<br />
bunk beds, sleeping up to 12 per room. Shared bathroom facilities include conventional toilets, showers and sinks.<br />
Room assignments will be determined by gender. Private rooms for couples are not available. Rooms have<br />
screened windows to allow for ocean breezes and limit nocturnal insect visitors; however we recommend you<br />
consider bringing your own personal mosquito netting, as well as duct tape and string for hanging net over your<br />
bunk.<br />
Spanish Lookout Caye is powered by solar, wind, and diesel<br />
generators. Fresh water is precious and we mandate water<br />
conservation. We capture rainwater and recycle all waste water<br />
using a state-of-the-art tertiary recycling plant. Volunteers are<br />
taught to conserve water by taking one quick (island) shower<br />
per day, turning off the faucet while lathering soap, brushing<br />
teeth, etc. Drinking water is purchased from Belize and<br />
available 24-7 at designated dispensers.<br />
There is sufficient power to recharge batteries. We encourage<br />
students to bring digital cameras. We rely on you to take<br />
pictures of the project and we like to download them during the<br />
expedition. Please bring the appropriate charging devices (110<br />
volts AC, 60 Hz, flat two-pin plugs) and your data cable to<br />
share images with your classmates and instructors. Also, please<br />
bring rechargeable batteries; used disposable batteries must<br />
return home with you! Note that there is NOT adequate electricity for high voltage appliances such as curling irons,<br />
blow dryers, or coffee makers. Internet access and phone usage on the caye is limited to emergency use by the<br />
instructors. Many US and UK cell phones (except Sprint) work in Belize, but the international fees are generally<br />
expensive. SMS Texting works well and costs about $0.50 per message.<br />
NOTE: In <strong>2012</strong>, we will also spend a few days on Corozal Bay in northern Belize with the<br />
Sarteneja Alliance for Conservation & Development Home Stay Program and may visit other<br />
conservation sites.<br />
FOOD<br />
At Spanish Bay, we will eat in the Belize Adventure Lodge dining area. Due to the logistics of living on a<br />
mangrove island, there is a limited amount of food variety during the expedition. Some special diets may be<br />
accommodated if advanced notice is given. Generally we eat a lot of chicken! BE SURE TO RECORD ANY<br />
DIETARY RESTRICTIONS ON YOUR REGISTRATION FORM…INCLUDING, BUT NOT LIMITED TO<br />
VEGETARIAN, VEGAN, LACTOSE INTOLERANT, GLUTEN INTOLERANT, FOOD OR OTHER<br />
ALLERGIES. The goal is to make our diet as traditional as possible, in order to make food a part of your<br />
experience! Weekly menus may include eggs and chicken, beef, pork, and local seafood (rarely), seasonal fruits and<br />
vegetables, tortillas, rice and beans, beans and rice (there is a difference), and pasta dishes. Water is the staple drink,<br />
supplemented with fruit juices, coffee, and tea. Also, if there are foods you do not eat for any reason please let us<br />
know so we don’t waste food by putting items on your plate that you won’t eat. For example, tomatoes are<br />
expensive so if you don’t eat them, just let us know and the staff will not put them on your plate!<br />
Comfort foods/drinks such as chips, cookies, candy, sodas, beer, rum, and wine are available (at your own expense)<br />
on the caye. A Gift Shop with limited supplies is located on our caye and is generally open daily from 9-4. Below<br />
are examples of the foods you might expect during the expedition. Please bear in mind that variety depends on<br />
availability.<br />
Breakfast: Eggs, beans, breakfast meats, tortillas, pancakes, oatmeal, fresh fruits<br />
Lunch: Tortillas, pasta, fresh veggies, rice, beans, chicken, boiled eggs, fresh fruit<br />
Dinner: Rice, beans, chicken, beef, pasta, veggies<br />
Snacks: Oranges, bananas, papaya, pineapple, watermelon<br />
Beverages: Water, coffee, tea, fruit juices<br />
Belize Field Course Briefing Page 7 4/4/<strong>2012</strong>
ENVIRONMENTAL CONDITIONS<br />
SBCRC is situated on a 184-acre twin mangrove island, partially cleared and filled, but with over<br />
80% of the mangrove ecosystem intact, including the mosquitoes and sand flies. A mangrove<br />
island harbors many little hazards such as broken shells to cut your feet, and mangrove roots to<br />
trip you up, so it’s important to pay close attention to this rustic environment, which is free of<br />
paved roads and sidewalks.<br />
The sun is very strong here, and<br />
brief periods of intense rain are not<br />
uncommon during the field season.<br />
More extreme tropical storms and<br />
hurricanes traditionally occur from<br />
June through November with late-<br />
August, September, and October as<br />
the most active periods.<br />
In the event of a hurricane, we may<br />
evacuate the island and move inland<br />
for the duration of the storm. This<br />
has occurred three times during the<br />
14 years of this project.<br />
Belize Field Course Briefing Page 8 4/4/<strong>2012</strong>
Potential Hazards<br />
We take pride in our experience, training, and track record with respect to students’ health and safety. So even when<br />
it seems that we are “mothering” you too much, we do expect you to follow our advice regarding your healthy and<br />
safety at all times!<br />
Hazard Type Associated Risks and Precautions<br />
Climate Students must be prepared to spend long hours in hot, humid, and wet conditions. The<br />
tropical sun is very strong. Dehydration, sunburn, and other heat related illnesses are a risk.<br />
Insects Sand flies and mosquitoes can be problematic. Sand flies are believed to be a vector for<br />
leishmaniasis in some regions. Mosquitoes may transmit a number of diseases (see<br />
Diseases below). Bot flies are also found in Belize, and mosquitoes may transmit their<br />
larvae to human hosts where the larvae will grow and develop. This is not life threatening,<br />
but can be painful and unpleasant.<br />
Marine life Fire corals, several species of stinging jellyfish, sharks, sea urchins, lionfish, and other<br />
potentially dangerous marine organisms can be found in the study area. These can all give<br />
painful and occasional severe stings or bites, which may become infected if untreated.<br />
Those with a dangerous allergy to bee or wasp stings may have a similarly dangerous<br />
reaction to corals and jellyfish. The best prevention is to avoid and not touch the animals.<br />
Please bring your epi-pen if you are allergic to anything.<br />
Snorkeling All the inherent risks of snorkeling are obviously present, including the effects of<br />
environmental conditions, marine life and other risks specific to your own physical/medical<br />
history. Snorkeling is optional and will be conducted in seagrass beds and on coral<br />
patches. Volunteers who chose to snorkel should know how to do so safely without<br />
hyperventilating or kicking up the substrate.<br />
Boats We will be aboard a boat for most fieldwork. All the fiberglass boats should have ladders<br />
and a bimini cover for sun protection. However, in some instances we may use a boat that<br />
lacks a ladder or shade. Deck surfaces of boats will become slippery and may place you at<br />
risk of slips, falls, and injuries that result from these accidents.<br />
Disease Diseases found in tropical regions include malaria, dengue fever, filariasis, leishmaniasis,<br />
onchocerciasis, trypanosomiasis (Chaga’s disease), schistosomiasis, leptospirosis, rabies,<br />
brucellosis, hepatitis, and typhoid. Most diseases are prevented with basic safety cautions.<br />
Driving Driving conditions are considered poor by western standards and pose inherent risks.<br />
Students will not be permitted to drive during the expedition.<br />
Medical Conditions of Special Concern<br />
Students should be physically fit, competent swimmers, and comfortable spending 3-4 hours on a boat. Those with<br />
chronic back problems or seasickness will find working and riding in small boats very uncomfortable. If you suffer<br />
from seasickness and intend to treat this with either over-the-counter or prescribed medication, please discuss the<br />
use and side effects with your physician and notify the PI before fielding. Be sure to tell your doctor that you will<br />
be spending all day in the sun! Please also let the PI know what medications, if any, you are taking for seasickness<br />
and/or malaria prevention. Some prophylactics have severe interactions with sun exposure and must be avoided.<br />
Any conditions that interfere with or limit stamina in the water, balance, swimming, or breathing should be carefully<br />
considered. If you have a current ear or sinus infection, it should be fully healed prior to participation in snorkeling<br />
or SCUBA. If you are allergic to bee stings, you may be allergic to Cnidaria stings (jellyfish and corals) and must<br />
bring an Epi-kit. Visual acuity (corrected via glasses or contacts is fine) and good hearing are important.<br />
Conditions or medications that increase one’s light sensitivity or sunburn risk should be discussed with a physician.<br />
Belize Field Course Briefing Page 9 4/4/<strong>2012</strong>
HEALTH INFORMATION<br />
Routine Immunizations<br />
All volunteers should make sure to have the following up-to-date immunizations: DPT (diphtheria, pertussis,<br />
tetanus), polio, MMR (measles, mumps, rubella) and varicella (if you have not already had chicken pox). Please be<br />
sure your tetanus shot is current.<br />
Project Inoculations & Prophylactics<br />
The following are recommendations only. Medical decisions are the responsibility of each student and their doctor.<br />
Note that health conditions around the world are constantly changing, so keep informed and consult your physician,<br />
a local travel health clinic, the US Center for Disease Control (www.cdc.gov), the World Health Organization<br />
(www.who.int). Please consult your physician for guidance on inoculations if you intend to travel to other parts of<br />
the country.<br />
Typhoid<br />
Hepatitis A<br />
Hepatitis B<br />
These inoculations are recommended by the CDC for health reasons whenever you are<br />
traveling outside the USA.<br />
Malaria A malaria prophylaxis that allows exposure to the sun is recommended by the CDC<br />
Other Advice / Information<br />
• Malaria: Malaria is not present at the research site, but it is found within Belize. A prophylaxis is<br />
recommended by the CDC for all areas except Belize City. The risk is highest in the western and southern<br />
regions of the country, which you may visit on a recreational day.<br />
PACKING CONSIDERATIONS<br />
Remember to review the Packing Checklist at the end of this briefing.<br />
General Considerations<br />
Do not bring more luggage than you can carry and handle on your own. Space is also extremely limited in the dorm<br />
rooms. We recommend that you pack a carry-on bag with an extra set of field clothing (shorts, t-shirt, hat, swim<br />
suit, mask and snorkel) and personal essentials in the event that your luggage is lost and/or takes several days to<br />
catch up with you.<br />
Remember to bring old t-shirts and shorts that you don’t mind getting dirty and possibly ruining. Expect to wear the<br />
same shorts and t-shirts repeatedly due to lack of laundry facilities. Clothes get ruined in the field; you will NOT<br />
need any good/nice clothes at the research camp. You might want to save a clean t-shirt and shorts for visiting<br />
inland sites, but, you don’t need dress slacks or skirts. If you have side trips planned before/after your research trip,<br />
plan accordingly.<br />
We have found some diversity among our previous students as to what they think should be mandatory. It varies not<br />
only with temporal conditions on the island, but also with student comfort when traveling within the developing<br />
world. The Packing Checklist at the end of this briefing contains recommendations based on our experience with<br />
over 500 previous students & volunteers. Some folks will require more creature comforts, and others can do without<br />
some of the recommended items. All volunteers should read this entire briefing carefully, and those who have done<br />
a lot of traveling can use the information to pack according to their experience. Less experienced travelers should<br />
bring everything we recommend!<br />
Note: As the project is stationed on a mangrove island, any trash produced must be burned. There are few recycling<br />
facilities in Belize. Please help protect the environment by leaving disposable products and plastic packaging at<br />
home.<br />
Belize Field Course Briefing Page 10 4/4/<strong>2012</strong>
Cultural Considerations<br />
Belize is predominately a Christian culture. Shorts and t-shirts are fine for both men and women. Swimwear is<br />
appropriate for beaches, but not for Belize City, where shirts and shoes are recommended at all times.<br />
Essential Items<br />
While it is recommended that you pack as light as possible, the following items are essential for participation: bug<br />
repellent for mosquitoes, oil (Avon Skin-so-Soft , baby oil, or olive oil is recommended) for sand flies, sunscreen<br />
(15-45 SPF), hat, long-sleeved shirt/cover up for boat, swimsuit, 1-2mm wetsuit or dive skin, and field clothes.<br />
Please see the Packing Checklist for a complete list of what you will need to take with you. We recommend<br />
going through the list with a pen or pencil and marking off each required item right before you leave for your<br />
expedition.<br />
EMERGENCIES IN THE FIELD<br />
The researchers and their host, Hugh Parkey’s Belize Dive Connection, are concerned with your health and safety<br />
during the field course. Minor injuries will be treated onsite using Red Cross First Aid and DAN Marine First Aid<br />
procedures. We take a precautionary approach to minor illnesses and injuries and will insist on scheduling an<br />
appointment with a local doctor if the situation does not improve within 24-48 hours after First Aid treatment.<br />
Volunteers with major illnesses or injuries will be transported to Belize City for medical advice and treatment.<br />
There is always a boat available for transport in case of an emergency. In case of a life-threatening illness/injury,<br />
we will take the following steps:<br />
1) Insure that all students/staff are safe from further injury<br />
2) Give First Aid/CPR as necessary to stabilize the victim(s)<br />
3) Contact HP's Belize Dive Connection to report the situation<br />
4) Transport the victim(s) by boat to Belize City or Contact DAN if indicated<br />
5) Notify victim’s Emergency Contact at first opportunity<br />
6) Arrange transportation from the dock in Belize City directly to an emergency care facility, Belize Medical<br />
Associates<br />
7) Follow up with HP's Belize Dive Connection as soon as the victim is under professional medical care<br />
8) Follow up with the rest of the class to keep them informed<br />
9) Follow-up with victim’s Emergency Contact<br />
OTHER USEFUL INFORMATION<br />
• Our Host: Ms. Teresa Parkey, Hugh Parkey's Belize Adventure Lodge, PO Box 1818, Belize City, Belize,<br />
Central America. Tel: ++501-223-4526 or ++501.223-5086 Fax: ++501-610-5235, E-mail:<br />
hugh@belizediving.com<br />
• Do not book your flight until your registration and the course has been confirmed! Expected confirmation date:<br />
60 days prior to fielding.<br />
• Airport Code BZE: American Airlines and Continental Airlines fly into Philip S. W. Goldson <strong>International</strong><br />
Airport in Belize City daily; Delta and US Airways also have a more limited schedule. If you are flying from<br />
the West Coast of the US, you might consider TACA. Round-trip airfare is currently running between US$600-<br />
$800.<br />
• You must have a Passport for entry into Belize; if you are a US Citizen you will automatically be granted a 30day<br />
tourist VISA upon arrival. If you are not a US Citizen, or for additional information, visit the Belize<br />
Tourism Board Website for additional information: http://www.travelbelize.org/<br />
• Consult the US State Department Website for current information regarding travel to Belize:<br />
http://www.state.gov/p/wha/ci/bh/index.htm<br />
Belize Field Course Briefing Page 11 4/4/<strong>2012</strong>
• Consult the CDC for immunization recommendations: http://wwwnc.cdc.gov/travel/destinations/belize.aspx<br />
• If you plan to arrive before the rendezvous date or remain in Belize after the departure date, please contact Dr.<br />
Self-Sullivan, for additional information on travel and accommodations in Belize. You may find the Moon<br />
Handbook Belize 8 th Edition, by my friend and colleague Josh Berman, a valuable resource, available online at<br />
Amazon.com.<br />
• Recommended Field and Travel Guides:<br />
Birds of Belize (The Corrie Herring Hooks Series), By H. Lee Jones<br />
Moon Handbooks Belize, 8 th Edition, by Joshua Berman<br />
Travellers' Wildlife Guides Belize & Northern Guatemala, By Les Beletsky<br />
Reef Creature Identification: Florida, Caribbean, Bahamas by Paul Humann and Ned DeLoach<br />
Reef Coral Identification: Florida, Caribbean, Bahamas (Reef Set, Vol. 3) by Paul Humann and Ned<br />
DeLoach<br />
Reef Fish Behavior: Florida, Caribbean, Bahamas by Paul Humann and Ned DeLoach<br />
Reef Fish Identification: Florida, Caribbean, Bahamas by Paul Humann and Ned DeLoach<br />
• BELIZE time zone: GMT/UTC -6:00; Daylight Savings Time is NOT observed in Belize.<br />
• Local currency: Belize dollars, however, US dollars are accepted everywhere for a fixed exchange rate of<br />
US$1 = BZ$2. There is NO NEED to change US$ into BZ$.<br />
• Electricity: 110 volts AC, 60 Hz, flat two-pin plugs (same as USA). Electricity is only minimally available at<br />
the project site, although charging of laptops, digital cameras and some small electronics is fine. Students DO<br />
NOT have Internet access during the course!<br />
• Language: English, also spoken: Spanish, Maya, Garifuna, Creole<br />
• Telephone dialing codes: When calling Belize from another country, dial the country’s international dialing<br />
code (e.g., 011 in the USA), followed by 501 (Belize Country Code) and the number (e.g., 223.4526). When<br />
calling within Belize, omit the 501, and just dial the number. When calling another country from Belize, dial<br />
00, followed by the other country’s country code and the number. For example, to call the USA, dial 001+area<br />
code+number. I will rent a local cell phone upon arrival and text you the number so you can contact me during<br />
your travel is necessary!<br />
• Personal conduct: The field course and project are able to remain in Belize due to the courtesy of the people<br />
and government of Belize. As such, you will be expected to conduct yourselves in a manner that is respectful of<br />
local sensitivities, customs, and laws. Any violations of Belizean law will be prosecuted in Belize with no<br />
recourse to foreign laws and attorneys. Any conduct that reflects negatively on project will be grounds for<br />
immediate deportation at the expense of those involved.<br />
• Personal funds: Past students have spent a wide range of funds during the course. You will have opportunities<br />
to shop on at least one of your recreational days. The airport also offers many gift shops in the departure area.<br />
Small bills are useful (US$1, US$5, US$10) as change will be given in BZ dollars. Traveler’s checks and credit<br />
cards are more difficult to use, but OK at more and more locations each year. Debit cards may be problematic,<br />
but VISA and MasterCard credit cards have been used to get cash advances from banks in Belize City. There<br />
are ATMs, but they appear tied to VISA credit cards only. Larger stores, restaurants and hotels take VISA and<br />
MasterCard credit cards, but few take American Express. Smaller shops and street vendors only take cash.<br />
• Tips: As a visitor, it is customary to tip for services in Belize. You are responsible for tipping anyone you think<br />
gave exceptional service, including but not limited to the field assistants, housekeepers, restaurant staff,<br />
bartenders, dive master, drivers, tour guides, and boat captains during recreational activities. We generally<br />
“pass the basket” for tips at the end of each team.<br />
Belize Field Course Briefing Page 12 4/4/<strong>2012</strong>
Essential Items<br />
PACKING CHECKLIST<br />
Photocopies of your passport, flight itinerary and credit cards in case the originals are lost or stolen; the<br />
copies should be packed separately from the original documents<br />
Passport and Driver’s License and other Photo ID such as your Student ID<br />
Required Items Necessary for Successful Completion of Coursework<br />
Laptop Computer with USB port (for pen drive)<br />
Digital Camera and Underwater Housing<br />
Copies of articles from your Literature Review<br />
Copies of all journal articles (included in this briefing)<br />
Clothing/Footwear for Fieldwork<br />
Old shorts & old t-shirts<br />
Lightweight, breathable long<strong>–</strong>sleeved shirts/pants for protection from sun/bugs<br />
Sweatshirt, sweatpants, socks in case you get chilled from being in water all day<br />
2-3 bathing suits (things don’t always dry over-night in the tropics)<br />
Well worn-in and comfortable walking shoes (sneakers or Teva-like sandals)<br />
Boat shoes (water shoes or bare feet are recommended on the boat)<br />
Rain gear or rain poncho<br />
Hat with wide brim to protect head from sun (very important!)<br />
Clothing/Footwear for Leisure<br />
One set of clothing to keep clean for day-off and end of expedition<br />
Field Supplies<br />
*****Polarized***** sunglasses with retaining strap<br />
Bound Field Journal & Copies of Journal Articles<br />
Small daypack/rucksack/backpack for inland trip<br />
Dry bag or heavy duty plastic “zip lock” bags for boat (for protecting equipment such as camera from sand,<br />
humidity, and water)<br />
Water bottle - 1 liter refillable, such as Nalgene<br />
Mask, snorkel, and fins if you want to participate in tasks requiring snorkeling, or for snorkeling during free<br />
time (we encourage you to invest in good quality mask and fins, not the department store variety)<br />
Dive skin or 1-2mm wetsuit<br />
Mosquito repellent (with Deet)<br />
Belize Field Course Briefing Page 13 4/4/<strong>2012</strong>
Oil or oil-based repellant (e.g. olive oil, AVON Skin-so-Soft Original Bath Oil, citronella oil repellent, Bit<br />
Blocker) for sandflies (oil creates a physical barrier from the sand flies; the researchers have found NOTHING<br />
except oil prevents the sand flies from feasting on you when/if the wind dies)<br />
A box of mosquito coils (commonly called “fish” in Belize) and a lighter for burning in your room at night to<br />
keep both mosquitoes and sandflies away<br />
Waterproof sunscreen with SPF 30 or higher<br />
Lip balm with SPF 30 or higher, also Blistex ointment if you are prone to fever blisters<br />
A notebook for your personal field notes and a couple of pencils (unlike ink pens, pencils continue to write even<br />
if they get wet)<br />
Personal Supplies<br />
Toiletries, such as a bar biodegradable soap, deodorant, toothpaste, toothbrush, shampoo, conditioner, and<br />
moisturizing lotion (no-fragrance types may help reduce the attraction of mosquitoes)<br />
Antibacterial wipes or lotion (good for “washing” hands while in the field)<br />
Personal medications, such as vitamins, prescription drugs, emergency allergy injections (Epi-kit), inhalants,<br />
motion sickness medication and over-the-counter antihistamines and anti-itch products<br />
Extra contact lenses and saline solution and/or spare glasses (you will need your reading glasses if you wear<br />
them)<br />
Miscellaneous<br />
Cash for snacks, gifts, tips, etc.<br />
Camera, film/memory cards, extra camera battery/battery charger, and computer cable for downloading images<br />
from digital cameras<br />
Laptop computer, power cord, power strip, flash drive or pen drive<br />
Cell phone with SMS Text Capability if you want to send and receive messages from home<br />
Optional Items<br />
Flashlight/torch or headlamp with extra batteries and extra bulb<br />
Earplugs (your roommates might snore!)<br />
Duct tape and roll of heavy string<br />
Very fine gauge mosquito net for hanging over your bunk (also to keep the sandflies out)<br />
Snack food<br />
Paperback books/field guides<br />
Small battery-operated fan with additional batteries<br />
Battery-operated tape/CD/MP3 player/recorder with headphones (NOT allowed on research boat)<br />
Binoculars<br />
Playing cards, board games, musical instruments, etc.<br />
Belize Field Course Briefing Page 14 4/4/<strong>2012</strong>
APPENDIX<br />
<strong>2012</strong> Course Flyer, Syllabus, Registration, & Policy Forms<br />
Field Journal Requirements<br />
Final Project Sample Fact Sheets<br />
Red Mangrove Root Crab<br />
Cushion Sea Star<br />
Fiddler Crab<br />
Patch Reef Habitats<br />
Brown Pelican<br />
Variegated Sea Urchin<br />
Red Fin Needlefish<br />
Fiddler and Hermit Crabs<br />
Cushion Sea Star<br />
Cushion Sea Star<br />
Magnificent Feather-Duster<br />
Marine Hermit Crab<br />
Variegated Sea Urchin<br />
Data Sheets for Manatee and Dolphin Research Project<br />
Faculty Curriculum Vitae<br />
Journal Club Readings<br />
Mortimer et al. 2007. Whose turtles are they, anyway? Molecular Ecology 16:<strong>17</strong>-18<br />
Blumenthal et al. 2009. Turtle groups or turtle soup: dispersal patterns of hawksbill turtles in<br />
the Caribbean. Molecular Ecology 18:4841-4853.<br />
Self-Sullivan. 2000. The Elusive Manatee: An ethological approach to understand behavior of<br />
the West Indian manatee in Belize.<br />
Self-Sullivan et al. 2003. Seasonal occurrence of male Antilllean manatees on the Belize<br />
Barrier Reef. Aquatic Mammals 29.3:342-354.<br />
Grimm. 2010. Is a Dolphin a Person? News Briefs, Science Magazine, 18-22 February 2010.<br />
Noren. 2008. Infant carrying behaviour in dolphins. Functional Ecology 22:284-288<br />
Miksis-Olds et al. 2007. Simulated vessel approaches elicit differential responses from<br />
manatees. Marine Mammal Science 23.3:629-649.<br />
Mann. 1999. Behavioral sampling methods for cetaceans: a review and critique. Marine<br />
Mammal Science 15.1:102-122.<br />
LaCommare et al. 2008. Distribution and habitat use of Antillean manatees in the Drowned<br />
Cayes, Belize. Aquatic Mammals 34.1:35-43.<br />
Kerr et al. 2005. Bottlenose dolphins in the Drowned Cayes, Belize. Caribbean Journal of<br />
Science 41.1:<strong>17</strong>2-<strong>17</strong>7.<br />
Jackson. 1997. Reefs since Columbus. Coral Reefs 16:S23-S32<br />
Harper & Schulte. 2005. Social interactions in captive female Florida manatees. Zoo Biology<br />
24:135-144.<br />
Belize Field Course Briefing Page 15 4/4/<strong>2012</strong>
Ecology, Behavior, and Conservation of Manatees & Dolphins<br />
A Unique Field Course in the Belize Barrier Reef Lagoon System<br />
<strong>2012</strong> Session III ~ 4 - <strong>17</strong> August <strong>2012</strong><br />
Lead Instructor & Principal Investigator: Caryn Self-Sullivan, Ph.D. 1,2<br />
Co-PI: Katie LaCommare, Ph.D. 2,3 | Visiting Faculty: TBA<br />
1 Nova Southeastern University, 2 <strong>Sirenian</strong> <strong>International</strong>, 3 Lansing Community College<br />
Want to be a Conservation Biologist, Behavioral Ecologist or Marine Mammalogist?<br />
Here's your chance to join our research team for two intense weeks of total immersion<br />
into the world of animal behavior, ecology & conservation, Antillean manatees,<br />
bottlenose dolphins, coral reefs, mangroves and seagrass beds in Belize!<br />
Course Overview: This is an experiential learning field course where you will live, work, and study<br />
from a marine science field station on a pristine, private island off the coast of Belize. Additionally, you<br />
will visit one or more Community Conservation Sites in Belize. Data collected during the course will<br />
contribute to our long-term manatee/dolphin research project. You will learn through a variety of<br />
learning activities, literature review and discussion, independent research projects, and actual field<br />
research. Be prepared to rise with the sun and spend 8-10 hours outdoors, including 3-4 hours on the<br />
water each day learning about the tropical Caribbean environment as we explore a maze of mangrove<br />
islands, seagrass beds, and coral patches searching for elusive manatees and charismatic dolphins.<br />
Location: Spanish Bay Conservation & Research Center at Hugh Parkey's Belize Adventure Lodge,<br />
http://belizeadventurelodge.com/ and Sarteneja Alliance for Conservation & Development,<br />
http://sartenejaconservation.org/. Passport required, immunizations as recommended by CDC<br />
Your Share of the Costs: US$2995 includes housing, meals, ground & water transfer fees, research &<br />
materials fees; DOES NOT include airfare, books, tips, and credit hours.<br />
Credit Hours: The course provides 100 experiential learning and lecture hours in the field, plus<br />
approximately 35 hours of pre-fielding research and preparation, is comparable to a 3 credit hour<br />
university course and meets the US DOE criteria in 34 CFR, §600.2. Deadlines: Early registration &<br />
and deposit due February 1st, <strong>2012</strong>; regular registration & deposit due March 1st, <strong>2012</strong>; balance due at<br />
least 60 days prior to field dates. Late payments and late registrations (if space available) incur a $100<br />
late fee.<br />
Information Contact: Caryn Self-Sullivan, Ph.D. | +1.540.287.8207 | cselfsullivan@sirenian.org<br />
=========================================================================<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207<br />
cselfsullivan@sirenian.org<br />
540.287.8207
Ecology, Behavior, and Conservation of Manatees & Dolphins in Belize<br />
A Unique Field Course in the Drowned Cayes, Belize<br />
Summer <strong>2012</strong> SYLLABUS & TENTATIVE SCHEDULE<br />
Session III: ~ 4 <strong>–</strong> <strong>17</strong> August, <strong>2012</strong><br />
Visit us on Facebook: http://www.facebook.com/event.php?eid=370432825564<br />
Lead Instructor & PI<br />
Caryn Self-Sullivan PhD<br />
NSU & <strong>Sirenian</strong> <strong>International</strong><br />
Phone: 540.287.8207<br />
Email: cselfsullivan@sirenian.org<br />
or cs<strong>17</strong>33@nova.edu<br />
Co-PI: Katherine LaCommare PhD<br />
Lansing Community College &<br />
<strong>Sirenian</strong> <strong>International</strong><br />
Phone: 248-756-3985<br />
Email: kslacommare@gmail.com<br />
or katie.lacommare@umb.edu<br />
Want to be a Conservation Biologist, Behavioral Ecologist or Marine Mammalogist? Here's your chance to<br />
join our research team for two intense weeks of total immersion into the world of animal behavior, ecology &<br />
conservation, Antillean manatees, bottlenose dolphins, coral reefs, mangroves and seagrass beds in Belize!<br />
Course Overview: This is a total immersion, experiential learning, field course where you will live, work, and<br />
study from a marine science field station on a pristine, private island off the coast of Belize. Additionally, you<br />
will visit one or more Community Conservation Sites in Belize. Data collected during the course will contribute<br />
to our long-term manatee/dolphin research project established in 1998. You will learn through a variety of<br />
learning activities, literature review and discussion, independent research projects, and actual field research. Be<br />
prepared to rise with the sun and spend 8-10 hours outdoors, including 3-4 hours on the water each day learning<br />
about the tropical Caribbean environment as we explore a maze of mangrove islands, seagrass beds, and coral<br />
patches searching for elusive manatees and charismatic dolphins. Optional SCUBA dives are available as time<br />
and weather permits; NOTE: additional costs are involved for SCUBA diving.<br />
PRIMARY LOCATION: Spanish Bay Conservation & Research Center at Hugh Parkey's Belize Adventure<br />
Lodge, http://belizeadventurelodge.com/ (Passport required, immunizations as recommended by CDC)<br />
SECONDARY LOCATION(s): Corozal Bay Wildlife Sanctuary, co-managed by the Sarteneja Alliance for<br />
Conservation & Development, Sarteneja. Alternative 2ndary locations include: Swallow Caye Wildlife<br />
Sanctuary, co-managed by Friends of Swallow Caye, Caye Caulker; Gales Point Wildlife Sanctuary, Gales Point<br />
Manatee & Southern Lagoon; South Water Caye Marine Reserve (& Placencia Lagoon), co-managed by SEA<br />
Belize (formerly Friends of Nature & TASTE); Port Honduras Marine Reserve, co-managed by the Toledo<br />
Institute for Development and the Environment.<br />
COSTS: US$2995 includes housing, meals, ground & water transfer fees, research & materials fees; DOES NOT<br />
include airfare, books, tips, and credit hours.<br />
OPTIONAL CREDIT HOURS: The course provides 100 experiential learning and lecture hours in the field,<br />
plus approximately 35 hours of pre-fielding research and preparation; at least 45 of the 135 total hours include<br />
direct instruction by faculty. This is comparable to a 3 credit hour university course and meets the US DOE<br />
criteria in 34 CFR, §600.2. You must make arrangements IN ADVANCE with BOTH your advising faculty and<br />
Dr. Self-Sullivan for credit to be earned through your home university. Credit hour fees must be paid directly to<br />
your school and you must fulfill any study abroad requirements of your school. This course is divided into 4<br />
major components: short lectures and learning activities (~1 hour per day), independent reading and assignments<br />
(~2 hour per day), supervised data collection in the field (~3 hours per day), independent project development &<br />
implementation (~2 hours per day), presentation of pre-field research (~1 hour per day), and debate/group<br />
discussion of reading materials (~1 hour per day).<br />
DEADLINES: Early registration & and deposit due February 1st, <strong>2012</strong>; regular registration & deposit due March<br />
1st, <strong>2012</strong>; balance due at least 60 days prior to field dates. Late payments and late registrations (if space<br />
available) incur a $100 late fee. If you are registering through your home university, earlier deadlines may<br />
exist—see your academic advisor!<br />
Marine Mammal Ecology, Behavior, Conservation Field Course, Summer <strong>2012</strong>, Page 1 of 5, Revised 4/4/<strong>2012</strong>
MINIMUM / MAXIMUM CLASS SIZE: 6-24 students<br />
WEBPAGE: http://www.sirenian.org/<strong>2012</strong>FieldCourse.html<br />
Course Objectives: The objectives of this course are to introduce you to the basic concepts animal behavior,<br />
ecology, and conservation through participation in an ongoing research project, development and implementation<br />
of your own independent research project; research and presentation on Belizean culture or natural history,<br />
reading and thinking critically about peer reviewed literature related to course content, and participation in a<br />
Service-Learning Project.<br />
<strong>COURSE</strong> CONTENT (Readings, Lectures, Learning Activities) SUBJECT TO CHANGE DEPENDING ON<br />
FACULTY<br />
• Introduction to Trichechus manatus & Tursiops truncatus<br />
• Participation in long-term research project, results to date, questions for the future<br />
• Introduction to coral reef, mangrove & seagrass ecology<br />
• Introduction to Animal Behavior & Conservation<br />
RESEARCH ACTIVITES<br />
• Manatees and Dolphins: point transect scans, focal follows, reconnaissance surveys, photo-id<br />
• Habitat Sampling: seagrass identification to species level, percent cover, density, comparison of<br />
epibionts, core sampling, feeding scars.<br />
• Individual or team research project: select from list of approved study subjects and topics below<br />
Daily schedule: Our schedule is closely tied to sunrise/sunset and tides, with a tentative schedule, below. You<br />
will be actively engaged from 06:00 until 21:00, daily. Morning and afternoon learning activities and research<br />
will be rotated as necessary.<br />
Time Daily Schedule<br />
6:00 Observing Wildlife, Individual Project Data Collection<br />
7:30 Breakfast<br />
8:30 Learning Activities (alternating AM & PM)<br />
12:30 Lunch<br />
13:30 Field Research (alternating AM & PM)<br />
18:00 Journal Club & Happy Hour<br />
19:00 Dinner<br />
20:00 Belize Sea to Stars Presentations (20-40 minutes)<br />
22:00 Quiet Time in the Dorms (Reading/Writing OK)<br />
23:00 Curfew and Lights Out<br />
Books & Tools<br />
We recommend the following books, which are available from Amazon.com:<br />
The Florida Manatee: Biology and Conservation, by Roger L. Reep and Robert K. Bonde, University Press of<br />
Florida<br />
The Bottlenose Dolphin: Biology and Conservation, by John E. Reynolds, Samantha D. Eide, and Randall S.<br />
Wells, University Press of Florida<br />
A selection of primary literature is included in the Field Briefing Document. Students are required to print<br />
out and read this document prior to fielding AND bring it with you to Belize. We encourage you to have this<br />
done at a print shop, such as FedExOffice-Kinkos and bound with a 1” wire ring (~$35). There is also a library of<br />
books and archived journals, including Marine Mammalogy, Animal Behavior, and Conservation Biology<br />
journals, in our library onsite in Belize.<br />
Marine Mammal Ecology, Behavior, Conservation Field Course, Summer 2011 Page 2 of 5, Revised 4/4/<strong>2012</strong>
Communication: If you have any questions or concerns, it is your responsibility to communicate with us via<br />
email or phone prior to travel and in person while at the field site. As your professors, we are here to help you<br />
learn. But learning is your responsibility.<br />
CONTACT: Dr. Caryn Self-Sullivan, cselfsullivan@sirenian.org, US +1.540.287.8207, BZ +501.636.3849<br />
Student Expectations: You should expect us to be enthusiastic about the course material and to be available for<br />
one-on-one help and feedback. You should expect us to challenge you to think about how & why the course<br />
material is relevant to you and your career goals. You should expect us to respect you as a person, regardless of<br />
your academic performance.<br />
Our Expectations: We expect you to complete all pre-field requirements and readings by the due dates posted in<br />
the briefing. We expect you to give respect to everyone in Belize, especially at our field site. We expect you to<br />
fully participate in every course activity, including being present and on time for all meals. We expect you to<br />
demonstrate enthusiasm and respect for all living organisms in the tropical marine environment.<br />
Grading Structure: Your final grade is based on a 500 point scale. To determine your “letter” grade, divide<br />
points earned by possible points: A > 90%; B > 80%; C > 70%; D > 60%.<br />
• *Independent Research Project (100 pts): Select a topic from approved list** before fielding; do<br />
background research before departure; design and implement a short-term research project during field<br />
course; write up results in format of Fact Sheet (see examples from previous students).<br />
• Field Journal & Reflections (100 pts): Purchase and bring a bound notebook to be used as a field<br />
journal during the course. Make daily entries following guidelines presented in course; include a 101<br />
species list.<br />
• Participation in field and course activities (100 pts): Includes evaluation of your participation in all<br />
course activities, including discussions, quiz bowls, and personal essay “Why Marine Mammals?” Be on<br />
time and demonstrate enthusiasm for learning for full points.<br />
• *Paper discussion and leading (100 pts): Read all papers prior to coming into the field; review in field<br />
and participate in all discussions; present and lead the discussion as assigned.<br />
• Service-Learning (50 pts): Either as a class or in mini-groups we will contribute the improvement of the<br />
island or regional environment. Projects could include picking up garbage in the mangroves, creating<br />
signage material for education, helping in the Swallow Caye refuge, participating in stranding or<br />
necropsy, giving a presentation at a local school or civic group, giving a talk to island staff.<br />
• *Belize from Sea to Stars (50 pts): Each student will prepare a presentation before arrival on some<br />
aspect of Belize. Topics could include the following**: Art, Astronomy, Culture, Economics, Ecology,<br />
Geology, History, Languages, Music, Politics and others. After dinner, a student will give an oral<br />
presentation on their topic. They may prepare handouts but this will not be a power point talk. For full<br />
points, there should be some interactive activity such as a quiz bowl, role playing, or crossword puzzle.<br />
These should be informative and interesting to a general audience. The island staff will be invited to join<br />
us when this is feasible.<br />
* Requires pre-course reading, research, and communication **See suggested topics at end of syllabus<br />
Missed and Late Assignments: Each Assignment (pre-fielding and in the field) will have a firm due date and<br />
time. If you miss the due date/time, you may turn in your assignment late. However, your grade will be reduced<br />
by 10% per day until the last day of the course. Any assignments not completed by the last day of the course will<br />
receive zero credit.<br />
Academic Dishonesty & Plagiarism: Students found violating the conditions of academic honesty will<br />
receive an F in the course. DO NOT copy & paste text from the Internet or other sources into your<br />
assignments; DO NOT copy work from your peers; DO NOT paraphrase someone's ideas without giving<br />
appropriate credit to the source. Plagiarism is a serious offense and will be treated as such. For more<br />
information on plagiarism see http://en.wikipedia.org/wiki/Plagiarism.<br />
Marine Mammal Ecology, Behavior, Conservation Field Course, Summer 2011 Page 3 of 5, Revised 4/4/<strong>2012</strong>
<strong>COURSE</strong> FEE & ADDITIONAL INFORMATION<br />
The course fee of $2995 includes all transportation, meals, and accommodations from your arrival at the PWG<br />
airport in Belize on the first day of the course to your departure on the last day of the course. Additionally, you<br />
are responsible for tips, insurance, credit hours, and round-trip airfare to Belize (airport code BZE).<br />
EARLY REGISTRATION DEADLINE: 1 February <strong>2012</strong><br />
REGULAR REGISTRATION DEADLINE: 1 March <strong>2012</strong><br />
LATE PAYMENTS OR LATE REGISTRATION: $100 Late Fee<br />
To register for course, complete, sign, and send these forms: http://sirenian.org/<strong>2012</strong>BelizeRegistration.pdf<br />
http://sirenian.org/<strong>2012</strong>BelizeSyllabus.pdf<br />
http://sirenian.org/<strong>2012</strong>BelizePolicy.pdf<br />
Download, print, bind, READ, and bring to Belize: http://sirenian.org/<strong>2012</strong>BelizeBriefing.pdf<br />
Recruit a friend: http://sirenian.org/<strong>2012</strong>BelizeFlyer.pdf<br />
For more information on the course, please email: cselfsullivan@sirenian.org<br />
For more information on the facilities, please visit: http://belizeadventurelodge.com<br />
http://sartenejaconservation.org/<br />
PRINT STUDENT NAME: ___________________________________________________________________<br />
I HEREBY CERTIFY THAT I HAVE READ AND FULLY UNDERSTAND THIS SYLLABUS AND THE<br />
ADDITIONAL REQUIRED DOCUMENTS LISTED ABOVE.<br />
STUDENT SIGNATURE: ____________________________________________________________________<br />
MY CHOICES FOR BELIZE SEA TO STARS AND INDEPENDENT RESEARCH TOPICS ARE:<br />
RESEARCH 1 ST CHOICE:<br />
RESEARCH 2 ND CHOICE:<br />
RESEARCH 3 RD CHOICE:<br />
BELIZE 1 ST CHOICE:<br />
BELIZE 2 ND CHOICE:<br />
BELIZE 3 RD CHOICE:<br />
Questions, Concerns, Feedback, Comments:<br />
Marine Mammal Ecology, Behavior, Conservation Field Course, Summer 2011 Page 4 of 5, Revised 4/4/<strong>2012</strong>
Topics for Independent Research Project (100 pts): Select a species and/or ecosystem from list below & get<br />
approved by Dr. C BEFORE fielding; do background research (primary literature search) in advance so<br />
that you can design and implement a short-term research project during field course; write up results in<br />
format of Fact Sheet (see examples from previous students).<br />
Seagrass ecosystem: Abundant seagrass beds dominated by Thalassia testudinum, with Syringodium filiform,<br />
Halodule spp., surround the caye, providing an excellent environment for snorkel based investigations. Potential<br />
projects include ecology or behavior of various organisms within these systems. Suggested species include sea<br />
stars (Oreaster reticulates), sea urchins (Lytechinus spp.), grunts (Haemulon spp.), needle fish (Strongylura spp.).<br />
Shallow subtidal mangrove peat banks surround the caye and have well-developed communities of fleshy algae<br />
(e.g., Caulerpa spp., Dictyota spp.) and calcified algae (e.g., Halimeda spp., Jania spp.) with sparse T.<br />
testudinum, rose corals (Manicina areolata), and corkscrew anemones (Bartholomea annulata) with symbiotic<br />
cleaner shrimp (Periclimenes spp.).<br />
Mangrove ecosystem: The caye is dominated by large perimeter and dwarf central red mangroves (Rhizophora<br />
mangle) and 2 significant black mangrove (Avicennia germinans) forests, with a scattering of white mangroves<br />
(Laguncularia racemosa), a few buttonwoods (Conocarpus erectus), and planted coconut palms, providing<br />
abundant habitat for hermit crabs (Coenobita clypeatus) and fiddler crabs (Uca spp.); we also sight mangrove<br />
crabs (Aratus pisonii), snails (Littoraria angulifera), mangrove root crabs (Goniopsis cruentata), and ghost crabs<br />
(Ucides cordatus), mangrove rivulus fish (Rivulus marmoratus), and giant termite nests (Nasutitermes spp.),<br />
unidentified bat species locally called rat bats that are thought to roost in the coconut palms during the day but can<br />
be observed foraging on insects at dusk and dawn, and many birds including the mangrove warbler, golden<br />
fronted woodpecker, osprey, common black hawk, egrets, and herons (see bird list). The central pond of on this<br />
caye is strongly influenced by diurnal tides, fluctuating from a mud flat at low tide to about 30cm of water at high<br />
tide and would make a good independent variable for natural experiments.<br />
Belize from Sea to Stars (50 pts): Each student will prepare a presentation before arrival on some aspect of<br />
Belize. Topics could include the following: Art, Astronomy, Culture, Economics, Ecology, Geology, History,<br />
Languages, Music, Politics and others. After dinner each night, one or more students will give oral presentations<br />
on their topic. This is NOT a Power Point presentation! Ideally, there should be some interactive nature such<br />
as stargazing, quiz bowl, role playing, or crossword puzzle. These should be approximately 20-40 minutes long,<br />
informative and interesting to a general audience. The island staff will be invited to join us when this is feasible.<br />
Once you have selected your topic, you must register it with Dr. C to prevent replication of topics by students.<br />
For some interesting issues in Belize search the following topics in association with Belize:<br />
Lethal Yellowing Disease, Coral Bleaching<br />
Business Ventures: Chalillo Dam, BEL, Fortis; Cruise<br />
Ship Industry explosion; sugar and banana industries<br />
Glow Worms and Bioluminescence (Odontosyllis<br />
luminosa, Dr. Gary Gaston); Pica Pica and Thimble<br />
Jellies<br />
Parliamentary Government: UDP vs. PUP<br />
Meteor Showers, the Southern Cross and other<br />
Astronomical Views<br />
Culture: Maya, Garifuna, Kriole/Creole, Mestizo,<br />
Mennonites<br />
Geography: 5 Oceanic Ridges and the Atolls<br />
Local NGOS: BACONGO, TIDE, TASTE, Friends of<br />
Nature, Belize Audubon, Wildtracks, Sharon Matola and<br />
the Belize Zoo<br />
Some Famous People from Belize: Baymen of Belize and<br />
the <strong>17</strong>98 Battle of St. George’s Caye, Baron Bliss, Andy<br />
Palacio, George C. Price, Marion Jones, Jamal Shyne<br />
Barrow, Evan X Hyde, Zee Edgell, Milt Palacio, Said<br />
Musa, Dean Barrow<br />
Conservation in Belize: Protected Areas, Hicatee Turtles,<br />
Coral Reefs, Harpy Eagle, etc.<br />
Xaté (sha-tay): leaves from three Chamaedorea palm<br />
species (C. elegans, C. oblongata and C. ernesti-augustii)<br />
used in the floral industry<br />
Belize-Guatemala Border Dispute<br />
Music: Stonetree Records<br />
Destinations: Caye Caulker, San Pedro, Placencia,<br />
Orange Walk, Punta Gorda, Dangriga, Belmopan,<br />
Corozal, Bermudian Landing, San Ignacio, Benque Viejo,<br />
Hopkins, Crooked Tree, Sarteneja<br />
Maya Archeological Sites: Xunantunich, Cahal Pech,<br />
Lamani, Altun Ha, Caracol, El Pilar, Lubaantun, Nim Li<br />
Punit, Actun Tunichil Muknal, Cerros, Tikal (Guatemala)<br />
Food: Sere, Hudut, Escabeche, Rice and Beans, Beans<br />
and Rice, Cowfoot Soup, Johnny Cakes, Creole Bread,<br />
Powder Buns, Black Cake, Dukunu, Meat Pies, Relleno,<br />
Stew Chicken, Garnaches, Salbutes, Panades, Tambrand<br />
Candies, Gibnut, Bamboo Chicken, Marie Sharp’s, Boilup,<br />
etc.<br />
Invasives: Lionfish Pterois volitans, ?P. miles<br />
Marine Mammal Ecology, Behavior, Conservation Field Course, Summer 2011 Page 5 of 5, Revised 4/4/<strong>2012</strong>
Ecology, Behavior & Conservation of Manatees and Dolphins<br />
A Unique Field Course in the Drowned Cayes, Belize<br />
Session III: August 4-<strong>17</strong> <strong>2012</strong><br />
Step 1: PRINT & COMPLETE THIS FORM - ALL PRICES ARE IN $US<br />
Step 2: Mail, Fax, or Scan & Email the SIGNED registration, syllabus, transcripts & policy forms<br />
Step 3: I will send you and invoice, which you can pay online or by mail<br />
Mailing Address, Email Address, and Fax:<br />
Caryn Self-Sullivan, 200 Stonewall Drive, Fredericksburg, VA 22401-2110<br />
Phone: 540.287.8207 | Fax: 888.371.4998 (Country Code 001)<br />
Email: Caryn Self-Sullivan (cselfsullivan@sirenian.org)<br />
Name:<br />
Mailing Address:<br />
City, State, Postal Code, Country:<br />
Email Address(es):<br />
Cell & Home Phone #s:<br />
(cell) (home)<br />
Why do you want to participate in this course? You do NOT need to be currently enrolled in school, however describe your academic status & highest<br />
degree. Don't forget to send unofficial copies of your transcripts and answer questions on page 2:<br />
Openings available only in Session III Fee<br />
Session I: 26 May - 8 June <strong>2012</strong> $2,995<br />
Session II: 16 June - 29 June <strong>2012</strong> $2,995<br />
Session III: August 4-<strong>17</strong> <strong>2012</strong> $2,995<br />
Note: If you register for more than one session, you will be responsible for expenses b/t sessions!<br />
Total Cost (excludes airfare, credit hours & tips)<br />
DEPOSIT (Minimum $500) Due now; I will invoice you upon receipt of registration!<br />
Low-income Country Scholarship* $500<br />
Early Registration & Recruiting Discounts* (Max $100)<br />
Late Payment / Late Registration Fee $100<br />
Balance due at least 60 days prior to fielding<br />
* Attach a 1500 word essay describing how this opportunity will enhance both your personal career goals AND<br />
sirenian research, educational outreach, and conservation in your home country<br />
** $100 discount if you register and submit $500 deposit by February 1st; $25 discount each for you and a friend, for<br />
each friend you recruit, provided you and your friend(s) register and submit $500 deposit by March 1st; $25 discount if<br />
you register and pay in full by March 1st ; Mix and Match discounts for up to a maximum total discount of $100.<br />
By signing below, I hereby promise to pay the price indicated above subject to qualified discounts<br />
and the refund policy as stated in the Syllabus. I understand that I will be invoiced for the deposit<br />
immediately, and balance is due at least 60 days prior to fielding. Invoices will be sent via PayPal<br />
from "MIB: Research & Educational Expeditions." A $100 fee will be assessed for returned checks<br />
or chargebacks.<br />
Signature ____________________________________________________________________<br />
Comments or Questions:
Please provide the following information about yourself. Use additional pages if necessary!<br />
* Gender:<br />
* Date of Birth:<br />
* Citizenship:<br />
male female (circle or highligh)<br />
** Passport Country and Number:<br />
** DAN Preferred Membership Number:<br />
*REQUIRED at time of application **REQUIRED at least 60 days prior to fielding<br />
Are you currently enrolled in school? Please give details about your status and program of study. Attach additional<br />
pages if necessary!<br />
Do you want/need credit for this course? How many hours?<br />
If yes, please provide contact information for your academic advisor:<br />
Name:<br />
Department:<br />
University:<br />
Phone:<br />
Email:<br />
Are you SCUBA Certified? YES NO<br />
If yes, provide a copy of your Certification Card and DAN Card.<br />
Tell us about other field courses or field work you have done:<br />
What should we know about your history (allergies, dietary restrictions, psychological conditions, physical or chronic<br />
conditions, etc.) that might affect your ability to participate in this course?<br />
Why did you chose this course? What are you expectations? Do you have any specific background, skills, or talents?<br />
Attach additional pages if necessary!<br />
Tell us about your experiences working with teams, with different cultures, or other situations that demonstrate your<br />
adaptability. Use additional pages if necessary!<br />
Circle or highlight your current swimming ability:<br />
non-swimmer recreational swimmer strong swimmer life-saving certificate<br />
Circle or highlight your comfort level in the ocean:<br />
not at all comfortable in calm seas comfortable in choppy seas<br />
Describe any life saving training you have done, include dates (Red Cross, First Aid, CPR, EMT, etc.)?<br />
What kind of snorkel, SCUBA, and/or boating experience (power, sail, kayak, etc.) do you have?
<strong>2012</strong> Field Course Policy & Liability Release Form<br />
Ecology, Behavior & Conservation of Manatees & Dolphins in Belize<br />
August 4-<strong>17</strong>, <strong>2012</strong><br />
Field Class Policy & Liability Release<br />
COMPLETE, SIGN, INITIAL, AND RETURN BY REGISTRATION DEADLINE<br />
As your instructors, it is our responsibility to ensure that students, staff, colleagues, and associates are familiar<br />
with and understand field course policies. As a student participant, you are required to read, fill out, and sign<br />
this document where indicated. Please feel free to contact me if you have any questions about our Field Class<br />
Policy & Liability Release Form.<br />
Return this signed document via mail, FAX, or Email at the same time you submit your Registration and Signed<br />
Syllabus (page 4 only). Also, be sure to send copies of your Passport, Dive Card, DAN Card, and Flight<br />
Itinerary ASAP, but not later than 60 days prior to fielding, to: Caryn Self-Sullivan, 200 Stonewall Drive,<br />
Fredericksburg, VA 22401-2110; Email: cselfsullivan@sirenian.org; Phone: 540.287.8207; FAX: 888.371.4998<br />
As Principal Investigator and Lead Instructor, I am responsible for:<br />
• ensuring policies are followed;<br />
• ensuring that risks are minimized to the best of my ability;<br />
• ensuring that students or staff are removed from the course if they, in my sole judgment, are in violation of these<br />
policies and/or present a danger to the safety and/or success of the course.<br />
Should a student become seriously injured or leave the course early for any reason, I will inform the student's emergency<br />
contact that they have left the course and why. Your signature below authorizes me to do so! Please complete your<br />
emergency contact information:<br />
In case of emergency or my removal from this course, please contact the following person(s):<br />
Name(s): __________________________________________________________________________________<br />
Address: __________________________________________________________________________________<br />
Phone(s): __________________________________________________________________________________<br />
Email(s): __________________________________________________________________________________<br />
Relationship to Student: ______________________________________________________________________<br />
Student Signature: ___________________________________________________________________________<br />
CONSENT TO INHERENT RISKS<br />
Some of the characteristics that make a field course attractive to students may also put the student or their property at risk.<br />
To reduce this risk, all services from arrival to and departure from Philip S. W. Goldson <strong>International</strong> Airport (PGIA,<br />
airport code BZE) are provided by Hugh Parkey's Belize Dive Connection (BDC), which has over 20 years of experience in<br />
Belize and also has liability insurance in place.<br />
Each student agrees to purchase international medical travel insurance in the form of a DAN Preferred Membership<br />
(Divers Alert Network) to cover his/her expenses in case of an emergency. This policy can be purchased online at<br />
http://www.diversalertnetwork.org/ under the MEMBERSHIP tab. Sign up as an Individual (not a student) and select the<br />
BEST: Preferred Membership Plan ($110). Note: you do not have to be a certified diver to purchase DAN Preferred<br />
Membership. Some of the potential risks inherent to this course are listed below. Initial Here: Yes ____ No ____<br />
Climate: Students must be prepared to spend long days in hot, humid, and wet conditions. The tropical sun is very strong.<br />
Risks include dehydration, sunburn, and other heat related illnesses. Students agree to take precautions as advised by<br />
instructor and staff to prevent illness. Students agree to notify instructor immediately if they are feeling ill.<br />
Insects: Sand flies and mosquitoes can be problematic on the caye. Sand flies are believed to be a vector for leishmaniasis<br />
in some regions of Belize, but this has not been reported for the cayes. Mosquitoes may transmit a number of diseases (see<br />
Diseases below). Botflies (Dermatobia hominis) are also reported in Belize, and mosquitoes may transmit their larvae to<br />
human hosts where the larvae will grow and develop. This is not life threatening, but can be painful and unpleasant.<br />
Students agree to keep instructor informed of insect bites.<br />
Page 1 of 4 Updated by CSS on 4/4/<strong>2012</strong>
<strong>2012</strong> Field Course Policy & Liability Release Form<br />
Marine life: Fire corals, several species of stinging jellyfish, sharks, sea urchins, lionfish, and other potentially dangerous<br />
marine organisms can be found in the study area. These can inflict painful and occasionally severe stings or bites, which<br />
may become infected if untreated. Those with a serious allergy to bee or wasp stings may have a similarly dangerous<br />
reaction to corals and jellyfish. The best prevention is to avoid touching unfamiliar plants & animals. Students agree to<br />
notify instructor immediately if they are injured as a result of an interaction with marine life. Students with known allergies<br />
must notify instructor on this form and bring their own medications and an 'epipen' as prescribed by their physician.<br />
Snorkeling: All the inherent risks of snorkeling are obviously present, including the effects of environmental conditions,<br />
marine life and other risks specific to your own physical/medical history. Snorkeling is optional. Students who chose to<br />
snorkel should bring their own gear (mask, snorkel, fins, wet suit or skin) and know how to snorkel safely without<br />
hyperventilating. A snorkel "check-out" by the instructor will be conducted on day 2 of the course.<br />
Boats: We will be aboard a small fiberglass boat for all fieldwork. All boats should have ladders and a bimini cover for<br />
sun protection; however, in some instances we may use a boat that lacks a ladder or shade. Deck surfaces of boats will<br />
become slippery and may place you at risk of slips, falls, and injuries that result from these accidents. All boats are operated<br />
by a licensed captain, employed by BDC. Students are not allowed to operate boats during the expedition.<br />
Disease: Diseases found in tropical regions include malaria, dengue fever, filariasis, leishmaniasis, onchocerciasis,<br />
trypanosomiasis (Chaga’s disease), schistosomiasis, leptospirosis, rabies, brucellosis, hepatitis, and typhoid. Most diseases<br />
are preventable with basic safety cautions. Students certify that they have sought professional advice from a physician,<br />
local health department, or international travel clinic regarding immunizations against diseases.<br />
Driving: Driving conditions in Belize are considered poor by US standards and pose inherent risks. All land transfers will<br />
be done by licensed drivers, employed by Hugh Parkey's Belize Adventures. Students will not be permitted to drive during<br />
the expedition.<br />
INTELLECTUAL PROPERTY RIGHTS<br />
Students may share photos, videos, and stories of the expedition with family, friends, local media, and in public forum for<br />
educational outreach purposes.<br />
However, all information (data and images) shared or gathered during the expedition remains the intellectual property of PI<br />
Caryn Self-Sullivan (CSS) and Co-PI Katherine S. LaCommare (KSL). Students are strictly forbidden from co-opting or<br />
plagiarism of data, images or information gathered during an expedition for use in a scientific thesis, masters or PhD work,<br />
for profit, or for the academic or business use of a third party without the express written permission of CSS & KSL.<br />
Students are aware that data gathered during the interviewing of local people becomes the intellectual property of CSS &<br />
KSL.<br />
Students are sometimes required to submit a written report reflecting what they have learned on a project, sometimes as a<br />
step toward receiving credit at their home institutions. Permission is hereby granted to students to use data and images for<br />
this purpose.<br />
LIFESTYLE CHOICES<br />
Our course does not discriminate on the basis of race, religion, ethnicity, or sexual orientation; and we respect students'<br />
rights to privacy. However, students should be warned that their lifestyle decisions may offend or clash with the<br />
sensibilities of local residents in our area of operations, or potentially violate local laws. Further, certain lifestyle decisions<br />
and behavior that impacts fellow students or the instructor may result in an uncomfortable, hostile and/or unproductive<br />
work environment.<br />
To ensure enjoyable and productive work conditions and smooth relations with local peoples, we have established the<br />
following code of conduct. Beyond practicing cultural sensitivity and showing common courtesy, please be mindful of the<br />
following limitations, and please remind and encourage all participants to practice sensitivity, courtesy, and compliance<br />
with these policies.<br />
Fraternization: Instructors, staff, colleagues, and associates are prohibited from becoming intimately involved with<br />
students during the duration of the field course. Romantic relationships between students that may otherwise seem<br />
permissible often create an unpleasant and/or unproductive work environment and are therefore strongly discouraged for<br />
the duration of the field course.<br />
Sexual Harassment: The inter-personal dynamics that exist between field course instructors, staff, and students are<br />
considered student-teacher relationships. Therefore, please be aware of the following policies.<br />
• Sexual harassment of students by the instructor or staff is prohibited.<br />
• Likewise, sexual harassment of the instructor, staff, other students, or local people by students is also prohibited.<br />
Page 2 of 4 Updated by CSS on 4/4/<strong>2012</strong>
<strong>2012</strong> Field Course Policy & Liability Release Form<br />
Sexual harassment infringes on an individual’s right to an environment free from unsolicited and unwelcome sexual<br />
overtones of conduct either verbal or physical. Sexual harassment does not mean occasional compliments of a socially<br />
acceptable nature. Sexual harassment refers to conduct which is offensive, which harms morale, or which interferes with<br />
the effectiveness of field courses and research expeditions. Such conduct is prohibited. Lewd or vulgar remarks,<br />
suggestive comments, displaying derogatory posters, cartoons or drawings, pressure for dates or sexual favors and<br />
unacceptable physical contact or exposure are examples of what can constitute harassment. It is important to realize that<br />
what may not be offensive to you, may be offensive to other students, the local population, staff members, or an instructor.<br />
Any individual who feels subjected to sexual harassment or has any knowledge of such behavior should report it at once to<br />
his or her instructor or staff members. All reports of sexual harassment will be handled with discretion and will be<br />
promptly and thoroughly investigated. Any student who is found to have engaged in conduct constituting sexual<br />
harassment will be removed from the course immediately. Any expenses involved in an early departure from the course are<br />
the responsibility of the student.<br />
Drugs: The manufacture, possession, use, purchase and/or sale of illegal drugs as defined by the United States Code and/or<br />
Belize Law is strictly forbidden while participating in this field course. Prescription drugs may only be purchased and used<br />
by the individual indicated on the prescription, in keeping with the intended use guidelines.<br />
Alcohol: Instructors and staff members are in a position of leadership and must therefore always be prepared to make<br />
responsible, timely decisions in any situation that may arise. If you decide to consume alcohol after course work and<br />
research has concluded for the day, please exercise your best judgment. Please respect local laws and customs.<br />
Intoxication can jeopardize your own safety, in addition to that of other students and staff.<br />
Students who are US Citizens must be a minimum of twenty-one (21) years old to consume alcohol. Be prepared to show<br />
your identification to the HP staff prior to the purchase of alcoholic drinks. Per day limits on alcohol consumption are in<br />
place during the course. In addition, restrictions on the use, possession, sale, or purchase of alcohol are solely at the<br />
discretion of the instructors. Local statutes, customs, practices, ordinances, and regulations with regard to the use,<br />
possession, sale, or purchase of alcohol are applicable to all students, instructors, and staff participating in this course.<br />
The instructor has the discretion to remove individuals from the course who consume alcohol in a time and manner that<br />
endangers the safety and/or productivity of other students, staff, the course, or any expedition.<br />
RECREATIONAL DAYS<br />
As an instructor and part-time resident of Belize since 1998, we (CSS & KSL) are familiar with local conditions and<br />
activities that may place the student at risk. Depending on the size of the student body, students may be given options for<br />
recreational day activities, but all activities must be conducted through BDC or their authorized agents. Potentially<br />
dangerous recreational day activities will not be allowed, as they may jeopardize your safety, the safety and well being of<br />
students, and ultimately, the success of this long-term research project.<br />
Students will occasionally have unsupervised free time before and after classes & discussions/field trips/research sessions.<br />
Activities undertaken during unsupervised time will be at your own risk.<br />
IN THE EVENT OF AN EMERGENCY: GOOD SAMARITAN ACTIONS<br />
In the event of emergencies, instructors, staff, and students must make certain judgments. While we have made an effort to<br />
ensure that qualified people make the most informed decisions possible, occasionally First Aid must be administered and<br />
other immediate steps taken by expedition participants who are not officially certified to make these decisions. We have<br />
safety protocols and emergency procedures in place. However, in rare, unforeseeable emergency situations, you are not<br />
restricted from exercising your best judgment with regard to your own safety. We do not restrict “good Samaritan”<br />
actions, or actions taken to assist fellow participants during emergency situations in the field. In the event of an unforeseen<br />
contingency that has not been provided for in BDC's emergency plan, we will take all reasonable steps to assist students<br />
and each other, with the understanding that we do not expect you to take steps that unreasonably jeopardize your own<br />
safety or that of others.<br />
STUDENTS AND DRIVING<br />
Students are not allowed to drive project vehicles or boats during the field course. Students may not transport themselves,<br />
other students, staff, or equipment in a vehicle of any kind, including their own. Students may only drive a project vehicle<br />
or boat, transport other students or perform errands in the field in case of an emergency and when no authorized drivers are<br />
available.<br />
Page 3 of 4 Updated by CSS on 4/4/<strong>2012</strong>
SCUBA, DIET, ALLERGIES<br />
<strong>2012</strong> Field Course Policy & Liability Release Form<br />
I (AM) or (AM NOT) (circle one) certified to dive using SCUBA equipment by _________________________<br />
_________________ (certifying organization). If certified, my certification level is ______________________;<br />
my certification number is ____________________; and I have attached a copy of my Dive Card to this form.<br />
Dietary Preferences (circle one): Vegan Vegetarian Carnivore Omnivore<br />
Allergies (circle one): NONE FOOD (list below) DRUGS (list below) OTHER (list below)<br />
REFUND POLICY<br />
I hereby acknowledge that if I cancel my registration before May 1 st , I will receive 100% of my deposit less a<br />
$50 service fee to cover bank and processing costs. If I cancel between May 1 st and June 1 st , I will forfeit my<br />
$500 deposit, but will receive a full refund of any balance paid above the deposit, less a $50 service fee. No<br />
refunds will be given after June 1 st . However, if the course is canceled by the instructor for any reason, I<br />
understand that I will receive a full refund of all payments made. I agree that I will NOT purchase my airfare<br />
until the instructor has confirmed that the course has made minimum enrollment. Initial Here: Yes ____ No ____<br />
PERMISSION TO RELEASE YOUR IMAGE AND INFORMATION<br />
I hereby consent to Caryn Self-Sullivan, the Hugh Parkey Foundation, and Belize Dive Connections releasing or<br />
using photographs of me taken during my field course & research expedition. Initial Here: Yes ____ No ____<br />
I hereby consent to Caryn Self-Sullivan releasing or using information gathered about me in the course of my<br />
field course & research expedition, except for contact and medical information and/or information that is<br />
otherwise agreed to be kept confidential. Initial Here: Yes ____ No ____<br />
ASSENT TO POLICIES<br />
I have read and understand the policies, rights, and responsibilities expressed in this document. I accept the<br />
policies above as a condition of participating on this field course and I accept the consequences described above<br />
for gross negligence or serious violations of course policies.<br />
RELEASE OF LIABILITY<br />
Student Signature: _____________________________________Date: ____/____/____<br />
I understand the potential hazards and risks attendant to the field course that is described in the syllabus, course<br />
briefing, and this policy document. By signing below I agree to participate in the field course at my own risk. I<br />
declare that I am in good health. I declare that I, and my heirs, in consideration of my participation in the<br />
Ecology, Behavior & Conservation of Manatees, Dolphins, Turtles field course from 18 th to 31 st May 2011 in<br />
Belize, hereby release Caryn Self-Sullivan, course instructors, and visiting scientists from any and all liability<br />
for damage to or loss of personal property, sickness or injury from whatever source, legal entanglements,<br />
imprisonment, death, or loss of money, which might occur while participating in this course. I am aware of the<br />
risks of participation, which include, but are not limited to, the possibility of injury or death. I hereby state that I<br />
am in sufficient physical condition to accept a normal level of physical activity during long-term exposure to the<br />
sun and heat. I understand that participation in this program is strictly voluntary and I freely chose to participate.<br />
I understand that the course does not provide medical coverage for me. I verify that I have purchased travel<br />
insurance equivalent to DAN Preferred Membership and that I will be responsible for any medical costs I incur<br />
as a result of my participation.<br />
However, I do not release Caryn Self-Sullivan, course instructors, and visiting scientists from liability on<br />
account of any injury, loss, or damage to me directly caused by the gross negligence or wanton or reckless<br />
misconduct of Caryn Self-Sullivan, course instructors, and visiting scientists.<br />
Student Signature: _____________________________________Date: ____/____/____<br />
Page 4 of 4 Updated by CSS on 4/4/<strong>2012</strong>
FIELD JOURNALS<br />
You are required to keep a permanently bound field journal or notebook (no loose leaf binders) in<br />
addition to having a notebook for your course notes. We encourage you to make daily entries following<br />
guidelines presented here and to include a “101 Species List” in the back of the journal. Your goal is to<br />
identify at least 101 organisms to the species level during the course including the 3 most prominent sea<br />
grass species, 4 mangrove species, common birds, fish, corals, reptiles, and insects. Your list should<br />
include common name, Genus and Specific Epithet, Phylum, identification keys, and a sketch. You will<br />
use information from your journal to complete your final exam, which will be in the form of a scavenger<br />
hunt!<br />
Memories are ephemeral and perceptions change with time, thus you should have your field journal with<br />
you at all times during to document your field experience, journal readings, and research efforts. Some<br />
researchers tend to write rough field notes in their journal while in the field, and then write a more<br />
complete description and expand on their observations each night after the fieldwork--in the same journal.<br />
This method has the advantage of allowing you to recognize what type of information you are omitting<br />
from your field notes and make sure you collect those data in the future.<br />
Your journals must be neat, legible and have appropriate spelling, etc. We recommend using a pencil or<br />
permanent ink pen (i.e., Pigma pens, Rite-in-the-Rain, etc.) since we may be in situations in which the<br />
pages of your journal will get wet. Most local stores that carry office supplies also have journals that are<br />
waterproof. While that type of journal is not required, it is encouraged.<br />
Drawings and sketches are an invaluable addition to your journal. Even the lousiest artist (like me :-) can<br />
make a decent field drawing in about 2 minutes. Start with the sea grasses and mangroves, then move on<br />
to the animals! A portion of your final grade will be based on the completeness, legibility, and accuracy<br />
of your journal. Also, do not erase incorrect entries but strike through them with a single line and write<br />
any correction above or below the original entry.<br />
A good field journal includes time, date, place, conditions, guests, professors, and other information at the<br />
heading of each entry. Entries should include information about each site and details about field<br />
techniques as well as information about your activities and data collection during your research project.<br />
In addition, entries should contain reflective thought about how you feel about the experience.<br />
Finally, you should always include contact information for anyone you meet in the class or the field as it<br />
may assist you in future endeavors. We are happy to review your journal at any time and it will be<br />
collected briefly and graded on the last day of class.<br />
Things that must be included in your journals:<br />
1. Your name, phone number, email and permanent address<br />
2. The names and contact information for your research partner<br />
3. Dates, times, weather conditions, and location of all field activities that are of sufficient detail to<br />
be replicated<br />
4. Name and contact information for every researcher or staff member that provides you information<br />
(including Belizeans!)<br />
5. Descriptions of field trips with detail on methodologies learned that would allow replication<br />
directly from your notebook<br />
6. Brief summaries of your daily activities<br />
7. All data and summaries of your thoughts regarding your individual research projects<br />
8. Your thoughts and reflections about each of the field experience<br />
9. Complete citations and notes from any sources (readings, etc.) so you can reference them in the<br />
future<br />
By J. Young (2010), Revised by C. Self-Sullivan (2011)
Red Mangrove Root Crab<br />
Goniopsis cruentata<br />
Hana Bucholz and Meagan Wise<br />
Edited by Caryn Self-Sullivan, Ph.D.<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2009 (contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Arthropoda<br />
Class Crustacea<br />
Order Decapoda<br />
Family Grapsidae<br />
Genus Goniopsis<br />
Species Goniopsis cruentata<br />
Ecology<br />
One of the more elusive crab species on Spanish<br />
Lookout Caye is the mangrove root crab<br />
(Goniopsis cruentata). Also known simply as<br />
the root crab, this brightly colored crab (purple,<br />
red and orange with white spots) lives among<br />
the red mangrove roots and branches. During<br />
low tide, you can find it beneath the red<br />
mangrove roots (Rhizophora mangle) and<br />
pimento stakes along the island’s central<br />
causeway. According to Feller (1996), these<br />
crabs have been observed feeding on mangrove<br />
propagules, insects, and organic material. In<br />
fact, mangrove root crabs may affect the<br />
distribution red mangroves. Little is known<br />
about the mangrove root crabs on Spanish<br />
Lookout Caye.<br />
Physical Description<br />
According to Kaplan (1984), the root crab is<br />
about 2.5 inches long and has stalked eyes, a<br />
brown carapace with white spots, and hairy red<br />
legs with white or yellow spots. Female crabs<br />
have a wide abdomen for egg storage, and males<br />
have a thin abdomen (Coulombe 1984). At<br />
sexual maturity, females and males are roughly<br />
the same size (Oliveira 2004).<br />
Conservation<br />
Any impact that significantly alters the red<br />
mangrove ecosystem may affect the red<br />
mangrove root crab because of its dependence<br />
on mangroves shelter and food. Worldwide,<br />
Figure 1. G. cruentata under red mangrove root on SLC<br />
mangroves are threatened by sea level rise,<br />
development, pollution, dredging, and disease.<br />
In Belize, the greatest threat is probably<br />
unsustainable development.<br />
Behavior<br />
When crabs mate, the female stores the male’s<br />
sperm until she releases her eggs. The eggs are<br />
kept in a mass (sponge) on the abdomen of the<br />
female and released simultaneously with the<br />
stored sperm, which flows over and fertilizes the<br />
eggs. Zygotes develop through four phases:<br />
fertilized egg, zoea larvae, megalops larvae, and<br />
adult (Coulombe 1984). Through personal<br />
observations we discovered that when mangrove<br />
root crabs are disturbed these crabs remain<br />
motionless for a short period of time before<br />
moving quickly sideways into a nearby burrow.<br />
If the crab does not automatically move away, it<br />
will bring its front claws forward in a defensive<br />
position.<br />
Figure 2. G. cruentata in Dominican Republic (photo by<br />
Pedro Genaro Rodriguez)<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
OUR INVESTIGATION<br />
Figure 2. G. crutentata on mud flat at SLC, 6 June 2009<br />
Introduction<br />
Red mangrove root crabs provide food for other<br />
species and reduce the insect population within<br />
the red mangrove ecosystem. By eating the leaf<br />
litter and organic material around the mangrove<br />
roots, they contribute to the recycling of<br />
nutrients. During our island orientation walk, we<br />
noticed these beautiful crabs and decided to<br />
investigate them further.<br />
Methods<br />
We conducted surveys each morning at 0600hrs<br />
for ten days during June 2009. For<br />
approximately one hour each morning, we<br />
walked the SLC central causeway, scanning the<br />
area on either side of the causeway, to observe<br />
the location and behavior of red mangrove root<br />
crabs.<br />
Results<br />
During low tides, G. crutentata were both near<br />
the mangrove roots and on the mud flat some<br />
distance away from the mangrove roots. At high<br />
tide, when there was no mud flat and the<br />
mangrove roots are partially submerged, crabs<br />
were observed only on structures above the<br />
water line, including roots, cement blocks,<br />
wooden planks, pimento stakes, and piles of<br />
emergent mud. They were predictably observed<br />
at certain locations and generally found alone or<br />
at least 30 cm away from any other crab.<br />
To better understand the distribution and<br />
behavior of G. crutentata on SLC, we propose<br />
the following sampling design for future<br />
students.<br />
Effect of Tidal Variation on Distribution and<br />
abundances of the Red Mangrove Root Crab<br />
on Spanish Lookout Caye, Belize<br />
Hypothesis: Tidal variations influence the<br />
abundance of the Red Mangrove Root Crab<br />
(Goniopsis cruentata) in the Spanish<br />
Lookout Cayes, Belize.<br />
Methods: Collect data during two complete<br />
lunar cycles, one during the dry season<br />
(February-April) and the other during the wet<br />
season (May-November). Conduct walking<br />
surveys three times a day at high tide, low tide,<br />
and intermediate tide. Each survey should<br />
include one transect along the central causeway<br />
in a single direction, one person on either side of<br />
the board walk scanning for crabs. Each person<br />
will observe both the area around the roots and<br />
in the water. The number and location of crabs<br />
will be recorded on a map of the area. At the<br />
beginning and end of each bridge along the<br />
causeway, water depth, salinity, temperature,<br />
and wind speed will be recorded. General<br />
observations, such as precipitation and debris<br />
will also be noted.<br />
If our hypothesis is true, then we expect more<br />
sightings of red mangrove root crabs at low tide<br />
than at high or intermediate tides.<br />
Literature Cited<br />
Coulombe, D. A. 1984. The Seaside Naturalist.<br />
Simon and Schuster. New York, New York. 139.<br />
Feller, I. C. 1996. Effects of nutrient enrichment<br />
on leaf anatomy of dwarf Rhizophora mangle<br />
L.(red mangrove). Biotropica 28: 13-22.<br />
Feller, I. C., Sitnik, M. 1996. Mangrove Ecology<br />
Workshop Manual.<br />
Kaplan, E. H. 1988. Southeastern and Caribbean<br />
Seashores. Houghton Mifflin Company. Boston,<br />
New York. Plate 22.<br />
Oliveira, N. F., Coelho, P. A. 2004. Maturidade<br />
sexual fisiológica em Goniopsis cruentata<br />
(Latreille) Crustacea, Brachyura, Grapsidae) no<br />
Estuário do Paripe, Pernambuco, Brasil. Revista<br />
Brasileira de Zoologia 21 (4): 1011<strong>–</strong>1015.<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
Cushion Sea Star<br />
Oreaster reticulates<br />
Jessica Barth and Aarin C. Allen<br />
Edited by Caryn Self-Sullivan, Ph.D.<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2009 (contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Echinodermata<br />
Class Asteroidea<br />
Order Valvatida<br />
Family Oreasteridae<br />
Genus Oreaster<br />
Species Oreaster reticulatus<br />
Ecology<br />
Cushion sea stars are found in the Atlantic<br />
waters from the Carolinas to Brazil (Puglisi,<br />
2008). Juveniles are green in color and can<br />
reach up 15cm in diameter (Kaplan 1982).<br />
Adults are tan, orange or rust in color and can<br />
grow to 50cm (Figure 1).<br />
O. reticulates is an omnivore, feeding on<br />
copepods, crab larvae and sponges (Puglisi<br />
2008). They are found in the calm shallow<br />
sandy flats around turtle grass, usually in<br />
shallow waters (Kaplan 1982). However, they<br />
have been found at depths of 37 meters (Puglisi<br />
2008).<br />
In Belize, they are ubiquitous in the seagrass<br />
beds where they are considered detritus feeders<br />
(C. Self-Sullivan, pers. comm.)<br />
Figure 1. Juvenile, adult, not to scale (Belize 2009).<br />
Behavior<br />
While feeding, cushion sea stars use their<br />
suction cup-like tube feet to wrap around their<br />
prey. In times when food availability is low the<br />
Figure 2. Six-armed specimen (Belize 2009).<br />
cushion sea star will prevent starvation by<br />
reabsorbing its own body tissue, which<br />
ultimately leads to a decrease in the sea stars<br />
overall size (Puglisi 2008).<br />
The species reproduces annually in the summer<br />
but can be found spawning continuously in the<br />
warm waters of the tropics (Puglisi 2008). The<br />
larvae can travel far distances in the ocean<br />
current before settling into a sea grass bed<br />
(Puglisi 2008).<br />
If a sea star is injured, it has the ability to<br />
regenerate its arms, as long as some portion of<br />
the central disk remains viable (Kaplan 1982),<br />
occasionally resulting in the production of an<br />
extra arm (Figure 2).<br />
Conservation<br />
O. reticulates was once the most common sea<br />
star in the Caribbean but is currently found most<br />
often in areas where the human population is<br />
relatively low. The species is considered to be<br />
rare in areas of high human populations due to<br />
being over harvested for souvenirs and<br />
aquariums (Puglisi 2008, Kaplan 1982).<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
OUR INVESTIGATION<br />
Introduction<br />
The Hugh Parkey Foundation for Marine<br />
Awareness and Education hosted our 2-week<br />
field course in June of 2009. At our base station<br />
on Spanish Lookout Caye (SLC, a 186 acre<br />
mangrove island just east of Belize City), we<br />
immediately noticed many bright orange sea<br />
stars in the seagrass beds just off the western<br />
shore of the caye. However, we didn’t see any<br />
off the eastern shore. This led us to ask why<br />
more sea stars were observed on the western<br />
than the eastern side of this pristine mangrove<br />
island.<br />
Methods<br />
The eastern shore of SLC faces windward,<br />
overlooking the Belize Barrier Reef and<br />
Caribbean Sea. This side of the caye is subject<br />
to more wind and wave action than the western<br />
shore, which overlooks the Belize mainland,<br />
about 10 miles west of SLC. To confirm our<br />
initial observation, we snorkeled along both<br />
sides of the caye once a day for ten days,<br />
counting cushion sea stars. Our snorkel<br />
transects were approximately 100m long and<br />
approximately 5m offshore.<br />
During our snorkels, we also noted the color,<br />
size, and behavioral state of each individual sea<br />
star, along with the depth and substrate at which<br />
they were found. We simulated turbulent<br />
conditions by placing individuals on their backs,<br />
and observing their behavior.<br />
Results<br />
During the surveys, we found no sea stars off the<br />
eastern side of the island, but many sea stars off<br />
the western side. On average, we observed 11<br />
sea stars per day off the western shore (Figure<br />
3). Smaller sea stars, presumably juveniles,<br />
were green in color. Medium, presumably subadults,<br />
were green with brown and orange<br />
patches. Large, presumably adults, were tan,<br />
orange, or rust in color. Sea stars were most<br />
often found on the sandy flats outside of the<br />
turtle grass (Thalassia testudinum) beds, in<br />
water depths of 1-3 meters. When turned over,<br />
most were holding onto some sort of shell or<br />
rock with their tube-feet.<br />
Number of Sea Stars<br />
Observed<br />
Sea Star Observations<br />
Figure 3. Counts off the western shore of SLC.<br />
The sea star’s body was somewhat pliable when<br />
picked up underwater; however the body<br />
became rigid if removed from the water. When<br />
picked up and held for a few minutes, they<br />
would attached to our<br />
hand using feet-like<br />
suction tubes.<br />
However, if we swam<br />
with one in our hand,<br />
it would become<br />
limp. When we<br />
inverted an individual<br />
to simulate conditions<br />
that may affect a sea star in a rough<br />
environment, it took approximately three<br />
minutes for the sea star to turn back over.<br />
Discussion<br />
With only ten days in the field, we were unable<br />
to follow-up on the results of our inductive,<br />
hypothesis generating survey. If given a chance<br />
to deductively test our hypothesis that “sea stars<br />
select the western shore to avoid more turbulent<br />
waters”, we would set up an experiment under<br />
laboratory conditions. Perhaps we could set up a<br />
large tank to simulate the 2 environments found<br />
around the caye. If a sea star placed in the more<br />
turbulent environment moved towards the<br />
calmer environment, then our hypothesis would<br />
be supported.<br />
Literature Cited<br />
Kaplan, Eugene H., 1982. A Field Guide to Coral<br />
Reefs Caribbean and Florida. New York: Houghton<br />
Mifflin. <strong>17</strong>6 pp.<br />
Puglisi, Melany P. 2008. Oreaster reticulatus<br />
http://www.sms.si.edu/IRLSpec/Oreaster_reticulatus.<br />
htm. Downloaded 06/01/2009.<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)<br />
16<br />
14<br />
12<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
1 2 3 4 5 6 7 8 9 10<br />
Day of Observation
Fiddler Crab<br />
Uca minax<br />
Whitney Montgomery & Daniel Hill<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2010 (contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Arthropoda<br />
Subphylum Crustacea<br />
Class Malacostraca<br />
Order Decapoda<br />
Family Ocypodidae<br />
Species Uca minax<br />
Ecology<br />
Fiddler crabs inhabit a wide range of tropical<br />
regions. In Belize, their habitats are mangroves<br />
and salt marshes. Fiddler crabs are extremely<br />
populous species. The semi-terrestrial species is<br />
diverse in color and habitat, with the most<br />
distinguishing characteristic being the males’<br />
asymmetrical claws (Fig. 1). They use the large<br />
claw to signal to other crabs within their<br />
vicinity, to attract a mate, and as a weapon used<br />
in combat between other males (Magnhagen,<br />
1991). These behaviors could draw attention to<br />
them from predators. The claw can be up to<br />
50% of an adult male’s body weight. If a claw is<br />
lost, the smaller claw of some species will grow<br />
larger and the crab will regenerate the lost claw<br />
with a new, smaller one.<br />
Figure 1. Fiddler crabs during experiment. (Belize, 2010)<br />
fiddler crab (Belize 2010)<br />
Figure<br />
2. Adult<br />
male<br />
Behavior<br />
Fiddler crabs tend to feed near their burrows, so<br />
that if they feel threatened, they will retreat into<br />
it to seek shelter. The crabs use their small<br />
claws to pick up pieces of sand, filtering out<br />
microscopic bits of algae, microbes, or fungus<br />
with their mouthparts. Crabs live about two<br />
years and molt their shells indeterminately (one<br />
to two times per year) as they continue to grow<br />
throughout their lives. Males wave their larger<br />
claw to attract female attention when ready to<br />
reproduce. Fiddler crab eggs are fertilized<br />
externally and the female fiddler crabs carry<br />
their eggs on their underside in a large mass.<br />
She keeps them safe in her burrow for a two<br />
week period and then journeys down to the edge<br />
of the sea to release the eggs into the receding<br />
tide. The larvae remain in a planktonic state for<br />
a further two weeks before venturing onto land<br />
and setting up a territory of their own.<br />
Conservation<br />
While fiddler crabs are not seriously threatened,<br />
their habitats are under pressure. As human<br />
populations continue to expand, they are<br />
destroying the mangroves and salt marshes<br />
where the crabs live. Fiddler crabs may serve as<br />
an important indicator species for these places.<br />
Fiddler crabs are important to the conservation<br />
of their ecosystems. Some experts believe that<br />
by continually sifting through the sand for food,<br />
their feeding habits aerate the substrate and<br />
prevent anaerobic conditions (Montague, D. L.<br />
1980).
OUR INVESTIGATION<br />
Introduction<br />
On a 13-day Field Biology class, the Hugh Parkey<br />
Foundation for Marine Awareness and Education<br />
was our host for the May 26 <strong>–</strong>June 8, 2010<br />
educational trip. Staying and studying on the<br />
Drowned Cayes (a mangrove island east of the<br />
Belize mainland), fiddler crabs are one of the easiest<br />
species to first recognize. After several days of<br />
observation of their behaviors and movements, the<br />
question of “Does the handedness of a fiddler crab<br />
affect its movement when threatened?” was<br />
developed. In other words, do crabs with a larger<br />
left claw tend to move towards the left side when<br />
threatened?<br />
Method<br />
After a tour of the island, we determined two prime<br />
spots for observation and collection of data. On the<br />
west side of the island, the mangroves and other<br />
vegetation provide protection for burrowing crabs<br />
with still access to sediments in which they eat. We<br />
studied the red-jointed fiddler crab (Uca minax) for a<br />
period of 2-h over the course of 2-d, with special<br />
emphasis on movement and behavior when<br />
threatened. We then collected a total of 20 fiddler<br />
crabs over another 2-d period (nleft claw=10 and nright<br />
claw=9). We used a ruler to determine the claw<br />
length of each crab (Fig. 2) and noted whether the<br />
larger claw was on the right or left (handedness).<br />
The crab was placed into a plastic container and<br />
allowed time for to calm down from the stress of<br />
human handling. We then “threatened” the crab by<br />
using the ruler to come at the crab facing its front<br />
(head) side. Using the same ruler, we measured the<br />
distance it traveled and noted the direction in which<br />
it moved. Ten of the fiddler crabs were collected<br />
from the West side of the island within an area and<br />
the other 10 were collected from a similar site on the<br />
East side of the island. They were taken from two<br />
sites to ensure we had a wide variety of crabs. The<br />
crabs were returned to their original sites upon<br />
completion of the experiment.<br />
Results<br />
Of the 20 crabs tested, 11 moved in the direction<br />
opposite their large claw (55%) and 9 moved in the<br />
direction of their large claw (45%). We found that<br />
the direction moved in relation to the larger claw<br />
was not significant. Using a chi-square test (df=1),<br />
we found crabs did not move significantly in the<br />
direction of their larger claw (Fig. 3, =0.2, p ><br />
0.05).<br />
<br />
<br />
<br />
Figure 3. The number of crabs and direction of crab movement s in<br />
relation to their larger claw.<br />
Discussion<br />
Our hypothesis that fiddler crabs would be more<br />
likely to move in the direction of their larger claw<br />
was not supported. The sample size was reasonable<br />
given the time constraint and if given the<br />
opportunity to redo the experiment, a larger sample<br />
size would allow for a stronger test of the<br />
hypothesis. While our findings showed no size<br />
sided preference in movement, a few observations<br />
suggested that fiddler crabs might move away from<br />
their large claw when threatened. Some research<br />
suggests that the large claw increases vulnerability<br />
to predation (Martin & Lopez, 1999), while other<br />
research suggests the claw might actually be used to<br />
ward off potential avian predators (Bildstein et all,<br />
1989). More research on the effect of fiddler crab<br />
movement while threatened could test this<br />
hypothesis.<br />
References<br />
Bildstein, K. L., McDowell, S. G. & Brisbin, I.L. 1989.<br />
Consequences of sexual dimorphism in sand fiddler<br />
crabs, Uca pugilator: differential vulnerability to<br />
avian predation. Animal Behavior, 37, 133-139.<br />
Magnhagen, C. 1991. Predation risk as a cost of<br />
reproduction. Trends in Ecology and Evolution, 6,<br />
183-186.<br />
Martin, J. & Lopez, P. 1999. When to come out from a<br />
refuge: risk-sensitive and state-dependent decisions<br />
in an alpine lizard. Behavioral Ecology, 10, 487-<br />
492.<br />
Montague, D. L. 1980. A natural history of temperate<br />
western Atlantic fiddler crabs (genus Uca) with<br />
reference to their impact on the salt marsh.<br />
Contributions in Marine Science, 23, 25-55.
Patch Reef Habitats<br />
Kelly Haun and Emy Rodriguez<br />
Edited by Caryn Self-Sullivan<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2010 (contact: caryns@sirenian.org)<br />
Ecology<br />
Corals are defined as small animals that attach<br />
themselves on hard surfaces to the sea floor.<br />
They colonize in the thousands and build<br />
skeletal structures of calcium carbonate, which<br />
produces the limestone foundation of coral<br />
reefs (Humann & Deloach, 2002). Hard<br />
corals are reef building corals that provide<br />
shelter and food for juvenile fish.<br />
Several types of coral reefs exist, which are<br />
defined by their general structure. Patch reefs<br />
represent one type of coral zone found near<br />
shore and are commonly surrounded by<br />
seagrass, sand, or algae (Mumby, 1999).<br />
These reefs are composed of hard and soft<br />
corals and rock or rubble, which form the<br />
concrete foundation on the sea floor.<br />
Patch reefs provide protective niches<br />
and food sources for many species of<br />
vertebrates and invertebrates (Kaplan, 1982).<br />
Conservation<br />
Coral reefs are in a state of decline around the<br />
world due to anthropogenic impacts that<br />
include overfishing, increased runoff and<br />
pollution from land, rising concentrations of<br />
carbon dioxide and invasive species (Turnball<br />
& Harborne, 2000). Overgrowth of algae,<br />
sponges and tunicates appears to have<br />
increased at many locations in the Caribbean.<br />
The growing problem can be partially<br />
contributed to the excess of nitrogen and other<br />
nutrients carried into the sea by runoff which<br />
stimulates the growth of plankton. The<br />
increase of plankton provides food to such<br />
organisms, which compete for space on corals<br />
and may eventually cause corals to die<br />
(Humann & Deloach, 2002). No previous<br />
studies have been made to analyze the reef<br />
system around the Drowned Cayes or<br />
document how these impacts may be affecting<br />
the reef ecosystem.<br />
View of reef patch from the surface.<br />
Introduction<br />
Our Investigation<br />
The Belize Barrier Reef System fringes along<br />
the eastern side of the Spanish Lookout Caye,<br />
which is a mangrove island surrounded by<br />
seagrass beds. Patch reefs are intermingled<br />
throughout and contribute to this complex and<br />
intricate environment.<br />
With the help of our guide, Denroy Usher, we<br />
surveyed the coastline around the caye and<br />
found several patch reefs varying in depth<br />
from 4 ft to 18 ft. These reefs have not been<br />
previously studied and an understanding of<br />
their structure and role can help contribute to<br />
conservation efforts in the area.<br />
Our objectives are to draw up preliminary data<br />
on the percentage of substrate types found in a<br />
coral patch and determine the relative<br />
abundance of fish in relation to the distance<br />
from a coral patch. We predict that the<br />
number of fish will decrease the further we<br />
move from the patch.<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
Method<br />
We selected a patch reef located on the south<br />
west point of the caye approximately 150 ft<br />
from the coastline. The depth of the coral<br />
patch was 4.7 ft with 100% water clarity<br />
which enabled ease of data collection. We<br />
randomly selected a patch head and set up a<br />
10 ft x 10ft study site. We randomly sampled<br />
four quadrants within the site to determine the<br />
percentage of substrates within the patch. A<br />
line transect was positioned from the center of<br />
the site east and west along the contour of the<br />
coastline each 40 feet from the center, totaling<br />
80 feet across. Fish counts were conducted<br />
every 10ft radiating from the center of the<br />
study site.<br />
Results<br />
Our findings show that substrate cover varied<br />
between quadrants 1 and 2. After calculating<br />
the total percent cover of both quadrants we<br />
determined the reef patch substrate to be<br />
composed of 40% rock/rubble, 22.5% hard<br />
coral and 16.25% algae/sponge (See Table 1<br />
in appendix).Figure 1 shows the data obtained<br />
during the fish line transect, showing a general<br />
trend whereby the total number of fish<br />
decreased as the distance from the coral patch<br />
increased. However, the east transect shows a<br />
decreased number of fish at 10 ft from the<br />
patch reef.<br />
Figure 1. Number of fish located at 10 ft increments from<br />
the center of the research site.<br />
Discussion<br />
This study attempted to draw up preliminary<br />
data on the percentage of substrate types<br />
found in a coral patches. However, due to<br />
limited time constraints we were unable to set<br />
up a control site with which to provide a<br />
comparative analysis.<br />
Our fish counts confirm our initial hypothesis<br />
that the relative abundance of fish decreased<br />
in relation to distance from the coral patch.<br />
This demonstrates the importance of patch<br />
reefs as a shelter and as a food source for fish,<br />
and suggests fish do not migrate far from these<br />
rocky/rubble and hard coral substrates.<br />
We believe that the slight discrepancy in data<br />
on the east transect line (10 ft distance) is a<br />
result of disturbance caused while performing<br />
the fish counts, as snorkeling became more<br />
challenging at shallower depths.<br />
We observed algal growth on many hard<br />
corals as well as a small amount of disease<br />
which may be indicative of increased nitrogen<br />
levels, as well as other anthropogenic stresses,<br />
and should be further investigated.<br />
Literature Cited<br />
Humann, P., & Deloach, N. (2002). Reef Coral<br />
Identification. Jacksonville: New World Publications.<br />
Kaplan, E. H. (1982). Ecology of the Coral Reef. In<br />
E. H. Kaplan, A Field Guide to Coral Reefs,<br />
Carribean and Florida (pp. 101-120). New York:<br />
Houghton Mifflin Company.<br />
Mumby, P. J. (1999). Classification Scheme for<br />
Manatee Habitats of Belize. London: Centre for<br />
Tropical Coastal Management Studies.<br />
Turnball, C., & Harborne, A. (2000). Summary of<br />
Coral Cay Conservation's Atlantic and Gulf Rapid<br />
Reef Assessment Data From Turneff Atoll, Belize.<br />
London: Coral Cay Conservation LTD.<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
Method<br />
We selected a patch reef located on the south<br />
west point of the caye approximately 150 ft<br />
from the coastline. The depth of the coral<br />
patch was 4.7 ft with 100% water clarity<br />
which enabled ease of data collection. We<br />
randomly selected a patch head and set up a<br />
10 ft x 10ft study site. We randomly sampled<br />
four quadrants within the site to determine the<br />
percentage of substrates within the patch. A<br />
line transect was positioned from the center of<br />
the site east and west along the contour of the<br />
coastline each 40 feet from the center, totaling<br />
80 feet across. Fish counts were conducted<br />
every 10ft radiating from the center of the<br />
study site.<br />
Results<br />
Our findings show that substrate cover varied<br />
between quadrants 1 and 2. After calculating<br />
the total percent cover of both quadrants we<br />
determined the reef patch substrate to be<br />
composed of 40% rock/rubble, 22.5% hard<br />
coral and 16.25% algae/sponge (See Table 1<br />
in appendix).Figure 1 shows the data obtained<br />
during the fish line transect, showing a general<br />
trend whereby the total number of fish<br />
decreased as the distance from the coral patch<br />
increased. However, the east transect shows a<br />
decreased number of fish at 10 ft from the<br />
patch reef.<br />
Figure 1. Number of fish located at 10 ft increments from<br />
the center of the research site.<br />
Discussion<br />
This study attempted to draw up preliminary<br />
data on the percentage of substrate types<br />
found in a coral patches. However, due to<br />
limited time constraints we were unable to set<br />
up a control site with which to provide a<br />
comparative analysis.<br />
Our fish counts confirm our initial hypothesis<br />
that the relative abundance of fish decreased<br />
in relation to distance from the coral patch.<br />
This demonstrates the importance of patch<br />
reefs as a shelter and as a food source for fish,<br />
and suggests fish do not migrate far from these<br />
rocky/rubble and hard coral substrates.<br />
We believe that the slight discrepancy in data<br />
on the east transect line (10 ft distance) is a<br />
result of disturbance caused while performing<br />
the fish counts, as snorkeling became more<br />
challenging at shallower depths.<br />
We observed algal growth on many hard<br />
corals as well as a small amount of disease<br />
which may be indicative of increased nitrogen<br />
levels, as well as other anthropogenic stresses,<br />
and should be further investigated.<br />
Literature Cited<br />
Humann, P., & Deloach, N. (2002). Reef Coral<br />
Identification. Jacksonville: New World Publications.<br />
Kaplan, E. H. (1982). Ecology of the Coral Reef. In<br />
E. H. Kaplan, A Field Guide to Coral Reefs,<br />
Carribean and Florida (pp. 101-120). New York:<br />
Houghton Mifflin Company.<br />
Mumby, P. J. (1999). Classification Scheme for<br />
Manatee Habitats of Belize. London: Centre for<br />
Tropical Coastal Management Studies.<br />
Turnball, C., & Harborne, A. (2000). Summary of<br />
Coral Cay Conservation's Atlantic and Gulf Rapid<br />
Reef Assessment Data From Turneff Atoll, Belize.<br />
London: Coral Cay Conservation LTD.<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventurelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
139<br />
140<br />
141<br />
142<br />
143<br />
144<br />
145<br />
146<br />
147<br />
148<br />
149<br />
Average BML Average AML
Variegated Sea Urchin<br />
Lytechinus variegatus<br />
Rachelle Boucher, Elizabeth Ferrell, and<br />
Samantha Egelhoff<br />
Edited by Caryn Self-Sullivan, Ph.D<br />
& Dr. Bruce Schulte, Ph. D<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2010 (contact:<br />
caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom: Animalia<br />
Phylum: Echinodermata<br />
Class: Echinoidea<br />
Order: Temnopleuroida<br />
Family: Toxopneustidae<br />
Genus: Lytechinus<br />
Species: Lytechinus variegatus<br />
Ecology<br />
Commonly known as the green sea urchin,<br />
Lytechinus variegatus is found in waters from<br />
Bermuda to Brazil, including the Caribbean Sea<br />
(Norris 2003). In their adult form, they are<br />
mainly white to green in color, although some<br />
have a pinkish-purple hue. They are covered in<br />
short spines that can regenerate. While their size<br />
usually varies from 1-12 cm, some can grow as<br />
large as 36 cm (Barnes 1982).<br />
Purple Variation of L. variegatus<br />
L. variegatus is normally found in calm, shallow<br />
waters, although they can be found in waters up<br />
to 50m deep (Norris 2003). It feeds on seagrass,<br />
mainly Thalassia sp., by using its tube feet and<br />
Aristotle’s Lantern (Norris 2003). The tube feet<br />
act as an anchor and hold the green sea urchin in<br />
White variation of L. variegatus<br />
place, while the mouth of the sea urchin, known<br />
as Aristotle’s Lantern, which is composed of<br />
strong joints and teeth, can scrape its food off of<br />
rocks or other hard surfaces (Barnes 1982).<br />
Behavior<br />
The green sea urchin reproduces by releasing<br />
unfertilized eggs and sperm into its water column,<br />
where the eggs are fertilized and develop into<br />
larvae (Norris 2003). The larvae undergo a<br />
complex metamorphosis to reach the adult form.<br />
L. variegatus use their spines and tube feet to<br />
cover their exterior with surrounding debris,<br />
serving as protection from predators and the sun’s<br />
rays.<br />
Conservation<br />
In Caribbean reef ecosystems, sea urchins have<br />
the capability of population explosion if one of<br />
their key predators is taken out of the system.<br />
Since their main food source is turtle grass<br />
(Thalassia sp.), a large urchin population can<br />
easily clear out whole patches of seagrass on<br />
which other animals, such as manatees and sea<br />
turtles, depend. However if sea urchins are taken<br />
out of a reef ecosystem, the seagrass patches that<br />
they normally keep under control can grow<br />
exponentially in some reef regions and kill off<br />
fragile corals. Jackson, however, mentions that<br />
Diadema urchins have always been abundant in<br />
their environment since pre-colonization, and<br />
predator control might not even be a factor<br />
(1997).
OUR INVESTIGATION<br />
Introduction<br />
We found L. variegatus to be quite common<br />
along the sea wall on the west side of Spanish<br />
Lookout Caye off the coast of Belize City,<br />
Belize. While observing them, we noticed that<br />
they were always hidden underneath rocks or<br />
covered with seagrass or shells, perhaps done in<br />
order to provide protection against predators or<br />
to protect them from the sun or other physical<br />
factors (Verling, 2004). The objective of this<br />
research was to see if the size of the sea urchin<br />
determined how fast the urchin recovered itself<br />
with debris, or if the size was related to how far<br />
they traveled away from their initial spot once<br />
uncovered. Smaller sea urchins are in need of<br />
more protection because they are more<br />
susceptible to threats. Therefore, we expected<br />
them to travel further or in less time in order to<br />
find protection.<br />
Methods<br />
Over a four-day period, we noted abiotic<br />
conditions and searched for different sized sea<br />
urchins along the shallow sandy and rocky<br />
bottoms. We removed all debris on 10 urchins<br />
using tweezers and tested 3 additional control<br />
urchins (n=13). Two controls were untouched,<br />
and one was touched but no debris was removed.<br />
Every two minutes for twenty minutes we<br />
measured the distance each traveled from its<br />
initial position. After observing for 20 minutes,<br />
we measured the diameter of each urchin, not<br />
including spines, and noted its coloration. Two<br />
graphs were plotted to see if the size of the<br />
urchins had an effect on the time of movement<br />
or distance traveled in order to recover<br />
themselves with debris.<br />
Results<br />
We found that there was no significant<br />
relationship between size and total distance<br />
(F(1,8)=1.32, P=0.28) (Fig 1) or size and time of<br />
movement (F(1,8)=0.38, P=0.55) (Fig 2). We<br />
tested a variety of juvenile-sized urchins, but<br />
each urchin had its own individual response. The<br />
control urchins suggested that there was no<br />
relationship between human interaction and<br />
individual urchin movement, only that they<br />
sought protection. All urchins either recovered<br />
themselves with debris or moved under rocks in<br />
place of coverings within the 20-minute time<br />
frame after we removed their initial debris.<br />
Figure 1. Total Distance Moved vs. Size<br />
Figure 2. Total Time of Movement vs. Size<br />
Discussion<br />
Time and distance traveled by each individual<br />
urchin appeared to be to dependent on how<br />
quickly they were able to recover themselves,<br />
whether with debris or under a rock; not their size.<br />
Further research should be done using a larger<br />
sample size including adult urchins in varying<br />
habitats. Since most of the urchins were juveniles<br />
and found in shallow water, this might have<br />
skewed our data and conclusions.<br />
Literature<br />
Barnes, Robert D. (1982). Invertebrate Zoology.<br />
Philadelphia, PA: Holt-Saunders <strong>International</strong>.<br />
Pp 961-981. ISBN 0-03-056747-5.<br />
Jackson, J.B.C. (1997). Reefs since Columbus. Coral<br />
Reefs: S23-S32.<br />
Norris, Amy. (2003). Green Sea Urchin Lytechinus<br />
variegatus. Marine Invertebrates of Bermuda:<br />
1-7.<br />
Verling, E. D.K.A. (2004). The dynamics of covering<br />
behavior in dominant Echinoid populations<br />
from American and European West coasts.<br />
Marine Biology: 191-206.
Red Fin Needlefish<br />
Strongylura natata<br />
Jamie Hennis and Carol Lacey<br />
Edited by Bruce A. Schulte, Ph.D. and<br />
Caryn Self-Sullivan, Ph.D.<br />
Ecology & Behavior of Manatees &<br />
Dolphins in Belize, Class of 2010.<br />
(contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Chordata<br />
Super Class Osteichthyes<br />
Class Actinopterygii<br />
Order Beloniformes<br />
Family Belonidae<br />
Genus Strongylura<br />
Species Strongylura natata<br />
Ecology<br />
The redfin needlefish is a marine fish<br />
located in warm, clear, coastal waters<br />
(Humann 2002). They can be seen<br />
solitary or in small groups in mangroves<br />
and sea grass ecosystems (Sweat 2009).<br />
Redfin needlefish are about 9 times their<br />
length than their width with an elongated<br />
needle-like bill. All species of needlefish<br />
are blue-green in color dorsally with<br />
silver sides. The redfin needlefish gets<br />
its name because the caudal and anal fins<br />
appear red in color. They are similar to<br />
other species in this family because their<br />
upper jaw is slightly shorter then the<br />
lower jaw, they have a short tail and the<br />
lower lobe is slightly longer than the<br />
upper lobe. We could find no<br />
information regarding conservation<br />
issues of needlefish.<br />
Figure 1. Needlefish at rest (photo by Carol<br />
Lacey).<br />
Behavior<br />
Needlefish are considered surface feeders.<br />
They are generally found just below the<br />
water surface where feeding is convenient for<br />
them due to their lateral line being located<br />
ventrally on their body (Lovejoy 2000). They<br />
are known as ram feeders. They capture their<br />
prey by striking them with their beak and<br />
pinning them in their jaws. The organism is<br />
then released and positioned properly for<br />
consumption in a head first position. Redfin<br />
needlefish have low predation risk because of<br />
their blue-green counter shading with the<br />
water, camouflaging them from overhead<br />
birds. Their nicknamed the skipper because<br />
when frightened, they tend to leap out of the<br />
water and skip across the surface.<br />
Figure 2. Needlefish foraging (photo by Carol<br />
Lacey).
OUR RESEARCH FINDINGS<br />
Introduction<br />
The Hugh Parkey Foundation for Marine<br />
Awareness and Education hosted our 13day<br />
field course (May 26-June 8, 2010).<br />
Our base station was located at Spanish<br />
Lookout Caye in the Drowned Cayes<br />
east of Belize City. This is a mangrove<br />
island made up of 186 acres. We found a<br />
population of redfin needlefish near the<br />
boat dock. We observed these fish for<br />
two days prior to collecting data and<br />
noticed several types of behavior. These<br />
observations suggested that activity<br />
levels depend on the time of day. Our<br />
objective was to determine if the<br />
occurrence of redfin needlefish<br />
behaviors varied with time of day.<br />
Method<br />
The sample area was 152.7 sq ft between<br />
the restaurant and the manatee museum.<br />
This area is full of marine life offering<br />
food and calm water (except waves from<br />
boats docking). During our observations<br />
we identified three behaviors as follows:<br />
1) Resting - needlefish oriented to the<br />
current in a head first position (Fig. 1).<br />
2) Foraging - needlefish swim in a<br />
random, non-organized manner looking<br />
for food (Fig. 2). 3) Socialization -<br />
needlefish appear to be aware of others<br />
in the population and interactions or<br />
non-aggressive swim takes place.<br />
We observed needlefish three times each<br />
day: morning 6:30-8 am, afternoon 12-4<br />
pm, and evening 6-9 pm. We scanned<br />
fish behavior once per minute for fifteen<br />
minutes, only including calculations in<br />
our results when fish were present. The<br />
abiotic factors of tidal depth, current<br />
speed, and weather conditions also were<br />
recorded.<br />
Results<br />
Needlefish were found to be most socially<br />
active in the morning and afternoon, with the<br />
three behaviors occurring equally at night<br />
(Fig. 3). The only abiotic factor appearing to<br />
have an effect on needlefish behavior was<br />
weather; needlefish were absent during rains.<br />
Figure 3. Number of fish resting, foraging, and<br />
socializing three times during the day.<br />
Discussion<br />
We found that fish numbers did not differ<br />
with time of day, but there was a difference<br />
in behaviors. Behaviors were equal in<br />
occurrence at night with mornings showing<br />
the highest socializing. With further research<br />
we could develop a better understanding of<br />
the effect time of day and other abiotic<br />
factors have on needlefish behavior and<br />
activity.<br />
Literature Cited<br />
Humann, Paul. 2002, Reef Fish Identification<br />
Florida Caribbean Bahamas, pg. 58.<br />
Lovejoy, Nathan R., 2000. Systematics of<br />
Needlefishes and their Allies (Teleostei:<br />
Beloniformes). Volume 54:pg 1349-1362.<br />
Sweat, L.H. 2009, Redfin Needlefishes.<br />
Smithsonian Marine Station at Fort Pierce.<br />
Downloaded May 20, 2010.<br />
www.sms.si.edu/irlspec/strong_hotata.htm.
Fiddler Crabs and Hermit Crabs<br />
Uca minax and Pagurus annulipes<br />
Rachel Calhoun and Molly Martin<br />
Ecology & Behavior of Manatees & Dolphins in Belize,<br />
Class of 2010 (contact:cselfsullivan@sirenian.org)<br />
Taxonomy<br />
Kingodom Animalia<br />
Phylum Arthropoda<br />
Subphylum Crustacea<br />
Class Malacostraca<br />
Order Decapoda<br />
Family Ocypodidae<br />
Species Uca minax<br />
Superfamily Paguroidea<br />
Species Pagurus annulipes<br />
Ecology<br />
Fiddler crabs are found in mangroves, salt marshes,<br />
and on sandy or muddy beaches in the Southeastern<br />
U.S., the Gulf of Mexico, and the Carribean (Kaplan<br />
1988). Their distinctively asymmetric claws easily<br />
identify them. U. minax is a detritivore, consuming<br />
algae, microbes, fungus, and other decaying materials.<br />
They appear to be a common resident along the<br />
brackish intertidal mud flats, lagoons, and swamps of<br />
Belize.<br />
Similarly, terrestrial hermit crabs are found in tropical<br />
areas within the intertidal zone. They also are<br />
detritivores, feed primarily on algae and decaying<br />
materials. They lack a carapace, or shell, so they<br />
“borrow” one from snails, such as periwinkles or<br />
oyster drills. As hermit crabs grow in size, they have<br />
to locate a larger shell, abandoning the previous one.<br />
Aside from this, their soft-coiled abdomens, two pairs<br />
of walking legs, and their asymmetric claws, are also<br />
used to identify them.<br />
Behavior<br />
Fiddler crabs create small burrows in the sediment<br />
that are used for mating, sleeping, refuge from<br />
predators, and ‘hibernating’ during the winter. They<br />
are very active during the day, foraging for food and<br />
digging burrows; they will return to their burrows at<br />
night and during high tide, plugging the entrance with<br />
mud or sand. Shells are an important resource for<br />
hermit crabs, but empty shells are generally low<br />
supply in nature (Kellogg 1976) and may be a limited<br />
resource for hermit crabs (Turra and Denadai 2004).<br />
As a result, hermit crabs will often compete with one<br />
another for shells. Competition may occur in two<br />
Figure 1. Hermit crab on the beach.<br />
ways: 1) hermit crabs may use chemical cues to locate<br />
newly available shells (Mesce 1982),or 2) they may<br />
display ritualized agonistic shell fighting behaviors<br />
and subordinate other individuals of the same species<br />
(Hazlett 1966). This is known as interference.<br />
However, many hermit crab species coexist in coastal<br />
areas and may cause various degrees of niche overlap<br />
(Turra and Denadai 2001).<br />
Introduction<br />
Our research took place during our 13-day research<br />
trip as part of a field marine biology class. We stayed<br />
on a mangrove island in the Drowned Cayes east of<br />
the Belize mainland with the Hugh Parkey Foundation<br />
for Marine Awareness and Education. Both hermit<br />
crabs and fiddler crabs are abundant on the island and<br />
are easily found. After much observation around the<br />
island of both crab species we asked the question of<br />
“How do fiddler and hermit crabs abundances change<br />
with the tides?”<br />
Figure 2. Fiddler crab near a burrow.<br />
Methods<br />
We chose two different areas to use for our research<br />
according to where we found the most of each crab<br />
species. Both areas were counted twice a day for three<br />
days. The area for the fiddler crabs was in a marshy<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventruelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
area on the north part of the island near the lagoon.<br />
The area for the hermit crabs was on the east part of<br />
the island only a few meters from the edge of the<br />
beach. Presence or absence of a breeze, temperature<br />
estimation, precipitation, the substrate moisture, and<br />
tide were taken before each data collection. We noted<br />
the presence and activity levels of each species in and<br />
around the study area before and after collecting data.<br />
We counted the number of fiddler crab burrows within<br />
a 10x10 ft area within the marshy area. This area had<br />
a small mangrove tree that marked one corner.<br />
Vegetation coverage in the area was estimated by<br />
looking at what percentage of the soil was covered by<br />
vegetation. The study area was divided in half and<br />
each person counted the number of burrows on each<br />
side to make counting easier and switched which side<br />
they counted in the afternoon.<br />
In the second area where hermit crabs were counted<br />
we demarcated a 15x15 ft study area with a palm tree<br />
in the eastern corner of the box and the number of<br />
hermit crabs on the beach was recorded. Next, the<br />
number of hermit crabs either climbing the tree or<br />
hidden in the braches of the tree were counted. Hermit<br />
crabs in the tree often crowded in crevices and would<br />
move when we got near making them more difficult to<br />
count. Therefore, each person counted and then<br />
shared numbers and an average number between<br />
counts was agreed upon. Also observations of hermit<br />
crabs in the area especially those near the water of the<br />
beach were recorded.<br />
Results<br />
The density of both crab species did not vary much<br />
with tide level (Fig. 3). The high tide for the fiddlers<br />
had the lowest density but it was also the only data<br />
taken during high tide whereas three data sets were<br />
taken during low tide and two taken during an ebb<br />
tide.<br />
The average density of fiddler crabs per square foot<br />
was 13 times greater than the average density of<br />
hermit crabs per square foot although the fiddler crab<br />
area was smaller. Most of the hermit crabs found were<br />
located in the crevices of the palm tree in the area and<br />
only a total of eight hermit crabs were found on the<br />
beach over all three days. The fiddler crab burrows in<br />
the designated area averaged 128.7 for each data<br />
collection and hermits found in the palm tree averaged<br />
21 for each data collection.<br />
Discussion<br />
We did not observe any major differences in the<br />
abundance of either crab species during different<br />
times of the tide. We did see six times more fiddler<br />
crabs than hermit crabs in a smaller area but neither<br />
species fluctuated too much with the tides. It is hard to<br />
draw any absolute conclusions from our data because<br />
we only took six data sets over three days and did not<br />
cover all the tides fairly. If we had more time to take<br />
more sets we would have more reliable data.<br />
Additionally, we did not have enough data to do any<br />
statistical analyses, so no certain conclusions can be<br />
made. However, this study seems to suggest that tides<br />
do not seem to directly affect either species’ numbers<br />
or their activity. Interestingly, the high tide did not<br />
reach the study site of the hermit crabs, which means<br />
that their decrease in density or relocation to the tree<br />
does not appear to be directly correlated with tides.<br />
There appears to be another reason as to why hermit<br />
crabs hide in trees, which could be predator<br />
avoidance.<br />
References<br />
Hazlett, B.A., 1966. Social behavior of the Paguridae and<br />
Diogenidae of Curacao. Stud. Fauna Curacao Other<br />
Caribb. Isl. Vol. 23, 1 <strong>–</strong>143.<br />
Kaplan, E.H. 1988. “Fiddler Crabs of the Southeastern<br />
U.S., Gulf of Mexico, and Carribean.” Southeastern<br />
and Caribbean Seashores. p 334-338<br />
Kellogg, C.W., 1976. Gastropod shells: a potentially<br />
limiting resource for hermit crabs. J. Exp. Mar. Biol.<br />
Ecol.Vol. 22, 101<strong>–</strong>111.<br />
Mesce, K.A., 1982. Calcium-bearing objects elicit shell<br />
selection behavior in a hermit crab. Science. Vol. 215,<br />
993<strong>–</strong>995.<br />
Turra, A., Denadai, M.R., 2001. Desiccation tolerance of<br />
four sympatric tropical intertidal hermit crabs<br />
(Decapoda Anomura). Mar. Freshw. Behav. Physiol.<br />
Vol. 34, 227<strong>–</strong> 238.<br />
Turra, A., Denadai, M.R., 2004. Interference and<br />
exploitation components in interespecific competition<br />
between sympatric intertidal hermit crabs. Journal of<br />
Experimental Marine Biology and Ecology Vol. 310,<br />
183<strong>–</strong> 193.<br />
<br />
Hugh Parkey Foundation for Marine Awareness & Education (belizeadventruelodge.com)<br />
<strong>Sirenian</strong> <strong>International</strong> (sirenian.org)
Cushion Sea Star<br />
Oreaster reticulates<br />
Whitney Allen and Rory McGonigle<br />
Edited by: Dr. Caryn Self-Sullivan, Ph.D.<br />
& Dr. Bruce Schulte, Ph.D.<br />
Ecology & Behavior of Manatees & Dolphins in<br />
Belize, Class of 2010 (contact: cselfsullivan@sirenian.org)<br />
Taxonomy<br />
Kingdom: Animalia<br />
Phylum: Echinodermata<br />
Class: Asteroidea<br />
Order: Valvatida<br />
Family: Oreasteridae<br />
Genus: Oreaster<br />
Species: Oreaster reticulates<br />
Ecology<br />
Cushion sea stars are found to be fairly common<br />
in Belize and other parts of the Caribbean. They<br />
are usually located in shallow waters inhabited<br />
with sea grasses and/or sand (Hummann and<br />
Deloach, 2002). Sea stars are normally fivearmed<br />
creatures varying in colors from green<br />
and brown to dark red and orange (Fig. 1).<br />
Juveniles tend to be green with spots of brown<br />
developing on them as they get older. Full adults<br />
may be yellow, orange, or red, normally mixing<br />
at least two of those together.<br />
Figure 1. Mid-range Adult Sea Star (Belize 2010)<br />
Some sea stars have six legs, which is almost<br />
always a cause of one previously being torn off.<br />
<br />
When growing back, it grows back mutated and<br />
forms two arms in the place of one (D. Usher,<br />
field assistant, pers. comm.).<br />
Behavior<br />
Sea stars have suction-like tube feet that allow<br />
them to stay on the sea floor (Fig. 2). Their<br />
connective tissue allows them “to change in<br />
response to passive mechanical properties” that<br />
can be caused by numerous environmental<br />
factors (Motokawa, 1984). When picked up or<br />
otherwise displaced, the tube feet close and are<br />
not opened until after the initial threat subsides.<br />
The center of their body is the central location of<br />
the digestive system, where all food is brought<br />
up into the body. The tube feet also may pick up<br />
food particles; those particles are then<br />
transferred down the arm to the mouth at the<br />
body’s center. Sea stars are not rapidly moving<br />
organisms. Our evidence suggests that they<br />
experience site fidelity, meaning they can be<br />
found in the same or near the same location on a<br />
day-to-day basis. However, their location can be<br />
affected by the water’s turbulence and weather<br />
conditions.<br />
Figure 2. Six-Armed Sea Star (Belize 2010)<br />
Conservation<br />
Sea stars, with their normally shallow location<br />
and non-movement along the sea floor, are<br />
typically taken for granted. With no visible<br />
breathing, many people forget they are even a<br />
living creature. Therefore, many sea stars lose<br />
their life when brought above water for too long,<br />
and from there, some are even kept for souvenirs<br />
and decorations.
Introduction<br />
Our Investigation<br />
Before our arrival to Hugh Parkey’s Adventure<br />
Lodge, we were asked to do pre-experimental<br />
research of information relating to a topic shared<br />
with a fellow student. After actual research<br />
began, we started by observing certain locations<br />
of sea stars along the western shore of the island.<br />
However, there did not appear to be a preference<br />
of their dwelling, so we eliminated that aspect<br />
out of our research. We noted that each time we<br />
picked up a sea star off the sea floor, it closed up<br />
its tube feet. It appeared that the time it took to<br />
completely close up the tube feet varied from<br />
one individual to another. This observation lead<br />
us to the main question of our project: Does the<br />
size of a sea star affect the amount of time it<br />
takes for it to close up its tube feet?<br />
Methods<br />
All of the data collected was done via snorkeling<br />
along the western shore (See appendix). Once a<br />
sea star was spotted, it was removed from the<br />
sea floor, and at the point of first contact the<br />
stopwatch was started. The star was flipped over<br />
in one’s hand and the time was recorded after<br />
complete closure. The sea star was measured<br />
from the end of one arm to the very center of the<br />
body. It was then measured again from the<br />
center of the body to a different leg to obtain a<br />
general diameter (two radii) for the organism.<br />
The following were also noted: location (sea<br />
grass or sand), color, number of legs and eating<br />
behavior at the time. After the data was<br />
collected, the sea star was gently placed back in<br />
its original position on the sea floor.<br />
Results<br />
No juveniles could be found on the western side<br />
of the island, but there was still a 16cm-27.5cm<br />
range of diameters in the sample pool (Figure 3).<br />
The average diameter was 24.0 cm (± 3.07).<br />
Four seconds was the most common closing<br />
time, consisting of 25% of our data. The fastest<br />
closing time was two seconds, which was a star<br />
measuring 25.5 cm in diameter. A 28.5 cm star<br />
closed the slowest, taking a total of nine<br />
<br />
seconds. Our evidence shows a weak but<br />
significant relationship between the size of a sea<br />
star and the time it takes to close (r 2 =0.20, df= 1,<br />
23, F=5.63, P=0.026; raw data located in<br />
appendix).<br />
Figure 3. Size of sea star and the time to completely close its tube<br />
feet when picked up and inverted underwater.<br />
Discussion<br />
Our hypothesis (“The bigger the sea star is the<br />
quicker it will close up its tube feet.”) was not<br />
supported by our data. We expected a negative<br />
correlation, while our results produced one that<br />
is positive. However, our acquired R square<br />
value shows that that the time it takes a sea star<br />
to close is somewhat size dependent, with larger<br />
stars taking longer. The rate at which it protects<br />
itself could also be caused by either the<br />
experience of turbulent waters or even predation.<br />
A follow-up analysis could be done to continue<br />
to further test our hypothesis. If each sea star in<br />
our sample pool had been in the exact same<br />
location or raised under the exact same<br />
conditions, the role of size in its closure speed<br />
would be easier to isolate.<br />
References<br />
Humann, P. and N. Deloach. (2002). Reef<br />
Creature Identification, Sea Stars, p.<br />
366. New World Publications.<br />
Motokawa, T. (1984). Connective tissue catch in<br />
echinoderms. Biol. Rev. 59, 255-270.
Cushion Sea Star<br />
Oreaster reticulatus<br />
Blakely Rice<br />
Edited by: Caryn Self-Sullivan, Ph.D.<br />
Ecology and Behavior of Manatees and<br />
Dolphins in Belize, Class of 2011<br />
(contact: cselfsullivan@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Echinodermata<br />
Class Asteroidea<br />
Order Valvatidas<br />
Family Oreasteridae<br />
Genus Oreaster<br />
Species Oreaster reticulatus<br />
Ecology<br />
The Cushion Sea Star (Oreaster<br />
reticulatus) is located throughout the<br />
Caribbean and are commonly found in<br />
shallow water seagrass and sand<br />
habitats (Scheibling, 1980). Adult sizes<br />
can range from 5 inches to 20 inches<br />
and juveniles from about 2 inches to 6<br />
inches (Kaplan, 1982; Rice, personal<br />
measurements). Coloration in adults is<br />
generally orange, yellow and red or any<br />
variation of the above. Juvenile<br />
coloration is a broad spectrum of greens<br />
(Figure 1).<br />
Oreaster is predominately a omnivorous<br />
grazer eating a variety of organisms<br />
including sponges and algae<br />
(Scheibling, 1982).<br />
Behavior<br />
Oreaster reticulates feeds by inserting<br />
its stomach into or onto its prey and<br />
secrets enzymes that digest the good,<br />
which is then absorbed (Kaplan, 1982).<br />
Their tube feet help them hold onto their<br />
prey. Oreasters many tube feet enable<br />
them to move quite quickly and cover<br />
significant distances in a short amount<br />
of time.<br />
Because sea stars reproduce asexually<br />
in a marine environment gametes meet<br />
mostly by chance (Kaplan, 1982). To<br />
increase chances of fertilization sea<br />
stars will release their gametes if other<br />
gametes are detected in the area<br />
(Kaplan, 1982).<br />
Conservation<br />
Even though Oreaster reticulatus is one<br />
of the most common sea stars in the<br />
Caribbean populations have become<br />
rare in some areas mainly due to over<br />
harvesting for the souvenir trade<br />
(Kaplan, 1982).
Introduction<br />
Sea stars have long been a favorite<br />
sight for snorkelers and divers, mainly<br />
because they are one of the only<br />
animals that does not bite or sting when<br />
handled. There only defensive behavior<br />
when being picked up is to become<br />
rigid, however after some time they will<br />
relax. This behavior inspired the<br />
question for my independent research<br />
project, how long does it take for a sea<br />
star to relax after becoming rigid and is<br />
there a difference between adults and<br />
juveniles. I hypothesized that juveniles<br />
will take more time to relax and the<br />
adults do.<br />
Methods<br />
I chose the western side of Spanish<br />
Lookout Caye (SLC) as my study<br />
location because of the persistence of<br />
sea stars in that area. To investigate my<br />
hypothesis I snorkeled along the sea<br />
wall and sea grass beds until I found a<br />
sea star. When a sea star was found it<br />
was picked up and turned over until all<br />
the tube feet had retracted and the<br />
animal was rigid. This was done to<br />
insure that the timing was accurate as it<br />
was found that some animals took<br />
longer to become rigid than others.<br />
Once fully rigid the animal was placed in<br />
my right, once it was on my hand timing<br />
began. Timing stopped when the first<br />
tube feet were present. The time was<br />
recorded on a slate and the animal was<br />
replaced in its original location, habitat<br />
location and size of the animal were also<br />
recorded.<br />
Results<br />
At the end of the study a total of 20<br />
adults and 6 juveniles were surveyed. A<br />
two-sample t-test for independent data<br />
was conducted to compare the latency<br />
(time) in adults and juveniles. There was<br />
a significant difference in the latency of<br />
adults and juveniles (t=2.34, df= 24,<br />
p=0.014). A one-way ANOVA was also<br />
conducted for post-HOC analysis<br />
regarding the latency with habitat type<br />
and latency with size. In regards to<br />
latency and habitat there was no<br />
significant difference in adults (f=0.45,<br />
df=18, p=0.6455) or juveniles (f=1.19,<br />
df= 5, p=0.3366). When looking at<br />
latency in regards to size there was a<br />
significant difference in adults (f=11.86,<br />
df=19, p=0.000596) but not a significant<br />
difference in juveniles (f=0.47,<br />
df=5,p=0.6644).<br />
Discussion<br />
My hypothesis that latency in juveniles<br />
would be longer than in adults was<br />
disproved. I found that latency in adults<br />
was actually longer than in juveniles.<br />
Continued studies need to be conducted<br />
to determine the reason for this<br />
behavior.<br />
Resources<br />
Kaplan, Eugene H., 1982. A field guide<br />
to Coral Reef. New York: Houghton<br />
Mifflin. P.<strong>17</strong>1-<strong>17</strong>6.<br />
Scheibling, R.E., 1982. Feeding habits<br />
of Oreaster reticulatus<br />
(Echinodermata: Asteroidea). Bulletin<br />
of Marine Sciences, 32, 504-510.<br />
Scheibling, R.E., 1980. Abundance,<br />
Spatial distribution and size structure<br />
of populations of Oreaster reticulatus<br />
(Echinodermata: Asterodiea) on Sand<br />
Bottoms. Marine Biology, 107-119.
Magnificent Feather-Duster<br />
Sabellastarte magnifica<br />
Sean Herbert<br />
Ecology, Behavior and Conservation of<br />
Manatees & Dolphins in Belize, Class of 2011<br />
(contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Annelida<br />
Class Polychaeta<br />
Order Sabellida<br />
Family Sabellidae<br />
Genus Sabellastarte<br />
Species S. magnifica<br />
Ecology<br />
Feather duster worms inhabit reefs, mangroves,<br />
rocks, wrecks, and sand or gravel bottoms. They<br />
often grow from coral heads (Humann & DeLoach<br />
2002), and in Belize, the magnificent variety is often<br />
found attached to mangrove roots. Whatever the<br />
substrate, they are usually found sharing the space<br />
with various sessile animals such as sponges,<br />
mollusks and tunicates. S. magnifica are among the<br />
largest and most common feather dusters found in<br />
Belize. Those who may cringe at the word ‘worm’<br />
need not fear; their bodies remain hidden, encased in<br />
a parchment-like tube attached to the substrate. The<br />
worms construct their tubes using fine sand and<br />
mucus secreted from glands just below their head.<br />
The only visible portion of the worm is the crown of<br />
feather-like appendages, which can be used to easily<br />
identify magnifica because of its exclusive doublecrown<br />
arrangement. The ‘feathers’ are usually<br />
banded and come in a variety of colors, including<br />
brown, tan, gold, reddish purple and white (Humann<br />
& DeLoach 2002).<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
Behavior<br />
Feather duster worms use their ‘feathers’, known<br />
as radioles, as both gills and filters to capture<br />
plankton <strong>–</strong> their cuisine of choice. Trapped<br />
microscopic food is brought to its mouth at the<br />
center of its crown of ‘feathers’ (Fitzsimons<br />
1965). As largely sessile animals, they rely on<br />
currents to provide them with their edible treats.<br />
Their soft bodies remain out of sight throughout<br />
their lives, slowly adding length to their tubes as<br />
they mature. Their most notable behavior is the<br />
near instantaneous retraction of their radioles back<br />
into the tube when disturbed. The worms can<br />
detect changes in water movement, light intensity<br />
white (Humann & DeLoach 2002), and of course,<br />
tactile stimulation. As such, predators are often<br />
sucker-feeding fish . They have evolved rows of<br />
upper-facing hooks along each segment of their<br />
body, which allow anchoring in the event a fish<br />
attempts to suck them out (Woodin 1987).<br />
Conservation<br />
Although feather dusters themselves are not<br />
seriously threatened, many populations are subject<br />
to ongoing anthropogenic impacts on the coral<br />
reef and mangrove ecosystems, including, but not<br />
limited to, pollution, increased ‘fin’ traffic as<br />
Belizean tourism expands (reef), ocean<br />
acidification (reef), development (mangrove), and<br />
invasive species (Turnball & Harborne 2000).
INVESTIGATION<br />
Introduction<br />
Retraction of radioles is its primary form of<br />
defense, yet they are necessary for both respiration<br />
and ingestion. Given that they are sessile, I would<br />
assume that the animal would achieve the most<br />
benefit from having its radioles extended as long<br />
and often as possible. Therefore, a predator<br />
physically touching the radioles should cause a<br />
full retraction, with a longer latency, versus a<br />
distant water disturbance causing a shorter<br />
retraction and latency. Two tests were devised<br />
for this hypothesis: is there a correlation between<br />
latency and disturbance type; and, do worms<br />
respond differently to tactile and distant<br />
disturbance?<br />
Methods<br />
S. magnifica were identified in two locations<br />
along western shore of Spanish Lookout Caye:<br />
randomly distributed along a concrete seawall and<br />
clustered against square concrete pillars once used<br />
for a pier. Fourteen individuals were found<br />
attached to the seawall, and sixteen were found<br />
attached to different pillars, often in groups of<br />
two. Traditional snorkel gear was used for data<br />
collection, including a floatation device so as to<br />
keep unwanted disturbance at a minimum.<br />
Individuals are disturbed physically by a light<br />
index finger touch. For water movement<br />
disturbance, a ruler was used to gauge a distance<br />
of about 5-6cm from the radioles, and then I<br />
waved my hand laterally until the worm retracted.<br />
Some individuals would not respond to the wave,<br />
in which case I would flick my four fingers from<br />
behind my thumb towards the animal. A static<br />
distance of 5cm was originally intended, though<br />
the sensitive nature of the animal prevented this.<br />
A full or partial retraction was noted. After<br />
retraction, a stopwatch is used to measure the<br />
latency until reemergence. Reemergence from full<br />
retraction is defined as radiole tips protruding<br />
from the rim of its tube, or visible anterior<br />
movement if it is partially retracted.<br />
Results<br />
Using a correlated one-tailed t-test for latency of<br />
each disturbance type, a significant difference was<br />
observed between tactile and distant disturbances in<br />
S. magnifica (t = +2.58, df=29, P < 0.01, n=30),<br />
supporting this portion of the hypothesis. However,<br />
using a 2x2 categorical analysis chi-square test, no<br />
significant difference was found between full and<br />
partial retraction ( 2 = 3.27, df=3, P = 0.3518).<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
Discussion<br />
The data collected only partially supports my<br />
hypothesis: tactile disturbance definitely results in<br />
a longer latency than distant water disturbance (P<br />
= 0.007), although full or partial retraction among<br />
individuals appears to be random. An alternate<br />
hypothesis could be that perhaps radioles grow<br />
more rapidly relative to tube construction, and<br />
may protrude even when the animal has retracted<br />
as far as it can. Another possibility is that<br />
individuals can detect the strength of the<br />
disturbance, either tactile or distant, and react<br />
accordingly. Further investigation using measured<br />
static force could be conducted to test this<br />
hypothesis.<br />
References<br />
Humann, P. and N. Deloach. (2002). Reef Creature<br />
Identification, Sea Stars, p. 140 & 148. New<br />
World Publications.<br />
Turnball, C., & Harborne, A. (2000). Summary of<br />
Coral Cay Conservation's Atlantic and Gulf<br />
Rapid Reef Assessment Data From Turneff<br />
Atoll, Belize. London: Coral Cay<br />
Conservation LTD.<br />
Fitzimons, G. (1965). Feeding and Tube-Building in<br />
Sabellastarte magnifica (Shaw). Bulletin of<br />
Marine Science, 15 (3) : 642-670
Marine Hermit Crab<br />
Clibanarius vittatus<br />
Michelle Cassidy<br />
Ecology and Behavior of Manatees and Dolphins in<br />
Belize, Class of 2011<br />
(contact: caryns@sirenian.org)<br />
Taxonomy<br />
Kingdom Animalia<br />
Phylum Arthopoda<br />
Class Crustacea<br />
Order Decapoda<br />
Family p<br />
Genus Clibinarius<br />
Species Clibinarius vittatus<br />
Ecology<br />
Common to most marine, estuarine, and<br />
shoreline environments, hermit crabs can<br />
be easily found worldwide. On Spanish<br />
Lookout Caye in Belize, Central America,<br />
both terrestrial and aquatic varieties of this<br />
crab can be easily spotted; the former found<br />
wandering the sandy shore, and the latter in<br />
the muddy shallows of the mangrove<br />
swamp. The mangrove species, C. vittatus,<br />
feeds on plant and animal detritus that sinks<br />
to the bottom.<br />
Physical Description<br />
Hermit crabs utilize shells (generally from<br />
gastropods) for protection. This results from<br />
their development of a non-calcified<br />
abdomen that would make them vulnerable<br />
to predation (Schejter & Mantelatto 2009).<br />
Crabs have large foreclaws and smaller<br />
hindclaws. Shell sizes and coloration vary<br />
significantly between individuals.<br />
Conservation<br />
Though the hermit crab itself is not<br />
endangered, human expansion threatens its<br />
habitat. These crabs play a key role in the<br />
Mangrove ecosystem via scavenging and<br />
consuming detritus. In order to protect these<br />
animals and the web of life they exist within,<br />
it is important to conserve Mangrove<br />
swamps, such as the ones on Spanish<br />
Lookout Caye. Plenty of Mangrove Cayes in<br />
Belize have been dredged and filled to<br />
provide land for resorts, the process of<br />
which destroys valuable Mangrove<br />
resources and hermit crab habitats.<br />
Promoting awareness of the numerous<br />
species that depend on Mangroves for<br />
survival is key to conserving these animals.<br />
Behavior<br />
Crabs retreat quickly into their shells at any<br />
sign of potential danger. They will warily<br />
emerge from their shells if not touched or<br />
approached for approximately one minute.<br />
Competition for particularly nutirient-rich<br />
patches can become aggressive, with larger<br />
crabs often strong-arming smaller<br />
individuals into leaving the area, so they can<br />
enjoy the food source (Ramsay et al. 1997).<br />
Hermit crabs reproduce seasonally, with<br />
intensive mating activity occurring in the<br />
warmer months. Studies show that the<br />
recruitment area for young crabs differs<br />
from the adult habitat, and individuals are<br />
already an intermediate size once they<br />
appear. (Sant´Anna et al. 2008, Turra &<br />
Leite 2000)<br />
Crabs can detect chemical signals of a dead<br />
or dying gastropod or conspecific, alerting<br />
them to the presence of a potential home<br />
(Rittschof 1980, Rittschof et al. 1992). They<br />
can also use chemical stimuli to identify<br />
predators (Rosen et al. 2009).
Introduction<br />
Shells serve as a limiting factor for hermit<br />
crabs in most environments (Tricarico &<br />
Gherardi 2006). As a result, crabs must<br />
develop strategic behaviors to procure<br />
viable shells as they grow. Synchronous<br />
vacancy chains (SVC) are adaptive<br />
behaviors that address this limitation by<br />
providing SVCs have been observed in<br />
terrestrial hermit crabs in the Drowned<br />
Cayes of Belize (Rotjan et al. 2010), but it is<br />
unknown if the marine species C. vittatus<br />
also implements SVCs. This study aims to<br />
answer the question of whether marine<br />
hermit crabs also utilize SVCs for shell<br />
acquisition.<br />
Closely related species of terrestrial Hermit Crabs<br />
(Coenobita clypeatus) in a SVC on SLC.<br />
Materials and Methods<br />
I gathered medium and large empty<br />
gastropod shells from the seagrass beds<br />
near the mangrove pond, then placed them<br />
within four 30 cm x 30 cm quadrats placed<br />
within the hermit crab habitat. I observed for<br />
two hour and a half sessions, making note<br />
of behavior in each quadrat every 10<br />
minutes. The number of hermit crabs<br />
present and the presence or absence of<br />
characteristic SVC behaviors were noted<br />
(piggybacking, waiting, tug of war). A SVC<br />
was present if the largest crab moves into<br />
an empty shell, while smaller crabs lined up<br />
behind move successively into the shell of<br />
the crab before it.<br />
Results<br />
Data collected couldnt be statistically<br />
analyzed, but some patterns emerged from<br />
observation. Crabs investigated empty<br />
shells, and one individual waited for<br />
approximately 30 minutes by a shell that<br />
was too large for it. On several occasions,<br />
crabs would “piggyback”, grasping to the<br />
shell of a larger individual while it moved<br />
around the area. The was one incident of<br />
“tug of war” in which two crabs pulled an<br />
empty shell back and forth between them<br />
for approximately ten minutes. Eventually<br />
the smaller crab walked away, and the<br />
larger crab switched into the empty shell.<br />
No other crabs came to take the shell left<br />
behind by that switch.<br />
Discussion<br />
While no SVCs were observed for C.<br />
vittatus, there was evidence of behaviors<br />
such as piggybacking and waiting that<br />
suggest that this species of hermit crab also<br />
utilizes this method of shell acquisition. It is<br />
possible that for this population, shells are<br />
not as vital as food—many individuals in the<br />
study area ignored the shells completely<br />
and spent their time feeding. The one shell<br />
switch I witnessed suggests that both<br />
synchronous and asynchronous vacany<br />
chains are useful methods of gaining new<br />
protective shells. It would require further<br />
investigation to evidence which method<br />
occurs more frequently for the marine hermit<br />
crabs of Spanish Lookout Caye.<br />
Literature Cited<br />
Rotjan D, Chabot J, and Lewis S. Social<br />
context of shell acquisition in Coenobita<br />
clypeatus hermit crabs. Behavioral Ecology.<br />
(2010) 21: 639-45.<br />
Tricarico E and Gherardi F. Shell acquisition<br />
by hermit crabs: which tactic is more<br />
efficient? Behavioral Ecology Sociobiology.<br />
(2006) 60: 492-500.
Variegated Sea Urchin<br />
Lytechinus variegatus<br />
Kalie Bishop<br />
Ecology and Behavior of Manatees and<br />
Dolphins in<br />
Belize, Class of 2011 (contact:<br />
cselfsullivan@sirenian.org)<br />
Taxonomy<br />
Kingdom: Anamalia<br />
Phylum: Echinodermata<br />
Class: Echinoidea<br />
Order: Temnopleuroida<br />
Family: Toxopneustida<br />
Genus: Lytechinus<br />
Species: Lytechinus variegates<br />
Ecology<br />
L. variegatus is commonly found in shallow,<br />
calm waters, such as seagrass beds. They<br />
have adapted to survive in many marine<br />
environments but primarily subsist within<br />
seagrass and kelp beds. L. variegatus<br />
inhabits the waters from North Carolina and<br />
Bermuda southward to the Caribbean and<br />
Brazil (Hendler et al. 1995). Although they<br />
cannot live in freshwater, they inhabit all<br />
depths and climates of the oceans.<br />
Physical Description<br />
L. variegatus vary in color, but the most<br />
frequently seen are green, purple and light<br />
pink. L. variegatus have radially symmetrical<br />
bodies divided into five equal parts. The<br />
skeletal structure of the sea urchin is a rigid<br />
test or theca that is made up of plates<br />
encircling the mouth in the center of the oral<br />
side, encompassing the urchins inner<br />
organs (Nichols 1962). Spines, which can<br />
regenerate, cover the entire surface of the<br />
sea urchin test and play a role in protection<br />
as well as movement. Among the spines are<br />
five paired rows of tiny tube feet with<br />
suckers that help with locomotion, capturing<br />
food, and holding onto the seafloor.<br />
White variation of L. variegatus<br />
Behavior<br />
L. variegatus feed on decaying matter, kelp,<br />
sea grass, algae and occasionally even sea<br />
urchin skeletons. Sea urchins reproduce by<br />
releasing the unfertilized eggs and sperm<br />
into the water column, where the eggs are<br />
fertilized and develop into free-swimming<br />
larvae known as plutei (Pechenik 2000;<br />
Harvey 1956). The larval form<br />
of L. variegatus, as with most echinoderms,<br />
is very different from the adult form and<br />
must undergo a complex metamorphosis to<br />
reach adult form.<br />
Conservation<br />
Being primary grazers of kelp, and algae, L.<br />
variegatus play an important role within the<br />
marine communities that they inhabit. The<br />
population of L. variegatus in an area is vital<br />
to the health of the coral reef, as they<br />
eliminate the seaweed and algae that can<br />
consume coral if a balance is not<br />
maintained within the environment. The<br />
opposite can also occur with an increase in<br />
L. variegatus causing erosion on coral reef.<br />
It is crucial to have balance within the<br />
environment to maintain healthy<br />
populations, such as the population on<br />
Spanish lookout Caye.
Introduction<br />
Lunar reproductive rhythms have been part<br />
of fishermans folklore since ancient times.<br />
As described in much detail by H. Monroe<br />
Fox in 1923, “Sizes of certain marine<br />
invertebrates, chiefly mollusks and<br />
echinoderms varies with the phases of the<br />
moon.” On a field biology course studying<br />
the Drowned Cayes, I noticed that L.<br />
variegatus was in abundance along the<br />
seawall on the North facing aspect of<br />
Spanish lookout Caye off the coast of Belize<br />
City, Belize. The objective of my research<br />
is to determine if L. variegatus mass<br />
increases in relation to the lunar cycle.<br />
Methods<br />
I selected the North facing concrete seawall<br />
on the West side of Spanish Lookout Caye.<br />
On 08/07/2011, 80/09,2011, and<br />
08/10/2011, I removed 14 individuals from<br />
the area by placing them in a bucket filled<br />
with seawater. All debris was removed from<br />
each individual by hand, then removed from<br />
the bucket for 15 seconds then placed on a<br />
scale to record each individuals body mass.<br />
Individual photos were taken and were<br />
reviewed later for photo identification in<br />
chronological order. Due to technical<br />
difficulties, I was unable to analyze<br />
individuals with photo I.D. from 08/10/2011.<br />
Results<br />
Although there was a trend there was no<br />
significant increase in body mass between<br />
day one and day two. (t=1.67), (df=4),<br />
(p=0.08). Despite failure to meet statistical<br />
criteria, each individual increased slightly<br />
from day one to day two. This suggests that<br />
it is plausible that increase in body mass of<br />
L. variegatus correlates to the lunar cycle.<br />
Green variation of L. variegatus<br />
Discussion<br />
Further investigation should been<br />
conducted using a larger sample size.<br />
Photo identification methods should be<br />
performed with the same high pixel lens<br />
throughout the study to ensure photo<br />
identification methods are correct, as this<br />
skewed my data and conclusions. Further<br />
research should be conducted throughout<br />
the entire duration of the lunar cycle, from<br />
new moon until full moon.<br />
Literature Cited<br />
Hendler, G., Kier, P.M., Miller, J.E.,<br />
Pawson, D.L. 1995. Sea Stars, Sea<br />
Urchins, and Allies: Echinoderms of<br />
Florida and the Caribbean. Washington:<br />
Smithsonian Institution Press; 390p<br />
Nichols, D. 1962. Echinoderms. London:<br />
Hutchinson & CO LTD; 192p.<br />
Pechenik, J.A. 2000. Biology of the<br />
Invertebrates, Fourth Edition. U.S.A.:<br />
McGraw-Hill Companies, Inc.; 578p.
Our Data Sheets<br />
We have designed our data sheets for long-term project use and built in redundancy to<br />
ensure we don’t miss data. They have been field tested over the past decade with high<br />
success.<br />
We organize the data sheets in field bins. Each bin should have sharpened pencils, a<br />
white eraser, and datasheets as described below. Please take the time to review these<br />
datasheets before fielding!<br />
We will use 5 bins total. Each volunteer/student will be responsible for one bin per day.<br />
After they master their data sheet, they will rotate jobs and teach each other, with<br />
oversight from the PI, Visiting Scientists, TAs, and Interns. The PI will personally teach<br />
and monitor for the first few days. If all goes as expected, interns and TAs will have<br />
mastered the forms with a few days in the field.<br />
One bin will have the Trip Summary (2-page data sheet). We should have 2 page-one<br />
of the Trip Summary per day (~20 per team) and 10 page-two per day (~100 per team).<br />
The team member who is responsible for this form is writing pretty much all day.<br />
The 2nd bin will have the Record of Effort (1-page data sheet) in it. We should have 4<br />
(front and back for a total 8) Record of Effort Sheets per day (~40/team). The team<br />
member who is responsible for this form is writing pretty much all day.<br />
The 3rd bin will have the Manatee Sighting and Scan (4-page data sheet). We should<br />
have 10 sets/day (~100/team). The team member who is responsible for this form is only<br />
writing during a scan or an opportunistic manatee sighting.<br />
The 4th bin will have the Habitat Sampling (4-page data sheet). We should have 10<br />
sets/day (~100/team). The team member who is responsible for this form is only writing<br />
during the habitat sample, which follows every scan.<br />
The 5th bin will have the Dolphin Survey (2-page data sheet). We should have 5<br />
sets/day (~50/team). The team member who is responsible for this form is documenting<br />
when we depart and return, but only writing down data when we encounter dolphins.<br />
Copies of our study site (map), marine life and bird field cards should also be kept in the<br />
bins.
TRIP SUMMARY SHEET (page 1) Day of the Week:<br />
Trip ID: Date: Julian Day:<br />
(yy-julianday-onedigittrip#) (dd-mon-yy) (001-365)<br />
EW Team #: # EW Vols: Total Obs:<br />
(yy-one digit team#)<br />
EW Team Members (first and last name):<br />
Researcher(s): Intern(s): Field Asst.:<br />
(first name) (firstname and initials) (firstname and initials)<br />
Effort Data Taken by:<br />
Sighting/Scan Data Taken by:<br />
Trip Summary Data Taken by:<br />
Behavioral Data Taken by:<br />
Seagrass Data Taken by:<br />
Locations (scan locations): ________________________________________________________<br />
_________________________________________________________________________<br />
Total # of Scans: Total # of Sightings (opp/scan): Total # Manatees:_______<br />
GENERAL WEATHER CONDITIONS<br />
DEPART DOCK: WEATHER CHECK I : WEATHER CHECK II:<br />
RETURN DOCK: Time: Time:<br />
*TIDES IN ORDER: Wind Speed: mph Wind Speed: mph<br />
High: Wind Direction: Wind Direction:<br />
Low: Sea State: Sea State:<br />
High: Swell Height: ft Swell Height: ft<br />
Low: Cloud Cover: Cloud Cover:<br />
High: Water Temp: °C Water Temp: °C<br />
Low: Air Temp: °C Air Temp: °C<br />
Rainfall Overnight: Salinity: PPT Salinity: PPT<br />
Rainfall Today: Barometer: HPa Barometer: HPa<br />
Cruise Ships (#):<br />
Comments:<br />
* If a high or low falls pre/post today’s date, put in parentheses in order of occurrence.<br />
Please Complete Trip Log on reverse side and additional pages!<br />
CSS/KSL ____________ Trip Summary Sheet Page_______Of_______<br />
(c) 2001-2004 Manatees in Belize - Earthwatch Project PIs: Katherine S. Lacommare & Caryn Self Sullivan -Updated 4/7/11
TRIP SUMMARY - LOG SHEET Date Page _____ of _____<br />
Record All Events during the Day<br />
TIME WPoint EVENT LOCATION LATITUDE LONGITUDE COMMENTS<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
<strong>17</strong>. 88.<br />
2001-2004 Manatees in Belize <strong>–</strong> Earthwatch Project PIs: Katherine S. LaCommare and Caryn Self Sullivan 4/7/2011
RECORD OF EFFORT DATA SHEET<br />
Event (include num.) Location or Route from/to: ____________________________<br />
Enter one of the following: Travel (trav)<strong>–</strong> Survey (surv)<strong>–</strong> Pt. Scan (pscn) <strong>–</strong> Sighting (sght)<strong>–</strong> Sampling (samp)- Other Work (othw) <strong>–</strong> Lunch (lunc)<strong>–</strong> Other Free (othf)<br />
Start time _________Stop time _______Average Speed (km/hr) Trip (km)______________<br />
Start WP (Garmin) _________ Stop WP (Garmin) ________EPE = _ # Manatees__________<br />
Comments:<br />
Event (include num.) Location or Route from/to: ____________________________<br />
Enter one of the following: Travel (trav)<strong>–</strong> Survey (surv)<strong>–</strong> Pt. Scan (pscn) <strong>–</strong> Sighting (sght)<strong>–</strong> Sampling (samp)- Other Work (othw) <strong>–</strong> Lunch (lunc)<strong>–</strong> Other Free (othf)<br />
Start time _________Stop time _______Average Speed (km/hr) Trip (km)_____________<br />
Start WP (Garmin) _________ Stop WP (Garmin) ________EPE = __ # Manatees_________<br />
Comments:<br />
Event (include num.) Location or Route from/to: ____________________________<br />
Enter one of the following: Travel (trav)<strong>–</strong> Survey (surv)<strong>–</strong> Pt. Scan (pscn) <strong>–</strong> Sighting (sght)<strong>–</strong> Sampling (samp)- Other Work (othw) <strong>–</strong> Lunch (lunc)<strong>–</strong> Other Free (othf)<br />
Start time _________Stop time _______Average Speed (km/hr) Trip (km)_____________<br />
Start WP (Garmin) _________ Stop WP (Garmin) ________EPE =__ # Manatees_________<br />
Comments:<br />
Event (include num.) Location or Route from/to: ____________________________<br />
Enter one of the following: Travel (trav)<strong>–</strong> Survey (surv)<strong>–</strong> Pt. Scan (pscn) <strong>–</strong> Sighting (sght)<strong>–</strong> Sampling (samp)- Other Work (othw) <strong>–</strong> Lunch (lunc)<strong>–</strong> Other Free (othf)<br />
Start time _________Stop time _______Average Speed (km/hr) Trip (km)_____________<br />
Start WP (Garmin) _________ Stop WP (Garmin) ________EPE = __ # Manatees_________<br />
Comments:<br />
DATE: ______________________ Effort Data Taken By? _____________________<br />
(dd-mon-yy)<br />
Trip ID: _____________________ RECORD OF EFFORT Page _______of________<br />
(yy-julianday-one digit trip#)<br />
(c) 2001-2007 Manatees in Belize - Earthwatch Project PIs: Caryn Self Sullivan & Katie LaCommare -Updated 4/7/11
MANATEE SIGHTINGS AND SCANS Page 1 of ____ ID # _______________________________<br />
(yy-julianday-sight#scan#)<br />
Sighting Number Sight Start (24 hr) Sight Stop<br />
Point Scan Number Scan Start (24 hr) Scan Stop<br />
Sighting Type (circle one): Opportunistic or Scan<br />
Wayppoint (Garmin): EPE:<br />
Location: Loc. Code (see list for proper code):<br />
MUST Enter for both sightings and scans ONLY Enter code if this is a scan (whether there is a sighting or not).<br />
# Of Other Boats in area during scan: Distance (closest point) (1) (2) ___(3)<br />
Comments (size, type, speed) (1) (2) (3)<br />
Manatee Sighted By__________________ Confirmed By________________________<br />
Enter one of the following: Researcher, Field Assistant, Intern, Volunteer, Team, Not confirmed<br />
*Total # Of Manatees No. Calves Calf Size NEW, YOY, OTHER<br />
1 ST 20 minutes of SCAN: Minimum: Maximum: Best Estimate: *TOTAL # DURING SIGHTING<br />
Total 30 minutes of SCAN: Minimum: Maximum: Best Estimate:<br />
__________________________________________________________________________________________<br />
Initial Distance to 1st Manatee at 1 st Sighting (m): Time: Initial Movement of 1st Manatee:<br />
Aw from boat Towards boat Milling No change Undetermined<br />
(CIRCLE Distance Detection Method: DIRECT INDIRECT ESTIMATE) (CIRCLE initial movement in relationship to observation boat)<br />
Habitat Type: (Circle habitat where the manatee is, NOT the scan pt. ) UNDETERMINED Resting hole-yes or no or unk<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Grassflat/UNK | Mud/Grass/Sand/Coral/UNK | Turtle/Shoal/Manatee/None/UNK<br />
DISTURBED? YES OR NO OR UNK Predominate Behavioral State (circle one):<br />
Feeding Resting Socializing Traveling Milling Undetermined Other (describe):<br />
__________________________________________________________________________________________<br />
Initial Distance to 2 nd Manatee at 1 st Sighting (m): Time: Initial Movement of 2 nd Manatee:<br />
Aw from boat Towards boat Milling No change Undetermined<br />
(CIRCLE Distance Detection Method: DIRECT INDIRECT ESTIMATE) (CIRCLE initial movement in relationship to observation boat)<br />
Habitat Type: (Circle habitat where the manatee is, NOT the scan pt. ) UNDETERMINED Resting hole-yes or no or unk<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Grassflat/UNK | Mud/Grass/Sand/Coral/UNK | Turtle/Shoal/Manatee/None/UNK<br />
DISTURBED? YES OR NO OR UNK Predominate Behavioral State (circle one):<br />
Feeding Resting Socializing Traveling Milling Undetermined Other (describe):<br />
__________________________________________________________________________________________<br />
Initial Distance to Center of Manatee Group at 1 st Sighting (m):____________________________________<br />
U/W Video Attempt : yes no; if yes, was it successful? _____ Seagrass Sample/Habitat Sample/No Sample<br />
SIGHTING/SCAN CONDITIONS<br />
Glare<br />
Yes<br />
No<br />
Glare<br />
Direction<br />
Cloud cover<br />
clear<br />
scattered<br />
partly<br />
mostly<br />
overcast<br />
Precipitation<br />
Dry<br />
Light rain<br />
Heavy rain<br />
Sea State<br />
Swell Height Tide State*<br />
High (+/- 60m)<br />
Low (+/- 60m)<br />
Flood Ebb<br />
* If Scan Start Time falls within 60 min of High or Low Tide, then circle High or Low; if not and Scan Start Time falls b/t High and Low, circle Ebb; if not and Scan<br />
Start Time falls b/t Low and High, circle Flood.<br />
DATE: ______________________ Sighting/Scan Data Taken By? ______________<br />
(dd-mon-yy)<br />
Trip ID: _____________________ SIGHTINGS AND SCANS Data Set _______of________<br />
(yy-julianday-one digit trip#)<br />
(c) 2001-2004 Manatees in Belize - Earthwatch Project PIs: Katherine S. Lacommare & Caryn Self Sullivan -Updated 4/7/11
MANATEE SIGHTINGS AND SCANS Page 2 of ____ ID # _______________________________<br />
(yy-julianday-sight#scan#)<br />
Site Map: On this page you will sketch the characteristics of the scan point in relationship to the<br />
Boat. Map Perspective: illustrator is standing on the driver’s seat and facing north when<br />
looking towards the top of this page. Sketch the above water characteristics and draw a vector<br />
(distance & direction) to each landmark and to each manatee sighted. Label each manatee vector<br />
with time, distance, & direction.<br />
N<br />
W E<br />
S<br />
(c) 2001-2004 Manatees in Belize - Earthwatch Project PIs: Katherine S. Lacommare & Caryn Self Sullivan -Updated 4/7/11
MANATEE SIGHTINGS AND SCANS Page 3 of ____ ID # _______________________________<br />
(yy-julianday-sight#scan#)<br />
Comments:_________________________________________________________________________________<br />
___________________________________________________________________________________________<br />
CatalogID # _______________________ Marked / Un-Marked / Undetermined Sex: M / F / U Prop Scars? Yes / No<br />
RESIGHT? _______________________ Tape Start _______Stop______ Animal _____ of _____ captured on tape<br />
Enter Feature Codes* 2 (see below)<br />
for each unique marking seen on animal<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
__ __ __ __ __ __ __ __ __ __<br />
Barnacles: Yes / No / Undetermined<br />
Number? ______ Size? ____________<br />
Remoras: Yes / No / Undetermined<br />
Number? ________________________<br />
Algae: Yes / No / Undetermined<br />
Color/Coverage? __________________<br />
Medial Notch: Yes / No / Undetermined<br />
Comments? ______________________<br />
Behavioral State: Rest / Feed / Travel / Social / Mill<br />
Play / Other / Undetermined<br />
Initial Reaction to diver: Approach / Retreat / No Change / Touch / Other<br />
Secondary Reaction to diver: Approach / Retreat / No Change / Touch / Other<br />
* 2 Identifying Feature Codes to use in blanks above, use one line for each unique marking:<br />
Type Region Position Number Size Color Shape<br />
S <strong>–</strong> scar D <strong>–</strong> dorsal F <strong>–</strong> flipper 1 <strong>–</strong> single L <strong>–</strong> large G <strong>–</strong> gray B <strong>–</strong> blotch<br />
M <strong>–</strong> mutilation L <strong>–</strong> left H <strong>–</strong> head 2 <strong>–</strong> 2 or 3 M <strong>–</strong> medium W <strong>–</strong> white L <strong>–</strong> line(s)<br />
D <strong>–</strong> deformity R <strong>–</strong> right A <strong>–</strong> ant. trunk 4 <strong>–</strong> 4 or more S <strong>–</strong> small<br />
F <strong>–</strong> freeze brand V <strong>–</strong> ventral B <strong>–</strong> mid. trunk<br />
N <strong>–</strong> medial notch C <strong>–</strong> post. trunk<br />
K <strong>–</strong> trunk plain D <strong>–</strong> peduncle<br />
L <strong>–</strong> tail plain X <strong>–</strong> ant. tail<br />
P <strong>–</strong> skin pigmentation Y <strong>–</strong> post. tail<br />
(c) 2001-2004 Manatees in Belize - Earthwatch Project PIs: Katherine S. Lacommare & Caryn Self Sullivan -Updated 4/7/11
MANATEE SIGHTINGS AND SCANS Page 4 of ____ ID # _______________________________<br />
(yy-julianday-sight#scan#)<br />
__________________________________________________________________________________________<br />
Initial Distance to 3 rd Manatee at 1 st Sighting ((m): Time: Initial Movement of 3 rd Manatee:<br />
Aw from boat Towards boat Milling No change Undetermined<br />
(CIRCLE Distance Detection Method: DIRECT INDIRECT ESTIMATE) (CIRCLE initial movement in relationship to observation boat)<br />
Habitat Type: (Circle habitat where the manatee is, NOT the scan pt. ) UNDETERMINED Resting hole-yes or no or unk<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Grassflat/UNK | Mud/Grass/Sand/Coral/UNK | Turtle/Shoal/Manatee/None/UNK<br />
DISTURBED? YES OR NO OR UNK Predominate Behavioral State (circle one):<br />
Feeding Resting Socializing Traveling Milling Undetermined Other (describe):<br />
__________________________________________________________________________________________<br />
Initial Distance to 4th Manatee at 1 st Sighting ((m): Time: Initial Movement of 4th Manatee:<br />
Aw from boat Towards boat Milling No change Undetermined<br />
(CIRCLE Distance Detection Method: DIRECT INDIRECT ESTIMATE) (CIRCLE initial movement in relationship to observation boat)<br />
Habitat Type: (Circle habitat where the manatee is, NOT the scan pt. ) UNDETERMINED Resting hole-yes or no or unk<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Grassflat/UNK | Mud/Grass/Sand/Coral/UNK | Turtle/Shoal/Manatee/None/UNK<br />
DISTURBED? YES OR NO OR UNK Predominate Behavioral State (circle one):<br />
Feeding Resting Socializing Traveling Milling Undetermined Other (describe):<br />
__________________________________________________________________________________________<br />
Sighting Comments (con’t):<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
____________________________________________________________________<br />
(c) 2001-2004 Manatees in Belize - Earthwatch Project PIs: Katherine S. Lacommare & Caryn Self Sullivan -Updated 4/7/11
Date: Page 1 of ____ Sight/Scan Record ID:<br />
(dd-mon-yy) (yy-julianday-sighting#scan#)<br />
HABITAT SAMPLING<br />
Sample ID: Data Recorded By:<br />
(yy-julianday-location code)<br />
LOCATION DATA<br />
LOC. CODE: WAYPOINT: LOCATION NAME:<br />
PHYSICAL DATA<br />
SECCHI READINGS (vertical secchi <strong>–</strong> take from center of boat; horizontal secchi <strong>–</strong> take from bow (0.5 meter)<br />
below the surface; secchi disk should face the sun.)<br />
Name (volunteer who took the data) Horiz. Vertical Depth Comments<br />
Air Temp<br />
°C<br />
H20 Temp<br />
Surface °C<br />
H20 Temp<br />
Bottom °C<br />
Salinity<br />
Surface ‰<br />
Salinity<br />
Bottom ‰<br />
Sea State<br />
Beaufort<br />
Swell<br />
Height (ft)<br />
Habitat Type: Circle the habitat characteristics of the scan point (where the boat is, NOT where the manatee<br />
was observed). This may be different from where the manatee was sighted or where the plot was set.<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Cove/Grassflat<br />
Mud/Sandy Mud/Muddy Sand/ Sand/Coral | Turtle/Shoal/Manatee/None<br />
Comments and other description:___________________________________________________________________<br />
______________________________________________________________________________________________<br />
______________________________________________________________________________________________<br />
PLOT DATA<br />
VECTOR (Distance in yards & Direction N, NE, E...) TO SAMPLE PLOT:<br />
Habitat Type: Circle the habitat characteristics of the plot NOT where the boat is but where the plot is. This<br />
may be different from where the boat is.<br />
InBogue/OutBogue/Reef | Chann/ChannEdge/DeadEndBogue/Lagoon/Cove/Grassflat | Mud/Sand/Coral | Turtle/Shoal/Manatee/None<br />
PLOT DEPTH at poles -- Center: North: South: East: West:<br />
Sediment Type (circle sediment type based on biomass cores): mud -- sand/coral <strong>–</strong> mud/sand -- -- other<br />
2001 Manatees in Belize <strong>–</strong> Earthwatch Project PIs: Katherine S. LaCommare and Caryn Self Sullivan 4/7/2011
Date: Page 2 of ____ Sight/Scan Record ID:<br />
(dd-mon-yy) (yy-julianday-sighting#scan#)<br />
BIOLOGICAL DATA<br />
BIOMASS CORES: (Enter zero if core would be empty; enter yes if core was taken; enter N/A if not taken. Take<br />
cores from the “a” corner of the small (density) quadrats. If you enter zero, a biomass datasheet must be filled out!!! This<br />
does not need to be entered into the database)<br />
N:__________S:__________E:__________W:__________<br />
DENSITY COUNTS<br />
4 samples:<br />
See plot design for sample layout.<br />
At least 2 Replicas: If possible, each replica should be taken by the same person. Please make comments about person<br />
who took data i.e., intern, volunteer etc.)<br />
Replica 1: Name (volunteer who took the data):<br />
A quarter B quarter C quarter D quarter Total<br />
N: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
S: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
E: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
W: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
Replica 2: Name (volunteer who took the data):<br />
A quarter B quarter C quarter D quarter Total<br />
N: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
S: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
E: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
W: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
Replica 3: Name (volunteer who took the data):<br />
A quarter B quarter C quarter D quarter Total<br />
N: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
S: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
E: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
W: T/ S/ M T/ S/ M T/ S/ M T/ S/ M T/ S/ M<br />
2001 Manatees in Belize <strong>–</strong> Earthwatch Project PIs: Katherine S. LaCommare and Caryn Self Sullivan 4/7/2011
Date: Page 3 of ____ Sight/Scan Record ID:<br />
(dd-mon-yy) (yy-julianday-sighting#scan#)<br />
PERCENT COVER (1 x 1 m QUADRAT)<br />
4 samples N,S,E,W: See plot design for sample layout. Each quadrat will be rolled 5 times.<br />
Replica 1: Name (volunteer who took the data):<br />
Sample N (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 1: Name (volunteer who took the data):<br />
Sample S (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 1: Name (volunteer who took the data):<br />
Sample E (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 1: Name (volunteer who took the data):<br />
Sample W (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
2001 Manatees in Belize <strong>–</strong> Earthwatch Project PIs: Katherine S. LaCommare and Caryn Self Sullivan 4/7/2011
Date: Page 4 of ____ Sight/Scan Record ID:<br />
(dd-mon-yy) (yy-julianday-sighting#scan#)<br />
Replica 2: Name (volunteer who took the data):<br />
Sample N (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 2: Name (volunteer who took the data):<br />
Sample S (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 2: Name (volunteer who took the data):<br />
Sample E (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
Replica 2: Name (volunteer who took the data):<br />
Sample W (grand total is entered into database) (rows 1-10)<br />
1<br />
2<br />
3<br />
4<br />
5<br />
Total<br />
Grand Total<br />
2001 Manatees in Belize <strong>–</strong> Earthwatch Project PIs: Katherine S. LaCommare and Caryn Self Sullivan 4/7/2011
DOLPHIN ENCOUNTER: Boat-Based Data Sheet Location: Drowned Cayes, Belize, Central America<br />
Survey # Encounter # Date: 20____ Boat: Captain: Photographer: # of images:<br />
(day month)<br />
Survey Start Time: End Time: Effort: Hours with Dolphins: Sunrise: Sunset: Moonrise: Moonset:<br />
Tide 1stHigh: 1stLow: High: Low: High: Total Survey Effort to Date: Total Hours with Dolphins to Date: Total Images to Date:<br />
Start Initial GPS Reading Sea Water Water Sighting Description Identified IDs:<br />
Time Tide State Swell Depth Temp Area Habitat Other Objects<br />
N W (m) (m) (C) General Comments:<br />
End Final GPS Reading Sea Water Water<br />
Sighting Description<br />
Time Tide State Swell Depth Temp Area Habitat Other Objects<br />
N W (m) (m) (C)<br />
Time Group Estimate ID Ratio<br />
Group Membership Marked:<br />
Calves Juveniles Adults Unmarked:<br />
< 1 month 1 - 3 mos 3 - 6 mos 6 - 10 mos Total < 1 yr 1 - 2 yr Total:<br />
Best Estimate ID-Ratio:<br />
SIGHTING LOG<br />
Way GPS Reading Area / Habitat Water Group Behavior Response to<br />
Time Pnt. N W Boats / Birds / Other depth Size State Research Vessel Notes and Comments<br />
Contact Details: Dr. Caryn Self-Sullivan, Ph.D., <strong>Sirenian</strong> <strong>International</strong>, 200 Stonewall Drive, Fredericksburg, VA 22401 Email: caryns@sirenian.org; Phone: 540.287.8207
Sighting Log Continued from Page 1<br />
Way GPS Reading Area / Habitat Water Group Behavior Response to<br />
Time Pnt. N W Boats / Birds / Other depth Size State Research Vessel Notes and Comments<br />
Contact Details: Dr. Caryn Self-Sullivan, Ph.D., <strong>Sirenian</strong> <strong>International</strong>, 200 Stonewall Drive, Fredericksburg, VA 22401 Email: caryns@sirenian.org; Phone: 540.287.8207
Caryn Self-Sullivan, Ph.D.<br />
200 Stonewall Drive, Fredericksburg, VA 22401-2110<br />
cselfsullivan@gmail.com | 540.287.8207<br />
Education<br />
Ph. D. 2008 Texas A&M University, College Station, TX, Department of Wildlife and Fisheries Sciences<br />
B. S. 1997 Coastal Carolina University, Conway, Department of Marine Science, minors in Biology and<br />
Mathematics<br />
2000 Pennsylvania State University, Graduate Program in Acoustics, non-degree seeking student (June)<br />
1996 Deakin University, Geelong, Victoria, Australia, exchange student (July-December)<br />
Professional Positions<br />
2011-Present Distance-Education Faculty (Adjunct), Fischler School of Education and the Oceanographic Center,<br />
Nova Southeastern University<br />
2010-2011 Assistant Professor (visiting, full-time faculty) Department of Fisheries & Wildlife, Virginia Tech<br />
2008-2010 Assistant Professor (temporary, full-time faculty) Department of Biology, Georgia Southern University<br />
2008 Substitute Teacher (Secondary Level) for Spotsylvania, Stafford, Fredericksburg, VA<br />
2010-Present Technical Expert to Friends of Swallow Caye, Swallow Caye Wildlife Sanctuary, Belize<br />
2010-Present Technical Expert to the Sarteneja Allicance for Conservation & Development, Corozol Bay Wildlife<br />
Sanctuary, Belize<br />
2007-Present Marine Science Advisor, Hugh Parkey Foundation for Marine Awareness & Education, Belize<br />
2005-Present Principal Advisor, NCRC Community Conservation of West African Manatees in Volta Lake, Ghana<br />
2005-Present Member, IUCN Species Survival Commission Sirenia Specialist Group<br />
2004-Present Member, Belize National Manatee Working Group<br />
2004-Present Member, Belize Marine Mammal Stranding Network<br />
2000-Present President & Co-founder, <strong>Sirenian</strong> <strong>International</strong>, Inc.<br />
1998-Present Principal Investigator, Ecology and Behavior of Antillean Manatees and Bottlenose Dolphins in the<br />
Drowned Cayes Area of Belize<br />
2004-2007 Director of Research & Education, Spanish Bay Conservation & Research Center and the Hugh Parkey<br />
Foundation for Marine Awareness and Education, Belize<br />
Professional Memberships: National Science Teachers Association (NSTA), North American Association for<br />
Environmental Education (NAAEE), Society for Marine Mammalogy (SMM), Society for<br />
Conservation Biology (SCB), Animal Behavior Society (ABS), American Pet Dog Trainers<br />
Association (APDT), Karen Pryor Academy Certified Training Partner (KPA CTP), <strong>Sirenian</strong><br />
<strong>International</strong> (SI)<br />
Teaching Experience<br />
2011-Present OCEE 540 Interpreting our Environment, Nova Southeastern University, Florida<br />
2011-Present OCEE 520 Teaching Environmental Concepts, Nova Southeastern University, Florida<br />
2010-2010 FIW 4314 Conservation Biology, Department of Fisheries & Wildlife Sciences, Virginia Tech<br />
2008-2010 BIOL 1130 General Biology, Department of Biology, Georgia Southern University<br />
2009-2010 BIOL 2108L Biology of Organisms Lab, Department of Biology, Georgia Southern University<br />
2009-2010 BIOL 2108SI Biology of Organisms Supplemental Instruction, Georgia Southern University<br />
2006-2011 Behavior, Ecology, and Conservation of Manatees & Dolphins in Belize, Independent Course with<br />
approved credit hours accepted by ~ a dozen accredited U.S. universities.<br />
2007-2009 EW/NCRC West African Fellows Manatee Conservation Workshop, Ghana, West Africa<br />
2008 Substitute for high school science classes in Spotsylvania, Stafford, & Fredericksburg, VA<br />
2004-2008 Marine Biology for Primary & Secondary Teachers Workshop, Belize Ministry of Education<br />
2004-2008 Biology of Mangroves, Sea Grass & Coral Reefs Field Course, Belize Ministry of Education<br />
1993-1997 Marine Biology, Chemistry, Mathematics, and Scientific Writing tutor, Coastal Carolina University<br />
1986-1993 Principals of Real Estate, Real Estate Brokerage, Real Estate Finance, Fair Housing, and Continuing<br />
Education Courses, Fredericksburg Area Association of Realtors, Virginia Real Estate Commission<br />
Additional Training in Education<br />
2011 KPA: Karen Pryor Academy Certified Training Partner<br />
2011 TTI: TAGteach <strong>International</strong> Certification<br />
2011 CCPDT: Real Solutions for Canine Behavior Problems by Pat Miller (June 11-12, 2011)<br />
2011 NSU: Blackboard Learn Distance-Education Training<br />
Caryn Self-Sullivan CV Page 1 of 1 Updated<br />
12/10/2011
2010 VT: Scholar, BANNER, Safe Zone Training<br />
2010 GSU: Teaching Large Classes, Teaching What You Don’t Know, Thinking Like Leonardo, Safe Space<br />
Training, Creating a Safe Learning Environment; Service-Learning "All for One and One for All<br />
Integrating Service with Teaching and Scholarship for Maximum Impact"; Successful Grant Writing<br />
Workshop; Inertia vs. Critical Mass Engaging Students in Active Discussions, Team-Based Learning-A<br />
Transformational Use of Small Groups in College Teaching; Scholarship of Teaching and Learning<br />
Commons; Teaching Large Classes FLC; Lab Safety Training, VISTA/WebCT/Blackboard/GeorgiaView;<br />
Service-Learning; CPS Classroom System; WINGS, BANNER, The Art of Changing the Brain; Teaching<br />
Large Classes FLC; POGIL Workshop; Active Learning in Large Classes; Active Learning for Students<br />
Graduate Students<br />
2011-Present Kristin Montour-Grubbs, Nova Southeastern University, M.S. Student, Committee Chair<br />
2011-Present Allen Conrad Allen, Nova Southeastern University, M.S. Student, Mentor<br />
2006-Present Haydee Dominguez, Duke University, Ph.D. Student, Mentor, Field Advisor, External Advisor. 2011<br />
Proposal: Manatees in the Dominican Republic.<br />
2006-2008 Marie-Lys Bacchus, Loma Linda University. M.S., Mentor, Field Advisor, Committee Member. 2007<br />
Thesis: Characterization of Resting Holes and Use by the Antillean Manatee (Trichechus manatus<br />
manatus).<br />
2009-2011 Kamla Aristide, University of Dschang, Cameroon, M.S. Student. Mentor, Co-advisor. 2011 Thesis:<br />
Activity Center, Habitat Use and Conservation of the West African Manatee (Trichechus senegalensis<br />
Link, <strong>17</strong>95) in the Douala-Edea Wildlife Reserve and Lake Ossa Wildlife Researve.<br />
2008-2009 Andrea Tanny, Georgia Southern University, M.S. Student, Mentor, Field Advisor, Committee Member.<br />
MS Proposal: 2008 Foraging Behavior in West Indian Manatees.<br />
Publications<br />
Appletons, W. et al. (in review). Magnitude of global marine biodiversity: one third of sea creatures discovered.<br />
Submitted to Science on 15 October 2010.<br />
Bacchus, M-L, S. G. Dunbar, C. Self-Sullivan. 2009. Characterization of resting holes and their use by the Antillean<br />
manatee (Trichechus manatus manatus) in the Drowned Cayes of Belize. Aquatic Mammals 35(1)62-71.<br />
Deutsch, C. J., C. Self-Sullivan, and A. A. Mignucci-Giannoni. 2007. Trichechus manatus. In 2007 IUCN Red List of<br />
Threatened Species (www.iucnredlist.org).<br />
Hoffman, M., et al. 2010. The impact and shortfall of conservation on the status of the world’s vertebrates. Published<br />
in Science Express online 10/26/2010; pending print publication date.<br />
LaCommare, K. S., C. Self-Sullivan, and S. Brault. 2008. Distribution and Habitat Use of Antillean Manatees<br />
(Trichechus manatus manatus) in the Drowned Cays Area of Belize, Central America. Aquatic Mammals<br />
34(1)34-43.<br />
LaCommare, K.S., S. Brault, C. Self-Sullivan, E.M. Hines. (in review). A Boat-based Method for Monitoring<br />
<strong>Sirenian</strong>s: Antillean Manatee Case Study.<br />
Schipper, J., et al. 2008. The status of the world's land and marine mammals: diversity, threat, and knowledge.<br />
Science 322225-230, 10 October 2008. DOI 10.1126/science.1165115.<br />
Self-Sullivan, C. 2004-2010. Manatee and mangrove special topics in the 6th, 7th, & 8th editions of Moon<br />
Handbooks Belize by Joshua Berman and Chicki Mallan, Avalon Travel Publishing, Emeryville, CA.<br />
Self-Sullivan, C. 2008. Conservation of Antillean Manatees in the Drowned Cayes Areas of Belize. Ph.D.<br />
Dissertation, Department of Wildlife and Fisheries Sciences, Texas A&M University.<br />
Self-Sullivan, C. (in press). West Indian Manatees (Trichechus manatus) in the Wider Caribbean Region. Book<br />
Chapter in <strong>Sirenian</strong> Conservation Issues and Strategies in Developing Countries. Edited by Ellen Hines, John<br />
Reynolds, Lemnuel Aragones, Antonio A. Mignucci-Giannoni, Miriam Marmontel. University of Florida<br />
Press.<br />
Self-Sullivan, C. and A. A. Mignucci-Giannoni. 2007. Trichechus manatus manatus. In 2007 IUCN Red List of<br />
Threatened Species (www.iucnredlist.org).<br />
Self-Sullivan, C., and L. Bird. 2007. Marine Science for Students Workbook. BRC Publishers, Benque, Belize.<br />
Self-Sullivan, C., Smith, G. W., Packard, J. M., and LaCommare, K. S. 2003. Seasonal occurrence of male Antillean<br />
manatees (Trichechus manatus manatus) on the Belize Barrier Reef. Aquatic Mammals 29(3)342-354.<br />
Williams, Jr., E. H., Mignucci-Giannoni, A. A., Bunkley-Williams, L., Bonde, R. K., Self-Sullivan, C., Preen, A., and<br />
Cockcroft, V. G. 2003. Echeneid-sirenian associations, with information on sharksucker diet. Journal of Fish<br />
Biology 631<strong>17</strong>6-1183.<br />
Caryn Self-Sullivan CV Page 2 of 2 Updated<br />
12/10/2011
Unpublished Reports<br />
Ofori-Danson, Patrick, Caryn Self-Sullivan, Victor M. Mombu, and Martin A. Yelibora. 2008. Enhancing<br />
Conservation of the West African Manatee in Ghana 2007 Annual Report. Prepared by Nature Conservation<br />
Research Centre (NCRC), P. O. Box KN925, Accra. June 2008.<br />
Self-Sullivan, C. 2006. Report to NCRC on an Insular Population of West African Manatees (Trichechus<br />
senegalensis) in the Upper Afram Arm of Volta Lake.<br />
Self-Sullivan, C., and K. S. LaCommare. 1998-2010. Annual Progress Reports to Belize Forestry Department, Belize<br />
Fisheries Department, Belize Coastal Zone Management Authority & Institute, Friends of Swallow Caye,<br />
Earthwatch Institute.<br />
Self-Sullivan, C., M. Fogel, and M. J. Wooller. 2005. Carbon and nitrogen stable isotope analyses of Antillean<br />
manatees from three distinct ecosystems in Belize inference about relative differences in diets and habitats.<br />
Preliminary Report to Wildlife Trust Belize, 30 October 2005.<br />
Conference Presentations & Posters<br />
Bonde, R. K., P. Lewis, D. A. Samuelson, C. Self-Sullivan, N. E. Auil, and J. A. Powell. 2006. Belize manatee<br />
(Trichechus manatus manatus) epibionts SEM viewing techniques. Poster Presentation at SMM 16 th Biennial<br />
Conference on the Biology of Marine Mammals, San Diego, 12-16 December 2005.<br />
LaCommare, K. S., C. Self-Sullivan, and S. Brault. 2005. Distribution, habitat use and seasonal occurrence of<br />
Antillean manatees in the Drowned Cayes area of Belize, Central America. Oral Presentation at the SMM<br />
16th Biennial Conference on the Biology of Marine Mammals, San Diego, 12-16 December 2005.<br />
LaCommare, K. S., Sullivan, C. S., Brault, S. 2001. Distribution and Foraging Ecology of Antillean Manatees<br />
(Trichechus manatus) in the Drowned Cays Area of Belize, Central America. Poster Presentation at the SMM<br />
14 th Biennial Conference on the Biology of Marine Mammals, Vancouver, British Columbia, 28 November - 3<br />
December 2001.<br />
Mignucci-Giannoni, A. A., and C. Self-Sullivan. 2005. Conservation Status of the Antillean Manatee (Trichechus<br />
manatus manatus) in the Wider Caribbean Region. Oral Presentation at the <strong>International</strong> <strong>Sirenian</strong> Workshop,<br />
SMM 16 th Biennial Conference on the Biology of Marine Mammals, San Diego, 12-16 December 2005.<br />
Self-Sullivan, C. 2009. Conservation of Antillean manatees in Belize 1998-2008. Oral presentation at the 2009<br />
<strong>International</strong> <strong>Sirenian</strong> Conservation Conference, 23-24 March 2009.<br />
Self-Sullivan, C. 2009. Triumph on the commons in Belize the importance of traditional knowledge and stakeholder<br />
input to successful MPAs. Oral Presentation at the <strong>Sirenian</strong> Workshop, Improving the Contribution of Marine<br />
Protected Areas to the Conservation of <strong>Sirenian</strong>s. Society for Conservation Biology Marine Section,<br />
<strong>International</strong> Marine Conservation Congress, May 21-24, 2009, Johnson Center, George Mason University,<br />
Fairfax, Virginia, USA.<br />
Self-Sullivan, C. 2008. Conservation of Antillean manatees in Belize. Seminar (3 hours) at Coastal Zone<br />
Management Authority and Institute, Belize City, Belize.<br />
Self-Sullivan, C. 2006. Non-lethal boat scars on manatees in Belize as a tool for evaluation of a Marine Protected<br />
Area. Oral Presentation at the 59th Annual Gulf and Caribbean Fisheries Institute Conference, Belize City, 6-<br />
11 November 2006 (Science and Management of Marine Protected Areas Session).<br />
Self-Sullivan, C. 2007. Community Conservation of West African Manatees in the Upper Afram Arm of Volta Lake,<br />
Ghana. Oral Presentation, Marine Mammals & Indigenous Communities Workshop, SMM <strong>17</strong>th Biennial<br />
Conference on the Biology of Marine Mammals, Cape Town, November 2007.<br />
Self-Sullivan, C. 2007. Conservation and Status of West African Manatees in Ghana, Senegal, Nigeria, Cameroon,<br />
and Sierra Leone. Oral Presentation, Marine Mammals in Africa Workshop, SMM <strong>17</strong>th Biennial Conference<br />
on the Biology of Marine Mammals, Cape Town, November 2007.<br />
Self-Sullivan, C. LaCommare, K.S. 2003. One Issue Related to Research and Conservation: Cruise Ship Tourism in<br />
Belize. Oral Presentation at the <strong>International</strong> <strong>Sirenian</strong> Workshop, SMM 15 th Biennial Conference on the<br />
Biology of Marine Mammals, Greensboro, North Carolina, 14 December 2003.<br />
Self-Sullivan, C., and A. A. Mignucci-Giannoni. 2006. Conservation Status of the Antillean Manatee (Trichechus<br />
manatus manatus) in the Wider Caribbean Region. Oral Presentation at IMC9, Sapporo, Japan, 31 July 2005.<br />
Self-Sullivan, Caryn. 1996. A Survey of Macroalgae in Discovery Bay, Jamaica Species Diversity and Zonation<br />
Differences in the Lagoon. Senior Thesis, Coastal Carolina University, SC Oral presentation to Discovery<br />
Bay Lab, Jamaica.<br />
Self-Sullivan, Caryn. 1996. The Effects of Benthos on Acartia tonsa Population Dynamics. National Conference on<br />
Undergraduate Research, Ashville, NC POSTER - Results of NSF-REU Fellowship research. Oral<br />
Caryn Self-Sullivan CV Page 3 of 3 Updated<br />
12/10/2011
presentation to Horn Point Environmental Lab and Coastal Carolina University.<br />
Sullivan, C. S., LaCommare, K. S., Packard, J. M., and Evans, W. E. 2001. Seasonal Occurrence of Male Antillean<br />
Manatees (Trichechus manatus manatus) on the Belize Barrier Reef and the Effect of Scale on Seasonal<br />
Distribution Studies. Poster Presentation at the SMM 14 th Biennial Conference on the Biology of Marine<br />
Mammals, Vancouver, British Columbia, 28 November - 3 December 2001.<br />
Sullivan, C. S., Packard, J. M., and Evans, W. E. 1999. Spring distribution and behavior of Antillean manatees in the<br />
Drowned Cayes, Belize. Poster Presentation at the SMM 13th Biennial Conference on the Biology of Marine<br />
Mammals, Maui, Hawaii, November 28 - December 3, 1999.<br />
Achievements, Awards, Honors<br />
• 2011 B.F. Skinner Scholarship, KPA<br />
• 2009 CCU COS Alumnus of the Year<br />
• 2003 Earthwatch Young Scientist Award<br />
• 1998-2001 NSF Graduate Fellowship<br />
• 1997 Outstanding Marine Science Senior<br />
• 1997 President's Excellence Nominee<br />
• 1996 Exchange student to Deakin<br />
University<br />
• 1995 NSF-REU Fellowship<br />
• 1994-1997 Nelson Scholarship<br />
• 1993 Freshman Essay Contest<br />
• 1993-1997 President's Honor List<br />
• 1993-1997 Dean's Honor List<br />
• Sigma Zeta Beta Mu Honor Society<br />
• Omicron Delta Kappa Society<br />
• Pi Mu Epsilon Honor Society<br />
Grants and Fellowships<br />
• Marine Mammal Commission 2009. <strong>Sirenian</strong>/MPA Workshop. Amount funded $12,105.<br />
• Earthwatch Institute, Center for Research 2001-2007. Field Research. Amount funded ~ $56,000/year<br />
• UNDP/GEF/COMPACT, 2006. HPF Educational Outreach Programme. Amount funded $40,000USD<br />
• Project AWARE Small Grant Program 2004. Field Research. Amount Funded $1,000<br />
• Earthwatch Institute, Center for Research 2003 Young Scientist Award. Amount funded $2,500<br />
• Conservation Action Fund, New England Aquarium 2001. Capacity Building Grant. Amount funded $5,000<br />
• American Museum of Natural History Lerner-Grey Fund for Marine Research 2000. Field Research. Amount<br />
funded $2,000.<br />
• Oceanic Society 1998-1999. Field Research. Amount funded undisclosed (9 months of field research in Belize)<br />
• National Science Foundation Graduate Fellowship 1998-2001. Tuition & Stipend. Amount funded $75,000<br />
Other Service Projects<br />
• Workshop Organizer Society for Conservation Biology, 2009 IMCC Conference, <strong>Sirenian</strong>s and MPAs<br />
• Peer reviewer for academic journals: Aquatic Mammals, Behaviour, Journal of Wildlife Management, USGS<br />
Sirenia Project, Journal of Environmental Studies and Sciences, Estuaries & Coasts<br />
• Scientific Advisor to field scientists from Guatemala, Ghana, and Dominican Republic regarding techniques for<br />
evaluating manatee populations in remote areas, with an emphasis on the incorporation of manatee behavior<br />
• Mentor young people (especially women) entering the sciences (M.S., Ph.D. students and interns)<br />
• Science Judge National Oceans Science Bowl Regional Competition, Texas A&M, 2006<br />
• Coordinator National Marine Science Symposium, Belize City, Belize 2006<br />
• Coordinator National Marine Science Quiz Bowl, Belize City, Belize 2006-2007.<br />
• Committee Member Society for Marine Mammalogy Conference Committee, 2003<br />
Educational Outreach Publications and Presentation: >100 public presentations on animal behavior, marine mammals,<br />
marine and environmental science through NGO and academic organizations including: public schools & dive clubs, the<br />
Sierra Club, Green Drinks, New England Aquarium, Coastal Carolina University, Mary Washington University, Texas<br />
A&M University, Universidad Autonoma de Santo Domingo, Rutgers University, Georgian Court University, Brookdale<br />
Community College, The Wetlands Institute, and at community festivals, etc.<br />
References<br />
Jane Dougan, Distance-Ed Coordinator, Nova Southeastern University, douganj@nova.edu, 866.229.2557<br />
Stephen P. Vives, Ph.D., Georgia Southern University, Statesboro, GA, svives@georgiasouthern.edu, 912.478.5487<br />
Bruce A. Schulte, Ph.D., Western Kentucky University, Bowling Green, KY, bruce.schulte@wku.edu, 270.745.5999<br />
Jessica R. Young, Ph.D., Western State College of Colorado, jyoung@western.edu, 970.943.3045<br />
Michelle Kopfer, Teaching Assistant, Virginia Tech, Blacksburg, VA, midavis1@vt.edu, 540.577.1081<br />
Paul J. Camp, Ph.D., Professor, Spelman College, pjcamp@gmail.com, 404.270.5864<br />
Jane M. Packard, Ph.D., Professor, Texas A&M University, College Station, TX, j-packard@tamu.edu, 979.845.1465<br />
Wyndylyn von Zharen, Ed.D., J.D., Professor, Texas A&M University, dr_vonzharen@msn.com, 409.740.4485<br />
Caryn Self-Sullivan CV Page 4 of 4 Updated<br />
12/10/2011
145 S. Tompkins Street<br />
Howell, MI 48843<br />
K ATHERINE S PENCER L A C OMMARE<br />
EDUCATION<br />
PhD. June 2011. Environmental Biology, University of Massachusetts, Boston, MA<br />
MSc. 1994. Forestry and Natural Resources, Purdue University, West Lafayette, IN<br />
B.Sc. 1989. Anthropology/Zoology, University of Michigan, Ann Arbor, MI<br />
Kslacommare@gmail.com<br />
(248) 756-3985<br />
TEACHING EXPERIENCE<br />
Instructor: Introduction to Environmental Science (2004, 2008-2011). Face-to-face; Hybrid,<br />
Online. Lansing Community College, Lansing, MI,<br />
Instructor: Organismal Biology (2010-2011). Face-to-face. Lansing Community College, Lansing,<br />
MI,<br />
Instructor: Behavior and Conservation of Marine Mammals Field Course. (2008-2010) Spanish<br />
Bay Research and Education Center, Belize, C.A.<br />
Field Instructor: Coastal Ecology and Marine Mammal Ecology (1998-2007) Oceanic Society,<br />
Earthwatch<br />
Teaching Assistant: Cell Biology (2001). University of Massachusetts, Boston<br />
Teaching Assistant: General Biology (2001). University of Massachusetts, Boston<br />
Writing Instructor: Introduction to Environmental Science (1999-2000), University of<br />
Massachusetts, Boston,<br />
PROFESSIONAL DEVELOPMENT AND CERTIFICATIONS<br />
University of Michigan Community College Biology Active Learning Workshop: August<br />
2011, Ann Arbor, MI<br />
Teaching Ecology to Undergraduates: Spring 2011. Pierce Creek Institute, Hastings MI.<br />
What should college biology look like? the first two years: Spring 2001. Michigan State<br />
University<br />
Leading Effective Discussions: Spring 2010. Lilly Seminar Series <strong>–</strong> Michigan State University<br />
Transforming Learning Through Teaching: Spring 2009. 14-week professional development<br />
course through Lansing Community College’s Center for Teaching Excellence.<br />
Michigan Virtual University Certification: Fall 2008<br />
PROFESSIONAL EXPERIENCE<br />
Principal Investigator: January 2001 <strong>–</strong> December 2003, Co-PI: January 2004 <strong>–</strong> Dec. 2007, Manatees in<br />
Belize, Earthwatch Project (part-time, contract)<br />
• Managed all project logistics: developed and managed the budget, hired and supervised staff,<br />
recruited and hired student interns, trained interns and volunteers, completed endangered species<br />
permit applications, wrote agency reports, wrote and received grants, hired vendors, conducted<br />
government relations and managed volunteer activities and developed and managed education<br />
activities.
• Developed and managed educational activities including: determining priorities, developing<br />
curriculum and leading expedition activities for students and volunteers.<br />
• Designed research protocol. Collected, analyzed, presented and published project data. Utilized<br />
Excel, SPSS, MatLab.<br />
• Managed data collection, training, education and field trips for 1-5 two-week teams per year of 8-24<br />
volunteers per team.<br />
• Managed project logistics: budgets ($43,000 per year), permits, agency reports, and vendors.<br />
• Managed 2-3 field assistants and student interns per team.<br />
• Networked with local agencies and NGO’s to develop an appropriate manatee research program.<br />
• Conducted 1-3 public/professional speaking events per year.<br />
• Authored or co-authored 4 scientific papers.<br />
• Designed and maintained an Access database to archive, organize and house individual manatee,<br />
manatee population, habitat and seagrass data.<br />
Aerial observer, (Part-time - contract) April - May 2001, National Marine Fisheries Service, Northeast<br />
Fisheries Science Center, Woods Hole, MA. Supervisor: Pat Gerrior.<br />
• Participated in line transect surveys from a twin engine Otter for right and other large whales off<br />
the Massachusetts coast.<br />
Fisheries Research Biologist, May 1999 <strong>–</strong> August 2000, National Marine Fisheries Service, Northeast<br />
Fisheries Science Center. CMER Assistantship program, Woods Hole, MA. Supervisor: Tim Smith<br />
• Analyzed distribution and habitat ecology of Brydes whales in the North Pacific with ArcView GIS<br />
and SPlus statistical package.<br />
Field Naturalist and Biologist: (Contract Position) October 1998 <strong>–</strong> March 2000, Oceanic Society<br />
Expeditions, Belize Dolphin/Manatee Project, Drowned Cayes, Belize, Central America<br />
• Designed research methods, collected and analyzed data on the habitat and foraging ecology of<br />
manatees.<br />
• Coordinated volunteer data collection.<br />
• Coordinated daily educational and natural history activities for volunteers.<br />
Land Protection Specialist: March 1995 <strong>–</strong> October 1998, The Nature Conservancy, Michigan Field<br />
Office, East Lansing, MI, Supervisor: Bill McCort<br />
• Developed and managed a landowner education program to promote conservation stewardship of<br />
private lands. Program activities included educating landowners about endangered habitats and<br />
assisting them in maintaining these habitats on their property.<br />
Natural Areas Preservation Assistant: October 1994 <strong>–</strong> March 1995, Ann Arbor Parks and Recreation<br />
Department, Ann Arbor, MI, Supervisor: David Borneman<br />
• Assisted Natural Areas Preservation Coordinator with general duties.<br />
• Managed plant inventory database.<br />
• Coordinated volunteer workdays in city parks.<br />
PUBLICATIONS<br />
LaCommare, Katherine et al. (in review). A Boat-based Method for monitoring <strong>Sirenian</strong>s: Antillean<br />
Manatee Case Study. Biological Conservation<br />
Aragones, Lemenuel; Katherine S. LaCommare, Sarita Kendall, Delma Nataly Castelblanco-<br />
Martinez, Daniel Gonzalez-Socoloske. In press. Boat and Land-based Surveys for <strong>Sirenian</strong>s in<br />
Developing Nations. In Hines et al. <strong>International</strong> Strategies for Manatee and Dugong<br />
Conservation. Florida University Press<br />
LaCommare, Katherine et al. 2008. Distribution and Habitat Use of Antillean Manatees in the Drowned<br />
Cayes Area of Belize, Central America. Aquatic Mammals Journal. 34(1): 35-42
Self Sullivan, Caryn, Gregory W. Smith, Jane M. Packard, Katherine S. LaCommare. 2003.<br />
Seasonal Occurrence of Male Antillean Manatees (Trichechus manatus manatus) on the Belize Barrier Reef.<br />
Aquatic Mammals. 29(3): 342-354.<br />
Weiler, Katherine S. and Joseph T. O’Leary. May 1994. Demographic Change and Forest Resources:<br />
Implications for the Lake States. Lake States Forest Resource Assessment.<br />
PUBLICATIONS IN PREPARATION<br />
LaCommare, Katherine et al. (in prep). Coastal Development in Belize: Impact on Manatees and<br />
Seagrass.. Aquatic Botany.<br />
LaCommare, Katherine et al. (in prep). Conservation Planning for Antillean Manatees in Belize.<br />
Conservation Biology.<br />
POPULAR PUBLICATIONS<br />
LaCommare, Katherine S. Fall 2000. <strong>Sirenian</strong> Song. Fall newsletter for the Evergreen Group’s Manatee<br />
ambassador program<br />
LaCommare, Katherine S. Spring 2000. <strong>Sirenian</strong> Song. Spring newsletter for the Evergreen Group’s<br />
Manatee ambassador program<br />
LaCommare, Katherine S. Fall 1999. <strong>Sirenian</strong> Song. Fall newsletter for the Evergreen Group’s Manatee<br />
ambassador program<br />
Adimey, Nicole M. and Katherine S. Weiler. October 1994. Orcas of Johnstone Strait. Waterway Times<br />
Weiler, Katherine. December 1994. The Bald Eagle Soars, But What About the Fuzzy Sandoze? The Bard<br />
PRESENTATIONS AT NATIONAL CONFERENCES<br />
LaCommare, Katherine et al May 2011. Oral Presentation. A Boat-based method for monitoring<br />
<strong>Sirenian</strong>s: Antillean Manatee Case Study. Oral Presentation. <strong>International</strong> Marine Conservation<br />
Congress. Victoria, BC Canada. May 2011.<br />
LaCommare, Katherine et al. May 2009. Mangrove Clearing and Bottom Dredging: How is Antillean<br />
Manatee (Trichechus manatus manatus) Habitat Use Affected by Coastal Development in Belize?. Oral<br />
Presentation. <strong>International</strong> Marine Conservation Congress. Washington, DC. May 2009.<br />
LaCommare, Katherine et al. December 2005. Distribution, Seasonal Occurrence and Habitat Use of<br />
Antillean Manatees in the Drowned Cayes Area of Belize, Central America. Oral Presentation. 16th Biennial Conference on the biology of Marine Mammals. San Diego, California. November<br />
2005.<br />
LaCommare, Katherine et al. December 2003. Habitat Use of Seagrass in the Drowned Cayes Area of<br />
Belize, Central America. Poster Presentation. 15th Biennial Conference on the biology of Marine<br />
Mammals. Greensborough, SC.<br />
LaCommare, Katherine et al. December 2001. Habitat Use of Antillean Manatees in the Drowned Cayes<br />
Area of Belize, Central America. Poster Presentation. 14th Biennial Conference on the biology of<br />
Marine Mammals. Vancouver, BC<br />
INVITED PRESENTATIONS<br />
LaCommare, Katherine S. April 2002. Habitat Ecology of Antillean Manatees in the Drowned Cayes Area<br />
of Belize. Earthwatch Chicago, Spring Event, Kent Culinary Institute, Chicago, IL<br />
LaCommare, Katherine S. March 2002. Habitat Ecology of Antillean Manatees in the Drowned Cayes<br />
Area of Belize. Science in Action Lecture Series, Maritime Aquarium, Norwalk, CT
LaCommare, Katherine S. November 2001. Habitat Ecology of Antillean Manatees in the Drowned Cayes<br />
Area of Belize. Lowell Lecture Series, New England Aquarium, Boston, MA<br />
FUNDING AND AWARDS<br />
June 2007 Dissertation Improvement Grant, University of Massachusetts, Office of Sponsored<br />
Programs. Project Title: Ecology and Behavior of Antillean Manatees in the Drowned Cays, Belize, Central<br />
America. Funding: $1,200<br />
January 2003 <strong>–</strong> December 2006. Co-Principal Investigator. Earthwatch Institute. Project Title: Ecology<br />
and Behavior of Antillean Manatees in the Drowned Cays, Belize, Central America. Funding: $43,000/year<br />
October 2002. Young Scientist Award. Earthwatch Institute. Project Title: Ecology and Behavior of<br />
Antillean Manatees in the Drowned Cays, Belize, Central America. Funding: $5,000<br />
January 2001 <strong>–</strong> December 2003. Principal Investigator. Project Title: Ecology and Behavior of Antillean<br />
Manatees in the Drowned Cays, Belize, Central America. Funding: $43,000/year<br />
December 2000. Conservation Action Fund. Project Title: Ecology and Behavior of Antillean Manatees in the<br />
Drowned Cays, Belize, Central America. Award: $5,120<br />
ADVISORY BOARDS/PROFESSIONAL SOCIETIES<br />
Livingston Land Conservancy, Board of Directors, Fundraising and Membership Chair, (2010present)<br />
Livingston Land Conservancy, Fundraising Committee (2004, 2010)<br />
<strong>Sirenian</strong> <strong>International</strong>, Inc., Co-founder<br />
<strong>Sirenian</strong> <strong>International</strong>, Inc.,Board of Directors, Secretary, (2000-2005)<br />
Michigan Science Teachers Association, Member<br />
National Science Teachers Association, Member<br />
Michigan Association of Community College Biology Teachers, Member<br />
Science Department Curriculum Committee, Adjunct Faculty Representative<br />
REFERENCES<br />
Meg Clark Elias, Professor, Science Department, Lansing Community College, Lansing, MI 48933. (5<strong>17</strong>)<br />
483-1557, Clarkm1@lcc.edu<br />
Solange Brault, Associate Professor, University of Massachusetts, Boston, Department of Biology, 100<br />
Morrissey Blvd., Boston, MA 02125. (6<strong>17</strong>) 287-6683, Solange.brault@umb.edu<br />
Robert Stevenson, Associate Professor, University of Massachusetts, Boston, Department of Biology, 100<br />
Morrissey Blvd., Boston, MA 02125. (6<strong>17</strong>) 287-6579, Robert.stevenson@umb.edu<br />
Caryn Self Sullivan, Adjunct Faculty, Nova Southeastern University, President, <strong>Sirenian</strong> <strong>International</strong>. (540)<br />
287-8207, cselfsullivan@sirenian.org<br />
William D. McCort, Ph.D., Director and Manager of Canine Professor LLC, Chair, Natural Areas Technical<br />
Advisory Committee, Natural Areas Preservation Program, Washtenaw County Parks & Recreation<br />
Department, Volunteer, Land Protection Committee, Washtenaw Land Trust, 18819 Bush Road,<br />
Chelsea, MI 48118-9749. 734-475-6908, billmccort@comcast.net
CURRICULUM VITAE<br />
HEATHER J. KALB<br />
Department of Natural Sciences and Mathematics, West Liberty University,<br />
West Liberty University, West Liberty, WV 26074<br />
(812) 461-8965 (C); heather.kalb@westliberty.edu<br />
Education<br />
Ph.D. in Zoology, 1999. Dept. Biology, Texas A&M University, College Station, TX. “Behavior and<br />
Physiology of Solitary and Arribada Nesting Olive Ridley Sea Turtles during the Internesting<br />
Period.” General areas of specialization: chelonians, conservation, reproductive physiology and<br />
behavior. Academic Advisor: Dr. David W. Owens<br />
B.A., 1989. Major Biology, Minor Chemistry, Wittenberg University, Springfield, OH.<br />
Positions Held<br />
Assistant Professor, Dept. of Natural Sciences and Mathematics, West Liberty University, Wheeling,<br />
WV 26074. Aug. 2010 - current<br />
Assistant Professor (temp.), Dept. of Biology, Georgia Southern University, Statesboro, GA 30460.<br />
Aug. 2008 - 2010<br />
Assistant Professor, Dept. of Biology, University of Evansville, Evansville, IN 47722.<br />
Aug. 2003 - May 2008<br />
Assistant Professor (temp.), Dept. of Biology, Arkansas State Univ., State University, AR 72467.<br />
Aug. 2002 - May 2003<br />
Lecturer, Dept. of Biology, West Chester University, West Chester, PA 19383.<br />
Aug. 2000-May 2002<br />
Teaching Assistant, Dept. of Biology, Texas A&M University, College Station, TX 77845.<br />
1989-92, 1994-99. Responsibilities: teach labs, write/grade quizzes & exams, hold office hours.<br />
Course Director / Assistant Lecturer for “Advanced Sea Turtle Biology” course. Jan. 1998.<br />
Méxican Sea Turtle Center in Mazunte, Oaxaca, México.<br />
Primary instructors: D. Owens (Dept. of Biology, Texas A. & M. Univ.), J. Flanagan<br />
(Veterinarian, Houston Zoo), R. Marquez (National Coordinator, Sea Turtle Research, Méxican<br />
National Fisheries Institute)<br />
Students: 15 Méxican/Cuban government researchers, veterinarians, and students.<br />
Range of topics: reproductive physiology & behavior, disease, and life history and hands-on<br />
training with laparoscopy, ultrasonography, blood sampling, and radio-telemetry.<br />
Assistant Lecturer/Organizer for “Reproductive Physiology of Sea Turtles” course.<br />
2-week course at Rancho Nuevo, México. May 1996.<br />
Courses Taught (locations where the course was taught are in parenthesis)<br />
Intro level courses, Majors<br />
• Animal Diversity/Introductory Zoology, lecture and lab (UE, ASU, WCU)<br />
• Biology 1 lab (GSU, UE, TAMU)<br />
• Biology 2 lab (GSU, TAMU)<br />
• Scientific Communications (UE)<br />
Upper level courses, Majors<br />
• Comparative Animal Physiology lecture and lab (GSU, UE, WLU)<br />
1
• Comparative Vertebrate Anatomy, lecture and lab (WCU, TAMU)<br />
• Animal Behavior, lecture (UE, WLU)<br />
• Vertebrate Zoology, lecture and lab (UE, WLU)<br />
• Conservation Biology, lecture (WCU)<br />
• Vertebrate Ecology, lecture and lab (WCU)<br />
• Undergraduate research (individual student) (GSU, UE)<br />
Biology courses for (usually) non-majors<br />
• Human Anatomy and Physiology 1, lecture and lab (GSU, WCU)<br />
• Human Anatomy & Physiology II, lecture and lab (WCU)<br />
• Microbiology lecture (summer 2010) (GSU)<br />
• Microbiology lab (scheduled for fall 2010) (GSU)<br />
• General Biology, lecture and lab (GSU, UE, ASU, WLU)<br />
• Environmental Science, lecture (UE)<br />
Honors, Fellowships, Other Recognitions<br />
Affiliate status for the Graduate Faculty at Georgia Southern University. 2010.<br />
Clare Booth Luce Fellowship in Science and Engineering ($25,000/yr). 1992-1994.<br />
Outstanding Teaching Performance, Dept. of Biology, Texas A&M University. 1992.<br />
Alpha Lambda Delta, Honor Society. 1985.<br />
Wittenberg University Scholar (top 5% of incoming freshman). 1985-1989.<br />
Reviewed Book Chapter, Peer Reviewed Publications, and Professional Reports<br />
S. Morreale, P. Plotkin, D. Shaver, and H. Kalb. 2007. Adult Migration and Habitat Utilization. In<br />
Biology and Conservation of Ridley Turtles. P. Plotkin and S. Morreale, Editors. John Hopkins<br />
University Press.<br />
G. Zug, H. Kalb, S. Lazar. 1997. Age and growth in wild Kemp’s ridley sea turtles (Lepidochelys<br />
kempii) from skeletochronological data. Biological Conservation 80: 261-268.<br />
H. Kalb and G. Zug. 1990. Age estimates for a population of American toads, Bufo americanus<br />
(Salientia: Bufonidae), in northern Virginia. Brimleyana 16:79-86.<br />
H. Kalb. 1992. Reproductive Physiology of the Spotted Turtle, Clemmys guttata at the Prairie Road<br />
Fen, in Springfield Ohio. Final Report to the Ohio Dept. of Natural Resources.<br />
H. Kalb and L. Laux. 1990. A Population Study of Clemmys guttata at the Prairie Road Fen, in<br />
Springfield Ohio. Final Report to the Ohio Dept. of Natural Resources.<br />
Editor<br />
H. Kalb, A. Rohde*, K. Gayheart, and K. Shanker. 2008. Proceedings of the 25th Annual<br />
Symposium on Sea Turtle Biology and Conservation. NOAA Tech. Memo.<br />
H. Kalb and T. Wibbels. 2000. Proceedings of the 19th Annual Symposium on Sea Turtle Biology<br />
and Conservation. NOAA Tech. Memo. NMFS-SEFSC-443. pp. 291.<br />
H. Kalb. Turtle and Tortoise Newsletter. 2000-2005. Official newsletter of the IUCN/SSC<br />
Freshwater Turtle and Tortoise Specialist Group. Published by Chelonian Research Foundation.<br />
1 st Edition, Feb. 2000. Editorial Board: Anders Rhodin, Peter C.H. Pritchard, John Behler, and<br />
Russell Mittermeier.<br />
H. Kalb. 1994-1999. Box Turtle Research and Conservation Newsletter. Initial funding: Savannah<br />
River Ecology Lab, later: Chelonian Research Foundation.<br />
*Names in bold are undergraduate students who were working directly with me.<br />
2
Grants and Internships<br />
Faculty Development Grant for travel to the Society for Integrative and Comparative Biology’s<br />
Annual Symposium in Seattle, WA. $1024. January 2010.<br />
The Effects of Incubation Temperatures on Embryonic Development and Post-hatching Morphology<br />
and Behavior in Malayan Box Turtles (Cuora amboinensis). Summer 2006. Funding by<br />
University of Evansville Undergraduate Research Program and Honor’s Program. Students:<br />
Alison Rohde, Katie Aldred, and Natalie Byars.<br />
Management of the Cuora amboinensis Taxon Management Group through Genetic Identification<br />
of Captive Stock. Spring 2006. $1,735. University of Evansville’s Undergraduate Research<br />
Program. Student: Sasha Rohde.<br />
Continued Support for the University of Evansville Captive Breeding Colonies of Asian Turtles.<br />
Spring 2005. $1,359. Univ. of Evansville Alumni Research and Scholarly Activity Fellowship.<br />
Subspecies Identification of Individuals in a Captive Colony of Malayan Box Turtles, Cuora<br />
amboinensis. Summer 2005. Funding Univ. of Evansville Honor’s Program. Student: Andrea<br />
Eyler.<br />
Treatment of Parasites in Malayan Box Turtles, Cuora amboinensis. Summer 2005. $9,810.<br />
Funding Univ. of Evansville Undergraduate Research Program. Student: Bozorgmehr Ouranos.<br />
Ultrasound Technology in the Research Lab and Classroom. 2004. $2,500. U.E. Arts, Research and<br />
Teaching Projects Grant.<br />
Maintenance and Reproduction in Two Asian Turtle Assurance Colonies. Fall 2003, Spring 2004.<br />
Total $5,330. University of Evansville’s Alumni Research and Scholarly Activity Fellowship.<br />
Behavior of the Olive Ridley in the Reproductive Patch. 1993. National Geographic Society. David<br />
Owens grantee.<br />
Reproductive Physiology of Clemmys guttata (Spotted Turtle). Grantee and Primary Investigator.<br />
1991-1992. Ohio Dept. of Natural Resources, Division of Parks and Recreation.<br />
Life History Study of Clemmys guttata. Primary Investigator. 1989.<br />
Ohio Dept. of Natural Resources, Division of Parks and Recreation.<br />
Vertebrate Zoology Internship, Smithsonian Institution, Washington, D.C. 1988. Project:<br />
Skeletochronological Age Estimates on the American Toad and the Kemp’s Ridley Sea Turtle.<br />
Travel Grants (2 @ $500) from Dept. of Biology, Texas A&M Univ. 1995 and 1997. To attend the<br />
77th Annual Meeting of Herpetologists’ League/SSAR and 15 th Annual Sea Turtle Symposium.<br />
Travel Grant ($500) to 2 nd <strong>International</strong> Congress on Chelonian Conservation. From Congress<br />
organizers. 1995.<br />
Research Presentations<br />
T. Hayes and H. Kalb. 2010. Visual Stages of Egg Development in the Malayan box turtle (Cuora<br />
amboinensis) as Observed with Ultrasound Technology. Annual Conference of the Society for<br />
Integrative and Comparative Biology, Seattle WA. (Poster presentation)<br />
N. Byars, A. Rohde, K. Aldred, and H. Kalb. 2007. Effects of Incubation Temperature on<br />
Incubation Duration and Hatchling Fitness in Malayan Box Turtles (Cuora amboinensis). 21 st<br />
National Conference on Undergraduate Research, California. Apr. 12-14. (Poster presentation)<br />
H. Kalb, N. Byars, and K. Aldred. 2006. Observations on Reproduction in a Captive Colony of<br />
Malayan Box Turtles, Cuora amboinensis. 4 th Annual Symposium on Conservation and Biology<br />
of Freshwater Turtles and Tortoises. St. Louis, Mo. Aug 10-13. (Poster presentation)<br />
A. Rohde, B. Ouranos, S. Patton and H. Kalb. 2006. Parasites Present in a Stable Breeding<br />
Population of Malayan Box Turtles (Cuora amboinensis). 4 th Annual Symposium on<br />
Conservation and Biology of Freshwater Turtles and Tortoises. St. Louis, Mo. Aug 10-13.<br />
(Poster presentation)<br />
3
B. Ouranos, C. Bursey, and H. Kalb. 2005. Preliminary Parasite Results in a Captive Colony of<br />
Malayan Box Turtles (Cuora amboinensis). Univ. of Evansville Fall Research Symposium.<br />
(Poster presentation)<br />
A. Eyler, A. Rohde, Y. Hsiang-Jui, and H. Kalb. 2005. Obtaining Blood Samples and Sequence<br />
from Cuora amboinensis: Preliminary Studies for Identifying Subspecies of a Local Collection.<br />
University of Evansville Fall Research Symposium. (Poster presentation)<br />
B. Hart, J. Legout, H. Kalb and M. Davis. 2005. Identification and Antibiotic Sensitivity Testing of<br />
Bacteria Infecting Threatened Asian Freshwater Turtles. 19 th National Conference on<br />
Undergraduate Research, Lexington, VA. Apr. 21-23. (Poster presentation)<br />
H. Kalb. 2005. Preliminary Data on Reproduction in Malayan Box Turtles (Cuora amboinensis). 25 th<br />
Annual Sea Turtle Symposium, Savannah GA. (Oral presentation)<br />
H. Kalb. 2004. Female Reproductive Physiology: a Review, Techniques, and Applications. Turtle<br />
Survival Alliance Symposium, Orlando FL. (Oral presentation)<br />
H. Kalb and D. Owens. 1997. The Olive Ridley Aggregation at Playa Nancite, Costa Rica. 77 th<br />
Annual Meeting of Herpetologists’ League/Society for the Study of Amphibian and Reptiles.<br />
Seattle, WA. (Oral presentation)<br />
H. Kalb and D. Owens. 1995. Internesting Movements and Conservation Concerns of the Olive<br />
Ridley Sea Turtle during the Reproductive Season at Nancite, Costa Rica. <strong>International</strong> Congress<br />
of Chelonian Conservation, Gonfaron, France. (Oral presentation)<br />
H. Kalb and D. Owens. 1995. Nancite to Ostional: Post-mating Movements of Female Olive Ridley<br />
Sea Turtles. 15th Annual Sea Turtle Symposium, Hilton Head, SC. (Oral presentation)<br />
H. Kalb, J. Kureen, P. Mayor, J. Peskin, and R. Phyliky. 1995. Conservation Concerns for the<br />
Nancite Olive Ridleys. 15 th Annual Sea Turtle Symposium, Hilton Head, SC. NOAA Tech.<br />
Memo. NMFS-SEFSC-387:141-142. (Poster presentation)<br />
H. Kalb. 1994. Reproductive Evaluation of Clemmys guttata, the Spotted Turtle. 74 th Annual<br />
Meeting of Herpetologists’ League/Society for the Study of Amphibian and Reptiles. Athens,<br />
GA. (Oral presentation)<br />
H. Kalb and D. Owens. 1994. Differences between Solitary and Arribada Nesting Olive Ridley<br />
Females during the Internesting Period. 14 th Annual Sea Turtle Symposium, Hilton Head, SC.<br />
NOAA Tech. Memo. NMFS-SEFSC-351:68. (Oral presentation)<br />
H. Kalb and D. Owens.1993. Arribada versus Solitary Nesters: Comparison of Internesting<br />
Behavior Patterns in Olive Ridley Sea Turtles, Lepidochelys olivacea. 2 nd World Congress of<br />
Herpetology, Adelaide, Australia. (Oral presentation)<br />
H. Kalb, R. Valverde, and D. Owens. 1992. What is the Reproductive Patch of the Olive Ridley Sea<br />
Turtle at Playa Nancite, Costa Rica? 12 th Annual Sea Turtle Symposium, Jekyll Island, GA.<br />
NOAA Tech. Memo. NMFS-SEFSC-361:57-60. (Oral presentation)<br />
L. Laux, C. Stanton, and H. Kalb. 1991. Spatial Relationship among a Restricted Spotted Turtle,<br />
Clemmys guttata, Population in a Small Prairie Fen. Ecological Society of America Annual<br />
Meeting, San Antonio, TX. (Poster presentation)<br />
D. Rostal, H. Kalb, J. Grumbles, P. Plotkin, and D. Owens. 1991. Application of Ultrasonography to<br />
Sea Turtle Reproduction. 11 th Annual Sea Turtle Symposium, Jekyll Island, GA. NOAA Tech.<br />
Memo. NMFS-SEFSC-302:181. (Poster presentation)<br />
H. Kalb. 1990. A Population Study of Clemmys guttata at the Prairie Road Fen, in Springfield,<br />
Ohio. 70 th Annual Meeting of Herpetologists’ League/SSAR, New Orleans, LA. (Poster<br />
presentation)<br />
G. Zug and H. Kalb. 1989. Skeletochronological Age Estimates for Juvenile Lepidochelys kempii<br />
from Atlantic Coast of North America. 9 th Annual Sea Turtle Symposium Jekyll Island, GA.<br />
NOAA Tech. Memo. NMFS-SEFSC-232:271-273. (Poster presentation)<br />
4
Invited Speaker<br />
The Asian Turtle Crisis and the UE Turtle Program. 2005. Hoosier Herpetological Society.<br />
Indianapolis, IN.<br />
The Asian Turtle Crisis. 2004. University of Southern Indiana, Dept. Biology Spring Seminar Series.<br />
Turtles. 2003. Women in Technology and Science Workshop for 6 th grade girls. Jonesboro, AR.<br />
Olive Ridley Sea Turtles of Playa Nancite, Costa Rica. 2001. Richard Stockton College Biology<br />
Seminar Series<br />
Olive Ridley Sea Turtles of Playa Nancite, Costa Rica. 2001. New York Turtle & Tortoise Seminar.<br />
Box Turtles in the Pet Trade and Sea Turtle Natural History and Conservation. 1997. Dept. of<br />
Recreational Sports, 1 st Annual Jamboree, T.A.M.U, College Station, TX.<br />
Box Turtles: Past, Present, and Future. 1997. Rio Brazos Audubon Society.<br />
Costa Rican Turtle Research & Conservation Implications. 1995. Society for Conservation Biology.<br />
Sea Turtle Natural History & Conservation. 1991. Texas Environmental Action Coalition<br />
Conference.<br />
Scientific American Frontiers TV show. 1991. 10 min. on my chemosensory imprinting research in<br />
Kemp’s ridleys.<br />
University Service Work<br />
University of Evansville, committee work<br />
Interdisciplinary Studies Committee, Chair<br />
Admissions and Standards Committee<br />
Biology library liaison<br />
Phi Eta Sigma (freshman honor’s society) advisor<br />
University of Evansville Turtle Program: Tours and Turtle Conservation Talks<br />
Fall, 2004 & 2005: Scott School.<br />
Spring, 2005: Memorial High School Environmental Science Classes. Contact Maryann Watson.<br />
Spring, 2004 & 2005: Luce Elementary School. Contact Pat Keith.<br />
University of Evansville Turtle Program Open House and Fund Raiser.<br />
Homecoming Weekend, 2004, 2005, 2006<br />
Biology representative. United Way Drive, Fall 2004.<br />
Community Service Work<br />
Babysat for parents attending English as a second language classes. Statesboro, GA 2009.<br />
Indiana Master Naturalist Program, Wesselman Woods Nature Preserve. Lecture on Reptile biology<br />
and conservation. Feb., 2007 & 2008.<br />
UE Turtle Ambassadors, Maryann Watson’s Environmental Science Class at Memorial High School.<br />
Six hatchling Malayan box turtles on loan. 2004 - 2008.<br />
Summer Turtle Classes with E-Academy.<br />
7 th -12 th graders at U.E. Two weeks, June 2005.<br />
1 st -3 rd graders at Evansville Day School. Two weeks, June 2004<br />
Turtle display at Reptile Invasion. Wesselman Woods Nature Preserve.<br />
June 14-15, 2008.<br />
June 11-12, 2005 with students Bo and Gollsheed Ouranos.<br />
June 12-13, 2004 with students Josh Yeager and Amanda Nelson.<br />
Turtle presentations at public schools.<br />
2008. Mater Dei summer program. Contact Shelly McFall.<br />
5
2006. Holy Redeemer Day Camp. Contact Becky Smither.<br />
2006. Chandler Elementary School. 5 th grade honor’s class. Contact Mrs. Strahle.<br />
2006. Evansville Day School. Middle School. Contact Mary Ann Kraft<br />
2004: 1 presentation. 3rd graders. Contact Lisa Oldham.<br />
2003: 3 presentations. Parkview Jr. High School (Lawrenceville, IL). Contact Joan Brian.<br />
2003: 3 presentations. Sharon Elementary School. 1st & 6th grade. Contact Carol Stalker.<br />
2003: 2 presentations. Evansville Day School. 1st graders. Contact Brooke Fuchs.<br />
Professional Society Memberships<br />
IUCN/SSC Tortoise and Freshwater Turtle Specialist Group<br />
Society for Integrative and Comparative Biology<br />
Turtle Survival Alliance<br />
Professional Volunteer Work<br />
Volunteer Coordinator. 25 th Annual Sea Turtle Symposium, Philadelphia, PA. 2005.<br />
Judge (student competition). 22nd Annual Sea Turtle Symposium, Miami, FL. 2002.<br />
Volunteer Chairman. 21 th Annual Sea Turtle Symposium, Philadelphia, PA. 2001.<br />
Program Co-chair. 19 th Annual Sea Turtle Symposium, South Padre Island, TX. 1999.<br />
Session Chair. 18 th and 19 th Annual Sea Turtle Symposium, 1998, 1999.<br />
Special Course Work and Skills<br />
Reading Roundtables, Georgia Southern University, Center for Teaching Excellence. 2009.<br />
Techniques in Molecular Biology. Biotechnology Institute, Penn. State University. 2002.<br />
College Teaching (graduate level course), Dept. of Education, Texas A&M University. 1996.<br />
Teacher Enhancement Workshops for Biology Lab Instructors. Dept. Biol., TAMU. 1989.<br />
Marine Biology and Oceanography Exchange Program, Duke Marine Lab, Beaufort, NC. 1988.<br />
6
Molecular Ecology (2007) 16,<br />
<strong>17</strong><strong>–</strong>18 doi: 10.1111/j.1365-294X.2006.03252.x<br />
Blackwell Publishing Ltd<br />
NEWS AND VIEWS<br />
PERSPECTIVE ARTICLE<br />
Whose turtles are they, anyway?<br />
JEANNE A. MORTIMER, *† PETER A. MEYLAN‡<br />
and MARYDELE DONNELLY§<br />
* Department of Zoology, University of Florida, Gainesville, Florida 32611, USA, † Island Conservation Society, Victoria, Mahe,<br />
Seychelles, ‡ Natural Sciences, Eckerd College, St. Petersburg, Florida 33711, USA, § Caribbean Conservation Corporation, Gainesville,<br />
Florida 32609, USA<br />
Abstract<br />
The hawksbill turtle ( Eretmochelys imbricata),<br />
listed since 1996 by the IUCN as Critically<br />
Endangered and by the Convention on <strong>International</strong> Trade in Endangered Species (CITES)<br />
as an Appendix I species, has been the subject of attention and controversy during the past<br />
10 years due to the efforts of some nations to re-open banned international trade. The most<br />
recent debate has centred on whether it is appropriate for Cuba to harvest hawksbills from<br />
shared foraging aggregations within her national waters. In this issue of Molecular Ecology,<br />
Bowen et al.<br />
have used molecular genetic data to show that such harvests are likely to have<br />
deleterious effects on the health of hawksbill populations throughout the Caribbean.<br />
Hawksbill trade involves ‘tortoiseshell’, the translucent<br />
scales covering the hawksbill plastron and carapace, which<br />
has been considered a precious material on par with ivory<br />
and rhinoceros horn since antiquity (according to legend,<br />
Cleopatra bathed in a tub made of tortoiseshell). Efforts<br />
to scientifically manage this resource were, in the past,<br />
hobbled by ignorance about key aspects of the hawksbill’s<br />
life cycle, most notably delayed sexual maturity (20<strong>–</strong><br />
30 years in the Caribbean and longer elsewhere) and the<br />
migratory nature of the species. Flipper tags, used in the<br />
context of much-reduced hawksbill populations and<br />
widely dispersed foraging habitats, revealed little about<br />
hawksbill migrations. Molecular genetics, however, have<br />
provided the technological breakthrough needed (Bass<br />
et al.<br />
1996). The study of Bowen et al.<br />
(2006) expands on that<br />
earlier work adding significantly to our understanding of<br />
the migrations of immature hawksbills. The authors use<br />
proven molecular techniques to examine the nesting beach<br />
origins of hawksbills on eight foraging grounds, four from<br />
previous literature and four newly sampled. The feeding<br />
grounds studied span the tropical West Atlantic from<br />
southern Texas to the US Virgin Islands, and include<br />
Inagua in the Bahamas and a site on the south coast of the<br />
Dominican Republic.<br />
The authors report that populations from 8 of 10 Caribbean<br />
nesting beaches studied are sufficiently distinct that<br />
their contribution to feeding aggregations can be estimated.<br />
The feeding aggregations show less genetic differentiation<br />
than nesting beaches but still exhibit significant differences.<br />
© 2006 The Authors<br />
Journal compilation © 2006 Blackwell Publishing Ltd<br />
Both maximum-likelihood and Bayesian mixed-stock analyses<br />
are used and yield similar results. Each of the feeding<br />
grounds (except Texas, which is the only sample consisting<br />
of pelagic-phase turtles) has significant contributions from<br />
multiple nesting beaches. Further analysis of the Bayesian<br />
results reveals strong relationships between feeding grounds<br />
and nesting beach populations: a direct relationship with<br />
nesting population size and an inverse relationship with<br />
distance between nesting and feeding sites. The management<br />
question that these authors ultimately address is how<br />
harvest of this species in one nation might affect populations<br />
elsewhere. Their mixed-stock analysis strongly corroborates<br />
evidence from tag returns, satellite tracking and<br />
previous genetic study, that harvests in any part of the<br />
Caribbean impact the species throughout the region.<br />
These results have important implications for conservation<br />
and for CITES. During 1970<strong>–</strong>92, Japan imported<br />
∼405<br />
<strong>17</strong>8 kg of ‘bekko’ (tortoiseshell) from 25 countries in<br />
Atlantic Latin America and the Caribbean (Milliken &<br />
Tokunaga 1987; Japanese Trade Statistics). This is equivalent<br />
to some 302 371 Caribbean hawksbill turtles (1.34 kg/<br />
turtle). By the mid-1980s many countries had acceded to<br />
CITES and stopped exporting shell, but Japan took a CITES<br />
reservation or exception on hawksbills, and continued to<br />
import shell until 1992 when international pressure forced<br />
her to drop the reservation and stop importing bekko. During<br />
those 22 years, Cuba harvested about 5000 turtles<br />
annually on their foraging grounds, contributing ∼33%<br />
of<br />
the total shell imports from the region (Carrillo et al.<br />
1999).
18 NEWS AND VIEWS<br />
By 1995, in response to the Japanese cessation of trade,<br />
Cuba reduced its annual official harvest to less than 500<br />
large turtles (Carrillo et al.<br />
1999), a level that continues to<br />
the present day. Since 1993, Cuba has stockpiled the shell<br />
from the harvest in the hope of eventually being able to sell<br />
it to Japan.<br />
In 1997 and 2000, Cuba’s interest in re-opening trade for<br />
‘Cuban hawksbills’ was presented to the biennial meeting<br />
of CITES. Cuba initially argued that hawksbills found in<br />
its waters were nonmigratory, part of a closed system and<br />
sought the rights to harvest 500 animals a year into perpetuity<br />
for the international trade and to sell off the stockpile<br />
of raw shell accumulated since 1993. Although these proposals<br />
were rejected, Cuba maintained an annual harvest<br />
of 500 or fewer large turtles. In early 2000 the stockpile was<br />
6900 kg (Cuban CITES proposal, 2000); since that time it<br />
has been increased by at ∼450<br />
turtles a year and is now<br />
estimated at ∼11<br />
200 kg of shell, representing some 7079<br />
Cuban-harvested turtles (1.59 kg/turtle).<br />
Meanwhile, Caribbean hawksbill populations are<br />
seriously depleted from their historic levels, largely as a<br />
result of centuries of trade; many populations in the region<br />
continue to decline (Mortimer & Donnelly, in press). From<br />
the 1970s onward, increasing numbers of Caribbean countries<br />
have enacted legislation that protects hawksbills on<br />
their beaches and in their waters. At some sites protective<br />
measures appear to have halted the decline of depleted populations<br />
and nesting numbers have stabilized. Remarkably,<br />
since the early 1990s, several such nesting populations<br />
increased significantly. These increases coincide with the<br />
90% decrease in the Cuban hawksbill fishery since 1994, a<br />
period during which more than 55 000 large hawksbills<br />
were spared from slaughter. This is significant given that<br />
fewer than 5000 female hawksbills are estimated to nest<br />
annually in the Caribbean region (Meylan & Donnelly<br />
1999; Mortimer & Donnelly, in press). The implications of<br />
the paper by Bowen et al.<br />
are that harvest of hawksbills<br />
from any nesting beach could impact multiple foreign<br />
feeding grounds, and that harvest on any feeding grounds<br />
could impact the nesting beach populations at multiple<br />
sites. Thus, any nation that opens or promotes harvest<br />
could be undermining badly needed efforts to conserve the<br />
species at other sites.<br />
There is more at stake than turtles. Hawksbills are spongivores<br />
(Leon & Bjorndal 2002) and spongivory helps to<br />
maintain healthy coral reefs by releasing corals from competition<br />
with sponges (Bjorndal & Jackson 2003). Historic<br />
consumption of sponges by hawksbill turtles is estimated<br />
to have been 800 times higher than it is now (McClenachan<br />
et al.<br />
2006). At a time when coral reefs are among the most<br />
endangered ecosystems on the planet (Wilkinson 2000),<br />
successful conservation and management of hawksbills<br />
may be an essential component for ecosystem restoration.<br />
References<br />
Bass AL, Good DA, Bjorndal KA et al. (1996) Testing models of<br />
female reproductive migratory behavior and population structure<br />
in the Caribbean hawksbill turtle, Eretmochelys imbricata,<br />
with mtDNA sequences. Molecular Ecology,<br />
5,<br />
321<strong>–</strong>328.<br />
Bjorndal KA, Jackson JBC (2003) Role of sea turtles in marine ecosystems<br />
— reconstructing the past. In: Biology of Sea Turtles (eds<br />
Lutz PL, Musick JA, Wyneken J), Vol. II, pp. 259<strong>–</strong>273. CRC Press,<br />
Boca Raton, Florida.<br />
Bowen BW, Grant WS, Hillis-Starr Z et al. (2006) Mixed-stock<br />
analysis reveals the migrations of juvenile hawksbill turtles<br />
( Eretmochelys imbricata)<br />
in the Caribbean Sea. Molecular Ecology,<br />
doi:10.1111/j.1365-294X.2006.03096.x.<br />
Carrillo CE, Webb GJW, Manolis SC (1999) Hawksbill turtles<br />
( Eretmochelys imbricata)<br />
in Cuba: an assessment of the historical<br />
harvest and its impacts. Chelonian Conservation and Biology,<br />
3,<br />
264<strong>–</strong>280.<br />
Leon YM, Bjorndal KA (2002) Selective feeding in the hawksbill<br />
turtle, an important predator in coral reef ecosystems. Marine<br />
Ecology Progress Series,<br />
245,<br />
249<strong>–</strong>258.<br />
McClenachan L, Jackson JBC, Newman MJH (2006) Conservation<br />
implications of historic sea turtle nesting beach loss. Front Ecological<br />
Environment,<br />
4,<br />
290<strong>–</strong>296.<br />
Meylan AB, Donnelly M (1999) Status justification for listing the<br />
hawksbill turtle ( Eretmochelys imbricata)<br />
as critically endangered<br />
on the 1996 IUCN Red List of Threatened Animals. Chelonian<br />
Conservation and Biology,<br />
3,<br />
<strong>17</strong>7<strong>–</strong>184.<br />
Milliken T, Tokunaga H (1987) The Japanese sea turtle trade<br />
1970<strong>–</strong>86. A Special Report Prepared by TRAFFIC (Japan) . Center<br />
for Environmental Education, Washington D.C.<br />
Mortimer JA, Donnelly M (in press)). IUCN 2006 Global Assessment<br />
for the Hawksbill Turtle (Eretmochelys imbricata) .<br />
Wilkinson CR (2000) Status of Coral Reefs of the World: 2000.<br />
Global<br />
Coral Reef Monitoring Network.<br />
Australian Institute of Marine<br />
Science.<br />
© 2006 The Authors<br />
Journal compilation © 2006 Blackwell Publishing Ltd
Molecular Ecology (2009) 18, 4841<strong>–</strong>4853 doi: 10.1111/j.1365-294X.2009.04403.x<br />
Turtle groups or turtle soup: dispersal patterns of<br />
hawksbill turtles in the Caribbean<br />
J. M. BLUMENTHAL,*† F. A. ABREU-GROBOIS,‡ T. J. AUSTIN,* A. C. BRODERICK,†<br />
M. W. BRUFORD,§ M. S. COYNE,†<strong>–</strong> G. EBANKS-PETRIE,* A. FORMIA,§ P. A. MEYLAN,**<br />
A. B. MEYLAN†† and B. J. GODLEY†<br />
*Department of Environment, Box 486, Grand Cayman KY1-1106, Cayman Islands, †Centre for Ecology and Conservation,<br />
School of Biosciences, University of Exeter Cornwall Campus, Penryn TR10 9EZ, UK, ‡Laboratorio de Genética, Unidad<br />
Académica Mazatlán, Instituto de Ciencias del Mar y Limnología, Mazatlán, Sinaloa 82040, México, §Biodiversity and<br />
Ecological Processes Research Group, School of Biosciences, Cardiff University, Cardiff CF10 3TL, UK, <strong>–</strong>SEATURTLE.ORG,<br />
Durham, NC 27705, USA, **Natural Sciences, Eckerd College, St Petersburg, FL 33711, USA, ††Florida Fish and Wildlife<br />
Conservation Commission, Fish and Wildlife Research Institute, St Petersburg, FL 33701, USA<br />
Introduction<br />
Abstract<br />
Despite intense interest in conservation of marine turtles, spatial ecology during the<br />
oceanic juvenile phase remains relatively unknown. Here, we used mixed stock analysis<br />
and examination of oceanic drift to elucidate movements of hawksbill turtles (Eretmochelys<br />
imbricata) and address management implications within the Caribbean. Among<br />
samples collected from 92 neritic juvenile hawksbills in the Cayman Islands we detected<br />
11 mtDNA control region haplotypes. To estimate contributions to the aggregation, we<br />
performed ‘many-to-many’ mixed stock analysis, incorporating published hawksbill<br />
genetic and population data. The Cayman Islands aggregation represents a diverse mixed<br />
stock: potentially contributing source rookeries spanned the Caribbean basin, delineating<br />
a scale of recruitment of 200<strong>–</strong>2500 km. As hawksbills undergo an extended phase of<br />
oceanic dispersal, ocean currents may drive patterns of genetic diversity observed on<br />
foraging aggregations. Therefore, using high-resolution Aviso ocean current data, we<br />
modelled movement of particles representing passively drifting oceanic juvenile<br />
hawksbills. Putative distribution patterns varied markedly by origin: particles from<br />
many rookeries were broadly distributed across the region, while others would appear to<br />
become entrained in local gyres. Overall, we detected a significant correlation between<br />
genetic profiles of foraging aggregations and patterns of particle distribution produced<br />
by a hatchling drift model (Mantel test, r = 0.77, P < 0.001; linear regression, r = 0.83,<br />
P < 0.001). Our results indicate that although there is a high degree of mixing across the<br />
Caribbean (a ‘turtle soup’), current patterns play a substantial role in determining genetic<br />
structure of foraging aggregations (forming turtle groups). Thus, for marine turtles and<br />
other widely distributed marine species, integration of genetic and oceanographic data<br />
may enhance understanding of population connectivity and management requirements.<br />
Keywords: conservation genetics, hawksbill, marine turtle, migratory species, mixed stock,<br />
ocean currents<br />
Received 29 June 2009; revision received 21 September 2009; accepted 25 September 2009<br />
Many marine vertebrates have life cycles that span wide<br />
spatiotemporal scales <strong>–</strong> complicating research and<br />
Correspondence: Brendan J. Godley, Fax: 44 (0) 1326 253638;<br />
E-mail: b.j.godley@exeter.ac.uk<br />
Ó 2009 Blackwell Publishing Ltd<br />
management. Recently, molecular techniques have provided<br />
insights into patterns of migration and stock resolution<br />
in species ranging from fish (Millar 1987) to<br />
porpoises (Andersen et al. 2001) and great whales<br />
(Baker et al. 1999; Witteveen et al. 2004). Such stock resolution<br />
issues are particularly important when species
4842 J. M. BLUMENTHAL ET AL.<br />
are commercially valuable, cross geopolitical boundaries<br />
or have been exploited to the point of endangerment.<br />
Marine turtles exemplify these concerns: a long history<br />
of commercial use has resulted in global declines<br />
(Baillie et al. 2004) and management of marine turtle<br />
populations is complex, as developmental and reproductive<br />
migrations may span the territorial waters of<br />
several nations and the high seas (Bowen et al. 1995;<br />
Bolten et al. 1998).<br />
Like many other marine organisms (Mora & Sale<br />
2002) most marine turtle species experience a phase of<br />
oceanic dispersal, which in marine turtles is known as<br />
the ‘lost year’ or ‘lost years’ (Carr 1987). This extended<br />
period of drifting in concert with initially small size<br />
and high mortality inhibits tracking movements of neonates<br />
from the time of entry into marine habitat off<br />
natal beaches to the time of recruitment into neritic foraging<br />
grounds (Bolten et al. 1998; Lahanas et al. 1998;<br />
Engstrom et al. 2002; Bowen et al. 2004; Luke et al.<br />
2004). Although foraging grounds are typically inhabited<br />
by individuals originating from multiple nesting<br />
beaches a behavioural barrier to genetic mixing at nesting<br />
beaches is maintained: generations of mature<br />
females return to their natal regions to reproduce (Meylan<br />
et al. 1990). This ‘natal homing’ leads to genetic differentiation<br />
at mitochondrial loci of nesting populations<br />
over time (hawksbills Eretmochelys imbricata: Broderick<br />
et al. 1994; Bass et al. 1996; loggerheads Caretta caretta:<br />
Encalada et al. 1998; Engstrom et al. 2002; green turtles<br />
Chelonia mydas: Meylan et al. 1990; Allard et al. 1994).<br />
Therefore, mitochondrial DNA haplotypes and haplotype<br />
frequencies characteristic of each region serve as<br />
a form of ‘genetic tag,’ linking juveniles in mixed<br />
aggregations with their specific nesting beach origins<br />
(Norman et al. 1994).<br />
In a method originally developed to resolve the origins<br />
of salmon stocks (Millar 1987), estimates of marine<br />
turtle origin can be made using a maximum likelihood<br />
(Pella & Milner 1987) or Bayesian (Pella & Masuda<br />
2001) mixed stock analysis (MSA), where haplotype frequencies<br />
for genetically mixed aggregations are compared<br />
with potential source populations. MSA has<br />
recently proven valuable in elucidating migratory patterns<br />
and range states in hawksbill (Bowen et al. 1996,<br />
2007a; Bass 1999; Diaz-Fernandez et al. 1999; Troëng<br />
et al. 2005; Velez-Zuazo et al. 2008), loggerhead (Bowen<br />
et al. 1995; Laurent et al. 1998; Engstrom et al. 2002;<br />
Maffucci et al. 2006) and green turtles (Lahanas et al.<br />
1998; Luke et al. 2004). While mixed stock analysis has<br />
traditionally been used to estimate contributions of<br />
many potential source rookeries to a single foraging<br />
ground (‘many-to-one’ analysis), a new approach has<br />
recently been developed which more realistically estimates<br />
the contributions of many rookeries to many for-<br />
aging grounds within a metapopulation context (‘manyto-many’<br />
analysis; Bolker et al. 2007). Caribbean hawksbills<br />
represent ideal candidates for many-to-many<br />
analysis: rookeries are sufficiently differentiated (Bass<br />
et al. 1996), populations may be relatively contained in<br />
the Caribbean region (Bowen et al. 1996) and data from<br />
multiple mixed stocks have recently been published<br />
(Bowen et al. 2007a; Velez-Zuazo et al. 2008).<br />
Tagging and satellite tracking have demonstrated<br />
that hawksbill developmental and reproductive migrations<br />
span the territorial waters of multiple jurisdictions<br />
(Meylan 1999; Horrocks et al. 2001; Troëng et al.<br />
2005; Whiting & Koch 2006; van Dam et al. 2008) and<br />
genetic research is beginning to elucidate links<br />
between nesting beaches and foraging grounds (Bowen<br />
et al. 1996, 2007a; Bass 1999; Diaz-Fernandez et al.<br />
1999; Troëng et al. 2005; Velez-Zuazo et al. 2008). It<br />
has also been determined that Caribbean hawksbill<br />
foraging aggregations show shallow but significant<br />
genetic structure (Bowen et al. 2007a). As in green turtles<br />
(Bass et al. 2006), this suggests that rather than the<br />
oceanic phase being a homogeneous mixture (the ‘turtle<br />
soup’ model: Engstrom et al. 2002), dispersal is<br />
non-random. However for many populations links<br />
between nesting beaches and foraging areas have not<br />
been identified (Bowen et al. 1996) and little is known<br />
regarding movements of oceanic juveniles during the<br />
lost year or years (Musick & Limpus 1997; Bolten<br />
2003) and factors which drive dispersal during this<br />
phase (Bowen et al. 2007a).<br />
In previous genetic studies of green (Lahanas et al.<br />
1998; Bass & Witzell 2000; Luke et al. 2004) and loggerhead<br />
turtles (Engstrom et al. 2002; Reece et al. 2006)<br />
source population size and geographic distance have<br />
proven inconsistent in determining recruitment from<br />
rookeries to foraging grounds. For Caribbean hawksbills,<br />
these factors were significantly correlated with foraging<br />
aggregation genetic composition, but correlations<br />
were weak and significant only for a small proportion<br />
of individual foraging grounds (Bowen et al. 2007a).<br />
Ocean currents have been postulated as playing a major<br />
role in distribution of oceanic juvenile marine turtles<br />
(Carr 1987) and recent studies have linked genetic composition<br />
of marine turtle foraging aggregations to ocean<br />
current patterns (Luke et al. 2004; Okuyama & Bolker<br />
2005; Bass et al. 2006; Carreras et al. 2006). However,<br />
for hawksbills, attempts to account for the role of ocean<br />
currents through incorporation of a current correction<br />
factor (increasing or decreasing geographic distances in<br />
a regression model by 10<strong>–</strong>20%) did not increase the<br />
strength of correlations (Bowen et al. 2007a).<br />
To evaluate population connectivity for marine organisms,<br />
empirical data must be linked with biophysical<br />
modelling of oceanic dispersal (Botsford et al. 2009).<br />
Ó 2009 Blackwell Publishing Ltd
Generalized ocean current diagrams can aid in consideration<br />
of the role of ocean currents in marine turtle<br />
dispersal (Bass et al. 2006; Carreras et al. 2006; Bowen<br />
et al. 2007a) but this method falls short when current<br />
patterns are complex. Trans-Atlantic drift has been<br />
modelled for oceanic juvenile loggerheads (Hays &<br />
Marsh 1997) and the availability of high-resolution<br />
ocean current data for the Caribbean and other regions<br />
now opens up the possibility of modelling dispersal of<br />
oceanic juvenile hawksbills in dynamic ocean currents.<br />
In this study, characterization of stock composition at<br />
the Cayman Islands hawksbill foraging site was<br />
extended by development of a hatchling drift model<br />
incorporating source population sizes and regional<br />
ocean current data to answer fundamental questions of<br />
hawksbill ecology: (i) how are oceanic juvenile hawksbills<br />
distributed during the lost year or years? (ii) how<br />
are foraging stocks geographically structured or regionally<br />
mixed? (iii) what role might ocean currents play in<br />
determining these patterns?<br />
Materials and Methods<br />
Study area<br />
The Cayman Islands are located in the Caribbean Sea,<br />
approximately 240 km south of Cuba (Fig. 1). The<br />
three low-lying islands are exposed peaks on the<br />
Cayman Ridge formation, with nearly vertical slopes<br />
extending to depths in excess of 2000 m on all sides<br />
(Roberts 1994). Hawksbill nesting, while described as<br />
abundant in historical records (Lewis 1940), appears to<br />
have been largely extirpated in recent decades (Bell<br />
et al. 2007). However, the islands host foraging aggregations<br />
of neritic juvenile hawksbill turtles, inhabiting<br />
colonized pavement, coral reef and reef wall habitats<br />
(Blumenthal et al. 2009a, b). For this study, two sampling<br />
sites were selected: Bloody Bay, Little Cayman<br />
(19.7°N, 81.1°W) and western Grand Cayman (19.3°N,<br />
81.4°W).<br />
Capture and sampling<br />
Juvenile hawksbills (20<strong>–</strong>60 cm straight carapace length)<br />
were hand-captured in neritic foraging habitat and flipper<br />
and PIT tagged according to standard protocols<br />
(Blumenthal et al. 2009b) to prevent repeated sampling<br />
of individuals. Genetic sampling for this study was<br />
undertaken year-round between 2000 and 2003. Skin<br />
biopsies were obtained from a rear flipper with a sterile<br />
4 mm biopsy punch and preserved in a solution of 20%<br />
DMSO saturated with NaCl (Dutton 1996). Blood samples<br />
were collected from the dorsal cervical sinus<br />
(Owens & Ruiz 1980) and preserved in lysis buffer<br />
Ó 2009 Blackwell Publishing Ltd<br />
HAWKSBILL TURTLE DISPERSAL 4843<br />
Fig. 1 Distribution of hawksbill aggregations studied in the<br />
Caribbean region. Circles represent locations of Caribbean<br />
hawksbill rookeries: filled circles are genetically characterized.<br />
Haplotype frequency data from Antigua, Barbados, Belize,<br />
Costa Rica, Cuba, Mexico, Puerto Rico and the US Virgin<br />
Islands were used in many-to-many mixed stock analysis;<br />
rookeries in Venezuela are presently insufficiently characterized<br />
to permit inclusion. Triangles represent genetically characterized<br />
hawksbill foraging grounds also incorporated in mixed<br />
stock analysis: Bahamas, Cayman, Cuba, Dominican Republic,<br />
Mexico, Puerto Rico, Texas and the US Virgin Islands. Marine<br />
ecoregions (1) Gulf of Mexico (2) western Caribbean (3) southwestern<br />
Caribbean (4) Greater Antilles (5) southern Caribbean<br />
(6) eastern Caribbean (7) Bahamian. Source of published<br />
genetic data: Antigua (Bass 1999), Barbados (Browne et al. in<br />
press), Belize (Bass 1999), Cuba (Diaz-Fernandez et al. 1999),<br />
Mexico (Bass 1999; Diaz-Fernandez et al. 1999), USVI (Bass<br />
1999; Bowen et al. 2007a), Costa Rica (Troëng et al. 2005;<br />
Bowen et al. 2007a) and Puerto Rico (Diaz-Fernandez et al.<br />
1999; Velez-Zuazo et al. 2008).<br />
(100mM Tris-HCl, pH 8, 100mM EDTA, pH 8, 10mM<br />
NaCl and 1<strong>–</strong>2% SDS: Dutton 1996).<br />
Molecular analysis<br />
Following overnight digestion at 55 °C with proteinase<br />
K, DNA extraction was conducted via standard phenol<br />
chloroform extraction (Milligan 1998), Qiagen DNEasy<br />
tissue kit, or a modification of a protocol by Allen et al.<br />
(1998) (Formia et al. 2006, 2007). Fragments of 486 or<br />
550 base pairs (bp) were amplified via polymerase<br />
chain reaction (PCR), using primer pairs TCR6 (Norman<br />
et al. 1994) ⁄ L15926 (Kocher et al. 1989) or LTEi3 ⁄ HDEi1<br />
(LTEi3 5¢-CCTAGAATAATCAAAAGAGAAGG-3¢;<br />
HDEi1 5¢- AGTTTCGTTAATTCGGCAG-3¢; Abreu-Grobois<br />
et al. 2006) respectively. PCR reactions [PCR<br />
Ready-To-Go Beads (Amersham) or 1.5 mM MgCl2, 1·<br />
PCR buffer, 200 lM of each dNTP, 0.5 lM of each<br />
primer, 0.5 U Invitrogen Taq DNA polymerase, 1 lL
4844 J. M. BLUMENTHAL ET AL.<br />
template DNA and sterile H2O to a volume of 10 lL)<br />
were carried out in a Perkin Elmer GeneAmp PCR<br />
system 9700 (3 min at 94 °C, 35 cycles of: 45 s at<br />
94 °C, 30 s at 55 °C and 1.5 min at 72 °C, followed by<br />
72 °C for 10 min; Formia et al. 2006). PCR products<br />
were cleaned with a Geneclean Turbo Kit (Qbiogene)<br />
or QIAquick spin columns (Qiagen), sequenced in<br />
both directions with a BigDye Terminator Cycle<br />
Sequencing Kit v. 2.0 (Applied Biosystems), purified<br />
via ethanol precipitation and run on an Applied Biosystems<br />
model 3100 automated DNA sequencer at<br />
Cardiff University, or analysed at the University of<br />
Florida Sequencing Core on an Amersham Pharmacia<br />
Biotech MegaBACE 1000 capillary array DNA<br />
sequencing unit. Negative controls (template-free PCR<br />
reactions) were utilized throughout to test for contamination.<br />
To identify haplotypes, sequences were aligned with<br />
published hawksbill mtDNA sequences (Bass et al.<br />
1996; Bowen et al. 1996; Bass 1999; Diaz-Fernandez<br />
et al. 1999; Troëng et al. 2005; Bowen et al. 2007a;<br />
Velez-Zuazo et al. 2008; Browne et. al. in press) in Sequencher<br />
2.1.2 (Gene Codes Corp). Haplotype diversity<br />
(h), nucleotide diversity (p, and population pairwise<br />
FST values (10 000 permutations; Kimura 2-parameter<br />
model, a = 0.50) were calculated in Arlequin v. 3.11<br />
(Excoffier et al. 2005).<br />
Mixed stock analysis<br />
Weighted and unweighted many-to-many analyses (Bolker<br />
et al. 2007) were also conducted, incorporating all<br />
previously published Caribbean hawksbill genetic data<br />
(haplotype frequencies: Antigua (Bass 1999), Barbados<br />
(Browne et al. in press; windward and leeward rookeries<br />
combined), Belize (Bass 1999), Cuba (Diaz-Fernandez<br />
et al. 1999; all foraging grounds combined), Mexico<br />
(Bass 1999; Diaz-Fernandez et al. 1999), USVI (Bass<br />
1999; Bowen et al. 2007a), Costa Rica (Troëng et al.<br />
2005; Bowen et al. 2007a) and Puerto Rico (Diaz-Fernandez<br />
et al. 1999; Velez-Zuazo et al. 2008). To enable comparison<br />
with previously published rookery haplotypes,<br />
sequences were truncated to 384 bp for analysis. Source<br />
population size (number of nests) was used as a strict<br />
constraint in the ‘weighted’ many-to-many model<br />
(Bolker et al. 2007), with population sizes taken from<br />
Mortimer & Donnelly (2007).<br />
Hatchling drift model<br />
Movement of particles was modelled using high resolution<br />
ocean current data from Aviso (Archiving, Validation<br />
and Interpretation of Satellite Oceanographic data,<br />
http://www.aviso.oceanobs.com). Aviso geostrophic<br />
velocity vector (GVV) data (maps of absolute geostrophic<br />
velocities derived from maps of Absolute<br />
Dynamic Topography combining all satellites, obtained<br />
from http://www.aviso.oceanobs.com) have a spatial<br />
resolution of 1 ⁄ 3 degree and a temporal resolution of<br />
1 day. To model the movement of passively drifting<br />
hawksbill hatchlings, post-hatchlings and oceanic juveniles,<br />
virtual particles were released from the coordinates<br />
of genetically characterized Caribbean hawksbill<br />
rookeries locations (Fig. 1). To simulate peak hatchling<br />
emergence, particles were released 60 days after the<br />
peak of the nesting season at each location (nesting season<br />
data: Chacón 2004; Garduño-Andrade 1999; Moncada<br />
et al. 1999), with one particle released per day over a<br />
60-day period and the resulting particle profiles<br />
weighted by source population size (taken from Mortimer<br />
& Donnelly 2007). Drift of virtual particles was initiated<br />
approximately 50 km offshore from each rookery to<br />
account for the phase of directed hatchling swimming<br />
that occurs prior to beginning passive drift (Hasbún<br />
2002; Chung et al. 2009a,b) and to bring particles into<br />
the area covered by Aviso data, as nearshore currents<br />
are too complex to map reliably. The u (eastward) and v<br />
(northward) ocean current vector components were sampled<br />
at the location of each particle from the daily GVV<br />
data file using a bilinear interpolation [Generic Mapping<br />
Tools (GMT 4.11; Wessel & Smith 1991)]. The u and v<br />
values (cm ⁄ s) were converted to m ⁄ day and the particle<br />
advected that distance in the x and y direction, respectively,<br />
and a new location obtained for that particle. Particles<br />
that left the GVV field (i.e. pushed towards the<br />
coastline) were removed from the model, as there is no<br />
information on how an oceanic juvenile hawksbill would<br />
behave in this situation. Also, while natural mortality<br />
during the hatchling, post-hatchling and oceanic juvenile<br />
phase is high, this factor was not incorporated into the<br />
model. One location per day was saved for each particle<br />
for analysis. All current modelling was carried out using<br />
custom perl scripts, the geod program, part of the<br />
PROJ.4 Cartographic Projections Library (http://<br />
www.remotesensing.org/proj/) and the GMT package.<br />
Comparison of the hatchling drift model with mixed<br />
stock analysis<br />
To examine the possible influence of ocean currents on<br />
patterns of dispersal, we compared foraging aggregation<br />
genetic profiles (determined via many-to-many<br />
analysis) with patterns of particle distribution (particle<br />
profiles) produced by the hatchling drift model<br />
(Table 1). To produce particle profiles, we used the Global<br />
Maritime Boundaries Database (GMBD) to delineate<br />
Exclusive Economic Zones of nations containing genetically<br />
characterized hawksbill foraging aggregations.<br />
Ó 2009 Blackwell Publishing Ltd
Table 1 Comparison of genetic profiles from many-to-many analysis and particle profiles derived from the hatchling drift model<br />
We then quantified the number of particles from each<br />
source to enter these territorial waters during the modelled<br />
first year of drifting (though it should be noted that<br />
a longer oceanic phase could occur; Musick & Limpus<br />
1997). Euclidean distance matrices were calculated for<br />
particle and genetic profiles (treating each estimate of<br />
proportional contribution as a character) and compared<br />
with a Mantel test (PopTools v. 3.0; Hood 2009). Additionally,<br />
to facilitate comparison with the results of<br />
Bowen et al. (2007a), we investigated the relationship<br />
between genetic and particle profiles with linear regression,<br />
regressing arcsine transformed proportions (Graph-<br />
Pad InStat 3.10, GraphPad Software).<br />
Results<br />
Cay Tex Bah Cuba PR USVI DR Mex<br />
MSA Drift MSA Drift MSA Drift MSA Drift MSA Drift MSA Drift MSA Drift MSA Drift<br />
Anti 6.0 13.1 1.9 0.0 5.2 13.2 2.7 4.3 5.3 18.5 10.2 50.2 4.4 23.1 6.7 3.8<br />
Barb 46.1 40.3 3.3 0.0 22.7 25.9 15.7 13.5 47.4 28.6 27.7 21.2 65.7 35.3 5.8 13.8<br />
Belz 0.7 0.0 0.6 0.0 0.8 0.0 0.9 0.0 1.5 0.0 0.7 0.0 2.1 0.0 0.7 0.6<br />
Cuba 21.9 45.7 2.4 0.0 26.1 0.0 44.3 64.6 8.0 0.0 23.0 0.0 10.1 0.0 5.8 0.0<br />
Mexi 13.2 0.0 85.3 100.0 28.4 46.6 10.5 <strong>17</strong>.1 8.5 0.0 14.1 0.0 7.7 0.0 71.6 81.3<br />
USVI 3.0 0.3 2.0 0.0 3.7 0.7 2.4 0.2 8.1 8.5 2.3 26.4 3.9 4.0 2.5 0.2<br />
CR 0.5 0.6 0.3 0.0 0.5 0.5 0.7 0.3 1.4 0.0 0.4 0.0 1.5 0.0 0.4 0.3<br />
PR 8.7 0.0 4.1 0.0 12.6 13.0 22.9 0.0 19.7 44.4 21.6 2.1 4.7 37.6 6.3 0.0<br />
Rows represent potential source rookeries <strong>–</strong> Antigua, Barbados, Belize, Cuba, Mexico, US Virgin Islands, Costa Rica and Puerto Rico;<br />
columns represent genetically characterized foraging grounds <strong>–</strong> Cayman Islands, Texas, Bahamas, Cuba, Puerto Rico, US Virgin<br />
Islands, Dominican Republic and Mexico (for sources of published genetic data see Fig. 1). Numbers are mean proportional<br />
contribution from each rookery to each foraging ground, estimated via weighted many-to-many mixed stock analysis (MSA) and<br />
proportion of particles from each rookery to enter each foraging ground during the first year of drifting, from the hatchling drift<br />
model (Drift).<br />
Genetic structure<br />
Among 92 neritic juvenile hawksbills from the Cayman<br />
Islands, we detected 11 mitochondrial DNA (mtDNA)<br />
control region haplotypes: Ei-A1, Ei-A2, Ei-A3, Ei-A9c,<br />
Ei-A11c, Ei-A18, Ei-A20, Ei-A24, EiA28, Ei-A29 and a<br />
previously undetected haplotype, Ei-A72 (GenBank<br />
accession number GQ925509). The Cayman Islands foraging<br />
aggregation represented a diverse mixed stock<br />
(h = 0.72 ± 0.04; p = 0.01 ± 0.005). Haplotype frequencies<br />
from Grand Cayman and Little Cayman were not<br />
significantly different (population pairwise FST = )0.01,<br />
P = 0.83). Therefore results for the two study sites were<br />
pooled for subsequent analyses.<br />
Rookeries of origin<br />
Using unweighted and weighted many-to-many analysis,<br />
we estimated the contributions of genetically character-<br />
Ó 2009 Blackwell Publishing Ltd<br />
HAWKSBILL TURTLE DISPERSAL 4845<br />
ized Caribbean rookeries to the Cayman Islands foraging<br />
aggregation. Shrink factors for all chains were
4846 J. M. BLUMENTHAL ET AL.<br />
Fig. 2 Genetic composition at the Cayman Islands foraging<br />
aggregation. Among the 92 individuals sequenced, 11 haplotypes<br />
were detected: Ei-A1 (44); Ei-A2 (2); Ei-A3 (1); Ei-A9c (8);<br />
Ei-A11c (<strong>17</strong>); Ei-A18 (1); Ei-A20 (1); Ei-A24 (11); Ei-A28 (4); Ei-<br />
A29 (1); Ei-A72 (2). (a) Using an unweighted many-to-many<br />
model (white bars), estimated contributions (mean proportion,<br />
2.5% and 97.5% confidence intervals) of genetically characterized<br />
hawksbills rookeries to the Cayman Islands foraging aggregation<br />
were Antigua (21.3%, 3.0<strong>–</strong>41.8%), Barbados (20.5%,<br />
1.1<strong>–</strong>48.1%), Belize (8.7%, 0.4<strong>–</strong>24.9%), Cuba (12.6%, 0.6<strong>–</strong>34.1%),<br />
Mexico (9.1%, 0.8<strong>–</strong>16.0%), USVI (11%, 0.4<strong>–</strong>27.6%), Costa Rica<br />
(10.7%, 2.9<strong>–</strong>21.3%), Puerto Rico (6.1%, 0.8<strong>–</strong>16.6%). When<br />
source population size was used as a constraint (grey bars), estimates<br />
were Antigua (6.0%, 0.7<strong>–</strong>15.5%), Barbados (46.1%, 20.6<strong>–</strong><br />
69.0%), Belize (0.7%, 0.0<strong>–</strong>2.7%), Cuba (21.9%, 3.7<strong>–</strong>43%), Mexico<br />
(13.2%, 7.4<strong>–</strong>20.6%), USVI (3%, 0.1<strong>–</strong>9.4%), Costa Rica (0.5%,<br />
0.0<strong>–</strong>2.0%), Puerto Rico (8.7%, 1.5<strong>–</strong>18.3%). (b) Arrows are scaled<br />
in proportion to estimated mean contribution (weighted manyto-many)<br />
to indicate potential scale of recruitment.<br />
365 days of drifting) and genetic profiles (many-tomany<br />
results for all genetically characterized Caribbean<br />
foraging aggregations) (Fig. 4).<br />
Discussion<br />
Because hawksbill turtles spend the vast majority of<br />
their lives at sea, for many populations links between<br />
nesting beaches and foraging grounds are unknown.<br />
To estimate origins of neritic juvenile hawksbill turtles<br />
foraging in the Cayman Islands, we applied many-tomany<br />
modelling using data from this study and other<br />
published genetic studies of Caribbean hawksbills. For<br />
the Cayman Islands foraging aggregation potentially<br />
contributing source rookeries spanned the Caribbean<br />
basin <strong>–</strong> delineating a scale of recruitment of 200<strong>–</strong><br />
2500 km (straight-line distance) and highlighting the<br />
complex spatial ecology of the species.<br />
We then investigated the role of ocean currents as a<br />
mechanism underlying patterns of dispersal. Bowen<br />
et al. (2007a) previously tested the role of geographic<br />
distance (both straight-line and modified in order to<br />
account for a likely influence of ocean currents) in determining<br />
recruitment of hawksbills from nesting populations<br />
to foraging aggregations. Here, we compared<br />
results of mixed stock analysis to particle profiles, which<br />
take source population size, geographic distance and<br />
oceanographic circulation patterns into account <strong>–</strong> an<br />
important consideration as oceanic juveniles are unlikely<br />
to follow a direct route between nesting beaches and foraging<br />
grounds. A strong correlation was detected<br />
between patterns of stock composition at Caribbean<br />
hawksbill foraging aggregations and patterns of particle<br />
distribution at these sites <strong>–</strong> indicating that ocean currents<br />
likely play a substantial role in determining geographic<br />
patterns of genetic diversity, and that models of<br />
oceanic drift may illustrate movements of oceanic juvenile<br />
turtles during the lost year or years. Putative dispersal<br />
patterns for oceanic juvenile hawksbills varied<br />
regionally: particles from many rookeries were broadly<br />
distributed across the Caribbean (a ‘turtle soup’), while<br />
particles from other rookeries appeared to be regionally<br />
constrained (forming ‘turtle groups’). Thus, because of<br />
the complexities of regional and local current patterns,<br />
some areas will be more diverse (turtles recruiting from<br />
multiple jurisdictions) while others will experience<br />
greater levels of proximate or local recruitment.<br />
Ultimately, population resolution may allow demographic<br />
parameters such as abundance, survival, sex<br />
ratio and growth to be linked across broad spatiotemporal<br />
scales and incorporated into population modelling<br />
(Bowen et al. 2004; Reece et al. 2006). However, at present,<br />
the accuracy of mixed stock estimates is limited by<br />
insufficient baseline data from nesting beaches (Troëng<br />
et al. 2005). Indeed, detection of a previously unknown<br />
haplotype at the Cayman Islands foraging aggregation<br />
indicates that nesting beaches in the Caribbean or source<br />
rookeries farther afield have been incompletely or insufficiently<br />
sampled. Longer periods of sampling and larger<br />
sample sizes (Reece et al. 2006), standardized use of<br />
primers which amplify larger mtDNA fragments (Abreu-<br />
Grobois et al. 2006), and expanded geographic coverage<br />
Ó 2009 Blackwell Publishing Ltd
HAWKSBILL TURTLE DISPERSAL 4847<br />
Fig. 3 Trajectories of modelled particles representing drifting oceanic stage juvenile turtles, released from hawksbill index beaches<br />
in (a) Gulf of Mexico (Isla Holbox, Punta Xen and Las Coloradas Mexico) (b) Greater Antilles (Doce Leguas Cuba, Mona Island<br />
Puerto Rico) (c) eastern Caribbean (Long Beach Antigua, Hilton Beach Barbados, Buck Island USVI) (d) southern Caribbean (Archipiélago<br />
Los Roques and Península de Paria Venezuela) (e) southwestern Caribbean (Tortuguero Costa Rica) (f) western Caribbean<br />
(Gales Point Belize) (g) depicts overall particle trajectory patterns in the wider Caribbean.<br />
Ó 2009 Blackwell Publishing Ltd
4848 J. M. BLUMENTHAL ET AL.<br />
Fig. 4 Comparison of hatchling drift (black bars) with weighted (grey bars) and unweighted (white bars) many-to-many results for<br />
the Cayman Islands and other genetically characterized foraging grounds (for sources of published genetic data see Fig. 1). Data are<br />
mean proportion (±2.5<strong>–</strong>97.5% confidence intervals for many-to-many analysis).<br />
(Engstrom et al. 2002) are priorities for improving estimates.<br />
Ocean current modelling may serve to identify<br />
potential locations of regional foraging aggregations<br />
and priorities for studies of unsurveyed source rookeries:<br />
for example, particles from the southern Caribbean<br />
(Venezuela) would appear to be widely dispersed, yet<br />
rookeries are presently insufficiently characterized to<br />
permit inclusion in mixed stock estimates.<br />
Within the context of developing a hatchling drift<br />
model, much remains to be resolved or refined regarding<br />
biological parameters (e.g. sizes of hawksbill source<br />
populations, timing of peak hatching, duration of the<br />
Ó 2009 Blackwell Publishing Ltd
oceanic phase, behaviour of oceanic juveniles pushed<br />
toward landmasses, and roles of active swimming and<br />
orientation in hatchlings, post-hatchlings, and oceanic<br />
juveniles) and modelling ocean currents (e.g. incorporating<br />
wind-induced components; Hays & Marsh 1997;<br />
Girard et al. 2009). In this study, our hatchling drift<br />
model was limited to determining exposure of foraging<br />
sites to particles during the first year of drift. This is<br />
useful as a measure of potential for recruitment from<br />
each of the sources, as the timeframes within which<br />
hawksbills begin actively swimming and recruit to foraging<br />
grounds is unknown. However, a longer oceanic<br />
phase may occur (Musick & Limpus 1997), calling for<br />
extended modelling of oceanic drift when such longterm<br />
high-resolution oceanographic data become available.<br />
Comparison of multiple years of Aviso data may<br />
also be informative in determining inter-annual variations<br />
in recruitment and implications for management:<br />
if, as in green turtles, there is temporal structuring in<br />
the genetic composition of neritic juvenile foraging<br />
aggregations, then marine turtle management plans<br />
should not be based on the ‘snapshot’ of short-term<br />
genetic studies (Bjorndal & Bolten 2008).<br />
In coral reef fishes, behaviour of larvae has been<br />
shown to influence population connectivity (Paris et al.<br />
2007) but to date, it has proven difficult to separate the<br />
role of oceanic currents from behaviour in determining<br />
distribution of oceanic juvenile marine turtles (Naro-<br />
Maciel et al. 2007). Under experimental conditions, loggerhead<br />
hatchlings have been shown to exhibit directed<br />
swimming which is consistent with that which would<br />
be needed to remain within oceanic gyres (Lohmann<br />
et al. 2001) and in the wild, oceanic juvenile loggerheads<br />
may actively select foraging areas at locations<br />
closely correlated with the latitudes of their natal beaches<br />
(Monzón-Argüello et al. 2009) and home toward<br />
their natal regions at the conclusion of the oceanic stage<br />
(Bowen et al. 2004). However, the ability to move<br />
against prevailing currents appears to be linked to<br />
increasing body size (Monzón-Argüello et al. 2009).<br />
Oceanic juvenile loggerheads in downwelling lines<br />
show minimal activity and positive buoyancy (Witherington<br />
2002) and as hawksbill hatchlings at least in<br />
some populations appear to be less active in the initial<br />
days of dispersal than other marine turtle species<br />
(Chung et al. 2009a) it is likely that small oceanic juvenile<br />
hawksbills are also relatively passive surface drifters.<br />
Therefore, while vertical movement of reef fish<br />
larvae limits the application of models based on surface<br />
current data (Fiksen et al. 2007) these models are especially<br />
applicable to marine turtles.<br />
Occurrence of active orientation, particularly at the<br />
end of the oceanic stage, may nevertheless explain<br />
contributions of hawksbill rookeries against prevailing<br />
Ó 2009 Blackwell Publishing Ltd<br />
HAWKSBILL TURTLE DISPERSAL 4849<br />
currents to foraging sites. Alternatively, unsurveyed<br />
hawksbill rookeries may contribute to foraging stocks<br />
(i.e. contributions assigned to a down-current source<br />
may instead originate from an unsurveyed or incompletely<br />
surveyed up-current source). It is also possible<br />
that a portion of Caribbean hawksbills could circle the<br />
Atlantic before re-entering the Caribbean, following a<br />
similar route to oceanic juvenile loggerheads <strong>–</strong> or perhaps<br />
a ‘short-cycle’ of the Atlantic could occasionally<br />
occur where hawksbills might become entrained in<br />
Atlantic eddies which more rapidly re-enter the Caribbean.<br />
In our comparisons between many-to-many<br />
results and the hatchling drift model, estimated contributions<br />
from some rookeries were high using genetic<br />
markers while the drift model predicted no contributions.<br />
These findings may be indicative of an Atlantic<br />
cycle or short-cycle: for instance, Mexican contributions<br />
to seemingly up-current foraging sites in the eastern<br />
Caribbean may be explained by hawksbills cycling a<br />
portion of the Atlantic before re-entering the Caribbean.<br />
Integration of better biological data with heuristic<br />
modelling of oceanic drift will assist in resolving migratory<br />
connectivity for marine turtles. Patterns of dispersal<br />
for oceanic juveniles may be confirmed by<br />
documentation of stranding patterns (Meylan & Redlow<br />
2006) and genetic studies (Carreras et al. 2006; Bowen<br />
et al. 2007a). In surveys to date, magnitude of hawksbill<br />
genetic diversity varies regionally <strong>–</strong> being lowest in the<br />
Gulf of Mexico, where mixed stock analysis indicates<br />
that oceanic juveniles originate almost exclusively from<br />
Mexican rookeries (Bowen et al. 2007a), and higher near<br />
the eastern Caribbean gyres and in foraging areas<br />
passed by the Caribbean current. Sightings and strandings<br />
of oceanic phase hawksbills in Florida (Meylan &<br />
Redlow 2006) and rarely on the east coast (Parker 1995)<br />
support transport from the Caribbean through the<br />
Florida Straits, and the presence of a neritic juvenile<br />
hawksbill foraging aggregation in Bermuda (Meylan<br />
et al. 2003) suggests some transport in the Gulf Stream.<br />
However, further genetic sampling of stranded and<br />
free-living oceanic hawksbills and genetic characterization<br />
of additional neritic foraging grounds is required.<br />
Efforts to resolve movements are particularly timely,<br />
as management of hawksbill turtles has been the subject<br />
of considerable controversy in recent years. Hawksbills<br />
are designated as ‘critically endangered’ by the IUCN<br />
(2007) and listing on Appendix I of the Convention on<br />
the <strong>International</strong> Trade in Endangered Species (CITES)<br />
bars international trade among parties to the Convention.<br />
Nevertheless, commercially valuable hawksbill<br />
products are still in high demand in some areas. Recent<br />
debate has centred on conservation status and priorities,<br />
scale and extent of transboundary movements (Bowen<br />
& Bass 1997; Mrosovsky 1997) and sustainable use of
4850 J. M. BLUMENTHAL ET AL.<br />
mixed stocks (Bowen et al. 2007b; Godfrey et al. 2007;<br />
Mortimer et al. 2007a,b). It has been argued that harvesting<br />
of hawksbills from foraging grounds should be<br />
prohibited, as exploitation of mixed stocks could potentially<br />
impact rookeries on a regional scale <strong>–</strong> or alternatively,<br />
that sustainability of harvesting on hawksbill<br />
foraging aggregations could be determined on a caseby-case<br />
basis (Godfrey et al. 2007).<br />
Many-to-many analysis and hatchling drift models<br />
facilitate an integrative, rookery-centric approach<br />
(Bolker et al. 2007) in which contributions of both small,<br />
vulnerable rookeries and large, regionally important<br />
rookeries are evaluated. Oceanic juveniles from some<br />
rookeries appear to be dispersed among multiple foraging<br />
grounds, while those from other rookeries appear to<br />
be more locally constrained. This diversity of distribution<br />
represents both a challenge and an opportunity for<br />
marine turtle management. Broadly dispersed stocks<br />
are difficult to manage, yet threats are often geographically<br />
widely dispersed and their impacts on individual<br />
stocks will depend on how they interact. In contrast, in<br />
areas with greater levels of local or proximate recruitment,<br />
threats <strong>–</strong> or conservation efforts <strong>–</strong> may have a<br />
concentrated effect (Bowen et al. 2004, 2005). Indeed,<br />
Caribbean hawksbill rookeries have experienced varying<br />
population trends, ranging from near-extirpation<br />
(Cayman Islands: Aiken et al. 2001) to dramatic increase<br />
(Antigua: Richardson et al. 2006; Barbados: Beggs et al.<br />
2007) <strong>–</strong> raising the possibility that variations in dispersal<br />
across an oceanographically complicated region<br />
may be a factor in these differing trajectories.<br />
While much remains to be determined regarding<br />
movements and management needs of hawksbill turtles,<br />
integration of many-to-many analysis and oceanographic<br />
data represents a promising approach. In this<br />
study, we delineated links between regional management<br />
units by conducting many-to-many mixed stock<br />
analysis across the range of Caribbean hawksbill rookeries<br />
and foraging aggregations, developed a tool with<br />
which to illustrate a possible role of ocean currents in<br />
determining distribution of oceanic juveniles, and highlighted<br />
gaps in knowledge and priorities for future<br />
sampling. Overall, results facilitate a broader understanding<br />
of Caribbean hawksbill movements <strong>–</strong> filling an<br />
important gap in knowledge of life-history and potentially<br />
informing regional management.<br />
Acknowledgments<br />
For invaluable logistical support and assistance with fieldwork,<br />
we thank Cayman Islands Department of Environment<br />
research, administration, operations and enforcement staff and<br />
numerous volunteers. For advice on analysis, we thank J. Hunt<br />
and W. Pitchers at University of Exeter, Cornwall campus.<br />
Vincente Guzmán (CONANP-México) and Eduardo Cuevas<br />
(Pronature-Península de Yucatán A.C.) generously provided<br />
information on population sizes of Campeche, Yucatán and<br />
Quintana Roo rookeries. Work in the Cayman Islands, the UK<br />
and the USA was generously supported by the European<br />
Social Fund, the Darwin Initiative, the Department of Environment,<br />
Food & Rural Affairs (DEFRA), Eckerd College, the Foreign<br />
and Commonwealth Office for the Overseas Territories,<br />
the Ford Foundation, the National Fish and Wildlife Foundation<br />
(NFWF) and the Turtles in the Caribbean Overseas Territories<br />
(TCOT) Project. We also acknowledge support to J.<br />
Blumenthal (University of Exeter postgraduate studentship and<br />
the Darwin Initiative). The manuscript was improved by the<br />
input of three anonymous reviewers.<br />
References<br />
Abreu-Grobois FA, Horrocks JA, Formia A et al. (2006) New<br />
mtDNA dloop primers which work for a variety of marine<br />
turtle species may increase the resolution capacity of mixed<br />
stock analyses. Poster presented at the 26th Annual<br />
Symposium on Sea Turtle Biology and Conservation, Crete,<br />
Greece, 2<strong>–</strong>8 April 2006. Available from http://www.<br />
iucnmtsg.org/genetics/meth/primers/abreu_grobois_etal_<br />
new_dloop_primers.pdf. Accessed 1 March 2008.<br />
Aiken JJ, Godley BJ, Broderick AC, Austin T, Ebanks-Petrie G,<br />
Hays GC (2001) Two hundred years after a commercial<br />
marine turtle fishery: the current status of marine turtles<br />
nesting in the Cayman Islands. Oryx, 35, 145<strong>–</strong>151.<br />
Allard MW, Miyamoto MM, Bjorndal KA, Bolten AB, Bowen<br />
BW (1994) Support for natal homing in green turtles from<br />
mitochondrial DNA sequences. Copeia, 1994, 34<strong>–</strong>41.<br />
Allen M, Engstrom AS, Meyers S et al. (1998) Mitochondrial<br />
DNA sequencing of shed hairs and saliva on robbery caps:<br />
sensitivity and matching probabilities. Journal of Forensic<br />
Science, 43, 453<strong>–</strong>464.<br />
Andersen LW, Ruzzante DE, Walton M et al. (2001)<br />
Conservation genetics of harbour porpoises, Phocoena<br />
phocoena, in eastern and central North Atlantic. Conservation<br />
Genetics, 2, 309<strong>–</strong>324.<br />
Baillie JEM, Hilton-Taylor C, Stuart SN (2004) IUCN red List of<br />
Threatened Species. A Global Species Assessment. IUCN, Gland<br />
Switzerland.<br />
Baker CS, Patenaude NJ, Bannister JL, Robins J, Kato H (1999)<br />
Distribution and diversity of mtDNA lineages among<br />
southern right whales (Eubalaena australis) from Australia<br />
and New Zealand. Marine Biology, 134, 1<strong>–</strong>7.<br />
Bass AL (1999) Genetic analysis to elucidate the natural history<br />
and behavior of hawksbill turtles (Eretmochelys imbricata) in<br />
the Wider Caribbean: a review and reanalysis. Chelonian<br />
Conservation and Biology, 3, 195<strong>–</strong>199.<br />
Bass AL, Witzell WN (2000) Demographic composition of<br />
immature green turtles (Chelonia mydas) from the east central<br />
Florida coast: evidence from mtDNA markers. Herpetologica,<br />
56, 357<strong>–</strong>367.<br />
Bass AL, Good DA, Bjorndal KA et al. (1996) Testing models<br />
of female reproductive migratory behavior and population<br />
structure in the Caribbean hawksbill turtle, Eretmochelys<br />
imbricata, with mtDNA sequences. Molecular Ecology, 5, 321<strong>–</strong><br />
328.<br />
Ó 2009 Blackwell Publishing Ltd
Bass AL, Epperly SP, Braun-McNeill J (2006) Green turtle<br />
(Chelonia mydas) foraging and nesting aggregations in the<br />
Caribbean and Atlantic: impact of currents and behavior on<br />
dispersal. Journal of Heredity, 97, 346<strong>–</strong>354.<br />
Beggs JA, Horrocks JA, Krueger BH (2007) Increase in hawksbill<br />
sea turtle Eretmochelys imbricata nesting in Barbados, West<br />
Indies. Endangered Species Research, 3, 159<strong>–</strong>168.<br />
Bell CD, Solomon JL, Blumenthal JM et al. (2007) Monitoring<br />
and conservation of critically-reduced marine turtle nesting<br />
populations: lessons from the Cayman Islands. Animal<br />
Conservation, 10, 39<strong>–</strong>47.<br />
Bjorndal KA, Bolten AB (2008) Annual variation in source<br />
contributions to a mixed stock: implications for quantifying<br />
connectivity. Molecular Ecology, <strong>17</strong>, 2185<strong>–</strong>2193.<br />
Blumenthal JM, Austin TJ, Bell CD et al. (2009a) Ecology of<br />
hawksbill turtles Eretmochelys imbricata on a western<br />
Caribbean foraging area. Chelonian Conservation and Biology,<br />
8, 1<strong>–</strong>10.<br />
Blumenthal JM, Austin TJ, Bothwell JB et al. (2009b) Diving<br />
behaviour and movements of juvenile hawksbill turtles<br />
Eretmochelys imbricata on a Caribbean coral reef. Coral Reefs,<br />
28, 55<strong>–</strong>65.<br />
Bolker BM, Okuyama T, Bjorndal KA, Bolten AB (2007)<br />
Incorporating multiple mixed stocks in mixed stock analysis:<br />
‘many-to-many’ analyses. Molecular Ecology, 16, 685<strong>–</strong>695.<br />
Bolten AB (2003) Variation in sea turtle life-history patterns:<br />
neritic vs. oceanic developmental stages. In:The Biology of Sea<br />
Turtles, Volume I (eds Lutz PL, Musick JA, Wyneken J). pp.<br />
243<strong>–</strong>257, CRC Press, Boca Raton.<br />
Bolten AB, Bjorndal KA, Martins HR et al. (1998) Trans-<br />
Atlantic developmental migrations of loggerhead sea turtles<br />
demonstrated by mtDNA sequence analyses. Ecological<br />
Applications, 8, 1<strong>–</strong>7.<br />
Botsford LW, White JW, Coffroth M-A et al. (2009)<br />
Connectivity and resilience of coral reef metapopulations in<br />
marine protected areas: matching empirical efforts to<br />
predictive needs. Coral Reefs, 28, 303<strong>–</strong>305.<br />
Bowen BW, Bass AL (1997) Movement of hawksbill turtles:<br />
what scale is relevant to conservation and what scale is<br />
resolvable with mtDNA data? Chelonian Conservation and<br />
Biology, 3, 440<strong>–</strong>442.<br />
Bowen BW, Abreu-Grobois FA, Balazs GH et al. (1995) Trans-<br />
Pacific migrations of the loggerhead sea turtle demonstrated<br />
with mitochondrial DNA markers. Proceedings of the National<br />
Academy of Sciences USA, 92, 3731<strong>–</strong>3734.<br />
Bowen BW, Bass AL, Garcia A et al. (1996) The origin of<br />
hawksbill turtles in a Caribbean feeding area as indicated by<br />
genetic markers. Ecological Applications, 6, 566<strong>–</strong>572.<br />
Bowen BW, Bass AL, Chow S et al. (2004) Natal homing in<br />
juvenile loggerhead turtles (Caretta caretta). Molecular Ecology,<br />
13, 3797<strong>–</strong>3808.<br />
Bowen BW, Bass AL, Soares L, Toonen RJ (2005) Conservation<br />
implications of complex population structure: lessons from<br />
the loggerhead turtle (Caretta caretta). Molecular Ecology, 14,<br />
2389<strong>–</strong>2402.<br />
Bowen BW, Grand WS, Hillis-Starr Z et al. (2007a) Mixed-stock<br />
analysis reveals the migrations of juvenile hawksbill turtles<br />
(Eretmochelys imbricata) in the Caribbean sea. Molecular<br />
Ecology, 16, 49<strong>–</strong>60.<br />
Bowen BW, Grant WS, Hillis-Starr Z et al. (2007b) The<br />
advocate and the scientist: debating the commercial<br />
Ó 2009 Blackwell Publishing Ltd<br />
HAWKSBILL TURTLE DISPERSAL 4851<br />
exploitation of endangered hawksbill turtles. Molecular<br />
Ecology, 16, 3514<strong>–</strong>3515.<br />
Broderick D, Moritz C, Miller JD, Guinea ML, Prince JR,<br />
Limpus CJ (1994) Genetic studies of the hawksbill turtle<br />
(Eretmochelys imbricata): evidence for multiple stocks and<br />
mixed feeding grounds in Australian waters. Pacific<br />
Conservation Biology, 1, 123<strong>–</strong>131.<br />
Browne DC, Horrocks JA, Abreu-Grobois FA (in press)<br />
Population subdivision in hawksbill turtles nesting on<br />
Barbados, West Indies, determined from mitochondrial DNA<br />
control region sequences. Conservation Genetics, DOI 10.1007/<br />
s10592-009-9883-3<br />
Carr AF (1987) New perspectives on the pelagic stage of sea<br />
turtle development. Conservation Biology, 1, 103<strong>–</strong>121.<br />
Carreras C, Pont S, Maffucci F et al. (2006) Genetic structuring<br />
of immature loggerhead sea turtles (Caretta caretta) in the<br />
Mediterranean Sea reflects water circulation patterns. Marine<br />
Biology, 149, 1269<strong>–</strong>1279.<br />
Chacón D (2004) Caribbean Hawksbills <strong>–</strong> An Introduction to<br />
Their Biology and Conservation Status. WWF-Regional<br />
Program for Latin America and the Caribbean, San Jose,<br />
Costa Rica.<br />
Chung FC, Pilcher NJ, Salmon M, Wyneken J (2009a) Offshore<br />
migratory activity of hawksbill turtle (Eretmochelys imbricata)<br />
hatchlings, I. Quantitative analysis of activity, with<br />
comparisons to green turtles (Chelonia mydas). Chelonian<br />
Conservation and Biology, 8, 28<strong>–</strong>34.<br />
Chung FC, Pilcher NJ, Salmon M, Wyneken J (2009b) Offshore<br />
migratory activity of hawksbill turtle (Eretmochelys imbricata)<br />
hatchlings II. Swimming gaits, swimming speed, and<br />
morphological comparisons. Chelonian Conservation and<br />
Biology, 8, 35<strong>–</strong>42.<br />
van Dam RP, Diez CE, Balazs GH et al. (2008) Sex-specific<br />
migration patterns of hawksbill turtles from Mona Island,<br />
Puerto Rico. Endangered Species Research, 4, 85<strong>–</strong>94.<br />
Diaz-Fernandez R, Okayama T, Uchiyama T et al. (1999)<br />
Genetic sourcing for the hawksbill turtle, Eretmochelys<br />
imbricata, in the northern Caribbean region. Chelonian<br />
Conservation and Biology, 3, 296<strong>–</strong>300.<br />
Dutton PH (1996) Methods for collection and preservation of<br />
samples for sea turtle genetic studies. In: Proceedings of the<br />
<strong>International</strong> Symposium on Sea Turtle Conservation Genetics<br />
(eds Bowen BW, Witzell WN), pp. <strong>17</strong><strong>–</strong>24. National Technical<br />
Information Service, Springfield, VA, USA, NOAA Technical<br />
Memorandum NMFS-SEFSC-396.<br />
Encalada SE, Bjorndal KA, Bolten AB et al. (1998) Population<br />
structure of loggerhead turtle (Caretta caretta) nesting<br />
colonies in the Atlantic and Mediterranean as inferred from<br />
mitochondrial DNA control region sequences. Marine<br />
Biology, 130, 567<strong>–</strong>575.<br />
Engstrom TN, Meylan PA, Meylan AB (2002) Origin of<br />
juvenile loggerhead turtles (Caretta caretta) in a tropical<br />
developmental habitat in Caribbean Panama. Animal<br />
Conservation, 5, 125<strong>–</strong>133.<br />
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an<br />
integrated software package for population genetics data<br />
analysis. Evolutionary Bioinformatics Online, 1, 47<strong>–</strong>50.<br />
Fiksen O, Jorgensen C, Kristiansen T, Vikebo F, Huse G (2007)<br />
Linking behavioural ecology and oceanography: larval<br />
behaviour determines growth, mortality and dispersal.<br />
Marine Ecology Progress Series, 347, 195<strong>–</strong>205.
4852 J. M. BLUMENTHAL ET AL.<br />
Formia A, Godley BJ, Dontaine J-F, Bruford MW (2006)<br />
Mitochondrial DNA diversity and phylogeography of<br />
endangered green turtle (Chelonia mydas) populations in<br />
Africa. Conservation Genetics, 7, 353<strong>–</strong>369.<br />
Formia A, Broderick AC, Glen F et al. (2007) Genetic<br />
composition of the Ascension Island green turtle rookery<br />
based on mitochondrial DNA: implications for sampling and<br />
diversity. Endangered Species Research, 3, 145<strong>–</strong>158.<br />
Garduño-Andrade M (1999) Nesting of the hawksbill turtle,<br />
Eretmochelys imbricata, in Rio Lagartos, Yucatan, Mexico,<br />
1990<strong>–</strong>1997. Chelonian Conservation and Biology, 3, 281<strong>–</strong>285.<br />
Girard C, Tucker AD, Calmettes B (2009) Post-nesting<br />
migrations of loggerhead sea turtles in the Gulf of Mexico:<br />
dispersal in highly dynamic conditions. Marine Biology, 156,<br />
1827<strong>–</strong>1839.<br />
Godfrey MH, Webb GJW, Manolis SC, Mrosovsky N (2007)<br />
Hawksbill sea turtles: can phylogenetics inform harvesting?<br />
Molecular Ecology, 16, 3511<strong>–</strong>3513.<br />
Hasbún CR (2002) Observations on first day dispersal of<br />
neonatal hawksbill turtles (Eretmochelys imbricata). Marine<br />
Turtle Newsletter, 96, 7<strong>–</strong>10.<br />
Hays GC, Marsh R (1997) Estimating the age of juvenile<br />
loggerhead sea turtles in the North Atlantic. Canadian Journal<br />
of Zoology, 75, 40<strong>–</strong>46.<br />
Hood GM (2009) PopTools version 3.1.1. Available on the<br />
internet. URL http://www.cse.csiro.au/poptools. Accessed<br />
20 July 2008.<br />
Horrocks JA, Vermeer LA, Krueger B et al. (2001) Migration<br />
routes and destination characteristics of post-nesting<br />
hawksbill turtles satellite-tracked from Barbados, West<br />
Indies. Chelonian Conservation and Biology, 4, 107<strong>–</strong>114.<br />
IUCN (2007) Red List of Threatened Species. Available from<br />
http://www.iucnredlist.org. Accessed 1 April 2008.<br />
Kocher TD, Thomas WK, Meyers A et al. (1989) Dynamics of<br />
mitochondrial DNA evolution in animals: amplification and<br />
sequencing with conserved primers. Proceedings of the<br />
National Academy of Sciences of the USA, 86, 6196<strong>–</strong>6200.<br />
Lahanas PN, Bjorndal KA, Bolten AB et al. (1998) Genetic<br />
composition of a green turtle (Chelonia mydas) feeding<br />
ground population: evidence for multiple origins. Marine<br />
Biology, 130, 345<strong>–</strong>352.<br />
Laurent L, Casale P, Bradai N et al. (1998) Molecular<br />
resolution of marine turtle stock composition in fishery<br />
bycatch: a case study in the Mediterannean. Molecular<br />
Ecology, 7, 1529<strong>–</strong>1542.<br />
Lewis CB (1940) The Cayman Islands and marine turtle.<br />
In:Herpetology of the Cayman Island (ed. Grant C). pp. 56<strong>–</strong>65,<br />
Bulletin of the Institute of Jamaica Science Series, Kingston.<br />
Lohmann KJ, Cain SD, Dodge SA, Lohmann CM (2001)<br />
Regional magnetic fields as navigational markers for sea<br />
turtles. Science, 294, 364<strong>–</strong>366.<br />
Luke K, Horrocks JA, LeRoux RA, Dutton PH (2004) Origins of<br />
green turtle (Chelonia mydas) feeding aggregations around<br />
Barbados, West Indies. Marine Biology, 144, 799<strong>–</strong>805.<br />
Maffucci F, Kooistra WHCF, Bentivegna F (2006) Natal origin<br />
of loggerhead turtles, Caretta caretta, in the neritic habitat off<br />
the Italian coasts, Central Mediterranean. Biological<br />
Conservation, 127, 183<strong>–</strong>189.<br />
Meylan AB (1999) <strong>International</strong> movements of immature and<br />
adult hawksbill turtles (Eretmochelys imbricata) in the Caribbean<br />
region. Chelonian Conservation and Biology, 3, 189<strong>–</strong>194.<br />
Meylan A, Redlow A (2006) Eretmochelys imbricata <strong>–</strong> Hawksbill<br />
turtle. In: Biology and Conservation of Florida Turtles (ed.<br />
Meylan PA). Chelonian Research Foundation, Lunenburg, MA,<br />
USA, 3, 105<strong>–</strong>127.<br />
Meylan AB, Bowen BW, Avise JC (1990) A genetic test of the<br />
natal homing versus social facilitation models for green<br />
turtle migration. Science, 248, 724<strong>–</strong>727.<br />
Meylan PA, Meylan AB, Gray J, Ward J (2003) The hawksbill<br />
turtle in Bermuda. In: Proceedings of the Twenty-Second Annual<br />
Symposium on Sea Turtle Biology and Conservation (compiler,<br />
Seminoff JA), p. 26. National Technical Information Service,<br />
Springfield, VA, USA. NOAA Technical Memorandum<br />
NMFS-SEFSC-503.<br />
Millar RB (1987) Maximum likelihood estimation of mixed<br />
stock fishery composition. Canadian Journal of Fisheries and<br />
Aquatic Science, 44, 583<strong>–</strong>590.<br />
Milligan BG (1998) Total DNA isolation. In:Molecular Genetic<br />
Analysis of Populations, a Practical Approach (ed Hoelzel AR).<br />
pp. 29<strong>–</strong>64, IRL Press, Oxford.<br />
Moncada F, Carillo E, Saenz A, Nodarse G (1999)<br />
Reproduction and nesting of the hawksbill turtle,<br />
Eretmochelys imbricata, in the Cuban archipelago. Chelonian<br />
Conservation and Biology, 3, 257<strong>–</strong>263.<br />
Monzón-Argüello C, Rico C, Carreras C, Calbuig P, Marco A,<br />
López-Jurado LF (2009) Variation in spatial distribution of<br />
juvenile loggerhead turtles in the eastern Atlantic and<br />
western Mediterranean sea. Journal of Experimental Marine<br />
Biology and Ecology, 373, 79<strong>–</strong>86.<br />
Mora C, Sale PF (2002) Are populations of coral reef fish open<br />
or closed? Trends in Ecology and Evolution, <strong>17</strong>, 422<strong>–</strong>428.<br />
Mortimer JA, Donnelly M (2007) Marine Turtle Specialist<br />
Group 2007 IUCN Red List Status Assessment Hawksbill<br />
Turtle (Eretmochelys imbricata). Available from http://<br />
www.iucn-mtsg.org/red_list/ei/index.shtml. Accessed 1<br />
March 2008.<br />
Mortimer JA, Donnelly M, Meylan AB, Meylan PA (2007a)<br />
Critically endangered hawksbill turtles: molecular genetics<br />
and the broad view of recovery. Molecular Ecology, 16, 3516<strong>–</strong><br />
35<strong>17</strong>.<br />
Mortimer JA, Meylan PA, Donnelly M (2007b) Whose turtles<br />
are they, anyway? Molecular Ecology, 16, <strong>17</strong><strong>–</strong>18.<br />
Mrosovsky N (1997) Movement of hawksbill turtles <strong>–</strong> a<br />
different perspective on the DNA data. Chelonian<br />
Conservation and Biology, 3, 438<strong>–</strong>439.<br />
Musick JA, Limpus CJ (1997) Habitat utilization and migration<br />
in juvenile sea turtles. In:The Biology of Sea Turtle (eds Lutz<br />
PL, Musick JA). pp. 137<strong>–</strong>163, CRC Press, Boca Raton.<br />
Naro-Maciel E, Becker HJ, Lima EHSM, Marcovaldi MA,<br />
DeSalle R (2007) Testing dispersal hypotheses in foraging<br />
green sea turtles (Chelonia mydas) of Brazil. Journal of<br />
Heredity, 98, 29<strong>–</strong>39.<br />
Norman JA, Moritz C, Limpus CJ (1994) Mitochondrial DNA<br />
control region polymorphisms: genetic markers for<br />
ecological studies of marine turtles. Molecular Ecology, 3,<br />
363<strong>–</strong>373.<br />
Okuyama T, Bolker BM (2005) Combining genetic and<br />
ecological data to estimate sea turtle origins. Ecological<br />
Applications, 15, 315<strong>–</strong>325.<br />
Owens DW, Ruiz GW (1980) New methods of obtaining blood<br />
and cerebrospinal fluid from marine turtles. Herpetologica, 36,<br />
<strong>17</strong><strong>–</strong>20.<br />
Ó 2009 Blackwell Publishing Ltd
Paris CB, Cherubin LM, Cowen RK (2007) Surfing, spinning, or<br />
diving from reef to reef: effects on population connectivity.<br />
Marine Ecology Progress Series, 347, 285<strong>–</strong>300.<br />
Parker LG (1995) Encounter with a juvenile hawksbill turtle<br />
offshore Sapelo Island, Georgia. Marine Turtle Newsletter, 71,<br />
19<strong>–</strong>22.<br />
Pella J, Masuda M (2001) Bayesian methods for analysis of<br />
stock mixtures from genetic characters. Fishery Bulletin, 99,<br />
151<strong>–</strong>167.<br />
Pella J, Milner GB (1987) Use of genetic marks in stock<br />
composition analyses. In:Population Genetics and Fishery<br />
Managemen (eds Ryman N, Utter F). pp. 247<strong>–</strong>276, University<br />
of Washington Press, Seattle.<br />
Reece JS, Ehrhart LM, Parkinson CL (2006) Mixed stock<br />
analysis of juvenile loggerheads (Caretta caretta) in Indian<br />
River lagoon, Florida: implications for conservation<br />
planning. Conservation Genetics, 7, 345<strong>–</strong>352.<br />
Richardson JI, Hall DB, Mason PA et al. (2006) Eighteen years<br />
of saturation tagging data reveal a significant increase in<br />
nesting hawksbill sea turtles (Eretmochelys imbricata) on Long<br />
Island, Antigua. Animal Conservation, 9, 302<strong>–</strong>307.<br />
Roberts H (1994) Reefs and lagoons of Grand Cayman. In:The<br />
Cayman Islands: Natural History and Biogeograph (eds Brunt<br />
MA, Davies JE). pp. 75<strong>–</strong>104, Kluwer Academic Publishers,<br />
Netherlands.<br />
Spalding MD, Fox HE, Allen GR et al. (2007) Marine<br />
ecoregions of the world: a bioregionalization of coast and<br />
shelf areas. BioScience, 57, 573<strong>–</strong>583.<br />
Troëng S, Dutton PH, Evans D (2005) Migration of hawksbill<br />
turtles Eretmochelys imbricata from Tortuguero, Costa Rica.<br />
Ecography, 28, 394<strong>–</strong>402.<br />
Velez-Zuazo X, Ramos WD, van Dam RP et al. (2008) Dispersal,<br />
recruitment and migratory behaviour in a hawksbill sea turtle<br />
aggregation. Molecular Ecology, <strong>17</strong>, 839<strong>–</strong>853.<br />
Wessel P, Smith HF (1991) Free software helps map and<br />
display data. Eos, Transactions of the American Geophysical<br />
Union, 72, 441.<br />
Ó 2009 Blackwell Publishing Ltd<br />
HAWKSBILL TURTLE DISPERSAL 4853<br />
Whiting SD, Koch AU (2006) Oceanic movement of a benthic<br />
foraging juvenile hawksbill turtle from the Cocos (Keeling)<br />
Islands. Marine Turtle Newsletter, 112, 15<strong>–</strong>16.<br />
Witherington BE (2002) Ecology of neonate loggerhead turtles<br />
inhabiting lines of downwelling near a Gulf Stream front.<br />
Marine Biology, 140, 843<strong>–</strong>853.<br />
Witteveen BH, Straley JM, von Ziegesar O, Steel D, Baker CS<br />
(2004) Abundance and mtDNA differentiation of humpback<br />
whales (Megaptera novaeangliae) in the Shumagin Islands,<br />
Alaska. Canadian Journal of Zoology, 82, 1352<strong>–</strong>1359.<br />
J.M.B. is a research officer at the Cayman Islands Department<br />
of Environment, where her work centres on using spatial ecology<br />
to aid in understanding management requirements of sea<br />
turtles and other marine species. G.E.-P. is the director of the<br />
Department of Environment and T.J.A. is the deputy director<br />
for the research section. M.S.C. coordinates a gobal sea turtle<br />
network as director of seaturtle.org. F.A.A.-G.’s research on<br />
conservation genetics spans marine turtle populations in Mexican<br />
and Wider Caribbean sites. P.A.M. is RR Hallin Professor<br />
of Natural Science at Eckerd College; A.B.M. leads the marine<br />
turtle research program for the Florida Fish and Wildlife Conservation<br />
Commission; together they study the reproductive<br />
biology, ecology and migrations of sea turtles in the Western<br />
Atlantic with long-term projects in Panama and Bermuda. A.F.<br />
works on marine turtle conservation genetics in West Africa<br />
and coordinates the Marine Turtle Partnership in Gabon.<br />
M.W.B. is head of the Biodiversity and Ecological Processes<br />
Group at Cardiff University. A.C.B. and B.J.G. coordinate the<br />
Marine Turtle Research Group at a range of sites around the<br />
world including Ascension Island, Northern Cyprus and the<br />
UK Overseas Territories.
The Elusive Manatee --<br />
An ethological approach to understanding behavior in the West Indian manatee<br />
Caryn Self Sullivan, Department of Wildlife and Fisheries Sciences, Texas A&M University, College Station, TX 77843-2258<br />
Last updated 19 October 2000<br />
CONTENTS<br />
Preface<br />
Part I: Introduction --<br />
Ethology, Proximate & Ultimate, and <strong>Sirenian</strong>s<br />
Part II: Problem Solving --<br />
Social, Reproductive & Physical<br />
References<br />
Recommended Reading<br />
Glossary<br />
Acknowledgments<br />
PREFACE<br />
This Species Brief originated in WFSC 422, an<br />
ethology course taught by my academic advisor, Dr. Jane<br />
M. Packard, at Texas A&M University in 1998. It was<br />
updated this year for distribution to Earthwatch Institute<br />
volunteers. Please consider it a work in progress <strong>–</strong> a “draft<br />
document” as it is continuously being revised and updated<br />
with better references and new information. Designed to be<br />
used by schools, zoos, wildlife parks, and oceanaria, it<br />
makes an excellent starting place for students, teachers, and<br />
others who are interested in learning about animal behavior<br />
and/or sirenians (manatees & dugongs). But, remember, it<br />
is only a briefing document. Use it to catapult yourself into<br />
the exciting world of animal behavior -- using manatees as<br />
an example. For more details, start with the<br />
Recommended Reading section; scientists and university<br />
students are encouraged to delve into the primary literature<br />
listed in the References section.<br />
PART I: INTRODUCTION<br />
As we idled around the corner of Swallow Caye I<br />
sighted two manatee noses in the distance. They were<br />
barely visible as they broke the surface of the clear<br />
Caribbean water. Patch, our boat operator, spotted the<br />
manatees at almost the same instant <strong>–</strong> he probably saw<br />
them first -- because before I could motion to him, he had<br />
already shut down the engine. We waited in silence, hoping<br />
they would surface again. So it goes with research on the<br />
elusive manatee. Most behavioral observations of manatees<br />
have been conducted on Florida manatees, either in<br />
captivity or in the clear spring waters of Florida during<br />
winter aggregations. Only recently have we attempted to<br />
observe behavior in Antillean manatees, which are sparsely<br />
distributed throughout the Caribbean, including the tropical<br />
waters of Belize.<br />
Five minutes passed - how long can these guys stay<br />
down? What are they doing down there? One reason we<br />
know so little about these incredibly well adapted animals<br />
is because they spend the majority of their time underwater,<br />
1<br />
regularly staying submerged 3-5 minutes between breaths.<br />
We heard them before we saw them. Both noses broke the<br />
water with a forceful exhalation at virtually the same<br />
moment. Then they were down again. I quietly entered the<br />
water and stealthily snorkeled the 50 meters towards their<br />
last location. Where did they go? Stop. Look. Listen. I<br />
heard them breathe again. When they finally came into<br />
view underwater, I thought, "Uh oh... a mother calf pair --<br />
they are going to run away".<br />
The larger animal was about 3 meters long, almost<br />
twice as big as the smaller. Readings and previous<br />
experience led me to assume a mother-calf relationship<br />
based on this size differential. But they didn't run. The<br />
larger animal was gently nuzzling the smaller one's back<br />
with its big prehensile lips. The next time they surfaced to<br />
breath, they were nose to nose in a manatee "kiss". When<br />
they noticed us, the nuzzling stopped and they both sank<br />
slowly to the soft muddy bottom. As they sank, I heard a<br />
few squeaks, similar to manatee vocalizations I’d recorded<br />
between mother-calf pairs last year. There they rested, side<br />
by side in typical mother-calf position, for three minutes.<br />
When the smaller one rose to the surface to breathe, I could<br />
tell it was a female by the location of a genital slit near the<br />
anus. But, what a surprise I had when the larger animal<br />
surfaced and I saw by a genital slit near the umbilicus scar<br />
that it was a male. I'll make no more assumptions about<br />
mother-calf pairs based on size differentials!<br />
Ethology<br />
Do manatees breath simultaneously? If so, why?<br />
Why was the large male manatee nuzzling the smaller<br />
female? Why do manatees "kiss"? How did they sink to<br />
the bottom and stay - without moving a muscle? Why did<br />
they vocalize during their descent? How do manatees<br />
create sound? Do manatees often lie side by side on the<br />
bottom? Are manatees usually found in pairs, or groups, or<br />
alone? Why? How long can manatees stay underwater<br />
without breathing? Why did the smaller animal surface to<br />
breath before the larger animal did? These are just a few of<br />
the questions raised by the brief observation. Some<br />
answers to these "how" and "why" questions are known;<br />
other answers may come through long-term ethological<br />
studies.<br />
Ethology is a relatively new, multi-perspective<br />
scientific approach to the study of animal behavior. Made<br />
famous by the work of 1973 Nobel Prize winners, Konrad<br />
Lorenz, Karl von Frisch, and Nikolaas Tinbergen<br />
(www.nobel.se/medicine/laureates/1973/index.html), it<br />
focuses on animal behavior in a natural setting. By using a<br />
Scientific Perspective, it differs from Folk Psychology,<br />
which is often used to explain animal behavior to the
general public. Folk psychology perspectives are intuitive<br />
in nature, usually based on personal experiences and<br />
observations. They are considered anthropomorphic<br />
because they describe and explain behavior in human terms<br />
<strong>–</strong> which are often the only terms we have to start with!<br />
These perspectives are appropriate and very useful when<br />
communicating with non-scientists, such as audiences in<br />
zoos, oceanaria, and wildlife parks. An interpreter will<br />
often use folk psychology to describe and explain animal<br />
behaviors based on the “model” that animals have desires,<br />
beliefs, and emotions like humans. Scientists also use these<br />
perspectives in developing hypotheses about specific<br />
behaviors. For example, I was using folk psychology when<br />
I assumed that the manatees in the anecdote above were a<br />
mother-calf pair. My intuition was based on the size<br />
differential and behavioral patterns typical of mother-calf<br />
pairs.<br />
Ethology encourages us to develop additional<br />
Scientific Perspectives in understanding, explaining,<br />
and/or describing animal behavior; the classical ethological<br />
perspectives include cause, development, evolution, and<br />
function (Martin & Bateson 1993, Lehner 1996). Modern<br />
ethologists agree that the behavior of an animal is the result<br />
of complex interactions between the genetic makeup of an<br />
individual and environmental factors that act upon the<br />
individual (Alcock 1998). However, many aspects of an<br />
animal's behavior can be explained from two very different<br />
perspectives: proximate and ultimate (Martin and Bateson<br />
1993, Lehner 1996). This often results in<br />
miscommunication among observers who are looking at<br />
behavior from different perspectives. Proximate and<br />
ultimate comparisons are equivalent to apple and orange<br />
comparisons <strong>–</strong> i.e. they are both valid fruits, but they are<br />
different things. "How" questions are usually asked from a<br />
proximate perspective; how questions seek explanations<br />
about the physical and chemical mechanisms that trigger an<br />
individual animal's behavior at any given point in time.<br />
"Why" questions, on the other hand, are usually based on<br />
ultimate perspectives; answers to these questions attempt<br />
to explain why certain behaviors exist within a population<br />
(or species) of animals. In other words, what pressures of<br />
natural selection led to the existence of a particular<br />
behavior within a population or species. Ethologists<br />
further divide proximate and ultimate into the subcategories<br />
of cause, development, evolution, and function<br />
based on the work of Niko Tinbergen (Martin and Bateson<br />
1993, Lehner 1996).<br />
Proximate: Cause and Development are<br />
proximate perspectives, which look at the behavior of an<br />
individual animal. Proximate Cause perspectives include<br />
looking at both internal mechanisms (hormones,<br />
neurotransmitters) and external stimuli (pheromones, photoperiod,<br />
temperature) that interact to trigger specific<br />
behaviors in a mature animal. Dr. Jane M. Packard<br />
explains it using the analogy of a camera, “Think of<br />
Proximate Cause as a snapshot in time that shows what is<br />
causing the behavior at that particular moment.” For<br />
2<br />
example, in our observation above, what “caused” the<br />
manatees to kiss? Was the female was giving off some<br />
signal (vocal, chemical, or behavioral) that attracted the<br />
male? Did the tactile stimulation by the male cause<br />
hormone production in the female, which triggered the<br />
“kiss”? Most likely, it was is a complex interaction<br />
between both the internal state of each animal and the<br />
resulting external behavioral stimuli. Proximate<br />
Development perspectives look at behavioral changes that<br />
occur as an individual animal matures. Think of Proximate<br />
Development as a "video" that shows how a behavior<br />
develops and changes over time as an individual animal<br />
matures. How might the behavior of manatees at different<br />
ages compare to the interactions we observed?<br />
Ultimate: Evolution and Function are ultimate<br />
perspectives, which look at specific behaviors present in a<br />
population of animals. These behaviors are thought to<br />
have evolved over time through the process called Natural<br />
selection. For Natural selection to act on a behavioral<br />
characteristic, the behavior must meet certain criteria <strong>–</strong> the<br />
same criteria necessary for natural selection to act on a<br />
physical trait such as coloration: (1) the trait must vary<br />
among individuals within a population; (2) the variation<br />
must be heritable; (3) if the heritable variation results in<br />
differential fitness (i.e. variations in the trait result in some<br />
individuals reproducing more successfully than others);<br />
then (4) we would expect the behavior to become<br />
genetically fixed in the population as the proportion of<br />
individuals displaying the trait increased (i.e. changes in<br />
the proportion of genotype and resulting phenotype).<br />
Ultimate Evolution perspectives include the<br />
comparison of behaviors among different, closely related<br />
species. This is our “video” perspective. From an<br />
Ultimate Evolution perspective, we hypothesize about how<br />
a behavior has changed (or remained the same) at the<br />
population and/or the species level over many generations.<br />
In my study of Antillean manatees, I will be comparing<br />
behavior to previous observations of behavior in Florida<br />
manatees. Although these sub-species are very closely<br />
related, they share different habitats. The Florida manatee<br />
inhabits a temporal/sub-tropic region where its behaviors<br />
are shaped by dramatic changes in water temperature during<br />
the year. On the other hand, the Antillean manatee inhabits<br />
a tropical region where the water temperature is relatively<br />
constant year round. We expect some behaviors to differ<br />
between the two sub-species as they evolved in different<br />
habitats. If behaviors, which we think are driven by water<br />
temperature today, exist in both sub-species, then perhaps<br />
they had some other function in the past.<br />
Ultimate Function perspectives attempt to explain<br />
what the function of a specific behavior is within a<br />
population, (i.e. why animals that display this behavioral<br />
trait are more reproductively successful than individuals<br />
who do not). This is our “snapshot” perspective. If<br />
variation exists within the behavior, Ultimate Function is<br />
the perspective used to explain why. Some Florida<br />
manatees travel long distances into more temperate regions
during the summer months while others stay in the same<br />
area year round <strong>–</strong> but we are not sure why the traveling<br />
behavior exists. The traveling animal must use more<br />
energy than the year round resident. Perhaps male animals<br />
that travel are exposed to more potential mates <strong>–</strong> thereby<br />
increasing their reproductive success. Chessie, a famous<br />
male manatee first sighted in the Chesapeake Bay in 1994,<br />
is known to have traveled between Florida in the winter and<br />
Rhode Island in the summer of 1995! Sweet Pea, a female<br />
rescued near Houston, Texas, later traveled along both<br />
coasts of Florida. Gina, a female manatee first sighted in<br />
Tampa Bay on the west coast of Florida is currently<br />
hanging out in the Bahamas! One of my questions is: Do<br />
Antillean manatees exhibit similar long distance traveling<br />
behaviors?<br />
Hypotheses about “why” these behaviors exist in<br />
manatees are different from hypotheses about “how” these<br />
behaviors are executed. The “why” questions are from<br />
ultimate perspectives of evolution and function, the “how”<br />
questions are from proximate perspectives of cause and<br />
development. These four basic concepts of ethology can<br />
be arranged in a 2 x 2 table comparing TIME FRAME and<br />
ANALYSIS PERSPECTIVES:<br />
TIME/<br />
ANALYSIS<br />
Proximate<br />
Perspective<br />
Individual Animals<br />
“How Questions”<br />
Ultimate<br />
Perspective<br />
Populations/Species<br />
“Why Questions”<br />
Pattern-Static<br />
“Snapshot”<br />
CAUSE<br />
(control)<br />
behavioral<br />
triggers:<br />
internal state/<br />
external<br />
stimuli<br />
FUNCTION<br />
adaptive<br />
significance:<br />
effect on<br />
reproductive<br />
fitness<br />
Process-Dynamic<br />
“Video”<br />
DEVELOPMENT<br />
(ontogeny)<br />
changes in<br />
behavior as an<br />
animal ages:<br />
maturation and/or<br />
learning<br />
EVOLUTION<br />
(phylogeny)<br />
changes in<br />
behavior<br />
(genotype) as<br />
populations/<br />
species diverge<br />
We can remember the concepts of ethology with the<br />
acronym AB=CDEF (Animal Behavior = Cause,<br />
Development, Evolution, Function). On the TIME axis,<br />
Cause and Function are “snapshot” perspectives that look<br />
at internal state and external stimuli in an individual animal<br />
or at reproductive success in a population of animals.<br />
Development and Evolution are “video” perspectives that<br />
look at changes in behaviors over time, either as the<br />
individual animal matures or as the population evolves.<br />
On the ANALYSIS axis, Cause and Development are<br />
proximate perspectives that attempt to answer "how"<br />
questions at the level of individual animals. Evolution and<br />
Function are ultimate perspectives that attempt to answer<br />
"why" questions at the level of populations and/or species.<br />
3<br />
For more information on Ethology, I encourage you to visit<br />
Dr. Packard’s website at canis.tamu.edu/wfscCourses/.<br />
<strong>Sirenian</strong>s<br />
So what are manatees anyway, and why should we<br />
study their behavior? Manatees belong to the order Sirenia<br />
of which there are only 4 extant species in 2 families,<br />
Trichechidae and Dugongidae. Although scientists often<br />
lump sirenians together with the order Cetacea (whales<br />
and dolphins) as totally aquatic marine mammals, manatees<br />
and the dugong are actually more closely related to<br />
elephants, hyraxes, and aardvarks than to any other marine<br />
mammal (Fischer 1990, Maluf 1995, Springer et al. 1997,<br />
Gaeth et al. 1999). Until a few years ago, very few people<br />
had ever heard of manatees, dugongs, or sea cows. But, as<br />
we learn more about these elusive and highly specialized<br />
creatures that share our coastal habitats, they are becoming<br />
more and more popular among both scientists and<br />
conservationists. The West Indian manatee (Trichechus<br />
manatus), the West African manatee (Trichechus<br />
senegalensis), and the Amazonian manatee (Trichechus<br />
inunguis) are members of the family Trichechidae. The<br />
dugong (Dugong dugon) is the only surviving member of<br />
the family Dugongidae (Reynolds and Odell 1991).<br />
Steller's sea cow (Hydrodamalis gigas) is usually included<br />
when we talk about modern sirenians; it was in the family<br />
Dugongidae (Reynolds and Odell 1991), but the species<br />
was extirpated by humans in <strong>17</strong>68, just 27 years after it<br />
was discovered by Russian explorers in <strong>17</strong>41 (Stejneger<br />
1887). Today, local and international laws protect all four<br />
living species, but they are also either threatened or<br />
endangered by humans wherever they exist.<br />
Florida Manatees: There is little evidence that<br />
Florida manatees were ever harvested commercially. But<br />
subsistence use, habitat destruction, and competition for<br />
space with recreational boaters have taken their toll on both<br />
prehistoric and modern populations. The USGS Sirenia<br />
Project, the U. S. Fish and Wildlife Service, and the Florida<br />
Fish and Wildlife Conservation Commission (formerly the<br />
Florida Department of Environmental Protection) have<br />
funded much of the manatee research in the United States.<br />
Over the past three decades, conservation efforts in Florida<br />
resulted in significant scientific research on the distribution,<br />
population biology, and behavior of the Florida manatee<br />
subspecies, T. m. latirostris (see O'Shea et al. 1995).<br />
Ecological and behavioral studies on localized populations<br />
such as those in Crystal and Homosassa Rivers (Hartman<br />
1979), St. John's River (Bengtson 1981), and Sarasota Bay<br />
(Koelsch 1997) have added to our understanding of Florida<br />
manatee behavior. However, relatively little research has<br />
been conducted outside of Florida. Therefore, most of the<br />
referenced information contained herein refers to the<br />
Florida subspecies.<br />
Antillean Manatees: Even before Russian sailors<br />
were exploiting Steller’s sea cow meat in the North Pacific,<br />
European explorers were provisioning their ships with<br />
Antillean manatee meat from the Caribbean area. Besides<br />
harvesting manatees for subsistence, some Native
Americans also sold manatee meat to the Europeans<br />
(O’Shea 1994). Today, the Antillean subspecies, T. m.<br />
manatus, is classified as endangered throughout its sparse<br />
distribution in the Caribbean Sea and Western Tropical<br />
Atlantic Ocean (Lefebvre et al. 1989). O'Shea and<br />
Salisbury (1991) suggest that Belize (formerly British<br />
Honduras), where manatees have been protected since the<br />
1930s (Auil 1998), may be the last stronghold for Antillean<br />
manatees in the Caribbean. Current research in Belize by<br />
James A. "Buddy" Powell, Nicole Auil, Greg Smith, Katie<br />
LaCommare, and myself is expanding our knowledge of the<br />
Antillean manatee. Most of the anecdotes included in this<br />
brief come from my personal experiences in Belize.<br />
PART II: PROBLEM-SOLVING<br />
One way to interpret animal behavior is as a method<br />
of problem solving. Over geological time, animals have<br />
evolved behaviors that enable them to solve problems.<br />
Some of these behaviors are identical from one individual<br />
to another. We describe such a behavior as a Fixed Action<br />
Pattern (FAP), because the individual’s genes control the<br />
trait. In other words, the genetic trait has become fixed in<br />
the population and all individuals perform the behavior in<br />
exactly the same way because they inherited the trait from<br />
their ancestors. At the other end of the behavioral scale, we<br />
find behaviors that vary a great deal among individuals.<br />
We describe such a behavior as a Variable Action Pattern<br />
(VAP), because the environment controls the trait. In other<br />
words, each individual performs the behavior differently<br />
due to different environmental factors during development.<br />
Additionally, any individual may perform the behavior<br />
differently at different times, depending its internal state<br />
(hormones, chemicals, neurons) and on external stimuli<br />
(environment). Of course FAP and VAP are not specific<br />
categories, but are the end points along a continuum. If a<br />
behavior falls near the middle of this continuum, we<br />
describe it as a Modal Action Pattern (MAP). In other<br />
words, the behavior is controlled in part by genetics and in<br />
part by the environment.<br />
We can divide problem solving into three major<br />
categories: reproductive, physical, and social. Manatees<br />
have evolved some interesting behaviors to overcome<br />
reproductive and physical problems. But, they are not<br />
considered to be very social animals. Although they tend<br />
to aggregate on resources, they do not appear to live in<br />
social groups and significant social behaviors have not been<br />
observed outside of reproductive activities. This species<br />
brief will use the concepts of ethology to introduce you to<br />
the behavioral methods manatees use to solve some of their<br />
reproductive and physical problems. We will examine<br />
specific behaviors using the different perspectives of<br />
proximate cause, proximate development, ultimate<br />
evolution, and ultimate function to answer a few questions<br />
regarding "how" and "why" manatees behave as the do.<br />
4<br />
Reproductive Problem Solving<br />
We idled into one of my favorite coves at the end of<br />
Bogue C in the Drowned Cayes near Belize City. I had<br />
taken volunteer researchers to this spot on previous<br />
occasions and ALWAYS, there had been a manatee resting<br />
in the manatee hole on the far side of the cove. As if on<br />
cue, before we could even cut the engine and anchor the<br />
boat, a single manatee surfaced in the vicinity of the<br />
manatee hole. Within minutes a second manatee surfaced<br />
in the middle of the cove. Two minutes later, a third animal<br />
entered the cove. "Gee...” I thought, "This must be a<br />
popular resting cove!" But, the two animals in the center of<br />
the cove were too active to be resting. I didn’t think they<br />
could be feeding, either, because previous habitat snorkels<br />
had found NO vegetation on the bottom. Another four<br />
minutes passed and two more manatees swam under the<br />
boat to join the active pair in the middle of the cove. "This<br />
is great,” I told the volunteers, "we'll be able to see how<br />
long it takes them to settle into a resting pattern". But they<br />
didn't settle. For the next hour we watched the four<br />
manatees in the middle of the cove breathe, roll, dive, and<br />
kiss while the first animal appeared oblivious to all the<br />
activity less than 50 meters away. Were we observing a<br />
mating herd?<br />
Mating System: One parameter of the West Indian<br />
manatee mating system is known as the mating herd. A<br />
mating system is the species-typical pattern of problem<br />
solving that includes how an individual finds a mate, how<br />
long it remains with the mate, and how much energy it<br />
invests in its offspring (Drickamer, et al. 1996). The West<br />
Indian manatee mating system can be broadly defined as<br />
promiscuous with the estrous female exhibiting<br />
polyandrous behavior and the male exhibiting polygynous<br />
behavior (Hartman 1979). A manatee-mating herd<br />
consists of a group of males in pursuit of an estrous female.<br />
The group is ephemeral, lasting only from a week to a<br />
month (Hartman 1979) and consisting of up to 20 males.<br />
The group does not remain together afterwards. Males will<br />
participate in multiple mating herds and attempt to<br />
copulate with many estrous females; similarly, females will<br />
copulate with multiple males among those in the herd.<br />
When we discuss why this mating system exists in<br />
manatees, we are using the ultimate function perspective.<br />
From the perspective of proximate cause, we do not<br />
know exactly what external signal stimulates males to<br />
aggregate around and attempt to copulate with the estrous<br />
female. Females must produce some sort of signal,<br />
possibly a chemical or acoustical signal, which stimulates<br />
an internal hormonal mechanism in males causing them to<br />
pursue her. Likewise, the male’s internal state must be such<br />
that he responds to the signal. Proximate development<br />
perspectives would look at how the males’ reaction to such<br />
a signal might differ at other stages of maturity.<br />
Daniel S. Hartman, one of the first scientists to<br />
make long-term observations of manatee behavior in the<br />
wild, found similarities between manatee mating herds and<br />
elephant mating, noting that female elephants are also
polyandrous - often mating with several males over a<br />
period of several hours. When Hartman (1979) compares<br />
the mating behavior of manatees to that of elephants, he is<br />
writing from an ultimate evolution perspective. In other<br />
words, he is hypothesizing that this aspect of the mating<br />
system evolved millions of years ago in an ancestor shared<br />
by both the manatee and the elephant. Since sirenians and<br />
proboscideans are two of only four extant orders that share<br />
a common ancestor among them, manatees are often<br />
compared to elephants using the ultimate evolution<br />
perspective [NOTE: The other two orders contain the<br />
hyraxes and the aardvarks, which are rarely compared to<br />
manatees in the literature]. Another promiscuous aspect of<br />
the manatee mating system is scramble-polygyny, where<br />
multiple males attempt to mate with the estrous female, but<br />
- without overt competition. Although males aggregate on<br />
the estrous female and jockey for the best position - they<br />
exhibit little agonistic behavior. Interestingly, male<br />
dugongs appear to be more agonistic during mating events.<br />
They set up territories and exhibit lek mating behaviors<br />
(Anderson 1997). What perspective would we use to<br />
compare mating strategies between manatees and dugongs?<br />
Timing: Let's assume that our observation was of a<br />
mating herd. That is, the group of manatees in the center<br />
of the cove consisted of 1 estrous female and 3 males.<br />
Why was the first manatee, the one originally sighted in the<br />
resting hole, not involved in the mating herd? Looking at<br />
the situation from a proximate perspective, there are<br />
several possibilities, and all involve timing. Suppose the<br />
resting manatee was a female. If she was sexually mature,<br />
but not in estrous, the mating herd would have no interest<br />
as she would not be producing an estrous signal. An<br />
estrous signal is the proximate cause of the mating herd<br />
behavior. It's "how" the males know the female is ready to<br />
conceive. Similarly, if the female were sexually immature,<br />
she could not be in estrous and therefore would not be<br />
sending a signal. Scientists have only recently answered<br />
the question of when a female manatee becomes sexually<br />
mature, thanks to the development of a new aging technique<br />
by Miriam Marmontel, et al. (1990). Since manatees<br />
continuously regenerate new teeth throughout their lives<br />
(Domning and Hayek 1984), they cannot be aged by their<br />
dentition like many other marine mammals. But, by<br />
looking at growth layers in manatee ear bones, we are now<br />
reasonably confident that female manatees in Florida reach<br />
sexual maturity between the age of 3 and 4 years - most<br />
giving birth to their first calf at age 4 (Marmontel 1995).<br />
Questions of "how" the behavior of signaling develops in<br />
females as they mature fall under the proximate<br />
development perspective.<br />
On the other hand, if the resting manatee was a<br />
male, why wasn't he attracted to the estrous female in the<br />
middle of the cove? He could have been either sexually<br />
immature or sexually inactive. Using the presence or<br />
absence of sperm in the testes as an indicator, Hernandez et<br />
al. (1995) found that sexual maturity (proximate<br />
development) varied among Florida male manatees with<br />
5<br />
both size and age with some males becoming<br />
physiologically mature as young as 2 years and as small as<br />
237 cm. But, from a proximate cause perspective, they<br />
also found that the reproductive system varied in<br />
functionality among mature male manatees depending on<br />
season in Florida, with little evidence of spermatogenesis<br />
present during winter months (December <strong>–</strong> February). In<br />
many mammals, reproductive activity varies seasonally<br />
with photoperiod, or the number of light hours per day.<br />
The pineal gland is usually the organ associated with<br />
behavioral changes affected by photoperiod. However, no<br />
pineal gland has ever been found in manatees or dugongs<br />
(Ralph et al. 1985, W. Welker personal communication<br />
2000). From an ultimate evolution perspective, it is<br />
interesting that the literature is unclear regarding the<br />
existence of a pineal gland in elephants (Ralph et al. 1985).<br />
Whether the resting manatee was inactive or<br />
immature, his timing would have been out of sync with the<br />
female and the estrous signal would have no effect on his<br />
behavior. From an ultimate function perspective, we say<br />
that those manatees whose sexual behavior is triggered at<br />
the appropriate time (i.e. when both the male and female are<br />
sexually mature, active, and receptive) are more<br />
reproductively successful than manatees that waste energy<br />
on futile sexual encounters. From an ultimate evolution<br />
perspective, there have been some behaviors observed in<br />
Antillean manatees that might be associated with seasonal<br />
spermatogenesis (G. Smith unpublished data). More<br />
studies are necessary before we can determine if seasonality<br />
affects the reproductive behavior of manatees in Belize.<br />
Parental Care: The final aspect of reproductive<br />
problem solving discussed here is parental care. Like many<br />
mammals, female manatees invest considerable time and<br />
energy into a relatively small number of offspring as their<br />
reproductive strategy. From an ultimate function<br />
perspective, data collected by scientists in Florida suggest<br />
that those females that invest 2 years of parental care in<br />
each offspring prior to becoming pregnant again are more<br />
successful than other females (Marmontel 1995). Although<br />
calves begin eating on their own within 3 months of birth,<br />
they continue to nurse periodically (Hartman 1979) as they<br />
grow and learn migration routes from their mothers (R.<br />
Bonde, personal communication 1999). When we ask<br />
"why" this behavior exists, we are asking ultimate function<br />
questions. Perhaps calves need the extra protein and fat<br />
provided by mother's milk during developmental years. Or,<br />
perhaps it takes calves almost two years to learn the routes<br />
to warm water effluents and good foraging grounds<br />
necessary for survival through the temperate winters in<br />
northern and central Florida.<br />
In an ongoing study of Antillean manatees in the<br />
Southern Lagoon of Belize, Buddy Powell is also seeing<br />
mother-calf pairs remain together for long periods of time<br />
(www.wesave.org/manatee/). When we compare this<br />
behavior between the Florida and Antillean subspecies, we<br />
are using the ultimate evolutionary perspective. On the<br />
other hand, "how" the mother-calf pair remains together
during this period is a proximate cause question. It<br />
appears that calves remain in close association with their<br />
mothers via vocalizations (Hartman 1979; Reynolds 1981;<br />
Bengtson and Fitzgerald 1985; personal observations).<br />
Although manatees usually have only one calf at a<br />
time, there are rare occurrences of manatees giving birth to<br />
twins (Hartman 1979; O'Shea and Hartley 1995; Rathbun et<br />
al. 1995). Twinning is often followed by the death of one<br />
(O'Shea and Hartley 1995) or both (L. Lefebvre, personal<br />
communication; personal observation) offspring. From an<br />
ultimate function perspective, this alternative behavior<br />
raises the questions (1)"why" aren't females that have twins<br />
more successful than those that have singles?" and (2) "why<br />
has the variation (single vs. twins) in birthing behavior<br />
persisted within the West Indian manatee?" From an<br />
ultimate evolution perspective, the variation could be<br />
studied by looking at closely related species.<br />
Unfortunately, there are little data available on the<br />
occurrence of twins within other manatee species, but<br />
Marsh (1995) references vague reports of twin fetuses in<br />
dugongs. This is one line of evidence that twinning is a<br />
characteristic shared with other sirenians and not recently<br />
derived within the Florida population.<br />
Before we leave parental care, we should ask, "could<br />
the observation described in the introduction have been a<br />
manatee father caring for his offspring?" Probably not,<br />
there is no evidence that male manatees participate in any<br />
form of parental care. A better hypothesis, based on what<br />
we now know about manatee reproductive behavior, is that<br />
perhaps the large male was attracted to little female because<br />
she was sending an estrous signal.<br />
Summary: We have learned much about Florida<br />
manatee reproductive strategies over the past three decades.<br />
Studies on free-ranging manatee populations at warm<br />
water effluents and data collected from carcasses through<br />
the salvage network agree on many life history traits<br />
(O’Shea et al. 1995). Female manatees appear to reach<br />
sexual maturity at age 3, producing their first calf at age 4.<br />
Single births are the norm with rare cases of twins. More<br />
twins are reported in carcasses than observed in live<br />
manatees, leading to the assumption that uniparity is the<br />
more successful behavior. Gestation is about 1 year and<br />
calves stay with their mothers for about 2 years, making the<br />
minimum interval between successful reproductions about 3<br />
years. However, females who abort fetuses or lose a young<br />
calf may reproduce again sooner. Florida males exhibit<br />
seasonal spermatogenesis, which correlates with seasonal<br />
observations of females with neonate calves.<br />
PHYSICAL PROBLEM-SOLVING<br />
<strong>Sirenian</strong>s (manatees and dugongs) belong to group<br />
of animals commonly referred to as marine mammals.<br />
Other marine mammals include whales and dolphins; seals,<br />
sea lions, and walruses; sea otters; and polar bears.<br />
Although they are not closely related to each other<br />
(remember, manatees are more closely related to elephants<br />
than to other marine mammals), these groups share<br />
6<br />
convergent characteristics that evolved as they solved the<br />
physical problems associated with adaptation from a<br />
terrestrial to a marine environment. For example, all<br />
marine mammals must breathe air, and they have evolved in<br />
various ways that enable them to survive in an aquatic<br />
environment. In the dolphins, nostrils have migrated up the<br />
rostrum to the top of the head and become a single<br />
blowhole. Many large whales and seals have physiological<br />
adaptations that enable them to remain underwater for hours<br />
at a time.<br />
From an ultimate evolutionary perspective, one<br />
interesting hypothesis is that the common ancestor between<br />
manatees and elephants was an aquatic (rather than a<br />
terrestrial) mammal. If true, this would make the elephant<br />
the only known animal to move from the sea to the land (as<br />
all mammal ancestors did when their remote ancestors<br />
evolved from fish to amphibians), back to the sea (as did<br />
the cetaceans, pinnipeds, and sirenians) and then back to<br />
the land again. This idea was originally based on the<br />
thought that the elephant's trunk evolved to enable the<br />
negatively buoyant animal to breath air from beneath the<br />
surface of the water. The longer the proboscis, the deeper<br />
the animal could forage - eventually resulting in the long,<br />
snorkel-like trunk we see today. This hypothesis has<br />
recently been supported by a study on elephant embryos,<br />
which indicates that the shared ancestor between sirenians<br />
and elephants was an aquatic mammal (Gaeth et al. 1999).<br />
All marine mammals have had to solve the problem<br />
of breathing in an aquatic environment, but the sirenians<br />
have additional unique physical problems: (1) the extant<br />
species are NOT well adapted to cold water; and (2)<br />
sirenians are the only marine mammal herbivores. Only<br />
one extinct species of sirenian was found in extreme cold<br />
waters: Hydrodamalis gigas, commonly known as Steller's<br />
giant sea cow. These animals were three times as large as<br />
manatees, ranging from 25 to 35 feet in length and<br />
weighing up to 8000 pounds. Their extremely large size<br />
enabled them to survive in the frigid Bering Sea where they<br />
feed on giant kelp. All extant species of sirenians are<br />
limited to tropical and sub-tropical waters year round; some<br />
individuals migrate into temperate areas during the<br />
summer. As herbivores, all sirenians are limited to<br />
shallow coastal waters, estuaries, and rivers where aquatic<br />
vegetation is abundant. The West Indian manatee inhabits<br />
riverine and marine systems from Florida to Brazil. But<br />
their distribution is patchy, and appears to be a function of<br />
physical problem solving such as thermoregulation,<br />
foraging, predator avoidance (Hartman 1979), and<br />
osmoregulation. These are the problems we will focus on<br />
here.<br />
Thermoregulation: Water temperature is well<br />
known to be a controlling factor in Florida manatee<br />
distribution in the United States (Hartman 1979, Irvin<br />
1983). Rarely, manatees have been sighted in North<br />
Carolina (Schwartz 1995); they are routinely sighted in<br />
South Carolina and Georgia; but the Chesapeake Bay has<br />
always been considered north of any expected range - even
in the summer. Chessie, a male Florida Manatee, earned<br />
his name by showing up in the Chesapeake Bay during the<br />
summer of 1994. <strong>Sirenian</strong> biologists agreed that Chessie<br />
would die from the cold of oncoming winter if he remained<br />
in the Bay, so they flew him back to Florida, put a satellite<br />
transmitter tag on him and let him go. When the water<br />
began to warm the next summer, Chessie headed north<br />
again. Not only did he temporarily enter the Chesapeake<br />
Bay, but when he came out, he continued his northern<br />
journey up the Atlantic seaboard. Upon reaching Port<br />
Judith, Rhode Island, he finally reversed direction and<br />
began working his way back to Florida for the winter!<br />
[NOTE: For more details go to<br />
www.sirenian.org/chessie.html]<br />
Each winter, hundreds of West Indian manatees<br />
aggregate in Crystal River, Florida, where they are soon<br />
joined by thousands of humans who want to swim with<br />
them. Why does the otherwise elusive manatee tolerate this<br />
human behavior? Why do they keep coming back, year<br />
after year?<br />
These are excellent examples of how manatees have<br />
adapted behaviors to solve the physical problem of<br />
thermoregulation. While most marine mammals have<br />
adapted to cold water by evolving high metabolic rates, the<br />
West Indian manatees have exceptionally low metabolic<br />
rates (Irvin 1983). When metabolic rates are graphed<br />
against body size, most mammals fall along a predictable<br />
curve where the rate decreases as the size increases. If we<br />
compare where manatees should fall on this curve to where<br />
they actually plot out on the graph, we find that manatee<br />
metabolism is only about 20% of what we would expect.<br />
The same comparison with other marine mammals shows<br />
that their metabolic rates are almost twice what we would<br />
expect - enabling them to easily live in cold water. This<br />
physiological problem should limit the West Indian<br />
manatee to warm tropical waters.<br />
In the United States, however, the Florida<br />
subspecies thermoregulates behaviorally by migrating to<br />
both natural and artificial warm water effluents during the<br />
winter months, enabling them to extend their habitat range.<br />
From a proximate cause perspective, the migration<br />
behavior is triggered when the water temperature drops<br />
below 20 degrees Celsius (Hartman 1979; Irvin 1983). But,<br />
from a proximate development perspective, how do adult<br />
manatees know where to find warm water effluents? We<br />
touched on this earlier, when we talked about why manatee<br />
calves remain dependent on their mothers for up to 2 years.<br />
Long term studies of radio tagged females with calves<br />
indicate that manatees initially learn migration routes from<br />
their mothers during the extended parental care period (R.<br />
Bonde, personal communication). Because of this learned<br />
behavior, interruption of man-made thermal effluents may<br />
have negative impacts on manatee survival in areas where<br />
no natural warm water effluents are nearby (Packard et al.<br />
1989).<br />
What about manatees in tropical habitats where<br />
water temperatures are relatively constant? If<br />
7<br />
thermoregulation were the only function of long distance<br />
travel, we would not expect Antillean manatees to exhibit<br />
the same degree of seasonal migration as Florida manatees.<br />
Buddy Powell (personal communication 2000) is working<br />
on that ultimate evolution question through telemetry<br />
studies of Antillean manatees in Belize (see Satellite<br />
Tracking of W.I. Manatees in Belize<br />
www.wesave.org/manatee/). After almost two years of<br />
data, it appears that several manatee mother-calf pairs<br />
remain in the Southern Lagoon area year round. On the<br />
other hand, Greg Smith (personal communication) finds<br />
male manatees seasonally absent on the reef at Basil Jones,<br />
Ambergris Caye, during the winter months of December -<br />
February. In my limited personal observations, I have not<br />
observed manatees on the reef at Gallows Point Reef, near<br />
Belize City, from November through April; but they were<br />
there daily in July and August 1999. Temperature changes<br />
between summer and winter on Gallows Reef vary from<br />
about 28 <strong>–</strong> 32 degrees Celsius (unpublished data), well<br />
above the 20 degree trigger found in Florida manatees. If<br />
temperature is not driving this apparent seasonal migration,<br />
what other factors could be the cause? Greg Smith<br />
hypothesizes that it is driven by seasonal reproductive<br />
cycles. Perhaps the males "hang out" at the reef during the<br />
summer months looking for an estrous female (see<br />
previous section on reproductive problem solving). More<br />
data are required before we can answer these questions.<br />
Foraging: As we explored the bogues (channels)<br />
that snake their way through the mangrove islands off the<br />
coast of Belize, it became apparent that Antillean manatees<br />
preferred certain micro-habitats within the larger habitat<br />
we call the Drowned Cayes. We reliably found manatees<br />
just west of the cayes feeding on turtle grass beds; and we<br />
always found manatees resting in narrow bogues or quiet<br />
coves among the cayes. They tended to travel in the deeper<br />
channels when moving between areas. One of the<br />
parameters of quality manatee habitat is the close<br />
availability of food (Hartman 1979). It appears that<br />
Antillean manatees using the Drowned Cayes are more<br />
likely to rest in areas that are linked to turtle grass beds by<br />
deep channels. Like all sirenians, manatees are<br />
opportunistic herbivores, feeding on a variety of fresh and<br />
saltwater vegetation. Although they may consume fish in<br />
some areas (Powell 1978) and incidentally ingest<br />
invertebrates (Powell 1978; Powell 1984; personal<br />
observations), the main component of their diet is aquatic<br />
vegetation: sea grasses in the marine and estuarine<br />
environment; floating, submerged, and emergent plants in<br />
the riverine environment. Sea grasses, like all plants,<br />
require sunlight for growth, which limits their presence to<br />
relatively shallow coastal waters. From an ultimate<br />
function perspective, the relationship between sea grasses<br />
and water depth has probably prevented manatees from<br />
dispersing into deeper oceanic waters.<br />
I snorkeled up to an Antillean manatee feeding<br />
underwater just west of Swallow Caye <strong>–</strong> at first, I thought it<br />
was dead... I can still recall the adrenalin rush as options
flashed through my mind regarding what to do with a dead<br />
manatee! It was lying perfectly still on the bottom in about<br />
3 meters of water and appeared to be missing its head. As I<br />
floated closer, (heart racing) I began to hear chewing<br />
noises and realized that the manatee had buried its head<br />
into the muddy substrate and was feeding on the sea grass<br />
roots. I soon learned that this is typical of how manatees<br />
feed on the sea grass beds near the Drowned Cayes - eating<br />
both roots and leaves and probably other benthic organisms<br />
living in the mud. Looking for muddy disturbances became<br />
another method of finding the elusive manatee.<br />
If you’ve ever tried to dive down and recover a lost<br />
item in deep water, you know that you, like most mammals,<br />
are positively buoyant and must work to get and stay<br />
submerged. Manatee bones are pachyostoic -- very dense<br />
and lacking marrow -- except in the vertebrae and sternum<br />
(Odell and Reynolds 1991). Because of this, manatees are<br />
negatively buoyant and can lie on the sea bottom without<br />
exerting any energy to stay down. The less energy they<br />
use, the longer manatees can remain submerged between<br />
breaths - making feeding more efficient. Indeed, we think<br />
manatees have the ability to control the volume of air their<br />
lungs, enabling them to rise to the surface, take a breath,<br />
and return to the bottom with no noticeable effort.<br />
Manatees exhibit different problem-solving<br />
behaviors related to foraging in different habitats. In rivers,<br />
manatees are often observed feeding on floating vegetation.<br />
They use their forelimbs like we use our hands to<br />
manipulate aquatic plants towards the mouth. The large<br />
prehensile upper lip is then used to work the plants into the<br />
mouth. When I was observing Georgia and Peaches in<br />
Florida, I was fascinated to see Georgia reach her head out<br />
of the water to feed on plants growing along the shoreline<br />
of the Hontoon Dead River. At times it looked as if she<br />
was going to climb out onto the bank! In fact, Florida<br />
manatees feeding on grasses have been sighted with up to<br />
one third of their body awash (Powell 1984).<br />
Although they are considered opportunistic feeders,<br />
manatees are known to prefer certain species of plants to<br />
others (Bengtson 1981). Hunters of West African manatees<br />
use cassava to lure the animals into box traps (O’Shea<br />
1994). Other traditional hunters attract manatees by<br />
dangling a favorite flower over the water's edge to entice<br />
their approach. While West Indian manatees in Florida and<br />
Belize feed on a variety of vegetation including floating,<br />
submergent, emergent, and over-hanging vegetation, the<br />
Amazonian and the West African manatees feed primarily<br />
on surface vegetation.<br />
In Gambia, when feeding on overhanging<br />
vegetation, West African manatees grasp the leaves of<br />
branches with their lip pads near the water's surface (Powell<br />
1984). They pull the branch into the water; use their<br />
forelimbs to hold it down; and then eat the leaves - but not<br />
the woody petiole or branches. They appear to prefer the<br />
young leaves and shoots of mangroves, as they are known<br />
to return to areas where they have previously fed to crop<br />
any new growth. Amazonian manatees face an even more<br />
8<br />
challenging problem. During the rainy season, floodwaters<br />
allow manatees to literally "forage among the tree tops", but<br />
in the dry season, they are often confined to isolated lakes<br />
and pools devoid of any vegetation. Robin Best calculated<br />
that Amazonian manatees could fast for up to seven months<br />
a year - surviving on stored fat reserves they build up<br />
during the rainy season (O'Shea 1994).<br />
Daryl Domning of Howard University describes<br />
morphological variation in the rostrum deflection among<br />
manatee species and subspecies and suggests that the<br />
variation results from differential foraging behavior among<br />
sirenian species (1980). For example, the rostral deflection<br />
in both the Amazonian and the West African species is<br />
relatively less than it is in the West Indian manatee. These<br />
comparisons of foraging behavior and rostral deflection<br />
represent ultimate perspectives. Comparing foraging<br />
behaviors among the species is an example of ultimate<br />
evolution. Hypothesizing about why the rostral deflection<br />
varies among species (because of different foraging<br />
behaviors) is an example of ultimate function. From an<br />
ultimate perspective, it is interesting to compare the<br />
foraging behavior of manatees to the dugongs. Dugongs<br />
forage exclusively on submerged marine sea grasses and<br />
their rostra are significantly deflected downward when<br />
compared to manatees. The divergence in foraging<br />
behavior between the two families of Sirenia may be due to<br />
the evolution of true grasses in the Caribbean and<br />
continuously regenerating teeth in manatees. Unique to the<br />
Trichechus genus, is the fact that manatees generate new<br />
molars throughout their lifespan (Domning and Hayek<br />
1984). As older molars are worn down from the abrasive<br />
silica content in true grasses, new molars gradually move<br />
forward at the rate of a few millimeters per month. The<br />
forward most molars eventually fall out and are replaced<br />
from the rear in a horizontal manner. This would be<br />
analogous to humans continuously generating new wisdom<br />
teeth at the rate of four - 2 lower and 2 upper - every few<br />
months! Daryl Domning (1982) convincingly argues that<br />
this trait enabled manatees to out compete dugongs in the<br />
Atlantic a few million years ago. While manatees are only<br />
found in the Atlantic and continue to forage on a variety of<br />
vegetation, dugongs are only found in the Indo-Pacific and<br />
forage exclusively on marine sea grasses.<br />
Predation: Although they may have existed in the past, we<br />
know of few natural predators on modern West Indian<br />
manatees...EXCEPT for humans. Large aquatic predators<br />
(crocodiles, alligators, sharks, and hippopotamus) have<br />
been hypothesized to take the occasional small or weak<br />
animal (Odell 1982; Powell 1984). But, in one of the few<br />
documented cases, Johnson (1937, as reported in Powell<br />
1984) found that only one out of one hundred crocodiles cut<br />
open contained the remains of a manatee.<br />
Amazonian manatees, on the other hand, still have to<br />
contend with predation by aquatic and terrestrial carnivores<br />
such as jaguars, caimans, and sharks (Reynolds and Odell<br />
1991) <strong>–</strong> especially during the dry season when then are<br />
stranded by receding flood waters. Both fossil and historical
ecords indicate that manatees have been hunted both for<br />
subsistence and commercially throughout the history of<br />
humans (Lefebvre et al. 1989, Reynolds and Odell 1991).<br />
While illegal poaching still exists <strong>–</strong> especially in remote<br />
areas -- most modern predation is incidental <strong>–</strong> resulting<br />
from entanglement in fishing gear, shark nets, and water<br />
control devices, and from collisions with watercraft. From<br />
an ultimate function perspective, we may hypothesize that<br />
the reason manatees are elusive creatures is to avoid<br />
predation. In other words, those animals that inherited a<br />
natural tendency towards elusive behavior were more<br />
reproductively successful.<br />
The story of Steller’s sea cow demonstrates how<br />
quickly humans can extirpate a species, particularly when<br />
population numbers are already reduced. The sea cow was<br />
discovered and described by Georg Wilhelm Steller, a<br />
German naturalist assigned to Captain Vitus Bering during<br />
a Russian Expedition to Alaska (Steller 1988). This giant<br />
sirenian was 25-35 feet long and weighed ~8000 pounds<br />
with flukes than spanned 8 feet. Unlike modern sirenians,<br />
it lived in extremely cold waters in the North Pacific. It had<br />
no teeth, but two grooved plates - one upper and one lower<br />
with which it crunched giant sea kelp. During the summer<br />
of <strong>17</strong>41, Captain Bering set sail from Kamchatka in NE<br />
Russia with 2 ships, the St. Peter and the St. Paul. During<br />
the voyage, a storm separated his ships; Bering, Steller, and<br />
the crew of the St. Peter were shipwrecked on an unknown<br />
island (later named Bering Island). Although Bering did<br />
not survive the winter, Steller and many of the crew did. In<br />
his book, A Voyage with Bering, Steller credits their<br />
survival to the giant sea cow. Only after the crew learned<br />
to hunt the sea cow did they begin to regain the strength to<br />
repair their ship. When they returned to Kamchatka in the<br />
summer of <strong>17</strong>42, they told of the wonderful sea cow meat.<br />
New hunting expeditions were formed almost immediately<br />
and every year thereafter. The expeditions would return to<br />
Bering Island where they spent 8-9 months hunting furanimals<br />
and eating sea cow meat to survive. Indeed, many<br />
of the expeditions are reported to have wintered on Bering<br />
Island for the express purpose of collecting sea cow meat to<br />
provision their ships for the rest of their 3-4 year voyage to<br />
America. As a result, the last sea cow was reported killed<br />
in <strong>17</strong>68, only 27 years after modern humans had discovered<br />
the island and the species.<br />
Even before Russian sailors were hunting Steller’s<br />
sea cow to extinction, European buccaneers and explorers<br />
were provisioning their ships with Antillean manatee meat,<br />
which they harvested themselves or purchased from the<br />
indigenous people (Reynolds and Odell 1991). In the<br />
Panama area, it is estimated that at one time, 7-8 thousand<br />
manatees were being harvested annually. The Amazonian<br />
manatee continued to be commercially harvested for it's<br />
tough skin (which was made into leather products) into the<br />
1950's. From an ultimate function perspective, it stands to<br />
reason that manatees that behaved in an elusive manner<br />
lived to produce more offspring during the last several<br />
centuries.<br />
9<br />
The greatest documented predation on manatees<br />
today occurs incidentally in Florida due to competition for<br />
space between manatees and humans. Government<br />
agencies have been documenting manatee mortality in<br />
Florida for almost three decades. The proportion of deaths<br />
related to collisions with watercraft tends to increase each<br />
year as more and more people move to Florida. For current<br />
statistics on Florida manatee mortality, visit the FMRI and<br />
FFWCC websites at www.fmri.usf.edu/manatees.htm and<br />
www.state.fl.us/fwc/psm/manatee/manatee.htm. Even the<br />
most elusive manatee has a difficult time avoiding<br />
interaction with people as the human population and<br />
development continue to grow.<br />
Osmoregulation: Unlike the Amazonian manatee,<br />
which is endemic to the fresh waters of the Amazon River<br />
basin, and the dugong, which is only found in marine<br />
habitats, the West Indian and West African manatees<br />
appear to move freely between fresh and marine<br />
environments. Although the Florida subspecies is usually<br />
associated with fresh or brackish water, it is occasionally<br />
found far offshore in high salinity water (Reynolds and<br />
Ferguson 1984). Large amounts of barnacle growth suggest<br />
that some individuals spend prolonged periods in marine<br />
environments (Husar 1977, Hartman 1979). Antillean<br />
manatees are found year round in totally marine<br />
environment of red mangrove islands in Belize (personal<br />
observation). How does this species osmoregulate as it<br />
moves among fresh, brackish, and saltwater? Two<br />
alternative hypotheses come to mind: (1) West Indian<br />
manatees have physiological adaptations that enable them<br />
to maintain water balance and/or (2) W. I. Manatees<br />
behaviorally maintain water balance by seeking out fresh<br />
water sources in marine environments. Graham Worthy, a<br />
physiologist at Texas A&M University, and his students are<br />
working on this and proximate cause questions involving<br />
physiology. Preliminary results indicate that W. I.<br />
Manatees are “good osmoregulators regardless of the<br />
environment” (Ortiz et al. 1998).<br />
A FEW THOUGHTS…<br />
As we examine problem-solving behavior in the<br />
West Indian manatee using the ethological perspectives of<br />
cause, development, evolution, and function, we begin to<br />
realize how many questions remain un-answered about this<br />
elusive marine mammal. Although we know (from the<br />
study of other mammals) that the proximate cause of<br />
behavior is a complex interaction between internal<br />
mechanisms and external stimuli, we don't know the<br />
specific triggers for many manatee behaviors. Proximate<br />
development has not been well studied for several reasons.<br />
Free ranging female manatees with new calves tend to<br />
isolate themselves in secluded areas making behavior<br />
difficult to observe. Observations of captive raised<br />
manatees may not be indicative of normal development.<br />
The West Indian manatee is an endangered species making<br />
experimental manipulation difficult, yet extremely<br />
important to the successful rehabilitation and release of
injured manatees. For example, how will a captive raised<br />
calf learn to find warm water effluents during cold spells?<br />
Can adult manatees learn successful migration routes or<br />
must they be learned during early development? From an<br />
ultimate perspective, manatees are also quite challenging<br />
due to the lack of closely related extant species and to the<br />
sparse fossil record. But, paleo-sirenian research by Daryl<br />
Domning, and others, continues to offer insight to the<br />
evolution and function of modern manatee behaviors.<br />
Why are there fewer sirenian species today than during the<br />
past? Does the evolution of the species Homo correlate<br />
with the decline of sirenians or were other environmental<br />
factors the reason for their extinctions? Will answers to<br />
these and other ultimate questions aid in our conservation<br />
efforts?<br />
Although manatees are generally considered elusive,<br />
there are cases where they appear to be curious and actually<br />
initiate contact with humans. Likewise, many behaviors<br />
tend vary between individuals, populations, and species.<br />
Because of the variable nature of manatee behavior, we<br />
must be careful in applying what we know about the Florida<br />
population to other areas. The Florida subspecies is<br />
fortunate to have the efforts of many US citizens and<br />
agencies working toward conservation issues and manatee<br />
behavior plays and important role in making management<br />
decisions in Florida. However, West Indian manatees are<br />
considered endangered throughout their range. Continued<br />
research effort on populations in the more tropical regions<br />
of the Caribbean is necessary for decision makers in those<br />
countries to make effective management decisions.<br />
REFERENCES<br />
Alcock, John. 1998. Animal Behavior. Sixth edition.<br />
Sunderland, Massachusetts: Sinauer Associates,<br />
Inc. 640 pp.<br />
Anderson, Paul K. 1997. Shark Bay dugongs in summer. I:<br />
Lek mating. Behaviour. 134(5-6):433-462.<br />
Auil, Nicole. 1998. Belize Manatee Recovery Plan.<br />
UNDP/GEF Coastal Zone Management Project,<br />
BZE/92/G31, Belize/UNEP Caribbean<br />
Environment Programme, Kingston, Jamaica. 67<br />
pp.<br />
Bengtson, John L. 1981. Ecology of Manatees Trichechus<br />
manatus) in the St. Johns River, Florida. Ph. D.<br />
thesis. University of Minnesota.<br />
Bengtson, John L., and Shannon M. Fitzgerald. 1985.<br />
Potential role of vocalizations in West Indian<br />
manatees. Journal of Mammalogy. 66(4):816-<br />
819.<br />
Domning, Daryl P. 1980. Feeding position preference in<br />
manatees (Trichechus). Journal of Mammalogy<br />
61(3):544-547.<br />
Domning, Daryl P. 1982. Evolution of manatees: a<br />
speculative history. Journal of Paleontology.<br />
56(3):599-619.<br />
10<br />
Domning, Daryl P., and Lee-Ann Hayek. 1984. Horizontal<br />
tooth replacement in the Amazonian manatee.<br />
Mammalia 48:105-127.<br />
Drickamer, Lee C., Stephen H. Vessey, and Doug Meikle.<br />
1996. Animal Behavior: Mechanisms, Ecology,<br />
Evolution. Fourth edition. Boston: William C.<br />
Brown Publishers. 447 pp.<br />
Fischer, M. S. 1990. The unique ear of elephants and<br />
manatees (Mammalia): a phylogenetic paradox.<br />
C. R. Acad. Sci., Paris, SER. III, vol. 31, no. 4, pp.<br />
157-162.<br />
Gaeth, A. P., R. V. Short and M. B. Renfree. 1999. The<br />
developing renal, reproductive, and respiratory<br />
systems of the Africa elephant suggest an aquatic<br />
ancestry. Proceedings of the National Academy of<br />
Sciences of the United States of America<br />
96(10):5555-5558.<br />
Hartman, Daniel S. 1979. Ecology and behavior of the<br />
manatee (Trichechus manatus) in Florida. Special<br />
publication no. 5. American Society of<br />
Mammalogists. 153 pp. ISBN 0-9436-1204-7.<br />
Hernandez, Patricia, John E. Reynolds, III, Helene Marsh,<br />
and Miriam Marmontel. 1995. Age and<br />
seasonality in spermatogenesis of Florida<br />
manatees. pp. 84-95 in Thomas J. O'Shea, Bruce<br />
B. Ackerman, and H. Franklin Percival, editors.<br />
Population biology of the Florida manatee.<br />
National Biological Service Information and<br />
Technology Report 1.<br />
Husar, Sandra L. 1977. The West Indian manatee<br />
Trichechus manatus. Wildlife Research Report<br />
No. 7. U. S. Department of the Interior, Fish and<br />
Wildlife Service, Washington, D.C. 22 pp.<br />
Irvin, A. Blair. 1983. Manatee metabolism and its<br />
influence on distribution in Florida. Biological<br />
Conservation. 25:315-334.<br />
Koelsch, Jessica K. 1997. The Seasonal Occurrence and<br />
Ecology of Florida Manatees (Trichechus manatus<br />
latirostris) in Coastal Waters near Sarasota,<br />
Florida. Masters thesis. University of South<br />
Florida. Tampa, Florida.<br />
Lefebvre, Lynn W., Thomas J. O'Shea, Galen B. Rathbun,<br />
and R. C. Best. 1989. Distribution, status, and<br />
biogeography of the West Indian manatee. pp.<br />
567-610 in Biogeography of the West Indies. ed.<br />
Charles A. Woods. Gainesville, FL: Sandhill<br />
Crane Press. ISBN 0-8493-2001-1.<br />
Lehner, Philip N. 1996. Handbook of Ethological<br />
Methods, 2 nd edition. Cambridge University Press.<br />
672 pp.<br />
Maluf, N. S. R. 1995. Kidney of elephants. Anatomical<br />
Record 242(4):491-514.<br />
Marmontel, Miriam. 1995. Age and reproduction in<br />
female Florida manatees. Pages 98-119 in Thomas<br />
J. O'Shea, Bruce B. Ackerman, and H. Franklin<br />
Percival, editors. Population biology of the
Florida manatee. National Biological Service<br />
Information and Technology Report 1.<br />
Marmontel, Miriam, Thomas J. O'Shea, and S.R.<br />
Humphrey. 1990. An evaluation of bone growthlayer<br />
counts as an age determination technique in<br />
Florida manatees. National Technical Information<br />
Service, Springfield, Va. Document PB 91-<br />
103564. 104 pp.<br />
Marsh, Helene. 1995. The life history, pattern of breeding,<br />
and population dynamics of the dugong. pp. 75-<br />
83 in Thomas J. O'Shea, Bruce B. Ackerman, and<br />
H. Franklin Percival, editors. Population<br />
biology of the Florida manatee. National<br />
Biological Service Information and Technology<br />
Report 1.<br />
Martin, Paul, and Patrick Bateson. 1993. Measuring<br />
Behaviour <strong>–</strong> An introductory guide, 2 nd edition.<br />
Cambridge University Press. 222 pp.<br />
Odell, Daniel K. 1982. West Indian Manatee Trichechus<br />
manatus. in Wild Mammals of North America<br />
Biology, Management, and Economics. Joseph A.<br />
Chapman and George A. Feldhamer, editors. The<br />
Johns Hopkins University Press. Baltimore and<br />
London.<br />
O'Shea, Thomas J. 1994. Manatees. Scientific American.<br />
July 1994.<br />
O'Shea, Thomas J. Bruce B. Ackerman, and H. Franklin<br />
Percival. 1995. Population Biology of the<br />
Florida manatee. National Biological Service<br />
Information and Technology Report 1.<br />
O’Shea, Thomas J., and Wayne Hartley. 1995.<br />
Reproduction and early-age survival of manatees<br />
at Blue Spring, upper St. Johns River, Florida. pp.<br />
157-<strong>17</strong>0 in Thomas J. O'Shea, Bruce B.<br />
Ackerman, and H. Franklin Percival, editors.<br />
Population biology of the Florida manatee.<br />
National Biological Service Information and<br />
Technology Report 1.<br />
O'Shea, Thomas J. and Charles A. 'Lex' Salisbury. 1991.<br />
Belize - a last stronghold for manatees in the<br />
Caribbean. Oryx. 25(3): 156-164.<br />
Ortiz, Rudy M., Graham A. J. Worthy and Duncan S.<br />
MacKenzie. 1998. Osmoregulation in wild and<br />
captive West Indian manatees (Trichechus<br />
manatus). Physiological Zoology 71(4):449.457.<br />
Packard, Jane M., R. Kipp Frohlich, John E. Reynolds, III,<br />
and J. Ross Wilcox. 1989. Manatee response to<br />
interruption of a thermal effluent. Journal of<br />
Wildlife Management 53(3):692-700.<br />
Powell, James A. 1978. Evidence of carnivory in manatees<br />
(Trichechus manatus). Journal of Mammalogy.<br />
59(2):442.<br />
Powell, James A. 1984. Manatees in the Gambia River<br />
Basin and potential impact of the Balingho<br />
Antisalt Dam with notes on Ivory Coast, West<br />
Africa. Trip Report. Institute for Marine Studies.<br />
11<br />
University of Washington. Seattle Washington<br />
98195.<br />
Ralph, C. L., S. Young, R. Gettinger and T. J. O’Shea.<br />
1985. Does the manatee have a pineal body?<br />
Acta Zoologica 66(1):55-60.<br />
Rathbun, Galen B., James P. Reid, Robert K. Bonde, and<br />
James A. Powell. 1995. Reproduction in freeranging<br />
Florida manatees. pp. 135-156 in T.J.<br />
O'Shea, B.B. Ackerman, and H.F. Percival,<br />
editors. Population biology of the Florida<br />
manatee. National Biological Service Information<br />
and Technology Report 1.<br />
Reynolds, John E., III. 1981. Aspects of the social<br />
behaviour and herd structure of a semi-isolated<br />
colony of West Indian manatees, Trichechus<br />
manatus. Mammalia. 45(4):431-451.<br />
Reynolds, John E., III, and John C. Ferguson. 1984.<br />
Implications of the presence of manatees<br />
(Trichechus manatus) near the Dry Tortugas<br />
islands. Florida Scientist 47(3):187-189.<br />
Reynolds, John E. III, and Daniel K. Odell. 1991.<br />
Manatees and Dugongs. New York: Facts on File,<br />
Inc. 192 pp. ISBN 0-8160-2436-7.<br />
Schwartz, F. J. 1995. Florida manatees, Trichechus<br />
manatus (Sirenia: Trichechidae) in North Carolina<br />
1919-1994. Brimleyana 22:53-60.<br />
Springer, M. S., G. C. Cleven, O. Madsen, W. W. De Jong,<br />
V. G. Waddell, H. M. Amrine, M. J. Stanhope.<br />
1997. Endemic African mammals shake the<br />
phylogenetic tree. Nature (London) 388<br />
(6637):61-64.<br />
Steller, Georg Wilhem. 1988. Journal of a Voyage with<br />
Bering <strong>17</strong>41-<strong>17</strong>42. Stanford, California: Stanford<br />
University Press. 252 pp.<br />
Stejneger, Leonhard. 1887. How the great northern seacow<br />
(Rytina) became exterminated. The American<br />
Naturalist 21(12):1047-1054.<br />
RECOMMENDED READING<br />
Caldwell, David K., and Melba C. Caldwell. 1985.<br />
Manatees - Trichechus manatus, Trichechus<br />
senegalensis, and Trichechus inunguis. pp. 33-36<br />
in Handbook of Marine Mammals, vol. 3, The<br />
<strong>Sirenian</strong>s and Baleen Whales. eds. S. H.<br />
Ridgway and R. J. Harrison. London: Academic<br />
Press. ISBN 0-1258-8503-2.<br />
Hartman, Daniel S. 1979. Ecology and behavior of the<br />
manatee (Trichechus manatus) in Florida. Special<br />
Publication no. 5. American Society of<br />
Mammalogists. 153 pp. ISBN 0-9436-1204-7.<br />
Reynolds, John E. III, and Daniel K. Odell. 1991.<br />
Manatees and Dugongs. New York: Facts on File,<br />
Inc. 192 pp. ISBN 0-8160-2436-7.
Zeiller, Warren. 1992. Introducing the Manatee.<br />
Gainesville, Florida: University Press of Florida.<br />
151 pp. ISBN 0-8130-1152-3.<br />
World Wide Web Internet URLs:<br />
Animal Behavior by Dr. Jane M. Packard:<br />
www.tamu.edu/ethology/<br />
Call of the Siren by Caryn Self Sullivan:<br />
www.sirenian.org/caryn.html<br />
The Nobel Prize in Physiology or Medicine 1973 <strong>–</strong> Karl<br />
von Frisch, Konrad Lorenz, Nikolaas Tinbergen, for their<br />
discoveries concerning organization and elicitation of<br />
individual and social behaviour patterns:<br />
www.nobel.se/medicine/laureates/1973/index.html<br />
Satellite Tracking West Indian Manatees in Belize:<br />
www.wesave.org/manatee/<br />
Florida Fish and Wildlife Conservation Commission:<br />
www.state.fl.us/fwc/psm/manatee/manatee.htm<br />
Florida Marine Research Institute:<br />
www.fmri.usf.edu/manatees.htm<br />
USGS Sirenia Project:<br />
www.fcsc.usgs.gov/sirenia/<br />
Sea World Education Department:<br />
www.seaworld.org/manatee/manatees.html<br />
Florida Power & Light Company:<br />
www.dep.state.fl.us/psm/webpages/manatees/booklet.html<br />
University of Wisconsin - The Brain of the Florida<br />
Manatee:<br />
www.neurophys.wisc.edu/manatee/<br />
University of Florida - Manatee Research Group:<br />
www.vetmed.ufl.edu/ufmrg/manatee/<br />
GLOSSARY<br />
agonistic: a general term that includes aggressive,<br />
submissive, and defensive behaviors that appear when the<br />
adrenal hormones are activated.<br />
anecdote: a short story or observation which has not been<br />
"tested" by the scientific method.<br />
anthropocentric: considering human beings as the most<br />
significant entity of the universe; interpreting or regarding<br />
the world in terms of human values and experiences.<br />
anthropogenic: of, relating to, or resulting from the<br />
influence of human beings on nature.<br />
anthropomorphic: described or thought of as having a<br />
human form or human attributes; ascribing human<br />
characteristics to nonhuman things.<br />
bogue: the local term for a channel of water that flows<br />
through a mangrove caye in Belize, C.A.<br />
cause: stimulus outside the animal plus the internal<br />
physiological state of the animal.<br />
Cetacea: the order of aquatic mostly marine mammals that<br />
include the whales, dolphins, porpoises, and related forms<br />
and that have a torpedo-shaped nearly hairless body,<br />
paddle-shaped forelimbs but no hind limbs, one or two<br />
nares opening externally at the top of the head, and a<br />
horizontally flattened tail used for locomotion.<br />
convergent: similar traits in species from very different<br />
genetic lineages, due to similar environmental functions.<br />
12<br />
development: changes in behavioral traits as an individual<br />
ages.<br />
divergent: different traits in species from similar genetic<br />
lineages due to differences in their environments.<br />
endemic: restricted or peculiar to a locality or region.<br />
Amazonian manatees are only found in the Amazon River<br />
Basin.<br />
ephemeral: short period of existence; opposite of eternal.<br />
estrous: the period during which a female has produced an<br />
egg ready to fertilize and is receptive to copulation.<br />
evolution: change in proportion of genotypes within the<br />
gene pool of a population over many many generations.<br />
extant: currently or actually existing; not destroyed or lost.<br />
extinct: no longer existing.<br />
extirpated: to destroy completely.<br />
folk psychology: a non-scientific way of talking about<br />
animal behavior. Uses human terms to describe non-human<br />
animal behavior. Intuitive <strong>–</strong> based on anecdotes,<br />
experiences, observations. Anthropomorphic. Assumes<br />
that animals have beliefs, feelings, desires.<br />
function: the meaning of a behavior in terms of survival<br />
and reproduction in a given environment.<br />
gestation: the carrying of young in the uterus.<br />
herbivores: animals that feed exclusively on plants.<br />
individual: a single organism; a way of understanding the<br />
biological hierarchy of concepts - those pertaining to<br />
physiological processes and experiences of each organism;<br />
in contrast to population, which focuses more on gene<br />
pools of groups of animals that interbreed.<br />
lactation: the secretion of milk.<br />
manatee hole: concave depression in sandy or muddy<br />
substrate where manatees habitually rest.<br />
mating herd: the name given an estrous female and the<br />
aggregation of males which follow her around attempting to<br />
mate.<br />
mating system: males strategies and female strategies as<br />
observed in a population.<br />
morphology: the form and structure of an organism or any<br />
of its parts.<br />
natural selection: the process that produces evolutionary<br />
changes in a population due to heritable traits in some<br />
individuals which result in those individuals being more<br />
reproductively successful; the process only occurs IF (1)<br />
there are variances in traits within a population, (2) the<br />
variances are heritable, (3) individuals with certain<br />
variances have greater reproductive success than<br />
individuals with other variances, then over time, we would<br />
expect to see the trait selected "for" become more common<br />
in the population.<br />
neonate: term used distinguish a newborn cetacean or<br />
sirenian calf from an older calf <strong>–</strong> you can still see the fetal<br />
folds on a neonate and the skin is usually darker than on an<br />
older calf.<br />
osmoregulation: regulation of osmotic pressure especially<br />
in the body of a living organism. how an organism<br />
regulates the amount of water entering and leaving its body.
parturition: the action or process of giving birth to<br />
offspring; birth, whelping.<br />
pinniped: any of a suborder (Pinnipedia) of aquatic<br />
carnivorous mammals (as a seal or walrus) with all four<br />
limbs modified into flippers.<br />
polyandry: a mating system where one female mates with<br />
two or more males at a time. (adj: polyandrous)<br />
population: a group of animals that interact and interbreed.<br />
postpardum: following parturition; after giving birth.<br />
prehensile: an appendage adapted for seizing or grasping<br />
especially by wrapping around - examples: a monkey has a<br />
prehensile tail and an elephant has a prehensile trunk.<br />
proboscis: the trunk of an elephant; also any long flexible<br />
snout; any of various elongated processes of the oral region<br />
of an invertebrate.<br />
promiscuity: a mating system where there is no prolonged<br />
association between the mating pair and at least one sex<br />
engages in multiple mates. (adj: promiscuous)<br />
proximate: immediately preceding or following as in a<br />
chain of events, causes, or effects; in ethology - the<br />
perspective that looks at the cause and development of<br />
behavior at the individual animal level.<br />
13<br />
scramble-polygyny: a mating system where many males<br />
try to mate with many females, but without overt<br />
competition among themselves.<br />
Sirenia: the taxonomic order of aquatic herbivorous<br />
mammals including the manatee, dugong, and Steller's sea<br />
cow. <strong>Sirenian</strong>s are members of the Order Sirenia.<br />
social animals: tending to form cooperative and<br />
interdependent relationships with others of one's kind;<br />
living and breeding in more or less organized communities.<br />
taxonomy: orderly classification of plants and animals<br />
according to their presumed natural relationships; the study<br />
of the general principles of scientific classification;<br />
systematics.<br />
thermoregulate: the maintenance or regulation of<br />
temperature; specifically the maintenance of a particular<br />
temperature of the living body.<br />
ultimate: most remote in space or time; last in a<br />
progression or series; in ethology - the perspectives of<br />
evolution and function that look at behavior at the<br />
population and/or species level.<br />
uniparous: producing one offspring at a time.<br />
ACKNOWLEDGMENTS<br />
Financial support for my research on manatees comes from the National Science Foundation Graduate Fellowship<br />
Program (1998-2001), the Oceanic Society of San Francisco (1998-1999), the Lerner-Gray Fund for Marine Research at the<br />
American Museum of Natural History (2000), and Earthwatch Institute (2001). A special thank you goes out to all the<br />
volunteers who have participated in field research with me in Belize. Many sirenian researchers from around the world have<br />
been quite generous in answering the multitude of questions I asked of them. Special thanks go to Jane Packard and Bill<br />
Evans of Texas A&M University, Bob Bonde of the USGS Sirenia Project, and Daryl Domning of Howard University, for<br />
their perpetual availability via email. I thank Nicole Auil of the Belize Coastal Zone Management Institute and Authority;<br />
Sidney Turton, Elda Cabellos, and the staff of Spanish Bay Resort; Greg Smith of the Belize National Manatee Working<br />
Group (BNMWG); and Buddy Powell of the Florida Marine Research Institute & the BNMWG for their continuing support.<br />
Most importantly, I offer an extra special thank you to my primary boat operator, Armando "Patch" Muñoz, for his hours of<br />
tolerance and patience as we searched for the Elusive Manatee among the Drowned Cayes in Belize. Many others have<br />
helped me along the way by taking me into the field or assisting me in the field, sharing their personal knowledge and<br />
experiences, donating slides to my educational programs, directing me towards the proper references, and just by being there<br />
to listen. In no particular order, they include: Maxine Monsanto, Heidi Petersen, Kate Schafer, Eleno “Landy” Requena,<br />
Kecia Kerr, Barbara Bilgre, Katie LaCommare, Leszek Karcmarski, Birgit Winning, Jaime Gilardi, Beth Wright, Lenisa<br />
Tipton, Tami* Gilbertson, Jessica Koelsch, Graham Worthy, Angela Garcia-Rodriguez, Mike Bragg, Edmund Gerstein, Rich<br />
Harris, Lizz Singh, Chocolate Heredia, Hans Rothauscher, Joe Olson, Patti Thompson, Bruce Ackerman, Cathy Beck, Chip<br />
Deutsch, Lynn Lefebvre, Tom O'Shea, Tom Pitchford, Roger Reep, Wally Welker, John Reynolds, Butch Rommel, Peter<br />
Tyack, Andrea Gill, Kevin Andrewin, Elaine Perez, Rob Young, and my extended family of relatives and friends.<br />
Want to learn more about manatees? Visit my homepage: www.sirenian.org/caryn.html
Aquatic Mammals 2003, 29.3, 342<strong>–</strong>354<br />
Seasonal occurrence of male Antillean manatees (Trichechus manatus<br />
manatus) on the Belize Barrier Reef<br />
Caryn Self-Sullivan 1,2 , Gregory W. Smith 3 , Jane M. Packard 1,2 and<br />
Katherine S. LaCommare 2,3<br />
1 Texas A&M University, Department of Wildlife & Fisheries Sciences, Mail Stop 2258, College Station,<br />
Texas 77843<strong>–</strong>2258, USA<br />
2 <strong>Sirenian</strong> <strong>International</strong>, Inc., 200 Stonewall Drive, Fredericksburg, Virginia 22401<strong>–</strong>2110, USA<br />
3 P.O. Box 142, San Pedro Town, Belize, Central America<br />
4 University of Massachusetts-Boston, Department of Biology, Boston, Massachusetts 02125, USA<br />
Abstract<br />
A fragment of manatee habitat that crosses the<br />
border of Belize and Mexico includes both activity<br />
centres and travel routes linking rivers, lagoons,<br />
seagrass beds and mangrove islands near Chetumal<br />
Bay. Little is known about how geophysical<br />
features like coral reefs may in uence manatee<br />
movements within and between habitat fragments<br />
like this. In this inductive study (1995<strong>–</strong>2001), we<br />
documented the seasonal occurrence of Antillean<br />
manatees at breaks in the northern Belize Barrier<br />
Reef. Survey locations were at: (1) Bacalar Chico<br />
National Park and Marine Reserve on Ambergris<br />
Caye (Basil Jones Cut) and (2) breaks in the reef<br />
70 km south near the Drowned Cayes (Gallows<br />
Reef). The probability of sighting at least one<br />
manatee on a 20-min point scan survey was 40%<br />
(n=382). Sighting probability was signi cantly<br />
higher during the summer season (May<strong>–</strong>August)<br />
compared to winter months (December<strong>–</strong>March).<br />
Group size ranged from one to ve manatees,<br />
peaking earlier (May) at the northern than southern<br />
site (August). Seventeen identi able individuals<br />
accounted for 87% of the sightings at Basil Jones<br />
Cut, with re-sightings within and between years.<br />
One individual from Basil Jones Cut was re-sighted<br />
at Gallows Reef. Of the manatees for which sex was<br />
determined, 100% were males. No calves were<br />
sighted. To better understand manatee activity<br />
centres and travel routes, we identi ed potential<br />
hypotheses relating seasonal in uences, stopover<br />
sites for travelling males, and habitat connectivity.<br />
To protect this highly vulnerable species, we recommend<br />
inclusion of the Belize Barrier Reef as an<br />
important component of manatee habitat within<br />
the coastal zone of Belize.<br />
Key words: Antillean manatee, Trichechus manatus<br />
manatus, Caribbean, Belize, habitat connectivity,<br />
? 2003 EAAM<br />
stopover sites, coastal zone management, Belize<br />
Barrier Reef, fragmented populations, seasonal<br />
habitat use.<br />
Introduction<br />
A sub-species of the West Indian manatee, the<br />
Antillean manatee (Trichechus manatus manatus)<br />
occurs in rivers and coastal marine systems of at<br />
least 19 countries in the Wider Caribbean Region,<br />
including the Greater Antilles, Mexico, Central<br />
America, and South America (CEP/UNEP, 1995;<br />
Lefebvre et al., 2001). It is listed by the IUCN<br />
(Hilton-Taylor, 2001) as vulnerable, in continuing<br />
decline, with severely fragmented populations (VU<br />
A1cd, C2a), and has been identi ed as one of the<br />
‘priority protected species of regional concern’<br />
(CEP/UNEP, 1995 pp. 1).<br />
One population of this focal species spans the<br />
border of Belize and Mexico, a diverse habitat<br />
including Chetumal Bay and the northern Belize<br />
Barrier Reef Lagoon System (Fig. 1). With a<br />
relatively short coastline extending from the Gulf<br />
of Honduras in the south to Chetumal Bay in<br />
the north, Belize reports the largest number of<br />
Antillean manatees in the Caribbean region<br />
(O’Shea & Salisbury, 1991). The contiguous<br />
Chetumal Bay is one of the most important areas<br />
for manatees in Mexico (Morales et al., 2000) and<br />
Northern Belize (Auil, 1998). Primary habitat consists<br />
of rivers, coastal lagoons and bays, and mangrove<br />
islands between the Belize Barrier Reef and<br />
the mainland (Bengtson & Magor, 1979; O’Shea &<br />
Salisbury 1991; Gibson, 1995; Auil, 1998; Morales-<br />
Vela et al., 2000). OVshore atoll systems, such as<br />
TurneVe Atoll, are considered secondary habitat for<br />
manatees (Gibson, 1995; Auil, 1998; Morales-Vela<br />
et al., 2000).
Manatees on the Belize Barrier Reef<br />
Figure 1. Northern Belize Barrier Reef Lagoon System and Southern Chetumal Bay showing survey<br />
locations and the 100-fathom line (100 fathoms=183 m, map modi ed from Purdy et al., 1975).<br />
343
344 C. Self-Sullivan et al.<br />
As identi ed by Packard & Wetterqvist (1986)<br />
in Florida, the components of manatee habitat<br />
systems include activity centres, travel routes,<br />
resources for expansion (potential feeding areas for<br />
recovering populations), essential areas (necessary<br />
to survive seasonal extremes), and the supporting<br />
ecosystem. For the purpose of the present study, we<br />
focused on the former two. We de ned activity<br />
centres as areas where manatees were frequently<br />
observed using resources such as vegetation and<br />
freshwater, in all seasons. For Belize, previous<br />
studies (Gibson, 1995; Lefebvre et al., 2001) indicated<br />
activity centres were located in silty substrates<br />
at 1<strong>–</strong>3 m depth. Activity centres could have<br />
attracted both residents and travellers (Reid et al.,<br />
1991; Koelsch, 1997). We de ned travel routes as<br />
connections between activity centres used by individual<br />
manatees for daily, seasonal, or migratory<br />
movements (Sanderson, 1966). We were interested<br />
in whether geophysical characteristics such as a reef<br />
could have been used as landmarks by traveling<br />
manatees.<br />
Traditional knowledge in local communities indicated<br />
manatees were often sighted on the Belize<br />
Barrier Reef in the summer. However, the reef had<br />
not been included in descriptions of manatee habitat<br />
in Belize, possibly due to a bias from research<br />
done in Florida where reefs were not present in<br />
areas where manatees had been studied. Only one<br />
section of Belize Barrier Reef was included in<br />
national standardized surveys i.e., in the north<br />
where it lies within several 100 m of Ambergris<br />
Caye and has been classi ed as the caye habitat<br />
type (Auil, 1998). As recommended by Weeks &<br />
Packard (1997), we listened to local residents and<br />
chose to document in a systematic manner the<br />
patterns of manatee occurrence that they perceived.<br />
Using an inductive approach, we examined whether<br />
the seasonal trend was robust, and explored the<br />
reasons that this component of the habitat system<br />
would be most attractive to manatees in the<br />
summer.<br />
In this paper, we describe the seasonal occurrence<br />
of predominately male manatees at two locations<br />
along the Belize Barrier Reef: (1) Basil Jones Cut,<br />
which was isolated by Ambergris Caye from a<br />
manatee activity centre in Chetumal Bay, and (2)<br />
Gallows Reef, which was adjacent to a manatee<br />
activity centre in the Belize Barrier Reef lagoon<br />
system (the Drowned Cayes) and exposed to frequent<br />
boat traYc. The study was initiated at Basil<br />
Jones Cut and extended to Gallows Reef. If<br />
manatee presence on the reef was in uenced by<br />
access to warm or fresh water during the winter, we<br />
expected to nd manatees at Gallows Reef when<br />
they were not at Basil Jones. Similarly, if the<br />
predominance of males was in uenced by isolation<br />
from an activity centre, we expected to nd female<br />
manatees and calves at Gallows Reef when they<br />
were not at Basil Jones.<br />
Materials and Methods<br />
Study site<br />
The Belize Barrier Reef extends from the Mexican<br />
border, where it is only a few meters from the coast,<br />
to the Gulf of Honduras where it is 50 km oVshore<br />
(Purdy et al., 1975). It forms an important geophysical<br />
barrier between the shallow coastal lagoon<br />
system and the deep Caribbean Sea. The Belize<br />
Barrier Reef, an extensive fringing and barrier<br />
reef, developed along an escarpment that abruptly<br />
terminates the 250 km-long Belize continental shelf;<br />
the sea oor plunges to over 183 m (100 fathoms)<br />
just beyond the reef crest (Fig. 1).<br />
In this tropical area of the Caribbean, seasons are<br />
less de ned by temperature and more by rainfall.<br />
The average air temperature ranges from 24 C in<br />
November<strong>–</strong>January, to 27 C in May<strong>–</strong>September<br />
(Purdy et al., 1975). The dry season extends from<br />
February through May; the rainy season extends<br />
from June through November (corresponding to<br />
the peak probability of hurricanes in July<strong>–</strong>October);<br />
December and January are referred to as the transition<br />
season (Auil, 1998). Average annual rainfall<br />
increases in a north to south direction, with 124 cm<br />
near Chetumal Bay, <strong>17</strong>8 cm at Belize City, and<br />
380 cm near the Gulf of Honduras (Purdy et al.,<br />
1975).<br />
The two sampling locations are approximately<br />
70 km apart (Fig. 1). These locations diVer substantially<br />
in both geophysical characteristics and human<br />
activity, as described in more detail below. The<br />
northern location, Basil Jones, is relatively far from<br />
boat traYc centres and manatee resources (abundant<br />
seagrass beds, freshwater, deep channels).<br />
Both locations could provide shelter from surf<br />
surge, with areas suitable for resting and socializing<br />
with other manatees. Compared to Gallows Reef,<br />
seagrass beds appear sparser near Basil Jones.<br />
Northern location<br />
Basil Jones Cut (Fig. 2a), is a few hundred<br />
metres east of Ambergris Caye (18 5 38 N,<br />
87 52 12 W). Inside the Bacalar Chico National<br />
Park and Marine Reserve, this cut is one of several<br />
dozen small breaks in a 50 km, continuous section<br />
of Belize Barrier Reef that hugs the windward shore<br />
(Purdy et al., 1975). Basil Jones Cut (>3 m) is used<br />
by powerboats travelling to a shrimp hatchery,<br />
about a dozen local residents, and a few shermen<br />
or tour operators. The reef lagoon is narrow<br />
(
(a)<br />
(b)<br />
Manatees on the Belize Barrier Reef<br />
Figure 2. Aerial photographs of (a) the northern survey location at Basil Jones Cut (manatees were<br />
observed resting in the deep water channel, i.e. the darker water in the photo), and (b) the southern<br />
survey location east of the Drowned Cayes (Gallows Reef is located along the right margin of<br />
photograph; mainland Belize is approximately 15 km to the west). Photographs by Jimmie C. Smith.<br />
345
346 C. Self-Sullivan et al.<br />
Caye is a solid landmass that blocks manatee travel<br />
from Belize Barrier Reef to Chetumal Bay, a core<br />
centre of manatee activity (Morales-Vela et al.,<br />
2000). Manatees at Basil Jones Cut could travel to<br />
activity centres via two routes (Fig. 1): (1) about<br />
7 km north via Bacalar Chico, a secluded narrow<br />
canal that connects Chetumal Bay to the Caribbean<br />
Sea along the border between Belize and Mexico, or<br />
(2) about 30 km south, around the southern tip of<br />
Ambergris Caye where San Pedro Town is located<br />
(a highly developed tourist destination).<br />
Southern location<br />
Gallows Reef (Fig. 2b), is a section of Belize Barrier<br />
Reef with two breaks: North Gallows Cut<br />
(<strong>17</strong> 30 32 N, 88 3 4 W) and South Gallows Cut<br />
(<strong>17</strong> 27 25 N, 88 2 <strong>17</strong> W). This central section of<br />
the Belize Barrier Reef is discontinuous, with large<br />
breaks in the reef crest, which provide many connections<br />
between deep water and the Belize Barrier<br />
Reef lagoon. Gallows Reef is about 2 km east of the<br />
Drowned Cayes (Fig. 1), an area of mangrove<br />
islands and associated seagrass beds used by<br />
manatees in both winter and summer (Auil, 1998;<br />
Sullivan et al., 1999; LaCommare et al., 2001).<br />
About 15 km due east of Belize City, this string of<br />
islands provides potential navigational ‘stepping<br />
stones’ from the reef to the Belize River, a longterm<br />
manatee activity centre identi ed from<br />
modern aerial survey data (Auil, 1998) and prehistoric<br />
archaeological data (McKillop, 1984).<br />
Throughout this shallow coastal zone, boat traYc is<br />
frequent, including shing boats, tugboats pulling<br />
sugar barges, recreational boats, tour boats, and<br />
water taxis. English Channel, a deep-water shipping<br />
route into the major port at Belize City, is used by<br />
cargo ships, tankers, and cruise ships. Cruise ships<br />
and sugar ships, which are too large for the port,<br />
have berths within the barrier reef lagoon. Small<br />
fast boats transport tourists in all directions from<br />
the cruise ships and Belize City, including welltravelled<br />
routes through the cuts at Gallows Reef<br />
to TurneVe Atoll. Tugboats tow barges of sugar<br />
from points north to temporary sites within the<br />
Drowned Cayes and then to the sugar ship.<br />
Sampling methods<br />
Based on year-round eVort at Basil Jones Cut,<br />
seasonal periods representing winter (December<br />
through March) and summer (June through<br />
August) were chosen for eYcient allocation of<br />
sampling eVort at Gallows Reef. At Basil Jones<br />
Cut, opportunistic observations of manatees by a<br />
local resident (the second author) began prior to<br />
1995 and extended beyond 1997 (Smith, 2000). A<br />
2-year period (April 1995<strong>–</strong>March 1997) of consistent<br />
eVort was selected for the purpose of the present<br />
analysis (Fig. 3). Preliminary studies indicated that<br />
manatees were not present in December-February;<br />
hence, more eVort was allocated to summer<br />
months. To determine whether manatees from Basil<br />
Jones Cut were re-sighted further south, surveys<br />
were extended to Gallows Reef during a study of<br />
the Drowned Cayes by the primary author (Sullivan<br />
et al., 1999). Sampling eVort was limited to winter<br />
(December<strong>–</strong>March) and summer (June<strong>–</strong>August) at<br />
Gallows Reef (1999<strong>–</strong>2001).<br />
Observation and recording procedures were<br />
similar at both locations, following a protocol that<br />
minimized disturbance by in-water observers<br />
(Smith, 2000). Each sample consisted of a 20-min<br />
continuous scan (Lehner, 1996) around a xed<br />
survey point. One to two snorkellers continuously<br />
scanned 360 around the survey point, while oating<br />
at the water’s surface. All samples were collected<br />
between 0800 and 1600 h local time. Only one<br />
sample was taken at each survey point on any given<br />
day at Gallows Reef; no more than two samples<br />
(one in the morning about 1000 h and one in the<br />
afternoon about 1600 h) were taken at the survey<br />
point at Basil Jones Cut.<br />
To determine sighting probability, any manatee<br />
observed during the scan, was recorded as a<br />
‘sighting’. If a manatee approached after the end of<br />
the scan, it was not recorded as a sighting although<br />
it could have been photographed, if feasible.<br />
‘Group size’ was recorded as the total number of<br />
individual manatees present in a scan. In other<br />
words the number of sightings within a 20-min scan<br />
was limited (0, 1); group size was unlimited (0, 1, 2,<br />
3, . . ., N). Group sizes greater than 1 were recorded<br />
only if more than one manatee was observed<br />
simultaneously or if sequential observations were of<br />
uniquely marked individuals. Sketches and photographs<br />
were used to record individual identities at<br />
Basil Jones Cut; photographs and video tapes were<br />
used to record individual identities at Gallows<br />
Reef. When visibility was too poor for positive<br />
identi cation of an individual, it was recorded as an<br />
‘unknown’.<br />
Behaviours (resting, feeding, socializing, milling,<br />
and travelling), body size (calf, adult) and gender<br />
(male, female) were recorded when possible.<br />
De nitions of behavioural activities were: (a) when<br />
‘resting’, the manatee was stationary, either in contact<br />
with the sea oor or at mid-water, occasionally<br />
rising in the vertical direction for breaths, but with<br />
no horizontal movement, no rooting or chewing,<br />
and no reaction to observer, (b) when ‘feeding’, the<br />
manatee was rooting or chewing in a seagrass bed<br />
or had seagrass parts trailing from it’s mouth when<br />
it rose above the bottom, (c) when ‘socializing’, one<br />
manatee touched or followed another, (d) when<br />
‘milling’, the direction of movement changed both<br />
vertically and horizontally with no consistent orientation<br />
to other manatees, to food, or in any one
direction, and (e) when ‘travelling’, the manatee was<br />
moving horizontally in one consistent direction,<br />
either towards or away from the survey point.<br />
To aid in interpretation of results at Gallows<br />
Reef, additional data were collected regarding the<br />
context of samples. As a check on visibility bias,<br />
assistants on a boat (anchored at the survey point)<br />
recorded surface behaviours of manatees relative to<br />
the in-water observer(s). Environmental measurements<br />
collected immediately following the sample<br />
included: (1) sea-surface water temperature using a<br />
thermometer (analogue or digital), (2) sea-surface<br />
salinity using a refractometer, and (3) vertical<br />
visibility using an eight-inch Secchi disk. In no<br />
instance was a manatee sighted by above-water<br />
assistants that was not also sighted by the in-water<br />
observer(s).<br />
The analyses were designed to account for diVerences<br />
in sampling eVort at Basil Jones (n=336) and<br />
Gallows Reef (n=45). Independent variables were<br />
season (winter, summer) at Gallows Reef and<br />
month (January through December) at Basil Jones.<br />
Dependent variables at both locations included: (a)<br />
group size, (b) frequency of surveys with manatees<br />
present or absent. Environmental measures of sea<br />
Manatees on the Belize Barrier Reef<br />
Figure 3. Individual manatee sightings at Basil Jones (April 1995<strong>–</strong>March 1997). Black boxes indicate the manatee was<br />
sighted at least once during the month; seasonal code indicates dry (white), rainy (dark grey), and transitional (light grey)<br />
months. Monthly probability of sighting (S/T) is de ned as the number of surveys that manatees were present, divided<br />
by the total number of surveys for each month. Note: BJ14 (an unmarked male) is included with unknowns in this gure.<br />
surface temperature, salinity, and visibility were<br />
also analysed at Gallows Reef. Non-parametric<br />
statistical tests of continuous variables included<br />
the Mann<strong>–</strong>Whitney U, and the Kruskal<strong>–</strong>Wallis<br />
(Lehner, 1996). Contingencies were tested using the<br />
Fisher’s exact test and the Freeman<strong>–</strong>Tukey deviate<br />
(Bishop et al., 1975).<br />
Results<br />
347<br />
Manatees were documented at both locations on<br />
the Belize Barrier Reef during the summer months,<br />
but not during the winter months. On 40% of all<br />
surveys, at least one manatee was sighted. As<br />
follows, analyses were speci c to each location.<br />
Basil Jones Cut<br />
At least one manatee was sighted on 42% of the 336<br />
surveys at Basil Jones Cut (Fig. 3), and mean group<br />
size varied signi cantly among survey months<br />
(Kruskal<strong>–</strong>Wallis H=119, df=6, P
348 C. Self-Sullivan et al.<br />
Figure 4. At Basil Jones, group size varied signi cantly by<br />
month (1995<strong>–</strong>1997). Horizontal bars indicate means and<br />
circles indicate the range of values.<br />
were present at Basil Jones during the rainy season<br />
(June through November) and absent from<br />
December through March (Fig. 3). They returned<br />
at the end of the dry season (April/May).<br />
Manatees stayed in the calm, deep water of the<br />
channel inside the reef crest. The primary activity<br />
was resting, with occasional socializing. During one<br />
sighting two manatees were observed with seagrass<br />
trailing from their mouths. During another sighting,<br />
two manatees were observed rooting in an area<br />
of sparse seagrass approximately 20 m east of the<br />
resting area. On one occasion, a bull shark was<br />
observed with the manatees, with no interaction.<br />
Travel into the shallower areas of the lagoon system<br />
was rare. Trends in direction of travel by manatees<br />
outside the reef crest could not be determined.<br />
For 87% of the manatees sighted, identity was<br />
determined (Fig. 3). Seventeen individual manatees<br />
had unique markings and were documented by<br />
sketches; most were photographed at least once.<br />
Fifteen of these individuals were observed to be<br />
males. No calves or females were sighted; the sex of<br />
only two identi able individuals was undetermined.<br />
The sex of unknowns (13% of the sightings for<br />
which individual identity could not be determined)<br />
was undetermined in most cases.<br />
Group composition was uid, with no detectable<br />
long-term associations among individuals (Fig. 3).<br />
Two males (BJ01 and BJ08) were re-sighted in each<br />
of 3 years sampled. All seven males that were rst<br />
identi ed in 1995 were re-sighted the next year. Ten<br />
individuals, eight males and two undetermined,<br />
were newly identi ed in 1996. The pattern of<br />
re-sightings within each year was variable; some<br />
individuals were present only one month each<br />
season, others departed and returned after a break<br />
Figure 5. At Gallows Reef, manatees were absent on all<br />
18 winter surveys and present on 12 of 27 summer<br />
surveys.<br />
of 1<strong>–</strong>3 months. One individual ‘White Patch’ (BJ01)<br />
came and went regularly (1994<strong>–</strong>1997), being sighted<br />
<strong>17</strong> times over seven consecutive months in 1996.<br />
The most frequently sighted individual (BJ03) was<br />
present in 41 surveys between April and August<br />
1996; however, he was not re-sighted again until<br />
June of 1998.<br />
Gallows Reef<br />
At least one manatee was sighted on 27% of the 45<br />
surveys at Gallows Reef (Fig. 5). Presence diVered<br />
signi cantly between seasons (Fisher’s exact<br />
phi=0.492, df=1, P=0.001). Manatees were absent<br />
on 100% of surveys during the winter season<br />
(Freeman<strong>–</strong>Tukey deviate z= "3.494) and present<br />
during 44% of the surveys in the summer season<br />
(Freeman<strong>–</strong>Tukey deviate z=1.611). Group size<br />
ranged from one to three, with the larger groups<br />
occurring only in late July and August. ‘White<br />
Patch’ (BJ01) who was frequently re-sighted at Basil<br />
Jones, was re-sighted at Gallows Reef in 1999.<br />
Sex was determined to be male for 10 of the <strong>17</strong><br />
manatees observed during all sightings. No calves<br />
or females were sighted, although the sex of seven<br />
manatees was undetermined.<br />
In general, manatees approached the in-water<br />
observer, paused momentarily and then retreated<br />
beyond visible range, remaining within view of<br />
assistants in the boat. On several occasions, the<br />
same manatee approached the in-water observer<br />
and retreated multiple times during a 20-min scan.<br />
On two occasions, more than one identi able<br />
manatee approached the in-water observer, both<br />
simultaneously and sequentially. Only once did a<br />
second individual approach the in-water observer<br />
after the end of the scan. The primary activity<br />
was milling, with occasional socializing. Rarely<br />
manatees were observed with seagrass trailing from<br />
their mouths; however, they were not observed<br />
rooting into the substrate for rhizomes.
Season signi cantly aVected mean sea surface<br />
temperature (Mann<strong>–</strong>Whitney U=33, n=18, 27,<br />
P20 C) such as natural springs and<br />
human-made attractions such as the warm-water<br />
eZuents of power plants (Packard & Wetterqvist<br />
1986). However, the concept of an essential area is<br />
open to critique based on more recent studies<br />
of individual manatee movements using satellite<br />
telemetry. Individual manatees from the Florida<br />
population vary widely in seasonal movements<br />
(Deutsch et al., 2000). Some travel long distances<br />
along the east coast of the USA, others travel short<br />
distances within Florida or between Florida and<br />
Georgia, and still others appear to be year-round<br />
residents remaining in one activity centre. Since<br />
some southern-most ranges overlap with other<br />
northern-most ranges of Florida manatees (T. m.<br />
latirostris), factors other than ambient water<br />
temperature appear to interact in determining<br />
seasonal movements. Perhaps ‘seasonal activity<br />
centre’ would be a better term for resident<br />
manatees.<br />
We hypothesize that access to freshwater is more<br />
of a directive factor than temperature in in uencing<br />
seasonal manatee use of the reef in the Belize<br />
coastal zone. Alternatively, individual manatee<br />
movements may be determined by a complex interaction<br />
of many factors experienced during a lifetime,<br />
including learning processes in uencing how<br />
travel routes are stored and retrieved from memory.<br />
Our reasoning is as follows.<br />
Even though manatee presence/absence on the<br />
reef was associated with water temperature, the<br />
same variation in seasonal water temperature was<br />
found in the adjacent Drowned Cayes where<br />
manatees were observed year-round during the<br />
same sampling period (Sullivan et al., 1999;<br />
LaCommare et al., 2001). Water temperature in the<br />
study area ranged between 25 C in the winter<br />
and 31 C in the summer, well above the incipient<br />
lethal level (as de ned by Fry, 1947) of cold<br />
tolerance for manatees (20 C c.f. Irvine, 1983).<br />
Temperature may have been correlated with<br />
another, undetermined, directive factor or gradient.<br />
An alternative hypothesis might be that manatees<br />
move further from estuaries during the summer<br />
rainy season when freshwater plumes from rivers<br />
are more likely to penetrate further into the coastal<br />
zone, meaning that manatees are more likely to be
350 C. Self-Sullivan et al.<br />
at the reef in the summer. Annual salinity range at<br />
Gallows Reef was 34.5<strong>–</strong>38.0 ppt. Osmoregulation<br />
experiments on Florida manatees indicate that: (1)<br />
they are good osmoregulators in both fresh (0‰)<br />
and marine (34‰) environments (Ortiz et al., 1998);<br />
and (2) although they drink large amounts of<br />
freshwater when available, they do not drink<br />
marine water, but possibly oxidize fat to meet their<br />
water needs when restricted to eating seagrass in the<br />
marine environment (Ortiz et al., 1999). Manatees<br />
are considered to be dependent on freshwater in<br />
Florida (Hartman, 1979), and periodic access to<br />
freshwater is thought to be important to manatees<br />
in Belize (Gibson, 1995). However, it is not known<br />
whether nor how often manatees might move from<br />
the oVshore activity centres to mainland sources of<br />
freshwater in Belize. Although this distance is all<br />
well within 1-day’s travel range, manatees within<br />
the Drowned Cayes area are often sighted with<br />
dozens of salt-water barnacles covering their bodies<br />
(Self-Sullivan, unpublished data), an indication<br />
of long periods of time spent in the marine<br />
environment (Husar, 1997).<br />
Alternatively, underwater springs near our study<br />
locations may dry up or become unpalatable during<br />
winter. One manatee, sighted three times at Basil<br />
Jones in January and February 2001, was near a<br />
spring adjacent to the channel (G. Smith, unpublished<br />
data). Whether the manatee was drinking<br />
from the underwater spring could not be determined;<br />
the water was sulphurous as determined by<br />
smell and colour. Since we documented higher<br />
surface salinity in the summer compared to the<br />
winter at Gallows Reef, we do not believe that<br />
rainfall provides a lens of fresh surface water at the<br />
reef. Possibly the higher salinity was related to<br />
evaporation during summer months.<br />
Another hypothesis might be that the breaks in<br />
the reef could serve as a stopover site for manatees<br />
travelling in search of fresh water, in an east<strong>–</strong>west<br />
direction to and from oVshore atolls. Gallows Reef<br />
is midway between the Belize River (a fresh<br />
water source) and TurneVe Atoll, located further<br />
oVshore. Manatees have been opportunistically<br />
sighted at TurneVe Atoll, a large complex mangrove<br />
island-seagrass-coral reef system similar to<br />
the Drowned Cayes (Auil, 1998; Barbara Bilgre,<br />
pers. comm.).<br />
Seasonal distribution also might be related to<br />
winter storms from the northwest. Cold air masses<br />
from North America frequently aVect both temperature<br />
and wind strength during October through<br />
January (Purdy et al., 1975). ‘Northers’, as the local<br />
people call these events, may lower the sea surface<br />
by as much as 0.8 m. These events have a greater<br />
eVect than spring tides (0.5 m) on water depth in<br />
shallow northern lagoons. If manatees move away<br />
from shallow areas inside the reef, then absence<br />
would be more likely after storm events in the<br />
winter.<br />
Alternatively, manatees may have learned that<br />
the reef provides shelter from high surf during the<br />
summer hurricane season. For example, three<br />
manatees were at Gallows Reef 2 days after<br />
Tropical Storm Chantal hit the coast of Belize in<br />
2001 (Self-Sullivan, unpublished data). This is the<br />
largest group sighted at Gallows Reef. If manatees<br />
that move into the protected lagoon system sense<br />
protection from strong currents created by storm<br />
surge, breaks in the reef might serve as stopover<br />
sites for individuals moving back out to the atolls or<br />
continuing their north-south travel outside the<br />
reef. For obvious logistical reasons, such movements<br />
during hurricanes would be better studied by<br />
satellite telemetry than boat-based surveys.<br />
Clearly, there are several physical factors that are<br />
correlated with seasonal changes in the habitat used<br />
by manatees in Belize. To tease out the relative<br />
importance of these factors, we would recommend<br />
a combination of regional studies of individual<br />
manatee movements using satellite telemetry as well<br />
as simultaneous site-speci c studies at breaks in the<br />
reef. This type of approach would also provide a<br />
better understanding of how external environmental<br />
conditions interact with physiological states,<br />
as outlined below.<br />
Relation between travel and seasonal male<br />
reproductive activity<br />
Noting the distinctive absence of females at the<br />
breaks in the reef that we studied, a hypothesis<br />
emerged related to potential seasonal changes in<br />
reproductive activity of males. The evidence that<br />
manatees show seasonal patterns in reproductive<br />
behaviour is still ambiguous. Early reports by<br />
Hartman (1979) and Marsh et al. (1978) found no<br />
evidence for strong seasonality in <strong>Sirenian</strong> reproductive<br />
behaviour. However, Best (1982) reported<br />
seasonal breeding in the Amazonian manatee (T.<br />
inunguis). More recently, seasonal spermatogenesis<br />
has been reported in both Florida manatees<br />
(Hernandez et al., 1995) and dugongs (Dugong<br />
dugon) (Marsh, 1995). Similarly, seasonal variation<br />
in female reproductive hormones has been shown in<br />
captive Florida manatees (Larkin, 2000). However,<br />
seasonal reproductive activities were not tightly<br />
coordinated within populations (Larkin, 2000; Best,<br />
1982; Marsh et al., 1984; Marsh, pers. comm.).<br />
By tracking manatees using VHF radio and<br />
satellite tags since 1997, Powell et al. (2001) found<br />
more variation in male than female movement<br />
patterns in coastal lagoons south of Belize City. For<br />
example, females and some males remained inside<br />
the enclosed estuary year-round. However, a few<br />
males wandered outside the lagoon system, north<br />
and south along the mainland coast of Belize; one
male travelled at least as far as Belize City, 33 km<br />
north (Powell, pers. comm.).<br />
Based on published literature and our observations<br />
of only males on the reef with seasonal<br />
peaks and absences, we hypothesized that male<br />
manatees use the reef as a landmark during searches<br />
for oestrous females throughout the coastal lagoon<br />
system. If there is a winter low in spermatogenesis,<br />
then movement rates might decline. If there is a<br />
peak in both oestrous females and searching behaviour<br />
of males in summer, then male movement rates<br />
may increase. Variations in seasonal reproductive<br />
activity within a population might explain the<br />
diVerence in sighting peaks between Basil Jones<br />
(May<strong>–</strong>June) and Gallows Reef (July<strong>–</strong>August).<br />
Alternatively, sighting peaks could indicate<br />
clustered movements by males along the reef.<br />
To test these alternate hypotheses, we would<br />
recommend satellite-telemetry studies of both males<br />
and females in activity centres adjacent to the reef<br />
(Drowned Cayes and Chetumal Bay), combined<br />
with simultaneous site-speci c studies of individually<br />
identi able manatees. Such studies could<br />
answer the critical question of the degree to which<br />
there is a seasonal peak in reproductive activity<br />
of manatees in Belize and the degree to which<br />
male manatees move along the reef during peak<br />
reproductive periods.<br />
Implications for expanding populations in<br />
fragmented habitat<br />
Based on our sightings of ‘White Patch’ (BJ01) at<br />
both the northern (1994<strong>–</strong>1997) and southern (1999)<br />
sampling locations, we hypothesized manatees<br />
move along the reef in a north-south direction.<br />
However, we were not able to monitor both locations<br />
simultaneously. Extensive studies based on<br />
re-sightings of individually marked manatees have<br />
been successful in Florida (Reid et al., 1991) and<br />
should be continued in Belize. Over the long-term,<br />
these studies can be combined with satellite-tagging<br />
studies to determine variation in movement patterns<br />
during the lifetimes of individuals within a<br />
population (Beck & Reid, 1995; Reid et al., 1995).<br />
By comparing records at TurneVe Atoll (Bilgre,<br />
unpublished data), the Drowned Cayes, Gallows<br />
Reef (Smith, 2000; Self-Sullivan, unpublished data),<br />
and Basil Jones (Smith, 2000), a more accurate<br />
model of what in uences manatee movements in the<br />
centre of the species’ range may emerge for comparison<br />
with what is known from the northern<br />
extreme of the range, as studied in Florida.<br />
Long distance movements in uence gene ow<br />
among subpopulations, as re ected in complex patterns<br />
of genetic variation among manatees in the<br />
greater Caribbean area (Garcia-Rodriguez et al.,<br />
1998). Movements of individual Florida manatees<br />
along the Atlantic coast of North America range<br />
Manatees on the Belize Barrier Reef<br />
351<br />
from 44 km to 2360 km (median=309 km) at rates<br />
of 25 km to 87 km per day (Deutsch et al., 2000). In<br />
comparison, the Belize Barrier Reef extends only<br />
about 220 km, with many breaks that could be used<br />
as stopover sites or for ingress to manatee activity<br />
centres such as Chetumal Bay, Cayes oV Belize<br />
City, Southern Lagoon, the Gulf of Honduras, and<br />
numerous coastal rivers and bays (Auil, 1998).<br />
Considering the long travel range of Florida<br />
manatees, Antillean manatees could move among<br />
population centres along the Yucatan Peninsula<br />
from Mexico through Belize to Guatemala and<br />
Honduras (Auil, 1998). During population expansion<br />
following historical exploitation, manatees<br />
from Belize may colonize unoccupied habitat<br />
between population centres in a wider region, with<br />
positive in uences on gene ow.<br />
The reef may also in uence connectivity within<br />
the Belize Barrier Reef lagoon system, decreasing<br />
the probability of inbreeding by increasing movements<br />
between southern areas and Chetumal Bay.<br />
The deep water just east of the barrier reef oVers an<br />
unobstructed north-south travel corridor and the<br />
reef oVers a physical feature that could be used for<br />
navigation. Use of this travel route by manatees<br />
could be an alternative to movement through the<br />
complex labyrinth created by the mangroveseagrass-coral<br />
lagoon system inside the barrier reef.<br />
An analogy might be a loop highway around a<br />
metropolitan maze of small roads.<br />
Consistent with knowledge of ‘bachelor males’ in<br />
other marine mammals species, we hypothesize that<br />
the turnover of individuals using stopover sites on<br />
the reef could be related to turnover in male access<br />
to breeding females. In 2001, local tour guides and<br />
staV at Bacalar Chico Reserve reported that they<br />
‘regularly’ observed manatees in a side lagoon<br />
adjacent to Bacalar Chico and at another cut in the<br />
reef between Bacalar Chico and Basil Jones (Smith,<br />
unpublished data). Another local tour guide<br />
reported that he observed a group of ve manatees<br />
inside the reef crest at GoV’s Caye (a sand caye on<br />
the Belize Barrier Reef just south of Gallows Reef)<br />
in November 2001 (Self-Sullivan, unpublished<br />
data). And a third tour guide reported seeing a<br />
manatee outside the reef crest at South Water Caye<br />
(a sand caye on the Belize Barrier Reef still further<br />
south) in June 2002 (Self-Sullivan, unpublished<br />
data). We recommend studies to determine if these<br />
are young dispersing males, prime males between<br />
periods of breeding activity, or old males unlikely to<br />
breed again. Answering these questions will provide<br />
insight to how genetic heterogeneity may be maintained<br />
in relatively isolated subpopulations of<br />
manatees.<br />
We recommend that the broader implications of<br />
this highly speci c study should be considered in the<br />
context of eVorts to design marine reserve systems
352 C. Self-Sullivan et al.<br />
within the Caribbean. The concept of ‘connectivity’<br />
was used by Roberts (1997:1454) to draw attention<br />
to how ‘local populations may depend on processes<br />
occurring elsewhere’. Although Roberts referred to<br />
patterns of larval transport among coral reefs, the<br />
concept also should be applied to large-bodied<br />
vulnerable species that use the lagoon systems protected<br />
by reefs e.g., manatees. We use the concept of<br />
‘stopover sites’ in a manner similar to its use in<br />
studies of migratory bird species. For example,<br />
Higuchi et al. (1996) identi ed stopover sites where<br />
migratory cranes paused for less than 10 days<br />
during travel between breeding and non-breeding<br />
locations. Although Antillean manatees clearly are<br />
not migratory as a species, the concept of a stopover<br />
site may be applied to a relatively short<br />
interruption of travel by individuals. Protecting<br />
stopover sites within local lagoon systems could be<br />
as important for conserving genetic heterogeneity,<br />
as protecting long-distance travel routes connecting<br />
fragmented populations on a regional scale.<br />
Conclusions<br />
Developing eVective strategies for conservation of<br />
manatee habitat requires knowledge of activity<br />
centres and movements among them. We documented<br />
manatee presence at breaks in the Belize<br />
Barrier Reef, with signi cantly more sightings in<br />
the summer (rainy) season than in the winter<br />
(transition/dry) season. Groups were small (average<br />
of two manatees), some individuals were repeatedly<br />
sighted and others were not. At least one individual<br />
moved between the northern and southern breaks<br />
that we monitored. All of the manatees for which<br />
sex was determined were male and no calves<br />
were sighted. However, we are cautious about<br />
interpretation of these data, due to the inductive<br />
nature of this study. We proposed several hypotheses<br />
related to physical (temperature, salinity,<br />
depth, storm surge) and physiological (thermoregulation,<br />
osmoregulation, reproductive status)<br />
factors potentially aVecting manatee use of breaks<br />
in the reef. To test these hypotheses, we recommend<br />
simultaneous studies of identi able manatees at<br />
several locations on the reef. In addition, satellitetagging<br />
techniques would be appropriate for determining<br />
the seasonality and extent of individual<br />
manatee movements among activity centres and<br />
stopover sites along travel routes associated with<br />
geophysical characteristics of the Belize Barrier<br />
Reef. Based on results of this study, we propose<br />
that reefs should be considered part of the network<br />
of travel routes and stopover sites identi ed as<br />
important components maintaining connectivity<br />
within manatee habitat in Belize and between fragmented<br />
populations of Antillean manatees in the<br />
Caribbean.<br />
Acknowledgments<br />
This research project was reviewed by the Belize<br />
National Manatee Working Group, Coastal Zone<br />
Management Authority and Institute, and conducted<br />
under Scienti c Research Permits issued by<br />
the Belize Forestry Department, Conservation<br />
Division. Funding and in-kind support were provided<br />
by The Oceanic Society of San Francisco, The<br />
Lerner<strong>–</strong>Gray Marine Science Research Fund, The<br />
Earthwatch Institute, Spanish Bay Resort, Texas<br />
A&M University, University of Massachusetts-<br />
Boston, and the National Science Foundation<br />
Graduate Fellowship Program. We are extremely<br />
grateful to our Belizean eld assistants Pach<br />
Muñoz, Landy Requena, Jerry Requena, and<br />
Gilroy Robinson; our Belizean student interns<br />
Maxine Monsanto, Seleem Chan, and Clifton<br />
Williams; our logistics interns Liz Johnstone<br />
and Pam Quayle; and our Oceanic Society and<br />
Earthwatch Institute volunteers for their assistance<br />
in data collection. We thank Jimmie C. Smith for<br />
providing aerial photographs of our study area.<br />
Our gratitude is extended to Nicole Auil, Miriam<br />
Marmontel, and Leon David Olivera-Gomez for<br />
comments on previous drafts of this manuscript.<br />
Literature Cited<br />
Auil, N. (1998) Belize Manatee Recovery Plan. UNDP/<br />
GEF Coastal Zone Management Project, BZE/92/G31,<br />
Belize/UNEP Caribbean Environment Programme,<br />
Kingston.<br />
Beck, C. A., & Reid, J. P. (1995) An automated photoidenti<br />
cation catalog for studies of the life history of<br />
the Florida manatee. In: T. J. O’Shea, B. B. Ackerman<br />
& H. F. Percival (eds.) Population Biology of the Florida<br />
Manatee, pp. 120<strong>–</strong>123 National Biological Service,<br />
Washington, D.C.<br />
Bengtson, J. L. & Magor, D. (1979) A survey of manatees<br />
in Belize. Journal of Mammalogy 60, 230<strong>–</strong>232.<br />
Best, R. C. (1982) Seasonal Breeding in the Amazonian<br />
Manatee, Trichechus inunguis (Mammalia: Sirenia).<br />
Biotropica 14, 76<strong>–</strong>78.<br />
Bishop, Y. M. M., Fienberg, S. E. & Holland, P. W.<br />
(1975) Discrete Multivariate Analysis. The MIT Press,<br />
Cambridge.<br />
CEP/UNEP. (1995) Regional Management Plan for<br />
the West Indian Manatee, Trichechus manatus.<br />
CEP Technical Report No. 35. UNEP Caribbean<br />
Environment Programme, Kingston.<br />
Deutsch, C. J., Reid, J. P. Bonde, R. K., Easton, D. E.,<br />
Kochman, H. I. & O’Shea, T. J. (2000) Seasonal<br />
movements, migratory behavior, and site delity of<br />
West Indian manatees along the Atlantic coast of the<br />
United States as determined by radio-telemetry. Final<br />
Report. Research Work Order No. 163. Florida<br />
Cooperative Fish and Wildlife Research Unit,<br />
U.S. Geological Survey and University of Florida,<br />
Gainesville.<br />
Fry, F. E. J. (1947) EVects of the environment on animal<br />
activity. University of Toronto Studies Biological Series
55. Ontario Fisheries Research Laboratory 68.<br />
University of Toronto Press, Toronto.<br />
Garcia-Rodrieguez, A. I., Bowen, B. W., Domning, D. P.,<br />
Mignucci-Giannoni, A. A., Marmontel, M., Montoya-<br />
Ospina, R. A., Morales-Vela, B., Rudin, M., Bonde,<br />
R. K. & McGuire, P. M. (1998) Phylogeography of the<br />
West Indian manatee (Trichechus manatus): how many<br />
populations and how many taxa? Molecular Ecology 7,<br />
1137<strong>–</strong>1149.<br />
Gibson, J. (1995) Managing Manatees in Belize. M.S.<br />
Thesis. Department of Marine Sciences and Coastal<br />
Management, University of Newcastle upon Tyne.<br />
Hartman, D. S. (1979) Ecology and Behavior of<br />
the Manatee (Trichechus manatus) in Florida.<br />
Special Publication No. 5, The American Society of<br />
Mammalogists.<br />
Hernandez, P., Reynolds, J. E., Marsh, H. & Marmontel,<br />
M. (1995) Age and seasonality in spermatogenesis of<br />
Florida manatees. In: T. J. O’Shea, B. B. Ackerman &<br />
H. F. Percival (eds.) Population Biology of the Florida<br />
Manatee, pp. 84<strong>–</strong>97 National Biological Service,<br />
Washington, D.C.<br />
Higuchi H., Ozaki K., Fijita G., Minton J., Ueta M.,<br />
Soma M., & Mita N. (1996) Satellite tracking of<br />
White-naped crane migration and the importance of the<br />
Korean demilitarized zone. Conservation Biology 10,<br />
806<strong>–</strong>812.<br />
Hilton-Taylor, C. (compiler). (2001) 2000 IUCN Red List<br />
of Threatened Species. IUCN, Gland, Switzerland and<br />
Cambridge, UK.<br />
Husar, S. L. (1977) The West Indian manatee (Trichechus<br />
manatus). Wildlife Research Report 7. U.S.<br />
Department of the Interior, Fish and Wildlife Service,<br />
Washington, D. C.<br />
Irvine, A. B., (1983) Manatee metabolism and its in uence<br />
on distribution in Florida. Biological Conservation 25,<br />
315<strong>–</strong>334.<br />
Koelsch, J. K. (1997) The seasonal occurrence and ecology<br />
of Florida manatees (Trichechus manatus<br />
latirostris) in coastal waters near Sarasota, Florida.<br />
Master’s Thesis, Department of Biology, University of<br />
South Florida.<br />
LaCommare, K. S., Sullivan, C. S. & Brault, S. (2001)<br />
Distribution and foraging ecology of Antillean<br />
manatees (Trichechus manatus) in the Drowned Cays<br />
area of Belize, Central America. Abstracts, 14th<br />
Biennial Conference on the Biology of Marine<br />
Mammals, November 29<strong>–</strong>December 3, 2001,<br />
Vancouver, Canada.<br />
Larkin, I. L. V. (2000) Reproductive endocrinology of the<br />
Florida manatee (Trichechus manatus latirostris):<br />
estrous cycles, seasonal patterns and behavior. Ph.D.<br />
Dissertation. University of Florida.<br />
Lefebvre, L. W., Marmontel, M., Reid, J. P., Rathbun,<br />
G. B. & Domning, D. P. (2001) Status and biogeography<br />
of the West Indian manatee. In: Charles A. Woods<br />
& Florence E. Sergile (eds.) Biogeography of the West<br />
Indies: Patterns and Perspectives, pp. 425<strong>–</strong>474. CRC<br />
Press: New York.<br />
Lehner, P. N. (1996) Handbook of Ethological Methods,<br />
2nd edition. Cambridge University Press.<br />
Marsh, H. (1995) The life history, pattern of breeding, and<br />
population dynamics of the dugong. In: T. J. O’Shea,<br />
Manatees on the Belize Barrier Reef<br />
353<br />
B. B. Ackerman & H. F. Percival (eds.) Population<br />
Biology of the Florida Manatee, pp. 75<strong>–</strong>83. National<br />
Biological Service, Washington, D.C.<br />
Marsh, H., Heinsohn, G. E. & Glover, T. D. (1984)<br />
Changes in the male reproductive organs of the<br />
dugong, Dugong dugon (Sirenia: Dugondidae) with age<br />
and reproductive activity. Australian Journal of<br />
Zoology 32, 721<strong>–</strong>742.<br />
Marsh, H., Spain, A. V. & Heinsohn, G. E. (1978)<br />
Minireview: physiology of the dugong. Comparative<br />
Biochemisty and Physiology 61A, 159<strong>–</strong>168.<br />
McKillop, H. I. (1984) Prehistoric Maya Reliance on<br />
Marine Resources: Analysis of a Midden from Moho<br />
Cay. Belize Journal of Field Archaeology 11, 25<strong>–</strong>35.<br />
Morales-Vela, B., Olivera-Gomez, D., Reynolds, J. E. &<br />
Rathbun, G. B. (2000) Distribution and habitat use by<br />
manatees (Trichechus manatus manatus) in Belize and<br />
Chetumal Bay, Mexico. Biological Conservation 95,<br />
67<strong>–</strong>75.<br />
Ortiz, R. M. Worthy, G. A. J. & Byers, F. M. (1999)<br />
Estimation of water turnover rates of captive West<br />
Indian manatees (Trichechus manatus) held in fresh and<br />
salt water. Journal of Experimental Biology 202, 33<strong>–</strong>38.<br />
Ortiz, R. M., Worthy, G. A. J. & MacKenzie, D. S. (1998)<br />
Osmoregulation in wild and captive West Indian<br />
manatees (Trichechus manatus). Physiological Zoology<br />
71, 449<strong>–</strong>457.<br />
O’Shea, T. J. & Salisbury, C. A. (1991) Belize - a last<br />
stronghold for manatees in the Caribbean. Oryx 25,<br />
156<strong>–</strong>164.<br />
Packard, J. M. & Wetterqvist, O. F. (1986) Evaluation of<br />
manatee habitat systems on the northwestern Florida<br />
coast. Coastal Zone Management Journal 14, 279<strong>–</strong>310.<br />
Powell, J. A., Bonde, R., Aguirre, A. A., Koontz, C.,<br />
Gough, M. & Auil, N. (2001) Biology and movements<br />
of manatees in Southern Lagoon, Belize. Abstract in<br />
Proceedings of the 14th Biennial Conference on the<br />
Biology of Marine Mammals, Vancouver.<br />
Purdy, E. G., Pusey, W. C., III & Wantland, K. F. (1975)<br />
Continental shelf of Belize <strong>–</strong> regional shelf attributes.<br />
In: K. F. Wantland & W. C. Pusey, III (eds.) Studies in<br />
Geology No. 2: Belize Shelf <strong>–</strong> Carbonate Sediments,<br />
Clastic Sediments, and Ecology and a paper on Petrology<br />
and Diagenesis of Carbonate Eolianites of Northeastern<br />
Yucatan Peninsula, Mexico, pp. 1<strong>–</strong>52. The American<br />
Association of Petroleum Geologists, Tulsa.<br />
Reid, J. P., Rathbun, G. B. & Wilcox, J. R. (1991)<br />
Distribution patterns of individually identi able<br />
West Indian manatees (Trichechus manatus) in Florida.<br />
Marine Mammal Science 7, 180<strong>–</strong>190.<br />
Reid, J. P., Bonde, R. K. & O’Shea, T. J. (1995)<br />
Reproduction and mortality of radio-tagged and recognizable<br />
manatees on the Atlantic Coast of Florida. In:<br />
T. J. O’Shea, B. B. Ackerman & H. F. Percival (eds.)<br />
Population Biology of the Florida manatee, pp. <strong>17</strong>1<strong>–</strong>191.<br />
National Biological Service, Washington, D.C.<br />
Roberts, C. M. (1997) Connectivity and management of<br />
Caribbean coral reefs. Science 278, 1454<strong>–</strong>1457.<br />
Sanderson, G. C. (1966) The study of mammal<br />
movements <strong>–</strong> a review. Journal of Wildlife Management<br />
30, 215<strong>–</strong>235.<br />
Sullivan, C. S., Packard, J. M. & Evans, W. E. (1999)<br />
Spring distribution and behavior of Antillean manatees<br />
(Trichechus manatus manatus) in the Drowned Cayes,
354 C. Self-Sullivan et al.<br />
Belize. Abstracts, 13th Biennial Conference on the<br />
Biology of Marine Mammals, November 28<strong>–</strong>December<br />
3, 1999, Wailea, Maui, Hawaii.<br />
Smith, G. W. (2000) Identi cation of Individual Manatee<br />
In the Basil Jones Area of the Bacalar Chico Marine<br />
Reserve and the Drowned Cays Area of Belize. Annual<br />
Report to the National Manatee Working Group,<br />
Coastal Zone Management Institute & Authority,<br />
Belize City.<br />
Weeks, P. & Packard, J. M. (1997) Acceptance of<br />
scienti c management by natural resource dependent<br />
communities. Conservation Biology 11, 236<strong>–</strong>245.
NEWS OF THE WEEK<br />
MEETINGBRIEFS>><br />
1070<br />
AAAS ANNUAL MEETING | 18 TO 22 FEBRUARY 2010 | SAN DIEGO, CALIFORNIA<br />
A symposium organized at the last minute at<br />
the annual meeting of the American Association<br />
for the Advancement of Science (the<br />
publisher of Science) by two of the world’s<br />
most prominent scientific organizations<br />
addressed recent attacks on an increasingly<br />
beleaguered climate science community. The<br />
panel met in the uncertain aftermath of the<br />
stolen e-mails affair and critiques of the<br />
Intergovernmental Panel on Climate Change<br />
(IPCC) (Science, 12 February, p. 768).<br />
The symposium was convened by U.S.<br />
National Academy of Sciences President<br />
Ralph Cicerone, in conjunction with AAAS, at<br />
a time when flaws in the latest<br />
IPCC report, and even the legitimacy<br />
of climate science, have<br />
made headlines. E-mails<br />
uncovered late last year<br />
revealed instances of scientists<br />
on the panel discussing<br />
withholding data and documents<br />
from those with opposing<br />
views, conspiring to keep<br />
contradictory papers out of<br />
influential reports, and<br />
encouraging colleagues to<br />
delete e-mails.<br />
Despite a drumbeat of studies that corroborate<br />
the conclusion that the planet is warming<br />
and human activities are largely responsible,<br />
these recent skirmishes “have really shaken<br />
the confidence of the public in the conduct of<br />
science [overall],” said Cicerone, citing a number<br />
of recent polls on the public perception of<br />
science. “The situation is completely out of<br />
hand,” said climate scientist Gerald North of<br />
Not so fast. Critics have hit IPCC over a false assertion<br />
that Himalayan glaciers would melt away by 2035.<br />
Scientists Grapple With ‘Completely<br />
Out of Hand’ Attacks on Climate Science<br />
E-mail etiquette. Gerald North<br />
says scientists should not sound<br />
like bloggers.<br />
Texas A&M University in College Station,<br />
who has served as an IPCC reviewer. “One<br />
guy e-mailed me to say I’m a ‘whore for the<br />
global warming crowd.’ ” His PowerPoint<br />
presentation at the meeting included a slide<br />
quoting conservative talk show host Glenn<br />
Beck, who suggested that scientists commit<br />
“hara-kari” to atone. “Scientists cannot use<br />
the same tone and rhetorical style as commentators<br />
and bloggers,” North said.<br />
Although much of the session at the<br />
meeting, titled “Ensuring the Transparency<br />
and Integrity of Scientific Research,” focused<br />
on what Harvard University oceanographer<br />
and former AAAS head James<br />
McCarthy called the “abominable”<br />
press coverage, scientists<br />
owned up to their share of<br />
the blame. Small errors in the<br />
2007 report were “careless,”<br />
said McCarthy, but IPCC<br />
should have done a full and<br />
public examination to describe<br />
how they had come about:<br />
“The names of the authors,<br />
who was on the review, what<br />
happened—it all should have<br />
been up there, and it wasn’t<br />
done. And I think that the institution was hurt<br />
as a result,” he said.<br />
The community allowed “the situation to<br />
get out of control,” said Sheila Jasanoff of<br />
Harvard University. She said in general scientists<br />
had to connect better to the public.<br />
“There is a kind of arrogance—we are scientists<br />
and we know best,” Jasanoff said.<br />
“That needs to change.” <strong>–</strong>ELI KINTISCH<br />
26 FEBRUARY 2010 VOL 327 SCIENCE www.sciencemag.org<br />
Published by AAAS<br />
The Latest on<br />
Geoengineering<br />
Preliminary findings presented here suggest<br />
that some proposed techniques to cool the<br />
planet manually may have fewer barriers<br />
than previously thought. But many technical<br />
and societal barriers remain.<br />
Even before they got to the sessions, the<br />
scientists had to contend with a smattering of<br />
activists with drums, cameras, and a megaphone<br />
alleging that the government is already<br />
performing geoengineering through the<br />
spraying of particles, in so-called chemtrails.<br />
Physicist David Keith of the University<br />
of Calgary in Canada addressed the concept<br />
of spreading aerosol droplets in the stratosphere,<br />
where they could block a small fraction<br />
of the sun’s rays. A paper published last<br />
year in Environmental Research Letters suggested<br />
that the leading proposal, spraying<br />
Is a Dolphin a Person?<br />
Are dolphins as smart as people? And if<br />
so, shouldn’t we be treating them a bit better?<br />
Those were the questions scientists<br />
and philosophers debated at a session here<br />
on Sunday.<br />
Dolphins, it turns out, are pretty darn smart.<br />
Panelist Lori Marino, an expert on cetacean<br />
neuroanatomy at Emory University in Atlanta,<br />
said they may be Earth’s second smartest creature,<br />
after humans, of course.<br />
Bottlenose dolphins have bigger brains than<br />
humans (1600 grams versus 1300 grams), and<br />
they have a brain-to-body-weight ratio greater<br />
than that of great apes (but smaller than that of<br />
humans), said Marino. “They are the second<br />
most encephalized beings on the planet.”<br />
But it’s not just size that matters. Dolphins<br />
also have a very complex neocortex, the part<br />
of the brain responsible for problem solving,<br />
self-awareness, and various other traits we<br />
associate with human intelligence. And<br />
researchers have found spindle neurons in<br />
dolphin brains called von Economo neurons<br />
that in humans and apes have been linked to<br />
emotions, social cognition, and even theory of<br />
mind: the ability to sense what others are<br />
thinking. Overall, said Marino, “dolphin<br />
brains stack up quite well to human brains.”<br />
What dolphins do with their brains is also<br />
impressive. Cognitive psychologist Diana<br />
CREDITS (TOP TO BOTTOM): PHOTOS.COM; CONSERVATION HISTORY ASSOCIATION OF TEXAS<br />
on March 11, 2010<br />
www.sciencemag.org<br />
Downloaded from
CREDIT: PHOTOS.COM<br />
sulfur dioxide gas, wouldn’t work. Sulfur<br />
dioxide is converted in the atmosphere into<br />
droplets of sulfuric acid, which would clump<br />
and fall out of the sky before they could have<br />
much cooling effect. To get around this<br />
problem, Keith and colleagues have proposed<br />
using airplanes to spray droplets of<br />
the acid itself, rather than sulfur dioxide. In<br />
unpublished data, the team found that injecting<br />
only “a few megatons per year” of sulfuric<br />
acid could be more than twice as effective<br />
at blocking radiation as starting with<br />
sulfur dioxide.<br />
While scientists are finding ways to overcome<br />
the engineering challenges, the environmental<br />
effects of planet-hacking techniques<br />
remain uncertain. One challenge in geoengineering<br />
a warmed planet is simultaneously<br />
restoring temperatures while minimizing disruption<br />
of rain and precipitation. (Stratospheric<br />
particles lower the total amount of energy striking<br />
Earth, the driver of precipitation.)<br />
In previous modeling efforts, adding<br />
sun-blocking particles uniformly across the<br />
globe has tended to undercool the poles<br />
while overcooling the equator. So Kenneth<br />
Reiss of Hunter College of the City University<br />
of New York has been working with dolphins in<br />
aquariums for most of her career, and she said<br />
their social intelligence rivals that of the great<br />
apes. Dolphins can recognize themselves in a<br />
mirror, a sign of self-awareness. They can<br />
understand complex gesture “sentences” from<br />
humans. And they can learn to poke an underwater<br />
keyboard to request toys. “Much of their<br />
learning is similar to what we see with young<br />
children,” said Reiss.<br />
So if dolphins are so similar to people,<br />
shouldn’t we be treating them more like people?<br />
“The very traits that make dolphins interesting<br />
to study,” said Marino, “make confining<br />
them in captivity unethical.” She noted, for<br />
example, that, in the wild, dolphins have a<br />
home range of about 100 square kilometers. In<br />
captivity, they roam one 10-thousandth of one<br />
percent of this area.<br />
Far worse, Reiss said, is the massive dolphin<br />
culling ongoing in some parts of the world,<br />
which she documented with a graphic video of<br />
dolphins being drowned and stabbed in places<br />
like the Japanese town of Taiji.<br />
Thomas White, a philosopher at Loyola<br />
Marymount University in Redondo Beach,<br />
California, suggested that dolphins aren’t<br />
merely like people—they may actually be people,<br />
or at least, “nonhuman persons.” Defining<br />
exactly what it means to be a person is difficult,<br />
White said, but dolphins seem to fit the check-<br />
Caldeira, a climate scientist at the Carnegie<br />
Institution for Science in Stanford, California,<br />
modeled various approaches to try to<br />
counteract a severe warming—the result of a<br />
doubling of preindustrial CO 2 concentration.<br />
In work yet to be published, he distributed<br />
the particles unevenly to try to minimize<br />
those effects; for example, by putting<br />
more at the poles versus the equator. (Global<br />
warming is greatest in the Arctic.) In models,<br />
that strategy helped fix the undercooling/overcooling<br />
problem, but it worsened<br />
the effects on precipitation. “There’s a<br />
complex problem of how do you balance<br />
the damage that you do against the benefit,”<br />
said Caldeira.<br />
That said, simulating either geoengineering<br />
approach to counteract global warming—<br />
distributing particles globally or focusing on<br />
the poles—suggests a cooler world with less<br />
disruption of rain patterns than one in which<br />
warming continues unabated. “In a highglobal-warming<br />
world, more people would be<br />
better off with geoengineering, but some people<br />
would be worse off,” he said.<br />
<strong>–</strong>ELI KINTISCH<br />
Social smarts. Dolphins display many of the<br />
same behaviors humans do.<br />
list many philosophers agree on. There are the<br />
obvious ones: They’re alive, aware of their environment,<br />
and have emotions; but they also<br />
seem to have personalities, exhibit self-control,<br />
and treat others appropriately, even ethically.<br />
When it comes to what defines a person, said<br />
White, “dolphins fit the bill.”<br />
Still, experts caution that the scientific<br />
case for dolphin intelligence is based on relatively<br />
little data. “It’s a pretty story, but it’s<br />
very speculative,” says Jacopo Annese, a<br />
neuroanatomist at the University of California,<br />
San Diego. Despite a long history of research,<br />
scientists still don’t agree on the roots of intelligence<br />
in the human brain, he says. “We don’t<br />
know, even in humans, the relationship<br />
between brain structure and function, let alone<br />
intelligence.” And, Annese says, far less is<br />
known about dolphins. <strong>–</strong>DAVID GRIMM<br />
With reporting by Greg Miller.<br />
More Highlights<br />
From AAAS 2010<br />
NEWS OF THE WEEK<br />
Science reporters posted more than two<br />
dozen blog entries and podcasts from the<br />
meeting. Here is a sample. For full coverage,<br />
see www.sciencenow.org. And to see what<br />
our guest bloggers had to say, see<br />
news.sciencemag.org/sciencebloggers.<br />
A Sexy Treatment for Traumatic Brain Injury<br />
The hormone progesterone is best known for<br />
its work in the female reproductive system,<br />
where it plays various roles in supporting<br />
pregnancy. But starting next month, it will be<br />
the focus of a phase III clinical trial for traumatic<br />
brain injury. Researchers hope an infusion<br />
of progesterone given within a few hours<br />
of a car accident or other trauma will help<br />
prevent brain damage.<br />
The Mathematics of Clumpy Crime<br />
Even in a sprawling city like Los Angeles,<br />
California, crimes still clump together. Mathematical<br />
models presented at the meeting<br />
show that such crime hot spots form when<br />
previous crimes attract more criminals to a<br />
neighborhood. By understanding how these<br />
blobs form, researchers hope to help police<br />
break them up.<br />
Are ‘Test Tube Babies’ Healthy?<br />
When Louise Brown was born on 25 July<br />
1978, she kicked off an era. The first “test<br />
tube baby” is a mother herself now, and she’s<br />
been joined by millions of others born with<br />
the help of in vitro fertilization (IVF). But are<br />
babies born via IVF the same as those born<br />
naturally? Researchers have discovered some<br />
subtle genetic differences.<br />
Drive Green, Make Money<br />
Widespread adoption of plug-in electric vehicles<br />
could dramatically cut greenhouse gas<br />
pollution and reduce U.S. dependence on<br />
foreign oil. And results of a new electric-car<br />
pilot project provide added incentive to go<br />
electric: Car owners could return unused<br />
electricity to the grid and make real money<br />
doing so.<br />
Science Is Kryptonite for Superheroes<br />
Hollywood has a message for scientists: If<br />
you want something that’s 100% accurate,<br />
go watch a documentary. A panel of screenwriters<br />
for superhero-driven movies and TV<br />
shows like Watchmen and Heroes said that<br />
their job is to get the characters right, not<br />
the science.<br />
www.sciencemag.org SCIENCE VOL 327 26 FEBRUARY 2010 1071<br />
Published by AAAS<br />
on March 11, 2010<br />
www.sciencemag.org<br />
Downloaded from
Functional Ecology 2008, 22,<br />
284<strong>–</strong>288 doi: 10.1111/j.1365-2435.2007.01354.x<br />
Blackwell Publishing Ltd<br />
Infant carrying behaviour in dolphins: costly parental<br />
care in an aquatic environment<br />
S. R. Noren*<br />
Institute of Marine Science, Center for Ocean Health, University of California at Santa Cruz, 100 Shaffer Road, Santa Cruz,<br />
CA 95060, USA<br />
Functional doi: 10.1111/j.1365-2435.2007.0@@@@.x<br />
Ecology (2007)<br />
xx,<br />
000<strong>–</strong>000<br />
Summary<br />
1. Infant carrying behaviour occurs across diverse taxa inhabiting arboreal, volant and aquatic<br />
environments. For mammals, it is considered to be the most expensive form of parental care after<br />
lactation, yet the effect of infant carrying on the energetics and performance of the carrier is<br />
virtually unknown.<br />
2. Echelon swimming in cetacean (dolphin and whale) mother<strong>–</strong>infant dyads, described as calf in<br />
very close proximity of its mother’s mid-lateral flank, appears to be a form of aquatic ‘infant<br />
carrying’ behaviour as indicated by the hydrodynamic benefits gained by calves in this position<br />
which enables them to maintain proximity of their travelling mothers. Although this behaviour<br />
provides a solution for minimizing separations of mother<strong>–</strong>infant dyads, it may be associated with<br />
maternal costs.<br />
3. Through kinematic analyses this study demonstrates empirically that ‘infant carrying’ impacts<br />
the locomotion of dolphin ( Tursiops truncatus)<br />
mothers as evident by decreased swim performance<br />
and increased effort.<br />
4. The mean maximum swim speed of mothers swimming in echelon only represented 76% of the<br />
mean maximum swim speed of these mothers swimming solitarily. In addition, there was a<br />
concomitant 13% reduction in distance per stroke for mothers swimming in echelon compared to<br />
periods of solitary swimming.<br />
5. Thus, ‘infant carrying’ in an aquatic environment is associated with maternal costs, and could<br />
ultimately impact maternal energy budgets, foraging efficiency and predator evasion.<br />
Introduction<br />
Key-words:<br />
cetacean, echelon position, hydrodynamics, kinematics, swimming<br />
Maternal investment in mammals includes gestation,<br />
lactation and other forms of parental care. Infant carrying is<br />
considered to be the most costly form of parental care after<br />
lactation (Altmann & Samuels 1992; Kramer 1998) and has<br />
been described in 6 of 19 eutherian mammalian orders (for<br />
review, see Ross 2001). This behaviour provides a solution for<br />
mothers of diverse taxa that must manoeuvre within their<br />
environment to forage and avoid predators while accompanied<br />
by their young offspring (Ross 2001), which are<br />
handicapped by small body size, undeveloped tissues and<br />
naïveté (Carrier 1996). This behaviour is only thought to<br />
evolve when offspring are unable to independently follow<br />
their mothers, as in arboreal (i.e. primates) and volant (i.e.<br />
bats) environments (Ross 2001).<br />
*Correspondence author. E-mail: snoren@biology.ucsc.edu<br />
The aquatic environment also appears to require ‘infant<br />
carrying’ to ensure that mother<strong>–</strong>infant dyads remain intact<br />
during travel as manatees and sea otters are observed<br />
physically carrying their young. Cetaceans (whale and dolphin),<br />
however, cannot physically carry their young. Yet similar to<br />
that observed for primates (Altmann & Samuels 1992; Doran<br />
1992; Wells & Turnquist 2001), mature locomotor performance<br />
in dolphins is precluded for several years postpartum (Noren,<br />
Biedenbach & Edwards 2006). Echelon position is the<br />
predominant behaviour displayed by cetacean mother<strong>–</strong>infant<br />
dyads (Fig. 1; McBride & Kritzler 1951; Tavolga & Essapian<br />
1957; Norris & Prescott 1961; Au & Perryman 1982; Taber &<br />
Thomas 1982; Mann & Smuts 1999; Noren & Edwards 2007)<br />
and appears to represent an aquatic form of ‘infant carrying’<br />
because it enables neonatal cetaceans to maintain close<br />
proximity to their mothers during travel (Norris & Prescott<br />
1961; Lang 1966) by increasing the swimming efficiency of the<br />
infant (Kelly 1959; Weihs 2004; Noren et al.<br />
2008). Thus<br />
cetaceans, like primates, appear to ‘carry’ their young.<br />
© 2007 The Author. Journal compilation © 2007 British Ecological Society
Fig. 1. Three bottlenose dolphin mother<strong>–</strong>calf pairs swimming in echelon<br />
position. Echelon position is described as calf in very close proximity<br />
with its mother’s mid-lateral flank in the region near her dorsal fin.<br />
Photo © and courtesy of Dolphin Quest Hawaii.<br />
Only a few studies have examined the energetic and<br />
locomotor consequences of infant carrying for the carrier,<br />
and these studies have focused on primates (Altmann &<br />
Samuels 1992; Schradin & Anzenberger 2001) and marsupials<br />
(Baudinette & Biewener 1998). Meanwhile the maternal<br />
consequences of infant carrying in other taxa and environments<br />
(i.e. volant and aquatic) remain unexplored. In view of this, I<br />
examined the kinematics of dolphin mothers swimming with<br />
their calf in echelon (Fig. 1) and swimming solitarily (> 1 m<br />
from their calf and all other dolphins) to elucidate the effect of<br />
‘infant carrying’ on maternal locomotor performance and<br />
effort in an aquatic environment. The advantages gained by<br />
calves in echelon position are examined in the accompanying<br />
paper (Noren et al.<br />
2008).<br />
Materials and methods<br />
Three captive bottlenose dolphin ( Tursiops truncatus)<br />
mother<strong>–</strong>calf<br />
pairs housed at Dolphin Quest Hawaii provided a controlled<br />
experimental approach to investigate cetacean locomotor performance<br />
and effort (Fish 1993; Skrovan et al.<br />
1999; Noren et al.<br />
2006, 2008).<br />
Methodologies regarding the dolphin enclosure, placement of the<br />
SCUBA diver-videographer, and placement of the dolphins in the<br />
water column are described in detail elsewhere (Noren et al.<br />
2008).<br />
Experimental swim sessions included both opportunistic (no reward)<br />
and directional swimming between two trainers (reward-based).<br />
Echelon swimming was recorded only when the mothers’ calves<br />
were 0<strong>–</strong>34 days postpartum. Thus, the present study only provides<br />
details regarding the maternal costs of echelon for mothers with<br />
very young calves (0<strong>–</strong>34 days postpartum). Solitary swimming was<br />
recorded at several intervals 0<strong>–</strong>2 years past parturition. Thirty-three<br />
hours of swimming were recorded and 334 short 1<strong>–</strong>6 s video clips<br />
were extracted and digitized (Fig. 2). Clips were divided into two<br />
association categories: (i) echelon position (Fig. 1); and (ii) solitary<br />
swimming (mother > 1 m away from calf and all other dolphins). In<br />
all clips the mothers were continuously stroking.<br />
A quantitative assessment of swim effort was obtained by<br />
calculating peak-to-peak fluke stroke amplitude and tailbeat<br />
oscillation frequency. Higher amplitudes and frequencies are<br />
associated with greater energy expenditure (Kooyman & Ponganis<br />
1998). Normalized tailbeat frequency (ratio of tailbeat frequency to<br />
© 2007 The Author. Journal compilation © 2007 British Ecological Society, Functional Ecology,<br />
22,<br />
284<strong>–</strong>288<br />
Infant carrying in dolphins<br />
285<br />
Fig. 2. A tracing from a digitized video clip of a solitarily swimming<br />
mother dolphin. Anatomical points of interest (rostrum tip, cranial<br />
insertion of the dorsal fin, and fluke tip) were digitized at a rate of 60<br />
fields per second of video using a motion-analysis system (Peak<br />
Motus 6·1; Peak Performance Technologies, Inc. Englewood, CO,<br />
USA) following methods similar to Skrovan et al. (1999) and Noren<br />
et al. (2006). A distinct trace represents the movements of each<br />
digitised anatomical point. From left to right, the trace from the<br />
rostrum leads (pink), followed by the trace from the cranial insertion<br />
of the dorsal fin (yellow), and last is the trace from the fluke tip (blue).<br />
The brown dot is a digitized reference point indicating that the<br />
camera was steady while filming this video clip.<br />
swim speed; Rohr & Fish 2004) and distance per stroke were also<br />
calculated. Methods for video analysis and swim effort calculations<br />
are described in detail elsewhere (Noren et al.<br />
2006).<br />
The goal of this study was not to address individual variation, but<br />
to quantify changes in locomotor performance associated with swim<br />
style (echelon vs. solitary), thus similar to other kinematic studies<br />
data across individuals were pooled (Fish 1993; Skrovan et al.<br />
1999;<br />
Noren et al.<br />
2006, 2008). Pearson product moment correlation<br />
coefficients were used to determine the correlations of peak-to-peak<br />
fluke stroke amplitude and tailbeat frequency with swim speed for<br />
echelon swimming and also for solitary swimming; linear regression<br />
analyses were then used to determine the relationship for the parameters<br />
that demonstrated a strong correlation. Swim speed, normalized<br />
tailbeat frequency, and distance per stroke during echelon swimming<br />
and solitary swimming were compared using student’s t-tests<br />
when normally distributed, or Mann<strong>–</strong>Whitney rank sum tests when<br />
normality failed ( α = 0·05). The maximum swim performance of<br />
each individual during echelon swimming and solitary swimming<br />
was compared using a paired t-test<br />
( α = 0·10). Statistical analyses<br />
were performed using Sigma<br />
Stat<br />
2·03 (Systat Software, Inc. Point<br />
Richmond, CA, USA). Means ± 1 SEM are presented.<br />
Results<br />
Our experimental approach adequately captured swimming<br />
behaviours representative of wild dolphins (Noren et al.<br />
2008). The average swim speed of mother dolphins was<br />
<strong>–</strong>1<br />
significantly slower during echelon swimming (2·11 ± 0·06 m s ;<br />
<strong>–</strong>1<br />
n = <strong>17</strong>8) than during solitary swimming (3·88 ± 0·09 m s ;<br />
n = 156; T = 36534·00, P < 0·001). Given that 53% of the<br />
echelon swim data were from directional swim trials, compared<br />
to 94% for the solitary swim data, swim speed data from<br />
directional trials only were also compared to ensure that the<br />
previous result was not due to experimental design. The result
286<br />
S. R. Noren<br />
Fig. 3. Swimming kinematics of the mother in relation to the<br />
swimming speed of the mother. Peak-to-peak fluke stroke amplitude<br />
(a) and tailbeat frequency (b) were both correlated with swim speed<br />
for mother dolphins swimming in echelon with their calf (white<br />
symbols; r = 0·400, P < 0·001, n = <strong>17</strong>8 and r = 0·799, P < 0·001, n =<br />
<strong>17</strong>8, respectively) and swimming solitarily (black symbols; r = 0·277,<br />
P < 0·001, n = 156 and r = 0·885, P < 0·001, n = 156, respectively).<br />
The relationship between swim speed and peak-to-peak fluke stroke<br />
amplitude appears to be nonlinear. Given the strong linear correlation<br />
between swim speed and tailbeat frequency (SF) linear regressions<br />
are provided for echelon (speed = 1·61 SF + 0·27; r 2 = 0·638, F =<br />
310·794, P < 0·001) and solitary (speed = 2·09 SF + 0·13; r 2 = 0·783,<br />
F = 555·610, P < 0·001) swimming. A different symbol is used for<br />
each of the three individual dolphins.<br />
was the same; average swim speed during echelon swimming<br />
was significantly slower than during solitary swimming<br />
( t = −11·644,<br />
P < 0·001, n = 94, 145). The absolute maximum<br />
swim speed for each mother swimming with their calf in<br />
<strong>–</strong>1<br />
echelon was 4·23, 4·39 and 4·32 m s for animals 1, 2 and 3,<br />
<strong>–</strong>1<br />
respectively. This compares to 5·65, 6·32 and 5·11 m s for<br />
animals 1, 2 and 3 swimming solitarily, respectively. As a<br />
result, mean maximum swim performance was significantly<br />
<strong>–</strong>1<br />
slower when mothers were swimming in echelon (4·31 ± 0·05 m s )<br />
<strong>–</strong>1<br />
compared to solitary swim periods (5·69 ± 0·35 m s ; t = −4·816,<br />
n = 3, P = 0·053).<br />
Peak-to-peak fluke stroke amplitudes (Fig. 3a) and tailbeat<br />
frequencies (Fig. 3b) were correlated with swim speed for<br />
mother dolphins swimming in echelon and swimming<br />
solitarily. Because average swim speed was significantly different<br />
between the two swim categories, swim effort was standardized<br />
for swim speed. Mothers in echelon demonstrated significant<br />
increases in effort compared to periods of solitary swimming<br />
as evident by greater normalized tailbeat frequency<br />
Fig. 4. ‘Infant carrying’ imparted maternal costs. Mother dolphins<br />
swimming in echelon with their calf (white bar) demonstrated<br />
significantly lower mean distance covered per stroke (t = −7·068,<br />
P < 0·001, n = <strong>17</strong>8, 156) compared to periods of solitary swimming<br />
(black bars).<br />
( T = 19949·00, P < 0·001, n = <strong>17</strong>8, 156) and reduced distance<br />
per stroke (Fig. 4). For a given speed, mothers swimming in<br />
echelon increased tailbeat frequency by <strong>17</strong>%, which resulted<br />
in a 13% decrease in distance covered per stroke compared to<br />
solitary swimming.<br />
Discussion<br />
Infant carrying provides a solution for mothers who must<br />
locomote in their environment while accompanied by their<br />
young offspring. Echelon swimming in cetacean mother<strong>–</strong><br />
infant dyads is a type of ‘infant carrying’ that improves calf<br />
swim performance (Noren et al.<br />
2008), but it appears to come<br />
with a maternal cost. Average and maximum swim speeds<br />
were significantly slower for mothers swimming in echelon<br />
with their calves compared to periods of solitary swimming.<br />
For example, the mean maximum swim speed of the mothers<br />
swimming in echelon only represented 76% of the mean<br />
maximum performance of these mothers swimming solitarily.<br />
The maximum speed was assumed to represent the animal’s<br />
extreme performance, which is the method used to qualify<br />
physiological capacity (Weibel et al.<br />
1987), thus this result<br />
implies that the presence of a calf is a detriment to maternal<br />
swim performance.<br />
An alternate hypothesis for the decreased performance of<br />
dolphin mothers swimming in echelon is that the mother can<br />
only travel as fast as the infant can actively swim because the<br />
infant must occasionally stroke during the echelon swim<br />
behaviour. However, 0<strong>–</strong>1 month-old calves are capable of<br />
<strong>–</strong>1<br />
independent swim speeds of 0·58<strong>–</strong>4·20 m s (Noren et al.<br />
2006) and this performance is increased when the calf is in<br />
echelon (Noren et al.<br />
2008) because it receives up to 60% of<br />
the thrust from its mother (Weihs 2004). Thus, it is unlikely<br />
that the swimming ability of an entrained calf constrains<br />
maternal echelon swim performance. Studies of chimpanzees<br />
( Pan troglodytes)<br />
also suggest that infant carrying decreases<br />
maternal travel speed (Wrangham 2000; Williams, Hsien-Yang<br />
& Pusey 2002).<br />
© 2007 The Author. Journal compilation © 2007 British Ecological Society, Functional Ecology,<br />
22,<br />
284<strong>–</strong>288
In addition to decreased performance, dolphin mothers<br />
swimming in echelon were required to increase effort compared<br />
to periods of solitary swimming. For a given speed,<br />
mothers swimming in echelon significantly increased tailbeat<br />
frequency with the result that distance covered per stroke<br />
significantly decreased by 13% compared to solitary<br />
swimming (Fig. 4). Interestingly, the proportion of decreased<br />
distance per stroke in ‘infant carrying’ dolphin mothers is<br />
strikingly similar to the <strong>17</strong>% decrease in distance per leap<br />
measured empirically for marmosets ( Callithrix jacchus)<br />
carrying weights equivalent to their newborn twin offspring<br />
(Schradin & Anzenberger 2001). Furthermore, although<br />
female wallabies carrying a load (approximating the mass of<br />
a fully developed offspring) did not alter stride frequency,<br />
they increased the time the foot applied force on the ground,<br />
which likely increased the stored elastic strain energy to levels<br />
necessary to transport the additional load (Baudinette &<br />
Biewener 1998). Regardless of habitat (aquatic vs. arboreal)<br />
or forces to overcome (hydrodynamic drag vs. gravity) in these<br />
systems maternal effort must change to support an increased load.<br />
The increase in maternal effort for dolphin mothers<br />
swimming in echelon compared to periods of solitary swimming<br />
may be associated with changes in water flow patterns and<br />
drag. The presence of the calf may disrupt the boundary flow<br />
around the mother causing it to separate, which would<br />
increase turbulent flow. In addition, the entrained calf could<br />
increase the surface area of the mother, which effectively<br />
increases the drag of the swimmer (Webb 1975). More power<br />
is required to overcome increased turbulent flow and drag<br />
(Webb 1975). As a greater proportion of maternal power<br />
output is utilized to accommodate increased turbulent flow<br />
and drag, there is less energy available to propel the animal<br />
forward. As a result, locomotor performance decreases<br />
because power output per stroke is limited by mechanical<br />
constraints (Fish & Hui 1991) and total work is limited by the<br />
animal’s metabolic scope (Weibel et al.<br />
1987). These relationships<br />
may explain the observed decrease in swim speed and<br />
distance covered per stroke for mothers swimming in echelon<br />
position. Although Weihs (2006) suggested that the gain in<br />
forward forces by the following body (calf) is larger than the<br />
added cost to the leading body (mother), a more detailed<br />
theoretical examination of the drag and flow patterns for the<br />
leading body in echelon position is warranted to validate the<br />
hypotheses for the proposed decreased maternal performance<br />
observed in the present study.<br />
Given that infant carrying mothers must forage and evade<br />
predators, one of many constraints on newborn offspring<br />
mass may be its impact on maternal locomotor performance.<br />
As a result, newborn dolphins and marmosets represent a<br />
similar proportion of maternal mass, 15% (mass data from a<br />
mother<strong>–</strong>calf pair in this study) and <strong>17</strong>% (Tardif, Harrison &<br />
Simek 1993), respectively. Ultimately as an infant ages and<br />
increases in size, maternal costs associated with infant<br />
carrying theoretically increase in aquatic (Weihs 2004, 2006)<br />
and arboreal (Schradin & Anzenberger 2001) environments.<br />
Optimal theory predicts that maternal carrying costs should<br />
not outweigh the sum of the costs of independent locomotion<br />
© 2007 The Author. Journal compilation © 2007 British Ecological Society, Functional Ecology,<br />
22,<br />
284<strong>–</strong>288<br />
Infant carrying in dolphins<br />
287<br />
by the mother and her offspring (Kramer 1998). Therefore,<br />
the transition to offspring locomotor independence may<br />
represent a parent<strong>–</strong>offspring conflict (Trivers 1974) as energy<br />
expenditure to carry an infant may limit the mother’s available<br />
energy to invest in future offspring (Kramer 1998). As<br />
cetacean and primate offspring increase in size, there is an<br />
increase in the active prevention and/or avoidance of infant<br />
carrying by mothers in both groups (Altmann 1980; Taber &<br />
Thomas 1982; Mann & Smuts 1999), such that with age there<br />
is a decrease in the time cetacean infants swim in echelon<br />
(Taber & Thomas 1982; Mann & Smuts 1999; Noren &<br />
Edwards 2007) and primate infants are carried (Altmann &<br />
Samuels 1992; Salvage et al.<br />
1996; Pontzer & Wrangham<br />
2006). Examination of human infant carrying behaviour also<br />
suggests that by a certain mass and age, mothers encourage<br />
their infants to walk independently (Kramer 1998).<br />
In summary, this study provides the first empirical evidence<br />
of the maternal consequences of ‘infant carrying’ in an aquatic<br />
environment. Although infant carrying provides a solution<br />
for mothers that must manoeuvre within their environment<br />
while accompanied by their underdeveloped offspring, this<br />
behaviour is associated with maternal costs regardless of environment,<br />
as evident in arboreal (Schradin & Anzenberger 2001)<br />
and aquatic (present study) regimes. The decreased locomotor<br />
performance and increased locomotor effort associated with<br />
infant carrying undoubtedly impacts maternal energy budgets,<br />
foraging efficiency and predator evasion. Given the prevalence<br />
of infant carrying behaviour across diverse taxa and habitats<br />
and the consequences of this behaviour, it is surprising that<br />
the energetics of infant carrying have largely been ignored.<br />
Future investigations are warranted, particularly in a volant<br />
environment, which remains unexplored.<br />
Acknowledgments<br />
I thank Dolphin Quest, particularly J. Sweeney and R. Stone, for providing the<br />
experimental facilities and animals and for funding portions of data collection<br />
and analyses. I also thank Southwest Fisheries Science Center (SWFSC),<br />
particularly S. Reilly and E. Edwards, for funding portions of data collection.<br />
In addition, I thank the staff at Dolphin Quest Hawaii (particularly G. Biedenbach<br />
and C. Buczyna), T. Williams of the University of California Santa Cruz for the<br />
use of her Peak Motus system, and J. Redfern of SWFSC for assistance with<br />
data management.<br />
References<br />
Altmann, J. (1980) Baboon Mothers and Infants.<br />
Harvard University Press,<br />
Cambridge.<br />
Altmann, J. & Samuels, A. (1992) Costs of maternal care: infant-carrying<br />
baboons. Behavioural Ecology and Sociobiology,<br />
29,<br />
391<strong>–</strong>398.<br />
Au, D. & Perryman, W. (1982) Movement and speed of dolphin schools<br />
responding to an approaching ship. Fishery Bulletin,<br />
80,<br />
371<strong>–</strong>379.<br />
Baudinette, R.V. & Biewener, A.A. (1998) Young wallabies get a free ride.<br />
Nature,<br />
395,<br />
653<strong>–</strong>654.<br />
Carrier, D.R. (1996) Ontogenetic limits on locomotor performance. Physiological<br />
Zoology,<br />
69,<br />
467<strong>–</strong>488.<br />
Doran, D.M. (1992) The ontogeny of chimpanzee and pygmy chimpanzee<br />
locomotor behavior: a case study of paedomorphism and its behavorial<br />
correlates. Journal of Human Evolution,<br />
23,<br />
139<strong>–</strong>158.<br />
Fish, F.E. (1993) Power output and propulsive efficiency of swimming bottlenose<br />
dolphins ( Tursiops truncatus).<br />
Journal of Experimental Biology,<br />
185,<br />
<strong>17</strong>9<strong>–</strong>193.<br />
Fish, F.E. & Hui, C.A. (1991) Dolphin swimming <strong>–</strong> a review. Mammal Review,<br />
21,<br />
181<strong>–</strong>195.
288<br />
S. R. Noren<br />
Kelly, H.R. (1959) A two body problem in the echelon swimming of porpoise.<br />
Naval Ordinance Test Station Technical Note 40606<strong>–</strong>1.<br />
Kooyman, G.L. & Ponganis, P.J. (1998) The physiological basis for diving at<br />
depth: birds and mammals. Annual Review of Physiology 60,<br />
19<strong>–</strong>32.<br />
Kramer, P.A. (1998) The costs of human locomotion: maternal investment in<br />
child transport. American Journal of Physical Anthropology,<br />
107,<br />
71<strong>–</strong>85.<br />
Lang, T.G. (1966) Hydrodynamic analysis of cetacean performance. Whales,<br />
Dolphins, and Porpoises (ed. K.S. Norris), pp. 410<strong>–</strong>432. University of<br />
California Press, Berkeley.<br />
Mann, J. & Smuts, B. (1999) Behavioral development in wild bottlenose<br />
dolphin newborns ( Tursiops sp.). Behavior,<br />
136,<br />
529<strong>–</strong>566.<br />
McBride, A.F. & Kritzler, H. (1951) Observations on pregnancy, parturition,<br />
and post-natal behavior in the bottlenose dolphin. Journal of Mammology,<br />
32,<br />
251<strong>–</strong>266.<br />
Noren, S.R. & Edwards, E.F. (2007) Physiological and behavioral development<br />
in dolphin calves: implications for calf separation and mortality due to tuna<br />
purse-seine sets. Marine Mammal Science,<br />
23(1),<br />
15<strong>–</strong>29.<br />
Noren, S.R., Biedenbach, G. & Edwards, E.F. (2006) The ontogeny of swim<br />
performance and mechanics in bottlenose dolphins ( Tursiops truncatus).<br />
Journal of Experimental Biology,<br />
209(23),<br />
4724<strong>–</strong>4731.<br />
Noren, S.R., Biedenbach, G., Redfern, J.V. & Edwards, E.F. (2008) Hitching a<br />
ride: the formation locomotion strategy of dolphin calves. Functional<br />
Ecology.<br />
doi: 10.1111/j.1365-2435.2007.01353.x.<br />
Norris, K.S. & Prescott, J.H. (1961) Observations on Pacific cetaceans of<br />
Californian and Mexican waters. University of California Publications in<br />
Zoology, 63, 291<strong>–</strong>402.<br />
Pontzer, H. & Wrangham, R.W. (2006) Ontogeny of ranging in wild chimpanzees.<br />
<strong>International</strong> Journal of Primatology, 27(1), 295<strong>–</strong>309.<br />
Rohr, J.R. & Fish, F.E. (2004) Strouhal numbers and optimization of swimming<br />
by odontocete cetaceans. Journal of Experimental Biology, 207, 1633<strong>–</strong>1642.<br />
Ross, C. (2001) Park or ride? Evolution of infant carrying in primates. <strong>International</strong><br />
Journal of Primatology, 22(5), 749<strong>–</strong>771.<br />
Salvage, A., Snowden, P.T., Giraldo, L.H. & Soto, L.H. (1996) Parental care<br />
patterns and vigilance in wild cotton-top tamarins (Saguinus oedipus).<br />
Adaptive Radiation of Neotropical Primates (eds N.A. Norconk, A.L.<br />
Rosenberg & P.A. Garber), pp. 187<strong>–</strong>199. Plenum Press, New York.<br />
Schradin, C. & Anzenberger, G. (2001) Costs of infant carrying in common<br />
marmosets, Callithrix jacchus: an experimental analysis. Animal Behavior,<br />
62, 289<strong>–</strong>295.<br />
Skrovan, R.C., Williams, T.M., Berry, P.S. & Moore, P.W. (1999) The diving<br />
physiology of bottlenose dolphins (Tursiops truncatus) II. Biomechanics and<br />
changes in buoyancy at depth. Journal of Experimental Biology, 202, 2749<strong>–</strong><br />
2761.<br />
Taber, S. & Thomas, P. (1982) Calf development and mother<strong>–</strong>calf spatial<br />
relationships in southern right whales. Animal Behaviour, 30, 1072<strong>–</strong>1083.<br />
Tardif, S.D., Harrison, M.L. & Simek, M.A. (1993) Communal infant care in<br />
marmosets and tamarins: relation to energetics, ecology, and social organization.<br />
Marmosets and Tamarins: Systematics, Behavior, and Ecology (ed.<br />
A.B. Rylands), pp. 200<strong>–</strong>219. Oxford Press, Oxford.<br />
Tavolga, M.C. & Essapian, F.S. (1957) The behavior of the bottle-nosed<br />
dolphin (Tursiops truncatus): mating, pregnancy, parturition and mother<strong>–</strong><br />
infant behavior. Zoologica, 42, 11<strong>–</strong>31.<br />
Trivers, R. (1974) Parent-offspring conflict. American Zoologist, 14, 249<strong>–</strong>264.<br />
Webb, P.W. (1975) Hydrodynamics and energetics of fish propulsion. Bulletin<br />
of the Fisheries Research Board of Canada, 190, 1<strong>–</strong>159.<br />
Weibel, E.R., Taylor, C.R., Hoppeler, H. & Karas, R.H. (1987) Adaptive<br />
variation in the mammalian respiratory system in relation to energetic<br />
demand: I. Introduction to problem and strategy. Respiration Physiology,<br />
69, 1<strong>–</strong>6.<br />
Weihs, D. (2004) The hydrodynamics of dolphin drafting. Journal of Biology,<br />
3(8), 1<strong>–</strong>23 (http://jbiol.com/content/3/2/8).<br />
Weihs, D. (2006) Aerodynamic interactions between adjacent slender bodies.<br />
American Institute of Aeronautics and Astronautics Journal, 44, 481<strong>–</strong>484.<br />
Wells, J.P. & Turnquist, J.E. (2001) Ontogeny of locomotion in rhesus macaques<br />
(Macaca mulatta): II postural and locomotor behavior and habitat use in a<br />
free-ranging colony. American Journal of Physical Anthropology, 115, 80<strong>–</strong><br />
94.<br />
Williams, J.M., Hsien-Yang, L. & Pusey, A.E. (2002) Costs and benefits of<br />
grouping for female chimpanzees at Gombe. Behavioural Diversity in<br />
Chimpanzees and Bonobos (ed. C. Boesch), pp. 192<strong>–</strong>203. Cambridge University<br />
Press, Cambridge.<br />
Wrangham, R.W. (2000) Why are male chimpanzees more gregarious than<br />
mothers? A scramble competition hypothesis. Primate Males, Causes and<br />
Consequences of Variation in Group Composition (ed. P.M. Kaeppler),<br />
pp. 248<strong>–</strong>258. Cambridge University Press, Cambridge.<br />
Received 12 June 2007; accepted 27 September 2007<br />
Handling Editor: Francisco Bozinovic<br />
© 2007 The Author. Journal compilation © 2007 British Ecological Society, Functional Ecology,<br />
22,<br />
284<strong>–</strong>288
MARINE MAMMAL SCIENCE, 23(3): 629<strong>–</strong>649 (July 2007)<br />
C○ 2007 by the Society for Marine Mammalogy<br />
DOI: 10.1111/j.<strong>17</strong>48-7692.2007.00133.x<br />
SIMULATED VESSEL APPROACHES ELICIT<br />
DIFFERENTIAL RESPONSES FROM MANATEES<br />
JENNIFER L. MIKSIS-OLDS 1<br />
PERCY L. DONAGHAY<br />
Graduate School of Oceanography,<br />
University of Rhode Island,<br />
Narragansett, Rhode Island 02882, U.S.A.<br />
E-mail: jmiksis@gso.uri.edu<br />
JAMES H. MILLER<br />
Department of Ocean Engineering<br />
and<br />
Graduate School of Oceanography,<br />
University of Rhode Island,<br />
Narragansett, Rhode Island 02882, U.S.A.<br />
PETER L. TYACK<br />
Woods Hole Oceanographic Institution,<br />
Woods Hole, Massachusetts 02543, U.S.A.<br />
JOHN E. REYNOLDS, III<br />
Mote Marine Laboratory,<br />
1600 Ken Thompson Parkway,<br />
Sarasota, Florida 34236, U.S.A.<br />
ABSTRACT<br />
One of the most pressing concerns associated with conservation of the endangered<br />
Florida manatee is mortality and serious injury due to collisions with watercraft.<br />
Watercraft collisions are the leading identified cause of manatee mortality, averaging<br />
25% and reaching 31% of deaths each year. The successful establishment and<br />
management of protected areas depend upon the acquisition of data assessing how<br />
manatees use different habitats, and identification of environmental characteristics<br />
influencing manatee behavior and habitat selection. Acoustic playback experiments<br />
were conducted to assess the behavioral responses of manatees to watercraft approaches.<br />
Playback stimuli made from prerecorded watercraft approaches were constructed<br />
to simulate a vessel approach to approximately 10 m in sea grass habitats.<br />
Stimulus categories were (1) silent control, (2) approach with outboard at idle speed,<br />
(3) vessel approach at planning speed, and (4) fast personal watercraft approach.<br />
Analyses of swim speed, changes in behavioral state, and respiration rate indicate<br />
1 Present address of corresponding author: Jennifer L. Miksis-Olds, The Pennsylvania State University,<br />
Applied Research Laboratory, P. O. Box 30, State College, Pennsylvania 16804, U.S.A.<br />
629
630 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
that the animals responded differentially to the playback categories. The most pronounced<br />
responses, relative to the controls, were elicited by personal watercraft.<br />
Quantitative documentation of response during playbacks provides data that may<br />
be used as the basis for future models to predict the impact of specific human<br />
activities on manatees and other marine mammal populations.<br />
Key words: manatee, Trichechus manatus latirostris, disturbance, avoidance, playback<br />
experiment, watercraft approaches.<br />
A pressing question regarding the conservation of Florida manatees (Trichechus<br />
manatus latirostris) ishow to minimize both the lethal and nonlethal impacts on<br />
manatees in areas where watercraft operate. The lethal impacts of watercraft account<br />
for an average of 25% to more than 30% of identified manatee deaths each year<br />
(Ackerman et al. 1995, Reynolds 1999). The nonlethal impacts of boating on the<br />
physical health of manatees, as well as an indirect impact on food availability and<br />
communication, are controversial topics for which available data are inadequate.<br />
Identifying specific environments or behaviors that put manatees at a greater risk<br />
for boat collisions and quantifying manatee reactions to varying speeds and motor<br />
types of approaching vessels are necessary steps toward achieving the overall goal<br />
of minimizing the negative, lethal, and nonlethal impacts of boats and associated<br />
noise.<br />
Manatees most commonly encounter relatively small boats: outboard or inboard/outboard<br />
leisure boats, personal watercraft (PWC), and fishing trawlers<br />
(Gorzelany 2004). PWC means a vessel
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 631<br />
Specifically, swimming speed increased and animals moved toward deeper channel<br />
waters in response to boat approaches. The distance from the manatees to the boat,<br />
water depth at the boat, and water depth at the manatees each had a significant effect<br />
on swimming speed, whereas the type of boat or boat speed did not have a significant<br />
effect.<br />
Accurate detection of an approaching boat is only part of the problem for a manatee.<br />
In order to swim out of the direct path of the boat, the manatee must accurately<br />
localize the boat and respond appropriately. The physical characteristics of sound<br />
propagation in shallow water are complicated, and this can make it difficult to locate<br />
a sound source. From an acoustic standpoint, shallow water refers to areas where sound<br />
is propagated to distances at least several times the water depth, under conditions<br />
where both the surface and sediment boundaries have an effect on transmission (Urick<br />
1983). Compared to deep-water environments, there is greater attenuation of sound<br />
in shallow water environments (Medwin and Clay 1998). Higher frequencies have<br />
shorter wavelengths and are therefore more directional than lower frequencies with<br />
respect to a sound source of a given size. High levels of sound reverberation in shallow<br />
water also make localization difficult for even those species that possess extremely<br />
accurate localization ability. For example, bottlenose dolphins (Tursiops truncatus),<br />
which have excellent acoustic localization capabilities, have been hit by boats in shallow<br />
water (Wells and Scott 1997, Buckstaff 2004). These observations suggest that<br />
successful localization of approaching boats is necessary, but not a sufficient component,<br />
for dolphins to safely avoid approaching watercraft. Localizing approaching<br />
boats may also be an important component for manatees to avoid collision.<br />
Manatees that do manage to avoid collisions with watercraft still need to cope<br />
with other effects of boats. Boats that produce noise over the same frequencies as<br />
manatee vocalizations (Nowacek et al. 2003) can potentially mask communication<br />
signals. Motor noise can potentially mask communication signals. For example, if<br />
boat noise interferes with communication, females may lose contact with their calves<br />
or other members of a group, which could affect survival of the calves (Bengtson and<br />
Fitzgerald 1985). Manatees also are displaced from critical habitats with chronic boat<br />
disturbance (Provancha and Provancha 1988, Buckingham et al. 1999). Heavy vessel<br />
traffic may also cause manatees to expend more energy than they normally would<br />
(Reynolds 1999). For example, manatees significantly increase swim speed and move<br />
from shallow habitats to deeper channels during boat approaches (Nowacek et al.<br />
2001a). Based on these findings, as well as observations of behavioral responses to<br />
opportunistic vessel approaches prior to the initiation of this study, three categories of<br />
response (slow swim, fast swim, and rolling dive) were defined prior to the initiation<br />
of the study. The magnitude and diversity of manatee responses to vessel traffic clearly<br />
indicate the need for playback experiments to demonstrate manatee responses to the<br />
acoustic component of vessel approaches.<br />
Stimulus Recordings<br />
MATERIALS AND METHODS<br />
All stimulus recordings were made using a single hydrophone suspended from<br />
a recording vessel anchored in a sea grass habitat adjacent to a boating channel.<br />
The recording hydrophone was a HTI-99-HF hydrophone with built-in preamplifier<br />
and had a 2 Hz<strong>–</strong>125 kHz frequency range and −<strong>17</strong>8 dB re 1V/Pa sensitivity.<br />
The recording system was a National Instruments PCMCIA DAQ Card-6062E used
632 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
Table 1. Characteristics of playback stimuli<br />
Category Exemplar<br />
Received level at<br />
10 m (peak SPL) Speed (mph) Motor size<br />
Idle 1 150 dB 5 115 hp<br />
2 151 dB 5 100 hp<br />
Planing 1 168 dB 35 115 hp<br />
2 163 dB 35 100 hp<br />
PWC 1 166 dB 40 1,235 cc<br />
2 158 dB 25 1,235 cc<br />
3 157 dB 25 1,235 cc<br />
4 162 dB 40 1,235 cc<br />
in conjunction with a Dell Inspiron 8100. The recording system had an overall<br />
frequency response of 20 Hz<strong>–</strong>22 kHz with a −<strong>17</strong>8 dB re 1 V/Pa sensitivity at 16bit<br />
resolution. For each stimulus, the sound was recorded as a vessel approached the<br />
recording hydrophone at a constant speed from approximately 500<strong>–</strong>1,000 m away.<br />
The vessel approached to exactly 10 m away from the recording hydrophone and<br />
continued away at a constant speed for another 500<strong>–</strong>1,000 m. This method enabled<br />
calculation of received levels of the watercraft at 10-m distance (Table 1). The 10-m<br />
distance was the closest point of approach. Acoustic recordings of vessel approaches<br />
were made at two boat speeds: idle and full-throttle planing. Multiple exemplars<br />
from each category were recorded in order to ensure the generality of results and<br />
reduce pseudoreplication, defined as generalization from a study due to an animal<br />
responding to or learning from a single exemplar (Kroodsma 1989). Characteristics<br />
of each exemplar are summarized in Table 1. All vessels recorded possessed a 4-stroke<br />
motor, which provided an element of standardization across playback categories,<br />
regardless of engine size or vessel type. Two vessels were recorded for idle approaches.<br />
One vessel had a 115-hp outboard, and the other vessel had 100 hp. The same vessels<br />
were used to record two exemplars during a planing approach. Two separate 4-stroke<br />
PWCs were used to produce two planing approach recordings from each vessel for a<br />
total of four exemplars. Each PWC was recorded at two planing speeds: 25 mph (40<br />
km/h) and 40 mph (64 km/h; Table 1).<br />
The duration of all stimulus exemplars, regardless of stimulus category, was edited<br />
to 3 min to ensure the same length for each exemplar. This was done to control<br />
for the amount of time the playback subject was exposed to any sound coming<br />
from the playback system. Three-minute durations were chosen based on the longest<br />
audible approach sequence, which was the slow moving idle approach. For the idle<br />
approaches, 45 s of ambient noise preceded and followed the onset and offset of the<br />
stimulus signal, resulting in a total exemplar duration of 3 min. For planing and<br />
PWC approaches, ambient noise preceding and following the stimulus signal was<br />
appropriately included to produce a final stimulus duration of 3 min. The duration<br />
of ambient recording added prior to the stimulus onset was equal to that added after<br />
the signal offset. A silent control of 3 min was also constructed by creating a null<br />
vector (sound file containing 3 min of silence) at the same sample rate as the other<br />
stimuli. A silent control was selected to expose the subjects to any extraneous noise<br />
added by the transmit system that could potentially elicit a reaction. A control of only<br />
background noise was not included in the playback sessions because ambient noise<br />
preceded boat noise stimuli in all stimulus categories. If the animals were significantly
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 633<br />
Figure 1. Playback categories aligned in time. Top row (A) shows the acoustic envelope<br />
difference among the categories. The middle row (B) shows the raw time series of a single<br />
exemplar within each category. The bottom row (C) shows the corresponding spectrograms<br />
for each exemplar. Spectrograms are plotted on a relative dB scale. Received levels of each<br />
vessel type are presented in Table 1.<br />
reacting to the background noise, there would have been a significant reaction to all<br />
category approaches. There was no significant response to the idle approach, which<br />
indicated that the animals were not reacting to either the boat noise or the background<br />
noise.<br />
Playback Categories<br />
The exemplars in the playback categories varied in several acoustic parameters. The<br />
stimulus categories differ not only in their acoustic envelopes, or overall amplitude<br />
shape, but also in their frequency characteristics (Fig. 1). The idle approaches had<br />
the longest stimulus duration but the lowest stimulus amplitude (Fig. 1a, b). The<br />
planing approaches had a more rapid rise time and higher amplitude than the idle<br />
approaches, but with a shorter acoustic envelope. The planing approaches also had a<br />
more gradual onset compared to the abrupt offset. The acoustic envelope of the PWC<br />
approaches was the shortest with the most rise time and approximately equal peak<br />
amplitude as the planing approaches.<br />
The differences in frequency parameters were most evident in the spectrograms<br />
of Figure 1c. The idle approaches lacked a clearly defined broadband peak at the<br />
closest point of approach, and the U-shaped bands indicated the beta effect of sound<br />
during the approach and retreat (Tang 2005). The beta effect results in a decrease<br />
in signal frequency prior to the vessel’s closest point of approach and an increase<br />
in signal frequency during the retreat. From the perspective of the manatee, the<br />
beta effect could provide information about the position of the vessel. The planing<br />
and PWC approaches had a clear broadband peak at the closest point of approach.
634 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
This was preceded by a strong tonal, harmonic signal. In the PWC approaches,<br />
the tonal component of the approach was vastly reduced compared to that of the<br />
planing approaches, and the broadband peak of the PWC approach was also much<br />
narrower. Figure 2 further illustrates the difference in frequency spectra among the<br />
playback categories. Fifteen seconds prior to the closest point of approach, the PWC<br />
was approximately 10 dB quieter than both the idle and planing signals at 2<strong>–</strong>3<br />
kHz (Fig. 2a). The magnitude of the idle and planing approaches was similar up<br />
to approximately 6 kHz. Above 6 kHz, the planing approach became about 10 dB<br />
louder than either the idle or PWC signals. Fifteen seconds prior to the closest<br />
point of approach, the planing approach transmitted the loudest signal in the higher<br />
frequencies. The planing approach was also clearly the loudest at the peak of approach<br />
by about 10 dB (Fig. 2b), with the exception of the PWC being the loudest between<br />
2 and 3 kHz. The idle approach was the quietest at all frequencies during the peak<br />
of approach.<br />
To better understand what the manatees detected during the vessel approaches,<br />
the power spectra of approaches were weighted by the only available manatee hearing<br />
thresholds as measured by Gerstein et al. (1999) (Fig. 3). Prior to the closest point of<br />
approach (Fig. 3a), planing signals were most salient above the background noise. The<br />
idle approach was above both the hearing threshold and noise floor for frequencies of<br />
approximately 2 kHz, whereas the PWC signature was either at or below detectable<br />
levels. At the closest point of approach (Fig. 3b), both the planing and PWC signals<br />
were well above threshold levels for all except the very lowest frequencies. The idle<br />
approach signal was loudest at approximately 2 kHz at the closest point of approach,<br />
but this level is 5 dB quieter than the planing and PWC levels over a band of 20 Hz<br />
to 20 kHz.<br />
Playback Experiments<br />
All playbacks were conducted in grass bed habitats that are commonly occupied<br />
by manatees. Restricting playbacks to one habitat type eliminated a confounding<br />
variable due to habitat type in the statistical analysis. In addition, playbacks were<br />
only performed with animals that were initially feeding or resting. The animals<br />
were relatively stationary during these two behaviors compared to traveling, milling,<br />
or social behaviors; therefore, changes in movement could be more easily observed<br />
and quantified. Restricting playbacks to two specific behavioral states also served to<br />
reduce the number of categories of analysis for statistical purposes, thus increasing<br />
the degrees of freedom within each category for each test.<br />
The playback protocol was designed to simulate a boat approaching a manatee to<br />
10 m. The same positional set-up was used for each playback regardless of which<br />
stimulus was used. The playback vessel was always positioned between the playback<br />
subject and the closest boating channel to simulate a boat approaching from deeper<br />
water and from the direction that a majority of boats would be traveling. Transmitted<br />
levels were adjusted for animal position so that the subject animal received<br />
appropriate received levels, as a function of stimulus type and exemplar (Table 1),<br />
regardless of the manatee’s distance to the playback vessel, which was between 2<br />
and 25 m from the manatee. Projection from a single, stationary transducer did not,<br />
however, allow for a directional motion component to the playback recording. If<br />
the manatees can localize the source accurately, the distance between the manatee<br />
and the playback vessel could have impacted the behavioral response. Due to the
A<br />
Power spectrum (dB re 1 Pa 2 /Hz)<br />
B<br />
Power spectrum (dB re 1 Pa 2 /Hz)<br />
130<br />
120<br />
110<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 635<br />
Power Spectrum of Vessel Approach 15 s Before Peak<br />
30<br />
0 5 10 15 20 25<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
Frequency (Hz)<br />
Idle<br />
Plane<br />
PWC<br />
Power Spectrum for Vessel Approaches at Closest Point of Approach<br />
130<br />
Idle<br />
120<br />
Plane<br />
PWC<br />
110<br />
30<br />
0 5 10 15 20 25<br />
Frequency (Hz)<br />
Figure 2. Power spectra of vessel approach stimuli: (a) the comparison of 2-s clips measured<br />
15 s before the closest point of approach; (b) the comparison of 2-s clips taken at the closest<br />
point of approach.
636 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
A<br />
Weighted Spectrum (dB)<br />
B<br />
Weighted Spectrum (dB)<br />
130<br />
120<br />
110<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
Weighted Spectrum for Vessel Approaches 15 s Before Peak<br />
30<br />
0 5 10 15 20 25 30 30<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
Frequency (kHz)<br />
Idle<br />
Plane<br />
PWC<br />
Noise<br />
Hearing Sensitivity<br />
Weighted Spectrum for Vessel Approaches at Closet Point of Approach<br />
130<br />
130<br />
120<br />
Idle<br />
Plane<br />
PWC<br />
120<br />
110<br />
Noise<br />
Hearing Sensitivity<br />
110<br />
100<br />
100<br />
30<br />
0 5 10 15 20 25 30 30<br />
Frequency (kHz)<br />
Figure 3. Spectra of vessel approaches weighted by the manatee hearing thresholds and<br />
smoothed in frequency plotted with ambient noise levels: (a) comparison of 2-s clips measured<br />
15 s before the closest point of approach; (b) comparison of 2-s clips taken at the closest point of<br />
approach. For illustration purposes the manatee hearing sensitivity values for the frequencies<br />
measured in this paper are shown by the dashed-dotted line (data taken from Gerstein et al.<br />
1999).<br />
130<br />
120<br />
110<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
Hearing Sensitivity (dB re 1 µPa)<br />
Hearing Sensitivity (dB re 1 µPa)
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 637<br />
logistical constraints of the study design, this impact could not be addressed. Additional<br />
tests with stimuli simulating a rapid change in either direction or speed during<br />
the approach would provide valuable information that this study design was unable to<br />
offer.<br />
During playback experiments, the focal animal (occurring as a single animal or<br />
in a pair) was observed for a minimum of 20 min before and after the exposure of<br />
the playback stimuli. Primary observers were blind to the stimulus presentation, as<br />
the transducer and observer were on opposite sides of the boat and the background<br />
noise in the boat masked the in-air component of the playback stimulus, which was<br />
audible at the amplifier. During this time the subject was either identified from<br />
previously catalogued animals or photographed for later identification. Playbacks<br />
were only conducted if no other animals were detected within visual range during<br />
the 20-min pre-exposure period. Pre- and poststimulus observation included 4-min<br />
interval sampling of focal animal course, heading, distance to boat, and behavioral<br />
state. Ventilation and vocalization rates were recorded continuously. Vocalization<br />
rates were not analyzed, however, because the transducer was on the same vessel as<br />
the recording hydrophone, and the playback saturated the hydrophone recording rendering<br />
the manatee vocalizations inaudible. Saturated vocalization recordings during<br />
the playback sessions did not allow for uninterrupted vocalization rate analysis. If<br />
the animals left the area during the pre-exposure period or during exposure to the<br />
playback stimuli, they were not followed.<br />
Five minutes prior to the playback of any stimulus, the research vessel was anchored<br />
within 25 m of the focal animal and the sound source was deployed. An Aiwa model<br />
XP-V516C compact disc player connected to a Rockwood Detonator AMP-400 CRX<br />
amplifier delivered sound to a Lubell 9162 transducer, which projected the playback<br />
stimuli. This system was capable of producing a source level of approximately 190 dB<br />
re 1 Pa at 1 m in the frequency range of 240 Hz to 20 kHz. Response due to multiple<br />
boat interactions was avoided by only performing a playback when no other vessel<br />
had entered a 1-km radius for a 15-min period prior to the start of playback session.<br />
Each playback session consisted of four playback stimulus presentations presented<br />
5 min apart. The 5-min stimulus presentation was chosen based on the knowledge<br />
that bottlenose dolphins in Sarasota Bay encounter a passing vessel every 6 min<br />
(Buckstaff 2004). Manatees would experience a similar or shorter interval between<br />
boat encounters depending upon the number of recreational vessels in the area at any<br />
particular time. One stimulus from each of the four categories (control, idle, plane,<br />
PWC) was presented in random order. If the focal animal moved outside of a 25-m<br />
radius before the presentation of the last stimulus, the research vessel re-anchored<br />
closer to the animal, and the remaining playback stimuli were presented followed by<br />
a 20-min poststimulus observation period. In an effort to reduce pseudoreplication,<br />
no manatee was a playback subject more than twice, and no animal ever received the<br />
same exemplar more than once (Table 2).<br />
During the playback session, observations included point sampling of focal animal<br />
course, heading, distance to boat, ventilation, and behavior at the time of each<br />
surfacing. Visually observed responses to the playback stimuli generally fell into four<br />
categories: (1) investigate boat, (2) slow swim, (3) rolling dive, and (4) fast swim. Animals<br />
investigating the boat swam directly to the boat and interacted with it in some<br />
way. Slow swims were characterized by the animals changing position relative to the<br />
playback vessel without any visible wake or fluke prints. Rolling dives were identified<br />
by the arching of the manatee’s back and entire fluke leaving the water prior to a<br />
dive. Fast swims were characterized by a visible wake and strong fluke prints, often
638 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
Table 2. Summary of playback subjects, stimulus sequence, and response. In the “age” and “group composition” categories, A indicates adult, SA<br />
indicates subadult, C indicates calf<br />
Number Group Stimulus 1 Stimulus 2 Stimulus 3 Stimulus 4<br />
Session Date ID Age in Group composition Response Response Response Response<br />
1 4 June 2004 Snuff A 2 A, A C Idle2 PWC4 —<br />
(SB047) None None FS, LA —<br />
2 10 June 2004 Phish LA A 1 Plane2 C Idle2 PWC2<br />
SS None SS RD<br />
3 15 June 2004 Phish A 1 PWC2 Plane1 Idle2 C<br />
SS None None None<br />
4 21 June 2004 DU723 A 2 A, C Plane2 Idle2 C PWC1<br />
SS None None FS<br />
5 23 June 2004 623BB A 2 A, A C Idle1 Plane1 PWC3<br />
None SS FS FS, LA<br />
6 25 June 2004 625A A 2 A, C PWC4 Idle2 Plane2 C<br />
FS SS SS None<br />
7 28 June 2004 628A A 2 A, C PWC1 C Plane1 Idle1<br />
FS SS RD None<br />
8 1 July 2004 DU723 A 2 A, C C Idle1 PWC2 Plane1<br />
None SS FS SS<br />
9 6 July 2004 706AA SA 1 PWC4 Idle1 C Plane1<br />
RD None None None<br />
10 21 July 2004 721A A 2 A, C Idle1 Plane2 C PWC1<br />
None None None FS, LA
Table 2. Continued.<br />
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 639<br />
11 22 July 2004 Clyde A 2 A, A C PWC4 Plane2 Idle2<br />
None SS None None<br />
12 23 July 2004 723B A 2 A, C Plane2 C PWC3 Idle1<br />
RD None RD RD<br />
13 23 July 2004 723A A 2 A, C PWC2 C Plane2 —<br />
RD None SS, LA —<br />
14 26 July 2004 726D A 2 A, C Plane1 Idle2 PWC1 C<br />
FS SS FS None<br />
15 26 July 2004 Phish LA A 2 A, A C Idle1 PWC4 Plane1<br />
None None FS SS, LA<br />
16 27 July 2004 Splotch A 2 A, A PWC2 Idle1 C Plane1<br />
FS None None None<br />
<strong>17</strong> 28 July 2004 727AA SA 1 C Idle2 — —<br />
SS None — —<br />
18 3 August 2004 803BB A 2 A, SA Plane2 C Idle1 PWC3<br />
RD None RD SS, LA<br />
19 4 August 2004 804C SA 2 SA, SA C PWC4 Idle2 Plane1<br />
None FS None None<br />
20 5 August 2004 805BB A 2 A, A Idle1 Plane1 PWC1 C<br />
None FS FS SS<br />
21 19 August 2004 823F SA 1 Idle2 PWC2 Plane1 C<br />
None SS None SS
640 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
accompanied by a mud cloud stirred up from the bottom. Two retreating behaviors<br />
were also observed in relation to the playback vessel: (1) retreat to deep water and<br />
(2) pass by boat on the way to deep water. Direct retreats were characterized by the<br />
animal increasing distance from the playback vessel in the direction of deeper water.<br />
Passing by the boat on the way to deep water was a separate classification because<br />
the animal had to initially approach the playback vessel before retreating to deeper<br />
water. Passing by the boat on the way to deep water differed from the “investigate<br />
boat” response because the passing by the boat did not include any direct contact or<br />
interaction with the playback vessel by the manatee.<br />
In addition to behavioral responses, a respiratory response to the playback stimuli<br />
was also calculated. An index of ventilation variability (index of variability) was<br />
calculated from the time series of visually observed surface breaths recorded continuously<br />
throughout the playback experiments. The poststimulus index of variability<br />
was calculated by adding the absolute values of the difference in time between breaths<br />
for the three consecutive breaths following the stimulus onset (poststimulus index<br />
of variability =|t 0 − t 1| +|t 1 − t 2| +|t 2 − t 3|). Higher values of poststimulus<br />
variability indicated a larger variability in ventilation. Traditional measures of ventilation<br />
rate and its associated variance were not utilized in this analysis because long<br />
submersions followed by quick successive breaths cancelled each other and masked<br />
the true breathing pattern. Inclusion of absolute values preserved the level of variability<br />
between breaths. Prestimulus index of variability was calculated in a similar<br />
manner. The index was calculated from a time series of triads consisting of three consecutive<br />
breaths preceding the first stimulus. The triad values were then averaged to<br />
produce a prestimulus index of variability (prestimulus index of variability ={(|p 0<br />
− p 1|+|p 1 − p 2|+|p 2 − p 3|)}/n). Multiple, consecutive triads were used in the<br />
prestimulus index in order to establish the most accurate baseline index of variability.<br />
The poststimulus index only contained a single triad because this was the minimum<br />
number of surfacings observed between stimulus presentations during the playback<br />
session.<br />
Statistical Analysis<br />
The first level of analysis compared the frequency of animals either responding<br />
or not responding to each playback category and the control condition. Statistical<br />
significance was based on the binomial distribution. Differences in the frequency of<br />
behavioral responses between categories were investigated with a R × C G-test for<br />
independence. Based on behavioral observations to opportunistic vessel approaches,<br />
the three categories of response (slow swim, fast swim, and rolling dive) were defined<br />
prior to the initiation of the playback study. A priori conditions were met, which<br />
justified the use of the R × C G-test for independence used to assess the frequency<br />
of behavioral responses to the playback stimuli. Differences in the deviation between<br />
prestimulus and poststimulus index of variability across playback categories were<br />
tested with a single-factor analysis of variance (ANOVA).<br />
RESULTS<br />
Twenty-one playback sessions were conducted with 19 different subjects encountered<br />
either as single animals or in pairs (Table 2). A total of eighty stimuli were<br />
played: 21 Control, 10 Idle1, 10 Idle2, 11 Planing1, 8 Planing2, 6 PWC1, 5 PWC2,<br />
3 PWC3, and 6 PWC4. Seventeen out of 21 animals (81%) showed no locomotor
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 641<br />
Figure 4. Proportion of responses elicited by playback treatments.<br />
response to the silent control (Table 2 and Figure 4). The difference between response<br />
or no response for all stimuli vs. the silent control was significant at the 95% significance<br />
level based on a binomial distribution (P < 0.001). This indicated that the<br />
manatees were not significantly reacting to the exposure of the broadcasting system<br />
provided by the control stimulus. Manatees showed a marked locomotor response<br />
to the playback stimuli compared to the silent control. Thirteen of twenty animals<br />
(65%) showed no response to the idle approach, whereas 35% showed some type<br />
of locomotor response (i.e., slow swim, approach boat, fast swim, etc.) (Figure 4).<br />
This 35% response rate was not significant at the 95% significance level. During<br />
the planing approaches there was a significant locomotor response rate of 63% (P =<br />
0.002). Of the twelve animals that showed a locomotor response to the planing approach,<br />
two abandoned the area. All animals showed a locomotor response to the<br />
PWC approach (P < 0.001). Four of the twenty animals that responded (20%) left<br />
the area.<br />
An analysis of response orientation and heading of those animals that did show<br />
a locomotor response to the playback stimuli revealed a striking pattern (Fig. 4).<br />
Of the four animals that did respond to the control, all four investigated the boat.<br />
Seven animals of twenty responded to the idle approach. Of the seven, four (57%)<br />
retreated directly to deep water, two (29%) passed by the playback vessel on the way<br />
to deeper water, and one animal (14%) retreated from the playback vessel to shallow<br />
water. During the planing and PWC approaches, the number of animals retreating<br />
directly to deep water increased, whereas the number of animals passing by the boat<br />
decreased. In general, manatees tended to respond to all approaches by retreating to<br />
deep water. The frequency of animals retreating directly to deep water increased in<br />
response to an increase in speed of the approaching vessel.<br />
Behavioral analysis of the retreating animals showed a graded response in behavior<br />
associated with playback category (Fig. 5). No animals retreated during the controls,<br />
so this category was not included in the analysis. The frequency of animals retreating<br />
with a slow swim decreased from 71% in response to the idle approach to 37% for the
642 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
Figure 5. Behavioral response of retreating manatees to the playback categories. There<br />
is no response for the controls because no animals retreated during control presentations. ∗<br />
shows the significant decrease in the slow swim response associated with playback category<br />
(0.01 < P < 0.025). ∗∗ shows the significance increase in fast swim responses associated with<br />
playback category (0.01 < P < 0.025).<br />
planing approach to finally 16% in response to the PWC approach (Fig. 5a). Analysis<br />
of frequency using a R × C G-test for independence revealed that the frequency of<br />
the slow swim response was dependent on playback category (0.01 < P < 0.025;<br />
Sokal and Rohlf 1995). Similarly, the increase seen in the frequency of fast swim<br />
response was dependent upon playback category (0.01 < P < 0.025). No animals<br />
responded to the idle approach with a fast swim, whereas 37% and 68% responded<br />
to the planing and PWC approaches, respectively (Fig. 5b). There was slight decrease<br />
in the frequency of rolling dive responses to the PWC approaches, but this decrease<br />
was not a significant pattern.<br />
An ANOVA of the poststimulus index of variability showed a significant overall<br />
effect of stimulus type (F3,82 = 2.72, P = 0.04; Fig. 6). Post hoc multiple comparisons<br />
indicated an increase in variability between the prestimulus and both planing<br />
Figure 6. Mean deviation from prestimulus/control variability for each playback category.<br />
The asterisk ( ∗ ) indicates categories that differed significantly from the prestimulus/control<br />
values. Error bars represent standard error.
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 643<br />
and PWC responses. Post hoc multiple comparisons were performed for all other<br />
category comparisons, but no other comparisons were significant. This suggests that<br />
the manatees responded to the planing and PWC approaches in a similar manner<br />
with an increase in ventilation variability.<br />
DISCUSSION<br />
The most pronounced responses to the playback stimuli, relative to the controls,<br />
were elicited by the PWC. Significant behavioral and physiological responses were also<br />
seen in response to planing boat approaches, indicating that rapid vessel approaches do<br />
affect manatee behavior. Approaches by fast-moving vessels resulted in the disruption<br />
of feeding activity, an increase in energy expenditure inferred from swim speed, and<br />
in some cases a short-term avoidance of the feeding area. Avoidance reactions to<br />
approaching vessels are not unique to manatees, as disturbance responses to motorized<br />
vehicles have been documented in both marine and terrestrial species. In the marine<br />
environment, avoidance to motorized watercraft had been reported in manatees,<br />
cetaceans, and pinnipeds (manatees: Provancha and Provancha [1988], Buckingham<br />
et al. [1999], Nowacek et al. [2004a]; cetaceans: bottlenose dolphins, Janik and<br />
Thompson [1996], Nowacek et al. [2001b], Hastie et al. [2003], Buckstaff [2004];<br />
killer whales, Orcinus orca, Kruse [1991], Williams et al. [2002a, b]; Hector’s dolphins,<br />
Cephalorhynchus hectori, Bejder et al. [1999]; beluga whales, Delphinapterus leucas, Finley<br />
et al. [1990]; pinnipeds: walruses, Odobenus rosmarus, Fay et al. [1984]; and harbor<br />
seals, Phoca vitulina, Reijnders [1981]). Documented disturbance reactions include<br />
increases in vocalization rate, increases in swim speed, longer dive durations, decreased<br />
interanimal distance, increased breathing synchrony, and displacement from haulout<br />
sites. Terrestrial animals (bighorn sheep, Ovis canadensis canadensis, MacArthur<br />
et al. [1979]; white-tailed deer, Odocoileus virginianus, Richens and Lavigne [1978];<br />
caribou, Rangifer tarandus, Murphy et al. [1993]; and penguins, Pygoscelis adelia, Culik<br />
et al. [1990]) were also found to avoid road vehicles, snowmobiles, and aircraft.<br />
The manatees studied here showed the ability to discriminate and differentially<br />
react to the two different engine types and speeds simulated in the playback experiments.<br />
Nowacek et al. (2001a, 2004a) reported a generalized response by manatees<br />
to approaching boats involving turning toward or into deep water without specific<br />
regard to boat type, boat speed, distance from the manatee, the kind of habitat the<br />
boat was operating in, or the kind of habitat occupied by the manatee. Increases in<br />
swim speed were most prevalent in shallow water grassbeds when boats approached<br />
to between 0 and9m(Nowacek et al. 2004a). This study also showed that manatees<br />
reacted to simulated vessel approaches to within 10 m with an increase in swim<br />
speed and directed movement toward the closest deep water. The findings here differed<br />
from the previous study, however, because boat type and boat speed in this<br />
study appear to have a significant effect on swimming speed. This effect may not<br />
have been detected by Nowacek et al. (2004a) due to differences in study design and<br />
categorization of visible responses. This study differentiated between responses based<br />
on two different changes in swim speeds and the presence of rolling dives, which<br />
indicate deeper dives. Results presented here show that the manatees responded to<br />
slower idle approaches with a greater number of slow swim responses and a larger<br />
number of retreat paths that intersected with the playback vessel. In contrast, responses<br />
to fast approaching outboard motorboats or PWCs elicited a significantly<br />
greater frequency of response for fast swim speeds and retreat paths that avoided
644 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
the playback vessel. Because playback experiments only introduce the frequency and<br />
amplitude components of vessel signals (other potential components include visual<br />
information above and below water, and changes in the bearing of the acoustic signal),<br />
the frequency and amplitude information available to the animal prior to the closest<br />
point of approach can be used to explain how the animals may be discriminating and<br />
ultimately reacting to the different playback categories.<br />
Fifteen seconds prior to the closest point of approach, the planing approaches were<br />
approximately 10 dB louder than the idle and PWC approaches for frequencies from<br />
6to22kHz. Similarly, the idle approach was approximately 12 dB louder than both<br />
the planing and PWC approaches at 2 kHz. Therefore, the slower rise times of the<br />
idle and planing approaches provide more information to the animals 15 s prior to<br />
arrival compared to the PWC approach. It is possible the manatees can extract the<br />
necessary information from these acoustic cues relating to speed, direction, and boat<br />
type in order to execute the most appropriate and energetically favorable response.<br />
The most desirable responses are those that enable manatees to avoid collisions with<br />
watercraft. The playbacks provided an opportunity to observe responses to specific<br />
acoustic components of vessels approaching at a constant speed under controlled<br />
circumstances. Responses indicated that manatees are able to detect approaching<br />
vessels and execute appropriate responses to avoid vessel collisions by retreating to<br />
deep water where they can swim beneath the boat. Furthermore, manatees are able to<br />
discriminate between vessel types and speeds, which appeared to influence the degree<br />
and direction of response. Manatees tended not to respond to idle approaches. When<br />
animals did respond, most did so at a slow speed. By contrast, responses to PWC and<br />
planning approaches elicited a greater proportion of fast swims. Slower responses to<br />
slowly approaching vessels and quicker responses to faster-moving vessels would be<br />
appropriate in real-life situations, provided the vessels did not change speed. Manatee<br />
response is in marked contrast to that of right whales, which tend not to respond<br />
to approaching ships even though they do respond to alert signals (Nowacek et al.<br />
2004b). Right whales responded to an alert signal during controlled exposures by<br />
swimming strongly to the surface, which is a response that is likely to increase rather<br />
than decrease the risk of collision.<br />
The most energetically favorable response for a manatee to any vessel approach<br />
would be to minimize locomotor costs by not moving at all. However, this would<br />
not always be the most appropriate response in terms of avoiding vessel collisions. If<br />
a change in location is necessary, a response at swim speeds at or near the minimum<br />
cost of transport would be most efficient. It is often found that birds and mammals<br />
swim underwater at or near the speed of minimum cost of transportation (Williams et<br />
al. 1993, Ropert-Coudert et al. 2001, Lavvorn et al. 2004). Manatees generally cruise<br />
at speeds of 2<strong>–</strong>6 mph (3<strong>–</strong>10 km/h), although they have been recorded at speeds of<br />
15 mph (24 km/h) for short bursts (Hartman 1979). Speeds of 2<strong>–</strong>6 mph would have<br />
been classified into the slow swim response in this study, so it appears that manatees<br />
responding to idle and many planing approaches acquire enough prior information to<br />
execute an energetically efficient response. The PWC acoustic signatures 15 s prior<br />
to arrival do not provide as much acoustic information compared to the idle and<br />
planing approaches.<br />
The short rise time signal associated with PWC approaches does not differ greatly<br />
from ambient noise levels until 5 s before the peak, so it is possible that the<br />
manatees do not perceive these approaches in enough time to execute an energetically<br />
favorable response. Consequently, faster, less efficient responses are necessary to
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 645<br />
retreat from a possible PWC collision. An alternative explanation is that the sharp<br />
rise time associated with the PWC approach elicits a startle response that causes<br />
manatees to retreat from the sound source without evoking a higher level of cognitive<br />
analysis. Avoidance responses to the short rise time signals have also been<br />
observed in sharks. Myrberg et al. (1978) reported that a silky shark (Carcharhinus<br />
falciformis) withdrew 10 m from a speaker broadcasting a 150<strong>–</strong>600-Hz sound<br />
with a sudden onset and a peak sound pressure level of 154 dB re 1 Pa. These<br />
sharks also avoided a pulsed attractive sound when its sound level was abruptly increased<br />
by >20 dB. Finally, through prior encounters with PWCs, manatees may<br />
have learned to associate the PWC acoustic signal with vessels that are less predictable<br />
and more likely to approach them in shallow areas, thus requiring a more extreme<br />
response.<br />
Regardless of the specific acoustic characteristic of the fast vessel approaches that<br />
elicit fast swim responses, these signals cause manatees to increase their swim speed.<br />
Swim speed, as well as breathing rates and heart rate, have been used to estimate the<br />
energies of free-ranging marine mammals (Sumich 1983, Kshatriya and Blake 1988,<br />
Williams et al. 1992, Hind and Gurney 1997). Assuming that manatees respond at<br />
maximum speeds and that the maximum aerobic energy used during locomotion can<br />
reach 4<strong>–</strong>11 times resting levels in marine mammals (Elsner 1986, Williams et al.<br />
1993), consistent responses to vessel approaches potentially affect the energy budget of<br />
manatees in a significant way. Responses detected in this study are consistent with an<br />
energy cost, and future work quantifying energy expenditures will determine whether<br />
multiple reactions could have a long-term effect at the individual or population<br />
level.<br />
Most playback experiments measure movement, visual, or vocal responses to the<br />
sound played. Behaviors most typically measured are orientation or movement relative<br />
to the sound source, vocalizations made in response to the playback, and previously<br />
defined behaviors or displays such as aggressive or sexual displays. Less frequently<br />
used, but possibly more objective, responses are changes in heart rate (birds: Davis<br />
[1986], Diehl [1992]; humans: Brown et al. [1976]; chimpanzees, Pan troglodytes:<br />
Berntson and Boysen [1989]; bottlenose dolphins: Miksis et al. [2001]) and hormone<br />
levels (Dufty 1982). Neither quantitative swim speed nor fluke rate or amplitude<br />
was measurable in this study. More accurate measurements of both swimming characteristics<br />
and physiological responses during playback responses are necessary in<br />
order to determine the degree to which repeated exposure to vessel approaches are<br />
affecting the manatee energy budget or stress levels. Technological advances in tag<br />
construction and measurement sensors may soon allow for the recording of these<br />
critical parameters (Johnson and Tyack 2003).<br />
In summary, the playback technique presented here permits the investigation of<br />
numerous questions associated with manatee disturbance, threshold level, etc. without<br />
the risk of injury associated with the unpredictable behavior of wild animals<br />
during directed vessel approaches. This methodology has identified that vessel approaches,<br />
especially by PWCs and fast approaching watercraft, are a cause of manatee<br />
disturbance. This may be of regulatory concern, as harassment is prohibited in the<br />
Unites States by the Marine Mammal Protection Act. Manatees were also shown to<br />
hear and respond to boats approaching at idle speeds. Much more information is<br />
needed to determine how to minimize this disturbance in order to meet the criteria<br />
for species downlisting as outlined in the Florida Manatee Recovery Plan (U.S. Fish<br />
and Wildlife Service 2003). For example, the measured responses in this study were
646 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
only made during feeding and resting behaviors in grass bed habitats. Behavioral<br />
state and habitat type may have significant impacts on the motivation to react, and<br />
responses in other behavioral contexts (e.g., actively socializing) and habitats should<br />
be investigated.<br />
ACKNOWLEDGMENTS<br />
The success of this project would not have been possible without the help of numerous<br />
people. The interns and staff in the Manatee Research Program at Mote Marine Laboratory were<br />
critical to the field efforts. Special thanks are extended to Doug Nowacek for his input on initial<br />
project design. Thanks are also extended to Gretchen Hurst, Joe Gaspard, and Marie Chapla<br />
for their help with stimulus recordings. All playbacks were conducted under the guidelines<br />
of U.S. Fish and Wildlife permit MA07<strong>17</strong>99-0 issued to Jennifer Miksis and comply with all<br />
current laws of the country. This project was funded by a National Defense for Science and<br />
Engineering Graduate (NDSEG) Fellowship, P.E.O. Scholar Award, and American Association<br />
for University Women (AAUW) Dissertation Writing Fellowship awarded to Jennifer Miksis.<br />
LITERATURE CITED<br />
ACKERMAN, B.B.,S. D. WRIGHT, R.K.BONDE, D.K.ODELL and D. J. BANOWETZ.<br />
1995. Trends and patterns in mortality of manatees in Florida. Pages 1974<strong>–</strong>1992 in T.<br />
J. O’Shea, B. B. Ackerman and H. F. Percival, eds. Population biology of the Florida<br />
manatee (Trichechus manatus latirostris). National Biological Service, Information and<br />
Technical Report 1. Washington, DC.<br />
BEJDER, L., S. DAWSON AND J. A. HARRAWAY. 1999. Responses by Hector’s dolphins to<br />
boats and swimmers in Porpoise Bay, New Zealand. Marine Mammal Science 15:738<strong>–</strong><br />
750.<br />
BENGTSON,J.L.,AND S. M. FITZGERALD. 1985. Potential role of vocalizations in West Indian<br />
manatees. Journal of Mammalogy 66:816<strong>–</strong>819.<br />
BERNTSON,G.G.,AND S. T. BOYSEN. 1989. Specificity of the cardiac response to conspecific<br />
vocalizations in chimpanzees. Behavioral Neuroscience 103:235<strong>–</strong>243.<br />
BROWN, J.W.,P.A.MORSE, L.A.LEAVITT AND F. K. GRAHAM. 1976. Specific attentional<br />
effects reflected in the cardiac orienting response. Bulletin of the Psychonomic Society<br />
7:1<strong>–</strong>4.<br />
BUCKINGHAM,C.A.,L. W. LEFEBVRE,J.M.SCHAEFER AND H. I. KOCHMAN. 1999. Manatee<br />
response to boating activity in a thermal refuge. Wildlife Society Bulletin 27:514<strong>–</strong>522.<br />
BUCKSTAFF, K.C.2004. Effects of watercraft noise on the acoustic behavior of bottlenose<br />
dolphins, Tursiops truncatus, inSarasota Bay, Florida. Marine Mammal Science 20:709<strong>–</strong><br />
725.<br />
CULIK,B.M.,D. ADELUNG AND A. J. WOAKES. 1990. Effects of disturbance on the heart rate<br />
and behavior of Adelie penguins (Pygoscelis adeliae) during the breeding season. Pages<br />
<strong>17</strong>7<strong>–</strong>282 in K. R. Kerry and G. HEMPEL, eds. Antarctic ecosystems. Ecological change<br />
and conservation. Springer, New York, NY.<br />
DAVIS,W.J.1986. Acoustic recognition in the belted kingfisher: Cardiac responses to playback<br />
vocalizations. Condor 88:505<strong>–</strong>512.<br />
DIEHL, P.1992. Radiotelemetric measurements of heart rate in singing blackbirds (Turdus<br />
merula). Ornithology 133:181<strong>–</strong>195.<br />
DUFTY,A.M.1982. Responses of brown-headed cowbirds to simulated conspecific intruders.<br />
Animal Behaviour 30:1043<strong>–</strong>1052.<br />
ELSNER,R.1986. Limits to exercise performance: Some ideas from comparative studies. Acta<br />
Physiologica Scandinavica 128:45<strong>–</strong>51.<br />
FAY, F.H.,B. P. KELLY, P.H.GEHNRICH, J.L.SEASE AND A. A. HOOVER. 1984. Modern<br />
populations, migrations, demography, trophics, and historical status of the Pacific walrus.
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 647<br />
Pages 231<strong>–</strong>376 in Outer Continental Shelf Environmental Assessment Program, Final<br />
Report 693. NOAA, Anchorage, AK.<br />
FINLEY, K.J.,G. W. MILLER, R.A.DAVIS AND C. R. GREENE. 1990. Reactions of belugas<br />
and narwhals to ice breaking ships in the Canadian high arctic. Canadian Bulletin of<br />
Fisheries Aquatic and Sciences 224:97<strong>–</strong>1<strong>17</strong>.<br />
GERSTEIN,E.R.2002. Manatees, bioacoustics and boats. American Scientist 90:154<strong>–</strong>163.<br />
GERSTEIN, E.R.,L. GERSTEIN, S.E.FORSYTHE AND J. E. BLUE. 1999. The underwater<br />
audiogram of the West Indian manatee (Trichechus manatus). Journal of the Acoustical<br />
Society of America 105:3575<strong>–</strong>3583.<br />
GORZELANY, J.F.2004. Evaluation of boater compliance with manatee speed zones along<br />
the Gulf coast of Florida. Coastal Management 32:215<strong>–</strong>226.<br />
HARTMAN, D.S.1979. Ecology and behavior of the manatee (Trichechus manatus) in Florida.<br />
Special Publication 5, American Society of Mammalogists, Lawrence, KS.<br />
HASTIE, G.D., B. WILSON, L.H.TUFFT AND P. M. THOMPSON. 2003. Bottlenose dolphins<br />
increase breathing synchrony in response to boat traffic. Marine Mammal Science 19:74<strong>–</strong><br />
84.<br />
HIND, A.T.,AND W. S. C. GURNEY. 1997. The metabolic cost of swimming in marine<br />
homeotherms. Journal of Experimental Biology 200:531<strong>–</strong>542.<br />
JANIK, V.M.,AND P. M. THOMPSON. 1996. Changes in surfacing patterns of bottlenose<br />
dolphins in response to boat traffic. Marine Mammal Science 12:597<strong>–</strong>602.<br />
JOHNSON, M.P.,AND P. L. TYACK. 2003. A digital acoustic recording tag for measuring<br />
the response of wild marine mammals to sound. IEEE Journal of Ocean Engineering<br />
28:3<strong>–</strong>12.<br />
KETTEN,D.R.,D. K. ODELL AND D. P. DOMNING. 1992. Structure, function, and adaptation<br />
of the manatee ear. Pages 77<strong>–</strong>95 in J. A. Thomas, R. A. Kaestelein, and A. Y. Supin,<br />
eds. Marine mammal sensory systems. Plenum Press, New York, NY.<br />
KROODSMA, D.E.1989. Suggested experimental designs for song playbacks. Animal Behaviour<br />
37:600<strong>–</strong>609.<br />
KRUSE, S.1991. The interactions between killer whales and boats in Johnstone Strait, B.C.<br />
Pages 149<strong>–</strong>159 in K. Pryor and K. S. NORRIS, eds. Dolphin societies: Discoveries and<br />
puzzles. University of California Press, Berkeley, CA.<br />
KSHATRIYA, M.,AND R. W. BLAKE. 1988. Theoretical model of migration energetics in the<br />
blue whale, Balaenoptera musculus. Journal of Theoretical Biology 133:479<strong>–</strong>498.<br />
LAVVORN,J.R.,Y. WATANUKI,A.KATO,Y.NAITO AND G. A. LIGGINS. 2004. Stroke patterns<br />
and regulation of swim speed and energy cost in free-ranging Brunnich’s guillemots.<br />
Journal of Experimental Biology 207:4679<strong>–</strong>4695.<br />
MACARTHUR, R.A.,R. H. JOHNSON AND V. GEIST. 1979. Factors influencing heart rate in<br />
free-ranging bighorn sheep: A physiological approach to the study of wildlife harassment.<br />
Canadian Journal of Zoology 57:2010<strong>–</strong>2021.<br />
MEDWIN, H., AND C. S. CLAY. 1998. Fundamentals of acoustical oceanography. Academic<br />
Press, New York, NY.<br />
MIKSIS, J.L.,M. D. GRUND, D.P.NOWACEK, A.R.SOLOW, R.C.CONNOR AND P.<br />
L. TYACK. 2001. Cardiac responses to acoustic playback experiments in the captive<br />
bottlenose dolphin (Tursiops truncatus). Journal of Comparative Psychology 115:227<strong>–</strong><br />
232.<br />
MURPHY, S.M.,R. G. WHITE, B.A.KUGLER, J.A.KITCHENS, M.D.SMITH AND D. S.<br />
BARBER. 1993. Behavioral effects of jet aircraft on caribou in Alaska. Proceedings of the<br />
6th <strong>International</strong> Congress Noise as a Public Health Problem 3:479<strong>–</strong>486.<br />
MYRBERG, Jr.,A.A.,C.R.GORDON AND A. P. KLIMLEY. 1978. Rapid withdrawal from<br />
a sound source by open-ocean sharks. Journal of the Acoustical Society of America<br />
64:1289<strong>–</strong>1297.<br />
NOWACEK,S.M.,R. S. WELLS,D.P.NOWACEK,E.C.G.OWEN,T.R.SPEAKMAN AND R. O.<br />
FLAMM. 2000. Manatee behavioral responses to vessel approaches. Final contract report<br />
to the Florida Fish and Wildlife Conservation Commission. Mote Marine Laboratory<br />
Technical Report No. 742. Sarasota, FL. 12 pp. + appendices.
648 MARINE MAMMAL SCIENCE, VOL. 23, NO. 3, 2007<br />
NOWACEK,S.M.,R. S. WELLS,D.P.NOWACEK,E.C.G.OWEN,T.R.SPEAKMAN AND R. O.<br />
FLAMM. 2001a. Manatee behavioral response to vessel traffic. 14th Biennial Conference<br />
on the Biology of Marine Mammals. Vancouver, BC, Canada, 28 Nov, 3 Dec, 2001.<br />
NOWACEK,S.M.,R. S. WELLS AND A. R. SOLOW. 2001b. Short-term effects of boat traffic on<br />
bottlenose dolphins, Tursiops truncatus,inSarasota Bay, Florida. Marine Mammal Science<br />
<strong>17</strong>:673-688.<br />
NOWACEK, D.P.,B.M.CASPER, R.S.WELLS, S.M.NOWACEK AND D. A. MANN. 2003.<br />
Intraspecific and geographic variation of West Indian manatee (Trichechus manatus spp.)<br />
vocalizations (L). Journal of the Acoustical Society of America 114:66<strong>–</strong>69.<br />
NOWACEK, S.M.,R. S. WELLS, E.C.G.OWEN, T.R.SPEAKMAN, R.O.FLAMM AND<br />
D. P. NOWACEK. 2004a. Florida manatees, Trichechus manatus latirostris, respond to<br />
approaching vessels. Biological Conservation 119:5<strong>17</strong><strong>–</strong>523.<br />
NOWACEK, D.P.,M.P.JOHNSON AND P. L. TYACK. 2004b. North Atlantic right whales<br />
(Eubalaena glacialis) ignore ships but respond to alerting stimuli. Proceedings of the<br />
Royal Society of London B 271:227<strong>–</strong>231.<br />
PROVANCHA, J.A.,AND M. J. PROVANCHA. 1988. Long-term trends in abundance and<br />
distribution of manatees in the northern Banana River, Brevard County, FL. Marine<br />
Mammal Science 4:323<strong>–</strong>338.<br />
REIJNDERS, P.J.H.1981. Management and conservation of the harbour seal, Phoca vitulina,<br />
population in the international Wadden Sea area. Biological Conservation 19:213<strong>–</strong>221.<br />
REYNOLDS,J.E., III. 1981. Behavior patterns in the West Indian manatee, with emphasis on<br />
feeding and diving. Florida Scientist 44:233<strong>–</strong>242.<br />
REYNOLDS, J.E.1999. Efforts to conserve manatees. Pages 267<strong>–</strong>295 in J.R. Twiss and R.R.<br />
Reeves, eds. Conservation and management of marine mammals. Smithsonian Institution<br />
Press, Washington, DC.<br />
RICHARDSON, W.,C. GREENE, C.MALME AND D. THOMSON. 1995. Marine mammals and<br />
noise. Academic Press, San Diego, CA.<br />
RICHENS, V.B.,AND G. R. LAVIGNE. 1978. Response of white-tailed deer to snowmobiles<br />
and snowmobile trails in Maine. Canadian Field-Naturalist 92:334<strong>–</strong>344.<br />
ROPERT-COUDERT,Y.,K. SATO,J.BAUDAT,C.A.BOST,Y.LE MAHO AND Y. NAITO. 2001.<br />
Feeding strategies of free-ranging Adelie penguins Pygoscelis adeliae analysed by multiple<br />
data recording. Polar Biology 24:460<strong>–</strong>466.<br />
SOKAL,R.R.,AND F. J. ROHLF. 1995. Biometry. 3rd edition. W. H. Freeman and Company,<br />
New York, NY.<br />
SUMICH, J.L.1983. Swimming velocities, breathing patterns, and estimated costs of locomotion<br />
in migrating gray whales, Eshrichitus robustus. Canadian Journal of Zoology<br />
61:647<strong>–</strong>652.<br />
TANG, D.2005. Inverting for sandy sediment sound speed in very shallow water using boat<br />
noise. Journal of the Acoustical Society of America 118:2503.<br />
U.S. FISH AND WILDLIFE SERVICE. 2003. Florida manatee recovery plan, (Trichechus manatus<br />
latirostris), 3rd revision. U.S. Fish and Wildlife Service, Atlanta, GA.<br />
URICK, R.J.1983. Principles of underwater sound. 3rd edition. Peninsula Publishing, Los<br />
Altos, CA.<br />
WEIGLE,B.L., I. E. WRIGHT AND J. A. HUFF. 1994. Responses of manatees to an approaching<br />
boat: A pilot study. First <strong>International</strong> Manatee and Dugong Research Conference, 11<strong>–</strong><br />
13 March 1994, Gainesville, FL.<br />
WELLS, R.S.,AND M. D. SCOTT. 1997. Seasonal incidence of boat strikes on bottlenose<br />
dolphins near Sarasota, Florida. Marine Mammal Science 13:475<strong>–</strong>480.<br />
WILLIAMS, T.M.,W. A. FRIEDL, M.L.FONG, R.M.YAMADA, P.SEDIVY AND J. E. HAUN.<br />
1992. Travel at low energetic cost by swimming and wave-riding of bottlenose dolphins.<br />
Nature 355:821<strong>–</strong>823.<br />
WILLIAMS, T.M.,W.A.FRIEDL AND J. E. HAUN. 1993. The physiology of the bottlenose<br />
dolphin (Tursiops truncatus): Heart rate, metabolic rate, and plasma lactate concentration<br />
during exercise. Journal of Experimental Biology <strong>17</strong>9:31<strong>–</strong>46.
MIKSIS-OLDS ET AL.: SIMULATED VESSEL APPROACHES 649<br />
WILLIAMS,R.,D. E. BAIN,J.K.B.FORD AND A. W. TRITES. 2002a. Behavioural responses of<br />
male killer whales to leapfrogging vessel. Journal of Cetacean Research and Management<br />
4:305<strong>–</strong>310.<br />
WILLIAMS, R.,A.W. TRITES AND D. E. BAIN. 2002b. Behavioural responses of killer whales<br />
(Orcinus orca) to whale-watching boats: Opportunistic observations and experimental<br />
approaches. Journal of Zoology 256:255<strong>–</strong>270.<br />
Received: 9 May 2006<br />
Accepted: 5 February 2007
MARINE MAMMAL SCIENCE, 15(1): 102-122 (January 1999)<br />
0 1999 by the Society for Marine Mammalogy<br />
BEHAVIORAL SAMPLING METHODS FOR<br />
CETACEANS: A REVIEW AND CRITIQUE<br />
JANET MANN<br />
Department of Psychology and Department of Biology<br />
Georgetown University<br />
Washington, DC 20057, U.S.A.<br />
E-mail: mannj2@gunet.georgetown.edu<br />
ABSTRACT<br />
Behavioral scientists have developed methods for sampling behavior in or-<br />
der to reduce observational biases and to facilitate comparisons between stud-<br />
ies. A review of 74 cetacean behavioral field studies published from 1989 to<br />
1995 in Marine Mammal Science and The Canadian Journal of Zoology suggests<br />
that cetacean researchers have not made optimal use of available methodology.<br />
The survey revealed that a large proportion of studies did not use reliable<br />
sampling methods. Ad libitum sampling was used most often (59%). When<br />
anecdotal studies were excluded, 45% of 53 behavioral studies used ad libitum<br />
as the predominant method. Other sampling methods were continuous, one-<br />
zero, incident, point, sequence, or scan sampling. Recommendations for sam-<br />
pling methods are made, depending on identifiability of animals, group sizes,<br />
dive durations, and change in group membership.<br />
Key words: methods, sampling, observations, behavior, vocalization, Ceracea,<br />
ethology, quantitative.<br />
Researchers studying the behavior of cetaceans at sea face several unusual<br />
methodological challenges. Many cetaceans swim rapidly, range over long dis-<br />
tances on a daily basis, and have seasonal migrations of thousands of kilome-<br />
ters. Cetaceans are difficult to follow because they disappear during dives and<br />
do not leave long-lasting traces, such as tracks, scats, or dens.<br />
To meet these challenges, cetacean researchers have developed photoiden-<br />
tification methods and some clever technical solutions. For example, individ-<br />
uals of many species can be identified by natural markings, allowing research-<br />
ers to link sightings separated by years and thousands of kilometers (Ham-<br />
mond et al. 1990). Individual animals can be followed in more detail by<br />
tagging them with radio, acoustic, or satellite tags. Diving behavior can be<br />
monitored using time-depth recorders. Hydrophones and underwater video<br />
cameras can provide continuous data on the behavior and communication sig-<br />
nals of animals for shorter periods of time.<br />
In spite of these impressive technical advances, methods for sampling ce-<br />
102
MANN: BEHAVIORAL SAMPLING METHODS 103<br />
tacean behavior have received less attention. In this paper, I review quantitative<br />
observational methods typically used in cetacean studies and offer specific sam-<br />
pling suggestions.<br />
Altmann (1974) points out that most field observations of behavior involve<br />
sampling decisions, whether those decisions are explicit or not.<br />
“Sampling decisions are made wbenever the stua’ent of social behavior cannot continuously<br />
observe and record all of the behavior of all of the members of a social group. . , . We<br />
suspect that the investigator ofen chooses a sampling procedure without being aware that<br />
he is making a choice. Of course, he does not thereby escape fiom the consequences of that<br />
choice.” (Altmann 1974:229).<br />
The lack of an explicit protocol for sampling was termed ad libitum sam-<br />
pling by Altmann (1974), and she pointed out that it typically entails scoring<br />
“as much as one can” or whatever is most readily observable of the behavior<br />
of a group (Altmann 1974:235-236). These kinds of observations may be<br />
necessary as one learns to identify behaviors, as one plans a study, or as one<br />
observes rare but significant events. However, ad libitum observations suffer<br />
from a variety of potential biases. Different individuals may be more or less<br />
visible. Some behaviors may be more salient and more readily recorded than<br />
others. Individual animals may alter their behavior depending on how visible<br />
they are to other animals (and to observers). The same observer may concen-<br />
trate on different behaviors during different observation periods, and different<br />
observers may notice or attend to entirely different behaviors during the same<br />
observation period. Such biases indicate that ad libitum data are not appropriate<br />
for estimating rates of behaviors or for comparing rates across subjects or across<br />
studies. The selection and appropriate use of sampling methods that yield<br />
unbiased estimates of behavior are critical to the scientific validity of any study.<br />
In this paper, first I review the methods employed in the majority of recent<br />
cetacean field studies, including sampling method and protocols, and discuss<br />
the general advantages and pitfalls of each sampling method. Second, I rec-<br />
ommend specific sampling methods for cetaceans, depending on identifiability<br />
of animals, group sizes, dive durations, and change in group membership.<br />
Literature Survey<br />
To evaluate what methods are currently used by cetologists, I surveyed<br />
papers on cetacean behavior published from 1989 through 1995 (see Appendix<br />
1 for details). To limit the survey, I selected two peer-reviewed journals that<br />
together published the majority of studies in wild cetacean behavior, The Ca-<br />
nadian Journal of Zoology (CJZ, 31% of studies) and Marine Mammal Science<br />
(MMS, 29% of studies).<br />
Seventy-four studies were reviewed, 38 in CJZ and 36 in MMS. The fol-<br />
lowing information was noted: species, age and sex classes, number of animals,<br />
number of observation or recording hours, group-size minima and maxima,<br />
whether animals were individually identified, types of behaviors recorded (in-<br />
cluding vocalizations), definition of group and definition of behaviors, and
104 MARINE MAMMAL SCIENCE, VOL. 15. NO. 1. 1999<br />
methods of observation and sampling. In many cases, essential information in<br />
these categories was not provided (Appendix 1).<br />
CRITIQUE OF METHODS<br />
For every behavioral study, two basic kinds of sampling decisions must be<br />
made. One choice concerns which subject(s) one watches and for how long;<br />
the other concerns the details of how behavior is recorded (Martin and Bateson<br />
1986). I will discriminate these as “follow protocol” and the “sampling meth-<br />
od.” “Follow protocol” refers to how long an observation extends and to wheth-<br />
er researchers follow a group or an individual animal. “Sampling method”<br />
refers to the procedures used to sample the behavior of individuals or groups.<br />
Such a distinction is necessary because the sources of error or bias differ ac-<br />
cording to each protocol and according to each method. For example, in the<br />
ethological literature the term “focal-animal sampling” is used to refer to data<br />
collection that involves sampling of behavior of one individual for a set period<br />
of time. However, there are really two separate components, the follow protocol<br />
(following an individual) and a sampling method (which could be continuous<br />
sampling, point sampling, etc.). To state only that one’s method is “focal-<br />
animal sampling” would be insufficient. For example, a researcher could collect<br />
data on a focal individual systematically at regular intervals or opportunisti-<br />
cally (ad libitum ) by irregularly noting behaviors of interest.<br />
In the following sections, the prevalence and the costs and benefits of dif-<br />
ferent follow protocols and sampling methods are discussed. Table 1 is a guide,<br />
illustrating how different techniques (follow protocol and sampling method)<br />
are likely or unlikely to be effective, depending on the characteristics of the<br />
animals under study. My use of the term “identifiable” in Table 1 refers to<br />
cases when observers can identify known individuals (such as in a longitudinal<br />
study) or can discriminate between animals sufficiently to keep track of the<br />
same animal. When I refer to “group,” both for group size and group mem-<br />
bership, this concerns the number of animals close enough together to be<br />
potentially confused with each other.<br />
Follow Protocol<br />
For the studies surveyed, five different follow protocols can be defined:<br />
survey, group-follow, individual-follow, tracking, and anecdote. Three studies used<br />
more than one protocol (e.g., group-follows and surveys). Two studies did not<br />
provide enough information to allow classification by protocol.<br />
Survey refers to encountering groups or individual animals and staying with<br />
those animals for brief periods to census, for example, the number of animals,<br />
identifications, location, and behavior. If observers typically monitor groups<br />
for 30 min or less, then their protocol is identified here as “survey.” Surveys<br />
comprised 16% (n = 12) of the studies reviewed. Surveys provide a “snapshot”<br />
of animal life and are very useful for tracking patterns of association and for<br />
analyzing demographic, reproductive, and ecological factors. Surveys are par-
MA”: BEHAVIORAL SAMPLING METHODS 105<br />
Table 1. Recommended uses for different sampling methods depending on subject<br />
characteristics.<br />
Group<br />
Rapidly Group membership Dive time<br />
identifiable? size rate of change<br />
Method yes no*
106 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
Appropriate sampling techniques applicable during group-follows are dis-<br />
cussed below (see Table 1).<br />
If observers monitor an individual regardless of whether the animal is in a<br />
group or not, their protocol is classified as indiuidaal-follow. Only 12% of the<br />
studies reviewed used this protocol. The indiuidad-follow is roughly equivalent<br />
to focal-animal sampling. The critical feature of this method is to focus on<br />
one animal and systematically record behavior that is defined a priori. When<br />
sampling individual behavior, researchers may still collect ad libitam data on<br />
other animals or scan the group at regular intervals to determine group mem-<br />
bership. However, with few exceptions, data collection on non-focal animals<br />
should not compromise focal data.<br />
The individual-follow enables the observer to focus on the individual ani-<br />
mal’s “perspective.” What is the day in the life of that animal like? Who do<br />
they approach, avoid, stay close to, interact with, mate with, and fight with?<br />
Observing the continuous stream of individual behavior in different contexts<br />
is central to the understanding of the dynamics of social relationships. Time-<br />
budget data for individual animals are more appropriate for most studies than<br />
group-activity data because individuals of varying age and sex have different<br />
behavioral and ecological strategies. Both methodological and theoretical ar-<br />
guments support focusing on the individual as the unit of analysis. Selection<br />
operates at the level of the individual (Williams 1966), and this perspective<br />
is critical to understanding the behavioral and reproductive strategies of ce-<br />
taceans. Group-living cetaceans may rely on each other for survival and repro-<br />
duction, but the costs and benefits of group living are unlikely to be shared<br />
equally among all group members. Measures of such variation are particularly<br />
relevant to the study of how natural and sexual selection has shaped the evo-<br />
lution of social systems and behavior patterns in a species.<br />
The success of systematic sampling with the individual-follow depends on<br />
how rapidly animals can be identified or discriminated from other group mem-<br />
bers when they surface (Table 1). Several factors influence this, including the<br />
size and stability of groups and the duration of dives or “out-of-sight’’ periods.<br />
The optimal situation for individual-follows is that of a small stable group<br />
where individuals can be identified as soon as they come into view. If indi-<br />
viduals are not readily identifiable upon surfacing, individual-follows may still<br />
be used if the animals are solitary much of the time (e.g., humpback whales).<br />
Group size is also a factor. No matter how identifiable an animal is, if there<br />
are hundreds of animals, it will be hard to keep track of a focal animal (Table<br />
1). Brief focal samples, spanning 5 min or less (one or two surfacing bouts),<br />
may. be appropriate under such conditions. Individual-follows might be facil-<br />
itated by focusing on a readily identifiable member (i.e., based on size, marks,<br />
tags, or radiotracking), although the potential biases inherent to such selection<br />
must be considered.<br />
Although sampling decisions depend on the observation conditions, long<br />
individual focal-follows are highly recommended where possible. Rogosa and<br />
Ghandour (1991) have argued (mathematically) that short samples are one of<br />
the biggest sources of unreliability in sampling uncommon behaviors. Long
MANN: BEHAVIORAL SAMPLING METHODS 107<br />
individual follows may also be practical, given the search-time costs of sam-<br />
pling different individuals on the same day. If the researcher is likely to lose<br />
track of the focal animal in certain behavioral contexts (e.g., foraging), then<br />
scans or other sampling techniques may be applied to prorate or correct for<br />
such biases. Individual follows often enable the observer to identify the sources<br />
of sampling bias (e.g., conditions that prove difficult for monitoring individual<br />
behavior). If the study design requires following individuals for long periods<br />
of time, but individuals are not rapidly identifiable, then visual, radio, or<br />
acoustic tags may be necessary.<br />
Tracking refers to studies that electronically monitor individuals’ locations<br />
or behavior (through hydrophones, transponder tags, or other devices). Track-<br />
ing is particularly valuable if researchers need to continuously record the be-<br />
havior of an animal over long periods. Six studies (8%) reviewed here tracked<br />
small numbers of animals using radio and transponder tags to record diving<br />
and vocal behavior. This approach can be relatively expensive, and the attach-<br />
ment process or device may affect behavior. Sample sizes tend to be small in<br />
tracking studies.<br />
An anecdote is a descriptive report of a single event or series of events, such<br />
as birth or predation. Twenty-eight percent (n = 21) of the studies analyzed<br />
here were classified as anecdotes. Anecdotes are a valuable means of describing<br />
rarely observed events.<br />
The survey and group-follow may be considered “group-protocols”-where<br />
groups are monitored unless animals are alone. The individual-follow is an<br />
“individual-protocol”-where a specific individual is monitored regardless of<br />
whether or not it is a member of a group. Tracking can involve groups or<br />
individuals. The merits of different protocols must be weighed against their<br />
impact or influence on the animal’s behavior. Few data are available on whether<br />
or how tagging, equipment noise (e.g., engine, tag, depth sounder), and pro-<br />
longed vs. brief follows directly affect the behavior of cetaceans. Indirect effects<br />
may also occur by alteration aQ the behavior of cetacean prey or predators. By<br />
combining different approachies, or directly testing for the effects of these<br />
approaches, observers can find ways to minimize their impact on behavior.<br />
Sampling Methods<br />
Given that a researcher has, chosen a protocol for following the animals, a<br />
number of sampling methods may be used. These include ad libitum, contin-<br />
uous, fical-group, one-zero, point, scan, predominant activity, sequence, and all-event1<br />
incident sampling (defined below). The success of each sampling method de-<br />
pends, in part, on whether a researcher is focusing on events (brief behaviors,<br />
measured in frequency) or states (long behaviors of measurable duration). Al-<br />
though the distinction is convenient, events and states are on a continuum<br />
(ie., all events have durations and all states have frequencies).<br />
Ad libitum sampling-Ad libitum samples are “typical field notes” (Altmann<br />
1974). The observer writes down what seems of interest. Ad libitum sampling<br />
does not involve systematic constraints on what is recorded and when it is
108 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
recorded. The initial phases of behavioral study often involve some ad libitzlm<br />
sampling in order to delineate and define behaviors and research questions.<br />
Most observers continue to collect some ad libitum data throughout the course<br />
of their study. In the survey of studies, the sampling method was classified as<br />
ad libitum only if it was the predominant method used. Ad libitum sampling<br />
was the most common technique used (59% of studies). Ad libitum sampling<br />
is useful for certain kinds of comparisons, but not for estimating rates of<br />
behavior or for comparing behavior patterns of different age or sex classes<br />
(Altmann 1974). As Altmann points out, some animals may be more notice-<br />
able, either because of their behavior (or activity level), size, or level of habit-<br />
uation. Ad libitum sampling may be applied during individual focal-follows<br />
(e.g., prey captures by focal animal) or group-follows (e.g., displays). Ad libitum<br />
data are valuable for looking at the direction of interactions or at what happens<br />
within a dyad if the direction of the interaction is unlikely to be influenced<br />
by the probability of seeing it. For example, ad libitum data are commonly<br />
used in ethology to assess dominance relationships, where one animal “wins,”<br />
and the other “loses” an agonistic interaction (e.g., see Samuels and Gifford<br />
1997). Rates of agonistic interactions cannot be acquired through ad libitum<br />
sampling, because the likelihood of seeing an interaction may be affected by<br />
the rank, size, sex, or other characteristics of the animals (e.g., conflicts between<br />
adult males are more likely to escalate and thus be observed than conflicts<br />
between adult females). However, given that two animals are fighting, the<br />
outcome of the fight is unlikely to influence or be influenced by the probability<br />
of seeing the interaction. Ad libitum sampling is effective for examining di-<br />
rected displays, in which both signaller(s) and recipient(s) can be determined.<br />
Sociomatrices are an excellent means of illustrating some patterns of interac-<br />
tion with ad libitum data. A sociomatrix has all potential interactants listed<br />
on the top and along the side, with one axis indicating actor or “winner” and<br />
the other as recipient or “loser.” Frequencies are tallied within each cell of<br />
interactants. When dominance hierarchies are linear, for example, then most<br />
entries will fall on one side of the diagonal. Sociomatrices can be applied to<br />
other behaviors sampled ad libitum. Mann and Smuts (in press) recorded 1,3 1 1<br />
rubbing events between nine wild Tursiops newborns and their mothers during<br />
focal individual-follows (189 h). Rubbing rate could not be determined, nor<br />
whether some infants rubbed more than others. But rubbings were extremely<br />
asymmetrical, with infants initiating and performing 99% of the rubbing on<br />
their mothers.<br />
Ad libitum data are a valuable part of any field study but should not be<br />
represented as rates, proportions, frequencies, or other unbiased estimates of<br />
behavior. The use of ad libitum data for anecdotes is perfectly appropriate. Rare<br />
events such as predation, a birth, or a lethal fight, can provide critical infor-<br />
mation and insights, and ad libitum sampling is likely to be the only way such<br />
events are recorded.<br />
Continuous sampling-Alternate names: event sampling, frequency sampling,<br />
focal-animal sampling. Continuous sampling is a systematic record of fre-<br />
quencies or durations for a specified set of behaviors. Researchers occasionally
MANN: BEHAVIORAL SAMPLING METHODS 109<br />
use the term “focal-animal sampling,” to indicate continuous sampling of<br />
behavior during an individual-follow. Measuring the exact times (and dura-<br />
tions for behavioral states) of every occurrence of a behavior is very demanding<br />
for the observer. Scoring event frequencies within time blocks can simplify<br />
data collection. The reliability of continuous data can be easily compromised<br />
if the observer tries to record too many behaviors at once, especially when the<br />
animals are active. Thus, Altimann (1974) recommends that this method be<br />
used for one or two animals at most.<br />
Continuous data on associa.tions, diving behavior (with transponder tags),<br />
other behaviors, or vocalizations were collected in 14% (n = 10) of the studies<br />
surveyed. In four of these studies researchers observed animals directly. Two<br />
studies used tags to track individuals. The remaining four collected continuous<br />
data on small groups of animals using videotape or multiple observers.<br />
Continuous sampling of behavior is relatively simple for activities at the<br />
surface (during surfacing bouts). Surfacing-bout durations, breath frequency,<br />
dive types, surface-display rates, and synchronous surfacings of the focal animal<br />
and others may be recorded on a continuous basis for many dolphin and whale<br />
species. Even deep-diving animals, such as sperm whales, may have prolonged<br />
bouts of socializing at the surface (Whitehead and Weilgart 1991). If the<br />
activities are brief, or difficult to time, then they can be scored as events<br />
(frequencies), rather than states (onset to offset). Records of behavior sequences<br />
are typically intact in continuous sampling. As Table 1 indicates, continuous<br />
sampling is most likely to succeed when animals are rapidly identifiable, live<br />
in small groups, and dive for short periods. Under these sampling conditions,<br />
observers can keep track of a specific individual for long periods of time.<br />
Continuous data are the richest source of information on social behavior and<br />
relationships, because such dlata include information on details, sequences,<br />
actors and recipients, rates and durations of behavior for individual animals.<br />
While it is simple to record vocalizations continuously using a tape recorder,<br />
for these data to qualify as continuous data for an individual follow, the ob-<br />
server must be able to identify which vocalization comes from the focal in-<br />
dividual. Use of tags that can record or transmit acoustic data is one technique<br />
for achieving this (e.g,, Tyack 1985, Tyack and Recchia 1991). Another tech-<br />
nique involves passive acoustic localization (e.g. Clark 1980, Freitag and Tyack<br />
1993, Frankel et al. 1995). Fusion of these acoustic and visual data remains a<br />
challenge to anyone interested in cetacean communication. As with behavioral<br />
studies, the study of cetacean vocalizations still hinges upon such basic infor-<br />
mation as the species, sex, or age class of the animal producing the call. The<br />
study of how these sounds are used in social communication will require<br />
increased application of methods to identify which animal produces which<br />
sound during interactions wit:hin and between groups.<br />
Focal group sampling-Alternate names: focal subgroup sampling, group<br />
sampling, predominant group activity sampling. Focal-group sampling is a<br />
continuous assessment of group activity. Group activity may be scored at in-<br />
tervals (e.g., indicate predominant group activity every 5 min) or continuously<br />
(e.g., the group rested from 1122-1147). This method is widely used in ce-
110 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
tacean research when observers follow groups, and Altmann (1974) has been<br />
cited in support of it (e.g., Shane 1990). Altmann (1974) used the term “focal<br />
sub-group sampling” to refer to continuous sampling with more than one<br />
individual. However, Altmann described focal sub-group sampling as appro-<br />
priate only under a very restricted set of conditions “in which all individuals<br />
in the sample group are continuously visible throughout the sample period<br />
. . . if one is working with observational conditions that are less than perfect,<br />
focal-animal sampling should be done on just one focal individual at a time,<br />
or at most, a pair (e.g., mother and young infant)” (Altmann 1974:243-244).<br />
Twenty-seven percent of the studies surveyed here (45% of ad libitum studies)<br />
used “focal group sampling.” These studies scored “group behavior.” Dozens<br />
of animals, not pairs, were sampled using focal-group sampling. I chose to<br />
classify these studies as ad libitum because none of the conditions necessary for<br />
unbiased focal-group-sampling were met. That is, observers could not have<br />
continuously observed all animals equally in a group regardless of activity.<br />
Some observers appeared to visually assess group activity, perhaps by infor-<br />
mally scanning the group. However, this method is not explicit enough in<br />
how subjects were sampled and does not differ in any substantial way from<br />
ad libitum sampling.<br />
A few studies identified their method as predominant group-activity sam-<br />
pling. This should not be confused with predominant activity sampling (de-<br />
fined below). Predominant group-activity sampling is essentially the same as<br />
“focal group sampling” and was coded as ad libitum in the survey. During<br />
predominant group-activity sampling, the observer defines group activity<br />
based on an assessment of what most (>50%) of the group is engaged in over<br />
an interval, focusing on the proportion of individuals estimated to be engaging<br />
in a behavior, not the proportion of time an individual engages in a behavior for<br />
that interval. An estimate of predominant group activity can be achieved by<br />
explicitly scan sampling over 50% of the individuals in the group, rather than<br />
by “watching the group.”<br />
Altmann (1974) prescribed a restricted set of conditions for focal sub-group<br />
sampling because of the many biases inherent to watching groups. First and<br />
foremost, the observer’s attention is naturally drawn to animals that are most<br />
visible, most active, or most interesting. No observer can possibly continuously<br />
track and record all behavior of all individuals simultaneously under all con-<br />
ditions; it is difficult enough to record the behavior of one animal. Group<br />
activity may be determined if all members of a small group are cohesive and<br />
engaging in the same behavior, such as resting closely at the surface, but it<br />
may be difficult to determine if some animals are engaging in other activities.<br />
The accuracy of focal-group sampling is dependent upon group size, cohe-<br />
siveness, and activities of the animals, thus potentially introducing biases into<br />
data collection. Some behaviors, such as socializing, may be particularly ob-<br />
vious to observers, and even though most of the animals are resting, one might<br />
score the more visible behaviors as group activity. While using group-sam-<br />
pling, observers are often making implicit assumptions about the relative im-
MA”: BEHAVIORAL SAMPLING METHODS 111<br />
portance of behaviors of different age and sex classes. If two mothers are hunt-<br />
ing, but their calves are traveling, should one call it hunting or traveling?<br />
With video, continuous sampling of groups is possible because observers<br />
can rewind the tape and code the behavior of different individuals separately<br />
(thus making the process equivalent to conducting multiple continuous focal-<br />
animal samples simultaneously). However, simultaneous behaviors of different<br />
individuals within the same group are likely to depend upon one another,<br />
which must be taken into account for statistical analyses. Focal-group sam-<br />
pling is not only used during the group-follow protocol, it is also used during<br />
surveys to identify “group activity.” However, this too is subject to the same<br />
criticisms.<br />
One study (Fragaszy et a/. 1992) has compared focal-group sampling with<br />
focal-individual point sampling and scan sampling (see definitions below).<br />
This study, although on primates, is particularly relevant for cetacean studies<br />
because, like cetaceans, capuchins and squirrel monkeys move in and out of<br />
view and are difficult to follow for continuous periods of time. Fragaszy et al.<br />
(1992) compared foraging time budgets of squirrel monkeys using both focal-<br />
group sampling and focal-animal point sampling methods and concluded that<br />
focal-group sampling overestimated foraging considerably. They came to this<br />
conclusion by correlating individual-focal point sampling rates with focal-<br />
group sampling rates. The focal-group sampling error rate (computed by JM)<br />
ranged between 39% and 63%. They also compared scan sampling of capuchin<br />
monkey groups to focal-animal point sampling. Scan sampling fared much<br />
better. When time budgets for scan sampling and focal-animal point sampling<br />
for foraging among capuchins were compared, the error rate was only 0.8%<br />
(computed by JM). For other, less frequent behavioral states, the error rates<br />
for scan sampling ranged from 10% to 26%. Focal-group sampling error rates<br />
for other behavioral states were not compared. However, error rates are lowest<br />
for more prevalent behaviors, and foraging was the most common state. Thus,<br />
error rates are likely to increase for other states. Such error rates should give<br />
pause to anyone considering group-sampling.<br />
One-zero sumpling-Alternate names: time-sampling, Hansen frequencies, in-<br />
terval sampling, partial interval sampling, method of repeated short samples,<br />
and modified frequencies. One-zero sampling entails scoring whether or not a<br />
behavior occurs during an interval (e.g., 30 sec), rather than scoring how fre-<br />
quently or how long the behavior occurred. For example, an observer may<br />
score whether or not an animal vocalized during 30 seconds, or whether or<br />
not it rubbed, rather than the frequency of vocalization events or the duration<br />
of rubbing (a state). Nine percent of the studies surveyed employed one-zero<br />
sampling. With few exceptions, this technique is not recommended, because<br />
one-zero scores do not represent frequency or duration and are prone to very<br />
high rates of sampling error (see Mann et al. 1991). Although one-zero sam-<br />
pling has no recommended uses (Altmann 1974), there are some natural types<br />
of one-zero data (categorical variables) that are not represented by the flow of<br />
continuous behavior but may occur over very long intervals. For example, if<br />
group fissions and fusions are rare, then whether or not an animal joins a
112 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
group during a day or field season is a significant piece of information and<br />
may be more biologically meaningful than the absolute rate. However, one<br />
must already know something about the rates and durations of behaviors of<br />
interest before employing one-zero sampling.<br />
Researchers sometimes use one-zero events to define states. That is, one-<br />
zero data may assist in the development of an ethogram but not be part of<br />
the sampling method. For example, if two animals are seen in contact during<br />
an interval, this may “define” socializing regardless of the duration or fre-<br />
quency of contact. Or, one might define foraging based on whether or not the<br />
animal flukes-up at the dive. It is important then to distinguish between the<br />
use of one-zero (categorical) information to help define behaviors that are pre-<br />
sumed to be continuous and treating one-zero scores as state behaviors them-<br />
selves.<br />
Point sampling-Alternate names: instantaneous sampling, on-the-dot sam-<br />
pling, time sampling. Point sampling entails scoring activity as a “snapshot”<br />
at a given moment (e.g., every 30 sec). The observer may score states, such as<br />
distance to others, activity, or other information on a point-sampling basis.<br />
Five percent of the studies surveyed used point sampling. It is a reliable<br />
method that is widely applied in ethological studies but has been rarely used<br />
for cetaceans. One possible explanation is that the “point” often happens when<br />
the animal is submerged. The few studies reviewed here that used point sam-<br />
pling did so by electronically tracking the animal. For direct observations,<br />
point sampling can still be applied with two basic strategies. (1) The animal’s<br />
behavior is sampled at the first surfacing after the interval point. Two factors<br />
may lead to sampling biases. One ends up sampling surface behavior rather<br />
than the behavior occurring underwater at the point. By examining the rela-<br />
tionships between surface and subsurface activities, these biases may be min-<br />
imized. Another bias is that the behavior is not sampled at regular intervals,<br />
but at intervals determined in part by the subject’s surfacing and diving be-<br />
havior. If the observer sampled at 1-min intervals, and animals tended to<br />
remain close to the surface during socializing and dive deeply for longer pe-<br />
riods during foraging, then the observer could successfully sample socializing<br />
but would miss several sampling intervals during foraging. Social behavior<br />
would be overrepresented in relation to foraging. This problem can be alle-<br />
viated by establishing a point interval greater than the typical long dives of<br />
the subject. (2) The second strategy is to use point sampling for behavioral<br />
states if one can assume that the state continues when the focal animal moves in and<br />
out of view. Observers often infer behavioral state based on surfacing pattern,<br />
proximity of other individuals, and other cues (e.g., flukeprints or bubbles at<br />
the surface). The observer may apply a convention such as scoring a behavioral<br />
state at the “point” if the animal engaged in the behavior prior to diving or<br />
disappearing from view and was engaged in the same behavior upon surfacing.<br />
This convention may be applied if the “out-of-sight” periods tend to be much<br />
shorter than the bout duration of the behavioral state. Another convention is<br />
to score the behavior as that last observed, or, alternatively, first seen upon<br />
surfacing. Such conventions must be explicitly defined, justified, and the po-
MANN: BEHAVIORAL SAMPLING METHODS 113<br />
tential biases considered. As Table 1 suggests, point sampling can be applied<br />
under many conditions but becomes problematic if dive times or out-of-sight<br />
periods are long and group sizes are large (making it difficult to keep track<br />
of an individual animal).<br />
Point sampling can be a very useful method to determine time budgets or<br />
diurnal patterns of behavior. This method is also a useful supplement to other<br />
focal-individual sampling methods. For example, the observer may record an<br />
animal’s speed at 5-min intervals, or note a focal animal’s nearest neighbor<br />
based on who surfaces closest to the focal animal at the first surfacing after<br />
each 10-min interval. Although point sampling is typically recommended for<br />
behaviors of appreciable duration, activity can be more difficult to score on a<br />
point-sampling basis. In reality, even with the animal in full view, it may take<br />
several seconds or more to “decide” what an animal is doing. It is important<br />
that the observer decide as quickly as possible, so shelhe does not inadvertently<br />
wait until the animal does something that is easier or more interesting to<br />
score. Brief behaviors or events are difficult to observe accurately with point<br />
sampling because such behaviors are often missed at the “point.”<br />
Scan sampling-Scan sampling entails taking a “point” or “instantaneous”<br />
sample of an individual’s behavior or location before moving to the next animal<br />
and doing the same. Scans are conducted either at regular intervals (e.g., sample<br />
each animal at 10-sec intervals), or as quickly as possible (ie., search for the<br />
next animal as soon as the last was sampled). Both point and scan samples are<br />
good techniques for measuring states, but brief events are likely to be missed.<br />
Three percent of the cetacean field studies reviewed here used scan sampling.<br />
Scan sampling is very valuable for sampling behavior when focal-individual<br />
observations are not possible or desirable, or if the researcher wishes to keep<br />
track of group activities. An explicit scan-sampling technique should also be<br />
considered for surveys of groups, rather than using ad libitam sampling to<br />
identify “group activity.” Observers sometimes use a random scan-sampling<br />
schedule, or just scan from one side of a group to the other. Alternatively, if<br />
group sizes tend to be large and the animals move rapidly, but age or sex<br />
classes are distinctive in size or markings, then the researcher may scan age<br />
and sex classes separately (e.g., scan large males, then females with dependent<br />
calves, then immature animals). This would leave some agehex classes unsam-<br />
pled but would enable the observer to scan a group of fast-moving animals<br />
more effectively.<br />
Scan sampling and point sampling share practical difficulties: the time it<br />
takes to decide the animal’s activity and out-of-sight or diving periods. One<br />
may employ the same strategies for scan sampling as for point sampling, but<br />
with scan sampling the observer samples each individual only once per session.<br />
In some cases, the observer may need to watch each individual for 3 min or<br />
more (approximating a short individual-follow) but uses the midpoint (e.g.,<br />
minute 2) to define the behavioral state. Thus, one data point is taken for<br />
each animal in succession, although each observation period is longer than for<br />
a typical scan. The same conventions used for point-sampling out-of-sight<br />
periods can be adapted to scan sampling.
114 MARINE MAMMAL SCIENCE. VOL. 15. NO. 1. 1999<br />
Potential uses for scan sampling are varied (see Table 1). First, an observer<br />
may scan a group at fixed intervals (e.g., every 10 min) to assess which is still<br />
present, which is close to a focal animal, etc. Second, an observer may scan a<br />
group to rapidly estimate group activity by identifying each animal’s activity<br />
as it surfaces. This is only possible for groups where one can keep track of<br />
each individual within the group during surfacing and depends in part on the<br />
degree of interindividual overlap in surfacing bouts. The observer may be able<br />
to scan rapidly enough to minimize the likelihood of resampling the same<br />
individual or scan from the front to the back of a group if direction changes<br />
are infrequent. If the scan procedure inadvertently allows for repeat sampling<br />
within a session, then the resampling of highly identifiable individuals would<br />
alert the observer to this problem. Third, an observer might use it to assess<br />
nearest neighbors for each animal. Fourth, one could record both activity and<br />
nearest neighbor for each group member. If group scans are conducted during<br />
individual-follows, then it is important that the scans can be completed very<br />
quickly to prevent the observer from being distracted from the focal individ-<br />
ual. With scan sampling, cetacean researchers can broaden their dataset to look<br />
at coordination of group activities and refined measures of association for a<br />
number of animals of different age and sex classes.<br />
Predominant activity sampling-Predominant Activity Sampling (PAS), de-<br />
veloped by Hutt and Hutt (1970), refers to scoring individual behavior as the<br />
predominant activity over some interval (e.g., 30 sec), only if that behavior<br />
occupied 50% or more of that interval (>15 sec). This method is only useful<br />
for measuring states. No studies in the survey used PAS. It is an empirically<br />
valid technique for estimating the proportion of time during which behavioral<br />
states occur (Tyler 1979). Very brief behaviors or displays (events) will not be<br />
represented in PAS data unless the observer uses very short intervals. It is<br />
important that the interval length is brief enough to capture the briefest states<br />
of interest.<br />
PAS is useful for sampling animals that go in and out of view for brief<br />
periods. In a sample of Ti~rsiops calves, some types of data have been collected<br />
using both focal individual point samples and focal PAS simultaneously (J.<br />
Mann, unpublished data). Thus calf activity budgets could be compared based<br />
on sampling type. For example, “calf position swimming” (when the calf is<br />
in contact under the mother) was measured using both point sampling (in-<br />
stantaneous measures of swim position at first surface after each 2.5-min in-<br />
terval) and PAS (calculated from the continuous data on observed onsets and<br />
terminations of calf position swimming). I compared the percent time swim-<br />
ming in calf position for 19 calves in my longitudinal sample. Each calf was<br />
observed during focal-individual follows for 10-15 h per yr for one or more<br />
years. The sum of observation hours for all calves across all years was 750.<br />
Each infant had a PAS and point sampling percentage for each year, for a total<br />
of 40 calf-years (an average of two years or samples per calf). The PAS per-<br />
centages and point sampling percentages were correlated by treating each in-<br />
fant observation year as independent. This yielded a significant value (Spear-<br />
man’s rho = 0.97, n = 40, P < 0.0001). The mean percent error rate for PAS
MA”: BEHAVIORAL SAMPLING METHODS 115<br />
was 3.2%. Several methodologists (e,g., Tyler 1979) have demonstrated the<br />
validity of PAS and point sampling using statistical models, but no one has<br />
contrasted these methods using actual observational data.<br />
Incident sampling-Alternate names: all-event sampling, all-occurence sam-<br />
pling, all-animals sampling. Incident sampling entails scoring all behavioral<br />
events of a specific type in a group. This method is not applicable for most<br />
behavioral states. The observability of the events is key to the success of this<br />
method. The behaviors themselves must be obvious enough to alert the ob-<br />
server (e.g., breaching). In addition, the observer must be able to record all<br />
the events regardless of how many animals are present. Thus, for incident<br />
sampling to be successful, the behavior must be sufficiently infrequent to allow<br />
complete recording for the group. As Altmann points out, this is a form of<br />
continuous sampling (referred to as focal animal sampling in her paper) of a<br />
group for a restricted set of behaviors. However, with continuous sampling,<br />
the individual animals are identified. During incident sampling, distinctions<br />
between subjects may or may not be made. Observers should still make every<br />
effort to determine whether the same or different animals are exhibiting the<br />
behavior(s). Two areas can be problematic. First, animals may tend to repeat<br />
displays, thus becoming overrepresented in the dataset. Second, if group com-<br />
position changes often (Table 1) it may be difficult to calculate event rate as<br />
a function of group size or structure.<br />
In the survey, 16% of the studies used incident sampling. Studies that<br />
identified their method as “focal-group-sampling” were sometimes coded as<br />
using incident sampling if they scored a very restricted set of obvious behav-<br />
iors. For example, breaches and lobtails are so visible that observers are likely<br />
to detect each occurrence within a group. Similarly, if observers could deter-<br />
mine how many animals were within acoustic range, then incident sampling<br />
could apply to studies of vocalizations.<br />
Incident or all-event sampling is valuable for cetacean researchers who wish<br />
to focus on specific dramatic or easily recognized events that involve few an-<br />
imals. For example, observers could use this method to compare successful and<br />
unsuccessful killer whale hunting attempts via beaching. Dramatic surface<br />
percussive displays, such as breaching or lobtailing, can also be observed re-<br />
liably in groups (Waters and Whitehead 1990). Incident sampling requires<br />
that the observer systematically record every event. Incident sampling may be<br />
adopted with a group protocol, but the observer should still be able to dis-<br />
tinguish which animal (or agehex class member) is engaging in the behavior<br />
to avoid misattributing the behavior pattern equally to all individuals.<br />
Sequence sampling-In sequence sampling, the observer focuses on sequences<br />
of behavior or on particular interactions, rather than individuals, and system-<br />
atically records all relevant behaviors that occur during the event(s), main-<br />
taining the sequences of behaviors in the record (Altmann 1974). Parameters<br />
defining how an interaction begins and terminates must be specified in se-<br />
quence sampling. For example, observers may score sequences of behaviors<br />
among surface-active groups of humpback whales by defining the beginning<br />
of the sequence as “male humpback whale challenges principal escort by ap-
116 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
proaching within 50 m” and terminating the sequence when either the chal-<br />
lenger or the principal escort separates more than 50 m from the female for<br />
30 min. It is distinct from incident sampling because during sequence sam-<br />
pling multiple events may occur in a group, but the observer focuses on the<br />
start of the first event seen and records that particular event or interaction to<br />
completion (even if other group members engage in similar behaviors at the<br />
same time). Only one of the studies surveyed used sequence sampling.<br />
Sequence sampling is recommended for very observable behaviors but can<br />
be applied under a broad range of field conditions (see Table 1). The researcher<br />
must be able to determine when the sequence begins and ends and be able to<br />
discriminate between, although not necessarily identify, the interactants. The<br />
individual animals may be hard to identify if they communicate over long<br />
ranges and move in and out of the observer’s view. For example, if the re-<br />
searcher wants to know whether breaching animals attract or repel others, s/<br />
he can use sequence sampling to test whether animals are likely to approach<br />
or leave the animal who breached (e.g., within 10 min of the first breach).<br />
Sequence sampling is excellent for determining the conditional probabilities<br />
of behavior sequences.<br />
RECOMMENDATIONS<br />
Although tightly focused and informative studies peppered the literature,<br />
the survey of CJZ and MMS revealed two shortcomings in studies of cetacean<br />
behavior. First, a large proportion of studies used ad libitum sampling, which<br />
is replete with bias. Second, there was little consistency in reporting behavioral<br />
data, sampling methods, hours of observation, behavioral and group defini-<br />
tions, and number of subjects. This information is critical to evaluation of the<br />
claims of any study. Furthermore, the comparability of results and hopes of<br />
replication are lost if this information is not provided. Information about<br />
equipment (cameras, film, boat, motor, tape recorders, etc. ) was frequently<br />
reported, but basic methodological information regarding subjects and pro-<br />
tocol was sometimes lacking.<br />
To allow evaluation of statistical power, it is important to indicate the<br />
number of animals observed and how long each animal was observed. Re-<br />
searchers usually reported the number of animals identified in the population<br />
but not the number of animals observed. Sample sizes can be determined only<br />
when animals are observed individually. Similarly, only a few studies treated<br />
each subject, rather than each observation, as independent. As Milinski (1997)<br />
points out, this is a form of pseudoreplication, one of the “deadly sins” in the<br />
study of behavior.<br />
Group size and definition of group are needed, as well as a statement of<br />
protocol used if group composition changed. I recommend proximity-based<br />
measures, because this method is quantifiable and does not rely on behavioral<br />
sampling to determine group membership. “Coordinated-behavior” definitions<br />
make implicit assumptions about proximity, because observers cannot assess<br />
the activities of animals who are kilometers away. Furthermore, individuals at
MANN: BEHAVIORAL SAMPLING METHODS 1<strong>17</strong><br />
close range may be engaged in more than one activity that is shared with<br />
those farther away. An explicit definition of group, and a rationale for the<br />
definition in terms of the study goals, is essential.<br />
Specific details of how behaviors were defined and how behaviors were sam-<br />
pled should be included, using either standardized terms from the ethological<br />
literature or descriptions. Only a few studies of those reviewed used standard<br />
sampling terms described by Altmann (1974) and others. All studies of ce-<br />
tacean behavior should provide the basic methodological information described<br />
in this review. Following are more specific recommendations regarding the<br />
uses of different sampling protocols and techniques, with special consideration<br />
of how to reconcile unbiased observations under ideal conditions with the<br />
imperfect conditions that cetacean biologists face. Additional detailed discus-<br />
sion of methods and statistical comparisons of different observational methods<br />
are available elsewhere (see Altmann 1974, Dunbar 1976, Tyler 1979, Martin<br />
and Bateson 1986, Mann et al. 1991, Rogosa and Ghandour 1991).<br />
Follow Protocol<br />
Sampling biases in group-follows and surveys can be minimized by using<br />
scan sampling, incident sampling, or sequence sampling methods. Monitoring<br />
groups (surveys and group follows) is valuable when the goal is to gain a<br />
“snapshot” of group life through surveys (e.g., habitat use by foraging com-<br />
pared to resting groups) when observation conditions do not permit tracking<br />
of individuals (but tracking of groups is possible), when research questions<br />
focus on the synchrony or coordination of group members, or when the study<br />
focuses on sequences of very observable behaviors. To identify the predominant<br />
group activity during a survey, observers can scan-sample >50% of the group<br />
members. This may be possible for large groups of dolphins or whales that<br />
are visible at the surface for long enough to be scanned. If group sizes are in<br />
the hundreds or thousands, then a protocol for random scan sampling of a<br />
subset of the group may be developed. Sampling biases, such as number of<br />
scans per individual as a function of group size, must be taken into account<br />
during data reduction. Researchers may track a group and record all occur-<br />
rences of certain very obvious behaviors, thus using incident sampling. Se-<br />
quence sampling may also be applied if observers are interested in sequences<br />
of behavior during specific types of interactions, such as cooperative hunting<br />
in killer whales. Focal-group sampling is not recommended, because this<br />
method does not explicitly or systematically sample individuals or behaviors<br />
in groups.<br />
Focal-individual sampling methods with the application of continuous,<br />
point, and/or predominant-activity sampling, are critical tools for insuring<br />
reliable estimates of behavior. The individual is the natural unit of analysis<br />
for behavior. Data can be pooled by individual, and data from different indi-<br />
viduals can often be treated as independent (Machlis et ul. 1985). When ob-<br />
servers cannot identify or discriminate which individual produces a behavior<br />
(e.g,, during scan- or all-event sampling), this makes it difficult to identify
118 MARINE MAMMAL SCIENCE. VOL. 15. NO. 1. 1999<br />
independent units for statistical analysis. This can present problems with the<br />
group-follow protocol.<br />
General Sampling Problems: Deep Divers, Large Groups, and Cowespondence<br />
Between Surface and Subsurface Behavior<br />
The correspondence between surface (observable) and subsurface (often<br />
unobservable) behavior is unknown for most studies of cetaceans, but such<br />
information could help in assessing the biases in relying on surface behavior<br />
alone. For example, some animals forage at depth and not during surfacing<br />
bouts. One option is to define the surfacing breaks between foraging dives as<br />
part of the continuous state of “foraging.” Other behaviors may not change at<br />
the surface (e.g., socializing, resting, traveling). Although it is still valuable<br />
to record surface behavior even though no correspondence to subsurface be-<br />
havior is implied, such sampling decisions should be indicated clearly in the<br />
methods section of the article.<br />
All ethologists face difficult decisions when their subjects disappear from<br />
view. These “out-of-sight’’ periods may be treated differently, depending on<br />
the animal under study. Nesting or burrowing animals may be most likely to<br />
disappear from view when feeding their young or engaging in sexual behavior.<br />
Some cetaceans may disappear for long periods when foraging. Deleting the<br />
time that animals are “out of sight” is desirable when animal visibility is not<br />
determined by or biased by animal activity. Other strategies for using “out-<br />
of-sight’’ periods are needed for cetaceans, because some of their activities co-<br />
vary with diving periods (when animals are most likely to be out of view).<br />
Cetaceans may rest, travel, socialize, or engage in other diverse behaviors at<br />
or near the surface. Certain types of social behavior may occur at depth, but<br />
certainly not all of them do. If the out-of-sight periods tend to be short relative<br />
to the bout lengths of the behavior, then ethologists might designate the<br />
activity for the out-of-sight period as either the last activity seen before the<br />
subject disappeared or the first activity seen when the animal reappears. Al-<br />
ternatively, the observer may consider a state continuous only if the last ac-<br />
tivity seen and the first activity seen after a brief out-of-sight period are the<br />
same. This strategy is not effective if some behaviors, such as foraging or<br />
hunting, consistently occur out of an observer’s view. In such circumstances,<br />
cetacean observers may detect foraging using other cues, including bubble<br />
patterns at the surface, echolocation clicks, or intermittent sightings of fish<br />
catches or chases. Sometimes there are signs at surfacing of what the animal<br />
was doing on the preceding dive. For example, bottom-feeding whales may<br />
surface with mud streaming from the mouth (Wiirsig et al. 1985). Thus,<br />
observers may use fleeting observations of events to define states or indirect<br />
cues (rapid surfacing, changing direction) to define “unobservable” states. One<br />
may also tag an individual to track behaviors at depth for comparison with<br />
cues visible at the surface. By combining these techniques, observers can de-<br />
velop behavioral definitions and systematic protocols to capture the range of<br />
cetacean behaviors.
MANN: BEHAVIORAL SAMPLING METHODS 119<br />
Sampling from large groups of hundreds or thousands (e.g., Scott and Per-<br />
ryman 1991) can prove to be difficult. However, Ostman (1994) has success-<br />
fully completed short focal follows on Stenella longirostris in Hawaii, suggesting<br />
that it is possible to stay with individuals long enough to get a 5-min sample<br />
using an underwater observation booth built into a vessel. Scan sampling of<br />
large groups of unidentified animals is another possible approach. In such<br />
circumstances, researchers would need to scan a subset, possibly by randomly<br />
selecting subgroups or particular agehex classes. Videotaping for later scoring<br />
or photogrammetric analysis (.g., Scott and Perryman 1991) can provide data<br />
on association and behavior.<br />
Although observation conditions for studying cetaceans are rarely ideal,<br />
proximity between animals, synchrony in surfacing, and basic activities can<br />
be recorded systematically under most sampling conditions. Cetacean behav-<br />
ioral research could benefit greatly from a wider use of quantitative sampling<br />
techniques. If applied carefully, such partial but systematic observations can<br />
help elucidate cetacean social and behavioral ecology.<br />
ACKNOWLEDGMENTS<br />
This paper was prepared while I was a fellow at the Center for Advanced Study in<br />
the Behavioral Sciences in Stanford, California, with support from the National Science<br />
Foundation (#SBR-9022 192). Helpful comments on the manuscript were provided by<br />
Robin Baird, Vincent Janik, Edward Miller, Amy Samuels, three anonymous reviewers,<br />
William F. Perrin, Douglas Wartzok, and especially Peter L. Tyack. I was also sup-<br />
ported by the Helen V. Brach Foundation, The Eppley Foundation for Research, and<br />
Georgetown University. I thank Jeanne Altmann for igniting my interest in observa-<br />
tional methods in 1980.<br />
LITERATURE CITED<br />
ALTMANN, J. 1974. Observational study of behavior: Sampling methods. Behaviour 49:<br />
227-267.<br />
CLARK, C. W. 1980. A real-time direction finding device for determining the bearing<br />
to the underwater sounds of southern right whales, (Eubalaena australis). Journal<br />
of the Acoustical Society of America 68:508-5 11.<br />
DUNBAR, R. I. M. 1976. Some aspects of research design and their implications in the<br />
observational study of behaviour. Behaviour 58:79-98.<br />
FRAGASZY, D. M., S. BOINSKI AND J. WHIPPLE. 1992. Behavioral sampling in the field:<br />
Comparison of individual and group sampling methods. American Journal of<br />
Primatology 26:259-275.<br />
FRANKEL, A. S., C.W. CLARK, L.M. HERMAN AND C.M. GABRIELE. 1995. Spatial distribution,<br />
habitat utilization, and social interactions of humpback whales, Megaptera<br />
novaeangliae, off Hawaii determined using acoustic and visual techniques. Canadian<br />
Journal of Zoology 73:1134-1146.<br />
FREITAG, L., AND P. L. TYACK. 1993. Passive acoustic localization of Atlantic bottlenose<br />
dolphin whistles and clicks. Journal of the Acoustical Society of America 93:<br />
2 197-2205.<br />
HAMMOND, P. S. , S. A. MIZROCH AND G.P. DONOVAN. EDS. 1990. Individual recognition<br />
of cetaceans: Use of photo-identification and other techniques to estimate popu-
120 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
lation parameters. Report of the <strong>International</strong> Whaling Commission (Special Issue<br />
12).<br />
Hum, S. J., AND C. HUTT. 1970. Direct observation and measurement of behavior.<br />
Charles C. Thomas, Springfield, IL.<br />
MACHLIS, L., P. W. D. DODD AND J. C. FENTRESS. 1985. The pooling fallacy: Problems<br />
arising when individuals contribute more than one observation to the data set.<br />
Zeitschrift fur Tierpsychologie 68:201-2 14.<br />
MANN, J. , AND B. SMUTS. In press. Behavioural development of wild bottlenose dolphin<br />
newborns. Behaviour.<br />
MANN, J., T. TEN HAVE, J. W. PLUNKETT AND S. J. MEISELS. 1991. Time sampling: A<br />
methodological critique. Child Development 62:227-241.<br />
MARTIN, P., AND P. BATESON. 1986. Measuring behaviour: An introductory guide. Cambridge<br />
University Press, Cambridge, UK.<br />
MILINSKI, M. 1997. How to avoid seven deadly sins in the study of behavior. Advances<br />
in the Study of Behavior 26:160-180.<br />
OSTMAN, J. S. 0. 1994. Social organization and social behavior of Hawai’ian spinner<br />
dolphins (Stenella longiristrzs). Ph.D. dissertation, The University of California,<br />
Santa Cruz, CA. 114 pp.<br />
ROGOSA, D., AND G. GHANDOUR. 1991. Statistical models for behavioral observations.<br />
Journal of Educational Statistics 16: 157-252.<br />
SAMUELS, A., AND T. GIFFORD. 1997. A quantitative assessment of dominance relations<br />
among bottlenose dolphins. Marine Mammal Science 13:70-99.<br />
SCOTT, M. D., AND W. L. PERRYMAN.<br />
1991. Using aerial photogrammetry to study<br />
dolphin school structure. Pages 227-241 in K. Pryor and K. S. Norris, eds.<br />
Dolphin societies: Discoveries and puzzles. University of California Press, Berkeley,<br />
CA.<br />
SHANE, S. H. 1990. Behavior and ecology of the bottlenose dolphin at Sanibel Island,<br />
Florida. Pages 245-265 in S. Leatherwood and R. R. Reeves, eds. The bottlenose<br />
dolphin. Academic Press, Inc., New York, NY.<br />
TYACK, P. 1985. An optical telemetry device to identify which dolphin produces a<br />
sound. Journal of the Acoustical Society of America 78:1892-1895.<br />
TYACK, P. L., AND C. A. RECCHIA. 1991. A datalogger to identify vocalizing dolphins.<br />
Journal of the Acoustical Society of America 90: 1668-167 1.<br />
TYLER, S. 1979. Time-sampling: A matter of convention. Animal Behaviour 272301-<br />
810.<br />
WATERS, S., AND H. WHITEHEAD. 1990. Aerial behaviour in sperm whales. Canadian<br />
Journal of Zoology 68:2076-2082.<br />
WHITEHEAD, H. 1995. Investigating structure and temporal scale in social organizations<br />
using identified individuals. Behavioral Ecology 6: 199-208.<br />
WHITEHEAD, H. 1997. Analyzing animal social structure. Animal Behaviour 53: 1053-<br />
1067.<br />
WHITEHEAD, H., AND L. WEILGART. 1991. Patterns of visually observable behaviour<br />
and vocalizations in groups of female sperm whales. Behaviour 118:275-296.<br />
WILLIAMS, G. C. 1966. Adaptation and natural selection. Princeton University Press,<br />
Princeton, NJ.<br />
WURSIG, B., E. M. DORSEY, M. A. FRAKER, R. S. PAYNE AND W. J. RICHARDSON. 1985.<br />
Behavior of bowhead whales, Balaena mysticetlrs, summering in the Beaufort Sea:<br />
A description. Fishery Bulletin, U.S. 83:357-377.<br />
Received: 10 July 1995<br />
Accepted: 14 January 1998
MA”: BEHAVIORAL SAMPLING METHODS 121<br />
APPENDIX 1<br />
I conducted a search of the Science Citation Index (Institute for Scientific Infor-<br />
mation, Inc., 1996) using all 39 cetacean genera for keywords. The search yielded 846<br />
studies. To determine whether the study focused on cetacean behavior in the wild, each<br />
title was examined for keywords: behavior, acoustics, observations, field, captive, wild,<br />
and a description of the geographic location. The 846 studies were classified according<br />
to 14 subjects (e.g., molecular, neurophysiological, conservation, behavior); many stud-<br />
ies received more than one classification. Of the total, 125 studies (14.8%) in 28<br />
journals focused on cetacean behavior (n = 106) or vocalizations (n = 19) in the wild.<br />
Studies that appeared to focus strictly on population assessment (e.g., aerial counts)<br />
were not classified as behavioral studies. CJZ and MMS published 60% of the studies.<br />
The next most popular journal for cetacean behavior studies was the Journal ofMam-<br />
malogy, which published only 5% (n = 6). Thus, CJZ and MMS clearly represent the<br />
majority of relevant peer-reviewed publications in the Science Citation Index (SCI) and<br />
were chosen for in-depth analysis for this reason. Based on a comparison of Impact<br />
Factors (Impact Factor = the average number of times a paper in a journal is cited in<br />
the Science Citation Index within two years of publication), MMS and CJZ represent<br />
mid-level quality in research. Approximately an equal number of cetacean studies<br />
ranked above (n = 21) and below (n = 22) MMS and CJZ, with six studies (published<br />
in the Journal of Mammalogy) ranking the same as CJZ and one study ranking below<br />
CJZ but above MMS.<br />
Because the SCI search may have missed studies that did not use genera for key-<br />
words, I surveyed all issues of CJZ and MMS for all studies of free-ranging cetacean<br />
behavior and vocalizations published during 1989-1995. The SCI search by genera<br />
did not miss any studies, but one CJZ study I classified as a “behavior study” from<br />
the title was excluded from the review because the article’s content did not include<br />
behavior or vocalizations.<br />
The basic unit of analysis was one published article. Of 144 authors, 21 (14.6%)<br />
published more than one study. Of these, 14 published two studies, six published<br />
between three and five studies and one researcher authored or co-authored nine studies<br />
(using five different sampling methods).<br />
Species studied-Twenty-four cetacean species were studied in the 74 papers I re-<br />
viewed. Studies of humpback whales (Megaptera novaeangliae) were the most common<br />
(24%), followed by those of sperm whales (Physeter macrocephalus, 16%), and killer<br />
whales (Orcinus orca, 16%). Bottlenose dolphins (TurJiops truncatus) were represented in<br />
9% of the studies. Bowhead (Balaena mysticetus) and gray whales (Eschrictius robustus)<br />
were each represented in 5% of the studies. Narwhals (Monodon monoceros), Bryde’s<br />
whales (Balaenoptera edeni) and belugas (Delphinapterus leucas) were each represented in<br />
4% of the studies. Seven studies included more than one species. Of the 15 remaining<br />
species, only one or two studies were published on each.<br />
Age and sex classes sampled-Twenty studies focused on animals of a particular age<br />
or sex class. Otherwise, observers either did not know the ages or sexes of animals, or<br />
they observed animals of all ages and both sexes. No studies made quantitative com-<br />
parisons between age or sex classes.<br />
Number of animals sampled-The range of animals sampled was one to >1,000.<br />
Typically, researchers reported the number of animals that they surveyed or identified<br />
but not how many animals were observed or sampled. In 45% of the studies, the<br />
number of animals identified and/or sampled could not be determined. When groups<br />
of animals were sampled (e.g., in surveys or group-follows), 59% of the time (n=44<br />
studies) I could not distinguish between how many animals were individually identified<br />
and how many were actually observed.<br />
Individual identifiation-In 66% (n = 49) of the studies, researchers could identify<br />
individual animals. In 34% (n = 25), individual animals were not identified, although<br />
in about half of those (48%, n = 12), physical features of the species and reference to
122 MARINE MAMMAL SCIENCE, VOL. 15, NO. 1, 1999<br />
other studies indicated that individual identification was possible. In all studies that<br />
could identify individuals, rates of individual behavior or vocalizations were not re-<br />
ported unless the study involved only one or two animals (n = 3 studies).<br />
Observationirecording hours-Twenty-eight studies (38%) did not report how much<br />
time was spent observing or recording behavior. Some of these studies reported the<br />
hours spent at sea, but this information is only useful (as an indicator of sighting<br />
effort) when observation hours are also presented. Sixty-two percent of the studies<br />
reported the number of surveys conducted or the number of hours spent observing<br />
animals. Only four studies treated each subject, rather than each observation, as in-<br />
dependent for behavioral analysis.<br />
Group definitions and group size-Definition of “group” is essential to assess the va-<br />
lidity of behavioral sampling from groups. I indicated whether or not “group” was<br />
defined, and if it was defined, I noted the definition. Group-size definition was not<br />
relevant in nineteen studies (e.g., only one animal was studied). For 55 studies (surveys,<br />
group-follows, anecdotes), groups of animals were observed. In 42 of these (76%),<br />
researchers did not define “group.” If group was defined (n = 13), proximity-based<br />
measures (e.g., within 10 m) or coordinated behavior (all visible animals engaged in<br />
the same activity) were the two most common definitions of group.<br />
Group-size means or ranges (minimum and maximum group size) were reported in<br />
38 studies. Seventeen studies did not report group size information.<br />
Types of behavior and vocalizations recordd-Behaviors recorded in each study were<br />
classified as breathing (14%), diving (30%), resting (24%), surface displays (8%), so-<br />
cializing (30%), feeding or foraging (43%), traveling (27%), or “other.” Association<br />
(proportion of time animals together) was recorded in 22% of the studies and vocali-<br />
zations were recorded in 38%. “Other” included births (1%) or other unusual events<br />
(16%). Over half of the studies (57%) provided definitions for one or more of the<br />
behaviors recorded. Studies were classified according to the behaviors that researchers<br />
recorded, not which data were analyzed. If an ethogram was given, this was noted. No<br />
studies reported measures of inter-observer or intra-observer reliability.<br />
Sound recording was used either to find or track whales or to describe the vocal<br />
repertoire of a species, group, or individual. Of the 29 papers that recorded dolphin<br />
or whale vocalizations, 28% identified species-specific repertoires or group-specific rep-<br />
ertoires. In one study, species-typical vocalizations were monitored to test for the pres-<br />
ence of the species. Identification of individual repertoires within groups was limited<br />
to anecdotes, but one study used passive acoustic localization to identify which animal<br />
was likely to be vocalizing. Fourteen percent of the studies that recorded cetacean<br />
vocalizations calculated vocalization rates for groups of animals but not for individuals.<br />
Solitary individuals were acoustically recorded for portions of one study.<br />
All 74 studies were classified according to sampling method: ad libitum, continuous,<br />
focal-group, one-zero, point, scan, predominant activity, sequence, and incident sampling. If a<br />
study used more than one sampling method (six studies), both were scored. The results<br />
of these classifications are reported in the main text.
Aquatic Mammals 2008, 34(1), 35-43, DOI 10.1578/AM.34.1.2008.35<br />
Distribution and Habitat Use of Antillean Manatees<br />
(Trichechus manatus manatus) in the Drowned Cayes Area of<br />
Belize, Central America<br />
Katherine S. LaCommare, 1 Caryn Self-Sullivan, 2, 3 and Solange Brault 1<br />
1 University of Massachusetts, Boston, Department of Biology, 100 Morrissey Boulevard, Boston, MA 02125, USA;<br />
E-mail: kslacommare1@hotmail.com<br />
2 Texas A&M University, Department of Wildlife and Fisheries, College Station, TX 77845, USA<br />
3 <strong>Sirenian</strong> <strong>International</strong>, Inc., 200 Stonewall Drive, Fredericksburg, VA 22401, USA<br />
Abstract<br />
Belize, Central America, has long been recognized<br />
as a stronghold for Antillean manatees (Trichechus<br />
manatus manatus) in the Caribbean (O’Shea &<br />
Salisbury, 1991). The Drowned Cayes area, in<br />
particular, has been noted as an important habitat<br />
(Bengston & Magor, 1979; O’Shea & Salisbury,<br />
1991; Auil, 1998, 2004; Morales-Vela et al., 2000).<br />
It is critical to evaluate habitat use and the relative<br />
importance of different habitat types within these<br />
cayes because this area is increasingly impacted<br />
by human activities (Auil, 1998). The two research<br />
objectives for this paper are (1) to document<br />
manatee distribution within the Drowned Cayes,<br />
Swallow Caye, and Gallows Reef, and (2) to examine<br />
habitat use patterns in order to identify habitat<br />
characteristics influencing the probability of sighting<br />
a manatee. Binary logistic regression was used<br />
to examine whether the probability of sighting a<br />
manatee varied in relation to several habitat variables.<br />
The probability of sighting a manatee across<br />
all points was 0.31 per scan (n = 795). Habitat<br />
category, seagrass category, and habitat category<br />
interaction with resting hole were the most important<br />
variables explaining the probability of sighting<br />
a manatee. The Drowned Cayes area clearly constitutes<br />
a manatee habitat area. Seagrass flats and<br />
cove habitats with resting holes were especially<br />
important habitat characteristics.<br />
Key Words: distribution, habitat use, Antillean<br />
manatee, Trichechus manatus manatus, Belize<br />
Introduction<br />
The Antillean subspecies of the West Indian manatee<br />
(Trichechus manatus manatus) is found in<br />
19 countries throughout the Caribbean, Central<br />
America, and South America (Lefebvre et al.,<br />
2001). It is listed as vulnerable to extinction by<br />
the <strong>International</strong> Union for the Conservation of<br />
Nature and Natural Resources (IUCN) (2004)<br />
because of its greatly reduced numbers, continued<br />
exploitation, and population fragmentation.<br />
However, Belize, Central America, has long been<br />
recognized as a stronghold for Antillean manatees<br />
in the Caribbean (O’Shea & Salisbury, 1991). In<br />
the 1960s, Charnock-Wilson (1968, 1970) conducted<br />
interviews and personal observations and<br />
reported that manatees were abundant in the country.<br />
From the late 1970s to the early 2000s, aerial<br />
surveys continued to document that Belize has a<br />
relatively large number of manatees in comparison<br />
to neighboring countries (Bengston & Magor,<br />
1979; O’Shea & Salisbury, 1991; Auil, 1998,<br />
2004; Morales-Vela et al., 2000).<br />
The Drowned Cayes, Swallow Caye, and<br />
Gallows Reef, in particular, are recognized as<br />
important manatee areas within Belize. Charnock-<br />
Wilson (1968) described manatees using drowned<br />
cayes or mangrove islands, in a general sense,<br />
stating that it was common to spot manatees<br />
around drowned cayes, their channels, lagoons,<br />
and surrounding seagrass beds. Subsequent aerial<br />
surveys have documented high concentrations of<br />
manatees in the cayes near Belize City, of which<br />
the Drowned Cayes, Swallow Caye, and Gallows<br />
Reef are a part (Bengston & Magor, 1979; O’Shea<br />
& Salisbury, 1991; Auil, 1998, 2004; Morales-<br />
Vela et al., 2000). Boat surveys that have been<br />
conducted since 1999 further corroborated<br />
that this area is consistently used by manatees<br />
(LaCommare et al., 2003).<br />
This important manatee area may become<br />
increasingly threatened by tourism and human<br />
population growth (Auil, 1998); growth in tourism<br />
by cruise ships has been particularly dramatic.<br />
Between 1998 and 2006, the number of tourists<br />
entering Belize via cruise ships increased from<br />
14,183 visitors per year to 851,436 visitors per<br />
year (Belize Tourism Board, 2007). During that<br />
same period, there was an 18% increase in the<br />
population of Belize City (Brinkhoff, 2005). Due
36 LaCommare et al.<br />
to the Drowned Cayes’ proximity to Belize City,<br />
and because these islands lie between the mainland,<br />
the reef, and developed cayes, manatees could be<br />
particularly prone to threats from these sources. As<br />
tourism increases, a corresponding increase in boat<br />
traffic as well as land development is expected.<br />
Boats pass through the Drowned Cayes when traveling<br />
from Belize City to the northern cayes, outer<br />
cayes, and the reef, causing a risk of watercraft collisions<br />
with manatees. Watercraft collisions are the<br />
leading source of human-caused manatee mortality<br />
in Belize (Auil & Valentine, 2004). The proximity<br />
of these islands to the barrier reef may result in<br />
increased land development that could result in the<br />
loss of mangrove habitat. In the last five years, new<br />
resort development and expansion have occurred.<br />
And finally, development in Belize City and along<br />
the Belize River is likely to cause increases in runoff,<br />
which may decrease the productivity, biomass,<br />
and percent bottom cover of seagrass (Hemminga<br />
& Duarte, 2000; Duarte, 2002)—a principal food<br />
source for manatees (Ledder, 1986; Provancha &<br />
Hall, 1991; Mignucci-Giannoni, 1998; Lefebvre<br />
et al., 2000; U.S. Fish and Wildlife Service, 2001).<br />
Because the Drowned Cayes area is an important<br />
manatee area within the country and because the<br />
islands are increasingly affected by human activities,<br />
it is critical to improve our understanding of<br />
how manatees utilize this area. Evaluating animal<br />
habitat use and the relative importance of habitat<br />
types informs habitat management and conservation<br />
(Garshelis, 2000). The two research objectives<br />
for this paper are (1) to document manatee distribution<br />
within the Drowned Cayes, Swallow Caye,<br />
and Gallows Reef, and (2) to examine habitat use<br />
patterns in order to identify habitat characteristics<br />
influencing the probability of sighting a manatee.<br />
Materials and Methods<br />
Study Area<br />
The Drowned Cayes and Swallow Caye are mangrove<br />
islands along the central coast of Belize,<br />
10 to 15 km east of Belize City and 5 km west of<br />
the Belize Barrier Reef (Figure 1). The Drowned<br />
Cayes are a string of mostly uninhabited islands<br />
that are 14 km long by 4 km at their widest<br />
point and are almost entirely comprised of red<br />
(Rhizophora mangle) ) and black (Avicennia ( ger-<br />
minans) mangrove stands. There is very little dry<br />
land, and the islands are interspersed with broad<br />
channels, narrow inlets, shallow lagoons, and protected<br />
coves. The entire complex is surrounded by<br />
seagrass beds; seagrass also grows in varying densities<br />
on the bottoms of the channels, lagoons, and<br />
coves. Turtle grass (Thalassia testudinum) is the<br />
predominant species with shoal (Halodule wrightii)<br />
and manatee grass (Syringodium filiforme) also<br />
very common. Water depth is less than 1 m in<br />
some places and is never greater than 6 m within<br />
the study area; the tidal range is less than 0.3 m. In<br />
2002, the northern portion of the Drowned Cayes<br />
and Swallow Caye were designated as the Swallow<br />
Caye Wildlife Sanctuary.<br />
The islands are in a marine environment. In<br />
open water areas, salinities range from 35 to 40<br />
ppt (mean salinity is 37 ppt), making the environment<br />
slightly hypersaline. Within the mangrove<br />
island complex, salinities have greater fluctuations<br />
as a result of larger run-off, evaporation, and tidal<br />
influences. Therefore, within the channels, coves,<br />
and lagoons of the mangrove islands, salinities<br />
can range from 30 to 42 ppt.<br />
Survey Design<br />
A point sampling survey design was devised for a<br />
small boat platform to quantify manatee distribution<br />
and habitat use. Fifty-four permanent points<br />
were established in all habitat types throughout<br />
the study area (Figure 2). Habitat types are<br />
defined in Table 1. These points were randomly<br />
sampled during January, February, March, June,<br />
July, and August from 2001 to 2004. Surveys were<br />
also conducted in June, July, and August 2005.<br />
Using a mix of experienced observers and volunteers,<br />
three to 13 observers searched for manatees<br />
at each point for 30 min. These searches are<br />
referred to as point scans. As the boat came within<br />
100 m of each point, observers started scanning<br />
for manatees in a 360º circle around the boat. The<br />
boat was then anchored in position using a pole.<br />
For each scan, the number of manatees and habitat<br />
characteristics were recorded. The latter consisted<br />
of habitat category, presence of a resting<br />
hole, presence and type of seagrass, temperature,<br />
salinity, and sea state (Table 2).<br />
Data Analysis<br />
Manatee Distribution—Manatee distribution was<br />
mapped within the Drowned Cayes area by calculating<br />
the probability of sighting a manatee for each<br />
point. Since most of the scans had no manatees and<br />
because location sample sizes were unequal, presence<br />
or absence of manatees was used to calculate<br />
the probability of sighting a manatee. Points that<br />
were visited less than five times were excluded<br />
from this analysis. This map is a snapshot of the<br />
overall variation in distribution throughout the<br />
study area. To provide the most comprehensive<br />
map of distribution, the largest number of points<br />
possible was included in the analysis. To do this,<br />
20-min scans were included in the analysis. To<br />
create parity between the 20- and 30-min scans,<br />
presence or absence of the 30-min scans was<br />
determined based on only the first 20 min of the<br />
sampling duration. This increased the sample size
Distribution and Habitat Use of Antillean Manatees in Belize 37<br />
Figure 1. Map of Drowned Cayes and surrounding area (Source: British Admiralty Navigation Chart No. 522, Belize City<br />
and Approaches, 1989)<br />
from 613 to 795 scans and increased the number of<br />
points used in the analysis from 39 to 47.<br />
Habitat Use—Binary logistic regression was used<br />
to examine whether the probability of sighting a manatee<br />
varied in relation to several habitat variables—<br />
habitat category, presence of a resting hole, presence<br />
and type of seagrass, temperature, salinity, and sea<br />
state. Due to the large number of zeros (268 zero<br />
counts of 430 samples) and unequal sample sizes<br />
across habitat categories, manatee presence/absence<br />
was used rather than the number of individual manatees<br />
sighted during each scan to calculate the probability<br />
of sighting a manatee. Habitat category was<br />
determined by placing each point into one of six<br />
categories based on mangrove shoreline features and<br />
the depth profile of the particular point. These habitat<br />
categories were qualitatively assigned and are mutually<br />
exclusive. It is recognized that not all points fit<br />
neatly into these designations. A complete description<br />
of each category is given in Table 1. A resting<br />
hole is a bottom feature that is a distinct, shallow<br />
depression in the seafloor. Although they are at least<br />
3 to 4 m wide by 3 to 4 m long, they can be larger and<br />
are not necessarily regularly shaped. In some cases,<br />
they appear to be natural features that are maintained<br />
by manatee use; in other cases, they may have been<br />
created by manatee use. Seagrass category connotes<br />
whether seagrass was present at a particular point<br />
and, if so, what species of seagrass were found there.<br />
There were six categories: (1) none, (2) Thalassia
38 LaCommare et al.<br />
Figure 2. Map of manatee sighting probability throughout the Drowned Cayes study area (n = 795) (Source: Defense<br />
Mapping Agency Chart, Belize City Harbor, 1996)
Distribution and Habitat Use of Antillean Manatees in Belize 39<br />
Table 1. List and description of habitat categories used in the logistic regression analysis; number of points equals the number<br />
of points in each habitat category type. Description is the definition of the habitat category type.<br />
Habitat category<br />
Table 2. Description and assessment of the model relating the presence/absence of manatees to habitat variables for the<br />
Drowned Cayes study area (n = 491, Nagelkerke R-Square = 0.22); significant variables were habitat category, seagrass<br />
category, and habitat category*resting hole interaction term. Backward stepwise procedure and log-likelihood function were<br />
used to determine the most parsimonious model. Hosmer and Lemeshow Goodness of Fit χ 2 = 3.757, p = .878.<br />
Variables in the model<br />
testudinum only, (3) Halodule wrightii only,<br />
(4) T. testudinum/H. /H. / wrightii mix, (5) Syringodium<br />
filiforme/T. testudinum mix, and (6) a mix of all<br />
three. Sea state was based on the Beaufort scale.<br />
Logistic regression accommodates both continuous<br />
and categorical predictor variables (Trexler<br />
& Travis, 1993; Floyd, 2001). Maximum likelihood<br />
method was used to fit the model to the data,<br />
and a backward stepwise procedure was used to<br />
determine the most parsimonious model. The likelihood<br />
ratio test and the Wald statistic were used<br />
to determine the significance of the parameters in<br />
explaining the variation in the dependent variable.<br />
To determine if the model was an adequate fit to the<br />
Change in -2 log<br />
likelihood df<br />
data and to rule out the undue influence of outliers,<br />
the Hosmer-Lemeshow test was used and observed<br />
and expected sighting probabilities were compared<br />
(Trexler & Travis, 1993). Plots of Cook’s distances<br />
and Studentized residuals vs predicted probabilities<br />
were used to examine the influence of outliers and<br />
bias in the data. All statistical analyses were performed<br />
using SPSS, Version 13.0 (SPSS, 2004).<br />
Results<br />
Significance<br />
of change<br />
Dependent variable<br />
Manatee presence/absence<br />
Independent variables<br />
Sea state (Beaufort scale) 0.55 3 NS<br />
Surface salinity (ppt) 1.76 1 NS<br />
Water temperature (ºC) 0.11 1 NS<br />
Habitat category 14.09 5 0.015*<br />
Resting hole 1.27 1 NS<br />
Seagrass category 23.21 5 0.0001*<br />
Habitat category*resting hole 10.46 3 0.015*<br />
Habitat category*seagrass category 10.84 8 NS<br />
NS = Not significant, * = significant<br />
Number<br />
of points Description<br />
Lagoon 5 A large open water area totally encompassed within the mangrove islands; the area<br />
has a uniform and shallow depth (< 3 m).<br />
Channel 12 An area of deeper water (3 to 6 m) that cuts between the mangrove islands;<br />
generally, there is mangrove shoreline on two sides.<br />
Channel edge 8 An area of deeper water that cuts through areas of shallower water; the point<br />
encompasses two habitat types: (1) channel and (2) seagrass bed. Depth ranges from<br />
1 to 5 m. The point may be adjacent to mangrove shoreline on one side or may be in<br />
open water.<br />
Seagrass bed 11 An area of shallow water, < 3 m, outside of the mangrove islands with a seagrass<br />
bottom.<br />
Cove 16 An area of very protected water that is at the end of a channel or off to the side of a<br />
channel; it is nearly enclosed by mangroves and has shoreline on at least three sides.<br />
Depth ranges from 0.6 to 4 m.<br />
Reef 2 This is an open water area on the back reef portion of the Belize Barrier Reef; there<br />
are no mangrove islands in the vicinity, and depth ranges from 1.5 to 4 m.<br />
Distribution Within the Drowned Cayes<br />
Manatees were sighted throughout the study area.<br />
The probability of sighting a manatee across all
40 LaCommare et al.<br />
points was 0.31/20-min scan (n = 795). Sighting<br />
probabilities for each location ranged from 0 to<br />
0.86/scan. Swallow Caye had the highest probability<br />
of sightings among its points, and the<br />
Gallows Reef points had the lowest probability of<br />
sightings. There does not appear to be a distinct<br />
spatial pattern of variability in sighting probability<br />
among the Drowned Cayes points (Figure 2).<br />
Habitat Use<br />
The probability of sighting a manatee was lowest<br />
at the reef (0.<strong>17</strong>/scan) and highest on seagrass<br />
flats (0.58/scan) (Figure 3). Habitat category,<br />
seagrass category, and habitat category interaction<br />
with resting hole were the most important<br />
variables explaining the probability of sighting<br />
a manatee (Table 2). The change in log likelihood<br />
for each of these three terms was 14.09 with<br />
p = 0.015, 23.21 with p = 0.0001, and 10.46 with<br />
p = 0.015, respectively. Within habitat categories,<br />
seagrass flats and coves with resting holes both<br />
explained a significant portion of the variation in<br />
manatee presence (Wald = 4.14 with p = 0.042<br />
and 4.79 with p = 0.029, respectively) (Figures<br />
3A & B). There was only one scan point categorized<br />
as a channel with a resting hole. The low<br />
sample size of this category type may explain<br />
why channels with resting holes had an opposite<br />
trend to other habitat types. Scan points with<br />
just T. testudinum and S. filiforme present or with<br />
just H. wrightii present had a significantly lower<br />
probability of sighting a manatee than other seagrass<br />
categories (Wald = 6.50 with p = 0.011 and<br />
6.4 with p = 0.011, respectively) (Figure 4). The<br />
Hosmer-Lemeshow chi-square goodness of fit<br />
was 1.486 with a p-value of 0.983, indicating that<br />
the overall model was a good fit to the data, and<br />
the Nagelkerke R-Square was 0.221 (Table 2).<br />
The expected sighting probability for each habitat<br />
and seagrass category was very similar to the<br />
observed sighting probability further indicating<br />
that the model was a good fit to the data (Figures<br />
3A & 4). Residual plots of change in deviance vs<br />
predicted probabilities indicated that samples with<br />
a low predicted probability of a presence were<br />
poorly fit by the model. In other words, manatees<br />
tended to be observed more often than predicted<br />
by the model at places where the probability of<br />
sighting a manatee was low.<br />
Discussion<br />
The Drowned Cayes area clearly constitutes a<br />
manatee habitat area. Manatees were sighted at<br />
least once at nearly all of the points that were<br />
surveyed (47 out of 54 points). There were<br />
some points that had particularly high sighting<br />
probabilities (≥ 0.5/20-min scan), and these points<br />
A.<br />
B.<br />
Figure 3. (A) Observed and predicted probability of sighting<br />
a manatee per 20-min scan by habitat; (B) Observed<br />
probability of sighting a manatee per 20-min scan by habitat<br />
with and without the presence of a resting hole; No =<br />
no resting hole, Yes = resting hole present (2001 to 2005,<br />
n = 491, ± 1 SE).<br />
Figure 4. Observed and predicted probability of sighting<br />
a manatee per 20-min scan by seagrass category (2001 to<br />
2005, n = 491)<br />
might be considered “hot spots” within the overall<br />
mangrove island complex. These points should<br />
be given particular consideration in management<br />
plans in terms of tourist activities, development,<br />
and watercraft traffic.
Distribution and Habitat Use of Antillean Manatees in Belize 41<br />
Based on the logistic regression procedure,<br />
habitat category was one of the three most important<br />
variables explaining variation in sighting<br />
probability across the Drowned Cayes. Variability<br />
in sighting conditions between habitat types did<br />
not result in differences in detection probability.<br />
It is unlikely that there were detection biases as a<br />
result of habitat types (LaCommare et al., unpub.<br />
data). Seagrass beds had the highest probability<br />
of sighting a manatee, and reefs had the lowest.<br />
Seagrass flats may be a particularly important<br />
habitat category, indicating the importance of the<br />
Drowned Cayes as a feeding area.<br />
However, manatees did utilize all habitat categories.<br />
Manatees use the entire area and many of<br />
its components to meet a variety, but probably not<br />
all, of their physiological and behavioral requirements.<br />
During the course of this study, manatees<br />
were observed feeding, socializing, resting,<br />
nursing calves, and moving from place to place<br />
(Self-Sullivan & LaCommare, unpub. data). Even<br />
resources that are used at a low frequency may be<br />
important components of the manatees’ overall<br />
habitat. For instance, the reef may be a seasonally<br />
important travel corridor and stopover site for male<br />
manatees during the mating season (Self-Sullivan<br />
et al., 2004). Behavioral and physiological studies<br />
indicate that manatees probably need regular<br />
access to freshwater sources to survive (Reynolds<br />
& Odell, 1991; Ortiz et al., 1998, 1999; Reynolds,<br />
1999; Deutsch et al., 2000, 2003). Only one source<br />
of hyposaline water (refractometer measurements<br />
ranged from 10 to <strong>17</strong> ppt) has been conclusively<br />
identified (Self-Sullivan & LaCommare, unpub.<br />
data) within the Drowned Cayes, which may be<br />
used for osmoregulation. Manatees may also be<br />
traveling from the Drowned Cayes to freshwater<br />
sources several kilometers away as they do in<br />
Florida (Deutsch et al., 2003). Management plans<br />
should recognize the importance of protecting a<br />
variety of habitat types.<br />
Seagrass is clearly an important component of<br />
the Drowned Cayes habitat. Seagrass category<br />
was an important variable explaining the probability<br />
of sighting a manatee. It contributed the<br />
largest improvement in model fit, and the parameter<br />
was highly significant. Specifically, sites that<br />
had a mix of T. testudinum and S. filiforme or sites<br />
with just H. wrightii present had a significantly<br />
lower probability of sighting a manatee than the<br />
other seagrass categories—no seagrass, a mix of<br />
all three species, just T. testudinum, , or a mix of T.<br />
testudinum and H. wrightii.<br />
How this relates to manatee foraging and feeding<br />
is not clear. The three most common species<br />
in the study area—T. testudinum, H. wrightii,<br />
and S. filiforme—are present in the manatee<br />
diet in both Florida and Puerto Rico (Packard,<br />
1984; Ledder, 1986; Provancha & Hall, 1991;<br />
Mignucci-Giannoni, 1998; Lefebvre et al., 2000;<br />
USFWS, 2001). The relative importance of these<br />
species in their diet is not fully understood in<br />
these places. In Indian River Lagoon, Florida, and<br />
in Puerto Rico, “manatees fed most often on the<br />
most frequently encountered seagrass” (Lefebvre<br />
et al., 2000, p. 295). In Indian River Lagoon, this<br />
was H. wrightii, and in Puerto Rico, this was T.<br />
testudinum (Lefebvre et al., 2000). However, in<br />
both locations, it is possible that manatees return<br />
to specific H. wrightii beds to feed on them<br />
(Lefebvre et al., 2000). Both T. testudinum and<br />
H. wrightii appear to be important food resources<br />
in the Drowned Cayes based on observations of<br />
feeding manatees, and certain areas appear to be<br />
more heavily foraged than others (LaCommare<br />
& Self-Sullivan, unpub. data), perhaps indicating<br />
foraging preferences.<br />
While certain types of seagrass sites, as well as<br />
species, may be particularly important to manatees<br />
in the Drowned Cayes, fully understanding<br />
manatee foraging will involve examining the<br />
extent of feeding behavior in particular places,<br />
specific physical and biological characteristics of<br />
important seagrass areas, and the juxtaposition of<br />
these areas to other important resources.<br />
Habitat category interacting with resting hole<br />
was also an important variable explaining variation<br />
in sighting probability. Cove habitats with<br />
resting holes were the most significant subcategory.<br />
Cove habitats are extremely sheltered places<br />
found at the end of narrow channels or off to the<br />
side of larger channels. In the Drowned Cayes,<br />
resting holes are a feature of the bottom topography<br />
and are associated with resting manatees<br />
(Self-Sullivan & LaCommare, unpub. data). In<br />
Florida, manatees often use secluded places for<br />
feeding, resting, mating, and calving (USFWS,<br />
2001). This appears to be true in the Drowned<br />
Cayes as well. The presence of a resting hole was<br />
a key feature distinguishing which particular cove<br />
area was utilized (0.25 to 0.49/scan increase in<br />
sighting probability). It is unclear at this point if<br />
the resting hole was the reason manatees used the<br />
site or if the resting hole depression is a result of<br />
repeated manatee use.<br />
While the Nagelkerke R-Square for the regression<br />
model was 0.22, indicating that only 22% of<br />
the variability in sighting probability was explained<br />
by the logistic regression model, the model was a<br />
good fit to the data, and the variability explained<br />
was the correct variability. The ability to explain<br />
variation in sighting probability will provide a<br />
better understanding of habitat use. Building a<br />
habitat model that includes variables that describe<br />
the spatial configuration and characteristics of habitat<br />
is an important follow-up to this analysis and
42 LaCommare et al.<br />
may even provide insight into habitat selection and<br />
preferences. Habitat use is dependent on behavior,<br />
and suitable habitat must include a variety of patch<br />
types or components so that animals can meet their<br />
essential behavioral and physiological requirements<br />
(Mysterud & Ims, 1998). In addition, habitat<br />
use, selection, and preferences are affected by habitat<br />
heterogeneity (Johnson, 1980; Li & Reynolds,<br />
1994; Kie et al., 2002). Heterogeneity can be<br />
defined in terms of number, proportion, spatial<br />
arrangement, shape, and contrast between patches<br />
and patch types (Kie et al., 2002). Analyses examining<br />
manatee habitat components in these terms<br />
and with respect to behavior could be important in<br />
understanding manatee habitat use patterns.<br />
Acknowledgments<br />
We thank Earthwatch Institute for substantial<br />
financial support and Earthwatch Volunteers for<br />
logistical support and data collection. We thank<br />
Conservation Action Fund, Henry Greenwalt,<br />
Project Aware, Virtual Explorers, and Jack Bert<br />
for additional financial support. Spanish Bay<br />
Resort provided logistical and in kind support<br />
from 2001 to 2004. Hugh Parkey Foundation for<br />
Marine Awareness and Education provided logistical<br />
and in kind support in 2005. We are grateful<br />
to the numerous interns and field assistants who<br />
were vital to data collection and field logistics. We<br />
are especially thankful for the help and guidance<br />
of our colleague, friend, and field assistant Gilroy<br />
Robinson. In addition, we thank the two anonymous<br />
reviewers for their insightful comments that<br />
led to the improvement of this manuscript.<br />
Literature Cited<br />
Auil, N. E. (1998). Belize manatee recovery plan (United<br />
Nations Development Program/Global Environment<br />
Facility, Coastal Zone Management Project No. BZE/92/<br />
G31). Kingston, Jamaica: United Nations Environment<br />
Program, Caribbean Environment Program.<br />
Auil, N. E. (2004). Abundance and distribution trends of<br />
the West Indian manatee in the coastal zone of Belize:<br />
Implications for conservation. Unpublished Master’s<br />
thesis, Texas A&M University, College Station. 83 pp.<br />
Auil, N. E., & Valentine, A. (2004). Manatee strandings<br />
along the coastal zone of Belize 1996-2003. Unpublished<br />
report, Coastal Zone Management Authority and<br />
Institute.<br />
Belize Tourism Board. (2007). Belize tourism statistics.<br />
Retrieved 11 March 2008 from www.belizetourism.org/<br />
belize-tourism/tourist-arrivals-by-mode-of-arrival.html.<br />
Bengston, J. L., & Magor, D. (1979). A survey of manatees<br />
in Belize. Journal of Mammalogy, 60(1), 230-232.<br />
Brinkhoff, T. (2005). City population. Retrieved 29<br />
February 2008 from www.citypopulation.de.<br />
Charnock-Wilson, J. (1968). The manatee in British<br />
Honduras. Oryx, 8(4), 293-294.<br />
Charnock-Wilson, J. (1970). Manatees and crocodiles.<br />
Oryx, 10(4), 236-238.<br />
Deutsch, C. J., Reid, J. P., Bonde, R. K., Easton, D. E.,<br />
Kochman, H. I., & O’Shea, T. J. (2000). Seasonal<br />
movements, migratory behavior, and site fidelity of<br />
West Indian manatees along the Atlantic coast of<br />
the United States as determined by radio-telemetry.<br />
Unpublished report, Florida Cooperative Fish and<br />
Wildlife Research Unit, USGS and University of<br />
Florida.<br />
Deutsch, C. J., Reid, J. P., Bonde, R. K., Easton, D. E.,<br />
Kochman, H. I., & O’Shea, T. J. (2003). Seasonal<br />
movements, migratory behavior, and site fidelity of<br />
West Indian manatees along the Atlantic coast of the<br />
United States. Wildlife Monographs, 67, 1-77.<br />
Duarte, C. M. (2002). The future of seagrass meadows.<br />
Environmental Conservation, 29(2), 192-206.<br />
Floyd, T. (2001). Logit modeling and logistic regression. In<br />
S. M. Scheiner & J. Gurevitch (Eds.), Design and analysis<br />
of ecological experiments (pp. 197-216). New York:<br />
Oxford University Press. 415 pp.<br />
Garshelis, D. L. (2000). Delusions in habitat evaluation:<br />
Measuring use, selection and importance. In L. Boitani<br />
& T. K. Fuller (Eds.), Research techniques in animal<br />
ecology (pp. 111-164). New York: Columbia University<br />
Press. 442 pp.<br />
Hemminga, M. A., & Duarte, C. M. (2000). Seagrass ecology.<br />
Cambridge, UK: Cambridge University Press.<br />
<strong>International</strong> Union for the Conservation of Nature and<br />
Natural Resources (IUCN). (2004). 2004 IUCN red list<br />
of threatened species. Retrieved 16 November 2005<br />
from www.redlist.org.<br />
Johnson, D. (1980). The comparison of usage and availability<br />
measurements for evaluating resource preference.<br />
Ecology, 61(1), 65-71.<br />
Kie, J. G., Bowyer, R. T., Nicholson, M. C., Boroski, B. B.,<br />
& Loft, E. R. (2002). Landscape heterogeneity at differing<br />
scales: Effects on spatial distribution of mule deer.<br />
Ecology, 83(2), 530-544.<br />
LaCommare, K. S., Self-Sullivan, C., & Brault, S. (2003).<br />
Habitat use of Antillean manatees (Trichechus manatus<br />
manatus) in the Drowned Cayes, Belize, Central<br />
America. Paper presented at the Fifteenth Biennial<br />
Conference on the Biology of Marine Mammals.<br />
Vancouver, Washington.<br />
Ledder, D. A. (1986). Food habits of the West Indian manatee,<br />
Trichechus manatus latirostris, in south Florida.<br />
Coral Gables, FL: University of Miami.<br />
Lefebvre, L. W., Reid, J. P., Kenworthy, W. J., & Powell,<br />
J. A. (2000). Characterizing manatee habitat use and seagrass<br />
grazing in Florida and Puerto Rico: Implications<br />
for conservation and management. Pacific Conservation<br />
Biology, 5, 289-298.<br />
Lefebvre, L. W., Marmontel, M., Reid, J. P., Rathbun,<br />
G. B., & Domning, D. P. (2001). Status and biogeography<br />
of the West Indian manatee. In C. A. Woods &
Distribution and Habitat Use of Antillean Manatees in Belize 43<br />
F. E. Sergile (Eds.), Biogeography of the West Indies:<br />
Patterns and perspectives (pp. 425-474). Boca Raton,<br />
FL: CRC Press.<br />
Li, H., & Reynolds, J. F. (1994). A simulation experiment<br />
to quantify spatial heterogeneity in categorical maps.<br />
Ecology, 75(8), 2446-2455.<br />
Mignucci-Giannoni, A. A. (1998). The diet of the manatee<br />
(Trichechus manatus) in Puerto Rico. Marine Mammal<br />
Science, 14(2), 394-397.<br />
Morales-Vela, B., Olivera-Gomez, D., Reynolds, J. E., III,<br />
& Rathbun, G. B. (2000). Distribution and habitat use<br />
by manatees (Trichechus manatus manatus) in Belize<br />
and Chetumal Bay, Mexico. Biological Conservation,<br />
95, 67-75.<br />
Mysterud, A., & Ims, R. A. (1998). Functional responses in<br />
habitat use: Availability influences relative use in tradeoff<br />
situations. Ecology, 79(4), 1435-1441.<br />
Ortiz, R. M., Worthy, G. A. J., & Byers, F. M. (1999).<br />
Estimation of water turnover rates of captive West<br />
Indian manatees (Trichechus manatus) held in fresh and<br />
salt water. The Journal of Experimental Biology, 202,<br />
33-38.<br />
Ortiz, R. M., Worthy, G. A. J., & MacKenzie, D. S. (1998).<br />
Osmoregulation in wild and captive West Indian manatees<br />
(Trichechus manatus). Physiological Zoology,<br />
71(4), 449-457.<br />
O’Shea, T. J., & Salisbury, C. A. L. (1991). Belize: A last<br />
stronghold for manatees in the Caribbean. Oryx, 25(3),<br />
156-164.<br />
Packard, J. M. (1984). Impact of manatees Trichechus<br />
manatus on seagrass communities in eastern Florida.<br />
Acta Zoologica Fennica, <strong>17</strong>2, 21-22.<br />
Provancha, J. A., & Hall, C. R. (1991). Observations of<br />
associations between seagrass beds and manatees in<br />
East Central Florida. Florida Scientist, 54(2), 86-98.<br />
Reynolds, J. E., III. (1999). Efforts to conserve manatees.<br />
In J. R. Twiss, Jr. & R. R. Reeves (Eds.), Conservation<br />
and management of marine mammals (pp. 267-293).<br />
Washington, DC: Smithsonian Institution Press.<br />
Reynolds, J. E., III, & Odell, D. K. (1991). Manatees and<br />
dugongs. New York: Facts on File, Inc.<br />
Self-Sullivan, C., Smith, G. W., Packard, J. M., &<br />
LaCommare, K. S. (2004). Seasonal occurrence of<br />
male Antillean manatees (Trichechus manatus manatus)<br />
on the Belize Barrier Reef. Aquatic Mammals, 29(3),<br />
342-354.<br />
SPSS. (2004). SPSS release 13.0 [Computer software].<br />
Apache Software Foundation.<br />
Trexler, J. C., & Travis, J. (1993). Nontraditional regression<br />
analyses. Ecology, 74(6), 1629-1637.<br />
U.S. Fish and Wildlife Service (USFWS). (2001).<br />
Technical/Agency draft: Florida manatee recovery plan<br />
(Trichechus manatus latirostris). Atlanta, GA: Author.
<strong>17</strong>2<br />
NOTES<br />
Caribbean Journal of Science, Vol. 41, No. 1, <strong>17</strong>2-<strong>17</strong>7, 2005<br />
Copyright 2005 College of Arts and Sciences<br />
University of Puerto Rico, Mayagüez<br />
Bottlenose Dolphins (Tursiops<br />
truncatus) in the Drowned Cayes,<br />
Belize: Group Size, Site Fidelity<br />
and Abundance<br />
KECIA A. KERR 1 ,R.H.DEFRAN 2 , AND GRE-<br />
GORY S. CAMPBELL Cetacean Behavior Laboratory<br />
- Department of Psychology, San Diego<br />
State University, San Diego, CA 92182-4611,<br />
USA<br />
ABSTRACT.—Group size, site fidelity and abundance<br />
of bottlenose dolphins, Tursiops truncatus,<br />
ms. submitted June 3, 2004; accepted December 28,<br />
2004<br />
1<br />
Current address: Whale Research Laboratory, Department<br />
of Geography, University of Victoria, PO<br />
Box 3050 STN CSC, Victoria, BC, V8W 3P5 Canada<br />
2<br />
Corresponding author: rdefran@sunstroke.<br />
sdsu.edu
were assessed during 392 photo-identification surveys<br />
conducted during 1997-1999 in the Drowned<br />
Cayes region, near Belize City, Belize, Central<br />
America. During this study 2155 dolphins were<br />
sighted across 736 groups. Mean group size was 2.9<br />
(SD = 2.32) which is one of the smallest reported for<br />
bottlenose dolphins. One hundred and fifteen individual<br />
dolphin were photographically identified,<br />
with sighting frequencies ranging from one to fifty<br />
(X ¯ = 8.1, SD = 9.05). Thirty percent of identified dolphins<br />
were judged to be residents, while 23% were<br />
photographed only once. Chao’s M th model for<br />
closed populations was used to derive an abundance<br />
estimate of 122 dolphins (95% CI = 114 -140). This<br />
low abundance estimate and a leveling trend in the<br />
rate of newly identified individuals, indicates that<br />
the Drowned Cayes dolphin population is both<br />
small and finite. Group size, abundance, and site<br />
fidelity comparisons were made with a 4-yr photoidentification<br />
study conducted at nearby Turneffe<br />
Atoll. Both the Drowned Cayes and Turneffe Atoll<br />
studies had similarly small group sizes, low and<br />
variable levels of site fidelity and low abundance<br />
estimates, but there was no overlap between individual<br />
sightings in the two areas. The observed behavioral<br />
patterns and similarities between the two<br />
studies raise concerns that increasing pressures on<br />
Belize’s marine resources may pose a threat to its<br />
bottlenose dolphins.<br />
KEYWORDS.—Residency patterns, population dynamics,<br />
cetaceans, Belize, Caribbean<br />
A recent 4-yr study at Turneffe Atoll in<br />
Belize, Central America (Campbell et al.<br />
2002), was one of the first to document the<br />
population dynamics of bottlenose dolphins<br />
(Tursiops truncatus) in a tropical offshore<br />
atoll containing coral reefs, sea grass<br />
beds and mangroves, and is one of the few<br />
published studies on this cetacean species<br />
in the Caribbean Sea (see also Grigg and<br />
Markowitz 1997; Rossbach and Herzing<br />
1999; Rogers et al. 2004). Turneffe bottlenose<br />
dolphins occurred in small groups<br />
that were larger when they contained<br />
calves, and this population had a high proportion<br />
of individuals sighted only once.<br />
Further, abundance estimates were low,<br />
and only a small proportion of these dolphins<br />
showed evidence of site fidelity<br />
(Campbell et al. 2002). The habitat characteristics<br />
of the Drowned Cayes, where the<br />
current study was conducted, are similar to<br />
those of Turneffe. Thus, the objectives of<br />
NOTES <strong>17</strong>3<br />
the current study were to examine the generality<br />
of the Turneffe findings and interpretations,<br />
identify possible overlap between<br />
these two dolphin populations, and<br />
extend the management and conservation<br />
implications of both studies.<br />
The Drowned Cayes are located in the<br />
western Caribbean, south of the Yucatan<br />
Peninsula, 6 km east of Belize City and 16<br />
km west of Turneffe Atoll (Fig. 1). The<br />
study area consisted of a 150 km 2 chain of<br />
mangrove islands extending from <strong>17</strong>°20’N-<br />
<strong>17</strong>°34’N, and 88°03’W-88°07’W. Like<br />
Turneffe Atoll, the Drowned Cayes area is<br />
characterized by a predominating substrate<br />
of sandy sea grass beds, but also includes a<br />
portion of the Belize Barrier Reef. Surface<br />
water temperature ranged from 25°-33°C<br />
(X ¯ = 29.2, SD = 1.62), and water depth, measured<br />
during 216 sightings, ranged from 0.9<br />
mto11.8m(X ¯ = 4.3 m). All aspects of the<br />
photo-identification methodology used in<br />
this study, including: survey methodology,<br />
criteria for groups and calves, image processing<br />
and analysis of photographic data<br />
(i.e., rate of discovery, sighting frequency,<br />
abundance), were the same as those described<br />
in Campbell et al. (2002). Briefly<br />
summarized, boat based surveys lasting<br />
approximately 4 h were carried out once or<br />
twice daily, and all portions of the study<br />
area were covered at least once a month.<br />
When dolphins were sighted, the survey<br />
vessel stopped while environmental data<br />
were recorded and group size was determined.<br />
Next, an attempt was made to obtain<br />
high quality dorsal fin photographs for<br />
each group member. Once photographic<br />
data were collected, the survey continued<br />
along a predetermined route to search for<br />
and photograph additional dolphins.<br />
During 1246 h expended over a 3-yr period<br />
(February-December in 1997 and 1998,<br />
and April-Dec in 1999), 392 surveys were<br />
completed, and 277 h were spent in direct<br />
observation of 2155 dolphins sighted across<br />
736 groups. Group size ranged from one to<br />
20 dolphins (X ¯ = 2.9, SD = 2.32), but was<br />
higher for groups with calves (n = 160, X ¯ =<br />
4.6 [includes calves], SD = 2.68) than without<br />
calves (n = 576, X ¯ = 2.5, SD =<br />
1.99)(Mann-Whitney U = 19540.0, Z =<br />
-11.433, p = < 0.001). Almost a third (31%)
<strong>17</strong>4<br />
FIG. 1. Drowned Cayes study area (within dashed outline). Insets show the location of the study area from a<br />
broad regional (top) and western Caribbean (bottom) perspective.<br />
of all sightings were solitary dolphins and<br />
99% of all groups contained 10 or fewer<br />
dolphins. Group sizes for Drowned Cayes<br />
(X ¯ = 2.9) and Turneffe dolphins (X ¯ = 3.8, SD<br />
= 3.55, Campbell et al. 2002) are among the<br />
smallest reported for any population of<br />
bottlenose dolphins (Connor et al. 2000).<br />
Group size in bottlenose dolphins, as in<br />
many other species, is a tradeoff between<br />
optimizing foraging efficiency and minimizing<br />
predation risk (Wells and Scott<br />
1999; Campbell et al. 2002). While there<br />
have been no studies of prey characteristics<br />
for Drowned Cayes or Turneffe dolphins to<br />
date, no observations or images indicated<br />
evidence of shark bites, suggesting that<br />
predatory threats from sharks are minimal.<br />
Low apparent predation at the Drowned<br />
Cayes suggests, as it did at Turneffe, that<br />
energy intake is a primary selective pressure<br />
on group size; and, the small group<br />
sizes in these areas are optimized for food<br />
resources that are likely low in density.<br />
NOTES<br />
Similarly, in both the Drowned Cayes and<br />
Turneffe, larger calf-groups may be an adaptation<br />
that facilitates allomaternal behavior,<br />
thereby increasing the foraging efficiency<br />
of nursing mothers who are<br />
constrained by maternal responsibilities<br />
(see review in Campbell et al. 2002).<br />
Across the study period, 115 dolphins<br />
were photographically identified. The<br />
slope of the Drowned Cayes discovery<br />
curve for newly photographed dolphins<br />
showed a steep rise until the 90 th survey<br />
when 76 dolphins (66%) had been identified<br />
(compare with Fig. 2, Campbell et al.<br />
2002). Few new dolphins were photographed<br />
until surveys 190-230 (Aug 1999)<br />
when an additional 23 dolphins (18%) were<br />
identified. No new dolphins were photographed<br />
during the remaining 162 surveys.<br />
Sighting frequencies for identified dolphins<br />
ranged from 1-50 (X ¯ = 8.1, SD = 9.02), with<br />
23% (n = 27) of these dolphins identified<br />
one time, 50% (n = 57) sighted between two
and nine times and 27% (n = 31) sighted 10<br />
or more times. Evidence for site fidelity<br />
was evaluated by examining sighting frequencies<br />
within and between the three<br />
study years. Dolphins sighted two or more<br />
times in each of the three study years, or<br />
four or more times in two successive years,<br />
were labeled as residents, and comprised<br />
30% of the identified population. The abundance<br />
estimate derived by Chao’s closed<br />
model M th was 122 (95% CI = 114-140)<br />
(Chao et al. 1992) and was quite similar to<br />
the number of identified dolphins (n = 115).<br />
Taken together, the discovery rate for<br />
new individuals, the high study effort<br />
(number of surveys) extending over a 3-yr<br />
period, the high proportion of individuals<br />
showing low sighting frequencies (vis-à-vis<br />
the high number of surveys), the small<br />
number of individuals showing evidence<br />
for site fidelity (“residents”), and the relatively<br />
low abundance estimate, suggests<br />
that, as at Turneffe, a small and finite population<br />
of dolphins uses this study area.<br />
These same data indicate that the Drowned<br />
Cayes area can only support a small number<br />
of dolphins, probably due to low prey<br />
item densities (group size interpretation<br />
suggested above). Some dolphins were<br />
seen throughout and across years, however,<br />
indicating that a degree of fidelity to<br />
the area does exist.<br />
Photographs of the 115 Drowned Cayes<br />
dolphins were compared to those of 81 individuals<br />
photographed during 549 surveys<br />
at Turneffe Atoll from 1992-1996<br />
(Campbell et al. 2002). Despite the close<br />
geographic proximity of the Drowned<br />
Cayes and Turneffe, the high survey effort<br />
in both studies, and the presence of a large<br />
number of individuals in each population<br />
not considered “residents,” there was no<br />
overlap documented between these populations.<br />
While the depth of the channel<br />
separating Turneffe Atoll from the Belize<br />
Barrier Reef and the Drowned Cayes (range<br />
= 274-305 m) may act as a physical barrier<br />
to regulate movements of dolphins between<br />
the study areas, the distance between<br />
Turneffe and the Drowned Cayes<br />
could be easily traveled by this species<br />
(Stoddart 1962; Tanaka 1987; Wood 1998;<br />
Defran et al. 1999). A tentative hypothesis,<br />
NOTES <strong>17</strong>5<br />
consistent with the existing photo-identification<br />
data, is that bottlenose dolphins in<br />
Belize may have overlapping ranges along<br />
the mainland coastline; however, exploitation<br />
of the offshore islands, such as the<br />
Drowned Cayes and Turneffe Atoll, is selective<br />
among subsets of this population.<br />
Similar selective partitioning among areas<br />
of their range were shown for bottlenose<br />
dolphins (Maze and Würsig 1999) as well<br />
as humpback (Megaptera novaeangilae) and<br />
sperm (Physeter macrocephalus) whales<br />
(Clapham 2000; Whitehead and Weilgart<br />
2000). To date, no mainland coastal photoidentification<br />
surveys have been carried out<br />
in Belize. Such surveys are needed, however,<br />
in order to clarify the home ranges of<br />
Drowned Cayes and Turneffee Atoll dolphins,<br />
as well as to evaluate the selective partitioning<br />
hypothesis we have proposed.<br />
Belize is rich in natural resources that<br />
could be vulnerable to ecologically unsustainable<br />
growth in tourism, fishing, and development.<br />
The accessibility of the<br />
Drowned Cayes to the coastal mainland of<br />
Belize, including Belize City and the Belize<br />
River, places an even higher pressure on<br />
levels of resource extraction, pollution, and<br />
boat traffic in this region than in other offshore<br />
locations such as Turneffe Atoll. Although<br />
fishing pressure in the Drowned<br />
Cayes, and the rest of Belize, may be considered<br />
low compared to highly populated<br />
areas in the Caribbean, the demands on<br />
fisheries resources are increasing and the<br />
effects of overfishing are already evident as<br />
changes in fish community structure (Sedberry<br />
et al. 1999). Some suggest that reefs in<br />
the Caribbean have been overfished for<br />
centuries, and the resulting changes in<br />
community structure have placed these<br />
ecosystems in a precarious state that is now<br />
collapsing (Jackson 1997; Jackson et al.<br />
2001). While fishing in the study area is<br />
mainly artisinal, even this type of fishing<br />
may be unsustainable (Coblentz 1997; Sedberry<br />
et al. 1999). Shifts in the ecological<br />
dominance from coral to fleshy algae have<br />
been found at some reef locations in Belize<br />
(Aronson and Precht 1997; McClanahan<br />
and Muthiga 1998). Since the first major<br />
bleaching event in the history of Belize’s<br />
coral reefs occurred in 1995, an even more
<strong>17</strong>6<br />
severe event occurred in 1998 (Aronson et<br />
al. 2000). Threats, such as overfishing, pollution,<br />
and the results of climatic change,<br />
have been shown to detrimentally affect<br />
marine mammals in a variety of ways (Harvell<br />
et al. 1999; Allen and Read 2000; Fair<br />
and Becker 2000; Wilson et al. 2000; Bossart<br />
et al. 2003). In the aggregate, these factors<br />
indicate that bottlenose dolphins in Belize<br />
could play a valuable role as a sentry species<br />
in this area. For example, shifts in<br />
bottlenose dolphin occurrence and abundance<br />
might indicate a further decline in<br />
prey (and other fish species) abundance.<br />
Similarly, increases in skin lesions (observed<br />
on at least two dolphins in the<br />
Drowned Cayes population, personal observations),<br />
may provide an early signal of<br />
compromised immunity to disease (Wilson<br />
et al. 2000; Bossart et al. 2003).<br />
Acknowledgments.—We thank the following<br />
individuals and institutions for their<br />
contributions to this research: B. Winning,<br />
Oceanic Society; Oceanic Society volunteers,<br />
the Belize Fisheries Department, the<br />
Belize Forestry Department, Coastal Zone<br />
Management Authority Institute and S.<br />
Turton in Belize; E. Requeña and staff at<br />
Spanish Bay Resort, K. Schafer, K. LaCommare,<br />
A. Sanders, B. Bilgre, S. Barclay, H.<br />
Petersen, and C. Sullivan for field support;<br />
B. Bilgre, L. Hinderstein and A. Sanders for<br />
contributions to the photographic dataset;<br />
K. Dudzik, B. Hancock, J. Schivone, L.<br />
Sousa, and A. Toperoff, as well as numerous<br />
interns, at the Cetacean Behavior Laboratory,<br />
San Diego State University, for assistance<br />
with photographic analysis. We<br />
appreciate the contributions of E. Bardin,<br />
A. Bradford, B. Hancock, D. Herzing, A.<br />
Mignucci-Giannoni and one anonymous<br />
reviewer who graciously offered helpful<br />
editorial remarks on earlier drafts of this<br />
manuscript. This research was conducted<br />
under a marine scientific research permit<br />
issued to Oceanic Society from the Ministry<br />
of Natural Resources, the Environment and<br />
Industry, Government of Belize.<br />
LITERATURE CITED<br />
Allen, M. C., and A. J. Read. 2000. Habitat selection of<br />
foraging bottlenose dolphins in relation to boat<br />
NOTES<br />
density near Clearwater, Florida. Mar. Mamm. Sci.<br />
16:815-824.<br />
Aronson, R. B., and W. F. Precht. 1997. Stasis, biological<br />
disturbance, and community structure of a Holocene<br />
coral reef. Paleobiology 23:336-346.<br />
Aronson, R. B., W. F. Precht, I. G. Macintyre, and T. J.<br />
Murdoch. 2000. Coral bleach-out in Belize. Nature<br />
405:36.<br />
Bossart, G. D., et al. 2003. Pathologic findings in<br />
stranded Atlantic bottlenose dolphins (Tursiops<br />
Truncatus) from the Indian River Lagoon, Florida.<br />
Florida Scient. 66:226-238.<br />
Campbell, G. S., B. A. Bilgre, and R. H. Defran. 2002.<br />
Bottlenose dolphins (Tursiops truncatus) in<br />
Turneffe Atoll, Belize: occurrence, site fidelity,<br />
group size, and abundance. Aquat. Mamm. 28:<strong>17</strong>0-<br />
180.<br />
Clapham, P. J. 2000. The humpback whale: seasonal<br />
feeding and breeding in a baleen whale. In Cetacean<br />
societies: field studies of dolphins and whales, ed. J.<br />
Mann, R. C. Conner, P. L. Tyack, and H. Whitehead,<br />
<strong>17</strong>3-198. Chicago and London: The University<br />
of Chicago Press.<br />
Chao, A., S-M. Lee, and S-L. Jeng. 1992. Estimating<br />
population size for capture-recapture data when<br />
capture probabilities vary by time and individual<br />
animal. Biometrics 48:201-216.<br />
Coblentz, B. E. 1997. Subsistence consumption of coral<br />
reef fish suggests non-sustainable extraction. Conserv.<br />
Biol. 11:559-561.<br />
Connor, R. C., R. S. Wells, J. Mann, and A. J. Read.<br />
2000. The bottlenose dolphin: social relationships<br />
in a fission-fusion society. In Cetacean societies: field<br />
studies of dolphins and whales, ed. J. Mann, R. C.<br />
Connor, P. L. Tyack, and H. Whitehead, 91-126.<br />
Chicago and London: The University of Chicago<br />
Press.<br />
Defran, R. H., D. W. Weller, D. Kelly, and M. A. Espinoza.<br />
1999. Range characteristic of Pacific Coast<br />
bottlenose dolphins in the Southern California<br />
Bight. Mar. Mamm. Sci. 15:381-393.<br />
Fair, P. A., and P. R. Becker. 2000. Review of stress in<br />
marine mammals. J. Aquat. Ecosys. Stress Recov. 7:<br />
335-354.<br />
Grigg, E., and H. Markowitz. 1997. Habitat use by<br />
bottlenose dolphins (Tursiops truncatus) at Turneffe<br />
Atoll, Belize. Aquat. Mamm. 23:163-<strong>17</strong>0.<br />
Harvell, C. D., et al. 1999. Emerging marine diseases—<br />
Climatic links and anthropogenic factors. Science<br />
285:1505-1510.<br />
Jackson, J. B. C. 1997. Reefs since Columbus. Coral<br />
Reefs 16, Suppl.: S23-S32.<br />
Jackson, J. B. C., et al. 2001. Historical overfishing and<br />
the recent collapse of coastal ecosystems. Science<br />
293:629-638.<br />
Maze, K.S., and B. Würsig. 1999. Bottlenose dolphins<br />
of San Luis Pass, Texas: Occurrence patterns, sitefidelity,<br />
and habitat use. Aquat. Mamm. 25:91-103.<br />
McClanahan, T. R., and N. A. Muthiga. 1998. An ecological<br />
shift in a remote coral atoll of Belize over 25<br />
years. Environ. Conserv. 25:122-130.
Rogers, C. A., B. J. Brunnick, D. L. Herzing, and J. D.<br />
Baldwin. 2004. The social structure of bottlenose<br />
dolphins, Tursiops truncatus, in the Bahamas. Mar.<br />
Mamm. Sci. 20:688-708.<br />
Rossbach, K. A., and D. L. Herzing. 1999. Inshore and<br />
offshore bottlenose dolphin (Tursiops truncatus)<br />
communities distinguished by association patterns<br />
near Grand Bahamas Island, Bahamas. Can. J. Zoolog.<br />
77:581-592.<br />
Sedberry, G. R., H. J. Carter, and P. A. Barrick. 1999. A<br />
comparison of fish communities between protected<br />
and unprotected areas of the Belize reef ecosystem:<br />
implications for conservation and management.<br />
Proceedings of the Gulf and Caribbean Fisheries<br />
Institute 45:95-127.<br />
Stoddart, D. R. 1962. Three Caribbean atolls: Turneffe<br />
Islands, Lighthouse Reef, and Glover’s Reef, British<br />
Honduras. Atoll Res. Bull. 87:1-147.<br />
Tanaka, S. 1987. Satellite radio tracking of bottlenose<br />
NOTES <strong>17</strong>7<br />
dolphins Tursiops truncatus. Nippon Suisan Gakkaishi<br />
53:1327-1338.<br />
Wells, R. S., and M. D. Scott. 1999. Bottlenose dolphin<br />
Tursiops truncatus. InHandbook of marine mammals:<br />
Volume VI, The second book of dolphins and porpoises,<br />
ed. S. H. Ridgeway and R. J. Harrison, 137-182. San<br />
Diego, CA Academic Press.<br />
Wilson, B., K. Grellier, P. S. Hammond, G. Brown, and<br />
P. M. Thompson. 2000. Changing occurrence of<br />
epidermal lesions in wild bottlenose dolphins.<br />
Mar. Ecol-Prog. Ser. 205:283-290.<br />
Whitehead, H., and L. Weilgart. 2000. The sperm whale:<br />
social females and roving males. In Cetacean societies:<br />
field studies of dolphins and whales, ed. J. Mann, R. C.<br />
Conner, P. L. Tyack, and H. Whitehead, 154-<strong>17</strong>2. Chicago<br />
and London: The University of Chicago Press.<br />
Wood, C. J. 1998. Movement of bottlenose dolphins<br />
around the south-west coast of Britain. J. Zool.,<br />
London 246:155-163.
Coral Reefs (1997) 16, Suppl.: S23—S32<br />
Reefs since Columbus<br />
J. B. C. Jackson<br />
Center for Tropical Paleoecology and Archeology, Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Republic of Panama<br />
Accepted: <strong>17</strong> January 1997<br />
Abstract. History shows that Caribbean coastal ecosystems<br />
were severely degraded long before ecologists began<br />
to study them. Large vertebrates such as the green turtle,<br />
hawksbill turtle, manatee and extinct Caribbean monk<br />
seal were decimated by about 1800 in the central and<br />
northern Caribbean, and by 1990 elsewhere. Subsistence<br />
over-fishing subsequently decimated reef fish populations.<br />
Local fisheries accounted for a small fraction of the fish<br />
consumed on Caribbean islands by about the mid nineteenth<br />
century when human populations were less than<br />
one fifth their numbers today. Herbivores and predators<br />
were reduced to very small fishes and sea urchins by the<br />
1950s when intensive scientific investigations began. These<br />
small consumers, most notably Diadema antillarum, were<br />
apparently always very abundant; contrary to speculation<br />
that their abundance had increased many-fold due to<br />
overfishing. Studying grazing and predation on reefs<br />
today is like trying to understand the ecology of the<br />
Serengeti by studying the termites and the locusts while<br />
ignoring the elephants and the wildebeeste. Green turtles,<br />
hawksbill turtles and manatees were almost certainly<br />
comparably important keystone species on reefs and<br />
seagrass beds. Small fishes and invertebrates feed very<br />
differently from turtles and manatees and could and can<br />
not compensate for their loss, despite their great abundance<br />
long before overfishing began. Loss of megavertebrates<br />
dramatically reduced and qualitatively changed<br />
grazing and excavation of seagrasses, predation on<br />
sponges, loss of production to adjacent ecosystems, and<br />
the structure of food chains. Megavertebrates are critical<br />
for reef conservation and, unlike land, there are no coral<br />
reef livestock to take their place.<br />
Introduction<br />
The status and trends of the world’s coral reefs are highly<br />
controversial and, until very recently, have attracted far<br />
less attention than has been lavished on the decline of<br />
tropical forests. This is partly a function of our greater<br />
ignorance about coral reefs and their briefer period of<br />
study. But it is also true that coral reef ecologists have<br />
been so devoted to dissecting small-scale processes that<br />
they have not seen the reefs for the corals.<br />
Ecology is a young science, with little reliable<br />
descriptive information from before the 1920s (Elton<br />
1927; Hutchinson 1978) and virtually no time series<br />
population data extending of back for more than a century.<br />
The situation is even worse for marine communities<br />
like coral reefs because of the difficulty of making observations<br />
underwater. Thus reef ecologists have been<br />
forced to try to explain patterns of distribution and<br />
abundance exclusively in terms of the events of the<br />
past few years using ‘‘real time’’ observations and<br />
experiments. Their success is manifest in our rapidly<br />
growing understanding of how environmental variation<br />
and biological interactions can shape reef communities<br />
(Connell 1978; Hughes 1989; Knowlton et al. 1990), although<br />
the importance of events rare on the scale of<br />
human lifetimes is only beginning to be understood<br />
(Woodley et al. 1981; Jackson 1991; Knowlton 1992;<br />
Hughes 1994).<br />
In the process of these endeavors, however, reef ecologists<br />
have turned their backs on history and assumed<br />
that what they were studying was ‘‘normal,’’ despite the<br />
lack of rigorous baseline data for the condition of coral<br />
reefs preceding the industrial revolution and the onset of<br />
worldwide, exponential human population growth. As<br />
a result, many reef ecologists have concluded that coral<br />
reefs are healthy in the face of overwhelmingly increasing<br />
evidence to the contrary. Indeed, the idea for this study<br />
arose from my feeling like Cassandra in the face of such<br />
denial during the 1993 Miami Colloquium on Global<br />
Aspects of Coral Reefs: Health, Hazards and History. Most<br />
felt at the beginning of the colloquium that the condition<br />
of coral reefs was on the whole rather good, despite<br />
Wilkinson’s (1992) plenary paper at the 7th <strong>International</strong><br />
Coral Reef Symposium about widespread devastation<br />
worldwide. However, by the end of the Colloquium, there<br />
was a beginning acceptance that the situation was perhaps<br />
bad in the Caribbean, but only recently, and probably not<br />
in the pacific.
S24<br />
The problem is that everyone, scientists included, believes<br />
that the way things were when they first saw them is<br />
natural. However, modern reef ecology only began in the<br />
Caribbean, for example, in the late 1950s (Goreau 1959;<br />
Randall et al. 1961; Randall 1965) when, enormous changes<br />
in coral reef ecosystems had already occurred. The same<br />
problem now extends on an even greater scale to the<br />
SCUBA diving public, with a whole new generation of<br />
sport divers who have never seen a ‘‘healthy’’reef, even by<br />
the standards of the 1960s. Thus there is no public perception<br />
of the magnitude of our loss.<br />
Another insidious consequence of this ‘‘shifting baseline<br />
syndrome’’ (Pauly 1995; Sheppard 1995) is a growing ecomanagement<br />
culture that accepts the status quo,andfiddles<br />
with it under the mantle of experimental design and statistical<br />
rigor, without any clear frame of reference of what it is<br />
they are trying to manage or conserve. These are the coral<br />
reef equivalents of European ‘‘hedgerow ecologists’’ arguing<br />
about the maintenance of diversity in the remnant<br />
tangle between fields where once there was only forest.<br />
Let me say from the start that I am not going to make<br />
some romantic appeal to set back the clock, nor propose<br />
draconian scenarios that ignore the realities of inexorable<br />
human population growth and underdevelopment. Instead,<br />
my goal is to set the stark realities we face in a deeper<br />
historical perspective than the last few decades in order to<br />
(1) silence absurd notions of multiuse sustainability and (2)<br />
help to define better the limited alternatives available. I will<br />
limit my discussion to the Caribbean, and principally to<br />
Jamaica, because I know the Caribbean best and Jamaica is<br />
arguably the worst case today in that region. Jamaica also<br />
offers the best historical record, which extends back nearly<br />
350 years, and provides startlingly good but unexploited<br />
information for ecological assessment.<br />
I will first work backward from the present to re-examine<br />
aspects of the classic story relating overfishing, the mass<br />
mortality of the long-spined sea urchin Diadema antillarum,<br />
and the collapse of Caribbean coral reefs in Jamaica and<br />
elsewhere (Lessios et al. 1984; Hay 1984; Hughes 1994). This<br />
is not to deny the important consequences of overfishing,<br />
but to show that we have been at least partly wrong about<br />
the historical role of the sea urchin. I will then work<br />
forward from 1492 to examine the extraordinarily rapid<br />
depletion of large consumers on coral reefs and their environs,<br />
which were once the equivalents of the wildebeeste<br />
and elephants of the Serengeti plains (Sinclair and Arcese<br />
1995). For this purpose, I will emphasize the Jamaican<br />
basedfisheryofthegreenturtleChelonia mydas because the<br />
data are the best; but the same sort of story applies to<br />
sharks, rays, groupers, manatees and the extinct Caribbean<br />
monk seal. I will then depart from the general theme of<br />
overfishing to briefly consider inputs from the land. These<br />
have almost certainly been of comparable importance to<br />
overfishing but the data are less complete. Finally, I will<br />
return to Jamaica to examine the fishing history there since<br />
Columbus, which clearly shows that coral reef ecosystems<br />
had begun to fall apart in the eighteenth century.<br />
Diadema, damselfish and overfishing<br />
The conventional story (Hay 1984; Lessios 1988; Hughes<br />
1994) is that intense overfishing allowed Diadema<br />
to increase because of reduced predation by fishes, and<br />
competitive release for algal food that was no longer<br />
consumed by larger herbivores. This increase in Diadema<br />
compensated for the loss of herbivorous fishes and kept<br />
down the growth of seaweeds on reefs. Then when the<br />
Diadema suddenly died, there were no other consumers<br />
capable of cropping the seaweeds which soon overgrew<br />
and killed most of the reef corals at depths down to about<br />
50 m.<br />
What are the facts and what is the inference in this<br />
story? The facts are that (1) overfishing was extreme by<br />
any standard, (2) Diadema was extremely abundant before<br />
the mass mortality in 1983, (3) mass mortality of Diadema<br />
allowed a dramatic increase in seaweeds which overgrew<br />
and smothered corals. The inference is that (1) decrease in<br />
large fish allowed a large increase in Diadema, (2) increase<br />
in people caused the decrease in fish in a roughly proportional<br />
fashion, and (3) the effects of people were relatively<br />
recent. Regarding the latter point, for example, Hughes’<br />
(1994) graph of human population increase starts in 1850<br />
and is introduced within a section titled ‘‘Overfishing:<br />
1960s to Present’’.<br />
Let us examine some problems with this inference<br />
before reviewing the historical data. Hay’s (1984) pioneering<br />
regional study of Diadema versus fish grazing confounds<br />
‘‘overfished’’ and ‘‘less fished’’ with geography. Hay<br />
was very careful to avoid terms like ‘‘pristine’’ and ‘‘unfished’’,<br />
unlike many who have referred to his work subsequently.<br />
However, all of his ‘‘overfished’’ reefs are on<br />
islands in the central Caribbean, and all but one of his<br />
‘‘less fished’’ reefs are on the mainland. Moreover, Levitan’s<br />
(1992) ingenious study of Diadema nutrition over the<br />
last century shows an even greater geographic effect. Levitan<br />
(1991) showed experimentally that the ratio of the size<br />
of the feeding apparatus to the size of the animal test<br />
increases in inverse proportion to the food supply. He<br />
then examined changes in the ratio through time using<br />
museum specimens, and found a significant but surprisingly<br />
small increase over the past century when Diadema<br />
populations were inferred to have exploded due to overfishing.<br />
In contrast, geographic variation accounted for<br />
much more (23%) of the total variation observed than<br />
that due to time. Ordination of coral abundance data<br />
from reefs around the Caribbean compiled by Liddell and<br />
Ohlhorst (1988) also shows strong island-mainland differences<br />
in overall reef community structure (Jackson et al.<br />
1996).<br />
So is it possible that Diadema were abundant in the<br />
Caribbean before Columbus arrived? It turns out that<br />
Diadema has long attracted commentary because of its<br />
great abundance and reputation of being dangerous<br />
(Table 1). Moreover, all of the authors cited except Young<br />
were professional or amateur naturalists, with extensive<br />
experience dredging or (in the case of Beebe 1928) diving<br />
in tropical waters. All were well known for the reliability<br />
and accuracy of their observations and were not the type<br />
to mistake an occasional aggregation or hearsay for genuinely<br />
great abundance. These sources make it clear that<br />
Diadema was indeed very abundant in the seventeenth<br />
century when human populations were very small, and<br />
therefore long before overfishing could have caused their<br />
increase.
Table 1. Caribbean Diadema lore<br />
Source Location Quoted text<br />
Beebe 1928 Haiti Under every bit of coral 2 in great abundance<br />
Clark 1919 Jamaica and Dry<br />
Tortugas<br />
Nutting 1919 Barbados Antigua<br />
and Bahamas<br />
On and about the coral reefs, the dreaded poisonous ‘‘black<br />
sea egg’’ (Centrechinus antillarum) is common and on certain<br />
areas it is so numerous that a person can scarcely move<br />
about without touching one.<br />
No one goes bathing or into the water for any purpose in<br />
this region without being warned against the danger of being<br />
wounded by the cruel black spines of this ubiquitous<br />
sea-urchin. It is found almost everywhere in shallow water,<br />
both on sandy and rocky bottom.<br />
The all too familiar black sea-egg Diadema antillarum is<br />
abundant here, as it is everywhere that I have collected in the<br />
West Indies (therefore includes his 1895 expedition to the<br />
Bahamas, not seen)<br />
Henderson 1914 Cuba We were then upon the inner edge of the main reef upon<br />
which any further progress would have been difficult on<br />
account of the rapidly increasing numbers of the long<br />
black-spined sea urchins, the diademas2<br />
2the usual presence of the net [dredging] of the diadema<br />
sea-urchin2<br />
During the hours of bright sunshine the diademas seek cover<br />
under the rocks2 In the late afternoon2they issue forth<br />
en masse in search of food and probably continue their slow<br />
wanderings throughout the night. In localities where hiding<br />
places are few, such as upon sandy patches in or near a reef,<br />
the diademas are always more or less in evidence.<br />
Field 1891 Jamaica The most strikingly conspicuous of all the creatures about<br />
the coral reefs. As one approaches the cay, far down in the<br />
water are seen numerous irregularly shaped black patches of<br />
varying extent.<br />
2around and under its branches (elkhorn coral) are<br />
crowded together great numbers of this large black urchin.<br />
2what formidable defense against attack they present<br />
when crowded so closely together.<br />
Aggasiz 1883 Dry Tortugas 2on somewhat deeper regions (of the reef) we find pockets<br />
filled with large Diadematidae.<br />
Young 1847<br />
(writing of his<br />
observations<br />
in 1839)<br />
Sloane <strong>17</strong>25<br />
(writing of his<br />
observations<br />
in 1688)<br />
Nicaragua and<br />
Honduras<br />
There is also strong paleontological evidence that Diadema<br />
were abundant on Jamaican reefs long before any<br />
humans arrived there (Gordon and Donovan 1992). Diadematoid<br />
plates and spines are the most abundant<br />
echinoid remains in the 125 000 year old Falmouth<br />
Formation, constituting 90% of all identified echinoid<br />
fragments. These are almost, certainly of Diadema antillarum<br />
because the only other Caribbean Pleistocene diadematoid<br />
is Astropyga magnifica which occurs today in<br />
S25<br />
2they were both badly cut by the coral and sea eggs in<br />
diving for the things that had been upset.<br />
Numbers of sea eggs were seen in all directions, and we well<br />
knew the danger of getting amongst them, as they have long<br />
and sharp pointed spines, which inflict deep and dangerous<br />
wounds on those who chance to tread on them.<br />
2the handsome sea-eggs inviting but to betray2<br />
Jamaica The great, long prickled Sea Egg2set about on every hand<br />
with prickles, the largest being three or four inches long, with<br />
membrane round their setting on to the shell2purple deep<br />
coloured2<br />
The prickles of this Sea Echinus are very rough and considered<br />
poisonous.<br />
I have found them in great numbers on the reef by Gun-Key,<br />
or, Cayos off the Port Royal Harbour in great numbers.<br />
deeper water than the backreef environment of the Falmouth<br />
Formation (Donovan and Gordon 1993).<br />
In conclusion, Diadema apparently has been the most<br />
abundant sea urchin on Caribbean reefs for at least<br />
125 000 years. Its abundance still may have increased<br />
historically due to overfishing, but we lack the quantitative<br />
data to tell. However, it now seems unlikely that any<br />
such increase was as great as that, for example, of<br />
Echinometra mathaei in response to over-fishing in
S26<br />
Kenyan lagoons (McClanahan and Muthiga 1988;<br />
McClanahan et al. 1996). Thus Diadema abundance is at<br />
best a secondary consequence of the degradation of Caribbean<br />
reefs, and we need to look for other indicators of<br />
faunal change in response to human interference. For this<br />
purpose, let us now turn to changes in the abundance of<br />
the really large consumers in reef environments, such as<br />
green turtles.<br />
How many turtles in 1492?<br />
Big animals are ecologically important, not only because<br />
of the amount of plants or animals each individual consumes,<br />
but also because of the physical and biological<br />
disturbance they cause, which fundamentally alters their<br />
environment and affects other species. Perhaps the best<br />
studied example is the Serengeti ecosystem of east Africa<br />
(Sinclair 1995a; Sinclair and Arcese 1995). Long distance<br />
migration (wildebeeste) and growth to very large size<br />
(elephant, hippopotamus, rhinoceros and buffalo) result in<br />
virtual escape from predation, so that such herbivores are<br />
‘‘bottom up’’ regulated by food limitation rather than ‘‘top<br />
down’’ by predators (Sinclair 1995a). The enormously<br />
abundant wildebeeste is a keystone species because its<br />
grazing and migration directly or indirectly affect almost<br />
everything else. Wildebeeste grazing alters the protein<br />
content of the grass and stimulates growth of new shoots,<br />
increasing the available food supply for smaller grazers,<br />
and possibly also for themselves (McNaughton and Banyikwa<br />
1995). There is also strong evidence for alternate<br />
stable states of vegetation (woodland versus grassland),<br />
maintained by grazing and disturbance by elephants and<br />
by fire (Dublin 1995; Sinclair 1995b).<br />
None of these topics have been studied as well on coral<br />
reefs and surrounding seagrass environments, but what we<br />
do know strongly suggests similarly important ecosystem<br />
effects of large species (Heinsohn et al. 1977; Ogden 1980;<br />
Thayer et al. 1984; Lanyon et al. 1989; Sheppard et al.<br />
1992). Green turtles crop ‘‘turtlegrass’’ ¹halassia testudinum<br />
to only a few centimeters above the bottom, and<br />
cause erosion and pits in the rhizomal mat of turtlegrass<br />
beds. Manatees and dugongs do much the same, sometimes<br />
ripping up entire beds of seagrasses and other<br />
aquatic vegetation, and thereby causing even greater<br />
physical disturbance. Stingrays excavate pits in seagrass<br />
beds foraging for mollusks beneath the rhizome mat,<br />
hawksbill turtles rip up sponges, and large parrotfishes<br />
occasionally bite to pieces entire coral colonies for unknown<br />
reasons.<br />
I have nothing new to add to such observations, except<br />
the comment that I have not even seen most of these large<br />
animals underwater for twenty years or more, and some of<br />
them never at all, despite thousands of hours SCUBA<br />
diving on and around coral reefs. Instead, I want to dwell<br />
on the past enormity of the populations of such creatures<br />
using green turtles as an example. My calculations are<br />
rough in ways that responsible turtle biologists have shied<br />
away from, and the data are 15 to 300 years old. Nevertheless,<br />
it is essential to try, because we have no conception of<br />
the way things were. Why, after all, are so many hundreds<br />
of sites around the Caribbean, such as the Dry ¹ortugas,<br />
named after turtles that almost no living person has<br />
ever seen?<br />
Calculations based on old hunting data from the Cayman<br />
Islands<br />
Estimates of the size of pre-columbian human populations<br />
in the Caribbean are controversial and depend on differing<br />
interpretations of archeological and historical sources<br />
(Roberts 1989). Nevertheless, populations of Hispaniola,<br />
Jamaica and Cuba certainly ranged in the hundreds of<br />
thousands, perhaps even in the millions, and were sustained<br />
by a highly productive agricultural system supplemented<br />
by fishing and hunting (Sauer 1966; Rouse 1992).<br />
These early Americans were reduced by conquest, slavery<br />
and disease to only a few thousand by 1600. Subsequent<br />
Spanish colonization was slow, so that there were only<br />
about five thousand people in Jamaica when the English<br />
captured the island in 1655 (Long <strong>17</strong>74). There was also<br />
no effective agricultural base, so the English turned immediately<br />
for food to the vast populations of green turtles<br />
that nested on Grand Cayman Island (Table 2). These<br />
abundant turtles were essential to the rapid growth and<br />
success of Jamaica as England’s most important colony of<br />
the time, and indeed provided most of the meat consumed<br />
there until the <strong>17</strong>30s (Sloane <strong>17</strong>07, <strong>17</strong>25; Long <strong>17</strong>74; Lewis<br />
1940; King 1982).<br />
Thirty years later, the Cayman Islands fishery had<br />
grown to approximately 40 sloops and 120 to 150 men<br />
who brought back to Jamaica some 13 000 turtles per year<br />
between 1688 and <strong>17</strong>30. Let us assume that the sex ratio<br />
and migration interval (time between years that females<br />
reproduce) were 1: 1 and 2.5 years respectively, just as they<br />
are today (Bjorndal 1982). Thus, the proportion of the<br />
adult population (N ) that are nesting females (N )is<br />
<br />
given by<br />
N "N /0.5 (sex ratio)0.4 (migration interval)"5N .<br />
<br />
Let us further assume that hunters in 1688 captured only<br />
1% of the nesting female turtles per year. This is an<br />
arbitrary but almost certainly conservative guess for the<br />
purpose of illustration, based on the impression from the<br />
early descriptions that the beaches all around Grand<br />
Cayman were literally covered by turtles, so that 13 000<br />
would have been a very small fraction of the total. Moreover,<br />
female green turtles require 40—60 years to reach<br />
reproductive maturity (Bjorndal and Zug 1995), so that<br />
harvested females could not have been replaced for half<br />
the century-long duration of the fishery. Despite this,<br />
however, 13 000 reproductively mature females were harvested<br />
annually over 42 y, for a total (with the above<br />
correction) of more than 2.5 million on this basis alone.<br />
Based on the assumption of an initial catch rate of 1%, the<br />
estimated total adult population (N ) based on the early<br />
<br />
hunting data is<br />
N "513 000 (number harvested)/0.01 (% N caught)<br />
<br />
"6.5 million.<br />
Further assume that there were five additional precolumbian<br />
green turtle rookeries roughly equal to Grand Cay-
Table 2. Historical accounts of<br />
the early great abundance of<br />
green turtles in the Caribbean<br />
Andres Bernaldez,<br />
writing about<br />
Columbus’<br />
2nd voyage in 1494<br />
Ferdinand Columbus,<br />
writing about the 4th<br />
voyage in 1503<br />
Edward Long (<strong>17</strong>74),<br />
writing of the late<br />
1600s<br />
not seen, cited in Lewis 1940<br />
Southeastern<br />
Cuba<br />
Cayman<br />
Islands<br />
man including Bermuda, Bahamas, Florida Keys,<br />
Costa Rica, and Isla Aves (now less than a mile long but<br />
historically much larger). Then the estimated total adult<br />
population for the entire precolumbian Caribbean is five<br />
to six times the Grand Cayman estimate, or about 33 to 39<br />
million. This is about 15 to 20 times the abundance and<br />
biomass of large ungulates in the Serengeti today (Sinclair<br />
1995a)!<br />
Calculations based on carrying capacity<br />
The following is based on Bjorndal’s (1982) study of green<br />
turtle nutrition and life history. There is apparently no<br />
reliable compilation of the total area of seagrasses in the<br />
wider Caribbean. However, we can assume that roughly<br />
ten percent of the total shelf area, which is 660 000 km<br />
excluding south Florida (Munroe 1983), is covered by<br />
seagrasses for a total of 66 000 km. This is probably<br />
conservative, since the mapped area of seagrasses for<br />
south Florida alone is 5500 km (Ziemann 1982). Bjorndal<br />
(1982) calculated that the carrying capacity of closely<br />
cropped (2.5 cm) ¹halassia is one 100 kg adult female per<br />
72 m per year, which rounding up to one turtle per<br />
100 m, gives 10 000 adult females per km. Assume further<br />
that the carrying capacity is the same for males as for<br />
females. Then the estimated total adult population (N )<br />
for the entire Caribbean is<br />
N "10 00066 000"660 million,<br />
which is about 20 times the estimate based on the old<br />
hunting data. Of course, sharks and other predators of<br />
principally juvenile turtles were also very much more<br />
abundant. However, Sinclair’s (1995a) generalization<br />
about the predominantly ‘‘bottom up’’ regulation of large,<br />
migratory herbivores suggests that the abundance of<br />
large, migrating adult green turtles would have approached<br />
carrying capacity even in the face of intense<br />
predation on juveniles.<br />
Differences in feeding between green turtles and other<br />
herbivores<br />
One adult green turtle consumes roughly the same<br />
amount of turtlegrass as 500 large sea urchins like Dia-<br />
West of the<br />
Cayman<br />
Islands<br />
S27<br />
But in those twenty leagues, they saw very many more, for<br />
the sea was thick with them, and they were of the very<br />
largest, so numerous that it seemed that the ships would run<br />
aground on them and were as if bathing in them.<br />
2in sight of two very small and low islands, full of tortoises,<br />
as was all the sea about, insomuch that they looked like little<br />
rocks2<br />
2it is affirmed, that vessels, which have lost their latitude in<br />
hazy weather, have steered entirely by the noise which these<br />
creatures make in swimming, to attain the Cayman isles.<br />
dema antillarum or ¹ripneustes ventricosus, which works<br />
out to a potential increase of about 5 sea urchins per m of<br />
seagrass beds throughout the Caribbean (calculations<br />
based on data in Thayer et al. 1984). Much more significantly,<br />
however, there are profound differences in the<br />
ways turtles versus sea urchins and herbivorous fishes<br />
graze on turtlegrass, and how well they process what they<br />
eat, with important consequences for the structure and<br />
function of the entire turtlegrass ecosystem (Ogden 1980;<br />
Thayer et al. 1982, 1984; Ogden et al. 1983). Sea urchins<br />
and fishes tend to feed indiscriminantly on turtlegrass, or<br />
on the older parts of the blades, whereas green turtles crop<br />
blades close to their base. They also return repeatedly to<br />
the same discrete grazing plots which may be maintained<br />
for a year or more. Moreover, grazing sea urchins and<br />
fishes feed principally on the cell contents of seagrass<br />
blades, due to lack of appropriate enzymes or microflora<br />
to digest cell walls, whereas green turtles rely on microbial<br />
fermentation in the hindgut to digest cell walls as well as<br />
their contents (Thayer et al. 1984).<br />
Repeated grazing of the same plots of turtlegrass by<br />
green turtles temporarily increases the nutritional quality<br />
of the blades for the turtles (Thayer et al. 1984). However,<br />
it also stresses the plants and eventually reduces turtlegrass<br />
productivity, when turtles presumably move on to<br />
feed elsewhere. Turtle grazing also results in a roughly<br />
15-fold decrease in the supply of nitrogen to seagrass roots<br />
and rhizomes, due to greatly decreased accumulation of<br />
detritus and digestion of cell walls (Thayer et al. 1984).<br />
Nitrogen in turtle feces and urine is also released over<br />
a much wider area with resulting net export to adjacent<br />
coral reefs and other adjacent ecosystems.<br />
These differences in the effects of grazing by small and<br />
large herbivores extend to dugongs and manatees that<br />
also possess hindgut microflora that digest cell walls<br />
(Thayer et al. 1984), and were also formerly very abundant<br />
throughout tropical seas (Dampier <strong>17</strong>29, Sheppard et al.<br />
1992). Similar ecosystem effects almost certainly transpired<br />
on coral reefs due to the virtual disappearance of<br />
large predators such as hawksbill turtles, groupers, and<br />
sharks that were also extremely abundant historically<br />
(Ibid, King 1982; Limpus 1995). For example, the hawksbill<br />
turtle, Eretmochelys imbricata, feeds almost exclusively<br />
on sponges which commonly display large, characteristic<br />
feeding scars in areas where the turtles are still common<br />
(Meylan 1985, 1988; van Dam and Diez, in press). Hawksbills<br />
can rip big sponges apart, and in the process facilitate
S28<br />
predation by other sponge feeders that cannot penetrate<br />
the heavy armor of many sponges such as Geodia; unlike<br />
the much smaller angelfishes that also feed almost exclusively<br />
on reef sponges (Randall and Hartman 1968). Hawksbills<br />
feed today mostly on non-toxic astrophorid and<br />
hadromerid sponges, but this may not have been true in<br />
the past when, by analogy to green turtles (King 1982;<br />
Limpus 1995; this study), hawksbills almost certainly<br />
numbered in the tens of millions. Thus non-toxic sponges<br />
may have been proportionally rarer before hawksbills<br />
were intensely harvested.<br />
Large herbivores and carnivores are ecologically extinct<br />
on Caribbean coral reefs and seagrass beds, where food<br />
chains are now dominated by small fishes and invertebrates<br />
(Hay 1984, 1991; Knowlton et al. 1990). Moreover,<br />
similar depletion of megavertebrates is almost complete<br />
throughout the Indo-Pacific (Sheppard et al. 1992; Limpus<br />
1995). Small consumers cannot fully compensate for<br />
the loss of megavertebrates because they cannot capture,<br />
consume or process their prey in the same ways as larger<br />
species. Many small herbivores are also feeding specialists,<br />
and live commonly on prey that are chemically defended<br />
against the much larger consumers that have now disappeared<br />
(Hay 1991, in press). Thus coral reef ecosystems<br />
must function in fundamentally different ways than only<br />
a few centuries ago. Similarly great changes are going on<br />
right now in east Africa (Sinclair and Arcese 1995). They<br />
also occurred 10 000 years ago in neotropical forests when<br />
over 15 genera of large herbivores became extinct (Janzen<br />
and Martin 1982), before which forests were probably<br />
more of a mixture of open forest and grassland than the<br />
dense tropical forest we imagine as more natural (Janzen<br />
and Wilson 1983). As a result, neotropical herbivorous<br />
food chains are now dominated by insects and small<br />
mammals, except where free-ranging livestock may have<br />
partially redressed the balance. But there are no such<br />
livestock on coral reefs!<br />
Effects from the land<br />
Sedimentation caused by deforestation and poor agricultural<br />
practice, eutrophication, and oil pollution have<br />
greatly increased along Caribbean coasts during the<br />
past few decades (Rodriguez 1981; Lugo et al. 1981),<br />
causing widespread and dramatic decline of coral reefs<br />
and associated marine communities throughout the<br />
region (Cortés and Risk 1985; Rogers 1985, 1990;<br />
Tomascik and Sander 1985, 1987a, b; Bak 1987; Jackson<br />
et al. 1989; Guzmán et al. 1991). A common assumption in<br />
such studies is that most reefs were not seriously affected<br />
by runoff from the land before the observations began.<br />
This allows the investigator to designate ‘‘unaffected’’ reefs<br />
or corals that can be used as a baseline to measure the<br />
effects of a particular source of pollution, such as an oil<br />
spill.<br />
New evidence, however, suggests that great ecological<br />
changes due to runoff began long before modern ecological<br />
analyses of Caribbean reefs. For example, Montastrea<br />
‘‘annularis’’ and Porites spp. were the dominant reef building<br />
corals around Barbados in the 1960s and 1970s (Lewis<br />
1960; Macintyre 1968; Stearn et al. 1977). These began to<br />
decline dramatically in the 1970s and 1980s, due primarily<br />
to runoff and eutrophication caused by exponential increase<br />
in populations of residents and tourists (Tomascik<br />
and Sander 1985, 1987a, b; Bell and Tomascik 1993).<br />
However, extensive deforestation began in Barbados in<br />
1627 for sugar plantations, after which the vegetation of<br />
the island was completely destroyed three times by hurricanes<br />
(references in Lewis 1984); with untold increases in<br />
runoff of sediments from the land. Moreover, shallow reefs<br />
at that time were dominated by the elkhorn coral Acropora<br />
palmata, which persisted in huge tracts all along the<br />
southern and western coasts of the island until the 1920s<br />
(Nutting 1919; Lewis 1984; Bell and Tomascik 1993), and<br />
the same was true throughout the Late Pleistocene (Mesolella<br />
1957; Jackson 1992). Degradation of these reefs and<br />
seagrass beds was clearly visible in aerial photographs<br />
taken in the 1950s (Lewsey 1978) when elkhorn corals<br />
were rare (Lewis 1960).<br />
Similar decline in Acropora palmata and A. cervicornis<br />
occurred all along the coast of lower Central America in<br />
the 1970s and 1980s due to disease, deforestation, algal<br />
overgrowth due to the decline of Diadema, coral bleaching,<br />
oil spills, and other factors (Cortés and Risk 1985;<br />
Cortés 1993; Guzmán et al. 1991; Ogden and Ogden 1993).<br />
Live coral cover along the Caribbean coast of central<br />
Panama has declined by 50—90% in the past ten years<br />
(Guzmán and Jackson, unpublished data). However, there<br />
were hidden signs of danger long before, as measured by<br />
a steady decline of nearly 50% in the growth rate of the<br />
massive coral Siderastrea siderea over the past century<br />
(H. Guzmán, unpublished data). This is all the more<br />
remarkable because S. siderea generally prospers in turbid<br />
coastal environments unsuitable to most other Caribbean<br />
coral species. Guzmán’s work obviously needs to be replicated<br />
elsewhere, but it strongly suggests that reef environments<br />
had begun to deteriorate at least 100 years<br />
before coral cover began to seriously decline. Isotopic<br />
ratios provide excellent proxy records of climate change,<br />
but simple growth rates and incidence of injuries are<br />
measures of coral fitness through time (Jackson 1982). As<br />
such, they may be among the best available measures of<br />
coral reef health (Dodge and Vaisnys 1977; Guzmán et al.<br />
1994; Jackson 1995).<br />
Fishing and people in Jamaica and San Blas<br />
The green turtle fishery in the Cayman Islands crashed in<br />
the latter half of the eighteenth century, and was entirely<br />
gone by 1800 when the Cayman islanders moved on to do<br />
the same thing to the turtles of the Moskito Coast (Long<br />
<strong>17</strong>74; Lewis 1940; Carr 1956; Neitschmann 1973, 1982;<br />
King 1982). Fishing on Jamaican reefs was inadequate to<br />
make up for the loss of turtles. By 1881 locally caught fish<br />
accounted for only 15% of the total consumed in Jamaica,<br />
with imported dried and preserved fish from the temperate<br />
zone making up the balance (Duerden 1901). This was<br />
probably true by the early 1800s, but there are not enough<br />
quantitative data. Moreover, extensive trials had clearly<br />
demonstrated that there was little prospect for improvement<br />
of local fisheries by trawling or longline fishing which<br />
are unsuitable for areas of coral reefs (Duerden 1901).
Table 3. Excerpts from Ernest F.<br />
Thompson’s 1945 report ¹he<br />
fisheries of Jamaica<br />
Despite these realities, Duerden’s (1901) official report<br />
was surprisingly optimistic and, in a still all too familiar<br />
tone, called for more ‘‘scientific investigations and encouragement’’<br />
to improve the marine resources of the Caribbean<br />
region. Half a century later the fisheries of Jamaica<br />
had not improved and were clearly unimprovable<br />
(Thompson 1945, Table 3). Thompson was far ahead of his<br />
time in recognizing the need for greatly reducing the<br />
numbers of fishermen, helping them to find alternative<br />
livelihoods, and focusing instead on the economic opportunities<br />
of fishing for tourism. But his report was apparently<br />
ignored and overfishing continued unabated.<br />
By far the most extensive coral reef fisheries research<br />
project in the Caribbean was carried out from 1969—1973<br />
all around Jamaica and in the Pedro Cays (Munroe 1983).<br />
Overfishing was accepted as an established fact, and it was<br />
‘‘shown that for most areas of the Jamaican Shelf, the<br />
fishing intensity is sufficient to ensure that extremely few<br />
fishes survive for more than a year after recruitment, and<br />
the proportion of fishes which survive to spawn must be<br />
extremely small.’’ The report goes on to recommend<br />
a mesh size for fish traps of 6.60 cm maximum aperture<br />
that would provide a maximum yield of barely reproductive<br />
juveniles with a maximum fishing intensity of 1.5<br />
canoes/km, without any consideration of the possible<br />
consequences for the health of the entire coral reef ecosystem.<br />
Of course, it is easy to be unfairly critical with the<br />
hindsight of the Diadema debacle and the collapse of<br />
Jamaican reefs (Hughes 1994), although even in the 1970s<br />
the consequences of not having any pretty reef fishes<br />
larger than sardines on lost tourist revenues were clear.<br />
The situation in Jamaica is a story repeated everywhere<br />
throughout the Caribbean, including the traditional fisheries<br />
of indigenous peoples commonly romanticized for exhibiting<br />
wise restraint from overharvesting not observed<br />
by others. The Comarca Kuna Yala, for example, extends<br />
some 250 km along the eastern Caribbean coast of Panamá<br />
and contains extensive coral reefs, seagrass beds<br />
and mangroves (Porter 1972; Glynn 1973; Clifton<br />
et al. 1996). The population of Kuna people within the<br />
Comarca increased from less than 9000 in 1904 to less<br />
than 24 000 in 1970, when marine biological research<br />
intensified in the Comarca, and had climbed to 41 000 in<br />
1989 (Francisco Herrera, personal communication). Kuna<br />
Locally caught fish represents less than 15% of the protein fish food consumed in Jamaica2 There is<br />
little prospect of any large increase in this local catch. In fact the probability is that the local areas are<br />
already overfished (p. 5)<br />
2the greatest need for Jamaican fishermen is more in the nature of welfare than development (p. 7,<br />
Thompson’s italics).<br />
In a great many countries, fishing has proved a very valuable asset as a tourist attraction. Development<br />
of the tourist potential would require that the present usages of some sections of the community must be<br />
controlled and restricted (p. 7).<br />
The fishing situation in Jamaica can be very briefly summarised. There are too many men trying to catch<br />
too few fish. As there seems little prospect of increasing the number of fish available, the only thing to do,<br />
if a decent living standard is to be attained, is to reduce the number of fishermen. Thus the chief problem<br />
for Jamaican fishermen is to organize them, stabilise their economy and assist about four-fifths of them<br />
to drift back to agriculture from whence they came (p. 83)<br />
S29<br />
artesanal fishing has always been a small-scale enterprise,<br />
and the Comarca is closely guarded against outside exploitation.<br />
Nevertheless, reefs were severely overfished by<br />
the 1970s when large fishes were uncommon and branching<br />
acroporid and poritid corals had been mined extensively<br />
for landfill. Thus 250 km of coast were already<br />
insufficient for 24 000 Kuna, ten years before lobster fishing<br />
for external markets reduced lobster populations to<br />
critically low levels within only ten years (Chapin 1995;<br />
Ventocilla et al. 1995).<br />
So let us now return to Jamaica to consider the history<br />
of Jamaican fisheries in light of human growth, much as<br />
Hughes (1994) did, but beginning 350 years before (Fig. 1).<br />
Depopulation by the Spaniards in the sixteenth century<br />
left the island almost uninhabited until the British invasion<br />
in 1655, when marine life may well have been at its<br />
apogee of the past 10 000 years. In 1688, when Sloane was<br />
in Jamaica, there were still only 40 000 Jamaicans and<br />
Diadema was the most abundant sea urchin on coral reefs.<br />
In <strong>17</strong>93, there were 300 000 Jamaicans and breadfruit had<br />
just been introduced following extensive research by the<br />
Royal Society and the mutiny on the Bounty, to help stave<br />
off the starvation of Jamaican slaves. In 1881, there were<br />
700 000 Jamaicans and local fish accounted for only 15%<br />
of the fish consumed. In 1945 there were 1 350 000 Jamaicans,<br />
and the 5 500 000 kg of fish caught locally was still<br />
only 15% of that consumed. In 1962 there were 1 700 000<br />
Jamaicans and the fish harvest peaked at 11 000 000 kg.<br />
This was also when Goreau (1959) published his first<br />
famous paper about Jamaican coral reefs and the modern<br />
ecological perspective was born. By 1968, when I began<br />
my own research in Jamaica, there were 1 900 000 Jamaicans,<br />
and the harvest of minnow-sized fishes from Jamaican<br />
coastal waters was back down to 5 500 000 kg.<br />
Munroe’s fisheries research project was only just beginning.<br />
It is obvious that any direct relationship between human<br />
population growth and fishing in Jamaica ended in<br />
the eighteenth century when human populations were<br />
only 10% of the present. Throughout this time, Jamaicans<br />
kept on eating fish courtesy of the now collapsed Grand<br />
Banks fisheries, and the same was true throughout the<br />
Caribbean. Thus, the causes of the present ecocatastrophe<br />
are deep and historical, not just the almost ‘‘current<br />
events’’ that have passed as history before.
S30<br />
Fig. 1. Jamaican human population growth since Columbus and the<br />
depletion of local fisheries resources. Fisheries became inadequate<br />
some time in the mid nineteenth century when the local population<br />
was about 15% of that today. Value for 1492 arbitraily set at 100 000<br />
which is almost certainly much too low (Sauer 1966). Sources for<br />
population size: 1658—<strong>17</strong>68 (Long <strong>17</strong>74), 1844—1871 (Gardner 1909).<br />
1901 (Duerden 1901), 1920—2000 (Hughes 1994), 1980 (National<br />
Geographic Society 1981)<br />
Concluding remarks<br />
I hope this brief discussion will put to rest two dangerous<br />
stupidities, at least within the scientific community. The<br />
first is the placebo of sustainable use for everyone and the<br />
second is the fallacy of a ‘‘pristine’’ coral reef. Forty<br />
thousand Kuna are too many fish eaters for 250 km of<br />
coastline (Ventocilla et al. 1995), just as a few hundred<br />
thousand Jamaicans were too many fish eaters for<br />
Jamaica. The same is true for reefs everywhere else, and<br />
not just the developing world (Wilkinson 1992). Even the<br />
Great Barrier Reef cannot be used sustainably at present<br />
levels, despite the best protection in the world (Bradbury<br />
et al. 1992), and virtually all other reefs are less sustainable.<br />
There is nothing new about all this, as described so<br />
eloquently in Peter Matthiessen’s (1975, 1986) eulogies to<br />
the turtle fishermen of Grand Cayman or the striped bass<br />
fishermen of the South Fork of Long Island. However,<br />
societies desperately need to set goals and priorities that<br />
reflect the realities of our lost and dying coral reef resources,<br />
and then follow them. For example, coral reef<br />
fishes may still provide sustainable luxury food and pleasure<br />
for tourists, just as Thompson (1945) proposed, and<br />
captive breeding of aquarium fishes is almost certainly<br />
a viable enterprise. Deciding on such options to the exclusion<br />
of others, and then enforcing them, involves extremely<br />
difficult economic, political and social issues beyond the<br />
bounds of ecological science. But some such decisions are<br />
long overdue, and no more research is required to get<br />
started, because the facts about ‘‘sustainability’’ have been<br />
clear for more than one hundred years.<br />
Getting things straight is all the more important because,<br />
as far as we can tell, almost all the reef species are<br />
still there almost everywhere. The Caribbean monk seal is<br />
extinct and manatees are nearly gone, but even green<br />
turtles still number in the tens of thousands in the Caribbean.<br />
This means that it is still possible, at least in principle,<br />
to save Caribbean coral reefs; although continued<br />
human population growth makes this more and more<br />
unlikely. For example, all the species of corals encountered<br />
previously on Panamanian Caribbean reefs are still<br />
there, even on some of the most devastated reefs, despite<br />
huge decreases in coral abundance (Guzmán and Jackson,<br />
unpublished data). Thus the situation is like the East<br />
African savannas where large animals are more and more<br />
restricted and diminished, but the great majority of species<br />
alive during the Pleistocene still survive (Sinclair and<br />
Arcese 1995).<br />
For this reason, and by analogy to the Serengeti, really<br />
large marine protected areas on the scale of hundreds to<br />
thousands of square kilometers are vital to any hope of<br />
conserving Caribbean coral reefs and coral reef species.<br />
Can we restore damaged reefs? Can we control inputs<br />
from the land and harvesting? Can we manage what we do<br />
decide to invest in and use? These are the questions that<br />
really do merit more research on a monumental scale<br />
(NRC 1995). The people trying to answer them are the<br />
heros of our discipline and the only chance we have got.<br />
Acknowledgements. N. Knowlton talked me into writing all this<br />
depressing stuff down and helped in innumerable other ways. Conversations<br />
with P. Dayton, L. D’Croz, H. Guzma´ n, J. Lang, J.<br />
Ogden, R. Robertson, A. Rodaniche, and J. Wulff helped greatly to<br />
focus my approach. A. Herrera and A. Aguirre tracked down countless<br />
obscure references which the Smithsonian Libraries obtained<br />
with their customary efficiency and grace. F. Herrera kindly provided<br />
data for Kuna populations inside the Comarca which are not<br />
separated from other mainland Kuna populations in the Government<br />
of Panama statistics. N. Knowlton, H. Lessios, D. Levitan and<br />
one other reviewer criticized the manuscript and made many helpful<br />
suggestions. To all I am very grateful.<br />
References<br />
Agassiz A (1883) Explorations of the surface fauna of the Gulf<br />
Stream, under the auspices of the United States Coast Survey. II.<br />
The Tortugas and Florida reefs. Mem Am Acad Arts Sci Centennial<br />
2 : 107—132<br />
Bak RPM (1987) Effects of chronic oil pollution on a Caribbean<br />
coral reef. Mar Poll Bull 18 : 534—539<br />
Beebe W (1928) Beneath tropic seas. Blue Ribbon Books, New York<br />
Bell PRF, Tomascik T (1993) The demise of the fringing coral reefs<br />
of Barbados and of regions in the Great Barrier Reef (GBR)<br />
lagoon-impacts of eutrophication. In: Ginsburg RN (comp) Proceedings<br />
of the colloquium on global aspects of coral reefs:<br />
health, hazards and history. Rosenstiel School Mar Atmos Sci,<br />
University of Miami, Florida, pp 319—325<br />
Bernáldez A (1988) History of the Catholic sovereigns, Don Ferdinand<br />
and Don a Isabella, chapters 123—131. In: Jane C (ed) The<br />
four voyages of Columbus. Dover Publications, New York,<br />
pp 115—188<br />
Bjorndal KA (1982) The consequences of herbivory for the life<br />
history pattern of the Caribbean green turtle, Chelonia mydas. In:<br />
Bjorndal KA (ed) Biology and conservation of sea turtles. Smithsonian<br />
Institution Press, Washington, DC, pp 111—116<br />
Bjorndal KA, Zug GR (1995) Growth and age of sea turtles (revised<br />
edn). In: Bjorndal KA (ed) Biology and conservation of<br />
sea turtles. Smithsonian Institution Press, Washington, DC,<br />
pp 599—600
Bradbury RH, Seymour RM, Antonelli PL (1992) Is the Great<br />
Barrier Reef ecologically sustainable? Proc 7th Int Coral Reef<br />
Symp 2 : 803<br />
Carr AF (1956) The windward road, Alfred Knopf, New York<br />
Chapin M (1995) Epilogue. In: Roeder H, (ed) Plants and animals<br />
in the life of the Kuna. University of Texas Press, Austin,<br />
pp 115—119<br />
Clark HL (1919) The distribution of the littoral echinoderms of the<br />
West Indies. Pap Dept Mar Biol Carnegie Inst Wash 13 : 51—73<br />
Clifton KE, Kim K, Wulff JL (in press) A field guide to the reefs of<br />
Caribbean Panama with an emphasis on western San Blas. Proc<br />
8th Int Coral Reef Symp<br />
Connell JH (1978) Diversity in tropical rain forests and coral reefs.<br />
Science 199 : 1302—1310<br />
Cortés J (1993) A reef under siltation stress: a decade of degradation.<br />
In: Ginsburg RN (comp) Proceedings of the colloquium on<br />
global aspects of coral reefs: health, hazards and history. Rosentiel<br />
School Mar Atmos Sci, University of Miami, Florida,<br />
pp 240—246<br />
Cortés J, Risk MJ (1985) A reef under siltation stress: Cahuita, Costa<br />
Rica. Bull Mar Sci 36 : 339—356<br />
Dampier W (<strong>17</strong>29) A new voyage around the world (1968 reprint).<br />
Dover, New York<br />
Dodge RE, Vaisnys JR (1977) Coral populations and growth patterns:<br />
responses to sedimentation and turbidity associated with<br />
dredging. J Mar Res 35 : 715—730<br />
Donovan SK, Gordon CM (1993) Echinoid taphonomy and the<br />
fossil record: supporting evidence from the Plio-Pleistocene of<br />
the Caribbean. Palaios 8 : 304—306<br />
Dublin HT (1995) Vegetation dynamics in the Serengeti-Mara ecosystem:<br />
the role of elephants, fire, and other factors. In: Sinclair<br />
ARE, Arcese P (eds) Serengeti II: dynamics, management, and<br />
conservation of an ecosystem. University of Chicago Press,<br />
Chicago, pp 71—90<br />
Duerden JE (1901) The marine resources of the British West Indies.<br />
West Indian Bull: J Imperial Agric Dept West Indies<br />
1901 : 121—141<br />
Elton C (1927) Animal ecology. Macmillan, New York<br />
Field GW (1891) Notes on the echinoderms of Kingston Harbor,<br />
Jamaica, W. I. The Johns Hopkins Univ Circ 2 : 83—84<br />
Gardner WJ (1909) A history of Jamaica. T. Fisher Unwin, London<br />
Glynn PW (1973) Aspects of the ecology of reefs in the western<br />
Atlantic region. In: Jones OA, Endean R (eds) Biology and<br />
geology of coral reefs, vol 2. Academic Press, New York,<br />
pp 271—324<br />
Gordon CM, Donovan SK (1992) Disarticulated echinoid ossicles in<br />
paleoecology and taphonomy: the last interglacial Falmouth<br />
Formation of Jamaica. Palaios 7 : 157—166<br />
Goreau TF (1959) The ecology of Jamaican coral reefs. I. Species<br />
composition and zonation. Ecology 40 : 67—90<br />
Guzmán HM, Weil E, Jackson JBC (1991) Short-term ecological<br />
consequences of a major oil spill on Panamanian subtidal reef<br />
corals. Coral Reefs 10 : 1—12<br />
Guzmán HM, Burns KA, Jackson JBC (1994) Injury, regeneration<br />
and growth of Caribbean reef corals after a major oil spill in<br />
Panama. Mar Ecol Prog Ser 105 : 231—241<br />
Hay ME (1984) Patterns of fish and urchin grazing on Caribbean<br />
coral reefs: are previous results typical? Ecology 65 : 446—454<br />
Hay ME (1991) Marine-terrestrial contrasts in the ecology of plant<br />
chemical defenses against herbivores. Trends Ecol Evol<br />
6 : 362—365<br />
Hay ME (in press) Seaweeds and the ecology and evolution of coral<br />
reefs. Coral Reefs<br />
Heinsohn GE, Wake J, Marsh H, Spain AV (1977) The dugong<br />
(Dugong dugon (Muller)). Biol Conserv 6 : 143—152<br />
Henderson JB (1914) The cruise of the Tomas Bárrera. GP Putnam’s<br />
Sons, New York<br />
Hughes TP (1989) Community structure and diversity of coral reefs:<br />
the role of history. Ecology 70 : 275—279<br />
Hughes TP (1994) Catastrophes, phase shifts, and large-scale degradation<br />
of a Caribbean coral reef. Science 265 : 1547—1551<br />
S31<br />
Hutchinson GE (1978) An introduction to population ecology. Yale<br />
University Press, New Haven, Connecticut<br />
Jackson JBC (1982) Biological determinants of present and past<br />
sessile animal distributions. In: Tevesz MJS, McCall PL (eds)<br />
Biotic interactions in recent and fossil benthic communities.<br />
Plenum, New York, pp 39—120<br />
Jackson JBC (1991) Adaptation and diversity of reef corals. Bio-<br />
Science 41 : 475—482<br />
Jackson JBC (1992) Pleistocene perspectives on coral reef community<br />
structure. Am Zool 32 : 719—731<br />
Jackson JBC (1995) The role of science in coral reef conservation<br />
and management. In: Partnership building and framework development.<br />
Final report, the <strong>International</strong> Coral Reef Initiative<br />
Workshop, Dumaguete City, Philippines, pp 5—9<br />
Jackson JBC, Cubit JD, Keller BD, Batista V, Burns K, Caffey HM,<br />
Caldwell RL, Garrity SD, Getter CD, Gonzalez C, Guzmán HM,<br />
Kaufmann KW, Knap AH, Levings SC, Marshall MJ, Steger R,<br />
Thompson RC, Weil E (1989) Ecological effects of a major oil<br />
spill on Panamanian coastal marine communities. Science<br />
243 : 37—44<br />
Jackson JBC, Budd AF, Pandolfi JM (1996) The shifting balance of<br />
natural communities? In: Jablouski D, Erwin DH, Lipps, JH<br />
(eds) Evolutionary paleobiology, Universiy of Chicago Press,<br />
Chicago, pp 89—122<br />
Janzen DH, Martin PS (1982) Neotropical anachronisms: the fruits<br />
the gomphotheres ate. Science 215 : 19—27<br />
Janzen DH, Wilson DE (1983) Mammals: introduction. In: Janzen<br />
DH (ed) Costa Rican natural history, University of Chicago<br />
Press, Chicago, pp 426—442<br />
King FW (1982) Historical review of the decline of the green turtle<br />
and the hawksbill. In: Bjorndahl KA (ed) Biology and conservation<br />
of sea turtles. Smithsonian Institution Press, Washington,<br />
DC, pp 183—188<br />
Knowlton N (1992) Thresholds and multiple stable states in coral<br />
reef community dynamics. Am Zool 32 : 674—682<br />
Knowlton N, Lang JC, Keller BD (1990) Case study of natural<br />
population collapse: post-hurricane predation on Jamaican<br />
staghorn corals. Smithson Contrib Mar Sci 31 : 1—25<br />
Lanyon JM, Limpus CJ, Marsh H (1989) Dugongs and turtles:<br />
grazers in the seagrass system. In: Larkum AWD, McComb AJ,<br />
Sheperd SA (eds) Biology of seagrasses: a treatise on the biology<br />
of seagrasses with special reference to the Australian region,<br />
Elsevier, Amsterdam, pp 610—634<br />
Lessios HA (1988) Mass mortality of Diadema antillarum in the<br />
Caribbean: what have we learned? Ann Rev Ecol Syst 19 : 371—393<br />
Lessios HA, Robertson DR, Cubit JD (1984) Spread of Diadema<br />
mass mortality through the Caribbean. Science 226 : 335—337<br />
Levitan DR (1991) Skeletal changes in the test and jaws of the sea<br />
urchin Diadema antillarum in response to food limitation. Mar<br />
Biol 111 : 431—435<br />
Levitan DR (1992) Community structure in times past: influence of<br />
human fishing pressure on algal-urchin interactions. Ecology<br />
73 : 1597—1605<br />
Lewis BC (1940) The Cayman Islands and marine turtle. Bull Inst<br />
Jamaica, Sci Ser 2 : 56—65<br />
Lewis JB (1960) The coral reefs and coral communities of Barbados,<br />
W. I. Can J Zool 38 : 1133—1145<br />
Lewis JB (1984) The Acropora inheritance: a reinterpretation of the<br />
development of fringing reefs in Barbados, West Indies. Coral<br />
Reefs 3 : 1<strong>17</strong>—122<br />
Lewsey CD (1978) Assessing the environmental effects of tourism on<br />
the carrying capacity of small island systems ‘‘The case for<br />
Barbados’’. ‘‘Unpubl. Ph.D Diss, Cornell University, Univ<br />
Microfilms Int, Ann Arbor, Michigan<br />
Liddell WD, Ohlhorst SL (1988) Comparison of western Atlantic<br />
coral reef communities. Proc 6th Int Coral Reef Symp 3 : 281—286<br />
Limpus CJ (1995) Global overview of the status of marine turtles:<br />
a 1995 viewpoint. In: Bjorndal KA (ed), Biology and conservation<br />
of sea turtles, 2nd edn. Smithsonian Institution Press, Washington<br />
DC, pp 605—609
S32<br />
Long E (<strong>17</strong>74) The history of Jamaica, or general survey of the<br />
antient and modern state of that island: with reflections on its<br />
situations, settlements, inhabitants, climate, products, commerce,<br />
laws and government (1970 Reprint). Frank Cass and Co,<br />
London<br />
Lugo A, Schmidt R, Brown S (1981) Tropical forests in the Caribbean.<br />
Ambio 10 : 318—324<br />
Macintyre IG (1968) Preliminary mapping of the insular shelf off the<br />
west coast of Barbados. Carib J Sci 8 : 95—100<br />
Matthiessen P (1975) Far Tortuga. Random House, New York<br />
Matthiessen P (1986) Men’s lives: the surfmen and the baymen of the<br />
South Fork. Random House, New York<br />
McClanahan TR, Muthiga NA (1988) Changes in Kenyan coral reef<br />
community structure and function due to exploitation.<br />
Hydrobiologia 166 : 269—276<br />
McClanahan TR, Kamukuru AT, Muthiga NA, Gilagabher Yebio<br />
M, Obura D (1996) Effect of sea urchin reductions on algae,<br />
coral, and fish populations. Conserv Biol 10 : 136—154<br />
McNaughton SJ, Banyikwa FF (1995) Plant communities and herbivory.<br />
In: Sinclair ARE, Arcese P (eds) Serengeti II: dynamics,<br />
management, and conservation of an ecosystem. University of<br />
Chicago Press, Chicago, pp 49—70<br />
Mesolella KJ (1957) Zonation of uplifted Pleistocene coral reefs on<br />
Barbados, West Indies. Science 156 : 638—640<br />
Meylan A (1985) Nutritional characteristics of sponges in the diet of<br />
the hawksbill turtle, Eretmochelys imbricata 3rd Int Sponge Conf:<br />
472—477<br />
Meylan A (1988) Spongivory in hawksbill turtles: a diet of glass.<br />
Science 239 : 393—395<br />
Munroe JL (ed) (1983) Caribbean coral reef fisheries, 2nd ed.<br />
ICLARM Stud Rev 7 : 1—276<br />
National Research Council (1995) Understanding marine biological<br />
diversity. National Academy Press, Washington, DC<br />
National Geographic Society (1981) Geographic atlas of the world,<br />
5th edn. Nat Geogr Soc, Washington, DC<br />
Neitschman B (1973) Between land and water: the subsistence ecology<br />
of the Moskito Indians, eastern Nicaragua. Seminar Press,<br />
New York<br />
Neitschman B (1982) The cultural context of sea turtle subsistence<br />
hunting in the Caribbean and problems caused by commercial<br />
exploitation. In: Bjorndal KA (ed), Biology and conservation<br />
of sea turtles. Smithsonian Institution Press, Washington, DC,<br />
pp 439—445<br />
Nutting CC (1919) Barbados-Antigua expedition. Univ Iowa Studies<br />
Nat Hist 8 : 1—274<br />
Ogden JC (1980) Faunal relationships in Caribbean seagrass beds.<br />
In: Phillips RC, McRoy CP (eds), Handbook of seagrass biology:<br />
an ecosystem perspective, Garland STPM Press, New York,<br />
pp <strong>17</strong>3—198<br />
Ogden JC, Robinson L, Whitlock H, Daganhart H, Cebula R (1983)<br />
Diel foraging patterns in juvenile green turtles (Chelonia mydas<br />
L) in St. Croix, United States, Virgin Islands. J Exp Mar Biol<br />
Ecol 66 : 199—205<br />
Ogden JC, Ogden NB (1993) The coral reefs of the San Blas Islands:<br />
revisited after 20 years. In: Ginsburg RN (comp) Proceedings of<br />
the colloquium on global aspects of coral reefs: health, hazards,<br />
and history. Rosentiel School Mar Atmos Sci, University of<br />
Miami, Florida, pp 267—272<br />
Pauly D (1995) Anecdotes and the shifting baseline syndrome of<br />
fisheries. Trends Ecol Evol 10 : 430<br />
Porter JW (1972) Ecology and species diversity of coral reefs on<br />
opposite sides of the Isthmus of Panamá. Bull Biol Soc Wash<br />
2:89—116<br />
Randall JE (1965) Grazing effect on sea grasses by herbivorous reef<br />
fishes in the West Indies. Ecology 46 : 255—260<br />
Randall JE, Schroeder RE, Starck WA, Jr (1961) Notes on the<br />
biology of Diadema antillarum. Carib J Sci 4 : 421—433<br />
Randall JE, Hartman WD (1968) Sponge-feeding fishes of the West<br />
Indies. Mar Biol 1 : 216—225<br />
Roberts L (1989) Disease and death in the New World. Science<br />
246 : 1245—1247<br />
Rodriguez A (1981) Marine and coastal environmental stress in the<br />
wider Caribbean region. Ambio 10 : 283—294<br />
Rogers CS (1985) Degradation of Caribbean and western Atlantic<br />
coral reefs and decline of associated fisheries. Proc 5th Int Coral<br />
Reef Congr 6 : 491—496<br />
Rogers CS (1990) Responses of coral reefs and reef organisms to<br />
sedimentation. Mar Ecol Prog Ser 62 : 185—202<br />
Rouse I (1992) The Tainos. Yale University Press, New Haven<br />
Sauer CO (1966) The early Spanish Main. University of California<br />
Press, Berkeley, California<br />
Sheppard C (1995) The shifting baseline syndrome. Mar Poll Bull<br />
30 : 766—767<br />
Sheppard C, Price A, Roberts C (1992) Marine ecology of the<br />
Arabian region. Academic Press, London<br />
Sinclair ARE (1995a) Serengeti past and present. In: Sinclair ARE,<br />
Arcese P (eds) Serengeti II: dynamics, management, and conservation<br />
of an ecosystem. University of Chicago Press, Chicago,<br />
pp 3—30<br />
Sinclair ARE (1995b) Equilibria in plant-herbivore interactions. In:<br />
Sinclair ARE, Arcese P (eds) Serengeti II: dynamics, management,<br />
and conservation of an ecosystem. University of Chicago<br />
Press, Chicago, pp 91—113<br />
Sinclair ARE, Arcese P (1995) Serengeti II: dynamics, management,<br />
and conservation of an ecosystem. University of Chicago Press,<br />
Chicago<br />
Sloane H (<strong>17</strong>07—<strong>17</strong>25) A voyage to the islands Madera, Barbadoes,<br />
Nieves, St Christophers, and Jamaica; with the natural history of<br />
the herbs and trees, four-footed beasts, fishes, birds, insects,<br />
reptiles, &c. of the last of those islands. In two volumes. Printed<br />
for the author, London<br />
Stearn CW, Scoffin TP, Martindale W (1977) Calcium, carbonate<br />
budget of a fringing reef on the west coast of Barbados.<br />
I. Zonation and productivity. Bull Mar Sci 27 : 479—510<br />
Thayer GW, Engel DW, Bjorndahl KA (1982) Evidence for shortcircuiting<br />
of the detritus cycle of seagrass beds by the green turtle,<br />
Chelonia mydas. J Exp Mar Biol Ecol 62 : <strong>17</strong>3—183<br />
Thayer GW, Bjorndahl KA, Ogden JC, Williams SL, Zieman JC<br />
(1984) Role of larger herbivores in seagrass communities. Estuaries<br />
7 : 351—376<br />
Thompson EF (1945) The fisheries of Jamaica. Development and,<br />
welfare in the West Indies, Bulletin 18, Bridgetown, Barbados<br />
Tomascik T, Sander F (1985) Effects of eutrophication on reefbuilding<br />
corals. I. Growth rate of the reef building coral Montastrea<br />
annularis. Mar Biol 87 : 143—155<br />
Tomascik T, Sander F (1987a) Effects of eutrophication on reefbuilding<br />
corals. II. Structure of scleractinian coral communities<br />
on fringing reefs, Barbados, West Indies. Mar Biol 944 : 3—75<br />
Tomascik T, Sander F (1987b) Effects of eutrophication on reefbuilding<br />
corals. III. Reproduction of the reef-building coral<br />
Porites porites. Mar Biol 94 : 77—94<br />
van Dam RP, Diez CE (in press) Predation by hawksbill turtles on<br />
sponges at Mona Island, Puerto Rico. Proc 8th Int Coral Reef<br />
Symp<br />
Ventocilla J, Herrera H, Nu´ nez V (1995) Plants and animals in the<br />
life of the Kuna, Roeder H (ed) University of Texas Press, Austin,<br />
150 pp<br />
Wilkinson CR (1992) Coral reefs of the world are facing widespread<br />
devastation: can we prevent this through sustainable management<br />
practices. Proc 7th Int Coral Reef Symp 1 : 11—21<br />
Woodley JD, Chornesky EA, Clifford PA, Jackson JBC, Kaufman<br />
LS, Knowlton N, Lang JC, Pearson MP, Porter JW, Rooney<br />
MC, Rylaarsdam KW, Tunnicliffe VJ, Wahle CM, Wulff JL,<br />
Curtis ASG, Dallmeyer MD, Jupp BJ, Koehl MAR, Neigel J,<br />
Sides EM (1981) Hurricane Allen’s impact on Jamaican coral<br />
reefs. Science 214 : 749—755<br />
Young T (1847) Narrative of a residence on the Mosquito shore:<br />
with an account of Truxillo, and the adjacent islands of Bonacca<br />
and Roatan; and a vocabulary of the Mosquitian language.<br />
Smith, Elder, London (1971 reprint). Krause Reprint, New York<br />
Ziemann (1982) The ecology of seagrasses of south Florida: a community<br />
profile. US Fish Wildlife Serv Prog FWS/OBS-82/25 124 : 1—26
Social Interactions in Captive Female<br />
Florida Manatees<br />
Jennifer Young Harper and Bruce A. Schulte n<br />
Zoo Biology 24:135<strong>–</strong>144 (2005)<br />
Department of Biology, Georgia Southern University, Statesboro, Georgia<br />
The Florida manatee (Trichechus manatus latirostris) is considered a semi-social<br />
species with strong bonds developed primarily between mother and offspring.<br />
Some field studies suggest sociality may be more developed and such social<br />
relationships may facilitate survival. Seven facilities in Florida house manatees,<br />
many of which were brought into captivity because of injury or illness sustained in<br />
the wild. Decisions to release such manatees consider individual history and<br />
health. We examined social interactions in adult female captive manatees to assess<br />
level of association and implications for manatee care and release. We<br />
investigated the degree of contact among 20 manatees in captivity at four<br />
facilities housing two to nine adult female manatees. We used all contact behavior<br />
occurrences sampling and continuous recording for 180 continuous minutes per<br />
day over 3 consecutive days at each facility. Virtually all contacts were nonaggressive.<br />
The number of contacts between manatees increased as the number of<br />
manatees per unit volume of water increased. Contacts did not fit a Poisson<br />
distribution, however, and were not random. When more than two manatees were<br />
present, manatees only associated with a subset of individuals in the aquarium.<br />
Relationships maintained in captivity indicate the potentially social nature of<br />
manatees, and suggest that further research is needed to examine the benefit of<br />
these relationships to the health and rehabilitation of manatees in captivity and<br />
conservation in the wild. Zoo Biol 24:135<strong>–</strong>144, 2005. c 2005 Wiley-Liss, Inc.<br />
Key words: aggression; association; conservation; ethogram; interaction; social behavior<br />
INTRODUCTION<br />
The organization of individuals into social groups is likely to evolve when the<br />
benefits of group living outweigh the costs [Alexander, 1974; Emlen 1997]. Benefits of<br />
Dr. Jennifer Young Harper’s present address is Coastal Georgia Community College, 3700 Altama<br />
Avenue, Brunswick, GA 31520.<br />
n<br />
Correspondence to: Bruce A. Schulte, Department of Biology PO Box 8042, Georgia Southern<br />
University, Statesboro, Georgia, 30460-8042. E-mail: bschulte@georgiasouthern.edu<br />
Received 2 April 2004; Accepted 8 November 2004<br />
DOI 10.1002/zoo.20044<br />
Published online in Wiley InterScience (www.interscience.wiley.com).<br />
c 2005 Wiley-Liss, Inc.
136 Harper and Schulte<br />
group living include greater access to resources and protection. Group living has<br />
ecological costs including an increase in competition for food, space, and mates and<br />
exposure to parasites, pathogens, and predators. Conflicts and aggression can<br />
become heightened in crowded areas, especially when there is competition over a<br />
limited resource. The resolution of conflict can occur when individuals leave a<br />
particular situation, defend territories, form small closely associated groups, or<br />
establish dominance hierarchies built upon aggression or affiliation [Maher and<br />
Lott, 1995, 2000; Pusey and Packer, 1999].<br />
Florida manatees (Trichechus manatus latirostris), a subspecies of the West<br />
Indian manatee, are considered semi-social or asocial animals, because they<br />
generally do not form organized social structures. Manatees are usually seen<br />
traveling in warmer coastal waters alone or in mother<strong>–</strong>calf pairs, with such bonds<br />
lasting until the calf is approximately 1<strong>–</strong>2 years old [Reynolds and Odell, 1991].<br />
From 1993<strong>–</strong>1996, Koelsch [1997] observed manatees in Sarasota Bay, Florida and<br />
noted some strong bonds among adult manatees. In addition, associations were<br />
stronger within than between the sexes. These non-random association patterns<br />
along with the social facilitation noted by Reynolds [1981] suggest that manatees<br />
may be more social than previously thought [Koelsch, 1997]. The current study<br />
investigated the type and extent of interactions among female manatees held in<br />
captivity to evaluate the strength of social bonds.<br />
Manatees in the wild face many problems in their aquatic environment<br />
including ingestion of foreign materials [Beck and Barros, 1991], exposure to toxins<br />
such as red tide [Bossart et al., 1998], and collisions with boats [Wright et al., 1995].<br />
Injured or sick manatees are taken into captivity for rehabilitation. Once an injured<br />
animal is out of critical care in a captive facility, it is usually housed with several<br />
other manatees of the same sex. We know relatively little, however, about how these<br />
supposed semi-social animals behave in the constant presence of multiple<br />
conspecifics. The specific objective of this research was to determine if manatees in<br />
captivity exhibit behaviors that are either aggressive, serve to reduce aggression or<br />
reflect the establishment of particular associations. Manatees were observed before,<br />
during, and after feeding periods because conflicts might arise in anticipation of, or<br />
during consumption of, food (a potentially limited resource). We examined two<br />
hypotheses on level of interaction: (1) manatees will rarely interact; and (2) manatees<br />
will interact more under higher density situations (i.e., more manatees per unit<br />
volume of aquarium space). In addition, we evaluated two hypotheses on the timing<br />
and type of interactions. Specifically, (3) most interactions will be aggressive and<br />
occur just before feeding or when food is becoming depleted, and (4) if manatees are<br />
more social than previously thought, most interactions will be non-aggressive and<br />
occur outside of feeding periods.<br />
MATERIALS AND METHODS<br />
Study Sites<br />
This research was conducted from January<strong>–</strong>March 2001 at four zoos and<br />
aquariums housing captive manatees, specifically Homosassa Springs Wildlife State<br />
Park (HSWSP, Homosassa, Florida), Lowry Park Zoo (Tampa, Florida), Sea World<br />
of Florida (Orlando, Florida), and Walt Disney World’s The Living Seas in Epcot
Captive Female Manatee Social Behavior 137<br />
(Orlando, Florida). Generally, manatees were housed in human-made enclosures<br />
approximately 3 m in depth and oval in shape. The exception was HSWSP, which is<br />
a naturally occurring spring that covers approximately 0.2 ha and reaches depths of<br />
13.5 m. Manatees (n ¼ 20) at all facilities were viewed from an underwater<br />
observation area or from above the enclosure. We estimated the volume of water<br />
for HSWSP from the dimensions of the exhibit. The other three facilities provided us<br />
with aquarium volumes. Water volumes at each facility are as follows: HSWSP (1.1<br />
10 7 L [9 manatees]), Sea World (1.4 10 6 L [6 manatees]), Epcot (760,000 L<br />
[2 manatees]), and Lowry Park Zoo (470,000 L [3 manatees]).<br />
Behavioral Observations<br />
Manatees were watched continuously for 180 min over 3 days around<br />
feeding times. The time duration was selected because there were constraints<br />
at the facilities that prevented longer observation periods including operating<br />
hours, interference in observations when visitor numbers were high, and<br />
logistical issues. Observations were made at the four facilities with 20 individuals<br />
for a total of 36 hr. No adult males were observed for this study, but three<br />
juvenile male manatees were observed. One male was at Lowry Park Zoo<br />
and two were at Sea World. All three males were with females. During the<br />
3 hr of continuous observations for all interactions, we recorded all aggressive<br />
and non-aggressive contacts between manatees (sender<strong>–</strong>receiver) and spatial<br />
displacement by manatees [Martin and Bateson, 1993]. Aggressive interactions were<br />
defined as contacts in which the sender’s head touched the receiver with some<br />
momentum or caused physical movement of the receiver. Displacement was occurred<br />
when one manatee (sender) caused another manatee (receiver) to move from its<br />
location without contact because of the presence or movement of the sender.<br />
Displacements were rare and not considered in the analysis. Other aggressive<br />
interactions occurred when the sender’s tail hit the receiver with force as it swam<br />
away. Non-aggressive interactions that might be affiliative in nature included all<br />
other contacts (head to different body parts, body to different body parts, fin to<br />
different body parts, and tail to different body parts without force). Incidental<br />
contacts transpired when an animal brushed against another animal. These contacts<br />
were not included in the analysis.<br />
Manatees at all facilities were fed 25<strong>–</strong>100 heads of romaine lettuce<br />
depending on the number of manatees in the aquarium (one head weighs<br />
approximately 488 grams; V. Burke, head keeper at Lowry Park Zoo,<br />
personal communication). Feeding occurred three to four times a day at regular<br />
intervals. Three observational periods were defined as pre-feeding, during-feeding,<br />
and post-feeding. The pre-feeding period occurred during the first 60 min of<br />
observations before the handlers provisioned any food. The feeding period occurred<br />
when the manatees were feeding and post-feeding was defined as when 10% or less of<br />
the food remained (heads and leaves combined). The number of heads of lettuce<br />
(or leaf equivalents) that remained in the aquarium were counted and recorded every<br />
5 min. Feeding periods lasted approximately 1 hr. Supplemental foods such as<br />
carrots, kale, and apples were sometimes provided to manatees by the aquarium<br />
staff, but not included in the determination of feeding or non-feeding observational<br />
periods.
138 Harper and Schulte<br />
Data Analysis<br />
The total number of contacts for each day and facility was counted. The<br />
proportion of aggressive or non-aggressive interactions by manatees at each facility<br />
was calculated by observational period over the 3 consecutive days of observations.<br />
Similar proportions also were calculated for individual manatees as sender and as<br />
receiver. Sender-receiver matrices were created for each facility. Non-aggressive<br />
contacts were used to construct sender-receiver matrices because of the low<br />
occurrence of aggressive contacts. Animals were ordered in the matrices based on the<br />
total number of contacts sent. Log-linear G-test analyses were used to examine if<br />
non-aggressive contacts were more likely before or after feeding at separate facilities<br />
and to compare the distribution among facilities of non-aggressive contacts before<br />
and after feeding. Linear regression was used to determine the effect of density on<br />
total number of contacts. We calculated the coefficient of dispersion (CD) using the<br />
number of contacts between unique pairs of manatees to determine if contacts fit a<br />
Poisson distribution, where values greater than one show a clumped distribution,<br />
whereas values less than one indicate repulsion. For the two facilities (HSWSP and<br />
SeaWorld) with greater than three manatees, we analyzed the distribution of these<br />
contacts using a G-test with the Williams correction factor [Sokal and Rohlf, 1995].<br />
The distribution is determined by the frequency of manatee pairs with zero to the<br />
maximum number of observed contacts. To maintain frequencies of at least four<br />
occurrences per cell (i.e., contacts/manatee pair) some categories were combined. In<br />
the two analyses, we used three categories. Because the parameter mean in each case<br />
was estimated from the sample, two degrees of freedom were removed, yielding a<br />
single degree of freedom for the G-statistic [Sokal and Rohlf, 1995]. Significant<br />
variation from a Poisson distribution indicated a non-random pattern of contacts.<br />
Analyses were carried out using JMP, version 3.02 (SAS Institute Inc., Cary, NC),<br />
and by methods described by Sokal and Rohlf [1995].<br />
RESULTS<br />
Manatees interacted regularly during the non-feeding periods of our study. We<br />
observed no contacts between manatees while they were feeding. Of the 218 nonaggressive<br />
encounters, 98 (45%) occurred in the pre-feeding period and 120 (55%) in<br />
the post-feeding period. The pre-feeding period was always 60 min long whereas the<br />
post period ranged from 30<strong>–</strong>75 min (55.773.62 min, mean7SE). The number of<br />
non-aggressive contacts did not differ in the pre- vs. the post-feeding observational<br />
period within each facility (all Go1.1, df ¼ 1, P40.05). The distribution of nonaggressive<br />
contacts also did not differ in the pre- and post-feeding periods among the<br />
four facilities (G ¼ 2.02, df ¼ 3, P40.05). The number of total contacts per day<br />
ranged from 10<strong>–</strong>34 at each of the four facilities (means7SE of contacts/hr: Day<br />
1 ¼ 6.7572.11, Day 2 ¼ 6.371.48, Day 3 ¼ 5.970.89).<br />
Of the 228 contacts recorded, only 10 (4.39%) were aggressive. The manatees<br />
at Lowry Park Zoo accounted for nine of 10 aggressive encounters and seven were<br />
observed during a single 180-min observation session. These contacts were head-tobody<br />
contacts mostly by the adult females to a 9-month-old orphaned male. Using<br />
sender-receiver matrices based upon non-aggressive contacts, we discuss manatee<br />
interactions at each facility.
Captive Female Manatee Social Behavior 139<br />
Epcot: The Living Seas<br />
Walt Disney World’s The Living Seas in Epcot housed two female manatees,<br />
Lydia and Mariah. Forty-three non-aggressive contacts (4.871.3 contacts/hr)<br />
occurred between these two manatees. Lydia initiated 20 contacts and Mariah<br />
initiated 23.<br />
Homosassa Springs Wildlife State Park<br />
Homosassa Springs Wildlife State Park housed nine female manatees:<br />
Amanda, Ariel, Betsy, Electra, Holly, Lorelei, Oakley, Rosie, and Willoughby.<br />
Thirty-one non-aggressive contacts (3.470.1 contacts/hr) occurred between the<br />
manatees. Eighteen contacts were made between three related individuals (Amanda,<br />
Ariel and Betsy) termed Group one, consisting of a mother and her two adult<br />
daughters. The remaining six manatees made 12 contacts to each other with only one<br />
contact made between the two groups of manatees (Table 1).<br />
Lowry Park Zoo<br />
Lowry Park Zoo housed two female manatees, Cinco and BB, and one<br />
orphaned male calf, Lowry. Eighty-three contacts occurred between these manatees;<br />
nine were aggressive (10.8%, 1.070.67 contacts/hr) and 74 were non-aggressive<br />
(89.2%, 8.270.9 contacts/hr). BB initiated five aggressive contacts to Lowry. Both<br />
Cinco and Lowry made two aggressive contacts (22.2%) toward each other. Lowry<br />
initiated the most non-aggressive contacts (38). BB and Cinco initiated 20 and 16<br />
non-aggressive contacts, respectively. For the adult females, BB initiated contact<br />
with Cinco eight times and Cinco contacted BB nine times.<br />
Sea World<br />
Sea World of Florida housed four adult female manatees: Charlotte, Rita,<br />
Sara, and Stubbie and two orphaned male calves: Brooks and Pistachio. Of the 71<br />
interactions that occurred, 70 (98.6%, 7.871.0 contacts/hr) were non-aggressive and<br />
one (1.4%, 0.<strong>17</strong>0.1 contact/hr) was aggressive. The juvenile males initiated more<br />
contacts than the adult females (52 of 71 or 73%). Brooks made 34 contacts with<br />
most directed toward Charlotte (15) and Sara (13). These were primarily play<br />
mounting contacts with the two females. Pistachio initiated four similar contacts<br />
TABLE 1. Matrix constructed from non-aggressive contacts at Homosassa Springs Wildlife<br />
State Park with nine adult females<br />
Amanda a Rosie Betsy a Ariel a Holly Will Oakley Lorelei Electra Total<br />
Amanda — 0 2 7 0 0 0 1 0 10<br />
Rosie 0 — 0 0 1 1 2 0 3 7<br />
Betsy 3 0 — 2 0 0 0 0 0 5<br />
Ariel 2 0 2 — 0 0 0 0 0 4<br />
Holly 0 0 0 0 — 1 0 0 1 2<br />
Will 0 0 0 0 0 — 1 0 0 1<br />
Oakley 0 0 0 0 0 1 — 0 0 1<br />
Lorelei 0 0 0 0 0 1 0 — 0 1<br />
Electra 0 0 0 0 0 0 0 0 — 0<br />
a Amanda, Betsy and Ariel are related as mother and two adult daughters.
140 Harper and Schulte<br />
each to Sara, Rita and Stubbie. Excluding the juvenile males from total contacts,<br />
Charlotte and Sara accounted for 11 of the 13 encounters (Table 2).<br />
All Facilities<br />
All manatees, with the exception of one, were involved in some type of<br />
interaction with other individuals. Density of manatees per unit volume explained<br />
the majority of variation in contact rates (Fig. 1; r 2 ¼ 0.96). The coefficient of<br />
TABLE 2. Matrix constructed from non-aggressive manatee contacts at Sea World of Florida<br />
Brooks a<br />
Pistachio b<br />
Charlotte c<br />
Sara c<br />
Rita c<br />
Stubbie c<br />
Total<br />
Brooks — 3 15 13 2 2 35<br />
Pistachio 3 — 1 4 4 4 16<br />
Charlotte 1 0 — 8 0 0 9<br />
Sara 2 0 3 — 0 0 5<br />
Rita 0 0 1 1 — 0 2<br />
Stubbie 1 0 0 0 0 — 1<br />
a Juvenile manatee (2 years old).<br />
b Juvenile manatee (1.5 years old).<br />
c Adult female manatee.<br />
Total Number of Contacts<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
HSSP<br />
Epcot<br />
SeaWorld<br />
y = 9.9482x + 21.887<br />
R 2 = 0.96<br />
0 1 2 3 4 5 6<br />
Number of Manatees per 1,000,000 L water<br />
Fig. 1. Number of contacts for each facility (Homosassa Springs Wildlife State Park, Walt<br />
Disney World’s The Living Seas at Epcot, Sea World of Florida, and Lowry Park Zoo)<br />
initiated by manatees per million L water within each aquarium.<br />
LPZ<br />
7
Captive Female Manatee Social Behavior 141<br />
dispersion (CD) at the three facilities with at least three manatees was always greater<br />
than one, indicating clumping of the distribution (HSWSP ¼ 3.9, SeaWorld ¼ 6.2,<br />
Lowry Park Zoo ¼ 1.9). The distribution of contacts was significantly different from<br />
a Poisson distribution for HSWSP (Gadj ¼ 9.24w 2 0.05[1] ¼ 3.84) and for SeaWorld<br />
(Gadj ¼ 13.74w 2 0.05[1] ¼ 3.84). We could not perform the analysis for Lowry Park<br />
Zoo because of the low number of manatees, resulting in single frequencies per cell.<br />
DISCUSSION<br />
Manatees in captivity associated and interacted with certain individuals. The<br />
occurrence of regular, non-aggressive contacts by manatees in captivity and the<br />
relative rarity of aggressive contacts support the contention that manatees may be<br />
more social than previously considered [Koelsch, 1997]. Hence, we reject our first<br />
hypothesis that manatees will rarely interact. The number of non-aggressive contacts<br />
increased with density (Fig. 1), supporting our second hypothesis. Aggression was<br />
rare and only occurred at the facilities with the greatest densities (Lowry Park Zoo<br />
and SeaWorld) but these facilities had one or two juvenile males in with the females.<br />
Of the ten total aggressive encounters observed, nine were recorded at Lowry Park<br />
Zoo. These aggressive contacts were directed mainly toward the 9-month-old male<br />
Lowry by the adult females, possibly to prevent the formation of a mother<strong>–</strong>calf<br />
bond. Lowry was born in captivity but his mother Ionia died from her injuries<br />
during rehabilitation when Lowry was introduced to the two resident adult females,<br />
BB and Cinco. Because aggression was rare and not related to feeding, we reject our<br />
third hypothesis and show support for our fourth hypothesis that most interactions<br />
will be non-aggressive and occur outside of feeding periods. A concurrent study<br />
showed that time in captivity or differences in facilities did not affect manatee<br />
activity patterns [Young, 2001].<br />
The density-dependent nature of non-aggressive interactions suggests that<br />
contacts were random and not selective. Interactions were not equally distributed,<br />
however, among manatees at least at the two facilities with more than three<br />
manatees. At HSWSP, greater association was apparent within two social<br />
subgroups. One subgroup consisted of Amanda and her two adult daughters, Ariel<br />
and Betsy. This observation in captivity fits nicely with field studies, where mother<br />
and offspring form the most apparent long-term bonds during calf dependency in the<br />
wild [Hartman, 1979; Bengtson, 1981; Reynolds, 1981]. All three animals were adults<br />
(Betsy is the youngest, born in captivity in 1990). This long-term close association<br />
may be common in the wild or an artifact of captivity. These three individuals<br />
contacted the other subgroup composed of six unrelated manatees only once. In<br />
contrast, this subgroup exhibited 12 contacts within the group and no contacts to<br />
Amanda, Ariel, or Betsy. Random interactions would predict more contacts between<br />
the two subgroups and the analysis supported a clumped distribution of contacts.<br />
This segregation of manatees is especially interesting at HSWSP, which represented<br />
the site with the lowest density of manatees.<br />
At Sea World of Florida, Charlotte and Sara associated regularly with Rita on<br />
the periphery of this pair. The two young males (Brooks and Pistachio) initiated a<br />
majority of the contacts, mostly as play mounting contacts with the females. Between<br />
the females, the overall distribution of contacts was clumped, not random. At Lowry<br />
Park Zoo, the two adult females, BB and Cinco initiated non-aggressive contacts to
142 Harper and Schulte<br />
each other in nearly equal number, while never initiating aggressive contacts to one<br />
another. This contrasted with the relationship to the orphaned, nine-month-old<br />
male, who received aggressive and non-aggressive contacts. The two adult females at<br />
Epcot showed a similarly reciprocal level of non-aggressive contact initiation as<br />
those at Lowry Park Zoo. Such equal initiation of contacts suggests an absence of<br />
dominance but with only a few animals, statistical analysis is not appropriate.<br />
In wild manatees, affiliative social interactions have been observed between<br />
adult females [Koelsch, 1997]. Mothers and calves associate regularly [Hartman,<br />
1979; Reynolds and Odell, 1991; Koelsch, 1997], but it is uncertain if this<br />
relationship lasts until adulthood as exhibited by the captive manatees in subgroup<br />
one at HSWSP. Elephants are the closest living relative of the Sirenia, and they form<br />
close mother<strong>–</strong>calf associations that last into adulthood for females [Buss and Smith,<br />
1966; Eisenberg et al., 1971; Douglas-Hamilton, 1972; Moss 1976; Dublin, 1983].<br />
The calf learns basic life skills from its mother and social group members [Lee and<br />
Moss, 1999], and young males have even been observed play mounting with their<br />
mothers or other females within their group [Gadgil and Nair, 1983]. Similar play<br />
behavior toward adult female manatees was observed with juvenile males at Sea<br />
World. In the wild, juveniles seem to be submissive to adult manatees [Hartman,<br />
1979], perhaps explaining the prevalence of aggressive contacts initiated by the adult<br />
females toward the juvenile male, Lowry. The types of associations observed with<br />
captive manatees are similar to those described by other mammals in captivity<br />
including elephants [Schulte, 2000], zebras [Schilder, 1992], captive spotted hyenas<br />
[Glickman et al., 1997], and bonobos [de Waal, 1995].<br />
In the case of manatees, understanding relationships in captivity and the wild<br />
becomes especially important in the rehabilitation and eventual release of manatees.<br />
Releasing manatees near where they were captured may be beneficial because<br />
tradition is probably important in manatee survival [Bengtson, 1981]. Manatees may<br />
learn feeding and resting areas among other information from older conspecifics<br />
during development. Like elephants [McComb et al., 2001] and cetaceans [Wells<br />
et al., 1999], manatees may have population level, behavioral diversity related to<br />
cultural traditions. Associations between individuals beyond weaning may be vital in<br />
manatee society and enhance survivorship. Such information could be helpful for<br />
rehabilitation and conservation. To our knowledge, in the instances of simultaneous<br />
releases of manatees from captivity, associations established in captivity have not<br />
been maintained (at least five: female and a surrogate calf, two pairs of males, one<br />
female pair and one mixed sex pair; R. Bonde, Sirenia Project, personal<br />
communication). The sample size, however, is relatively small. Our data suggest<br />
that females do establish some particular association groups in captivity, but further<br />
study is needed to assess to what degree such associations are a result of captivity. In<br />
terms of captive manatee management, density did not seem to have any negative<br />
effects on manatee behavior and social aggregation of manatees seems to be a<br />
normal condition of captive living.<br />
CONCLUSIONS<br />
Captive female manatees at different facilities carried out regular non-aggressive<br />
contacts among conspecifics. Aggressive interactions were rare and not associated<br />
with feeding periods. The rates of non-aggressive contacts were density dependent
ut not random. Behaviors were not observed in the first 3 months of captivity nor<br />
were interactions with human handlers documented. Additional studies of captive<br />
and wild manatees are needed to understand association patterns and the role that<br />
relatedness and natal range might play in the level and types of interactions.<br />
ACKNOWLEDGMENTS<br />
We thank Walt Disney World The Living Seas at Epcot Center, Homosassa<br />
Springs Wildlife State Park, Lowry Park Zoo, Miami Seaquarium, Mote Marine<br />
Laboratory, Parker Aquarium in Bradenton and Sea World of Florida for granting<br />
permission to do this research. We also thank the following organizations for helping<br />
support this research: Georgia Southern University’s Academic Excellence Award,<br />
the Graduate Student Professional Development Fund at Georgia Southern<br />
University, and the Jane Smith Turner Foundation and the Eppley Foundation<br />
(awards to B.A.S.). This research was conducted with IACUC approval by Georgia<br />
Southern University and approval from each facility participating in the study.<br />
R. Bonde, J. Koelsch and Drs. R. Chandler, D. Gleason, and L. Leege provided<br />
valuable comments on the manuscript. R. Bonde has approved of the personal<br />
communication reference in the discussion. The research was conducted with the<br />
cooperation of the USGS Sirenia Project (MA79<strong>17</strong>21-2) and we thank their<br />
personnel for cooperation in this study.<br />
REFERENCES<br />
Alexander RD. 1974. The evolution of social<br />
behaviour. Ann Rev Ecol Syst 5:325<strong>–</strong>83.<br />
Beck C, Barros N. 1991. The impact of debris on<br />
the Florida manatee. Mar Poll Bull 22:508<strong>–</strong>10.<br />
Bengtson JL. 1981. Ecology of manatees (Trichechus<br />
manatus) in the St. Johns River, Florida.<br />
[DPhil dissertation]. Twin Cities: University of<br />
Minnesota.<br />
Bossart G, Baden D, Ewing R, Roberts B, Wright<br />
S. 1998. Brevetoxicosis in manatees (Trichechus<br />
manatus latirostris) from the 1996 epizootic:<br />
gross, histologic, and immunohistochemical<br />
features. Toxicol Pathol 26:276<strong>–</strong>82.<br />
Buss IO, Smith, NS. 1966. Observations on<br />
reproduction and breeding behavior of the<br />
African elephant. J Wildl Manage 30:375<strong>–</strong>88.<br />
de Waal FBM. 1995. Bonobo sex and society. Sci<br />
Am 272:82<strong>–</strong>8.<br />
Douglas-Hamilton I. 1972. On the ecology and<br />
behaviour of the African elephant. [DPhil<br />
dissertation]. Oxford: University of Oxford.<br />
Dublin HT. 1983. Cooperation and reproductive<br />
competition among female African elephants.<br />
In: Wasser SK, editor. Social behaviour of<br />
female vertebrates. New York: Academic Press.<br />
p 291<strong>–</strong>313.<br />
Eisenberg JF, McKay GM, Jainudeen, MR. 1971.<br />
Reproductive behavior of the Asiatic elephant<br />
Captive Female Manatee Social Behavior 143<br />
(Elephas maximus maximus L.). Behaviour<br />
38:193<strong>–</strong>225.<br />
Emlen ST. 1997. Predicting family dynamics in<br />
social vertebrates. In: Krebs J, Davies N, editors.<br />
Behavioral ecology, an evolutionary approach.<br />
Cambridge, MA: Blackwell Science. p 228<strong>–</strong>53.<br />
Gadgil M, Nair P. 1983. Observations on the social<br />
behaviour of free ranging groups of tame Asiatic<br />
elephants (Elephas maximus). Anim Sci 93:<br />
225<strong>–</strong>33.<br />
Glickman S, Zabel C, Yoerg S, Weldele M, Drea<br />
C, Frank L. 1997. Social affiliation, and dominance<br />
in the social life of spotted hyenas. In:<br />
Carter C, Lederhendler I, Kirkpatrick B, editors.<br />
The integrative neurobiology of affiliation: animals<br />
of the New York Academy of Science.<br />
Cambridge, MA: MIT Press. p 131<strong>–</strong>9.<br />
Hartman D. 1979. Ecology and behavior of the<br />
manatee (Trichechus manatus) in Florida. Special<br />
publication no. 5. Am Soc Mamm 153 p.<br />
Koelsch J. 1997. The seasonal occurrence and<br />
ecology of Florida manatees (Trichechus manatus<br />
latirostris) in coastal waters near Sarasota,<br />
Florida. [M.S. dissertation]. Sarasota: University<br />
of South Florida.<br />
Lee PC, Moss CJ. 1999. The social context for<br />
learning and behavioural development among<br />
wild African elephants. In: Box HO, Gibson KR,
144 Harper and Schulte<br />
editors. Mammalian social learning: comparative<br />
and ecological perspectives. New York: Cambridge<br />
University Press. p 102<strong>–</strong>25.<br />
Maher CR, Lott DF. 1995. Definitions of territoriality<br />
used in the study of variation in<br />
vertebrate spacing systems. Anim Behav<br />
49:1581<strong>–</strong>97.<br />
Maher CR, Lott DF. 2000. A review of ecological<br />
determinants of territoriality within vertebrate<br />
species. Am Midl Nat 143:1<strong>–</strong>29.<br />
Martin P, Bateson P. 1993. Measuring behaviour:<br />
an introductory guide. Second edition. Cambridge:<br />
Cambridge University Press. 222 p.<br />
McComb K, Moss C, Durant SM, Baker L,<br />
Soyialel S. 2001. Matriarchs as repositories of<br />
social knowledge in African elephants. Science<br />
292:491<strong>–</strong>4.<br />
Moss CJ. 1976. Portraits in the wild: behavior<br />
studies of East African mammals. Chicago (IL):<br />
University of Chicago Press. 363 p.<br />
Pusey A, Packer C. 1999. The ecology of relationships.<br />
In: Krebs J, Davies N, editors. Behavioral<br />
ecology, an evolutionary approach. Cambridge:<br />
Blackwell Science. p 254<strong>–</strong>83.<br />
Reynolds J. 1981. Behavior patterns in the West<br />
Indian manatee, with emphasis on feeding and<br />
diving. Fla Sci 44:233<strong>–</strong>42.<br />
Reynolds J, Odell D. 1991. Manatees and dugongs.<br />
New York: Facts on File, Inc. 192 p.<br />
Schilder MB. 1992. Stability and dynamics of<br />
group composition in a herd of captive plains<br />
zebras. Ethology 90:154<strong>–</strong>68.<br />
Schulte BA. 2000. Social structure and helping<br />
behavior in captive elephants. Zoo Biol 19:447<strong>–</strong>60.<br />
Sokal RR, Rohlf FJ. 1995. Biometry. 3rd edition.<br />
New York (NY): W.H. Freeman and Company.<br />
887 p.<br />
Wells RS, Boness DJ, Rathbun GB. 1999. Behavior.<br />
In: Reynolds JE III, Rommel SA, editors.<br />
Biology of marine mammals. Washington DC:<br />
Smithsonian Institution Press. p 324<strong>–</strong>422.<br />
Wright SD, Ackerman BB, Bonde RK, Beck CA,<br />
Banowetz DJ. 1995. Analysis of watercraftrelated<br />
mortality of manatees in Florida, 1979<strong>–</strong><br />
1991. In: O’Shea T, Ackerman B, Percival H,<br />
editors. Population of biology of the Florida<br />
manatee (Trichechus manatus latirostris).<br />
National Biological Service Information and<br />
Technology Report 1. Washington, DC: US<br />
Department of the Interior. p 259<strong>–</strong>68.<br />
Young JS. 2001. An investigation of captive<br />
Florida manatee (Trichechus manatus latirostris)<br />
behavior and social interactions. [M.S. dissertation].<br />
Statesboro: Georgia Southern University.
Community Structure, Population Control, and Competition<br />
Nelson G. Hairston; Frederick E. Smith; Lawrence B. Slobodkin<br />
The American Naturalist, Vol. 94, No. 879. (Nov. - Dec., 1960), pp. 421-425.<br />
Stable URL:<br />
http://links.jstor.org/sici?sici=0003-0147%28196011%2F12%2994%3A879%3C421%3ACSPCAC%3E2.0.CO%3B2-A<br />
The American Naturalist is currently published by The University of Chicago Press.<br />
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at<br />
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained<br />
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in<br />
the JSTOR archive only for your personal, non-commercial use.<br />
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at<br />
http://www.jstor.org/journals/ucpress.html.<br />
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed<br />
page of such transmission.<br />
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic<br />
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,<br />
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take<br />
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.<br />
http://www.jstor.org<br />
Tue Nov 6 13:08:48 2007
Vol. XCIV, No. 879 The American Naturalist November-December, 1960<br />
COMMUNITY STRUCTURE, POPULATION CONTROL,<br />
AND COMPETITION<br />
NELSON G. HAIRSTON, FREDERICK E. SMITH,<br />
AND LARIRENCE B. SLOBODKIN<br />
Department of Zoology, The University of Michigan, Ann Arbor, Michigan<br />
The methods whereby natural populations are limited in size have been<br />
debated with vigor during three decades, particularly during the last few<br />
years (see papers by Nicholson, Birch, Andrewartha, Milne, Reynoldson,<br />
and Hutchinson, and ensuing discussions in the Cold Spring Harbor Symposium,<br />
1957). Few ecologists will deny the importance of the subject,<br />
since the method of regulation of populations must be known before we can<br />
understand nature and predict its behavior-. Although discussion of the subject<br />
has usually been confined to single species populations, it is equally<br />
important in situations where two or more species are involved.<br />
The purpose of this note is to demonstrate a pattern of population control<br />
in many communities which derives easily from a series of general, widely<br />
accepted observations. The logic used is not easily refuted. Furthermore,<br />
the pattern reconciles conflicting interpretations by showing that populations<br />
in different trophic levels are expected to differ in their methods of<br />
control.<br />
Our first observation is that the accumulation of fossil fuels occurs at a<br />
rate that is negligible when compared with the rate of energy fixaeion<br />
through photosynthesis in the biosphere. Apparent exceptions to this observation,<br />
such as bogs and ponds, are successionai stages, in which the<br />
failure of decomposition hastens the termination of the stage. The rate of<br />
accumulation when compared with that of photosynthesis has also been<br />
shown to be negligible over geologic time (Hutchinson, 1948).<br />
If virtually all of the energy fixed in photosynthesis does indeed flow<br />
through he biosphere, it must follow that all organisms taken together are<br />
limited by the amount of energy fixed. In particular, the decomposers as a<br />
group must be food-limited, since by definition they comprise the trophic<br />
level which degrades organic debris. There is no a priori reason why predators,<br />
behavior, physiological changes induced by high densities, etc., could<br />
not limit decomposer populations. In fact, some decomposer populations<br />
may be limited in such ways. If so, however, others must consume the<br />
tt<br />
left-over" food, so that the group as a whole remains food limited; otherwise<br />
fossil fuel would accumulate rapidly.<br />
Any population which is not resource-limited must, of course, be limited<br />
to a level below that set by its resources.<br />
Our next three observations are interrelated. They apply primarily to terrestrial<br />
communities. The first of these is that cases of obvious depletion<br />
of green plants by herbivores are exceptions to the general picture, in which
422 THE AMERICAN NATURALIST<br />
the plants are abundant and largely intact. Moreover, cases of,obvious<br />
mass destruction by meteorological catastrophes are exceptional in most<br />
areas. Taken together, these two observations mean that producers are<br />
neither herbivore-limited nor catastrophe-limited, and must therefore be<br />
limited by their own exhaustion of a resource. In many areas, the limiting<br />
resource is obviously light, but in arid regions water may be the critical<br />
factor, and there are spectacular cases of limitation through the exhaustion<br />
of a critical mineral. The final observation in this group is that there are<br />
temporary exceptions to the general lack of depletion of green plants by<br />
herbivores. This occurs when herbivores are protected either by man or<br />
natural events, and it indicates that the herbivores are able to deplete the<br />
vegetation whenever they become numerous enough, as in the cases of the<br />
Kaibab deer herd, rodent plagues, and many insect outbreaks. It therefore<br />
follows that the usual condition is for populations of herbivores not to be<br />
limited by their food supply.<br />
The vagaries of weather have been suggested as an adequate method of<br />
control for herbivore populations. The best factual clues related to this<br />
argument are to be found in the analysis of the exceptional cases where<br />
terrestrial herbivores have become numerous enough to deplete the vegeta-<br />
tion. This often occurs with introduced rather than native species. It is<br />
most difficult to suppose that a species had been unable to adapt so as to<br />
escape control by the weather to which it was exposed, and at the same<br />
time by sheer chance to be able to escape this control from weather to<br />
which it had not been previously exposed. This assumption is especially<br />
difficult when mutual invasions by different herbivores between two coun-<br />
tries may in both cases result in pests. Even more difficult to accept, how-<br />
ever, is the implication regarding the native herbivores. The assumption<br />
that the hundreds or thousands of species native to a forest have failed to<br />
escape from control by the weather despite long exposure and much selec-<br />
tion, when an invader is able to defoliate without this past history, implies<br />
that "pre-adaptation" is more likely than ordinary adaptation. This we can-<br />
not accept.<br />
The remaining general method of herbivore control is predation (in its<br />
broadest sense, including parasitism, etc.). It is important to note that<br />
this hypothesis is not denied by the presence of introduced pests, since it<br />
is necessary only to suppose that either their natural predators have been<br />
left behind, or that while the herbivore is abIe to exist in the new climate,<br />
its enemies are not. There are, furthermore, numerous examples of the di-<br />
rect effect of predator removal. The history of the Kaibab deer is the best<br />
known example, although deer across the northern portions of the country<br />
are in repeated danger of winter starvation as a result of protection and<br />
predator removal. Several rodent plagues have been attributed to the local<br />
destruction of predators. More recently, the extensive spraying of forests<br />
to kill caterpillars has resulted in outbreaks of scale insects. The latter<br />
are protected from the spray, while their beetle predators and other insect<br />
enemies are not.
POPULATION CONTROL AND COMPETITION<br />
Thus, although rigorous proof that herbivores are generally controlled by<br />
predation is lacking, supporting evidence is available, and the alternate<br />
hypothesis of control by weather leads to false or untenable implications.<br />
The foregoing conclusion has an important implication in the mechanism<br />
of control of the predator populations. The predators and parasites, in con-<br />
trolling the populations of herbivores, must thereby limit their own re-<br />
sources, and as a group they must be food-limited. Although the populations<br />
of some carnivores are obviously limited by territoriality, this kind of in-<br />
ternal check cannot operate for all carnivores taken together. If it did, the<br />
herbivores would normally expand to the point of depletion of the vegeta-<br />
tion, as they do in the absence of their normal predators and parasites.<br />
There thus exists either direct proof or a great preponderance of factual<br />
evidence that in terrestrial communities decomposers, producers, and preda-<br />
tors, as whole trophic levels, are resource-limited in the classical density-<br />
dependent fashion. Each of these three can and does expand toward the<br />
limit of the appropriate resource. We may now examine the reasons why this<br />
is a frequent situation in nature.<br />
Whatever the resource for which a set of terrestrial plant species com-<br />
pete, the competition ultimately expresses itself as competition for space.<br />
A community in which this space is frequently emptied through depletion<br />
by herbivores would run the continual risk of replacement by another assemblage<br />
of species in which the herbivores are held down in numbers by<br />
predation below the level at which they damage the vegetation. That space<br />
once held by a group of terrestrial plant species is not readily given up is<br />
shown by the cases where relict stands exist under climates no longer suit-<br />
able for their return following deliberate or accidental destruction. Hence,<br />
the community in which herbivores are held down in numbers, and in which<br />
the producers are resource-limited will be the most persistent. The develop-<br />
ment of this pattern is less likely where high producer mortalities are in-<br />
evitable. In lakes, for example, algal populations are prone to crash whether<br />
grazed or not. In the same environment, grazing depfetion is much more<br />
common than in communities where the major producers are rooted plants.<br />
A second general conclusion follows from the resource limitation of the<br />
species of three uophic levels. This conclusion is that if more than one<br />
species exists in one of these levels, they may avoid competition only if<br />
each species is limited by factors completely unutilized by any of the other<br />
species. It is a fact, of course, that many species occupy each level in<br />
most communities. It is also a fact that they are not sufficiently segregated<br />
in their needs to escape competition. Although isolated cases of nonoverlap<br />
have been described, this has never been observed for an entire<br />
assemblage. Therefore, interspecific competition for resources exists<br />
among producers, among carnivores, and arllong decomposers.<br />
It is satisfying to note the number of observations that fall into line with<br />
the foregoing deductions. Interspecific competition is a powerful selective<br />
force, and we should expect to find evidence of its operation. Moreover,<br />
the evidence should be most conclusive in trophic levels where it is neces-<br />
423
424 THE AMERICAN NATURALIST<br />
sarily present. Among decomposers we find the most obvious specific<br />
mechanisms for reducing populations of competitors. The abundance of<br />
antibiotic substances attests to the frequency with which these mechanisms<br />
have been developed in the trophic level in which interspecific competition<br />
is inevitable. The producer species are the next most likcly to reveal evi-<br />
dence of competition, and here we find such phenomena as crowding, shad-<br />
ing, and vegetational zonation.<br />
Among the carnivores, however, obvious adaptations for interspecific<br />
competition are less common. Active competition in the form of mutual<br />
habitat-exclusion has been noted in the cases of flatworms (Beauchamp and<br />
Ullyott, 1932) and salamanders (Hairston, 1951). The commonest situation<br />
takes the form of niche diversification as the result of interspecific compe-<br />
tition. This has been noted in birds (Lack, 1945; MacArthur, 1958), sala-<br />
manders (Hairston, 1949), and other groups of carnivores. Quite likely,<br />
host specificity in parasites and parasitoid insects is at least partly due<br />
to the influence of interspecific competition.<br />
Of equal significance is the frequent occurrence among herbivores of ap-<br />
parent exceptions to the influence of density-dependent factors. The grass-<br />
hoppers described by Birch (1957) and the thrips described by Davidson and<br />
Andrewartha (1948) are well known examples. Moreover, it is among herbi-<br />
vores that we find cited examples of coexistence without evidence of com-<br />
petition for resources, such as the leafhoppers reported by Ross (1957), and<br />
the psocids described by Broadhead (19>8). It should be pointed out that<br />
in these latter cases coexistence applies primarily to an identity of food<br />
and place, and other aspects of the niches of these organisms are not known<br />
to be identical.<br />
SUMMARY<br />
In summary, then, our general conclusions are: (1) Populations of producers,<br />
carnivores, and decomposers are limited by their respective resources<br />
in the classical density-dependent fashion. (2) Interspecific com-<br />
petition must necessarily exist among the members of each of these three<br />
trophic levels. (3) Herbivores are seldom food-limited, appear most often<br />
to be predator-limited, and therefore are not likely to compete for common<br />
resources.<br />
LITERATURE CITED<br />
Andrewartha, H. G., 1957, The use of conceptual models in population ecol-<br />
ogy. Cold Spring Harbor Symp. Quant. Biol. 22: 219-232.<br />
Beauchamp, R. S. A., and P. Ullyott, 1937, Competitive relationships be-<br />
tween certain species of fresh-water triclads. J. Ecology 20: 200-<br />
208.<br />
Birch, L. C., 1957, The role of weather in determining the distribution and<br />
abundance of animals. Cold Spring Harbor Symp. Quant. Biol. 22:<br />
2<strong>17</strong>-263.<br />
Broadhead, E., 1958, The psocid fauna of larch trees in northern England.<br />
J. Anim. Ecol. 27: 2<strong>17</strong>-263.
POPULATION CONTROL AND COMPETITION<br />
425<br />
Davidson, J., and H. G. Andrewartha, 1948, The influence of rainfall, evaporation<br />
and atmospheric temperature on fluctuations in the size of a<br />
natural population of Thrzps zmagznis (Thysanoptera). j. him.<br />
Ecol. <strong>17</strong>: 200-222.<br />
Hairston, N. G., 1949, The local distribution and ecology of the Plethodontid<br />
salamanders of the southern Appalachians. Ecol. hlonog.<br />
19: 47-73.<br />
1951, Interspecies competition and its probable influence upon the vertical<br />
distribution of Appalachian salamanders of the genus Plethodon.<br />
Ecology 32: 266-274.<br />
Hutchinson, G. E., 1948, Circular causal systems in ecology. Ann. M.Y.<br />
Acad. Sci. 50: 221-246.<br />
1957, Concluding remarks. Cold Spring Harbor Symp. Quant. Biol. 22:<br />
415-427.<br />
Lack, D., 1945, The ecology of closely related species with special reference<br />
to cormorant (Phalacrocorax carbo) and shag (P, arzstotelzs).<br />
J. Anim, Ecol. 14: 12-16.<br />
MacArthur, R. H., 1958, Population ecology of some warblers of northeastern<br />
coniferous forests. Ecology 39: 599-619.<br />
Milne, A., 1957, Theories of riatural control of insect populations. Cold<br />
Spring Harbor Symp. Quane. Biol. 22: 253-271.<br />
Nicholson, A. J., 1957, The self-adjustment of populations to change, Cold<br />
Spring Harbor Symp. Ouant. Biol. 22: 153-<strong>17</strong>2.<br />
Reynoldson, T. B., 1957, Population fluctuations in Urceolarza rnzt~a (Peritricha)<br />
and Enchytraeus nlbzdus (Oligochaeta) and their bearing on<br />
regulation. Cold Spring Hat bor Symp. Quant. Biol. 22: 313-327.<br />
Ross, H.H., 1957, Principles of natural coexistence indicated by leafhopper<br />
populations. Evolutaon 11 : 113-129.
Environmental Conservation 29 (2): 192<strong>–</strong>206 © 2002 Foundation for Environmental Conservation DOI:10.10<strong>17</strong>/S0376892902000127<br />
The future of seagrass meadows<br />
CARLOS M. DUARTE*<br />
IMEDEA (CSIC <strong>–</strong> UiB), C/ Miquel Marqués 21, 07190 Esporles, (Islas Baleares), Spain<br />
Date submitted: 29 May 2001 Date accepted: 15 February 2002<br />
SUMMARY<br />
Seagrasses cover about 0.1<strong>–</strong>0.2% of the global ocean, and<br />
develop highly productive ecosystems which fulfil a key<br />
role in the coastal ecosystem. Widespread seagrass loss<br />
results from direct human impacts, including mechanical<br />
damage (by dredging, fishing, and anchoring),<br />
eutrophication, aquaculture, siltation, effects of coastal<br />
constructions, and food web alterations; and indirect<br />
human impacts, including negative effects of climate<br />
change (erosion by rising sea level, increased storms,<br />
increased ultraviolet irradiance), as well as from natural<br />
causes, such as cyclones and floods. The present review<br />
summarizes such threats and trends and considers<br />
likely changes to the 2025 time horizon. Present losses<br />
are expected to accelerate, particularly in South-east<br />
Asia and the Caribbean, as human pressure on the<br />
coastal zone grows. Positive human effects include<br />
increased legislation to protect seagrass, increased<br />
protection of coastal ecosystems, and enhanced efforts<br />
to monitor and restore the marine ecosystem. However,<br />
these positive effects are unlikely to balance the negative<br />
impacts, which are expected to be particularly prominent<br />
in developing tropical regions, where the capacity<br />
to implement conservation policies is limited.<br />
Uncertainties as to the present loss rate, derived from<br />
the paucity of coherent monitoring programmes, and<br />
the present inability to formulate reliable predictions as<br />
to the future rate of loss, represent a major barrier to the<br />
formulation of global conservation policies. Three key<br />
actions are needed to ensure the effective conservation<br />
of seagrass ecosystems: (1) the development of a<br />
coherent worldwide monitoring network, (2) the development<br />
of quantitative models predicting the responses<br />
of seagrasses to disturbance, and (3) the education of the<br />
public on the functions of seagrass meadows and the<br />
impacts of human activity.<br />
Keywords: seagrass, conservation, status, perspectives, global<br />
change<br />
INTRODUCTION<br />
Seagrasses ecosystems are widely recognized as key ecosystems<br />
in the coastal zone, with important functions in the<br />
* Correspondence: Professor Carlos M. Duarte Fax: +34 971<br />
61<strong>17</strong>61 e-mail: cduarte@uib.es<br />
marine ecosystem (Hemminga & Duarte 2000). Yet, there is<br />
growing evidence that seagrass meadows are presently experiencing<br />
worldwide decline primarily because of human<br />
disturbance, such as direct physical damage and deterioration<br />
of water quality (Short & Wyllie-Echeverria 1996;<br />
Hemminga & Duarte 2000). There is, therefore, concern that<br />
the functions seagrasses have performed in the marine<br />
ecosystem will be reduced or, in some places, lost altogether.<br />
While efforts are being made to conserve seagrass ecosystems,<br />
these are not guided by a clear forecast of the changes<br />
expected and the threats to come. There is, therefore, a need<br />
for a prospective examination of the expected status of<br />
seagrass ecosystems in the future, which would provide guidance<br />
for the implementation of effective conservation<br />
policies. However, such an exercise requires reliable forecasts<br />
as to the expected progression of the relevant properties<br />
affecting the future status of seagrass ecosystems. Over the<br />
time horizon of 2025, there are estimates of human population<br />
growth and global climate change (Foundation for<br />
Environmental Conservation 2001), and the goal of this<br />
review is to provide a forecast of the likely status of seagrass<br />
ecosystems over that period. However, as Noble laureate Nils<br />
Bohr once stated, ‘prediction is very difficult, particularly if<br />
it concerns the future’, particularly given the many gaps in<br />
the present knowledge on the status of seagrass ecosystems.<br />
Hence, the future scenario depicted here should be<br />
considered as an ecological forecast, rather than a prediction,<br />
driven by the projected trends in the environmental factors<br />
that shape the seagrass habitat. While such forecasts must<br />
necessarily entail considerable uncertainty, their formulation<br />
requires a revision of our understanding of the response of<br />
seagrass ecosystems to a changing ecosystem that may help<br />
identify the areas where more robust knowledge is most<br />
needed. Hence, independently of the reliability of the forecast<br />
issued here, this review will deliver an identification of<br />
the research required to increase the reliability of the predictions<br />
about the response of seagrass ecosystems to a changing<br />
environment.<br />
This review differs from previous attempts to forecast the<br />
future status of seagrass ecosystems in that these focused on<br />
their response to climate change (e.g. Brouns 1994; Beer &<br />
Koch 1996; Short & Neckles 1999), whereas the present exercise<br />
also considers the responses to the pressures derived<br />
from a growing human population. First, seagrasses and the<br />
functions they perform in the marine ecosystem are discussed<br />
together with their response to environmental forcing factors,
including direct and indirect human impacts. Future threats<br />
and the consequences of seagrass loss are used to identify<br />
long-term trends and the likely status of seagrass meadows in<br />
2025. This review concludes by discussing the managerial<br />
frameworks that must be implemented to anticipate the<br />
trends forecasted and optimize the conservation status of<br />
seagrass meadows by year 2025. Indeed, the forecast<br />
provided here reflects a projection from a ‘business as usual’<br />
situation, whereas the adaptive nature of society to change,<br />
through the development of technologies and practices that<br />
minimize human impacts on the coastal environment, may<br />
render the forecasts issued here wrong. The required change<br />
of attitudes and technological developments may be,<br />
however, encouraged by the consideration of the future scenarios<br />
depicted here.<br />
SEAGRASSES AND THEIR FUNCTIONS IN THE<br />
ECOSYSTEM<br />
Seagrasses are angiosperms restricted to growth in marine<br />
environments. These plants evolved early in the history of<br />
angiosperms, but, unlike their land counterparts, showed a<br />
comparative evolutionary stasis (sensu Burt 2001), with their<br />
species richness being unlikely to exceed one hundred species<br />
at any one time through their evolutionary history (den<br />
Hartog 1970; Duarte 2001). All seagrass species are rhizomatous,<br />
clonal plants, occupying space through the reiteration<br />
of shoots, with their leaves and roots produced as a result of<br />
rhizome extension (Marbá & Duarte 1998; Hemminga &<br />
Duarte 2000). This asexual process appears to be the mechanism<br />
for seagrass proliferation, although some species such<br />
as Zostera marina (Olesen 1999) and Enhalus acoroides<br />
(Duarte et al. 1997a) reproduce a lot sexually, a mode of<br />
reproduction which is uncommon in some other species like<br />
Cymodocea serrulata and Posidonia oceanica (Hemminga &<br />
Duarte 2000). Seagrass species vary about 100-fold in growth<br />
rate and lifespan, which is inversely related to size (Duarte<br />
1991a; Marbá & Duarte 1998; Hemminga & Duarte 2000).<br />
Seagrasses require an underwater irradiance generally in<br />
excess of 11% of that incident in the water surface for<br />
growth, a requirement that typically sets their depth limit<br />
(Duarte 1991b). The upslope limit of seagrasses is imposed<br />
by their requirement for sufficient immersion in seawater or<br />
tolerable disturbance by waves and, in northern latitudes, ice<br />
scour (Hemminga & Duarte 2000). Most seagrass species<br />
grow subtidally, although species within some genera such as<br />
Zostera spp., Phyllospadix spp. and Halophila spp. can grow<br />
intertidally (den Hartog 1970; Hemminga & Duarte 2000).<br />
Seagrasses can grow in estuarine and brackish waters, but<br />
require salinity in excess of 5 ‰ to develop (Hemminga &<br />
Duarte 2000). Although some species such as Phyllospadix<br />
spp. grow on rocky shores, most grow on sediments, ranging<br />
from sandy to muddy, with organic contents
194 C.M. Duarte<br />
well as more recent massive losses such as that in Florida Bay<br />
(Robblee et al. 1991), which is one of the largest areas of<br />
seagrass ecosystem worldwide (Fourqurean & Robblee 1999).<br />
The causes and the possible role of human-derived impacts<br />
in such losses are still uncertain (cf. Hemming & Duarte<br />
2000). Strong disturbances, such as damage by hurricanes,<br />
can also lead to major seagrass losses (Preen et al. 1995;<br />
Poiner et al. 1989). Smaller-scale, more recurrent disturbances,<br />
such as that caused by the motion of sand waves in<br />
and out of seagrass patches (Marbá & Duarte 1995) and that<br />
caused by large predators, such as dugongs (Nakaoka & Aioi<br />
1999), represent the main factor structuring some seagrass<br />
landscapes, which are characteristically patchy (Marbá &<br />
Duarte 1995). In contrast, some seagrass species have been<br />
able to form long-lasting meadows, with meadows of the<br />
long-lived Posidonia oceanica dated to >4000 years old<br />
(Mateo et al. 1997), and single clones of Zostera marina dated,<br />
using molecular techniques at 3000 years old (Reusch et al.<br />
1999).<br />
In addition to natural variability of the seagrass habitat,<br />
human intervention is becoming a major source of change to<br />
seagrass ecosystems, whether by direct physical modification<br />
of the habitat by the growing human activity in the coastal<br />
zone (e.g. boating, fishing, construction; Walker et al. 1989),<br />
or through their impact on the quality of waters and sediments<br />
to support seagrass growth (Short & Wyllie-<br />
Echeverria 1996; Hemminga & Duarte 2000), as well as<br />
changes in the marine food webs linked to the seagrasses<br />
(Aragones & Marsh 2000; Jackson 2001; Jackson et al. 2001).<br />
Human impacts<br />
Humans impact seagrass ecosystems, both through direct<br />
proximal impacts, affecting seagrass meadows locally, and<br />
indirect impacts, which may affect seagrass meadows far<br />
away from the sources of the disturbance (Table 1). Proximal<br />
impacts include mechanical damage and damage created by<br />
the construction and maintenance of infrastructures in the<br />
coastal zone, as well as effects of eutrophication, siltation,<br />
coastal engineering and aquaculture (Table 1; Fig. 1).<br />
Indirect impacts include those from global anthropogenic<br />
changes (Fig. 2), such as global warming, sea-level rise, CO 2<br />
and ultraviolet (UV) increase, and anthropogenic impacts on<br />
marine biodiversity, such as the large-scale modification of<br />
the oceanic food web through fisheries ( Jackson 2001;<br />
Jackson et al. 2001). Indirect impacts are already becoming<br />
evident at present (e.g. Beer & Koch 1996).<br />
The most unambiguous source of human impact to<br />
seagrass ecosystems is physical disturbance. This susceptibility<br />
derives from multiple causes, all linked to increasing<br />
human usage of the coastal zone for transportation, recreation<br />
and food production. The coastal zone is becoming an<br />
important focus for services to society, since about 40% of<br />
the human population presently inhabit the coastal zone<br />
(Independent World Commission on the Oceans 1998).<br />
Direct habitat destruction by land reclamation and port<br />
construction is a major source of disturbance to seagrass<br />
meadows, due to dredging and landfill activities, as well as<br />
the reduction in water transparency associated with both<br />
these activities. The construction of new ports is associated<br />
with changes in sediment transport patterns, involving both<br />
increased erosion and sediment accumulation along the adjacent<br />
coast. These changes can exert significant damage on<br />
seagrass ecosystems kilometres away, which can be impacted<br />
by both erosion and burial associated with the changing sedimentary<br />
dynamics (e.g. Pascualini et al. 1999). The operation<br />
of the ports also entails substantial stress to the neighbouring<br />
seagrass meadows, due to reduced transparency and nutrient<br />
Table 1 Impacts of direct and indirect human forcing on seagrass ecosystems.<br />
Type Forcing Possible consequences Mechanisms<br />
Direct impacts Mechanical damage (e.g. trawling, Seagrass loss Mechanical removal and sediment<br />
dredging, push nets, anchoring,<br />
dynamite fishing)<br />
erosion<br />
Eutrophication Seagrass loss Deterioration of light and sediment conditions<br />
Salinity changes Seagrass loss, changes in community Osmotic shock<br />
structure<br />
Shoreline development Seagrass loss due to burial or erosion Seagrass uprooting<br />
Land reclamation Seagrass loss Seagrass burial and shading<br />
Aquaculture Seagrass loss Deterioration of light and sediment conditions<br />
Siltation Seagrass loss and changes in<br />
community structure<br />
Deterioration of light and sediment conditions<br />
Indirect impacts Seawater temperature rise Altered functions and distributions Increased respiration, growth and flowering,<br />
increased microbial metabolism<br />
Increased CO concentration 2 Increased depth limits and Increased photosynthesis, eventual decline of<br />
production calcifying organisms<br />
Sea level rise and shoreline erosion Seagrass loss Seagrass uprooting<br />
Increased wave action and storms Seagrass loss Seagrass uprooting<br />
Food web alterations Changes in community structure Changes in sediment conditions and<br />
disturbance regimes
Nutrient inputs<br />
+<br />
+<br />
+<br />
Phytoplankton<br />
Epiphytic and<br />
opportunistic<br />
algae (+)<br />
- (shading)<br />
(Sediment anoxia)<br />
Grazers<br />
and contaminant inputs associated with ship traffic and<br />
servicing, as well as dredging activities associated with port<br />
and navigation-channel maintenance. Rapid increases in seabased<br />
transport, as well as recreational boating activities have<br />
led to a major increase in the number and size of ports worldwide<br />
(Independent World Commission on the Oceans 1998),<br />
with a parallel increase in the combined disturbance to<br />
seagrass meadows. Ship activity also causes disturbance to<br />
seagrass through anchoring damage, which can be rather<br />
extensive at popular mooring sites (Walker et al. 1989), as<br />
well as fisheries operation, particularly shallow trawling<br />
(Ramos-Esplá et al. 1993; Pascualini et al. 1999) and smallerscale<br />
activities linked to fisheries, such as clam digging and<br />
use of push nets over intertidal and shallow areas and, in<br />
extreme cases, dynamite fishing (Kirkman & Kirkman 2000).<br />
The exponential growth of aquaculture, the fastest-growing<br />
food production industry, has also led to impacts on<br />
seagrasses through shading and physical damage to the<br />
-<br />
-<br />
-<br />
Sediment resuspension<br />
-<br />
Figure 1 Effects of eutrophication derived from increased nutrient<br />
loading on seagrass ecosystems. Positive and negative indicate the<br />
sign of the effects. Chained positive and negative effects affect<br />
seagrasses negatively, whereas chained negative effects affect<br />
seagrasses positively.<br />
New habitat<br />
Sea level rise<br />
+<br />
- Submarine<br />
erosion<br />
deep regression<br />
Increased CO2<br />
Photosynthesis+ - reduced calcification<br />
Seawater warming<br />
Reproduction<br />
habitat<br />
expansion +<br />
-Increased respiration<br />
Figure 2 Forecasted effects of climate change on seagrass<br />
ecosystems. All components include positive and negative effects.<br />
Future of seagrass meadows 195<br />
seagrass beds, as well as deterioration of water and sediment<br />
quality leading to seagrass loss (Delgado et al. 1997, 1999;<br />
Pergent et al. 1999; Cancemi et al. 2000).<br />
The coastal zone also supports increasing infrastructure,<br />
such as pipes and cables for transport of gas, water, energy<br />
and communications, deployment and maintenance of which<br />
also entail disturbance to adjacent seagrass meadows. The<br />
development of coastal tourism, the fastest-growing industry<br />
in the world, has also led to a major transformation of the<br />
coastal zone in areas with pleasant climates. For instance,<br />
about two-thirds of the Mediterranean coastline is urbanized<br />
at the present time, with this fraction exceeding 75% in the<br />
regions with the most developed tourism industry (UNEP<br />
[United Nations Environment Programme] 1989), with<br />
harbours and ports occupying 1250 km of the European<br />
Mediterranean coastline (European Environmental Agency<br />
1999). Urbanization of the coastline often involves destruction<br />
of dunes and sand deposits, promoting beach erosion, a<br />
major problem for beach tourism. Beach erosion, however,<br />
does not only affect the emerged beach, and is usually propagated<br />
to the submarine sand colonized by seagrass,<br />
eventually causing seagrass loss (Medina et al. 2001).<br />
Groynes constructed to prevent beach erosion often create<br />
extensive problems, by altering longshore sediment transport<br />
patterns, further impacting the seagrass ecosystem.<br />
Extraction of marine sand for beach replenishment is only<br />
economically feasible at the shallow depths inhabited by<br />
seagrasses, which are often impacted by these extraction<br />
activities (Medina et al. 2001). The threats coastal tourism<br />
poses to seagrasses are sometimes direct, as in some cases of<br />
purposeful removal of seagrass from beach areas to ‘improve’<br />
beach conditions. Fortunately, there are indications that<br />
coastal tourism is attempting, at least in some areas, to<br />
embrace sustainable principles, including the maintenance of<br />
ecosystem services, such as those provided by seagrasses, and<br />
could well play a role in the future as an agent pressing for<br />
seagrass conservation.<br />
Widespread eutrophication of coastal waters (Vidal et al.<br />
1999) derived from the excessive nutrient input to the sea, is<br />
leading to global deterioration in the quality of coastal waters<br />
(Cloern 2001), which is identified as a major loss factor for<br />
seagrass meadows worldwide (Duarte 1995; Short & Wyllie-<br />
Echeverria 1996; Hemminga & Duarte 2000). Human<br />
activity presently dominates the global nitrogen cycle, with<br />
anthropogenic fixation of atmospheric nitrogen now<br />
exceeding natural sources (Vitousek et al. 1997), and anthropogenic<br />
nitrogen now dominating the reactive nitrogen pools<br />
in the atmosphere, and therefore rainwater, of industrialized<br />
and agricultural areas (Nixon et al. 1996). Hence, anthropogenic<br />
nitrogen dominates the nitrogen inputs to<br />
watersheds, with the human domination of nitrogen fluxes<br />
being reflected in a close relationship between nitrate export<br />
rate and human population in the world’s watersheds (Cole et<br />
al. 1993). Tertiary water-treatment plants only achieve a<br />
partial reduction in nitrogen inputs to the sea, for nitrogen<br />
inputs to the coastal zone are already dominated by direct
196 C.M. Duarte<br />
atmospheric inputs in heavily industrialized or agricultural<br />
areas (Paerl 1995).<br />
Although seagrass meadows are often nutrient limited<br />
(e.g. Duarte 1990), increased nutrient inputs can only be<br />
expected to enhance seagrass primary production at very<br />
moderate levels at best (Borum & Sand-Jensen 1996).<br />
Whereas seagrasses, through their low nutrient requirements<br />
for growth and their high capacity for internal nutrient recycling<br />
(Hemminga & Duarte 2000), are well fitted to cope with<br />
low nutrient availability, other primary producers, both<br />
micro- and macroalgal, are more efficient, because of greater<br />
affinity and higher uptake rates, in using excess nutrient<br />
inputs (Duarte 1995; Hein et al. 1995). Coastal eutrophication<br />
promotes phytoplankton biomass, which deteriorates the<br />
underwater light climate, and the stimulation of the growth<br />
of epiphytes and opportunistic macroalgae, which further<br />
shade and suffocate seagrasses (Duarte 1995; den Hartog<br />
1994; Hemminga & Duarte 2000; Hauxwell et al. 2001). The<br />
alleviation of nutrient limitation, together with the proliferation<br />
of phytoplankton and epiphyte biomass as a result of<br />
increased nutrient inputs imply that coastal eutrophication<br />
leads to a shift from nutrient limitation to light limitation of<br />
ecosystem production (Cloern 2001), enhanced through<br />
competitive interactions between different types of primary<br />
producers for light (Sand-Jensen & Borum 1991; Duarte<br />
1995). These effects result in seagrass loss, particularly in the<br />
deeper portions of the meadows (Sand-Jensen & Borum<br />
1991; Duarte 1995; Borum 1996). Heavy grazing, which can<br />
buffer the negative effects of eutrophication, may attenuate<br />
the effects of overgrowth by phytoplankton, epiphytes and<br />
macroalgae (cf. Heck et al. 2000; Hemminga & Duarte 2000).<br />
Eutrophication may, furthermore, have negative effects<br />
directly derived from the high resulting nutrient concentration,<br />
for high nitrate and ammonium concentrations may<br />
be toxic to seagrasses (e.g. Van Katwijk et al. 1997).<br />
Whereas research on the effects of eutrophication on<br />
seagrass meadows has focused on the effects of reduced light<br />
quality (Duarte 1995), the deterioration of the sediment<br />
conditions may also play a critical role in enhancing the loss<br />
of seagrasses. Seagrass sediments are typically rich in organic<br />
materials, due to the enhanced particle deposition and trapping<br />
under seagrass canopies (Terrados & Duarte 1999;<br />
Gacia & Duarte 2001) compared to adjacent bare sediments.<br />
Microbial processes are, therefore, stimulated in the seagrass<br />
rhizosphere (Hemminga & Duarte 2000), which, if sufficiently<br />
intense, lead to the depletion of oxygen and the<br />
development of bacterial communities with anaerobic metabolism,<br />
which release by-products, such as sulphide and<br />
methane, that may be toxic to seagrasses (e.g. Carlson et al.<br />
1994; Terrados et al. 1999). In order to avoid such toxicity<br />
effects, seagrasses pump a significant fraction of the photosynthetic<br />
oxygen produced to the roots, which release oxygen<br />
to maintain an oxidized microlayer at the root surface<br />
(Pedersen et al. 1998). However, eutrophication reduces<br />
seagrass primary production both through shading and<br />
seagrass loss, thereby reducing the oxygen seagrass roots may<br />
release. This allows anaerobic processes and the resulting<br />
metabolites to accumulate closer to the root surface,<br />
increasing the chances of toxic effects to seagrass. At the same<br />
time, the increased pelagic primary production leads to a<br />
greater input of organic matter to the sediments, enhancing<br />
microbial activity and the sediment oxygen deficit, which<br />
may increase the production of metabolites from anaerobic<br />
microbial metabolism. Both these processes result in the<br />
deterioration of the sediment environment to support<br />
seagrass growth, leading, through its interaction with the<br />
consequences of reduced light availability, to accelerated<br />
seagrass loss (Hemminga & Duarte 2000). Eutrophication<br />
effects on sediments may be more acute where sewage is the<br />
dominant source of nutrients, for this is discharged along<br />
with a high organic load, stimulating microbial activity<br />
( Jorgensen 1996). Aquaculture activities are becoming<br />
increasingly prominent in the shallow, sheltered coastal<br />
waters where seagrass meadows abound. Shading and high<br />
inputs of organic matter from fish cages have been shown to<br />
lead to seagrass decline below and around fish cages (Delgado<br />
et al. 1997, 1999; Pergent et al. 1999, Cancemi et al. 2000),<br />
through processes comparable to those of the eutrophication<br />
outlined above.<br />
Increased siltation of coastal waters is also a major human<br />
impact on seagrass ecosystems, which derives from changes<br />
in land use leading to increased erosion rates and silt export<br />
from watersheds. Siltation is a particularly acute problem in<br />
South-east Asian coastal waters, which receive the highest<br />
sediment delivery in the world as a result of high soil erosion<br />
rates derived from extensive deforestation and other changes<br />
in land use (Miliman & Meade 1993). Siltation severely<br />
impacts South-east Asian seagrass meadows through<br />
increased light attenuation (Bach et al. 1998) and burial<br />
(Duarte et al. 1997b), leading to seagrass loss and, where less<br />
intense siltation occurs, a decline in seagrass diversity,<br />
biomass and production (Terrados et al. 1998).<br />
Large-scale coastal engineering often alters circulation<br />
and salinity distributions, leading to seagrass loss. Hence,<br />
seagrass meadows, previously abundant in Dutch coastal<br />
areas, are now much reduced in surface, partially related to<br />
shifts of coastal waters from marine to brackish or freshwater<br />
regimes (e.g. Nienhuis et al. 1996).<br />
Pollution, other than that from nutrients and organic<br />
inputs, may be an additional source of human impacts on<br />
seagrass ecosystems, although, seagrass appears to be rather<br />
resistant to pollution by organic and heavy metal contaminants<br />
(Hemminga & Duarte 2000). These substances may<br />
possibly harm some components of the seagrass ecosystem,<br />
although such responses have not been examined to a significant<br />
extent.<br />
Human activity in the coastal zone has greatly impacted<br />
biodiversity at regional scales, whether by removing or<br />
adding species. These changes to marine biodiversity impact<br />
food webs and competitive interactions between organisms,<br />
directly and indirectly affecting seagrass beds ( Jackson 2001;<br />
Jackson et al. 2001). Hunting has led to a major decline in the
abundance of dugongs and sea turtles, which are agents of<br />
small-scale disturbance influencing the structure and<br />
dynamics of seagrass ecosystems (e.g. Aragones & Marsh<br />
2000). Indeed, there is concern that the depletion of dugong<br />
and sea turtle populations to levels where they can no longer<br />
exert an impact at the basin scale, may lead to major shifts in<br />
seagrass community structure, with the decline of the species<br />
dependent on recurrent small-scale disturbance, such as a<br />
Halophila spp. (Preen 1995; De Iongh 1996). These concerns<br />
stem from evidence from the fossil record that extinctions<br />
within the one diverse syrenid fauna were associated with<br />
extinctions of seagrass species (Domning 2001). It has been<br />
suggested recently that the decline in mega-grazers may have<br />
caused an increase in organic matter deposition in seagrass<br />
sediments, possibly explaining recent large-scale decline of<br />
seagrasses in systems such as Florida Bay, USA, and<br />
Moreton Bay, Australia ( Jackson 2001; Jackson et al. 2001).<br />
However, the evidence to suggest that the loss of megagrazers<br />
results or has resulted in a deterioration of habitat<br />
conditions to support seagrass is meagre at best ( Jackson<br />
2001; Jackson et al. 2001), so that these suggestions should be<br />
considered as interesting hypotheses pending confirmation,<br />
rather than established facts. In addition, high grazing<br />
pressure in the past, such as implied by this hypothesis,<br />
stresses the plants and ultimately results in a decline of leaf<br />
production, which has been ascribed to a depletion in the<br />
sediment nutrients available for plant growth (Zieman et al.<br />
1984).<br />
Global trade has increased the mobility of marine species,<br />
whether purposely, such as aquarium specimens, or inadvertently,<br />
such as organisms carried in ballast waters. The<br />
increased human-mediated transport of species between<br />
geographically distant locations has increased the incidence<br />
of invasive species. A case affecting Mediterranean, and<br />
probably soon Eastern Pacific seagrasses, is the invasion of<br />
the Mediterranean by the tropical algal species Caulerpa taxifolia,<br />
which first invaded the French Mediterranean in the<br />
early 1980s, apparently released from an aquarium, and has<br />
been reported to have expanded since along the French coast<br />
to reach the Italian and Spanish (Majorca Island) coasts<br />
(Meinesz & Hesse 1991). C. taxifolia grows rapidly and<br />
appears largely to colonize areas devoid of seagrasses, but has<br />
been reported to compete for space and resources with<br />
Posidonia oceanica off Monaco (Meinesz & Hesse 1991), being<br />
able to damage the Posidonia oceanica meadows, particularly<br />
when these are already under stress (de Villéle & Verlaque<br />
1995; Jaubert et al. 1999). The species has recently been<br />
reported on the Californian coast, raising concerns that it<br />
could also cause problems for the seagrass beds there. The<br />
Mediterranean Sea has also been invaded by Halophila stipulacea<br />
(Biliotti & Abdelahad 1990), across the Suez Canal, but<br />
no damage to the local seagrass meadows has been reported.<br />
Direct or indirect human intervention locally causes most<br />
impacts. However, at the regional or global scale, human<br />
activity also exerts an important impact on seagrass ecosystems.<br />
These effects are remarkably difficult to separate from<br />
Future of seagrass meadows 197<br />
responses to background natural changes in the highly<br />
dynamic coastal ecosystem. These impacts involve the effects<br />
of the realized and predicted climate change (Table 1), and<br />
result from changes in sea level, water temperature, UV irradiance,<br />
and CO 2 concentration.<br />
The mean global sea level has risen about 10<strong>–</strong>25 cm over<br />
the 20th century, which should have generated an average<br />
recession of the global coastline by 10<strong>–</strong>25 m (Bruun 1962),<br />
and, therefore, a large-scale erosion of shallow marine sediments.<br />
There is little doubt that such changes must have<br />
affected seagrasses, which are very sensitive to sediment<br />
erosion (e.g. Marbá & Duarte 1995), although there is little<br />
or no direct observational evidence for these changes. The<br />
effects on seagrasses of seawater warming of 0.3<strong>–</strong>0.6 o C over<br />
the 20th century (Mackenzie 1998) are generally less<br />
evident than those produced by sea-level changes.<br />
Temperature affects many processes that determine seagrass<br />
growth and reproduction, including photosynthesis, respiration,<br />
nutrient uptake, flowering and seed germination<br />
(Short & Neckles 1999). Although observations of increased<br />
flowering frequency of Posidonia oceanica in the<br />
Mediterranean have been tentatively linked to the seawater<br />
temperature increase (Francour et al. 1994), there is little<br />
evidence at present to suggest any impact of increased<br />
temperature as a result of global warming. The temperature<br />
increase may further impact seagrass ecosystems through<br />
effects on other components, such as an increased respiratory<br />
rate of the associated microbial communities.<br />
Stimulation of microbial respiration would further enhance<br />
the problems derived from high organic inputs to seagrass<br />
sediments (Terrados et al. 1999). Summer UV irradiance<br />
has greatly increased at high latitudes, and an increase in<br />
the north-temperate zone is also becoming evident.<br />
Increased UV levels are expected to impact negatively<br />
shallow, particularly intertidal, seagrasses (Dawson &<br />
Dennison 1996). The photosynthetic rates of light-saturated<br />
seagrass leaves are often limited by the availability of<br />
dissolved inorganic carbon, and, since the concentration of<br />
CO 2 in well-mixed, shallow coastal waters is in equilibrium<br />
with the atmosphere, the increase in atmospheric CO 2<br />
concentration by 25%, from 290 ppm to 360 ppm, over the<br />
20th century (Mackenzie 1998), may have led to an increase<br />
in light-saturated seagrass photosynthesis by as much as<br />
20% (recalculated from information in Brouns 1994 and<br />
Beer & Koch 1996). There is, however, little evidence that<br />
such physiological responses have led to observable changes<br />
in seagrass ecosystems at present.<br />
Natural versus anthropogenic influences<br />
Some cause-effect relationships between local seagrass loss<br />
and direct human activities, such as increased nutrient and<br />
organic loading, constructions on the coastline or boating<br />
activity, can be readily demonstrated. For instance, the<br />
loss of Zostera marina in a bay on the Atlantic coast of the<br />
USA has been shown to be closely correlated to housing
198 C.M. Duarte<br />
development (Short & Burdick 1996). However, the link<br />
between seagrass losses and indirect human influences is<br />
more elusive, since the coastal zone is a highly dynamic<br />
ecosystem, where many conditions vary simultaneously.<br />
Disturbances such as strong storms, hurricanes and<br />
typhoons, severely impact seagrass beds, to the point that<br />
they may be essential components of the dynamics of seagrass<br />
meadows (Duarte et al. 1997b). The difficulties of discriminating<br />
between sources of seagrass loss are best illustrated by<br />
example. A wasting disease decimated Zostera marina<br />
meadows in the 1930s on both sides of the Atlantic. The<br />
proximal cause for the loss seems to have been an infection by<br />
a fungus, Labyrinthula zosterae, although it may have affected<br />
Zostera marina meadows that were already stressed (den<br />
Hartog 1987). Hypotheses to account for this widespread loss<br />
also point to natural changes, such as unusual seawater<br />
warming, as possible triggers for the decline (den Hartog<br />
1987). Whether indirect human impacts on global processes<br />
may have played a role remains untested. The causes for the<br />
more recent loss of vast areas of seagrass, mostly Thalassia<br />
testudinum, in Florida Bay (Robblee et al. 1991) continue to be<br />
debated, and include, among others, increased anthropogenic<br />
nutrient loading, the effects of climatic changes involving a<br />
long time interval without hurricanes affecting the area also<br />
causing unusually low freshwater discharge (Fourqurean &<br />
Robblee 1999; Zieman et al. 1999), and the effects of the<br />
increased accumulation of detritus derived from loss of<br />
mega-grazers ( Jackson 2001; Jackson et al. 2001). Difficulties<br />
in experimenting at the appropriate scale of whole meadows<br />
to test these hypotheses have precluded elucidation of<br />
whether the decline was due to human or natural causes, or a<br />
combination of both.<br />
The challenge lies, therefore, in separating local loss<br />
processes, which would suggest a local, likely humanderived,<br />
source of disturbance, from indirect effects, both<br />
natural and anthropogenic, acting at a much larger spatial<br />
scale. The ability to reconstruct the growth history of the<br />
long-lived Mediterranean seagrass Posidonia oceanica allowed<br />
the examination of the coherence of past growth patterns<br />
along the Spanish Mediterranean coast (Marbá & Duarte<br />
1997). This examination revealed coherent growth patterns<br />
in populations 200<strong>–</strong>300 km along the coast, as well as a<br />
general tendency towards a decline in vertical growth, which<br />
indicates a tendency towards sediment erosion (Marbá &<br />
Duarte 1997). These trends were present even in meadows<br />
distant from any identifiable human disturbance, and were<br />
related to decadal changes in climate (Marbá & Duarte<br />
1997).<br />
Hence, despite clear signals of anthropogenic effects on<br />
climate components, the responses of the seagrass ecosystem<br />
are still unclear, probably due to the still modest size of the<br />
changes experienced but also, and perhaps to a greater<br />
extent, to the lack of adequate long-term monitoring systems<br />
allowing the detection of responses in the seagrass<br />
ecosystem.<br />
Future threats<br />
Future threats to seagrass meadows evolve from the development<br />
and expansion of those threats already operating.<br />
Increasing human use of the coastal zone predicates an uncertain<br />
future for seagrass meadows. Coastal populations, both<br />
resident and tourist, continue to grow rapidly, with those in<br />
the Mediterranean doubling at 30-year and 15-year intervals,<br />
respectively (UNEP 1989). Increased human activity in the<br />
coastal zone is associated with increased physical disturbance<br />
of seagrass meadows, through the construction of coastal<br />
infrastructures, and increased boating and shipping.<br />
Aquaculture is growing exponentially worldwide (World<br />
Resources Institute 1998), and parallel depletion of natural<br />
fish populations should lead to a maintenance, or even acceleration,<br />
of the expansion of coastal aquaculture, with<br />
associated potential impacts on seagrass meadows. The<br />
present trend towards increasing per caput water use, the<br />
expansion of waterborne sewage systems, and the growing<br />
use of fertilizers (World Resources Institute 1998) should<br />
lead to increasing nutrient and sewage discharge to coastal<br />
waters, and therefore a further expansion of the already widespread<br />
eutrophication problems. Although sewage treatment<br />
is also growing, the volume treated is growing much more<br />
slowly than that of total sewage and nutrient discharge<br />
(Nixon 1995). Similarly, there are no symptoms that land-use<br />
changes and deforestation are declining in the developing<br />
world, so that associated problems, such as siltation of coastal<br />
waters (see above), are likely to continue to increase in the<br />
future.<br />
The extent of indirect human impacts on seagrass ecosystems<br />
should also increase in the future, due to increasing<br />
emissions of CO 2 and other greenhouse gases. These activities<br />
should lead to an increase in the concentration of CO 2 in<br />
seawater, with a resulting decline in seawater pH (e.g.<br />
Kleypas et al. 1999), increased water temperature, and sealevel<br />
rise, together with an increased frequency of storms and<br />
wave action (Carter 1988; Bijlsma et al. 1996). Whereas an<br />
increase in the concentration of CO 2 in seawater may have<br />
positive effects on seagrass photosynthesis, possibly<br />
enhancing seagrass primary production worldwide, the<br />
decline in seawater pH should negatively affect calcifying<br />
organisms (Kleypas et al. 1999), including those present in<br />
seagrass meadows. The increasing sea level and storm activity<br />
(Carter 1988; Bijlsma et al. 1996) should be conducive to<br />
increasing coastal erosion and subsequent loss of seagrass<br />
meadows. Sea-level rise leads to a recession of the coastline,<br />
which can be roughly estimated as a horizontal recession rate<br />
of 100 times the consolidated sea-level rise (applying Bruun’s<br />
1962 rule of 1 m recession per cm increase in sea level).<br />
Sea-level rise creates, in the long term, new habitat for<br />
seagrass growth, which should lead to colonization of the<br />
inundated land, although the rates of sea-level rise are too<br />
slow to allow direct observation of the progression of the<br />
upslope limit of seagrass distribution; however, consideration<br />
of the sea-level rise of 10<strong>–</strong>25 cm over the 20th century
suggests the inundation (applying Bruun’s 1962 rule) of<br />
10<strong>–</strong>25 m of land, with an equivalent potential upslope<br />
progression of the vegetation. The downslope limit of the<br />
seagrass, which is generally set by the water transparency<br />
(Duarte 1991b), is likely to experience a similar upslope<br />
regression (i.e. 1<strong>–</strong>2.5 m horizontally yr -1 ): this assumes the<br />
transparency to remain constant, so that the bathymetric<br />
range occupied by seagrasses should not necessarily increase<br />
with the colonization of new inundated areas upslope. These<br />
processes introduce a dynamic perspective on the upslope<br />
and downslope distribution of seagrass meadows, which<br />
needs be considered in monitoring programmes, which may<br />
interpret the expected horizontal regression of 1<strong>–</strong>2.5 m per<br />
decade at the downslope limit as decline, when it may well<br />
represent a readjustment of the entire meadow to sea-level<br />
rise.<br />
Consequences of seagrass loss<br />
Seagrass loss leads to a loss of the associated functions and<br />
services in the coastal zone. The consequences of seagrass<br />
loss are well documented, through observations of the<br />
changes in the ecosystem upon large-scale seagrass losses<br />
(Hemminga & Duarte 2000). Seagrass loss involves a shift in<br />
the dominance of different primary producers in the coastal<br />
ecosystem, which can only partially compensate for the loss<br />
of primary production. For instance, the increased planktonic<br />
primary production with increasing nutrient inputs<br />
does not compensate for the lost seagrass production, so that<br />
there is no clear relationship between increased nutrient<br />
loading and ecosystem primary production (Borum 1996).<br />
The loss of the sediment protection offered by the seagrass<br />
canopy enhances sediment resuspension, leading to a further<br />
deterioration of light conditions for the remaining seagrass<br />
plants (Olesen 1996). The extent of resuspension can be so<br />
severe following large-scale losses, such as that experienced<br />
during the Zostera marina wasting disease, that the shoreline<br />
may be altered (Christiansen et al. 1981). The loss of<br />
seagrasses will also involve the loss of the oxygenation of sediment<br />
by seagrass roots, promoting anoxic conditions in the<br />
sediments. Seagrass loss has been shown to result in significant<br />
loss of coastal biodiversity, leading to a modification of<br />
food webs and loss of harvestable resources (Young 1978;<br />
Hemminga & Duarte 2000). In summary, seagrass loss represents<br />
a major loss of ecological as well as economic value to<br />
the coastal ecosystems, and is therefore, a major source of<br />
concern for coastal managers.<br />
IDENTIFIED LONG-TERM TRENDS<br />
Whereas records of change in seagrass ecosystems are few,<br />
the existing reports point to a tendency for widespread<br />
decline in both temperate and tropical ecosystems. There<br />
have been reports of large-scale seagrass decline at over 40<br />
locations (Hemminga & Duarte 2000), 70% of which were<br />
Future of seagrass meadows 199<br />
unambiguously attributable to human-induced disturbance<br />
(Short & Wyllie-Echeverria 1996), and reports of seagrass<br />
loss have multiplied by 10 in the last decade, which can only<br />
partially be accounted for by the increased effort in seagrass<br />
research and monitoring (Duarte 1999). Whereas some<br />
recovery has been demonstrated (e.g. Kendrick et al. 1999;<br />
Preen et al. 1995), most records indicate this to be either nonexistent<br />
or incomplete, so that the perceived long-term trend<br />
of seagrass points to a global decline, the extent of which is<br />
largely unknown. Short and Wyllie-Echevarria (1996) estimated,<br />
extrapolating from the reports available, the area of<br />
seagrass meadows globally lost during the 1990s at about<br />
12 000 km 2 , corresponding to the loss during this period alone<br />
of about 2% of the area originally covered by seagrasses.<br />
There is mounting evidence that Posidonia oceanica, which<br />
was estimated to cover about 50 000 km 2 in the<br />
Mediterranean (Bethoux & Copin-Montégut 1986), has been<br />
suffering widespread decline, at least in the north-west<br />
Mediterranean (Marbá et al. 1996; Pascualini et al. 1999;<br />
Boudouresque et al. 2000). An analysis of the age structure of<br />
Posidonia oceanica meadows in the Spanish Mediterranean<br />
showed a tendency for decline in 57% of the meadows examined<br />
(Marbá et al. 1996). Whereas some of those meadows<br />
still persist, some others have already been lost (C.M.<br />
Duarte, unpublished results 1999). The loss is not evenly<br />
distributed along the Spanish Mediterranean, but is concentrated<br />
in approximately 400 km of coastline, where only<br />
traces of some meadows recorded in the late 1980s remain<br />
and many have been lost already (C.M. Duarte, unpublished<br />
results 1999). Whereas local human impacts are often evident<br />
as causes for the decline of Posidonia oceanica meadows, this<br />
is not always the case, and, together with significant correlations<br />
between decadal growth and climate variation (Marbá &<br />
Duarte 1997), points to climate change effects and possibly<br />
indirect human impacts on the observed decline. For<br />
instance, the gradual decline in Zostera marina growing in the<br />
French coast noticed during the 1980s and 1990s was also<br />
related to elevated sea-surface temperatures observed since<br />
the 1980s (Glémarec 1997). These observations suggest that<br />
seagrass decline and, possibly expansion, may be triggered by<br />
large-scale climate change (Fig. 2). If so, rapidly changing<br />
climatic patterns may lead to major changes in seagrass cover<br />
in the future.<br />
Along the Danish coast, eelgrass has shown an important<br />
net reduction in cover over the 20th century, with important<br />
declines (die-off in the 1930s), rebounds and secondary<br />
declines, in between (P.B. Christensen, D. Krause-Jensen &<br />
J. S. Laursen, unpublished data 2000). These observations<br />
portray eelgrass ecosystems as highly dynamic at time scales<br />
longer than decadal. If undisturbed, seagrass communities<br />
can be rather stable, with records of continuous presence of<br />
specific seagrass beds for several millennia (e.g. Zostera<br />
marina, Reusch et al. 1999; Posidonia oceanica, Mateo et<br />
al.1997), and the extent and time scales of fluctuations in<br />
seagrass meadows may differ greatly across species and<br />
locations.
200 C.M. Duarte<br />
POTENTIAL STATUS IN 2025<br />
The prospective trends in both human forcing and climate<br />
change for the next 25 years suggest that the status of seagrass<br />
ecosystems may somewhat improve in some developed countries,<br />
because legislation is being rapidly implemented to<br />
protect seagrass meadows. The zero loss policy in the USA<br />
and Australia (Coles & Fortes 2001) means that any losses<br />
caused by direct human intervention should be compensated<br />
by the creation, generally by transplanting, of a similar extent<br />
of new seagrass meadow. Indeed, the general public is<br />
becoming increasingly aware of the existence of seagrasses<br />
and their important services and functions in coastal ecosystems,<br />
which can only raise the prospect of implementing<br />
conservation management policies further. Tourism, which<br />
has often acted in the past as a vector of degradation of coastal<br />
ecosystems, is shifting towards a more sustainable perspective<br />
and is becoming, in many countries, an agent promoting<br />
an improved quality of coastal ecosystems.<br />
Regrettably, legal protection against seagrass losses is only<br />
possible where disturbance derives from proximal causes,<br />
and difficulties of assigning responsibility for more diffuse<br />
impacts imply that these cannot be corrected through a<br />
posteriori legal actions. This includes eutrophication-derived<br />
seagrass loss, for nutrient inputs are mostly diffuse in nature.<br />
Nutrient reduction plans are being implemented in most<br />
developed countries, although these seem to have had little<br />
impact on the extent of coastal eutrophication, which is likely<br />
to expand in the future, although hopefully at a reduced rate,<br />
despite efforts to reduce nutrient inputs. Moreover, the<br />
trends outlined above only encompass a fraction of the<br />
world’s nations, where losses over the 20th century have<br />
already been large, and include only a modest fraction of the<br />
world’s coastline and seagrass beds.<br />
The bulk of the extant seagrass meadows are found on the<br />
coastline of developing tropical countries, which are experiencing<br />
the greatest rate of environmental degradation at<br />
present, and will continue to do so in the future. The impacts<br />
on seagrass meadows there will derive from land-use changes,<br />
particularly deforestation, and the associated siltation of<br />
coastal waters and growing nutrient and sewage inputs, as<br />
populations, fertilizer use and sewage implementation<br />
expand, and the impacts of the rapidly-growing aquaculture<br />
operations arise. What the loss will be by the year 2025<br />
remains uncertain, since the total extent of these impacts is as<br />
yet unknown. However, the geographic extent of these major<br />
impacts can be readily delimited to the developing world,<br />
particularly affecting seagrass meadows in South-east Asia,<br />
East Africa, and, to a lesser extent, the Caribbean. The<br />
coastal areas of other developing regions, such as South<br />
America and West Africa, support seagrass meadows that are<br />
much more limited in extent, so that the local impacts will<br />
affect smaller areas.<br />
Oceania is the only region with a major seagrass belt where<br />
losses are expected to be relatively small, due to the enforcement<br />
of zero seagrass loss policies and the small population<br />
(29.5 million) relative to the region’s coastline (30 663 km, i.e.<br />
968 humans km -1 ) compared at the other extreme to Asia<br />
(14 197 humans km -1 ; World Resources Institute 1998). Even<br />
so, the coastal zone near Australian cities has already experienced<br />
a significant loss (e.g. Cambridge & McComb 1984;<br />
Clarke & Kirkman 1989), so that seagrasses are considered<br />
amongst the most stressed of Australian ecosystems<br />
(Kirkman 1992; Kirkman & Kirkman 2000).<br />
Seagrass meadows in both developed and developing<br />
regions will be affected by changes in global processes. The<br />
predicted sea-level rise of about 0.5 cm yr -1 (Mackenzie 1998)<br />
implies a further average regression of the global coastline by<br />
12 m by the year 2025 and therefore important submarine<br />
erosion, which should impact seagrass meadows globally.<br />
Quantitative models relating coastline regression to submarine<br />
erosion and seagrass loss are, however, lacking, so that<br />
sensible predictions cannot be formulated. The sea-level rise<br />
may be faster than anticipated due to fast ice melt in polar<br />
regions, land subsidence, and the release of the fresh water<br />
retained in reservoirs reaching the end of their operative<br />
lives. Reservoirs presently hold an amount of water equivalent<br />
to 7 cm of sea-level rise (Carter 1988), and most of them<br />
will have exceeded their operative lives by the year 2025.<br />
This effect may be compensated by construction of new<br />
reservoirs. Submarine erosion may proceed further than<br />
expected because of the predicted increased frequency of<br />
storms and wave energy in the coastal zone with climate<br />
change (Carter 1988; Bijlsma et al. 1996). For similar reasons<br />
the consequences of the predicted increase of 0.5 o C in global<br />
temperature by the year 2025 (Mackenzie 1998) cannot be<br />
quantitatively formulated, although they are likely to be<br />
minor compared to other sources of change. Based on the<br />
calculations in Brouns (1994) and Beer and Koch (1996) the<br />
elevation of CO 2 concentration to almost 400 ppm by the year<br />
2025 may further stimulate seagrass photosynthesis by<br />
10<strong>–</strong>20%. In addition, Zimmerman et al. (1997) calculated<br />
that this increased photosynthesis may increase the depth<br />
limit of seagrasses, which may lead to an expansion in<br />
seagrass cover.<br />
The limitations of knowledge are evident from the fact<br />
that only qualitative predictions on the status of seagrass<br />
ecosystems by year 2025 are possible, because of (1) undefined<br />
rates of change in most relevant forcing factors, (2) lack<br />
of quantitative dose-response models that predict seagrass<br />
response to changes in environmental factors, and (3) absence<br />
of a reliable knowledge of the present extent of seagrass<br />
ecosystems. Even for responses where quantitative predictions<br />
may be formulated, such as the response of seagrass<br />
photosynthesis to increasing CO 2 concentrations, the forecasts<br />
are suspect, for they assume everything else to be held<br />
constant. The predictions therefore lose reliability when the<br />
plants respond in an integrative manner to multiple simultaneous<br />
changes (Chapin et al. 1987), and when these<br />
responses may be affected by responses of other components<br />
of the complex communities that comprise the seagrass<br />
ecosystem. For instance, the positive effect of increased CO 2
concentrations on seagrasses predicted from photosynthetic<br />
rates may be obscured by similar responses by the epiphytes,<br />
buffering the seagrass response due to shading. The<br />
predicted increase in water temperature may enhance<br />
community respiration, resulting in CO 2 concentrations<br />
higher than predicted if in equilibrium with the atmosphere<br />
in some ecosystems. However, enhanced seagrass respiration<br />
may also partially offset the possible increased photosynthetic<br />
gains. One of the major effects of climate change on terrestrial<br />
vegetation is the shift in vegetation ranges with global<br />
warming. A shift in the range of seagrass species, involving a<br />
displacement of subtropical and tropical species towards<br />
higher latitudes, similarly may be expected. This shift should<br />
not affect a poleward extension of the range of temperate<br />
species, for there is no evidence for temperature-dependence<br />
of the latitudinal limit of seagrasses (Hemminga & Duarte<br />
2000; Duarte et al. 2002). In contrast to terrestrial vegetation,<br />
quantitative predictions on the possible shifts in seagrass<br />
ranges with global warming are, however, not possible at the<br />
moment, since these would be dependent on changes in the<br />
oceanic currents and fronts that are themselves difficult to<br />
forecast. The response of seagrass ecosystems to climate<br />
change integrate partial responses at a number of levels, from<br />
physiological to organismal and ecosystem levels, which often<br />
act in opposite directions and interact with one another. The<br />
complex nature of the responses renders our capacity to reliably<br />
predict the response of seagrass ecosystems to climate<br />
change weak.<br />
Provided the recent trends and anticipated threats identified,<br />
it is clear that the most likely scenario is one of major<br />
loss of seagrass ecosystems, although quantitative forecasts<br />
cannot be issued. In a scenario of present and future loss of<br />
seagrass ecosystems, the question of the reversibility of<br />
decline becomes critical, particularly because loss and<br />
recovery of seagrass may be also associated with loss and<br />
recovery of important associated fauna and resources. The<br />
partial recovery of North Atlantic eelgrass populations<br />
affected by the wasting disease within a couple of decades<br />
(Short & Wyllie-Echeverria 1996) shows that recovery from<br />
certain impacts may be possible. However, this is only so<br />
provided that suitable conditions to support seagrass growth<br />
are re-established. Yet, the forcing factors identified above as<br />
likely to cause seagrass decline by year 2025 are forecasted to<br />
increase even further beyond that date so that global-scale<br />
seagrass recovery may not even occur within the present<br />
century. Whereas recovery is possible for species with fast<br />
growth and/or sexual output, such as Halophila spp. and<br />
Enhalus acoroides (Hemminga & Duarte 2000), recovery is<br />
slow in slow-growing species relying largely on clonal expansion,<br />
like the Mediterranean Posidonia oceanica, where<br />
colonization is thought to take place over several centuries<br />
(Duarte 1995). In such species, present and future loss will<br />
therefore affect many generations to come. Recovery may<br />
involve a shift in species composition, towards either fastgrowing,<br />
pioneer species, such as Halophila and Halodule<br />
spp. (Marbá & Duarte 1998), and/or species with high repro-<br />
Future of seagrass meadows 201<br />
ductive outputs, like Enhalus acoroides in the Indo-Pacific<br />
region (Duarte et al. 1997a).<br />
CONCLUSIONS AND MANAGEMENT<br />
Provided the largely qualitative nature of the possible predictions,<br />
the future of seagrass ecosystems remains uncertain.<br />
However, since their present status is already plagued with<br />
problems, and the future global environmental scenario is<br />
one of progressive change, the most likely prospect is a<br />
mounting global seagrass decline throughout the 21st<br />
century. The main threats to seagrass ecosystems on a large<br />
scale will continue to derive from human activity, both<br />
through direct disruption of the coastal zone, changes in land<br />
use and inputs of silt, nutrients and sewage to the coastal<br />
zone, as well as diffuse effects of human activity on climate<br />
(Figs. 1 and 2). Sea-level rise, in particular, is projected to be,<br />
and may already be (e.g. Marbá & Duarte 1997), a major<br />
cause of seagrass decline (Hemminga & Duarte 2000).<br />
Despite this alarm, knowledge of seagrass ecosystems is<br />
still insufficient to even document their possible present and<br />
likely future decline. The 600 000 km2 area global cover of<br />
seagrasses (Charpy-Roubaud & Sournia 1990) is only a best<br />
guess, derived from the application of a few rules of thumb,<br />
in contrast with the relatively adequate knowledge on the area<br />
cover of other key marine ecosystems that can be observed<br />
more easily from space (e.g. mangroves, coral reefs).<br />
Knowledge of the present area seagrasses cover globally is<br />
uncertain, and the situation is probably not much better for<br />
the bulk of the individual nations with an extended coastline.<br />
These uncertainties preclude any reliable quantitative forecast<br />
of future seagrass cover, and only qualitative forecasts<br />
can be formulated based on established trends and expected<br />
changes, supported by anecdotal evidence and opportunistic<br />
observations.<br />
Remote sensing techniques using optical or acoustic<br />
methods allow the monitoring of the surface seagrasses<br />
occupy, allowing the detection of seagrass regression or<br />
expansion (Kendrick et al. 1999; McKenzie et al. 2001) but<br />
cannot resolve changes in the internal density of the vegetated<br />
area. There is therefore a need to improve on the<br />
capacity to monitor seagrass remotely, such as improvements<br />
in acoustic techniques and use of hyperespectral imagerybased<br />
tools to discriminate between seagrass and other<br />
vegetated sediments. Monitoring programmes, typically<br />
based on the assessment of cover and shoot density along<br />
transects and quadrats, generally have an associated error<br />
greater than 30% about the mean (Heidelbaugh & Nelson<br />
1996), making it difficult to reliably detect changes; a reliable<br />
assessment of decline may only be possible for losses already<br />
amounting to as much as 50<strong>–</strong>80%. The low power of monitoring<br />
techniques implies that most monitoring programmes<br />
can only detect a reliable tendency towards seagrass loss<br />
when the seagrass meadows monitored have already experienced<br />
substantial damage. There is, therefore, an urgent need<br />
to design more effective monitoring approaches, capable of
202 C.M. Duarte<br />
detecting losses of 10% or less, as well as to develop early<br />
warning indicators of decline. Repeated census of marked<br />
shoots in plots may allow such precision for some seagrass<br />
species (e.g. Short & Duarte 2001). Use of reconstruction<br />
techniques to assess past seagrass demography (cf. Duarte et<br />
al. 1994) in monitoring programmes has allowed assessment<br />
of local tendencies for decline or expansion and the examination<br />
of the overall demographic balance of very large<br />
seagrass ecosystems (Peterson & Fourqurean 2001). Yet,<br />
even an optimistic analysis clearly indicates that monitoring<br />
efforts cannot possibly encompass but a minimum (< 0.01%)<br />
fraction of the world’s seagrass extent, although the<br />
implementation of seagrass monitoring programmes has<br />
increased over the past decades. The information derived<br />
from individual monitoring efforts could, however, be used<br />
to derive knowledge of trends at larger spatial scales than that<br />
encompassed by the individual monitoring efforts if these<br />
were networked into a coherent programme. The implementation<br />
of coordinated monitoring networks at the national,<br />
regional and global scales would prove instrumental to diagnosis<br />
of the large-scale status and trends of seagrass<br />
ecosystems.<br />
Management practices to address problems affecting<br />
seagrass ecosystems once detected are also insufficiently<br />
developed, for management intervention often fails to revert<br />
an ongoing seagrass decline (e.g. Delgado et al. 1999). The<br />
easiest impacts to regulate are those pertaining to direct<br />
mechanical effects on seagrass ecosystems, as well as some<br />
improvement of seagrass health following regulation of<br />
anchoring in recreational areas (e.g. Cabrera National Park,<br />
Spain; N. Marbá, C.M. Duarte, M. Holmer, R. Martínez, G.<br />
Basterretxea, A. Orfila, A. Jordi & J. Tintoré, personal<br />
communication 2002). Impacts related to changes in water<br />
quality are, however, more difficult to alleviate, although<br />
some successes have been reported, such as the reversal of<br />
seagrass losses following wastewater treatment on the French<br />
Society<br />
- Improved education<br />
- Improved awareness<br />
Managerial<br />
- Monitoring programmes<br />
- Accurate maps<br />
- Impact prevention<br />
- Precautionary principle<br />
- Networking<br />
- Restoration<br />
- Legislation<br />
Mediterranean coast (Pergent-Martini & Pascualini 2000).<br />
Empirical (e.g. Duarte 1991b) aswell as mechanistic (e.g.<br />
Gallegos 1994) models relating water quality to seagrass cover<br />
have been developed, which can be used to build scenarios of<br />
the effectiveness of different management strategies to<br />
improve water quality, and have also led to the development<br />
of metrics based on seagrass performance to assess water<br />
quality (Dennison et al. 1993). Direct management intervention<br />
in the proximal, in addition to the ultimate, causes of<br />
seagrass decline may be possible, such as intervention on<br />
sediment processes. For instance, iron additions to carbonate<br />
sediments may help prevent oxygen consumption by<br />
sulphide oxidation, possibly alleviating seagrass stress from<br />
anoxic sediments with high sulphide concentrations (cf.<br />
Hemminga & Duarte 2000; Chambers et al. 2001). Such<br />
direct intervention practices, however, must be based on<br />
robust knowledge of the proximal causes of seagrass decline,<br />
which requires further research emphasizing experimental<br />
tests of mechanisms and predictive models.<br />
Transplanting techniques to restore seagrasses have had<br />
mixed success (Fonseca 1992; Kirkman 1992), however,<br />
seagrass transplantation is too costly to be implemented at a<br />
large scale, which is a particularly critical drawback provided<br />
the fact that the most important losses are expected in developing<br />
countries. Hence, even though nationwide mangrove<br />
afforestation programmes have been implemented in some<br />
developing countries such as Thailand (Aksornkoae 1993),<br />
similar initiatives for seagrass meadows are as yet not feasible.<br />
In addition, seagrass transplantation programmes often<br />
require damage to a donor population, thereby reducing the<br />
value of transplantation as a management option.<br />
The threats and stresses to seagrass ecosystems are so<br />
diverse that management practices are unlikely to be effective<br />
unless accompanied by changing public attitudes and awareness<br />
(Fig. 3). Hence, the mid- to long-term conservation of<br />
seagrass ecosystems must be based on an adequate education<br />
Improving seagrass ecosystems<br />
Scientific<br />
- Increased knowledge<br />
- Improved predictive<br />
capacity<br />
- Interdisciplinary<br />
approaches<br />
- Improved monitoring<br />
technologies<br />
- Efficient restoration and<br />
intervention techniques<br />
Figure 3 Cooperative elements required to prevent present trend towards seagrass decline and efficiently conserve seagrass ecosystems.
of the public as to the nature of seagrasses, the functions they<br />
perform in nature, and the services these functions provide to<br />
society, and on the formulation and dissemination of best<br />
management practices to conserve seagrass ecosystems, all of<br />
this based on solid, interdisciplinary understanding of<br />
seagrass ecology and how to manage the health of seagrass<br />
ecosystems (Fig. 3). Education campaigns must, therefore,<br />
occupy a prominent place on the agenda for the conservation<br />
of seagrass ecosystems (Kirkman & Kirkman 2000), mobilizing<br />
seagrass scientists, managers and environmental<br />
educators to convey effectively the importance of seagrass<br />
ecosystems and the threats to them. Public awareness alone<br />
cannot be effective unless accompanied by sufficient understanding<br />
of the causes of stress to seagrass, and how these are<br />
integrated at the different levels, from physiological to demographic<br />
and landscape scales, to yield responses to possible<br />
intervention alternatives. In addition, our capacity to predict<br />
seagrass recovery from stress must be developed in order to<br />
allow reliable forecasts to be issued. The future of seagrass<br />
ecosystems will be largely dictated, therefore, by the social<br />
and scientific responses to the challenges ahead.<br />
ACKNOWLEDGEMENTS<br />
The writing of this paper was supported by the European<br />
Commission Managing and Monitoring of Seagrass Beds<br />
project (contract # EVK3-CT-2000-00044). I thank G.<br />
Kendrick and G. Harris for valuable discussion, J. Borum,<br />
D.I. Walker, J. Fourqurean and N.V.C. Polunin for valuable<br />
revisions.<br />
References<br />
Aksornkoae, S. (1993) Ecology and Management of Mangroves.<br />
Bangkok, Thailand: IUCN.<br />
Aragones, L. & Marsh, H. (2000) Impact of dugong grazing and<br />
turtle cropping on tropical seagrass communities. Pacific<br />
Conservation Biology 5: 277<strong>–</strong>288.<br />
Bach, S.S., Borum, J., Fortes, M.D. & Duarte, C.M. (1998) Species<br />
composition and plant performance of mixed seagrass beds along<br />
a siltation gradient at Cape Bolinao, the Philippines. Marine<br />
Ecology Progress Series <strong>17</strong>4: 247<strong>–</strong>256.<br />
Beer, S. & Koch, E. (1996) Photosynthesis of seagrasses vs. marine<br />
macroalgae in globally changing CO2 environments. Marine<br />
Ecology Progress Series 141: 199<strong>–</strong>204.<br />
Bethoux, J.P. & Copin-Montégut, G. (1986) Biological fixation of<br />
atmospheric nitrogen in the Mediterranean Sea. Limnology and<br />
Oceanography 31: 1353<strong>–</strong>8.<br />
Bijlsma, L., Ehler, C.N., Kelin, R.J.L., Kulshrestha, S.M.,<br />
McLean, R.F., Mimumra, N., Nichols, R.J., Nurse, L.A., Pérez<br />
Nieto, H., Stakhiv, E.Z., Turner, K. & Warrick, R.A. (1996)<br />
Coastal zones and small islands. In: Climate Change 1995: Impacts,<br />
Adaptations and Mitigation of Climate Change, ed.<br />
Intergovernmental Panel on Climate Change, pp. 289<strong>–</strong>324.<br />
Cambridge, UK: Cambridge University Press.<br />
Biliotti, A. & Abdelahad, N. (1990) Halophila stipulacea (Frossk.)<br />
Aschers. (Hydrocharticaeae): éspèce nouvelle pour l’Italie.<br />
Posidonia Newsletter 3: 23<strong>–</strong>26.<br />
Future of seagrass meadows 203<br />
Borum, J. (1996) Shallow waters and land/sea boundaries. In:<br />
Eutrophication in Coastal Marine Ecosystems, ed. K. Richardson &<br />
B.B. Jorgensen, pp. <strong>17</strong>9<strong>–</strong>203. Washington DC, USA: American<br />
Geophysical Union.<br />
Borum, J. & Sand-Jensen, K. (1996) Is total primary production in<br />
shallow coastal marine waters stimulated by nitrogen loading?<br />
Oikos 76: 406<strong>–</strong>10.<br />
Boudouresque, C.F., Charbonel, E., Meinesz, A., Pergent, G.,<br />
Pergent-Martini, C., Cadiou, G., Bertrandy, M.C., Foret, P.,<br />
Ragazzi, M. & Rico-Raimondino, V. (2000) A monitoring network<br />
based on the seagrass Posidonia oceanica in the Northwestern<br />
Mediterranean Sea. Biologia Marina Mediterranea 7: 328<strong>–</strong>331.<br />
Brouns, J.J. (1994) Seagrasses and climate change. In: Impacts of<br />
Climate Change on Ecosystems and Species: Marine and Coastal<br />
Systems, ed. J.C. Pernetta, R. Leemans, D. Elder & S. Humphrey,<br />
pp. 59<strong>–</strong>71. Gland, Switzerland: IUCN.<br />
Bruun, P. (1962) Sea level rise as a cause of shore erosion. Journal<br />
Waterways Harbours Division, Proceedings American Society Civil<br />
Engineering 88: 1<strong>17</strong><strong>–</strong>130.<br />
Burt, L. (2001) Evolutionary stasis, contraints, and other terminology<br />
describing evolutionary patterns. Biologial Journal of the<br />
Linnean Society 72: 509<strong>–</strong>5<strong>17</strong>.<br />
Cambridge, M.L. & McComb, A.J. (1984) The loss of seagrasses in<br />
Cockburn Sound, Western Australia. I. The time course and<br />
magnitude of seagrass decline in relation to industrial development.<br />
Aquatic Botany 24: 269<strong>–</strong>285.<br />
Cancemi, G., De Falco, G. & Pergent, G. (2000) Impact of a fish<br />
farming facility on a Posidonia oceanica meadow. Biologia Marina<br />
Mediterranea 7: 341<strong>–</strong>344.<br />
Carlson, P.R., Yabro, L.A. & Barber, T.R. (1994) Relationship of<br />
sediment sulfide to mortality of Thalassia testudinum in Florida<br />
Bay. Marine Science Bulletin 54: 733<strong>–</strong>746.<br />
Carter, R.W.G. (1988) Coastal Environments. London, UK:<br />
Academic Press.<br />
Cebrián, J. & Duarte, C.M. (1997) Patterns in leaf herbivory on<br />
seagrasses. Aquatic Botany 60: 67<strong>–</strong>82.<br />
Chambers, R.M., Fourqurean, J.W., Macko, S.A., Hoppenot, R.<br />
(2001) Biogeochemical effects of iron availability on primary<br />
producers in a shallow marine carbonate sediment. Limnology and<br />
Oceanography 46: 1278<strong>–</strong>1286.<br />
Chapin, F.S., III, Bloom, A.J., Field, C.B. & Waring, R.H. (1987)<br />
Plant responses to multiple environmental factors. BioScience 37:<br />
49<strong>–</strong>87.<br />
Charpy-Roubaud, C. & Sournia, A. (1990) The comparative estimation<br />
of phytoplanktonic and microphytobenthic production in<br />
the oceans. Marine Microbial Food Webs 4: 31<strong>–</strong>57.<br />
Christiansen, C., Christoffersen, H., Dalsgaard, J. & Norberg, R.<br />
(1981) Coastal and nearshore changes correlated with die-back in<br />
eelgrass (Zostera marina L.). Sedimentary Geology 28: 163<strong>–</strong>73.<br />
Clarke, S.M. & Kirkman, H. (1989) Seagrass dynamics. In:<br />
Seagrasses: a Treatise on the Biology of Seagrasses with Special<br />
Reference to the Australian Region, ed. A.W.D. Larkum, A.J.<br />
McComb & S.A. Shepherd, pp. 304<strong>–</strong>34, Amsterdam, the<br />
Netherlands: Elsevier.<br />
Cloern, J.E. (2001) Our evolving conceptual model of the coastal<br />
eutrophication problem. Marine Ecology Progress Series 210:<br />
223<strong>–</strong>253.<br />
Cole, J.J., Peierls, B.L., Caraco, N.F. & Pace, M.L. (1993) Nitrogen<br />
loading of rivers as a human-driven process. In: Humans as<br />
Components of Ecosystems. ed. M.J. McDowell & S.T.A. Rickett<br />
pp. 141<strong>–</strong>157. Springer-Verlag.
204 C.M. Duarte<br />
Coles, R. & Fortes, M. (2001) Protecting seagrass <strong>–</strong> approaches and<br />
methods. In: Global Seagrass Research Methods, ed. F.T. Short &<br />
R.G. Coles, pp. 445<strong>–</strong>463. Amsterdam, the Netherlands: Elsevier.<br />
Costanza, R., d’Arge, R., De Groot, R., Fraber, S., Grasso, M.,<br />
Hannon, B., Limburg, K., Naeem, S., O’Neill, R.V., Paruelo, J.,<br />
Raskin, R.G., Sutton, P. & Van den Belt, M. (1997) The value of<br />
the world’s ecosystem services and natural capital. Nature 387:<br />
253<strong>–</strong>60.<br />
Dawson, S.P. & Dennison, W.C. (1996) Effects of ultraviolet and<br />
photosynthetically active radiation on five seagrass species.<br />
Marine Biology 125: 629<strong>–</strong>638.<br />
De Iongh, H.H. (1996). Plant-herbivore interactions between<br />
seagrasses and dugongs in a tropical small island ecosystem.<br />
Ph.D. Thesis, Catholic University Nijmegen, the Netherlands.<br />
De Villéle, X. & Verlaque, M. (1995) Changes and degradation in a<br />
Posidonia oceanica bed invaded by the introduced tropical alga<br />
Caulerpa taxifolia in the north western Mediterrnean. Botanica<br />
Marina 38: 79<strong>–</strong>87.<br />
Delgado, O., Grau, A., Pou, S., Riera, F., Massuti, C., Zabala, M.<br />
& Ballesteros, E. (1997) Seagrass regression caused by fish<br />
cultures in Fornells Bay, Menorca, western Mediterranean.<br />
Oceanologica Acta 20: 557<strong>–</strong>63.<br />
Delgado, O., Ruiz, J., Pérez, M., Romero, J. & Ballesteros, E. (1999)<br />
Effects of fish farming on seagrass (Posidonia oceanica) in a<br />
Mediterranean bay: seagrass decline after organic loading cessation.<br />
Oceanologica Acta 22: 109<strong>–</strong><strong>17</strong>.<br />
den Hartog, C. (1970) The Seagrasses of the World. Amsterdam, the<br />
Netherlands: North Holland.<br />
den Hartog, C. (1987) ‘Wasting disease’ and other dynamic<br />
phenomena in Zostera beds. Aquatic Botany 27: 3<strong>–</strong>14.<br />
den Hartog, C. (1994) Suffocation of a littoral Zostera bed by<br />
Enteromorpha radiata. Aquatic Botany 47: 21<strong>–</strong>8.<br />
Dennison, W.C., Orth, R.J., Moore, K.A., Stevenson, J.C., Carter,<br />
V., Kollar, S., Bergstrom, P.W. & Batiuk, R. (1993) Assessing<br />
water quality with submersed aquatic vegetation. Bioscience 43:<br />
86<strong>–</strong>91.<br />
Domning, D.P. (2001) <strong>Sirenian</strong>s, seagrasses, and Cenozoic<br />
ecological change in the Caribbean. Palaeogeography,<br />
Palaeoclimatology, Palaeoecology 166: 27<strong>–</strong>50.<br />
Duarte, C.M. (1990) Seagrass nutrient content. Marine Ecology<br />
Progress Series 67: 201<strong>–</strong>207.<br />
Duarte, C.M. (1991a) Allometric scaling of seagrass form and<br />
productivity. Marine Ecology Progress Series 77: 289<strong>–</strong>300.<br />
Duarte, C.M. (1991b) Seagrass depth limits. Aquatic Botany 40:<br />
363<strong>–</strong>377.<br />
Duarte, C.M. (1995) Submerged aquatic vegetation in relation to<br />
different nutrient regimes. Ophelia 41: 87<strong>–</strong>112.<br />
Duarte, C.M. (1999) Seagrass ecology at the turn of the millennium:<br />
challenges for the new century. Aquatic Botany 65: 7<strong>–</strong>20.<br />
Duarte, C.M. (2001) Seagrass ecosystems. In: Encyclopedia of<br />
Biodiversity, Volume 5, ed. S.L. Levin, pp. 255-268. San Diego,<br />
USA: Academic Press.<br />
Duarte, C.M. & Cebrián, J. (1996) The fate of marine autotrophic<br />
production. Limnology and Oceanography 41: <strong>17</strong>58<strong>–</strong>66.<br />
Duarte, C.M. & Chiscano, C.L. (1999) Seagrass biomass and<br />
production: a reassessment. Aquatic Botany 65: 159<strong>–</strong><strong>17</strong>4.<br />
Duarte, C.M., Marbà, N., Agawin, N.S.R., Cebrián, J., Enríquez,<br />
S., Fortes, M.D., Gallegos, M.E., Merino, M., Olesen, B., Sand-<br />
Jensen, K., Uri, J. & Vermaat, J. (1994) Reconstruction of<br />
seagrass dynamics: age determinations and associated tools for<br />
the seagrass ecologist. Marine Ecology Progress Series 107:<br />
195<strong>–</strong>209.<br />
Duarte, C.M., Martínez, R. & Barrón, C. (2002) Biomass, production<br />
and rhizome growth near the northern limit of seagrass<br />
(Zostera marina L.) distribution. Aquatic Botany (in press).<br />
Duarte, C.M., Uri, J., Agawin, N.S.R., Fortes, M.D., Vermaat, J.E.<br />
& Marbá, N. (1997a) Flowering frequency of Philippine<br />
seagrasses. Botanica Marina 40: 497<strong>–</strong>500.<br />
Duarte, C.M., Terrados, J., Agawin, N.S.W., Fortes, M.D., Bach,<br />
S. & Kenworthy, W.J. (1997b) Response of a mixed Philippine<br />
seagrass meadow to experimental burial. Marine Ecology Progress<br />
Series 147: 285<strong>–</strong>294.<br />
Enríquez S., Duarte, C.M. & Sand-Jensen, K. (1993) Patterns in<br />
decomposition rates among photosynthetic organisms: the<br />
importance of detritus C:N:P content. Oecologia 94: 457<strong>–</strong>471.<br />
European Environmental Agency (1999) State and Pressures of the<br />
Marine and Coastal Mediterranean Environment. Luxembourg:<br />
European Environmental Agency.<br />
Fonseca, M.S. (1992) Restoring seagrass systems in the United<br />
States. In: Restoring the Nation’s Marine Environment, ed. G.W.<br />
Thayer, pp. 79<strong>–</strong>110. College Park, Maryland, USA: Maryland<br />
Sea Grant College.<br />
Foundation for Environmental Conservation (2001) An introduction<br />
to long-term environmental trends [WWW document]. URL<br />
http://www.ncl.ac.uk/icef<br />
Fourqurean, J.W. & Robblee, M.B. (1999) Florida Bay: a history of<br />
recent ecological changes. Estuaries 22: 345<strong>–</strong>357.<br />
Francour, P., Bouderesque, C.F., Harmelin, J.G., Harmelin-<br />
Vivien, M.L. & Quignard, J.P. (1994) Are the Mediterranean<br />
waters becoming warmer: information from biological indicators.<br />
Marine Pollution Bulletin 28: 523<strong>–</strong>526.<br />
Gacia, E. & Duarte, C.M. (2001) Elucidating sediment retention by<br />
seagrasses: sediment deposition and resuspension in a<br />
Mediterranean (Posidonia oceanica) meadow. Estuarine Coastal<br />
and Shelf Science 52: 505<strong>–</strong>514.<br />
Gallegos, C.L. (1994) Refining habitat requirements of submersed<br />
aquatic vegetation: role of optical models. Estuaries <strong>17</strong>: 187<strong>–</strong>199.<br />
Glémarec, M. (1979) Les fluctuations temporelles des peuplements<br />
benthiques liées aux fluctuations climatiques. Oceanologica Acta 2:<br />
365<strong>–</strong>71.<br />
Harrison, P.G. (1989) Detrital processing in seagrass systems: a<br />
review of factors affecting decay rates, remineralization and detritivory.<br />
Aquatic Botany 23: 263<strong>–</strong>288.<br />
Hauxwell, J., Cebrián, J., Furlong, C., Valiela, I. (2001) Macroalgal<br />
canopies contribute to eelgrass (Zostera marina) decline in<br />
temperate estuarine ecosystems. Ecology 82: 1007<strong>–</strong>1022.<br />
Heck, K.L., Pennock, J.R., Valentine, J.F., Cohen, L.D. & Sklenar,<br />
S.A. (2000) Effects of nutrient enrichment and small predator<br />
density on seagrass ecosystems: an experimental assessment.<br />
Limnology Oceanography 45: 1041<strong>–</strong>1057.<br />
Heidelbaugh, W.S. & Nelson, W.G. (1996) A power analsis of<br />
methods for assessment of change in seagrass cover. Aquatic<br />
Botany 53: 227<strong>–</strong>233.<br />
Hein, M., Pedersen, M.F. & Sand-Jensen, K. (1995) Size-dependent<br />
nitrogen uptake in micro- and macroalgae. Marine Ecology<br />
Progress Series 118: 247<strong>–</strong>253.<br />
Hemminga, M. & Duarte, C.M. (2000) Seagrass Ecology.<br />
Cambridge, UK: Cambridge University Press.<br />
Independent World Commission on the Oceans (1998) The Ocean.<br />
Our Future. Cambridge, UK: Cambridge University Press.<br />
Jackson, J.B.C. (2001) What was natural in the coastal ocean?<br />
Proceedings of the National Academy of Sciences USA 98:<br />
5411<strong>–</strong>5418.<br />
Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A.,
Botsford, L.W., Bourque, B.J., Bradbury, R.H., Cooke, R.,<br />
Erlandson, J., Estes, J.A., Hughes, T.P., Kidewell, S., Lange,<br />
C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H.,<br />
Steneck, R.S., Tegner, M.J. & Warner, R. (2001) Historical overfishing<br />
and the recent collapse of coastal ecosystems. Science 293:<br />
629<strong>–</strong>638.<br />
Jaubert, J.M., Chisholm, J.R.M., Ducrort, D., Ripley, H.T., Roy,<br />
L. & Passeron-Seitre, G. (1999) No deleterious alterations in<br />
Posidonia beds in the Bay of Menton (France) eight years after<br />
Caulerpa taxifolia colonization. Journal of Phycology 35:<br />
1113<strong>–</strong>1119.<br />
Jorgensen, B.B. (1996) Material flux in the sediment. In:<br />
Eutrophication in Coastal Marine Ecosystems, ed. K. Richardson &<br />
B.B. Jorgensen, pp. 115<strong>–</strong>135. Washington DC, USA: American<br />
Geophysical Union.<br />
Kendrick, G.A., Eckersley, J. & Walker, D.I. (1999) Landscapescale<br />
changes in seagrass distribution over time: a case study from<br />
Success Bank, Western Australia. Aquatic Botany 65: 293<strong>–</strong>309.<br />
Kirkman, H. (1992) Large-scale restoration of seagrass meadows.<br />
In: Restoring the Nation’s Marine Environment, ed. G.W. Thayer,<br />
pp. 111<strong>–</strong>140. College Park, Maryland, USA: Maryland Sea Grant<br />
College.<br />
Kirkman, H. & Kirkman, J.A. (2000) The management of seagrasses<br />
in South East Asia and Australia. Biologia Marina Mediterranea 7:<br />
305<strong>–</strong>319.<br />
Kleypas, J. A., Buddemeier, R. W., Archer, D., Gattuso, J.-P.,<br />
Langdon, C. & Opdyke, B. N. (1999) Geochemical consequences<br />
of increased atmospheric CO 2 on coral reefs. Science 284:<br />
118<strong>–</strong>120.<br />
Koch, E.W. (2001) Beyond light: physical, geological and geochemical<br />
parameters as possible submersed aquatic vegetation habitat<br />
requirements. Estuaries 24: 1<strong>–</strong><strong>17</strong>.<br />
Mackenzie, F.T. (1998) Our Changing Planet. Second edition.<br />
Upper Saddle River, NJ, USA: Prentice Hall.<br />
Marbà, N. & Duarte, C.M. (1995) Coupling of seagrass (Cymodocea<br />
nodosa) patch dynamics to subaqueous dune migration. Journal of<br />
Ecology 83: 381<strong>–</strong>389.<br />
Marbá, N. & Duarte, C.M. (1997) Interannual changes in seagrass<br />
(Posidonia oceanica) growth and environmental change in the<br />
Spanish Mediterranean littoral. Limnology and Oceanography 42:<br />
800<strong>–</strong>810.<br />
Marbá, N. & Duarte, C.M. (1998) Rhizome elongation and seagrass<br />
clonal growth. Marine Ecology Progress Series <strong>17</strong>4: 269<strong>–</strong>280.<br />
Marbá, N., Duarte, C.M., Cebrián, J., Enríquez, S., Gallegos,<br />
M.E., Olesen, B. & Sand-Jensen, K. (1996) Growth and population<br />
dynamics of Posidonia oceanica on the Spanish<br />
Mediterranean coast: elucidating seagrass decline. Marine Ecology<br />
Progress Series 137: 203<strong>–</strong>213.<br />
Mateo, M.A., Romero, J., Pérez, M., Littler, M.M. & Littler, D.S.<br />
(1997) Dynamics of millenary organic deposits resulting from the<br />
growth of the Mediterranean seagrass Posidonia oceanica.<br />
Estuarine Coastal Shelf Science 44: 103<strong>–</strong>110.<br />
McKenzie, L.J., Finkbeiner, M.A. & Kirkman, H. (2001) Methods<br />
for mapping seagrass distribution. In: Global Seagrass Research<br />
Methods, ed. F.T. Short & R.G. Coles, pp. 101<strong>–</strong>121. Amsterdam,<br />
the Netherlands: Elsevier.<br />
Medina, J.R., Tintoré, J. & Duarte, C.M. (2001) Las praderas de<br />
Posidonia y la regeneración de playas. Revista de Obras Públicas<br />
3409: 31<strong>–</strong>43.<br />
Meinesz, A. & Hesse, B. (1991) Introduction et invasion de l’algue<br />
Caulerpa taxifolia en Méditerranée nord-occidentale.<br />
Oceanologica Acta 14: 415<strong>–</strong>426.<br />
Future of seagrass meadows 205<br />
Menzies, R.J., Zaneveld, J.S. & Pratt, R.M. (1967) Transported<br />
turtle grass as a source of organic enrichment of abyssal sediments<br />
off North Carolina. Deep-Sea Research 14: 111<strong>–</strong>2.<br />
Miliman, J.D. & Meade, R.H. (1993) World-wide delivery of river<br />
sediment to the oceans. Journal of Geology 91: 1<strong>–</strong>21.<br />
Nakaoka, M. & Aioi, K. (1999) Growth of the seagrass Halophila<br />
ovalis at dugong trails compared to existing within-patch variation<br />
in a Thailand intertidal flat. Marine Ecology Progress Series<br />
184: 97<strong>–</strong>103.<br />
Nixon, S.W. (1995) Coastal marine eutrophication: a definition,<br />
social causes, and future concerns. Ophelia 41: 199<strong>–</strong>219.<br />
Nixon, S.W., Ammerman, J., Atkinson, L., Berounsky, V., Billen,<br />
G., Boicourt, W., Boynton, W., Church, T., DiToro, D.,<br />
Elmgren, R., Garber, J., Giblin, A., Jahnke, R., Owens, N.,<br />
Pilson, M.E.Q. & Seitzinger, S. (1996) The fate of nitrogen and<br />
phosphorus at the land-sea margin of the North Atlantic Ocean.<br />
Biogeochemistry 35: 141<strong>–</strong>180.<br />
Nienhuis, P.H., De Bree, B.H.H., Herman, P.M.J., Holland,<br />
A.M.B., Verschuure, J.M. & Wessel, E.G.J. (1996) Twenty five<br />
years of changes in the distribution and biomass of eelgrass,<br />
Zostera marina, in Grevelingen Lagoon, The Netherlands.<br />
Netherlands Journal of Aquatic Ecology 30: 107<strong>–</strong><strong>17</strong>.<br />
Ochieng, C.A. & Erftemeijer, P.L.A. (1999) Accumulation of<br />
seagrass beach cast along the Kenyan coast: a quantitative assessment.<br />
Aquatic Botany 65: 221<strong>–</strong>238.<br />
Olesen, B. (1996) Regulation of light attenuation and eelgrass<br />
Zostera marina depth distribution in a Danish embayment. Marine<br />
Ecology Progress Series 134: 187<strong>–</strong>94.<br />
Olesen, B. (1999) Reproduction in Danish eelgrass (Zostera marina<br />
L.) stands: size-dependence and biomass partitioning. Aquatic<br />
Botany 65: 209<strong>–</strong>219.<br />
Paerl, H.W. (1995) Coastal eutrophication in relation to atmospheric<br />
nitrogen deposition: current perspectives. Ophelia 41:<br />
237<strong>–</strong>259.<br />
Pascualini, V., Pergent-Martini, C. & Pergent, G. (1999)<br />
Environmental impact identification along the Corsican coast<br />
(Mediterranean sea) using image processing. Aquatic Botany 65:<br />
311<strong>–</strong>320.<br />
Pedersen, O., Borum, J., Duarte, C.M. & Fortes, M.D. (1998)<br />
Oxygen dynamics in the rhizosphere of Cymodocea rotundata.<br />
Marine Ecology Progress Series 169: 283<strong>–</strong>288.<br />
Pergent-Martini, C. & Pascualini, V. (2000) Seagrass population<br />
dynamics before and after the setting up of a wastewater treatment<br />
plant. Biologia Marina Mediterranea 7: 405<strong>–</strong>408.<br />
Pergent, G., Mendez, S., Perget-Martini, C. & Pascualini, V. (1999)<br />
Preliminary data on the impact of fish farming facilities on<br />
Posidonia oceanica meadows in the Mediterranean. Oceanologica<br />
Acta 22: 95<strong>–</strong>107.<br />
Peterson, B.J. & J.W. Fourqurean (2001) Large-scale patterns in<br />
seagrass (Thalassia testudinum) demographics in south Florida.<br />
Limnology and Oceanography 46: 1077<strong>–</strong>1090.<br />
Poiner, I.R., Walker, D.I. & Coles, R.G. (1989) Regional studies <strong>–</strong><br />
seagrasses of tropical Australia. In: Biology of Seagrasses, ed.<br />
A.W.D Larkum, A.J. McComb & S.A. Shepherd, pp. 279<strong>–</strong>303.<br />
Amsterdam, the Netherlands: Elsevier.<br />
Preen, A. (1995) Impacts of dugong foraging on seagrass habitats:<br />
observational and experimental evidence for cultivation grazing.<br />
Marine Ecology Progress Series 124: 201<strong>–</strong>13.<br />
Preen, A.R., Lee Long, W.J. & Coles, R.G. (1995) Flood and<br />
cyclone related loss, and partial recovery, of more than 1000 km 2<br />
of seagrass in Hervey Bay, Queensland, Australia. Aquatic Botany<br />
52: 3<strong>–</strong><strong>17</strong>.
206 C.M. Duarte<br />
Ramos-Esplá, A., Martínez-Pérez, L., Aranda, A., Guillen, J.E.,<br />
Sánchez-Jerez, P. & Sánchez-Lizaso, J.L. (1993) Protección de la<br />
praderas de Posidonia oceanica (L.) Delile mediante arrecifes artificiales<br />
disuasorios frente a la pesca de arrastre ilegal: el caso de El<br />
Campello (SE Ibérico). Publicaciones Especiales Instituto Español<br />
Oceanografía 11: 431<strong>–</strong>439.<br />
Reusch, T.B.H., Borström, C., Stam, W.T. & Olsen, J.L. (1999) An<br />
ancient eelgrass clone in the Baltic. Marine Ecology Progress Series<br />
183: 301<strong>–</strong>304.<br />
Robblee, M.B., Barber, T.R., Carlson, P.R. Jr, Durako, M.J.,<br />
Fourqurean, J.W., Muehlstein, L.K., Porter, D., Yarbro, L.A.,<br />
Zieman, R.T. & Zieman, J.C. (1991) Mass mortality of the tropical<br />
seagrass Thalassia testudinum in Florida Bay (USA). Marine<br />
Ecology Progress Series 71: 297<strong>–</strong>9.<br />
Sand-Jensen, K. & Borum, J. (1991) Interactions among phytoplankton,<br />
periphyton, and macrophytes in temperate freshwaters<br />
and estuaries. Aquatic Botany 41: 137<strong>–</strong>76.<br />
Short, F.T. & Burdick, D.M. (1996) Quantifying seagrass habitat<br />
loss in relation to housing development and nitrogen loading in<br />
Waquoit Bay, Massachusetts. Estuaries 19: 730<strong>–</strong>739.<br />
Short, F.T. & Duarte, C.M. (2001) Methods for the measurement<br />
of seagrass growth and production. In: Global Seagrass Research<br />
Methods, ed. F.T. Short & R.G. Coles, pp: 155-182. Amsterdam,<br />
the Netherlands: Elsevier.<br />
Short, F.T. & Neckles, H.A. (1999) The effects of global climate<br />
change on seagrasses. Aquatic Botany 63: 169<strong>–</strong>196.<br />
Short, F.T. & Wyllie-Echeverria, S. (1996) Natural and humaninduced<br />
disturbance of seagrasses. Environmental Conservation 23:<br />
<strong>17</strong><strong>–</strong>27.<br />
Terrados, J. & Duarte, C.M. (1999) Experimental evidence of<br />
reduced particle resuspension within a seagrass (Posidonia<br />
oceanica L.) meadow. Journal of Experimental Marine Biology and<br />
Ecology 243: 45<strong>–</strong>53.<br />
Terrados, J., Duarte, C.M., Fortes, M.D., Borum, J., Agawin,<br />
N.S.R., Bach, S., Thampanya, U., Kamp-Nielsen, L.,<br />
Kenworthy, W.J., Geertz-Hansen, O. & Vermaat, J. (1998)<br />
Changes in community structure and biomass of seagrass<br />
communities along gradients of siltation in SE Asia. Estuarine,<br />
Coastal and Shelf Science 46: 757<strong>–</strong>768.<br />
Terrados, J., Duarte, C.M., Kamp-Nielsen, L., Borum, J., Agawin,<br />
N.S.R., Fortes, M.D., Gacia, E., Lacap, D., Lubanski, M. &<br />
Greve, T. (1999) Are seagrass growth and survival affected by<br />
reducing conditions in the sediment? Aquatic Botany 65: <strong>17</strong>5<strong>–</strong>197.<br />
UNEP (1989) State of the Mediterranean Marine Environment.<br />
Athens, Greece: MAP Technical Report Series, Volume 28.<br />
Van Katwijk, M.M., Vergeer, L.H.T., Schmitz, G.H.W. & Roelofs,<br />
J.G.M. (1997) Ammonium toxicity in eelgrass Zostera marina.<br />
Marine Ecology Progress Series 157: 159<strong>–</strong>73.<br />
Vidal, M., Duarte, C.M. & Sánchez, M.C. (1999) Coastal eutrophication<br />
research in Europe: progress and imbalances. Marine<br />
Pollution Bulletin 38: 851<strong>–</strong>854.<br />
Vitousek, P.M., Mooney, H.A., Lubchenco, J. & Melillo, J.M.<br />
(1997) Human domination of Earth’s ecosystems. Science 277:<br />
494<strong>–</strong>499.<br />
Walker, D.I., Lukatelich, R.J., Bastyan, G. & McComb, A.J. (1989)<br />
Effect of boat moorings on seagrass beds near Perth, Western<br />
Australia. Aquatic Botany 36: 69<strong>–</strong>77.<br />
World Resources Institute (1998) World Resources 1998<strong>–</strong>99. Oxford,<br />
UK: Oxford University Press: 369 pp.<br />
Young, P.C. (1978) Moreton Bay, Queensland: a nursey area for<br />
juvenile penaeid prawns. Australian Journal Marine Freshwater<br />
Research 29: 55<strong>–</strong>57.<br />
Zieman, J.C., Fourqurean, J.W., Frankovich, T.A. (1999) Seagrass<br />
dieoff in Florida Bay (USA): long-term trends in abundance and<br />
growth of turtle grass, Thalassia testudinum. Estuaries 22:<br />
460<strong>–</strong>470.<br />
Zieman, J.C., Iverson, R.L. & Ogden, J.C. (1984). Herbivory effects<br />
on Thalassia testudinum leaf growth and nitrogen content. Marine<br />
Ecology Progress Series 15: 151<strong>–</strong>8.<br />
Zimmerman, R.C., Kohrs, D.G., Steller, D.L. & Alberte, R.S.<br />
1997. Impacts of CO 2 enrichment on productivity and light<br />
requirements of eelgrass. Plant Physiology 115: 599<strong>–</strong>607.
OUTLOOK ON EVOLUTION AND SOCIETY<br />
doi:10.1111/j.1558-5646.2009.00911.x<br />
IS THE AGE OF THE EARTH ONE OF OUR<br />
“SOREST TROUBLES?” STUDENTS’<br />
PERCEPTIONS ABOUT DEEP TIME AFFECT<br />
THEIR ACCEPTANCE OF EVOLUTIONARY<br />
THEORY<br />
Sehoya Cotner, 1,2 D. Christopher Brooks, 3 and Randy Moore 1<br />
1 Biology Program, University of Minnesota, Minneapolis, Minnesota 55455<br />
2 E-mail: harri054@umn.edu<br />
3 Office of Information Technology, University of Minnesota, Minneapolis, Minnesota 55455<br />
Received July 20, 2009<br />
Accepted November 10, 2009<br />
KEY WORDS: Age of the Earth, evolution education, science education.<br />
When Charles Darwin was developing his ideas for On the Origin<br />
of Species, the most widely accepted estimates of Earth’s<br />
age were those of William Thomson (later Lord Kelvin). Kelvin<br />
used calculations involving thermodynamics to argue that Earth is<br />
only 20<strong>–</strong>100 million years old—an age far too brief to accommodate<br />
evolution by natural selection. Darwin referred to Thomson’s<br />
claim as one of his “sorest troubles,” for Darwin understood that<br />
the history of life on Earth ultimately relies on geology. Darwin<br />
suspected that Earth was much older than Thomson claimed, but<br />
Thomson’s enormous stature as a scientist obliged Darwin to reconcile<br />
his claims with Kelvin’s data. To accommodate Kelvin’s<br />
timeline, Darwin proposed pangenesis as an explanation of inheritance<br />
(i.e., every sperm and egg contained “gemmules thrown off<br />
from each different unit throughout the body”). Darwin’s explanation<br />
sped evolution while avoiding Lamarck’s quasi-spiritual<br />
sources of acquired traits. However, Darwin’s explanation of inheritance<br />
was wrong (see discussion in Moore et al. 2009a).<br />
The age of Earth remains a divisive topic in the modern<br />
evolution<strong>–</strong>creationism controversy. Whereas mainstream science<br />
has long acknowledged that Earth is approximately 4.5 billion<br />
years old, a vocal group of citizens and religious activists con-<br />
858<br />
tinue to insist that Earth is less than 10,000 years old. Although<br />
most geocentrists and flat-Earth advocates have capitulated to<br />
scientific evidence, young-Earth creationists continue to reject<br />
scientific evidence in favor of religious dictum to claim that Earth<br />
is less than 10,000 years old. These antiscience claims have been<br />
surprisingly popular with the public. For example, a Gallup Poll<br />
in early 2009 reported that “On Darwin’s [200th] Birthday, Only<br />
4 in 10 Believe in Evolution” (Newport 2009), and Berkman et al.<br />
(2008) noted that “16% [of biology teachers] believed that human<br />
beings were created by God in their present form at one time<br />
within the last 10,000 years.” In another study, 12.5% of students<br />
were young-Earth creationists (Rutledge and Warden 2000), as are<br />
10%<strong>–</strong>14% of biology majors (Moore and Cotner 2009). Answers<br />
in Genesis’ (AiG) Creation Museum, along with the $27 million in<br />
donations required to build it, attest to the appeal of young-Earth<br />
creationism. Indeed, AiG’s income for 2005 exceeded $13 million,<br />
and that of the Institute for Creation Research (ICR, another<br />
religious organization based on young-Earth creationism) exceeded<br />
$7 million. For comparison, the 2005 income of National<br />
Center for Science Education—the nation’s leading organization<br />
that defends the teaching of evolution in public schools—was<br />
C○ 2010 The Author(s). Journal compilation C○ 2010 The Society for the Study of Evolution.<br />
Evolution 64-3: 858<strong>–</strong>864
$1.2 million (Moore et al. 2009a). Clearly, Earth’s age remains<br />
one of the “sorest troubles” for many people today, just as it did<br />
for Darwin.<br />
In this study, we examined how college students’ selfdescribed<br />
religious and political views influence their beliefs<br />
about Earth’s age and how this may affect their knowledge and<br />
acceptance of evolution. To our knowledge, this is the first study<br />
to examine these factors in college students.<br />
Methods<br />
POPULATION AND SURVEY INSTRUMENT<br />
In 2009, we electronically surveyed 400 students enrolled in several<br />
sections of a nonmajors introductory biology course at the<br />
University of Minnesota. Because the course is one of a few options<br />
required of all undergraduates at the University, we assumed<br />
the survey population is representative of all students (politically,<br />
religiously, and demographically), except for those enrolled in the<br />
College of Biological Sciences. The optional survey, which was<br />
completed before the start of classes, consisted of the 20-item<br />
Measure of Acceptance of the Theory of Evolution (MATE) developed<br />
and validated by Rutledge and Sadler (2007), our own 10item<br />
Knowledge of Evolution Exam (KEE; Moore et al. 2009a),<br />
and several items intended to gauge students’ religious and political<br />
preferences. The MATE consists of statements such as “the age<br />
of the Earth is less than 20,000 years” and “humans are the product<br />
of evolutionary processes,” to which students noted their level of<br />
agreement on a five-point Likert scale (from “strongly agree” to<br />
“strongly disagree”). The KEE questions were modified from an<br />
internal exam database, were authentic to the nonmajors’ test experience,<br />
and were designed to evaluate students’ understanding<br />
of basic tenets of evolutionary theory—for example, how fitness<br />
is measured, natural selection as a mechanism for evolutionary<br />
change, etc. (Appendix I). None of the KEE items specifically addresses<br />
the age of the Earth or evolutionary time. We also asked<br />
students whether they were politically liberal, conservative, or<br />
middle-of-the-road, and whether they were not religious (atheist<br />
or agnostic), or, if religious, were they progressive/liberal, orthodox,<br />
or middle-of-the-road. Students could ignore questions, and<br />
their responses had no impact on students’ grades. Response rate<br />
varied by survey item, with as many as 195 students (almost 50%<br />
of the targeted group) completing the KEE, and as few as 124 responding<br />
to some of the MATE items. The survey and subsequent<br />
data collection were approved by the University of Minnesota’s<br />
Institutional Review Board.<br />
DATA ANALYSIS<br />
We extracted six variables from the survey to explore the factors<br />
that contribute to holding young-Earth and old-Earth beliefs about<br />
the origins of the world, on the one hand, and the relation of those<br />
OUTLOOK ON EVOLUTION AND SOCIETY<br />
beliefs to students’ knowledge of evolution and their presidential<br />
vote.<br />
The first two variables are self-reported measures of religious<br />
and political beliefs. The variable measuring students’ religious<br />
views (MYRELVIEW) is based on the question, “In general, I<br />
would describe my religious views as conservative, middle-ofthe-road,<br />
liberal/progressive, or none of the above/I’m not religious.”<br />
Responses were coded on a four-point scale from conservative<br />
(1) to not religious (4). Similarly, students’ political views<br />
(MYPOLVIEW) were determined by the question, “In general,<br />
I would describe my political views as conservative, middle-ofthe-road,<br />
or liberal.” MYPOLVIEW consists of a three-point scale<br />
ranging from conservative (1) to liberal (3).<br />
We constructed the young-Earth and old-Earth variables from<br />
MATE items that represent perspectives aligning with those beliefs.<br />
For the young-Earth variable (YOUNGEARTH), we averaged<br />
responses to the following two items: “The age of the Earth<br />
is less than 20,000 years,” and “The theory of evolution cannot be<br />
correct because it disagrees with the Biblical account of creation.”<br />
The old-Earth variable (OLDEARTH) was constructed by averaging<br />
responses to how much students agreed or disagreed with the<br />
statements that “Organisms existing today are the result of evolutionary<br />
processes that have occurred over millions of years,”<br />
and “The age of the earth is at least 4 billion years.” Scales for<br />
YOUNGEARTH and OLDEARTH range from 1 (Strongly Disagree)<br />
to 4 (Strongly Agree). Factor analysis confirms that each<br />
pair of items load onto a single dimension representing young-<br />
Earth creationist and old-Earth evolutionist beliefs with rotated<br />
factor loadings of 0.6979 and 0.7787, respectively.<br />
Students’ recollections of their high school biology courses<br />
(HSBIO) were captured in responses to an item asking them to<br />
identify whether or not evolution or creationism was taught in<br />
their courses. For the purposes of our analysis, HSBIO is coded<br />
according the following scheme: included neither evolution nor<br />
creationism = 1; included creationism, but not evolution = 2;<br />
included both evolution and creationism = 3; and included evolution,<br />
but not creationism = 4.<br />
The variable measuring students’ level of evolutionary biology<br />
knowledge (EVOGRADE) is a summative index of the<br />
number of questions answered correctly about various facets of<br />
evolutionary theory. EVOGRADE ranges in value from zero to 10<br />
with each increment representing a correctly answered question.<br />
The 2008 presidential candidate supported by each student was<br />
collected via the self-reported, retrospective response to the statement,<br />
“In the past presidential election, I voted/would have voted<br />
for John McCain or Barack Obama.” For analytical purposes, we<br />
constructed the dichotomous variable VOTE in which support for<br />
John McCain = 0 and support for Barack Obama = 1. Students<br />
who voted for other candidates constituted a small minority and<br />
were excluded from analysis.<br />
EVOLUTION MARCH 2010 859
OUTLOOK ON EVOLUTION AND SOCIETY<br />
Figure 1. Structural equation model demonstrating the nature of the relationships among the variables identified in Table 1. Note that<br />
“+” and“−” refer to positive and negative relationships, respectively, and asterisks follow the code ∗P < 0.05, ∗∗P< 0.01, ∗∗∗P < 0.001.<br />
Although our initial goals included the construction of a<br />
structural equation model (SEM) that employed simultaneous<br />
equations, the nature of the data is prohibitive (none of our variables<br />
is either linear or continuous). Therefore, we employed<br />
several individual models, each of which is best suited to the<br />
particular dependent variable under consideration. We then developed<br />
a structural model that demonstrates the nature of the<br />
relationships between and among our variables (Fig. 1). With<br />
the exception of models with VOTE as the dependent variable,<br />
all models are ordered logistical regression. We employed<br />
logistic regression models when the dichotomous VOTE variable<br />
was dependent. A Monte Carlo simulation was used to<br />
transform the ordered logistical regression results into predicted<br />
probabilities.<br />
Summary statistics for the variables under consideration are<br />
reported in Table 1.<br />
Table 1. Summary statistics.<br />
Variable N Mean Std. Dev. Min. Max.<br />
Myrelview 180 2.8111 0.9501 1 4<br />
Mypolview <strong>17</strong>3 2.3468 0.7361 1 3<br />
Youngearth 124 1.4839 0.6806 1 4<br />
Oldearth 132 3.3674 0.6958 1 4<br />
Hsbio 194 3.3402 1.0270 1 4<br />
Evograde 195 5.3026 2.2328 0 10<br />
Vote <strong>17</strong>4 0.7816 0.4143 0 1<br />
860 EVOLUTION MARCH 2010<br />
Results<br />
The first set of relationships includes the exogenous variables<br />
related to students’ religious and political views. These variables<br />
mutually reinforce one another at statistically significant levels<br />
(P < 0.001). That is, the more liberal one’s political views, the<br />
more likely one is to be liberal, agnostic, or atheistic in their<br />
religious views and vice versa.<br />
Students’ religious views also served as significant predictors<br />
of their beliefs about the origins of the world, their knowledge<br />
of evolutionary theory, and for whom they cast their presidential<br />
votes. Specifically, the more conservative a student’s religious<br />
views, the greater the likelihood of endorsing young-Earth beliefs<br />
(P < 0.05) and the less likely they are to endorse old-Earth<br />
evolutionist beliefs (P < 0.01). Additionally, more liberal, agnostic,<br />
or atheistic religious students were significantly more likely<br />
(P < 0.05) to correctly answer knowledge-based questions about<br />
theories and facts related to evolution. Finally, students holding<br />
less conservative religious views were considerably more likely<br />
to have voted for or supported Barack Obama in the 2008 presidential<br />
election (P < 0.001).<br />
The political views held by students contribute significantly<br />
to their disposition toward evolutionary theory and to their choice<br />
of presidential candidate. Here, the more liberal a student’s political<br />
views, the more likely they are to hold old-Earth beliefs about<br />
the origins of the world (P < 0.05) and to have cast their vote<br />
for Obama (P < 0.001). However, although more liberal political<br />
views are negatively related to acceptance of young-Earth views
and positively related to knowledge of evolutionary theory, political<br />
views fail to predict significantly either of those variables.<br />
Turning to the intermediate variables measuring the impact<br />
of beliefs about the origins of life on Earth, we find mutually<br />
reinforced inverse relationships between young-Earth and old-<br />
Earth views that are highly significant (P < 0.001). Specifically,<br />
those who hold young-Earth views are significantly less likely to<br />
accept an old-Earth rooted in evolutionary theory and vice versa.<br />
We also find that holding old-Earth beliefs contributes significantly<br />
to the ability of students to comprehend complex theoretical<br />
and factual tenets of evolutionary theory (P < 0.05).<br />
Furthermore, acceptance of old-Earth beliefs also significantly<br />
increased the probability of voting for or supporting Obama.<br />
Conversely, holding young-Earth beliefs appears to impact<br />
negatively, but not significantly, one’s ability to grasp cognitively<br />
evolutionary theory. And while holding creationist beliefs did<br />
predict a greater likelihood of voting for John McCain, they did<br />
not do so in a statistically significant manner.<br />
Finally, in keeping with previous findings (Moore and Cotner<br />
2009), the content of students’ high school biology courses<br />
was a significant predictor of their acceptance of evolutionary theory.<br />
The content of students’ high school biology course affects<br />
knowledge of evolution as well: students whose high school biology<br />
course included only evolutionary theory had approximately<br />
a 70% chance of answering half of the questions or more correctly<br />
whereas those with courses teaching only creationism had an approximately<br />
50% chance of doing so (for discussion, see Moore<br />
et al. 2009b); those with neither evolution nor creationism had a<br />
42% chance of scoring 50% or above, suggesting that creationismonly<br />
education is comparable to no education at all with respect to<br />
evolutionary biology. In combination, students who recall being<br />
taught evolution only in high school, and who are religiously and<br />
politically liberal, were more likely to earn any score above 50%<br />
on the exam than were their counterparts with more conservative<br />
educational, religious, and political backgrounds.<br />
Discussion<br />
College students have a variety of religious and political views that<br />
have been shaped collectively by their families, their communities,<br />
and the institutions with which they have come into contact<br />
during the course of their lives. These deeply rooted views influence<br />
students’ receptiveness to theories about life’s origins.<br />
And because these beliefs about creationism and evolution are<br />
firmly grounded in, and are expressions of, their worldviews, as<br />
instructors we are naïve to assume that students in college biology<br />
courses are necessarily open to scientific inquiry.<br />
Although student views of evolution are subject to multiple<br />
influences, our data indicate that their views about Earth’s age<br />
are a strong predictor of several different factors. But is the age<br />
OUTLOOK ON EVOLUTION AND SOCIETY<br />
of Earth one of our “sorest troubles?” Our research and historical<br />
and contemporary literature suggest the following:<br />
(1) Deep time is conceptually difficult to grasp, for the general<br />
public, science educators, and students throughout the educational<br />
spectrum.<br />
Charles Darwin himself had a hard time grasping “deep<br />
time”—the geologic concept of time, requiring billions of years—<br />
but knew that it was essential for his proposed evolutionary mechanism.<br />
Darwin didn’t publish any estimates of Earth’s age, but in<br />
the first edition of On the Origin of Species he did estimate the<br />
time needed to erode the Weald (a region in south England stretching<br />
from Kent to Surrey), “say three hundred million years.” He<br />
then explained: “I have made these few remarks because it is<br />
highly important for us to gain some notion, however, imperfect,<br />
of the lapse of years. During each of these years, over the whole<br />
world, the land and the water has been peopled by hosts of living<br />
forms. What infinite number of generations, which the mind<br />
cannot grasp, must have succeeded each other in the long roll of<br />
years” (Darwin 1859).<br />
More recently, attention has focused on the inability of<br />
scientists (Brush 2001), science educators (Petcovic and Ruhf<br />
2008), college students (Catley and Novick 2009; Libarkin 2006;<br />
Truscott et al. 2006), and the general public (Hofstadter 1996)<br />
to comprehend time in billions of years. Our deficiencies involve<br />
concepts about both absolute time (time in years; Catley and<br />
Novick 2009) and relative time (events in an accurate sequence;<br />
Libarkin 2006).<br />
(2) Students’ inability to accept an old Earth is a barrier to<br />
evolution acceptance.<br />
Not only is an appreciation of deep time foundational to<br />
our full understanding of life’s origin and diversification (Catley<br />
and Novick 2009; Hillis 2007), it is a critical concept for understanding<br />
geology, physics, and astrophysics (see review in Trend<br />
2002). Creationists themselves acknowledge that “An old age for<br />
the Earth is the heart of evolution” (Henry 2003), and have, in<br />
recent years, focused their argument to undermine an ancient<br />
Earth (Humphreys 1999; Baumgardner 2003; Humphreys et al.<br />
2004; DeYoung et al. 2005). Investing energy in debunking an<br />
ancient Earth is only logical if in fact Earth’s age is key to the<br />
Darwinian revolution, and is, in the words of Ernst Mayr (1972),<br />
“the snowball that started the whole avalanche.”<br />
Furthermore, evolution instruction is often independent of<br />
attempts to teach about deep time (Libarkin et al. 2005), a practice<br />
that enables students to learn about evolution without fully<br />
realizing its value as foundational to modern science. For example,<br />
Libarkin et al. (2005) describe students who can cite Earth’s<br />
age correctly but fail to comprehend the span of time Earth was<br />
uninhabitable, or the types of primitive organisms that likely arose<br />
first. These disconnects may speak to a need to incorporate historical<br />
arguments (Jensen and Finley 1995) or, more generally, an<br />
EVOLUTION MARCH 2010 861
OUTLOOK ON EVOLUTION AND SOCIETY<br />
emphasis on the nature of science (Lombrozo et al. 2008) into a<br />
more multidisciplinary strategy for teaching about evolution and<br />
the age of Earth.<br />
(3) Creationists’ explanations for life’s origin are easier to<br />
teach, learn and internalize than are scientific explanations that<br />
rely on an understanding of deep time.<br />
For evidence of creationism’s staying power, one can look to<br />
the recycled “argument for design” throughout history: In 45 BCE,<br />
Marcus Tullius Cicero invoked a sundial to claim evidence of a<br />
creator; in 1691, Anglican clergyman John Ray upgraded to a<br />
clock; and in 1802, William Paley’s treatise “Natural Theology”<br />
invoked a watch and its associated watchmaker (“every indication<br />
of contrivance, every manifestation of design, which existed in<br />
the watch, exists in the works of nature ...”). Technology<strong>–</strong>and<br />
the scientific advancements required to progress from sundials<br />
to watches—has changed, but the basic argument—evidence of<br />
design—remains unchanged (Moore et al. 2009a).<br />
Public-opinion polls provide more contemporary evidence<br />
of the power of creationists’ claims. In 1982, Philip Abelson<br />
wrote that scientific creationists “have no substantial body of<br />
experimental data to back their prejudices. Truth is not on their<br />
side. In the end, their activities must bring only harm to their<br />
cause.” At that time, 44% of the American public believed that<br />
living organisms were created in their present form within the<br />
last 10,000 years (Gallup 1982). After more than two decades of<br />
science education reform, that percentage remains unchanged.<br />
Whether we are constrained by teleology or intentionality<br />
(Sinatra et al. 2008), humans tend to believe in a universe designed<br />
for a purpose (Kelemen 2004; Evans 2001). This tendency<br />
lends a natural authority to supernatural explanations, which, in<br />
combination with a more intuitive model of Earth’s age (thousands,<br />
rather than billions, of years; see DeYoung et al. 2005),<br />
severely jeopardizes a realization of deep time. In asking our students<br />
to forsake the young, the intended, and the concrete for the<br />
old, the accidental, and the abstract, we may be fighting our own<br />
evolutionary history.<br />
(4) Teaching about time requires teaching for conceptual<br />
change.<br />
Learning about deep time is a conceptual change for most<br />
students, whereby instructors must realize the real barrier that<br />
Earth’s age presents to student understanding of evolution. In<br />
addition, and in accordance with studies of teaching for conceptual<br />
change (e.g., Sinatra et al. 2008), teachers may need to guide<br />
students to a realization that their understandings of Earth’s age,<br />
both absolute and relative, are faulty and in need of revision.<br />
Although several authors have suggested strategies for addressing<br />
relative time (e.g., James and Clark’s “ticking toilet paper roll”<br />
[2006]), input on conquering the absolute-time barrier is scarce.<br />
(5) Failure to accept an ancient Earth has real-world implications.<br />
862 EVOLUTION MARCH 2010<br />
We do not doubt that adherence to creationist explanations<br />
for natural phenomena has far-reaching significance. However,<br />
in this study, we specifically explored available information on<br />
voting activity in November 2008, and demonstrate a significant<br />
link between presidential candidate choice and Earth’s-age<br />
dimensions. This association is likely bolstered by contemporary<br />
partisan politics; for example, Republican party platforms<br />
in several states support some form of teaching creationism (as<br />
“Creation Science,” “Intelligent Design,” etc.) as a viable alternative<br />
theory to evolution by natural selection (e.g., Governor Sarah<br />
Palin’s Alaska: “We support giving Creation Science equal representation<br />
with other theories of the origin of life. If evolution is<br />
taught, it should be presented as only a theory”).<br />
The political motivation for embracing creationism is clear:<br />
Recent Gallup polls highlight the different levels of acceptance<br />
of evolution among Republicans, Democrats, and Independents<br />
(Newport 2008, 2009). In one poll, 60% of Republicans claim that<br />
humans were created in their present form by God within the last<br />
10,000 years, a belief shared by 40% of Independents and 38% of<br />
Democrats (Gallup 2008). Of those polled agreeing that humans<br />
evolved and God had no part, 4% are Republicans, 19% are Independents,<br />
and <strong>17</strong>% are Democrats. These data are consistent with<br />
findings that Republicans are more likely to attend church regularly,<br />
and Americans who attend church weekly are highly likely<br />
to believe in creationist explanations for human origins (Newport<br />
2006, 2008). For example, of those who attend church weekly,<br />
70% believe that God created humans in their present form (a<br />
central tenet of young-Earth creationists, or YEC); of those who<br />
seldom or never attend church, 24% are YEC (Newport 2008).<br />
These data suggest that religiosity correlates with one’s tendency<br />
to vote Republican and one’s likelihood to doubt evolutionary<br />
interpretations of origins, specifically, as highlighted herein,<br />
an Earth that is 4<strong>–</strong>5 billion years old. We do not intend to claim that<br />
ignorance of science is restricted to Republicans, nor that there<br />
are no creationist Democrats; rather, Republicans frequently embrace<br />
creationism more explicitly than do their counterparts and<br />
have benefited in recent years by making creationism a campaign<br />
issue.<br />
Implications<br />
Our research suggests that students who are liberal, agnostic,<br />
or atheistic in their religious views, are politically liberal, were<br />
taught evolution in high school, and accept the science behind<br />
evolutionary theory are more likely to understand the theoretical<br />
concepts and empirical findings related to evolution than those<br />
who are more conservative politically and religiously, received<br />
either no evolution or a diluted form of evolution instruction in<br />
high school, and who do not accept an old Earth. However, our<br />
findings also reveal a possible exception to this latter formulation:<br />
holding young-Earth views may not significantly impede a
student’s ability to learn facts about evolutionary theory. Thus,<br />
although it is not the role of biology instructors to engage in<br />
political or religious proselytizing, there remains the possibility<br />
of changing what students know about evolution via academic<br />
instruction.<br />
A student’s vote is likely a proxy for religious views, and<br />
therefore the association between their voting and their acceptance<br />
of evolution appears robust and bears further investigation.<br />
Given that adherence to young-Earth beliefs requires a refutation<br />
not only of modern biology, but also geology, paleontology,<br />
and physics, such convictions may serve as a proxy for scientific<br />
ignorance in general. Consequently, a commitment to comprehensive<br />
conceptual-change instruction in Earth’s age, the scientific<br />
method, and evolutionary theory could have practical, real-world<br />
implications that include nothing less than who we elect for political<br />
office.<br />
LITERATURE CITED<br />
Abelson, P. H. 1982. Creationism and the age of the earth. Science 215:119.<br />
Baumgardner, J. R. 2003. Catastrophic plate tectonics: the physics behind<br />
the genesis flood. Pp. 1<strong>–</strong>10 in In Proceedings of the Fifth <strong>International</strong><br />
Conference on Creationism, Pittsbugh, PA. Creation Science Fellowship,<br />
Inc. Pittsburgh, PA.<br />
Berkman, M. B., J. S. Pacheco, and E. Plutzer. 2008. Evolution and creationism<br />
in america’s classrooms: a national portrait. Plos Biol. 6:e124.<br />
Brush, S. G. 2001. Is the earth too old? The impact of geochronology on<br />
cosmology, 1929<strong>–</strong>1952. Geol. Soc. Lond. Spec. Publ. 190:157<strong>–</strong><strong>17</strong>5.<br />
Catley, K. M., and L. R. Novick. 2009. Digging deep: exploring college students’<br />
knowledge of macroevolutionary time. J. Res. Sci. Teach. 46:311<strong>–</strong><br />
332.<br />
Darwin, C. 1859. The origin of species. John Murray Publishers, London.<br />
DeYoung, D. B., J. Baumgardner, D. R. Humphreys, A. Snelling, S. A. Austin,<br />
E. F. Chaffin, S. W. Boyd, and L. Vardiman. 2005. Thousands, not<br />
billions: challenging an icon of evolution: questioning the age of the<br />
Earth. New Leaf Publishing Group, Green Forest, AR.<br />
Evans, E. M. 2001. Cognitive and contextual factors in the emergence of diverse<br />
belief systems: creation versus evolution. Cogn. Psychol. 42:2<strong>17</strong><strong>–</strong><br />
266.<br />
Gallup, Poll 1982. Evolution, Creationism, Intelligent Design. Available at:<br />
http://www.gallup.com/poll/21814/Evolution-Creationism-Intelligent-<br />
Design.aspx. Accessed December 7, 2009.<br />
Henry, J. 2003. An old age for the earth is the heart of evolution. Creation<br />
Res. Soc. Q. 40:164<strong>–</strong><strong>17</strong>2.<br />
Hillis, D. M. 2007. Making evolution relevant and exciting to biology students.<br />
Evolution 61:1261<strong>–</strong>1264.<br />
Hofstadter, D. R. 1996. Metamagical themas: questing for the essence of mind<br />
and pattern. illustrated ed. Basic Books. Web, New York.<br />
Humphreys, D. R. 1999. Evidence for a young world. Creation Matters 4:1<strong>–</strong>4.<br />
Humphreys, D. R., S. A. Austin, J. R. Baumgardner, and A. A. Snelling. 2004.<br />
Helium diffusion age of 6,000 years supports accelerated nuclear decay.<br />
Creation Res. Soc. Q. 41:1<strong>–</strong>16.<br />
James, P., and I. Clark. 2006. Overcoming geological misconceptions. Planet<br />
<strong>17</strong>:10<strong>–</strong>13.<br />
Jensen, M. S., and F. N. Finley. 1995. Teaching evolution using historical<br />
arguments in a conceptual change strategy. Sci. Educ. 79:147<strong>–</strong>167.<br />
Kelemen, D. 2004. Are children “intuitive theists”? Reasoning about purpose<br />
and design in nature. Psychol. Sci. 15:295<strong>–</strong>301.<br />
OUTLOOK ON EVOLUTION AND SOCIETY<br />
Libarkin, J. C. 2006. College student conceptions of geological phenomena<br />
and their importance in classroom instruction. Planet <strong>17</strong>:6<strong>–</strong>9.<br />
Libarkin, J. C., S. W. Anderson, J. Dahl, M. Beilfuss, W. Boone, and J. P.<br />
Kurdziel. 2005. Qualitative analysis of college students’ ideas about<br />
the earth: interviews and open-ended questionnaires. J. Geosci. Educ.<br />
53:<strong>17</strong><strong>–</strong>26.<br />
Lombrozo, T., A. Thanukos, and M. Weisberg. 2008. The importance of<br />
understanding the nature of science for accepting evolution. Evol. Educ.<br />
Outreach 1:290<strong>–</strong>298.<br />
Mayr, E. 1972. The nature of the Darwinian revolution. Science <strong>17</strong>6:981<strong>–</strong>989.<br />
Moore, R., and S. Cotner. 2009. The creationist down the hall: does it matter<br />
when teachers teach creationism? BioScience 59:429<strong>–</strong>435.<br />
Moore, R., S. Cotner, and A. Bates. 2009a. The influence of religion and high<br />
school biology courses on students’ knowledge of evolution when they<br />
enter college. J. Excellent Teach. (Special Issue on Evolution Education),<br />
9:3<strong>–</strong>11.<br />
Moore, R., M. D. Decker, and S. Cotner. 2009b. No prospect of an end: a<br />
chronology of the evolution-creationism controversy. Greenwood Press,<br />
Westport, CT.<br />
Newport, F. 2006. Almost half of Americans believe humans did not<br />
evolve. Available at: http://www.gallup.com/poll/23200/Almost-Half-<br />
Americans-Believe-Humans-Did-Evolve.aspx. Accessed December 7,<br />
2009.<br />
———. 2008. Republicans, democrats differ on creationism. Available<br />
at: http://www.gallup.com/poll/108226/Republicans-Democrats-Differ-<br />
Creationism.aspx. Accessed December 7, 2009.<br />
———. 2009. On Darwin’s birthday, only 4 in 10 believe in evolution.<br />
Retrieved May 3, 2009, from http://www.gallup.com/poll/<br />
114544/Darwin_Believe-Evolution.aspx<br />
Petcovic, H. L., and R. J. Ruhf. 2008. Geoscience conceptual knowledge<br />
of preservice elementary teachers: results from the geoscience concept<br />
inventory. J. Geosci. Educ. 56:251<strong>–</strong>260.<br />
Rutledge, M. L., and M. A. Warden. 2000. Evolutionary theory, the nature<br />
of science and high school biology teachers: critical relationships. Am.<br />
Biol. Teacher 62:23<strong>–</strong>31.<br />
Rutledge, M. L., and K. C. Sadler. 2007. Reliability of the Measure of Acceptance<br />
of the Theory of Evolution (MATE) instrument with university<br />
students. Am. Biol. Teacher 69:332<strong>–</strong>335.<br />
Sinatra, G. M., S. K. Brem, and E. M. Evans. 2008. Changing minds? Implications<br />
of conceptual change for teaching and learning about biological<br />
evolution. Evol. Educ. Outreach 1:189<strong>–</strong>195.<br />
Trend, R. D. 2002. Developing the concept of deep time. Pp. 187<strong>–</strong>201 in V.<br />
J. Mayer, ed. Global science literacy. Springer, New York.<br />
Truscott, J. B., A. Boyle, S. Burkill, J. Libarkin, and J. Lonsdale. 2006. The<br />
concept of time: can it be fully realised and taught? Planet <strong>17</strong>:21<strong>–</strong>23.<br />
Associate Editor: T. Meagher<br />
Appendix I<br />
KNOWLEDGE OF EVOLUTION EXAM (KEE)<br />
(1) Which of the following support the theory of evolution?<br />
A. artificial selection (also known as selective breeding), an<br />
analogue of natural selection<br />
B. comparative biochemistry, where similarities and differences<br />
of DNA among species can be quantified<br />
C. vestigial structures that serve no apparent purpose<br />
D. comparative embryology, where the evolutionary history<br />
of similar structures can often be traced<br />
EVOLUTION MARCH 2010 863
OUTLOOK ON EVOLUTION AND SOCIETY<br />
E. all of the above provide evidence to support the theory of<br />
evolution<br />
(2) Resistance to a wide variety of insecticides has recently<br />
evolved in many species of insects. Why?<br />
A. mutations are on the rise<br />
B. humans are altering the environments of these organisms,<br />
and the organisms are evolving by natural selection<br />
C. no new species are evolving, just resistant strains or varieties.<br />
This is not evolution by natural selection<br />
D. humans have better health practices, so these organisms<br />
are trying to keep up<br />
E. insects are smarter than humans<br />
(3) Which of the following is the most fit in an evolutionary sense?<br />
A. a lion who is successful at capturing prey but has no cubs<br />
B. a lion who has many cubs, eight of which live to adulthood<br />
C. a lion who overcomes a disease and lives to have three<br />
cubs<br />
D. a lion who cares for his cubs, two of whom live to adulthood<br />
E. a lion who has a harem of many lionesses and one cub<br />
(4) How might a biologist explain why a species of birds has<br />
evolved a larger beak size?<br />
A. large beak size occurred as a result of mutation in each<br />
member of the population<br />
B. the ancestors of this bird species encountered a tree with<br />
larger than average sized seeds. They needed to develop<br />
larger beaks to eat the larger seeds, and over time, they<br />
adapted to meet this need<br />
C. some members of the ancestral population had larger<br />
beaks than others. If larger beak size was advantageous,<br />
they would be more likely to survive and reproduce. As<br />
such, large beaked birds increased in frequency relative<br />
to small beaked birds<br />
D. the ancestors of this bird species encountered a tree with<br />
larger than average sized seeds. They discovered that by<br />
stretching their beaks, the beaks would get longer, and<br />
this increase was passed on to their offspring. Over time,<br />
the bird beaks became larger<br />
E. none of the above<br />
(5) Which of the following statements about natural selection is<br />
true?<br />
A. natural selection causes variation to arise within a population<br />
B. natural selection leads to increase likelihood of survival<br />
for certain individuals based on variation. The variation<br />
comes from outside the population<br />
C. all individuals within a population have an equal chance<br />
of survival and reproduction. Survival is based on choice<br />
864 EVOLUTION MARCH 2010<br />
D. natural selection results in those individuals within a population<br />
who are best-adapted surviving and producing<br />
more offspring<br />
E. natural selection leads to extinction<br />
(6) All organisms share the same genetic code. This commonality<br />
is evidence that<br />
A. evolution is occurring now<br />
B. convergent evolution has occurred<br />
C. evolution occurs gradually<br />
D. all organisms are descended from a common ancestor<br />
E. life began millions of years ago<br />
(7) Which of the following statements regarding evolution by<br />
natural selection is FALSE?<br />
A. natural selection acts on individuals<br />
B. natural selection is a random process<br />
C. very small selective advantages can produce large effects<br />
through time<br />
D. natural selection can result in the elimination of certain<br />
alleles from a population’s gene pool<br />
E. mutations are important as the ultimate source of genetic<br />
variability upon which natural selection can act<br />
(8) A change in the genetic makeup of a population of organisms<br />
through time is<br />
A. adaptive radiation<br />
B. biological evolution<br />
C. LaMarckian evolution<br />
D. natural selection<br />
E. genetic recombination<br />
(9) Which of the following is the ultimate source of new variation<br />
in natural populations?<br />
A. recombination<br />
B. mutation<br />
C. hybridization<br />
D. gene flow<br />
E. natural selection<br />
(10) Which of the following best describes the relationship between<br />
evolution and natural selection?<br />
A. natural selection is one mechanism that can result in the<br />
process of evolution<br />
B. natural selection produces small-scale changes in populations,<br />
whereas evolution produces large-scale ones<br />
C. natural selection is a random process whereas evolution<br />
proceeds toward a specific goal<br />
D. natural selection is differential survival of populations or<br />
groups, resulting in the evolution of individual organisms<br />
E. they are equivalent terms describing the same process
Conservation and<br />
Behaviour<br />
Richard Buchholz, Department of Biology, University of Mississippi, Mississippi, USA<br />
Patrick Yamnik, Department of Biology, University of Mississippi, Mississippi, USA<br />
Cassan N Pulaski, Department of Biology, University of Mississippi, Mississippi, USA<br />
Chelsea A Campbell, Department of Biology, University of Mississippi, Mississippi, USA<br />
Conservation behaviour is an area of conservation biology particularly suited to<br />
investigate problems of species endangerment associated with managing animals in<br />
fragmented habitats and isolated parks. It employs a theoretical framework that<br />
examines the mechanisms, development, function and phylogeny of behavioural<br />
variation in order to develop practical tools for preventing extinction. This framework<br />
can be used to attract animals to suitable habitat, restore migratory movements,<br />
identify individual reproductive strategies affecting conservation and focus protection<br />
on species most susceptible to extinction. The future success of conservation behaviour<br />
requires that behaviourists link individual variation in behaviour to processes that<br />
determine population viability.<br />
Introduction<br />
Conservation behaviour is an emerging discipline that utilizes<br />
the framework and methods of ethology to help solve<br />
issues important to the biodiversity crisis (Buchholz, 2007).<br />
Species are becoming extinct or threatened with extinction<br />
at an accelerating rate due to human population growth<br />
and our unsustainable use of natural resources. The relatively<br />
new field of conservation behaviour aims to use a<br />
scientific understanding of animal behaviour to both determine<br />
the reasons for differential susceptibility to extinction,<br />
and to manage threatened animals and their habitats<br />
to promote population viability. Conservation behaviour<br />
is not merely the description of the natural history of an<br />
animal. Instead conservation behaviour uses the framework<br />
of ethology, originally posed by the Nobel laureate<br />
Niko Tinbergen, to allow conservation issues to be investigated<br />
from four complementary behavioural perspectives:<br />
mechanisms, ontogeny, adaptive function and<br />
phylogeny. The proximate mechanisms of behaviour include<br />
the physiological and neuro-endocrine underpinnings<br />
of animal action. The ontogeny of behaviour<br />
ELS subject area: Ecology<br />
How to cite:<br />
Buchholz, Richard; Yamnik, Patrick; Pulaski, Cassan N; and, Campbell,<br />
Chelsea A (December 2008) Conservation and Behaviour. In:<br />
Encyclopedia of Life Sciences (ELS). John Wiley & Sons, Ltd: Chichester.<br />
DOI: 10.1002/9780470015902.a00212<strong>17</strong><br />
. Introduction<br />
. Habitat Selection<br />
Advanced article<br />
Article Contents<br />
. Dispersal, Migration and Conservation<br />
. Reproductive Behaviours and Population Growth<br />
. Effects of Small Population Size on Behaviour<br />
. Managing Human<strong>–</strong>Wildlife Conflict<br />
. Implications for Management<br />
Online posting date: 15 th December 2008<br />
explores the genetic and learned basis of behavioural development<br />
over the lifetime of individual animals. Ultimately<br />
behaviour patterns may cause differential survival<br />
or reproduction of individuals, and thus serve an adaptive<br />
function. Finally the ultimate origins of inter-specific variation<br />
in the behaviour may be investigated by exploring<br />
how behaviour has evolved among a set of phylogenetically<br />
related species. These four approaches can be integrated to<br />
allow conservation behaviourists to propose novel solutions<br />
to conservation problems. See also: Biodiversity <strong>–</strong><br />
Threats; Modern Extinction; Tinbergen, Nikolaas<br />
Although the interdisciplinary field of conservation biology<br />
was formalized as a science in the mid-1980s with the<br />
formation of the Society for Conservation Biology, recognition<br />
that behavioural biology could contribute uniquely<br />
to conservation did not coalesce for another decade, as<br />
evidenced by a flurry of symposia and publications on the<br />
subject (listed by Caro, 2007). Conservation behaviour<br />
came about as there was an increasing realization that<br />
population-level investigations of genetics and ecology<br />
were insufficient to explain some rather dramatic species<br />
declines. For example, cheetah Acinonyx jubatus population<br />
decline was linked to cheetah-cub killing by lions<br />
Panthera leo and other large carnivores (Laurenson et al.,<br />
1995). Because cheetah-cub killing is not simply a form of<br />
predation for food, a more complex behavioural investigation<br />
of the origins and causes of this behaviour by lions is<br />
warranted. A conservation behaviour approach to this<br />
problem would include an investigation of the sensory cues<br />
that lions use to find cheetah litters (mechanism), the ability<br />
of cheetah mothers to learn to reduce their risk of litter<br />
predation (ontogeny), the effect of cheetah-cub killing on<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 1
the lion’s survival and reproduction (adaptive function),<br />
and a comparative analysis of the ecological and life-history<br />
factors associated with inter-specific infanticide<br />
among species in the order Carnivora (phylogeny). This<br />
comprehensive approach to conservation holds promise<br />
for ameliorating conservation problems if it can be translated<br />
into management actions that help threatened populations<br />
grow to a sustainable equilibrium (Beissinger,<br />
1997).<br />
When a species is reduced to just a few animals, the behaviour<br />
of every individual has a direct and obvious impact<br />
on the species’ survival. When animals are endangered, but<br />
not at the brink of extinction, the role of behaviour may be<br />
less obvious but is no less important. Natural selection is<br />
expected to mould the decision-making of individual animals<br />
so that they weigh the potential lifetime fitness costs<br />
and benefits of their options before choosing a course of<br />
action. The conservation behaviourist’s objective is to understand<br />
how animal decisions may have become maladaptive<br />
in a human-dominated landscape, and to<br />
participate in management efforts that correct or compensate<br />
for behaviour that reduces survival or reproduction.<br />
Although there are myriad ways that behaviour might impact<br />
population dynamics, five behavioural topics will play<br />
major roles in protecting biodiversity: habitat selection,<br />
dispersal and migratory movements, reproduction, living<br />
in small populations and human<strong>–</strong>animal conflict over resource<br />
use. See also: Conservation of Populations and<br />
Species<br />
Habitat Selection<br />
Habitats that look fairly homogeneous to the untrained<br />
human eye may be heterogeneous in features important to<br />
the survival and reproduction of a particular animal species.<br />
Even within species, suitable habitat can vary with the<br />
age, sex and condition of the individual. Thus we expect<br />
animals to be choosy about where they live, and to move to<br />
places that maximize their lifetime reproductive success. If<br />
landscape changes by humans cause animals to bypass<br />
good habitats, or to become attracted to poor habitats,<br />
population decline may occur. The framework of conservation<br />
behaviour provides us with potential management<br />
tools to manipulate the habitat use of animals so that they<br />
use habitats optimally in light of anthropogenic alterations<br />
to which they have not had time to adapt in an evolutionary<br />
sense.<br />
Getting animals past human obstacles<br />
Goal-oriented movements of animals might cover very little<br />
geographic distance or alternatively occur for just a few<br />
minutes of the animal’s life, but be of great importance to<br />
the survival and reproductive success of individuals. For<br />
example, some species have a critically important transition<br />
from the location where they are born or hatched, and<br />
the place where they will be reared. Understanding the cues<br />
2<br />
Conservation and Behaviour<br />
that these naïve, fledgling animals use to find their way<br />
allows the conservationist to avoid conflicts between this<br />
critical transition and anthropogenic changes. For example,<br />
newly hatched sea turtles and fledgling petrels must<br />
move from their nests to the ocean, but can become disoriented<br />
by some wavelengths of artificial night lighting<br />
(Witherington, 1997). Likewise, roadways and hydroelectric<br />
dams can function as barriers to animal movement that<br />
could be circumvented through management of the cues<br />
that imperilled species use to find their way. There will be<br />
times when the findings of conservation behaviourists may<br />
run into conflict with other efforts at environmental protection.<br />
For example, migratory tree bats (Lasiurus species)<br />
appear to be using wind turbines as mating sites along migration<br />
routes. Cryan and Brown (2007) have suggested<br />
that these normally solitary bats may use a behavioural rule<br />
such as ‘go to the tallest structure’ to rendezvous during fall<br />
migration. As a result dozens or hundreds of bats are killed<br />
during a several week period, and the protection of biodiversity<br />
finds itself at odds with efforts to use alternative<br />
energy sources to reduce global climate change. Conservation<br />
behaviourists must be involved in efforts to make<br />
wind turbines less attractive or less deadly to bats, as well as<br />
participate in other situations where acute animal contact<br />
with human landscape structures cause mortality and interrupt<br />
crucial events in a species’ life cycle.<br />
Getting animals to use good habitat<br />
In human-dominated landscapes habitat patches are likely<br />
to be relatively small and susceptible to local extinctions.<br />
Small populations in those habitat fragments may decline<br />
for a number of reasons, including inbreeding depression<br />
and greater exposure to climatic extremes. Suitable patches<br />
in some cases may remain unoccupied if cues used in habitat<br />
selection are absent, if the patch is outside of an organism’s<br />
perceptual range or if misleading cues are present<br />
that deter settlement (Gilroy and Sutherland, 2007).<br />
Behavioural knowledge may be used to lure individuals<br />
to settle in these unoccupied or underutilized habitats.<br />
Many species display conspecific attraction wherein the<br />
presence of conspecifics is likely to be indicative of suitable<br />
habitat (Ahlering and Faaborg, 2006). Ward and<br />
Schlossberg (2004) successfully attracted endangered territorial<br />
black-capped vireos Vireo atricapilla to unoccupied<br />
habitat using recorded vocalizations. Heterospecific cues<br />
are also used in habitat selection; least flycatchers, Empidonax<br />
minimus were attracted to habitat utilized by<br />
American redstarts Setophaga ruticilla (Fletcher, 2007).<br />
Similarly, marbled newts Triturus marmoratus oriented towards<br />
the calls of natterjack toads Bufo calamita (Diego-<br />
Rasilla and Luengo, 2004). In addition to conspecific or<br />
heterospecific auditory cues, olfactory, chemical and visual<br />
cues may also be used to attract individuals to unoccupied<br />
habitat (Gilroy and Sutherland, 2007). Habitat cues such as<br />
light and vegetation are also used by organisms in selecting<br />
suitable habitat, and these cues may be manipulated to<br />
attract individuals to unoccupied patches. Similarly<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net
animals may adopt microhabitat choices by observing the<br />
behaviour of experienced individuals.<br />
Getting animals to avoid bad habitat<br />
Alternatively it may be necessary to dissuade animals from<br />
using an attractive habitat patch that is actually harmful.<br />
Such ‘ecological traps’ occur when cues that historically<br />
indicated suitable habitat have become disconnected from<br />
habitat quality due to human disturbance. An ecological<br />
trap functions as a population sink (Schlaepfer et al., 2002).<br />
The conservation behaviourist may develop tools for<br />
masking misleading cues to eliminate the sensory allure<br />
or introduce additional cues to indicate the habitat’s poor<br />
suitability for settlement. The majority of ecological trap<br />
studies have focused on birds, and a number of these have<br />
found forest edge habitats to function as traps due to increased<br />
predation and nest-parasitism (Battini, 2004).<br />
Edge-effect traps may be countered by creating openings<br />
in the canopy of the forest interior which may attract birds<br />
away from the habitat edge (Battini, 2004). Heterospecific<br />
cues may also be used to dissuade individuals from settling<br />
in unsuitable habitat. American redstarts S. ruticilla were<br />
more likely to avoid habitats with least flycatcher, E. minimus<br />
vocalizations than control habitat patches (Fletcher,<br />
2007). Understanding behavioural cues and mechanisms<br />
resulting in ecological traps will allow for the management<br />
of these problematic habitats and their associated wildlife.<br />
Dispersal, Migration and Conservation<br />
The direction, rate and motivational basis for animal<br />
movement varies throughout the day, seasonally and with<br />
life stage. If we employ the Tinbergen framework to the<br />
problem of an animal’s movement out of its traditional<br />
home range, conservation behaviour might consider, for<br />
example, the hormonal (mechanistic) basis for an individual’s<br />
decision to move elsewhere, whether it learned about<br />
the direction of movements from its parents (ontogeny)<br />
and the fitness consequences of the move (adaptive function).<br />
Comparing the relationship between life-history<br />
traits and the movement patterns of a group of related<br />
species might be useful in anticipating conservation conflicts.<br />
Individual differences in the distance that an animal<br />
moves provides additional structure for our discussion of<br />
how animal locomotion relates to conservation of biodiversity.<br />
Geographical scales of animal movement can be<br />
divided into two broad categories: dispersal over relatively<br />
short distances, and migration over much longer distances.<br />
Short distances<br />
Short distance dispersal entails emigration from the natal<br />
or home range wherein an individual’s objective for moving<br />
may be to avoid inbreeding, reduce competition with kin or<br />
to locate better food resources. There are a wide range of<br />
abiotic and biotic factors that determine dispersal. In<br />
fragmented or isolated habitat patches, dispersing individuals<br />
may suffer high mortality and reduce the viability of<br />
the local population. Dispersing animals, particularly large<br />
carnivores, may come into conflict with humans when they<br />
travel through human-dominated habitats, reducing public<br />
support for wildlife preservation. Thus investigating the<br />
motivation behind individual dispersal can be critical to<br />
species conservation. For example, Zedrosser et al. (2007)<br />
demonstrated that male brown bears Ursus arctos will migrate<br />
up to 119 km to avoid inbred mating, exposing them<br />
to a greater risk for human contact. Proper management of<br />
the brown bear might involve translocation of breedingage<br />
animals to create local breeding opportunities for bears<br />
that would otherwise disperse. Alternatively manipulating<br />
environmental cues to redirect the movement of dispersing<br />
bears through protected corridors may reduce bear mortality<br />
and conflict with humans. See also: Range Limits<br />
The causes of animal dispersal will be of increasing concern<br />
as disease epidemics and emerging zoonotic pathogens<br />
threaten both human and wildlife health. The social systems<br />
of chimpanzees Pan troglydytes and gorillas Gorilla<br />
gorilla, for example, may explain the differential impact of<br />
the ebola virus on these great ape species. In Great Britain<br />
Eurasian badgers Meles meles serve as reservoirs of bovine<br />
tuberculosis, and areas with high densities of badgers typically<br />
have higher incidences of bovine tuberculosis<br />
(McDonald et al., 2008). Reducing the density of badgers<br />
through culling, however, disrupts the social and territorial<br />
structure of the badgers. This resulted in the surviving<br />
badgers travelling greater distances and utilizing larger<br />
areas, effectively spreading bovine tuberculosis. The latter<br />
example highlights the need for a behaviourist’s understanding<br />
of individual behavioural strategies for anticipating<br />
the complexities of disease transmission in landscapes.<br />
Long distances<br />
Conservation and Behaviour<br />
Seasonal migration may involve non-stop movement<br />
across thousands of kilometres of unsuitable habitat, resulting<br />
in the mortality of individuals in poor condition or<br />
those unable to orient correctly. Unfortunately there is little<br />
information known about how the different stages of<br />
migratory movement affect survival and reproduction. For<br />
example, how might the migratory routes or duration of<br />
stopovers at resting sites affect fecundity in the subsequent<br />
breeding seasons? These migratory options are behavioural<br />
decisions suitable for investigation by conservation behaviourists.<br />
The most information available is for birds, with<br />
virtually no data available for the highly threatened migrations<br />
of hoofed mammals (Bolger et al., 2008). Comparative<br />
studies in birds have revealed that migratory<br />
species are more susceptible to decline or extinction than<br />
non-migratory species (references in Thomas et al., 2006).<br />
In a phylogenetically controlled analysis of population<br />
status in North American shorebirds, Thomas et al. (2006)<br />
found that species that opted for inland (rather than<br />
coastal) migratory routes were more threatened. The authors<br />
suggest that additional protections for ephemeral<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 3
Number declining or extinct<br />
wetlands and upland area stopover sites might stem the<br />
decline of these bird species. Green sea turtles Chelonia<br />
mydas that both reproduce and forage in the Cocos Islands<br />
have very brief migrations compared to other populations,<br />
and are thus hypothesized to have spared energetic resources<br />
that should result in greater fecundity or survival<br />
(Whiting et al., 2008). If confirmed, this population of turtles<br />
might be more resilient to human disturbance, allowing<br />
conservationists to focus their preservation efforts on the<br />
more susceptible long-distance migrants.<br />
Migratory motivation and compass orientation can have<br />
both learned and instinctual components. New migratory<br />
routes seem to evolve rather quickly, and thus may not be<br />
reflected in broad brush genetic determinations of ‘evolutionary<br />
significant units’, a measure of population uniqueness<br />
favoured by policy makers (Davis et al., 2006). Efforts<br />
to recover the endangered whooping crane Grus americana<br />
in North America have entailed using behavioural biology<br />
to stop migration in some reintroduced populations (central<br />
Florida, USA), and establish new migratory routes<br />
(between north Florida and Wisconsin, USA) in others<br />
(Canadian Wildlife Service and U.S. Fish and Wildlife<br />
Service, 2005), in both cases through the manipulation of<br />
learned cues in the rearing environment. Virtually nothing<br />
is known about the mechanistic and developmental basis of<br />
migration in ungulates (Bolger et al., 2008); however, a<br />
better understanding of how their behaviour might be<br />
managed so that they can circumvent anthropogenic<br />
obstacles will be critical to the conservation of these land-bound<br />
mammals (Figure 1). See also: Migration: Orientation<br />
and Navigation<br />
Wildlife corridors<br />
14<br />
12<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
Dispersal behaviour is of particular interest to conservation<br />
biologists due to its potential impacts on population<br />
Overhunting Habitat loss<br />
Threats<br />
Migration obstacle<br />
Figure 1 More migratory populations and species of hoofed mammals are endangered by anthropogenic obstacles to animal migration such as fences, roads<br />
and reservoirs, than by overhunting or agricultural development of natural areas. Data from Bolger et al. (2008).<br />
4<br />
Conservation and Behaviour<br />
genetics. Animals confined to small, disconnected populations<br />
have only genetic relatives to reproduce with, resulting<br />
in low genetic variability, inbreeding depression and<br />
poor disease resistance. Dispersal behaviour significantly<br />
affects population connectivity, and may be facilitated by<br />
the creation of suitable habitat corridors within the fragmented<br />
landscape. Behavioural studies can provide important<br />
insight into designing corridors based on<br />
behavioural responses of wildlife to habitat edges and corridors.<br />
See also: Landscape Ecology<br />
Behavioural studies will provide valuable insight into<br />
determining when corridors are needed and how to design<br />
them. Behavioural studies have demonstrated that many<br />
species, including spotted salamanders Ambystoma maculatum,<br />
and two butterfly species, Phoebis sennae and<br />
Eurema nicippe, detect and avoid habitat edges, and this<br />
behaviour leads to a greater probability of corridor use<br />
(Haddad, 1999; Rittenhouse and Semlitsch, 2006). The<br />
habitat quality within the corridor itself may determine its<br />
effectiveness. The Chucao tapaculo Scelorchilus rubecula<br />
disperses well through low-quality shrub habitat to access<br />
forest patches but avoids large open gaps between habitat<br />
patches (Castello´ n and Sieving, 2005). The salamander<br />
Ensatina eschscholtzii travels more rapidly between suitable<br />
habitat patches if corridor quality is poor (Rosenberg<br />
et al., 1998). Thus, behavioural studies can greatly inform<br />
the use and design of management by understanding which<br />
organisms are likely to use corridors effectively based on<br />
their response to habitat edges, their behaviour within corridors<br />
and other factors.<br />
An important application of corridor theory is the use of<br />
highway overpasses and underpasses to provide connectivity<br />
between habitat patches that are bisected by roads.<br />
Understanding how animals respond to roads, traffic and<br />
the structural design of under- and overpasses will allow<br />
managers to design corridors that maximize wildlife use<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net
and minimize traffic-induced wildlife mortality. For example,<br />
highway crossings by elk Cervus elaphus in Arizona,<br />
USA, decreased with increasing traffic volume, although<br />
crossings were more likely to occur near important resources<br />
and depended on the season (Gagnon et al., 2007).<br />
Importantly Ng et al. (2004) found that the structural design<br />
of crossings can greatly affect their utility. Over- and<br />
underpass design may be the most important factor affecting<br />
which, if any, species are likely to use them, with some<br />
species such as grizzly bears, U. arctos preferring wide,<br />
short and tall crossings, whereas other species such as cougars<br />
Puma concolor prefer more constricted passages<br />
(Clevenger and Waltho, 2005; Mata et al., 2005). Understanding<br />
an organism’s behaviour will allow managers to<br />
most efficiently plan highway crossings that will suit wildlife<br />
needs.<br />
Reproductive Behaviours and<br />
Population Growth<br />
The rate and productivity with which animals breed determine<br />
in part whether their population can compensate for<br />
losses due to anthropogenic changes such as climate<br />
change, hunting and competition with invasive species.<br />
Differential mating success, or sexual selection, occurs<br />
through mate choice and intra-sexual competition, with<br />
myriad implications for conservation (Quader, 2005).<br />
See also: Sexual Selection<br />
Mating systems and genetic diversity<br />
Mating systems provide broadly defined categories encompassing<br />
how the sexual strategies of females and males<br />
are integrated under natural conditions. Polygamous mating<br />
systems have been linked to an increased risk of extinction<br />
in birds (Doherty et al., 2003), perhaps because the<br />
effective population size (N e) is lower in mating systems<br />
where only a few males are mating with the majority of<br />
females (Parker and Waite, 1997). In small populations in<br />
particular low genetic diversity may be exacerbated by extreme<br />
differences in male mating success, an effect that may<br />
be ameliorated by behavioural management. For example,<br />
Alberts et al. (2002) found that the temporary removal of<br />
dominant males in the threatened Cuban iguana (Cyclura<br />
nubile) allowed other males to mate, theoretically increasing<br />
genetic variability and population persistence. If female<br />
choice is the primary determinant of male mating success,<br />
however, it might be appropriate to manipulate positively<br />
the apparent quality of a male not normally attractive to<br />
females. Fisher et al. (2003) showed that by redistributing<br />
scent markings of individual male loris in captivity, it was<br />
possible to increase the likelihood that females would subsequently<br />
mate with them.<br />
Human efforts to promote conservation can have inadvertent<br />
negative effects on population viability when they<br />
focus on natural selection and ignore sexual selection.<br />
Trophy hunting has been traditionally thought of as a relatively<br />
benign way to both remove surplus males from<br />
populations and provide financial resources to human<br />
communities surrounding protected areas. Infanticide by<br />
males that replace trophy hunted males, unfortunately,<br />
may destabilize lion prides. Thus the intra-sexual competition<br />
of male lions enabled by trophy hunting may result in<br />
unsustainable use if male age is not considered (Whitman<br />
et al., 2004). In the case of bighorn sheep Ovis canadensis<br />
removal of males with trophy size horns removes fastgrowing<br />
males whose ‘good genes’ may normally benefit<br />
female reproductive success indirectly through the survival<br />
of their offspring (Coltman et al., 2003). Even when wildlife<br />
managers attempt to positively affect female condition, for<br />
example, through supplemental feeding, sexual selection<br />
can cause unexpected population consequences. Supplemental<br />
feeding of female endangered kakapo parrots Strigops<br />
habroptilus in New Zealand resulted in male-biased<br />
clutches of offpring (Clout et al., 2002). Because males must<br />
be large to compete with other males for matings, mothers<br />
apparently produce more sons if they have an abundance of<br />
food. See also: Natural Selection: Sex Ratio<br />
Parental behaviour<br />
Conservation and Behaviour<br />
Parental care of offspring must also be considered in managing<br />
endangered animal populations. In some species,<br />
adults may be present at nesting sites and breeding burrows,<br />
but are not contributing their genes to the next generation.<br />
In cooperatively breeding, or ‘helping’, species,<br />
offspring from previous breeding seasons may stay with<br />
their parents and assist them in rearing additional siblings.<br />
Most helpers appear to help because no breeding opportunities<br />
are available elsewhere. A population census<br />
would mistakenly incorporate them in an analysis of population<br />
viability, when in fact the N e is much lower than it<br />
appears. To reduce the risk of extinction of the Seychelles<br />
warbler Acrocephalus sechellensis, which was restricted to a<br />
single island, Komdeur et al. (1995) translocated helpers to<br />
an unoccupied island. These helpers bred successfully and<br />
eventually accumulated helpers of their own when the new<br />
island’s territories became fully occupied. Thus knowledge<br />
of the adaptive function of cooperative breeding was used<br />
to found a new population of this endangered species without<br />
reducing the effective population size of the single<br />
source population. See also: Eusociality and Cooperation<br />
In other cases, wildlife management intended to protect<br />
endangered species has had unintended negative consequences<br />
on parental care. Poaching of rhinos for their<br />
horns resulted in drastic declines of African rhino species.<br />
In Namibia wildlife managers captured and removed the<br />
horns of their few remaining rhinos to make them worthless<br />
to horn-poachers. Unfortunately the calves of de-horned<br />
mother rhinos had low survival, probably due to the reduced<br />
ability of their mothers to protect their offspring<br />
from predators (Berger and Cunningham, 1994).<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 5
Effects of Small Population Size on<br />
Behaviour<br />
The reduced reproductive success of individuals in small<br />
populations, called the ‘Allee Effect’, is of great concern to<br />
conservationists. Sometimes this reduced reproduction is<br />
simply a consequence of low population densities and the<br />
inability of individuals to find one another for mating. In<br />
other cases the adaptive benefits of living in social groups<br />
are reversed below a threshold group size. For example, in<br />
some normally social species it might be more costly to<br />
lifetime reproductive success to live in a too small group<br />
than to live an entirely solitary existence. Allee effects are<br />
not readily predicted if a ‘species typical’ viewpoint of behaviour<br />
is adopted, but will be anticipated by conservation<br />
behaviourists because they focus at the level of individual<br />
differences in decision-making behaviour. We explore several<br />
examples of the Allee effect in endangered mammal<br />
species.<br />
African wild dogs<br />
Small population sizes may have particularly profound<br />
effects on the behaviour and population viability of highly<br />
social animals such as the African wild dog, Lycaon pictus.<br />
Courchamp et al. (2002) demonstrated that small African<br />
wild dog populations may suffer reduced viability due to a<br />
trade-off between hunting and pup-guarding. Wild dog<br />
hunting is performed predominantly by the pack, and<br />
larger hunting groups are typically more successful at capturing<br />
prey. Additionally, although the majority of the<br />
pack is hunting, one adult usually remains to guard pups<br />
from predation. Thus, when packs reach a threshold level<br />
of fewer than about five individuals, the trade-off between<br />
hunting success and pup protection may have serious<br />
effects on population viability. Either hunting success or<br />
pup survival will be reduced at low group sizes. African<br />
wild dog populations composed of very small groups can be<br />
viable only if the risks of pup predation are also very low.<br />
Saiga antelope<br />
Behavioural shifts associated with reduced population size<br />
have also been demonstrated in the saiga antelope (Saiga<br />
tatarica) in Asia. Saiga populations have seen a catastrophic<br />
decrease in the number of males, due to an increase<br />
in poaching for saiga meat and horns. Poaching of adult<br />
males resulted in an extremely female-biased sex ratio,<br />
which in turn led to a decline in the number of pregnancies<br />
(Milner-Gulland, 2003). Normally, in the polygnous mating<br />
system used by saiga antelopes, males will defend a<br />
harem of 12<strong>–</strong>30 females. In female-biased populations,<br />
dominant females aggressively excluded subdominant females<br />
from mating with the few remaining males (Milner-<br />
Gulland, 2003). As a consequence first-year females in<br />
poached populations do not conceive, and the populations<br />
decline due to the direct mortality of males by poaching,<br />
6<br />
Conservation and Behaviour<br />
and the reduced fecundity of females due to a behavioural<br />
Allee effect.<br />
Hawaiian monk seals<br />
In contrast to the saiga, small populations of the endangered<br />
Hawaiian monk seal Monachus schauinslandi have<br />
unusually high numbers of adult males. Although the reasons<br />
for the male-biased sex ratios seen in this species are<br />
unknown, small population size may be a contributing<br />
factor (Starfield et al., 1995). Male-biased sex ratios are<br />
thought to increase the probability of mobbing, a form of<br />
intra-sexual competition in which individuals or groups of<br />
males may injure and/or kill adult females and immature<br />
pups during mating attempts. Mobbing exacerbates the<br />
male-biased sex ratios by further reducing the numbers of<br />
breeding females. Conservation behaviourists have effectively<br />
reduced mobbing by translocating surplus males to<br />
islands unoccupied by seals, allowing females to mate and<br />
rear pups more successfully.<br />
Managing Human<strong>–</strong>Wildlife Conflict<br />
Utilitarian value through sustainable use of wildlife by humans<br />
is thought to be an essential incentive for nature<br />
preservation (Hutton and Leader-Williams, 2003). An understanding<br />
of how and why animals behave the way they<br />
do will be instrumental in managing the population-level<br />
consequences of the direct and indirect effects of harvesting<br />
animals. Examples of wildlife (or their habitat) use by humans<br />
that benefit from a conservation behaviour approach<br />
include bycatch reduction in marine fisheries, management<br />
of ecotourism, reducing crop raiding by elephants and disruption<br />
of animal communication by pollution.<br />
Bycatch reduction<br />
An estimated 27 million tons of non-target species are discarded<br />
annually in fisheries worldwide (Alverson et al.,<br />
1994). Of greatest concern is the destruction of individuals<br />
of long-lived, slow breeding (also called ‘K-selected’) species<br />
such as porpoises, sea turtles and albatrosses, whose<br />
populations cannot recover quickly from high rates of<br />
adult mortality. Understanding how and why non-target<br />
species are being caught in these fisheries are intrinsically<br />
behavioural questions. The development of sensory deterrents,<br />
such as acoustic ‘pingers’ to warn porpoises away<br />
from gill nets (Kastelein et al., 2006), and olfactory repellants<br />
to keep sea birds from pursuing baited hooks in longline<br />
fisheries (Pierre and Norden, 2006) will require<br />
continued study to understand how their effectiveness is<br />
impacted by individual learning, and how evolutionary<br />
history determines which species respond to these bycatch<br />
reduction methods. Bycatch reduction devices that allow<br />
non-target species to exit trawl nets without reducing the<br />
capture of the target organism will require further investigation<br />
of the sensory world of these species and the<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net
decision rules they use in response to those stimuli. Differential<br />
light preferences of target and non-target species may<br />
also be useful in developing more selective fisheries<br />
(Marchesan et al., 2005). See also: Reproductive Strategies<br />
Ecotourism<br />
Generally, ecotourism is thought to change wildlife behaviour,<br />
but some studies have been able to link human disturbance<br />
with positive effects on animal populations. for<br />
example, the presence of bear viewing tours in British Columbia<br />
temporarily displaces adult male brown bears<br />
U. arctos, which in turn allows for females with cubs to<br />
spend more time fishing and foraging due to the fact that<br />
female bears appear to be less vigilant around humans than<br />
adult males. The displacement of large males by the presence<br />
of ecotourists is thought to create a temporal refuge<br />
that enhances feeding opportunities for subordinate age/<br />
sex classes (Nevin and Gilbert, 2005). More commonly investigators<br />
conclude that human disturbance has a neutral<br />
or only slightly negative effect on wildlife. Minimal reaction<br />
to human observers occurs in eastern chimpanzees<br />
(Pan troglodytes schweinfurthi) in the Kibale Forest,<br />
Uganda, provided that tourists visit in small groups<br />
(Johns, 1996). Magellanic penguins Spheniscus magellanicus<br />
in Argentina show both behavioural and physiological<br />
habituation to tourist visitation (Walker et al.,<br />
2006). It is not known, however, if the ontogenetic habituation<br />
of these penguins serves an adaptive function. What<br />
if becoming tamer around humans puts penguins at greater<br />
risk of ignoring other predators? Inter-specific comparisons<br />
of stress responses to disturbance of penguin species<br />
that evolved in the absence of native terrestrial predators,<br />
such as the Galapagos penguin Spheniscus mendiculus (Figure<br />
2), with those that show fearful reactions would be<br />
helpful to predict the conservation consequences of expanding<br />
ecotourism to additional penguin species. The<br />
challenge of ecotourism has always been to combine the<br />
demands of tourists with the needs of local populations and<br />
the conservation of protected areas. Not all behavioural<br />
change by animals visited by ecotourists is harmful to<br />
population viability. Sustainable ecotourism requires that<br />
conservationists minimize disturbance that impairs the<br />
lifetime reproductive success of individual animals.<br />
Crop raiding by elephants<br />
Human<strong>–</strong>elephant conflict over crop raiding is a significant<br />
threat to elephant conservation efforts in Africa and Asia.<br />
In addition to imposing significant economic burdens on<br />
humans that farm near elephant reserves, elephants may<br />
harm or even kill humans who attempt to repel them.<br />
Conservation behaviourists are attempting to properly understand<br />
the proximate and ultimate factors that promote<br />
crop raiding. The benefits of crop raiding to the elephants<br />
are dependent on the proximity of natural habitats to the<br />
crop fields, the timing of crop ripening and the spatial distribution<br />
of wild forage items (Rode et al., 2006). Males<br />
between the ages of 10 and 14 years and between the ages of<br />
20 and 24 years are associated with crop raiding, potentially<br />
due to three factors: social independence, risk tolerance<br />
and male<strong>–</strong>male competition (Chiyo et al., 2005). In<br />
addition to practical methods for preventing elephants<br />
from reaching crop fields (such as trenches, electric fences<br />
and grass buffers), behaviourists must develop management<br />
techniques that co-opt male elephant strategies that<br />
promote conflict with their human neighbours.<br />
Pollution<br />
Conservation and Behaviour<br />
Figure 2 Sustainable ecotourism provides financial incentives to protect<br />
natural areas. Conservation behaviour studies are needed to determine if<br />
island species, such as this Galapagos penguin, habituate more easily to<br />
human disturbance because they have evolved without terrestrial predators.<br />
Photo courtesy of Clifford Ochs.<br />
Human activities contaminate nearby animal habitats with<br />
a wide range of pollutants that may interfere with the<br />
communication systems of animal species. In the presence<br />
of such information disruptors, organisms may fail to detect<br />
important environmental cues required for the acquisition<br />
of food, mates, habitats and predators. For example,<br />
pesticides and heavy metals may interfere with antipredator<br />
responses (Lu¨ rling and Scheffer, 2007). Excess<br />
artificial light (light pollution) disrupts orientation cues in<br />
hatchling marine turtles (Witherington, 1997) and other<br />
nocturnally active wildlife. The nocturnal beach mouse is<br />
unable to forage efficiently in areas illuminated by artificial<br />
lights (Bird et al., 2004). Man-made noises, also serve as<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 7
stressors to animal populations causing increased hypothalamic<br />
response resulting in physiological stress that may<br />
cause dispersal or impair reproduction.<br />
Implications for Management<br />
Problems of animal behaviour occur in most areas of conservation,<br />
and conservation behaviour is likely to become<br />
an even more pertinent tool to prevent biodiversity loss as<br />
wildlife becomes restricted to parks and preserves surrounded<br />
by anthropogenic environments (Buchholz,<br />
2007). Unfortunately most conservation biologists and<br />
wildlife managers have little training in the ethological<br />
framework or methodology that underpins conservation<br />
behaviour. Despite calls for the inclusion of behaviourists<br />
on conservation planning teams (Arcese et al., 1997), conservation<br />
behaviour remains poorly integrated with conservation<br />
biology (Caro, 2007; Angeloni et al., 2008). Our<br />
ability to protect biodiversity despite global climate<br />
change, exponential human population growth and unsustainable<br />
resource use will require that ethologists participate<br />
in conservation management, and demonstrate more<br />
effectively that the mechanisms, ontogeny, adaptive function<br />
and evolutionary history of animal behaviour have<br />
practical import to conserving nature.<br />
References<br />
Ahlering MA and Faaborg J (2006) Avian habitat management<br />
meets conspecific attraction: if you build it, will they come? Auk<br />
23: 301<strong>–</strong>312.<br />
Alberts AC, Lemm JM, Perry AM, Morici LA and Phillips JA<br />
(2002) Temporary alteration of local social structure in a<br />
threatened population of Cuban iguanas (Cyclura nubila). Behavioral<br />
Ecology and Sociobiology 51: 324<strong>–</strong>335.<br />
Alverson DL, Freeberg MH, Murawski SA and Pope JG (1994)<br />
A global assessment of fisheries bycatch and discards. FAO<br />
Fisheries Technical Paper 339, Food and Agriculture Organization<br />
of the United Nations, Rome, http://www.fao.org/docrep/003/T4890E/T4890E00.HTM<br />
Angeloni L, Schlaepfer MA, Lawler JJ and Crooks KR (2008)<br />
A reassessment of the interface between conservation and behaviour.<br />
Animal Behaviour 75: 731<strong>–</strong>737.<br />
Arcese P, Keller LF and Cary JR (1997) Why hire a behaviorist into<br />
a conservation or management team? In: Clemmons JR and<br />
Buchholz R (eds) Behavioral Approaches to Conservation in the<br />
Wild, pp. 48<strong>–</strong>71. Cambridge, UK: Cambridge University Press.<br />
Battini J (2004) When good animals love bad habitats: ecological<br />
traps and the conservation of animal populations. Conservation<br />
Biology 18: 1482<strong>–</strong>1491.<br />
Beissinger SR (1997) Integrating behavior into conservation<br />
biology: potentials and limitations. In: Clemmons JR and<br />
Buchholz R (eds) Behavioral Approaches to Conservation in the<br />
Wild, pp. 23<strong>–</strong>47. Cambridge, UK: Cambridge University Press.<br />
Berger J and Cunningham C (1994) Active intervention and conservation:<br />
Africa’s pachyderm problem. Science 263:<br />
1241<strong>–</strong>1242.<br />
8<br />
Conservation and Behaviour<br />
Bird B, Branch L and Miller D (2004) Effects of coastal lighting<br />
on foraging behavior of beach mice. Conservation Biology<br />
18: 1435<strong>–</strong>1439.<br />
Bolger DT, Newmark WD, Morrison TA and Doak DF (2008)<br />
The need for integrative approaches to understand and conserve<br />
migratory ungulates. Ecology Letters 11: 63<strong>–</strong>77.<br />
Buchholz R (2007) Behavioural biology: an effective and<br />
relevant conservation tool. Trends in Ecology and Evolution<br />
22: 401<strong>–</strong>407.<br />
Canadian Wildlife Service and U.S. Fish and Wildlife Service<br />
(2005) <strong>International</strong> Recovery Plan for the Whooping Crane,<br />
162pp. Ottawa: Recovery of Nationally Endangered Wildlife<br />
(RENEW), and Albuquerque, NM: U.S. Fish and Wildlife<br />
Service.<br />
Caro T (2007) Behavior and conservation: a bridge too far? Trends<br />
in Ecology and Evolution 22: 394<strong>–</strong>400.<br />
Castello´ n TD and Sieving KE (2005) An experimental test of matrix<br />
permeability and corridor use by an endemic understory<br />
bird. Conservation Biology 20: 135<strong>–</strong>145.<br />
Chiyo P and Cochrane E (2005) Population structure and behavior<br />
of crop-raiding elephants in Kibale National Park,<br />
Uganda. African Journal of Ecology 43: 233<strong>–</strong>241.<br />
Clevenger AP and Waltho N (2005) Performance indices to identify<br />
attributes of highway crossing structures facilitating<br />
movement of large mammals. Biological Conservation 121:<br />
453<strong>–</strong>464.<br />
Clout MN, Elliott GP and Robertson BC (2002) Effects of supplementary<br />
feeding on the offspring sex ratio of kakapo: a dilemma<br />
for the conservation of a polygynous parrot. Biological<br />
Conservation 107: 13<strong>–</strong>18.<br />
Coltman DW, O’Donoghue P, Jorgenson JT et al. (2003) Undesirable<br />
evolutionary consequences of trophy hunting. Nature<br />
426: 655<strong>–</strong>658.<br />
Courchamp F, Rasmussen GSA and Macdonald DW (2002)<br />
Small pack size imposes a trade-off between hunting and pupguarding<br />
in the painted hunting dog Lycaon pictus. Behavioral<br />
Ecology 13: 20<strong>–</strong>27.<br />
Cryan PM and Brown AC (2007) Migration of bats past a remote<br />
island offers clues toward the problem of bat fatalities at wind<br />
turbines. Biological Conservation 139: 1<strong>–</strong>11.<br />
Davis LA, Roalson EH, Cornell KL, McClanahan KD and<br />
Webster MS (2006) Genetic divergence and migration patterns<br />
in a North American passerine bird: implications for evolution<br />
and conservation. Molecular Ecology 15: 2141<strong>–</strong>2152.<br />
Diego-Rasilla FJ and Luengo RM (2004) Heterospecific call recognition<br />
and phonotaxis in the orientation behavior of the<br />
marbled newt, Triturus marmoratus. Behavioral Ecology and<br />
Sociobiology 55: 556<strong>–</strong>560.<br />
Doherty PF Jr, Sorci G, Royle JA et al. (2003) Sexual selection<br />
affects local extinction and turnover in bird communities. Proceedings<br />
of the National Academy of Sciences of the USA 100:<br />
5858<strong>–</strong>5862.<br />
Fisher HS, Swaisgood RR and Fitch-Snyder H (2003) Odor familiarity<br />
and female preferences for males in a threatened primate,<br />
the pygmy loris Nycticebus pygmaeus: applications for<br />
genetic management of small populations. Naturwissenschaften<br />
90: 509<strong>–</strong>512.<br />
Fletcher RJ Jr (2007) Species interactions and population density<br />
mediate the use of social cues for habitat selection. Journal of<br />
Animal Ecology 76: 598<strong>–</strong>606.<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net
Gagnon JW, Theimer TC, Dodd NL, Boe S and Schweinsburg RE<br />
(2007) Traffic volume alters elk distribution and highway crossings<br />
in Arizona. Journal of Wildlife Management 71: 2318<strong>–</strong>2323.<br />
Gilroy JJ and Sutherland WJ (2007) Beyond ecological traps:<br />
perceptual errors and undervalued resources. Trends in Ecology<br />
and Evolution 22: 351<strong>–</strong>356.<br />
Haddad NM (1999) Corridor use predicted from behaviors at<br />
habitat boundaries. American Naturalist 153: 215<strong>–</strong>227.<br />
Hutton JM and Leader-Williams N (2003) Sustainable use and<br />
incentive-driven conservation: realigning human and conservation<br />
interests. Oryx 37: 215<strong>–</strong>226.<br />
Johns BG (1996) Responses of chimpanzees to habituation and<br />
tourism in the Kibale Forest, Uganda. Biological Conservation<br />
78: 257<strong>–</strong>262.<br />
Kastelein RA, Jennings N, Verboom WC, de Haan D and Schooneman<br />
NM (2006) Differences in the response of a striped<br />
dolphin (Stenella coeruleoalba) and a harbour porpoise<br />
(Phocoena phocoena) to an acoustic alarm. Marine Environmental<br />
Research 61: 363<strong>–</strong>378.<br />
Komdeur J, Castle G, Huffstad A et al. (1995) Transfer experiments<br />
of Seychelles warblers to new islands: changes in dispersal<br />
and helping behaviour. Animal Behaviour 49: 695<strong>–</strong>708.<br />
Laurenson MK, Wielebnowski N and Caro TM (1995) Extrinsic<br />
factors and juvenile mortality in cheetahs. Conservation Biology<br />
9: 1329<strong>–</strong>1331.<br />
Lu¨ rling M and Scheffer M (2007) Info-disruption: pollution and<br />
the transfer of chemical information between organisms. Trends<br />
in Ecology and Evolution 22(7): 374<strong>–</strong>379.<br />
Marchesan M, Spoto M, Verginella L and Ferrero EA (2005)<br />
Behavioural effects of artificial light on fish species of commercial<br />
interest. Fisheries Research 73: <strong>17</strong>1<strong>–</strong>185.<br />
Mata C, Hervás I, Herranz J, Suárez F and Malo JE (2005) Complementary<br />
use by vertebrates of crossing structures along a<br />
fenced Spanish motorway. Biological Conservation 124: 397<strong>–</strong>405.<br />
McDonald RA, Delahay RJ, Carter SP, Smith GC and Cheeseman<br />
CL (2008) Perturbing implications of wildlife ecology for<br />
disease control. Trends in Ecology and Evolution 23: 53<strong>–</strong>56.<br />
Milner-Gulland EJ (2003) Reproductive collapse in saiga antelope<br />
harems. Nature 422: 135.<br />
Nevin OT and Gilbert BK (2005) Perceived risk, displacement and<br />
refuging in brown bears: positive impacts of ecotourism? Biological<br />
Conservation 121: 611<strong>–</strong>622.<br />
Ng SJ, Dole JW, Sauvajot RM, Riley SPD and Valone TJ (2004)<br />
Use of highway undercrossings by wildlife in southern California.<br />
Biological Conservation 115: 499<strong>–</strong>507.<br />
Parker PG and Waite TA (1997) Mating systems, effective<br />
population size, and conservation of natural populations. In:<br />
Clemmons JR and Buchholz R (eds) Behavioral Approaches to<br />
Conservation in the Wild, pp. 243<strong>–</strong>261. Cambridge, UK: Cambridge<br />
University Press.<br />
Pierre JP and Norden WS (2006) Reducing seabird bycatch in<br />
longline fisheries using a natural olfactory deterrent. Biological<br />
Conservation 130: 406<strong>–</strong>415.<br />
Quader S (2005) Mate choice and its implications for conservation<br />
and management. Current Science 89: 1220<strong>–</strong>1229.<br />
Rittenhouse TAG and Semlitsch RD (2006) Grasslands as movement<br />
barriers for a forest-associated salamander: migration<br />
behavior of adult and juvenile salamanders at a distinct habitat<br />
edge. Biological Conservation 131: 14<strong>–</strong>22.<br />
Rode K, Chiyo P, Chapman C and McDowell L (2006) Nutritional<br />
ecology of elephants in Kibale National Park, Uganda,<br />
and its relationship with crop-raiding behavior. Journal of<br />
Tropical Ecology 22: 441<strong>–</strong>449.<br />
Rosenberg DK, Noon BR, Megahan JW and Meslow EC (1998)<br />
Compensatory behavior of Ensatina eschscholtzii in biological<br />
corridors: a field experiment. Canadian Journal of Zoology<br />
76: 1<strong>17</strong><strong>–</strong>133.<br />
Schlaepfer MA, Runge MC and Sherman PW (2002) Ecological<br />
and evolutionary traps. Trends in Ecology and Evolution<br />
<strong>17</strong>: 474<strong>–</strong>480.<br />
Starfield AM, Roth JD and Ralls K (1995) Mobbing in Hawaiian<br />
monk seals (Monachus schauinlandi): the value of simulation<br />
modeling in the absence of apparently crucial data. Conservation<br />
Biology 9: 166<strong>–</strong><strong>17</strong>4.<br />
Thomas GH, Lanctot RB and Sze´ kely T (2006) Can intrinsic factors<br />
explain population decline in North American breeding<br />
shorebirds? A comparative analysis. Animal Conservation<br />
9: 252<strong>–</strong>258.<br />
Walker BG, Boersma PD and Wingfield JC (2006) Habituation of<br />
adult magellanic penguins to human visitation as expressed<br />
through behavior and corticosterone secretion. Conservation<br />
Biology 20: 146<strong>–</strong>154.<br />
Ward MP and Schlossberg S (2004) Conspecific attraction and the<br />
conservation of territorial songbirds. Conservation Biology<br />
18: 519<strong>–</strong>525.<br />
Whiting SD, Murray W, Macrae I et al. (2008) Non-migratory<br />
breeding by isolated green sea turtles (Chelonia mydas) in the<br />
Indian Ocean: biological and conservation implications.<br />
Naturwissenschaften 95: 355<strong>–</strong>360.<br />
Whitman K, Starfield AM, Quadling HS and Packer C (2004)<br />
Sustainable trophy hunting of African lions. Nature 428:<br />
<strong>17</strong>5<strong>–</strong><strong>17</strong>8.<br />
Witherington BE (1997) The problem of photopollution for sea<br />
turtles and other nocturnal animals. In: Clemmons JR and<br />
Buchholz R (eds) Behavioral Approaches to Conservation in the<br />
Wild, pp. 303<strong>–</strong>328. Cambridge, UK: Cambridge University<br />
Press.<br />
Zedrosser A, Støen O, Saebø S and Swenson JE (2007) Should I<br />
stay or should I go? Natal dispersal in the brown bear. Animal<br />
Behaviour 74: 369<strong>–</strong>376.<br />
Further Reading<br />
Conservation and Behaviour<br />
Caro T (1998) Behavioral Ecology and Conservation Biology. New<br />
York: Oxford University Press, New York.<br />
Festa-Bianchet M and Apollonio M (2003) Animal Behavior and<br />
Wildlife Conservation. Washington, DC: Island Press.<br />
Gosling LM and Sutherland WJ (2000) Behaviour and Conservation.<br />
New York: Cambridge University Press.<br />
Harcourt AH and Stewart KJ (2007) Socio-ecology and gorilla<br />
conservation. Chap. 14. In: Gorilla Society. Chicago: University<br />
of Chicago Press.<br />
Sutherland WJ (1996) From Individual Behaviour to Population<br />
Ecology. Oxford Series in Ecology and Evolution. New York:<br />
Oxford University Press.<br />
ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 9
Behavioural biology: an effective and<br />
relevant conservation tool<br />
Richard Buchholz<br />
Opinion TRENDS in Ecology and Evolution Vol.22 No.8<br />
Department of Biology, University of Mississippi, University, MS 38677-1848, USA<br />
‘Conservation behaviour’ is a young discipline that<br />
investigates how proximate and ultimate aspects of<br />
the behaviour of an animal can be of value in preventing<br />
the loss of biodiversity. Rumours of its demise are<br />
unfounded. Conservation behaviour is quickly building<br />
a capacity to positively influence environmental decision<br />
making. The theoretical framework used by animal behaviourists<br />
is uniquely valuable to elucidating integrative<br />
solutions to human-wildlife conflicts, efforts to reintroduce<br />
endangered species and reducing the deleterious<br />
effects of ecotourism. Conservation behaviourists must<br />
join with other scientists under the multidisciplinary<br />
umbrella of conservation biology without giving up on<br />
their focus: the mechanisms, development, function and<br />
evolutionary history of individual differences in behaviour.<br />
Conservation behaviour is an increasingly relevant<br />
tool in the preservation of nature.<br />
The origins of conservation behaviour<br />
Behavioural biology did not rank among the fields included<br />
under the multidisciplinary umbrella of conservation<br />
biology when this new science was created in 1985 [1].<br />
Perhaps as a consequence, behavioural study was not<br />
incorporated into the first conservation biology textbooks<br />
and conservation was not mentioned in the animal behaviour<br />
texts of the era [2]. It required another decade for the<br />
nascent field of ‘conservation behaviour’ to coalesce from<br />
symposia, workshops and the resulting multi-authored<br />
volumes that appeared after the mid-1990s (citations in<br />
[3]). Some have suggested that behavioural biologists<br />
became interested in conservation only because they<br />
anticipated a new source of research funding [3,4], but<br />
those of us who were graduate students at the time know<br />
the real reason that conservation behaviour arose: in the<br />
face of biodiversity loss and widespread habitat destruction,<br />
we wanted our science to be relevant to saving the<br />
natural world.<br />
Wishing for behavioural biology to be useful to<br />
conservation, however, is not evidence that it is. Indeed,<br />
some early proponents of conservation behaviour recently<br />
seem to have lost their faith in the discipline * . Likewise, I<br />
have observed that some aspiring conservation behaviourists<br />
are questioning the success of conservation behaviour<br />
at integrating with mainstream conservation efforts. Their<br />
Corresponding author: Buchholz, R. (byrb@olemiss.edu).<br />
* Caro, T. (2006) Challenges and opportunities for behavioural ecologists to save<br />
Planet Earth. 25 July Plenary session, 11th Congress of the <strong>International</strong> Society for<br />
Behavioral Ecology, Tours, France.<br />
Available online 27 June 2007.<br />
www.sciencedirect.com 0169-5347/$ <strong>–</strong> see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2007.06.002<br />
pessimism is in sharp contrast to my own view: a decade<br />
after I co-edited the first book on the subject [5], Iam<br />
pleased with recent developments in the adolescence of<br />
conservation behaviour. After only a decade of existence, it<br />
is premature to dismiss the relevancy of the conservation<br />
behaviourist to saving biodiversity. Here, I show evidence<br />
of the vibrancy of this growing field (Box 1) and then<br />
recount how the theoretical framework of conservation<br />
behaviourists is already positioning them to help solve<br />
the types of conservation issues that will be especially<br />
vexing in the coming decades.<br />
Defining conservation behaviour<br />
It is obvious that species-typical patterns of animal<br />
movement, feeding and mating must be considered in<br />
conservation planning. Thus, the debate over the role of<br />
modern animal behaviour studies in conservation is not a<br />
question of whether major differences in species comportment<br />
(e.g. seasonal migration versus year-round residency)<br />
are important to conservation planning. Instead,<br />
it is a disagreement over whether the discipline-specific<br />
training of animal behaviourists, sensu stricto, is a valuable<br />
addition to conservation action teams [6]. I have<br />
observed two possible reasons why conservation behaviour<br />
has not made major in-roads in traditional conservation<br />
biology circles: conservation ecologists believe that they<br />
already ‘do’ behaviour, and many think that animal behaviourists<br />
work at scales of minor value to protecting entire<br />
landscapes, the most cost-effective means of conservation.<br />
Behaviour thinking<br />
On the surface, ‘doing’ behaviour seems relatively easy. For<br />
example, conservation biologists might walk through a<br />
forest quantifying territorial vocalizations to census a<br />
population of songbirds. Such superficial uses of behavioural<br />
biology are no doubt useful, but they, like<br />
species-typical behavioural descriptions by natural historians<br />
[7], are not conservation behaviour. Conservation<br />
behaviour takes advantage of an investigatory framework<br />
implicit to all modern animal behaviour studies. This<br />
framework is commonly referred to as Tinbergen’s ‘four<br />
questions’, because it was first proposed by Nobel-Prizewinning<br />
ethologist Niko Tinbergen as a way to guide<br />
behavioural research [8]. Tinbergen suggested that we<br />
can ask four mutually exclusive questions about any one<br />
behaviour by considering both proximate and ultimate<br />
explanations for the cause and origin of that behaviour<br />
pattern (see Table I in Box 2). Some refer to this use of<br />
Tinbergen’s framework as ‘behaviour thinking’, and to
402 Opinion TRENDS in Ecology and Evolution Vol.22 No.8<br />
Box 1. The influence of conservation behaviour is growing<br />
This tenth anniversary of the publication of the first books on<br />
conservation behaviour [5,44] is an arbitrary date on which to assess<br />
this young discipline’s success at integrating with conservation<br />
biology. Rather than expecting conservation behaviour to be fully<br />
fledged at this point, we should look for signs that it continues to<br />
strengthen in terms of training resources, pervasiveness in the<br />
published literature, and acceptance by grant-awarding agencies<br />
and career mentors. Supportive evidence includes the following.<br />
Eighty percent of recently published animal behaviour texts<br />
address conservation in some fashion [2]. Page content that<br />
contains reference to conservation now averages 2% for texts<br />
published during the past five years, compared with 0% in the<br />
decades before that.<br />
Conservation behaviour training resources are readily available<br />
on the Internet (http://www.animalbehavior.org/Committees/<br />
Conservation). It is now easier for aspiring conservation behaviourists<br />
(and their mentors) to access background information,<br />
including a general bibliography, funding sources and graduate<br />
programs. Also available for download is ‘The Conservation<br />
Behaviorist’, a journal written for both behaviourists and nonscientist<br />
decision makers.<br />
The use of behavioural biology in published conservation studies<br />
is not uncommon; in fact, it is growing. Linklater [45] discovered<br />
that the percentage of conservation publications that mention<br />
behaviour has increased nearly threefold since the conception of<br />
conservation behaviour in the early 1990s.<br />
The EO Wilson Student Conservation Research Grant Award of<br />
the Animal Behavior Society (ABS) was created to provide an<br />
annual small grant to foster the career development of conservation<br />
behaviourists. Conservation grant awardees have<br />
studied logging effects on tree shrew mating systems, stream<br />
pollution interference with fish communication and dragonfly<br />
oviposition requirements as bioindicators of wetland quality.<br />
Perhaps more significant than these new conservation behaviour<br />
research funds are the types of research that ‘regular’ student<br />
behaviourists are doing. In 2006, 15% of the standard Student<br />
Research Awards from the ABS were for projects with conservation<br />
themes.<br />
The ‘father’ of conservation biology, Michael Soulé, said that<br />
the new discipline: ‘‘should attract and penetrate every field<br />
that could possibly benefit and protect the diversity of life’’ [46].<br />
These demographics suggest that the population of conservation<br />
behaviourists is indeed growing.<br />
proximate and ultimate explanations as ‘how’ and ‘why’<br />
questions.<br />
Proximate questions about behaviour consider how<br />
an individual is able to perform an activity. They ask<br />
about the mechanisms within an organism that make it<br />
possible for it to behave in a certain way. The proximate<br />
causes of behaviour include the sensory and endocrine<br />
mechanisms that regulate behaviour. However, we know<br />
that these mechanisms can be modified by individual<br />
experience; thus, we must consider the proximate origins<br />
of behaviour as well (i.e. how learning modifies behaviour).<br />
Ultimate questions about behaviour, on the other hand,<br />
ask why animal species have evolved the proximate systems<br />
that enable them to behave the way they do. The<br />
ultimate cause of a behaviour must explain how it helps the<br />
individual survive and reproduce. If we ponder the ultimate<br />
origin of that same behaviour pattern, we are examining<br />
its evolutionary history by comparing how that<br />
behaviour differs across a group of related species. Tinbergen’s<br />
framework is most easily explained by applying it to a<br />
behavioural example (Box 2).<br />
www.sciencedirect.com<br />
Box 2. Understanding Tinbergen’s framework for studying<br />
behaviour<br />
Tinbergen’s four complementary approaches to studying a behaviour<br />
pattern are implicit to modern animal behaviour studies (see<br />
Table I). This framework is most easily explained by applying it to an<br />
example of human behaviour: automobile drivers stop their cars<br />
when a traffic light turns red.<br />
Mechanisms<br />
We can ask how automobile drivers are able to perceive the red<br />
colour of the traffic light, the pattern of neural depolarization that<br />
allows a decision to be made in the central nervous system and how<br />
that decision is conveyed to the muscles that control the placement<br />
of the foot on the brake. The problem of ‘road rage’ suggests that<br />
latency to apply the brake may be affected by the hormonal milieu of<br />
the driver.<br />
Ontogeny<br />
Humans are not born knowing how to drive an automobile. We can<br />
ask questions about how humans go about learning to stop at a red<br />
light and how it is that development affects the ability to learn. For<br />
example, accident statistics suggest that teenagers and senior<br />
citizens may have greater difficulty stopping at red lights than other<br />
drivers.<br />
Function<br />
We can investigate why stopping at red lights allows drivers to live<br />
longer and reproduce successfully. The legal and financial costs and<br />
benefits that contribute to the individual fitness of drivers are<br />
worthy of study. We might also consider why some individuals are<br />
more likely to cooperate with other drivers in their society.<br />
Phylogeny<br />
An investigation of the history of traffic signals might tell us how<br />
responses to red lights have changed over time. In humans, we can<br />
look at how populations have independently developed means of<br />
making intersections safer. If we compare humans to other species,<br />
we might ask how the colour red has evolved as a warning signal.<br />
Table I. Tinbergen’s framework for organizing the study of<br />
behaviour proposes that any behaviour pattern should be<br />
investigated from four complementary perspectives<br />
Cause Origin<br />
Proximate Mechanisms Ontogeny<br />
Ultimate Function Phylogeny<br />
The beauty of Tinbergen’s four questions is that they<br />
force us to consider multiple, complementary explanations<br />
for the same behaviour [9]. In terms of saving biodiversity,<br />
this framework is especially effective in situations in which<br />
the behavioural adaptations of wildlife are at odds with<br />
anthropogenic landscapes [10]. Conservationists might<br />
eventually stumble upon the need for knowing both<br />
the ‘how’ and the ‘why’ of animal behaviour (Box 3), but<br />
it would be much more cost effective and time efficient<br />
if Tinbergen’s framework was applied at the onset of<br />
conservation research.<br />
Where does conservation behaviour fit?<br />
It would be preposterous to claim that behavioural biology<br />
is the conservation ‘cure all’. Behavioural study is neither<br />
appropriate nor a priority at all levels of conservation<br />
action (Table 1 and [11]). The need for behaviourists will<br />
be greatest when we are confronted with the challenge of<br />
maintaining animals marooned in protected islands of<br />
habitat, isolated in a sea of humanity, and managing
Box 3. Harbour porpoises and gill nets: application of<br />
Tinbergen’s framework<br />
In 1994 y , Marquez and I used the example of a ‘real-world’<br />
conservation problem, the drowning of harbour porpoises Phocoena<br />
phocoena following entanglement in the gill nets of commercial<br />
fisherman, to explain how Tinbergen’s four questions could guide<br />
conservation research. Read and colleagues [47] had called for<br />
improved understanding of the behaviour of harbour porpoises<br />
around gill nets. Incidental bycatch of harbour porpoises in the US<br />
Gulf of Maine groundfish gill net fishery alone averaged over 2000<br />
individuals per year in the early 1990s, more than twice the<br />
allowable take rate.<br />
For heuristic purposes, we suggested that behavioural aspects<br />
of gill net bycatch research could be approached simultaneously<br />
from mechanistic, ontogenetic, functional and phylogenetic<br />
perspectives. Revisiting the problem now, it is rewarding to see<br />
that anti-entanglement research evolved within Tinbergen’s<br />
framework.<br />
Harbour porpoises appear to be in the vicinity of nets because<br />
nets are placed where porpoise prey are abundant [48]. Making nets<br />
from material that is more reflective of porpoise sonar did reduce<br />
bycatch significantly, but perhaps only because the nets were stiffer<br />
and less likely to entangle the porpoises [49]. Adding acoustic<br />
alarms, called ‘pingers’, to gill nets reduces the bycatch mortality of<br />
harbour porpoises [50]. Harbour porpoises show aversive reactions<br />
to pinging, including spatial avoidance and increasing respiration<br />
rate [51].<br />
Other small cetaceans, however, do not necessarily react the<br />
same way [51]. In a field experiment, bottlenose dolphins Tursiops<br />
truncatus (Figure I) appeared to use fish caught in the gill net as a<br />
food source, but dolphin groups were less likely to approach the<br />
net’s ‘zone of vulnerability’ when the pingers were activated [50].<br />
There is considerable interest in investigating how these species will<br />
react to long-term use of acoustic alarms. Will cetaceans become<br />
sensitized to pingers and avoid them more often, or will pingers<br />
become a ‘dinner bell’ of sorts, attracting porpoises and dolphins to<br />
pre-caught fish?<br />
Although the gill net responses of relatively few cetaceans<br />
have been studied so far, some researchers have hypothesized<br />
that vulnerability to natural predators, such as sharks and killer<br />
whales Orcinus orca, might explain species differences in the<br />
efficacy of using acoustic alarms to reduce incidental bycatch<br />
mortality.<br />
I do not mean to suggest that these studies are the result of our<br />
Society for Conservation Biology poster; these works developed<br />
organically from the need to solve the gill net entanglement<br />
problem. Nevertheless, this conservation problem demonstrates<br />
how a priori use of a conservation behaviour framework would<br />
foster an efficient conservation research plan [52].<br />
Figure I. Bottlenose dolphins are attracted to the fish caught in commercial gill<br />
nets (photo by Jill Frank).<br />
y Marquez, M. and Buchholz, R. (1994) An Ethological Framework for<br />
Conservation Biology. Poster at the annual meeting of the Society for<br />
Conservation Biology, University of Guadalajara, Jalisco, Mexico.<br />
www.sciencedirect.com<br />
Opinion TRENDS in Ecology and Evolution Vol.22 No.8 403<br />
the inevitable conflicts between the daily needs of humans<br />
and other animals. It is not pleasant to think of a future in<br />
which most of nature is drastically altered by humans, but<br />
the reality is undeniable [12].<br />
Global warming<br />
Conservation behaviourists are developing predictive tools<br />
for understanding which species in pristine communities<br />
will need behavioural management when their habitats<br />
are altered by direct and indirect human disturbance. Paz<br />
y Miño [13] has termed these circumstances ‘behavioural<br />
unknowns’ (after Myers ‘environmental unknowns’ [14]).<br />
The most pressing of these, perhaps, is the need to understand<br />
how the mechanisms, ontogeny, adaptiveness and<br />
phylogenetic diversity of animal behaviour will respond to<br />
climate change due to global warming. Conservation behaviourists<br />
have already begun to develop a body of literature<br />
that addresses behavioural responses to rapid climatic<br />
alterations (Box 4).<br />
Restoring balance to ecosystems<br />
Other behavioural unknowns may have less to do with how<br />
we are destroying habitat and more to do with our attempts<br />
to restore ecological integrity to human-altered landscapes.<br />
The reintroduction of large carnivores appears to<br />
enhance ecosystem biodiversity and stability [15]. Nonetheless,<br />
game managers fear that huntable (by humans)<br />
prey species will be decimated because of their naiveté<br />
after generations without non-human predation. By experimentally<br />
investigating the anti-predator responses of<br />
ungulates to predator cues before and after carnivore<br />
reintroduction, Berger [16] found that prey typically return<br />
to ‘normal’ anti-predator strategies within one generation<br />
of carnivore return. If politically feasible, conservation<br />
behaviour data such as these will be used to support the<br />
repatriation of carnivores so that balanced ecosystems are<br />
restored.<br />
Ecological prediction is a mainstay of traditional<br />
conservationists [<strong>17</strong>]. Game theory, and other skills in<br />
the behaviourist’s toolbox that take advantage of individual<br />
differences in behaviour, are currently being used to<br />
anticipate the population consequences of management<br />
options [18]. In addition to predicting how animals will<br />
respond to anthropogenic disturbances, animal behaviourists<br />
are influencing conservation management directly.<br />
Conservation behaviourists in action<br />
To date, the contributions of conservation behaviourists<br />
are much more than theoretical. Conservation behaviourists<br />
are already involved in hands-on efforts to restore and<br />
protect animal species. Here, I briefly review some recent<br />
contributions of note.<br />
Species reintroduction<br />
Although the future of biodiversity is in the wild, captive<br />
breeding of endangered species is sometimes an irreplaceable<br />
component of the conservationist’s toolbox [11,19].<br />
Conservation behaviourists have concentrated on two<br />
important aspects of captive propagation of endangered<br />
species: preventing ‘captive selection’ (i.e. maladaptive<br />
heritable changes in behaviour), and behavioural training
404 Opinion TRENDS in Ecology and Evolution Vol.22 No.8<br />
Table 1. Behavioural biology is relevant to multiple conservation contexts<br />
Conservation context Conservation tool Example of use<br />
Preventing biodiversity loss Reserve design a<br />
[37]<br />
Ecosystem management b<br />
[15]<br />
Population viability analysis b<br />
[43]<br />
Compromises with economic development Sustainable use b<br />
[29]<br />
Species and habitat restoration Field recovery of endangered species a<br />
[23]<br />
Captive breeding and reintroduction a<br />
[20]<br />
Ecosystem restoration a<br />
a<br />
Tools with current behaviour usage.<br />
[16]<br />
b<br />
Tools that will need more behaviourist input as habitat degradation continues (after Beissinger [11]).<br />
Box 4. Etho-conservation tackles global warming<br />
The weather systems of the Earth are expected to become more<br />
extreme, perhaps suddenly and with great spatial heterogeneity,<br />
owing to the atmospheric retention of heat energy from anthropogenic<br />
‘greenhouse gas’ production. How animals might react to<br />
climate change is one of the many ‘behavioural unknowns’ caused<br />
by environmental degradation [13] that is being investigated using<br />
Tinbergen’s four questions.<br />
Mechanisms<br />
Cactus-living Drosophila species might respond to climate<br />
change via selection on the timing of their active period in the<br />
circadian clock mechanism [53]. By becoming active at cooler<br />
times of the day, they are able to avoid deleterious exposure to<br />
heat extremes. The physiological stress responses of vertebrate<br />
organisms are likely to be dependent on individual differences in<br />
social status and the social function of dominance in that species<br />
[31,54].<br />
Ontogeny<br />
Culturally determined foraging movements of sperm whale Physeter<br />
macrocephalus clans might make some song clans predictably<br />
susceptible to changes in ocean-current-dependent food sources<br />
[55]. For species with temperature-mediated sex determination,<br />
such as turtles and crocodilians, behavioural commitment to<br />
reusing traditional nest sites will impede adaptive population-level<br />
responses in these long-lived animals. For example, if global<br />
warming occurs as forecast, modelling suggests that natal site<br />
philopatry by egg-laying painted turtles will condemn them to<br />
producing such biased sex ratios that the population becomes<br />
inviable [43].<br />
Function<br />
Species with poor mobility might experience rapid evolutionary<br />
behavioural adaptation in response to micro-climatic divergence.<br />
For example, wood frog tadpoles in shaded, cool ponds appear to<br />
evolve adaptive thermal preferences quickly [56]. Populations in<br />
warmer sun-lit ponds, however, seem to lack growth-benefiting<br />
thermal preferences, suggesting that global warming would cause<br />
meta-population sinks with predictable localized patterns of species<br />
extirpation. Hawksbill turtle Eretmochelys imbricata females might<br />
be able to adjust to a hotter climate through existing preferences for<br />
tree-shaded beach nesting sites. Unfortunately, beach deforestation<br />
may prevent this endangered species from avoiding climateinduced<br />
biased sex ratios [57].<br />
Phylogeny<br />
Sexual selection may have an impact on the response of animals<br />
to climate change. The degree of advancement of spring migration<br />
in birds is associated with the strength of female choice across<br />
species [58]. Colonizing warmer habitats appears to release<br />
energetic constraints on sexual selection in dark-eyed juncos Junco<br />
hyemalis [59]. The relaxation of cold climate extremes might select<br />
against isolating mechanisms, such as reproductive diapause in<br />
mustelids [60].<br />
www.sciencedirect.com<br />
and management to maximize post-release survival of<br />
reintroduced individuals.<br />
Often, captive conditions impose different selection<br />
pressures on animal genomes than natural selection,<br />
resulting in behaviour that is advantageous to the survival<br />
and reproduction of individuals in captivity, but maladaptive<br />
should they be reintroduced to the wild. For example,<br />
captivity selects for aggressive behaviour in place of foraging<br />
behaviour in breeding colonies of the endangered<br />
butterfly splitfin fish Ameca splendens [20]. Behaviourists<br />
are helping to modify aquaculture programs to produce fish<br />
that will forage rather than fight when released to their<br />
restored habitat.<br />
In other cases, the protected nature of captive<br />
environments might allow genetic release from directional<br />
selection. The escape behaviour of captive-bred oldfield<br />
mice Peromyscus polionotus, for example, is never subjected<br />
to selection by owls, stoats, snakes or any of the<br />
predators that normally threaten the survival of free-living<br />
rodents. As a result, they have long escape latencies that<br />
would make them highly susceptible to predators in the<br />
wild [21]. By considering the genetic mechanisms underlying<br />
variation in behaviour, McPhee and Silverman [22]<br />
conclude that we need not abandon reintroduction of such<br />
individuals. Their solution is to simply use the variance in<br />
escape behaviour to recalculate the number of released<br />
animals sufficient to ensure a surviving nucleus of breeders.<br />
By understanding the genetic mechanisms underlying<br />
behaviour, conservation behaviourists are able to<br />
provide release and survival estimates that are more<br />
realistic. This would thus engender more reasonable expectations<br />
and cost planning by wildlife managers and<br />
political decision-makers, and more patience for success<br />
from the general public.<br />
Similarly careful behavioural preparation of captiveraised<br />
animals will lessen animal welfare concerns over<br />
the poor survival of individuals released to the wild for<br />
conservation purposes. Anti-predator training of captivereared<br />
prairie dogs Cynomys ludovicianus appears to<br />
increase survival upon reintroduction [23]. But the value<br />
of behavioural manipulation is not limited to potential prey<br />
species. Poor attention to the social management of African<br />
wild dog Lycaon pictus groups during ‘soft’ releases to the<br />
wild might explain several costly reintroduction failures<br />
[24]. If individuals are chosen to minimize dominance<br />
conflicts among members of the released wild dog pack,<br />
the reintroduced animals are more likely to behave cooperatively<br />
and hopefully survive to reproduce.
Conservation behaviour and natural populations<br />
There are two major areas of inquiry in which ‘behaviour<br />
thinking’ is already being applied to the conservation of<br />
wild populations: improving population viability by adaptively<br />
managing individual survival and reproductive success<br />
in isolated populations, and identifying and managing<br />
deleterious effects of ecotourism.<br />
Managing survival and reproduction<br />
Re-establishing prairie dog populations in protected areas<br />
is important to ecosystem functioning [25], and is crucial to<br />
efforts to establish a prey base to grow viable populations<br />
of the highly endangered black-footed ferret Mustela<br />
nigripes. Behaviourist Debra Shier took advantage of existing<br />
behavioural information on the adaptive nature of kin<br />
groups to demonstrate experimentally that translocating<br />
entire wild prairie dog families achieves conservation goals<br />
more successfully and efficiently than moving unrelated<br />
animals [26]. The red-cockaded woodpecker Picoides borealis<br />
is another endangered US species that has benefited<br />
from a behaviourist’s understanding of kin recognition<br />
mechanisms. Wallace and I showed that, by exchanging<br />
nestlings between nesting cavities at an age young enough<br />
that parents had not yet learned to identify their offspring,<br />
we could overcome genetic isolation in a fragmented<br />
habitat without reducing survivorship of translocated<br />
individuals [27].<br />
Behaviourists have shown how evolutionary conflicts<br />
among individual animals can give us insight into management<br />
methods that would not be apparent to a<br />
traditionally trained wildlife ecologist or conservation<br />
geneticist. When it comes to population viability, larger<br />
population size is usually better. Unfortunately, small<br />
populations do not always have room to grow. A case in<br />
point is the threatened Cuban iguana Cyclura nubile<br />
population confined to the US Naval base at Guantanamo<br />
Bay, Cuba. In the absence of opportunities to<br />
increase the overall lizard population, Alberts et al.<br />
[28] showed how the negative impact of male dominance<br />
on population-wide genetic variation (N e) could be overcome<br />
through behavioural management. Temporarily<br />
removing dominant males allowed other adult males to<br />
obtain mates.<br />
Sustainable ecotourism<br />
Ecotourism, the other doyen of sustainable-use advocates,<br />
must also consider the behaviourally mediated impact of<br />
human disturbance on population viability. Descriptive<br />
behavioural studies are likely to find that wildlife change<br />
their behaviour in response to human visitors, but that<br />
changes in behaviour are not necessarily bad for the<br />
animal under observation. For example, although foraging<br />
brown bears Ursus arctos alter their feeding activities so<br />
that they can be vigilant in the presence of tourists, this<br />
behavioural change has no apparent effects on body condition<br />
[29]. Therefore, human disturbance probably does<br />
not translate into reduced survival and lower population<br />
viability in this case. Nesting Magellanic penguins Spheniscus<br />
magellanicus appear to habituate to frequent tourists,<br />
but previously undisturbed colonies are markedly<br />
stressed by the arrival of human visitors [30]. Because<br />
www.sciencedirect.com<br />
Opinion TRENDS in Ecology and Evolution Vol.22 No.8 405<br />
stressed animals are likely to experience tradeoffs in<br />
reproductive investment or survival [31], opening new<br />
colonies to tourism is not recommended without careful<br />
determination of the human activities that cause the most<br />
stress to nesting penguins. The complexity of the interspecific<br />
and intraspecific responses of wild animals to<br />
anthropogenic and natural stressors benefits from being<br />
managed in a conservation behaviour framework.<br />
Should conservation behaviour conform to the<br />
emphases of conservation biology?<br />
I think the only way that behavioural biology makes<br />
sense for conservation is if we retain our unique<br />
perspective on animal diversity. The ecologists that<br />
helped found the Society for Conservation Biology [32]<br />
did not abandon island biogeography theory to work in<br />
conservation; they applied the concept to the habitat<br />
islands that are nature reserves. Likewise, population<br />
geneticists did not ignore Hardy<strong>–</strong>Weinberg equilibrium<br />
to promote the conservation of bottlenecked populations;<br />
rather, they found ways to inform us of the practical<br />
importance of theoretical genetics. Conservation behaviourists<br />
are conservation biologists; we should be<br />
constructively critical of endangered species recovery<br />
plans [33]. Decisions are often made with very little or<br />
faulty evidence. For example, decisions to allow habitat<br />
destruction in the dwindling range of the endangered<br />
Florida panther Puma concolor coryi were based on<br />
questionable evidence and the illogical opinion of one<br />
researcher that this subspecies is a forest obligate (for a<br />
shocking review, see [34]).<br />
It has become a conservation platitude of sorts that<br />
conservation behaviourists must ‘‘translate behavior into<br />
currencies relevant to conservation at large spatial scales’’<br />
[11]. But conflicts with large carnivores and other species of<br />
potential harm to humans attract much popular press<br />
and political interest. It is the young panther that<br />
disperses 350 miles [34] or the few African elephants<br />
Loxodonta africana that invade villages or kill rhinos<br />
[35,36] that threaten to scuttle goodwill for conservation,<br />
not the average behaviour of the population. It is ridiculous<br />
to suggest that individualistic responses of animals are<br />
unimportant to conservation.<br />
Conservation behaviourists will rise to the challenge<br />
of the important discoveries being made at the interface<br />
of conservation ecology and genetics. For example, Riley<br />
and colleagues [37] recently used microsatellite analysis<br />
combined with radiotelemetry to discover that carnivore<br />
territories tend to pile up at habitat edges along an<br />
automotive freeway in California, USA. The intense<br />
social challenges faced by individual bobcats Lynx rufus<br />
and coyotes Canis latrans attempting to disperse<br />
through the gauntlet of territories concentrated near<br />
large roadways accentuated the limits on gene flow<br />
imposed by the physical barrier of the road itself. Populations<br />
on either side of the highway show evidence of<br />
genetic differentiation as a result. The problem of<br />
‘territory piling’ [37] along roadways is one well suited<br />
to Tinbergen’s framework. These crowded carnivores<br />
should expect a visit from some local conservation<br />
behaviourists [38].
406 Opinion TRENDS in Ecology and Evolution Vol.22 No.8<br />
The next step<br />
Caro’s [3] charges of irrelevancy levied against conservation<br />
behaviour are genuine, but not unique to our field.<br />
There are problems with the application of any science to<br />
public policy. Academic scientists get caught up in grant<br />
writing, hypotheses testing and data collection, whereas<br />
conservation practitioners need practical advice today [39].<br />
The reverse is also true; conservation monitoring programs<br />
often take on a life of their own and do not achieve conservation<br />
objectives [40].<br />
Nearly 100 years ago, conservationist William T Hornaday<br />
[41] declared: ‘‘We will endeavor to avoid the discussion<br />
of academic questions, because the business of<br />
conservation is replete with urgent practical demands’’. I<br />
believe that it is this sort of concern, that there is an<br />
opportunity cost to the ‘theoretical’ concerns of conservation<br />
behaviour, that has led some of my colleagues to<br />
retreat from conservation behaviour. Instead, they must<br />
come to realize that we approach a time of desperate need<br />
for applied behaviourists. One need only look at costly<br />
efforts to recover species listed under the US Endangered<br />
Species Act [42] to see that, although all is not lost, all is not<br />
well either. Wildlife managers have stopped many endangered<br />
species from going extinct, but few species have<br />
recovered sufficiently to be de-listed. We can improve<br />
conservation management decisions. The behaviour of<br />
individual animals matters to conservation. The growing<br />
pains of conservation behaviour are not symptoms of dysfunction,<br />
but rather positive signs of a thriving adolescence.<br />
Conservation behaviour is relevant right now and<br />
the time is ripe for the conservation behaviourist to make a<br />
difference.<br />
Acknowledgements<br />
I am grateful to Cliff Ochs, Guillermo Paz y Miño and Suhel Quader for<br />
commenting on an earlier draft of this paper. Detailed suggestions by Dan<br />
Blumstein, Marco Festa-Bianchet and an anonymous reviewer improved<br />
the manuscript. Communications with Steve Beissinger and Joel Berger<br />
were helpful ten years ago and now too. Tim Caro suggested that I put it<br />
in writing, and I am very appreciative of his and the editors’ assistance<br />
and generous patience. Jill Frank and Glenn Parsons kindly offered the<br />
use of their dolphin photos.<br />
References<br />
1 Soulé, M.E. (1985) What is conservation biology? Bioscience 35, 727<strong>–</strong><br />
734<br />
2 Buchholz, R. (2006) Should animal behaviorists teach conservation?<br />
Conserv. Behav. 4, 3<strong>–</strong>4<br />
3 Caro, T. (2007) Behaviour and conservation: a bridge too far? Trends<br />
Ecol. Evol. (in press)<br />
4 Knight, J. (2001) If they could talk to the animals.... Nature 414, 246<strong>–</strong><br />
247<br />
5 Clemmons, J.R. and Buchholz, R., eds (1997) Behavioral Approaches<br />
to Conservation in the Wild, Cambridge University Press<br />
6 Arcese, P. et al. (1997) Why hire a behaviorist into a conservation or<br />
management team? In Behavioral Approaches to Conservation in the<br />
Wild (Clemmons, J.R. and Buchholz, R., eds), pp. 48<strong>–</strong>71, Cambridge<br />
University Press<br />
7 Greene, H.W. (2005) Organisms in nature as a central focus for biology.<br />
Trends Ecol. Evol. 20, 23<strong>–</strong>27<br />
8 Tinbergen, N. (1963) On aims and methods of ethology. Z. Tierpsychol.<br />
20, 410<strong>–</strong>433<br />
9 Blumstein, D.T. (2007) Tinbergen’s four questions. In The Encyclopedia<br />
of Applied Animal Behaviour and Welfare (Mills, D. ed), (in press), CAB<br />
<strong>International</strong><br />
www.sciencedirect.com<br />
10 Schlaepfer, M.A. et al. (2002) Ecological and evolutionary traps. Trends<br />
Ecol. Evol. <strong>17</strong>, 474<strong>–</strong>480<br />
11 Beissinger, S.R. (1997) Integrating behavior into conservation biology:<br />
potentials and limitations. In Behavioral Approaches to Conservation<br />
in the Wild (Clemmons, J.R. and Buchholz, R., eds), pp. 23<strong>–</strong>47,<br />
Cambridge University Press<br />
12 Sanderson, E.W. et al. (2002) The human footprint and the last of the<br />
wild. Bioscience 52, 891<strong>–</strong>904<br />
13 Paz-y-Miño, C. (2006) Behavioral unknowns: an emerging challenge for<br />
conservation. Conserv. Behav. 4, 2<br />
14 Myers, N. (1995) Environmental unknowns. Science 269, 358<strong>–</strong>360<br />
15 Fortin, D.L. et al. (2005) Wolves influence elk movements: behavior<br />
shapes a trophic cascade in Yellowstone National Park. Ecology 86,<br />
1320<strong>–</strong>1330<br />
16 Berger, J. Carnivore repatriation and Holarctic prey: narrowing the<br />
deficit in ecological effectiveness. Conserv. Biol. (in press)<br />
<strong>17</strong> Purvis, A. et al. (2000) Predicting extinction risk in declining species.<br />
Proc. Biol. Sci. 267, 1947<strong>–</strong>1952<br />
18 Sutherland, W.J. (2006) Predicting the ecological consequences of<br />
environmental change: a review of the methods. J. Appl. Ecol. 43,<br />
599<strong>–</strong>616<br />
19 Snyder, N.F.R. et al. (1996) Limitations of captive breeding in<br />
endangered species recovery. Conserv. Biol. 10, 338<strong>–</strong>348<br />
20 Kelley, J.L. (2006) Captive breeding promotes aggression in an<br />
endangered Mexican fish. Biol. Conserv. 133, 169<strong>–</strong><strong>17</strong>7<br />
21 McPhee, M.E. (2004) Generations in captivity increases behavioral<br />
variance: considerations for captive breeding and reintroduction<br />
programs. Biol. Conserv. 115, 71<strong>–</strong>77<br />
22 McPhee, M.E. and Silverman, E.D. (2004) Increased behavioral<br />
variation and the calculation of release numbers for reintroduction<br />
programs. Conserv. Biol. 18, 705<strong>–</strong>715<br />
23 Shier, D.M. and Owings, D.H. (2006) Effects of predator training on<br />
behavior and post-release survival of captive prairie dogs (Cynomys<br />
ludovicianus). Biol. Conserv. 132, 126<strong>–</strong>135<br />
24 Gusset, M. et al. (2006) Divided we fail: the importance of social<br />
integration for the re-introduction of endangered African wild dogs<br />
(Lycaon pictus). J. Zool. 270, 502<strong>–</strong>511<br />
25 Davidson, A.D. and Lightfoot, D.C. (2006) Keystone rodent<br />
interactions: prairie dogs and kangaroo rats structure the biotic<br />
composition of a desertified grassland. Ecography 29, 755<strong>–</strong>765<br />
26 Shier, D.M. (2006) Effect of family support on the success of<br />
translocated blacktailed prairie dogs. Conserv. Biol. 20, <strong>17</strong>80<strong>–</strong><strong>17</strong>90<br />
27 Wallace, M.T. and Buchholz, R. (2001) Translocation of red-cockaded<br />
woodpeckers by reciprocal fostering of nestlings. J. Wildl. Manage. 65,<br />
327<strong>–</strong>333<br />
28 Alberts, A.C. et al. (2003) Temporary alteration of local social structure<br />
in a threatened population of Cuban iguanas (Cyclura nubila). Behav.<br />
Ecol. Sociobiol. 51, 324<strong>–</strong>335<br />
29 Rode, K.D. et al. (2006) Behavioural responses of brown bears mediate<br />
nutritional effects of experimentally introduced tourism. Biol. Conserv.<br />
133, 70<strong>–</strong>80<br />
30 Walker, B.G. et al. (2006) Habituation of adult Magellanic penguins to<br />
human visitation as expressed through behaviour and corticosterone<br />
secretion. Conserv. Biol. 20, 146<strong>–</strong>154<br />
31 Wingfield, J.C. and Sapolsky, R.M. (2003) Reproduction and resistance<br />
to stress: when and how. J. Neuroendocrinol. 15, 711<strong>–</strong>724<br />
32 Ralls, K. (1997) On becoming a conservation biologist. In Behavioral<br />
Approaches to Conservation in the Wild (Clemmons, J.R. and Buchholz,<br />
R., eds), pp. 356<strong>–</strong>372, Cambridge University Press<br />
33 Boersma, P.D. et al. (2001) How good are endangered species recovery<br />
plans? Bioscience 51, 643<strong>–</strong>649<br />
34 Gross, L. (2005) Why not the best? How science failed the Florida<br />
panther. PLoS Biol. 3, e333<br />
35 Rode, K.D. et al. (2006) Nutritional ecology of elephants in Kibale<br />
National Park, Uganda, and its relationship with crop-raiding<br />
behaviour. J. Trop. Ecol. 22, 441<strong>–</strong>449<br />
36 Slowtow, R. and van Dyk, G. (2001) Role of delinquent young ‘‘orphan’’<br />
male elephants in high mortality of white rhinoceros in Pilanesberg<br />
National Park, South Africa. Koedoe 44, 85<strong>–</strong>94<br />
37 Riley, S.P.D. et al. (2006) A southern California freeway is a physical<br />
and social barrier to gene flow in carnivores. Mol. Ecol. 15, <strong>17</strong>33<strong>–</strong><strong>17</strong>41<br />
38 Blumstein, D.T. and Fernández-Juricic, E. (2004) The emergence of<br />
conservation behavior. Conserv. Biol. 18, 1<strong>17</strong>5<strong>–</strong>1<strong>17</strong>7
39 Cabin, R.J. (2007) Science-driven restoration: a square grid on a round<br />
Earth? Restor. Ecol. 15, 1<strong>–</strong>7<br />
40 Nichols, J.D. and Williams, B.K. (2006) Monitoring for conservation.<br />
Trends Ecol. Evol. 21, 668<strong>–</strong>673<br />
41 Hornaday, W.T. (1914) Wild Life Conservation in Theory and Practice,<br />
Yale University Press<br />
42 Male, T.D. and Bean, M.J. (2005) Measuring progress in US<br />
endangered species conservation. Ecol. Lett. 8, 986<strong>–</strong>992<br />
43 Morjan, C. (2003) How rapidly can maternal behavior affecting<br />
primary sex ratio evolve in a reptile with environmental sex<br />
determination? Am. Nat. 162, 205<strong>–</strong>219<br />
44 Caro, T. ed. (1998) Behavioural Ecology and Conservation Biology,<br />
Oxford University Press<br />
45 Linklater, W.L. (2004) Wanted for conservation research:<br />
behavioral ecologists with a broader perspective. Bioscience 54,<br />
352<strong>–</strong>360<br />
46 Soulé, M.E. (1986) Conservation biology and the ‘‘real world’’. In<br />
Conservation Biology: the Science of Scarcity and Diversity (Soulé,<br />
M.E., ed.), pp. 1<strong>–</strong>12, Sinauer Associates<br />
47 Read, A.J. et al. (1993) Harbor porpoises and gill nets in the Gulf of<br />
Maine. Conserv. Biol. 7, 189<strong>–</strong>193<br />
48 Carlstrom, J. et al. (2002) A field experiment using acoustic alarms<br />
(pingers) to reduce harbour porpoise by-catch in bottom-set gillnets.<br />
ICES J. Mar. Sci. 59, 816<strong>–</strong>824<br />
49 Larsen, F. et al. (2002) Reduction in harbour porpoise by-catch in the<br />
North Sea by high density gillnets. Paper SC/54/SM30, <strong>International</strong><br />
Whaling Commission Scientific Committee (www.cetaceanbycatch.org/<br />
Papers/larse02.pdf)<br />
A Special Issue of Current Biology<br />
21st August 2007<br />
50 Cox, T.M. et al. (2003) Behavioral responses of bottlenose dolphins,<br />
Tursiops truncatus, to gillnets and acoustic alarms. Biol. Conserv. 115,<br />
203<strong>–</strong>212<br />
51 Kastelein, R.A. et al. (2006) Differences in the response of a striped<br />
dolphin (Stenella coeruleoalba) and a harbour porpoise (Phocoena<br />
phocoena) to an acoustic alarm. Mar. Environ. Res. 61, 363<strong>–</strong>378<br />
52 Naidoo, R. et al. (2006) Integrating economic costs into conservation<br />
planning. Trends Ecol. Evol. 21, 681<strong>–</strong>687<br />
53 Sørensen, G.K. and Loeschcke, V. (2002) Natural adaptation to<br />
environmental stress via physiological clock-regulation of stress<br />
resistance in Drosophila. Ecol. Lett. 5, 16<strong>–</strong>19<br />
54 Cockrem, J.F. (2005) Conservation and behavioral neuroendocrinology.<br />
Horm. Behav. 48, 492<strong>–</strong>501<br />
55 Whitehead, H. and Rendell, L. (2004) Movements, habitat use and<br />
feeding success of cultural clans of South Pacific sperm whales.<br />
J. Anim. Ecol. 73, 190<strong>–</strong>196<br />
56 Freidenburg, L.K. and Skelly, D.K. (2004) Microgeographical variation<br />
in thermal preference by an amphibian. Ecol. Lett. 7, 369<strong>–</strong>373<br />
57 Kamel, S.J. and Mrosovsky, N. (2006) Deforestation: risk of sex ratio<br />
distortion in Hawksbill sea turtles. Ecol. Appl. 16, 923<strong>–</strong>931<br />
58 Spottiswoode, C.N. et al. (2006) Sexual selection predicts advancement<br />
of avian migration in response to climate change. Proc. R. Soc. Lond. B.<br />
Biol. Sci. 273, 3023<strong>–</strong>3029<br />
59 Yeh, P.J. (2004) Rapid evolution of a sexually selected trait following<br />
population establishment in a novel habitat. Evolution 58, 166<strong>–</strong><strong>17</strong>4<br />
60 Thom, M.D. et al. (2004) The evolution and maintenance of delayed<br />
implantation in the Mustelidae (Mammalia: Carnivora). Evolution 58,<br />
<strong>17</strong>5<strong>–</strong>183<br />
This special issue of Current Biology takes a broad look at the importance of sociality in biology, from the evolution of<br />
cooperation to the way that social living might have facilitated the evolution of human intelligence. The issue<br />
illustrates the fascinating variety of biological societies with articles on social amoebae, social spiders, eusocial<br />
insets, crows, hyenas and humans.<br />
Guest Editorial<br />
All life is social<br />
Steve Frank<br />
Reviews<br />
Evolutionary explanations for cooperation<br />
Stuart West, Ashleigh Griffin and Andy Gardner<br />
Kin selection versus sexual selection in eusocial<br />
insects<br />
Jacobus Boomsma<br />
Social learning<br />
Lars Chittka<br />
The Cold War of the social amoebae<br />
Gad Shaulsky and Richard Kessin<br />
Social cognition in humans<br />
Chris Frith<br />
Sociality, evolution and cognition<br />
Richard Byrne and Lucy Bates<br />
www.sciencedirect.com<br />
Opinion TRENDS in Ecology and Evolution Vol.22 No.8 407<br />
Social immune systems<br />
Sylvia Cremer<br />
Primers<br />
The social life of corvids<br />
Nicky Clayton and Nathan Emery<br />
Hyena societies<br />
Heather Watts and Kay Holekamp<br />
Social spiders<br />
Duncan Jackson<br />
Quick Guide
Anim Cogn (2009) 12:43<strong>–</strong>53<br />
DOI 10.1007/s10071-008-0169-9<br />
ORIGINAL PAPER<br />
Evidence of teaching in atlantic spotted dolphins<br />
(Stenella frontalis) by mother dolphins foraging in the presence<br />
of their calves<br />
Courtney E. Bender Æ Denise L. Herzing Æ<br />
David F. Bjorklund<br />
Received: 10 January 2008 / Revised: 18 March 2008 / Accepted: 5 June 2008 / Published online: 29 July 2008<br />
Ó Springer-Verlag 2008<br />
Abstract Teaching is a powerful form of social learning,<br />
but there is little systematic evidence that it occurs in<br />
species other than humans. Using long-term video archives<br />
the foraging behaviors by mother Atlantic spotted dolphins<br />
(Stenella frontalis) were observed when their calves were<br />
present and when their calves were not present, including<br />
in the presence of non-calf conspecifics. The nine mothers<br />
we observed chased prey significantly longer and made<br />
significantly more referential body-orienting movements in<br />
the direction of the prey during foraging events when their<br />
calves were present than when their calves were not present,<br />
regardless of whether they were foraging alone or with<br />
another non-calf dolphin. Although further research into<br />
the potential consequences for the naïve calves is still<br />
warranted, these data based on the maternal foraging<br />
behavior are suggestive of teaching as a social-learning<br />
mechanism in nonhuman animals.<br />
Keywords Teaching Dolphins Social learning<br />
Foraging<br />
Electronic supplementary material The online version of this<br />
article (doi:10.1007/s10071-008-0169-9) contains supplementary<br />
material, which is available to authorized users.<br />
C. E. Bender (&) D. F. Bjorklund<br />
Department of Biological Sciences, Florida Atlantic University,<br />
777 Glades Road, Boca Raton, FL 33431-0991, USA<br />
e-mail: courtbender@yahoo.com; cgreen21@fau.edu<br />
D. L. Herzing<br />
Department of Biological Sciences, Wild Dolphin Project,<br />
Florida Atlantic University, 777 Glades Road, Boca Raton,<br />
FL 33431-0991, USA<br />
Introduction<br />
Although one of the most potent forms of social learning in<br />
humans, there has been little evidence to suggest that<br />
teaching occurs in nonhuman animals. Some theorists have<br />
suggested that this may be because teaching requires<br />
advanced social-cognitive skills, including the ability to<br />
take the perspective of another and theory of mind, the<br />
ability to appreciate that an individual’s behavior is based<br />
on its knowledge and its desires (Boesch and Tomasello<br />
1998; Tomasello 1996, 2000; Tomasello et al. 1993). For<br />
example, previous examples of suggested teaching behavior<br />
by meerkats, cheetahs, and domestic cats seem to<br />
benefit the prey-handling abilities of the young, but do not<br />
require the use of higher cognitive mechanisms (Thornton<br />
and McAuliffe 2006; Caro and Hauser 1992). However,<br />
Caro and Hauser (1992) provided a definition of teaching<br />
that may be more inclusive for nonhuman animals, defining<br />
it as, ‘‘An individual actor A can be said to teach if it<br />
modifies its behavior only in the presence of a naive<br />
observer, B, at some cost or at least without obtaining an<br />
immediate benefit for itself. A’s behavior thereby encourages<br />
or punishes B’s behavior, or provides B with<br />
experience, or sets an example for B. As a result, B<br />
acquires knowledge or learns a skill earlier in life or more<br />
rapidly or efficiently than it might otherwise do, or that it<br />
would not learn at all’’ (p. 153).<br />
Previous studies that suggested teaching in primates and<br />
cetaceans, although promising, lacked systematic measurement<br />
of the behavior. Probably the best evidence of<br />
teaching in nonhuman primates to date is Boesch’s studies<br />
of mother chimpanzees (Pan troglodytes) in the Tai<br />
National Park of the Ivory Coast (Boesch 1991, 1993;<br />
Greenfield et al. 2000). Boesch suggested that the<br />
chimpanzee mothers facilitated the development of their<br />
123
44 Anim Cogn (2009) 12:43<strong>–</strong>53<br />
offspring’s nut-cracking skills by means of stimulation,<br />
facilitation, and active teaching. Nut cracking is observed<br />
in only a few populations of chimpanzees, despite the<br />
availability of nuts and appropriate tools (i.e., rocks<br />
appropriate for use as anvils and hammers), qualifying, by<br />
some definitions, as an example of culturally-transmitted<br />
behavior (Whiten 2005; Whiten et al. 1999). The interpretation<br />
of such episodes as ‘‘teaching’’ has been<br />
questioned, however (Bering 2001; Bering and Povinelli<br />
2003). Moreover, such episodes are rarely observed, suggesting<br />
that direct teaching is not a common form of<br />
cultural transmission in chimpanzees.<br />
Although evidence of social learning is easier to document<br />
in these terrestrial great apes than it is in marine<br />
mammals, nongenetic transmission of behavior across<br />
generations has also been observed for cetaceans (Kruetzen<br />
et al. 2005; Rendell and Whitehead 2001), suggesting that,<br />
like the great apes, these large-brained, slow-developing,<br />
and socially complex species (Bjorklund and Bering 2003)<br />
have evolved powerful social-learning mechanisms.<br />
Similar to research with great apes, little is known about<br />
the actual mechanism of transmission across generations in<br />
cetaceans. Herzing (2005) described some potential scenarios<br />
and mechanisms observed for a group of free-ranging<br />
Atlantic spotted dolphins (Stenella frontalis), including<br />
implications of vertical, horizontal, and oblique directions<br />
of transmission of information during various behavioral<br />
contexts. Recently, Spininelli et al. (2006) described preytransfer<br />
between mother and calf in the marine tucuxi dolphin<br />
(Sotalia fluvialis). Observations such as these suggest<br />
that the mother-calf relationship may be one of the most<br />
important sources of information in the young calf’s life.<br />
There is some evidence of presumed maternal teaching<br />
behavior associated with stranding behavior as a foraging<br />
specialization used by part of the population of killer whales<br />
(Orcinus orca) in the Crozet Islands and off Punta Norte,<br />
Argentina to capture seal pups on pinniped breeding beaches.<br />
Adult females demonstrated a modification of their strand<br />
foraging behavior in the presence of naïve juvenile observers<br />
(presumably their calves), suggesting that teaching may be<br />
involved in the development and the rate of success of calves<br />
in mastering these behaviors (Guinet and Bouvier 1995).<br />
This comparison between purported teaching in chimpanzees<br />
and killer whales is interesting because any<br />
commonalities would have been derived through convergent<br />
evolution, as the last common ancestor of primates with<br />
cetaceans is estimated to have lived over 90 million years ago<br />
(Marino et al. 2007). However, despite the multitude of<br />
observations of chimpanzees and killer whales in the wild,<br />
incidences of mother<strong>–</strong>infant teaching are scarce and<br />
anecdotal in nature.<br />
Foraging behavior is a likely candidate for social<br />
learning among wild dolphins, particularly between mother<br />
123<br />
and calf. The mother/calf relationship is the strongest<br />
association that Atlantic spotted dolphins have in their<br />
lifetimes (Herzing and Brunnick 1997). The prolonged<br />
developmental period provides both ample time and situational<br />
possibilities for a calf to learn foraging strategies<br />
from its mother (Herzing 1996). The majority of daytime<br />
feeding behavior of Atlantic spotted dolphins is benthic<br />
foraging, in which dolphins use echolocation to locate fish<br />
in the sandy bottom and then dig prey items out of the sand<br />
in order to catch and eat them (Herzing 1996).<br />
Much of what we know about marine mammal social<br />
learning comes from research with captive animals due, in<br />
part, to the difficulty of studying such phenomena in the<br />
wild. Attempts to assess teaching among captive animals<br />
often involve contrived situations, which may affect the<br />
animals’ success or failure (Kuczaj et al. 2005). Studies<br />
that assess social learning in wild populations of marine<br />
mammals are needed to validate and supplement the findings<br />
from captive animals and to better understand the<br />
spontaneous occurrence of social learning in natural<br />
settings.<br />
In the present study, using video archives from a longterm<br />
naturalistic study, we investigated social learning and<br />
possible teaching behavior by the Atlantic spotted dolphin.<br />
Unlike most previous research examining social learning in<br />
nonhuman mammals in which the focus is on the observer/<br />
learner (e.g., Bjorklund et al. 2002; Guinet and Bouvier<br />
1995; Herman 2002; Tomasello et al. 1993), the present<br />
study shifted the focus from the observer (in our case, the<br />
calf observing the foraging behavior) to the model (the<br />
mother performing the foraging behavior) to explore the<br />
possibility of teaching behavior. We examined the foraging<br />
behavior by mother dolphins when foraging in the presence<br />
of their young (less than 3-year-old) calves in comparison<br />
to when the calves were not present. Additionally, comparisons<br />
were then made between mothers foraging alone<br />
and when they were foraging with non-calf conspecifics.<br />
Should the mothers alter their foraging behavior in the<br />
presence of their young, it would be suggestive of teaching<br />
and provide a possible mechanism for cultural transmission<br />
within this dolphin species.<br />
Methods and analysis<br />
Natural history<br />
This study was performed using underwater video recordings<br />
of the Atlantic spotted dolphin collected by the Wild<br />
Dolphin Project (Herzing 1997) in the study area north of<br />
Grand Bahama Island, Bahamas during summer field seasons<br />
between 1991 and 2004 (Fig. 1). Unlike many other<br />
marine mammal habitats, this area is optimal for behavioral
Anim Cogn (2009) 12:43<strong>–</strong>53 45<br />
Fig. 1 Map of study area for the wild dolphin project underwater<br />
video recordings collected over sandbanks north of Grand Bahama<br />
Island Study population of Atlantic spotted dolphins ranges over an<br />
area of approximately 500 km 2<br />
observation with clear, warm waters that allow for excellent<br />
visibility up to 90 ft and long observational periods<br />
(Herzing 1996). The dolphins in this study have been<br />
observed since 1985, and include over 200 dolphins that<br />
have been individually recognized and sexed. Atlantic<br />
spotted dolphins can be categorized into four age classes,<br />
based on the pigmentation of the individual (Herzing<br />
1996). The number of spots on the individual is correlated<br />
with age, with a newborn having no spots. Although most<br />
individuals in the population are tracked from birth, this<br />
allows for approximating the individual’s age when it is not<br />
known from previous sightings. Calves in the sampled<br />
foraging events 3-years-old or younger were still observed<br />
to be nursing during the encounter year. The year of the<br />
calf’s birth was determined from previous sightings of the<br />
pregnant female followed by a sighting with a closely<br />
associated calf or a sighting of the mother with a suckling<br />
calf (Herzing 1997).<br />
Apparatus and procedure<br />
Video was recorded using various underwater cameras<br />
(Sony CCDV9 8-mm, Yashica KXV Hi8-mm with attached<br />
Labcore 76 hydrophone, Sony DCR-SC100 NTSC, or Sony<br />
DCR-PC110 NTSC Digital Video). Underwater video<br />
sequences were analyzed using focal follow (of the mother<br />
with and without the calf) as a sampling rule and continuous<br />
sampling as a recording rule. Video sequences from<br />
the long-term video archives from the Wild Dolphin Project<br />
Ò between 1991 and 2004 were assessed for the<br />
presence of benthic foraging events by individual mothers,<br />
either with or without their calves present, based on a<br />
behavioral ethogram designed to measure the individual<br />
benthic foraging behavior of the mother dolphins (see<br />
Fig. 2). Each benthic fish catch was broken down into a<br />
series of behaviors that made up a typical foraging event.<br />
For the purpose of this study, a foraging event was<br />
defined as the series of behaviors performed by the dolphin<br />
in order to catch the prey animal. For benthic feeding, the<br />
series of events was as follows:<br />
scanning ! rooting ! chase ! ingestion<br />
Scanning is observed when the dolphin moves its head<br />
horizontally or vertically repeatedly while performing a<br />
directional swim, and usually occurs near the sea floor and<br />
can be followed by a dig or fish catch (Miles and Herzing<br />
2003). Rooting or digging is observed when the dolphin<br />
inserts the rostrum into the sea floor or sandy bottom to dig<br />
the prey out of the substrate in the attempt to capture the<br />
prey. Chasing is observed as swimming in the direction of<br />
the prey object, as in pursuit of the prey. For our purposes,<br />
ingestion was defined as the food going into mouth and<br />
never being seen again. The categories of foraging<br />
behaviors measured were chase latency and number of<br />
body-orienting movements during pursuit. Chase latency,<br />
the length of time the prey was pursued, was operationally<br />
defined as the period of time when the fish appeared out of<br />
the sand (was rooted out) until the time of ingestion by the<br />
dolphin, or the dolphin no longer pursued prey (lost interest<br />
in prey). Body-orienting movements were measured to<br />
examine the dolphin’s attention to the prey object. A bodyorienting<br />
movement was measured in foraging events as a<br />
movement of the body reorienting in direction of prey<br />
object, often seen during pursuit of prey, from the time the<br />
prey was rooted out of the sand. Body-orienting<br />
movements were particularly interesting as they appeared<br />
to be exaggerated movements in the direction of the prey,<br />
which may be an attention-directed referential behavior<br />
similar to the spontaneous pointing observed by dolphins<br />
during experiments in captivity (Xitco et al. 2001).<br />
Figure 2 is a visual ethogram describing the sequence of<br />
the benthic foraging behavior (Miles and Herzing 2003).<br />
123
46 Anim Cogn (2009) 12:43<strong>–</strong>53<br />
Fig. 2 Visual ethogram of<br />
select foraging behavioral<br />
events (Miles and Herzing<br />
2003). The visual ethogram<br />
includes the possible positions<br />
in which calves were observed<br />
during foraging events in<br />
relation to their mother and the<br />
sequence of behaviors for a<br />
benthic foraging event<br />
The segments from the video archives in which mothers<br />
were observed engaged in foraging behavior with or<br />
without their calves were then viewed and then examined<br />
for further criteria. Thirty-eight video segments were used<br />
in the study based on the selection criteria: fourteen video<br />
segments of mothers foraging with their calves present and<br />
24 video segments of those mothers foraging without their<br />
calves present. Videos were selected based on the presence<br />
of a target female performing an individual foraging event,<br />
as well as on the following designated video acceptance<br />
criteria: (a) the individuals were identifiable; (b) it was<br />
possible to identify the beginning and end of the chase<br />
sequence; (c) the prey was visible, or if the prey was not<br />
completely visible, it was possible to identify the position<br />
of the prey based on the behavior of the dolphin; (d) if the<br />
calf was present, the calf was in a nearby position (within a<br />
proximity of two body lengths) from which it was capable<br />
of observing its mother; and (e) if the calf was present, it<br />
was possible to identify the position of the calf relative to<br />
the mother during the foraging event. The position of the<br />
calf relative to the mother was recorded as Infant, Headunder-head,<br />
Echelon, Observation, or Other, with comments<br />
where necessary, as depicted in Fig. 2.<br />
The mothers’ foraging behaviors were then individually<br />
measured for the variables of chase latency and body-orienting<br />
movements, both when foraging alone (or with other<br />
juvenile or adult dolphins) and when foraging in the<br />
presence of their calves. Other variables, such as types of<br />
play involved in that foraging encounter, position of calf<br />
relative to the mother, directionality of the calf, age of the<br />
mothers and their calves, whether the mother eventually ate<br />
the prey; the prey species, when identifiable, were also<br />
recorded for each foraging event.<br />
Participants<br />
Positions of Calf Relative to Mother<br />
The foraging behaviors of nine mother dolphins were<br />
recorded both with (n = 14) and without (n = 22) their<br />
123<br />
Echelon Position Infant Position Head-under-head Observation Position<br />
Sequence of Behaviors for<br />
Benthic Foraging Event<br />
Position<br />
Individual Scanning Dig/Root Chase<br />
calves present. Ten different calves were observed with the<br />
nine mothers in the 14 foraging events, with one mother<br />
observed during separate events with two different calves.<br />
Calves ranged in age from neonate to 3 years old. All<br />
calves were observed to be nursing within the same field<br />
season as the foraging event. Ages of the mothers were<br />
known, or were estimated based on their age class. It is<br />
important to note that some foraging events without calves<br />
present (10 of the 22) occurred while the target female was<br />
still a juvenile (prior to sexual maturity); however, during<br />
those events, the ‘‘mothers to be’’ were already past the age<br />
of weaning and were independently foraging. The minimum<br />
age of any female during foraging events over the<br />
12-year period was 10 years for mothers foraging with<br />
their calves and 4 years for mothers-to-be foraging without<br />
calves present. Of the observations without calves present,<br />
four of the females were observed foraging as juveniles,<br />
prior to becoming mothers. Of the nine mothers observed,<br />
foraging events were observed for one female both when<br />
she was a calf with her mother, and later as a mother<br />
herself with her own calf.<br />
The video segments of the foraging events were shortened<br />
to within 1 min of the beginning and end of the<br />
foraging event and labeled with the foraging event number<br />
and the individuals involved. Segments were then watched<br />
by the first author and two independent observers to measure<br />
the desired behaviors. Of the 38 total foraging events<br />
measured, 32 were measured by the first author and two<br />
independent observers; the other six were measured by the<br />
first author and only one independent observer. For the<br />
measurement of chase latencies of foraging events, there<br />
was significant correlation between the author and the first<br />
independent observer, r 36 = 1.0, P \ 0.001, between the<br />
author and the second independent observer, r30 = 0.999,<br />
P \ 0.001, and between the first and second observer,<br />
r30 = 0.998, P \ 0.001. For the measurement of number of<br />
body-orienting movements, there was significant correlation<br />
between the author and the first independent observer,
Anim Cogn (2009) 12:43<strong>–</strong>53 47<br />
Table 1 Mean chase latency of mothers with and without calf<br />
Mother Little Gash Mugsy Nassau Nippy PR1 PR2 Rosemole Trimy Uno Mean SD<br />
Mean chase latency<br />
without calf (n)<br />
Mean chase latency<br />
with calf (n)<br />
Mean number of BOM<br />
without calf (n)<br />
Mean number of BOM<br />
with calf (n)<br />
r36 = 1.0, P \ 0.001, between the author and the second<br />
independent observer, r30 = 0.988, P \ 0.001, and<br />
between the first and second observer, r30 = 0.988,<br />
P \ 0.001.<br />
The videos were watched and timed using Windows<br />
Media Player version 10 on a Hewlett Packard laptop<br />
computer and a projector in order to enlarge the viewing<br />
area. Due to the restrictions of the media software used for<br />
editing and playing the video, a ‘‘second or less’’ rule was<br />
instituted for measurement of latencies in which the<br />
latencies appearing to be less than 1 s were rounded up to<br />
1 s in duration.<br />
Results<br />
Chase latencies<br />
Mean chase latencies for each of the nine mothers, both<br />
when foraging with and without their calves, is presented in<br />
Table 1. The mothers chased the prey significantly longer<br />
when their calves were present (M = 22.24 s, SD = 9.36)<br />
than when their calves were not present (M = 2.74 s,<br />
SD = 1.47), t8 = 6.57, P \ 0.001, d = 1.14. Mean chase<br />
latencies were longer when foraging with their calves than<br />
without their calves for each of the nine mothers (Fig. 3).<br />
Body-orienting movements<br />
5.6 (5) 2.25 (4) 2.00 (3) 2.00 (1) 2.<strong>17</strong> (2) 4.00 (1) 1.67 (3) 4.00 (4) 1.00 (1) 2.74 1.47<br />
38.33 (1) 23.00 (2) 16.33 (2) 36.00 (1) 24.33 (3) 19.67 (1) 16.50 (2) 12.00 (1) 14.00 (1) 22.24 9.36<br />
0.80 (5) 0.00 (4) 0.33 (3) 0.67 (1) 0.50 (2) 0.00 (1) 0.00 (3) 0.25 (4) 0.00 (1) 0.28 0.31<br />
0.00 (1) 2.50 (2) 1.50 (2) 0.00 (1) 2.33 (3) 2.00 (1) 0.00 (2) 2.00 (1) 1.00 (1) 1.26 1.04<br />
Bold indicates mothers who performed significantly more body-orienting movements in the presence of their calves. Chase latencies were<br />
significantly longer for all nine mothers<br />
Mean number of body-orienting movements for each of the<br />
nine mothers, both when foraging with and without their<br />
calves, is presented in Table 1. The mothers made significantly<br />
more body-orienting movements when their calves<br />
were present (M = 1.26, SD = 1.04) than when their<br />
calves were not present (M = 0.28, SD = 0.31), t8 = 2.46,<br />
P = 0.04, d = 2.31. Six of nine mothers made more bodyorienting<br />
movements when foraging in the presence of<br />
their calves than when not foraging with their calves.<br />
45.00<br />
40.00<br />
35.00<br />
30.00<br />
25.00<br />
20.00<br />
15.00<br />
10.00<br />
5.00<br />
0.00<br />
38.33<br />
5.60<br />
23.00<br />
Chase Latency of Mothers<br />
16.33<br />
36.00<br />
Cross-generational comparisons<br />
Two consecutive generations of the mother/calf pairs in the<br />
study showed the same altered foraging behavior. In the<br />
first generation, one mother, Nippy, and her calf, Nassau,<br />
demonstrated the presumed teaching behavior. In the second<br />
generation, Nassau, now a mother, showed the same<br />
altered foraging behavior as her mother with her calf,<br />
Neptune.<br />
Foraging with non-calf individuals<br />
Of the foraging events without calves, three mothers were<br />
observed foraging both alone and with other individuals<br />
that were not calves, at least juveniles or older. The chase<br />
latencies and body-orienting movements for these mothers<br />
were compared for the two conditions (alone and with noncalf<br />
individuals). There were too few subjects (n = 3) to<br />
perform a statistical test; however, the chase latencies of<br />
these three mothers foraging alone (M = 1.33, SD = 0.58)<br />
were comparable to the chase latencies of these same<br />
mothers foraging with non-calf individuals (M = 3.44,<br />
SD = 1.50), and both were much lower than the chase<br />
24.33<br />
2.25 2.00 2.00 2.<strong>17</strong><br />
Mother<br />
19.67<br />
4.00<br />
Mothers Chasing<br />
With Calves<br />
Mothers Chasing<br />
Without Calves<br />
Fig. 3 Mean chase latencies of mothers foraging with and without<br />
their calves present<br />
16.50<br />
1.67<br />
12.00<br />
4.00<br />
14.00<br />
1.00<br />
123
48 Anim Cogn (2009) 12:43<strong>–</strong>53<br />
latencies of those mothers foraging with their calves<br />
present (M = <strong>17</strong>.56, SD = 6.26). Additionally, the number<br />
of body-orienting movements of these three mothers foraging<br />
alone (M = 0.<strong>17</strong>, SD = 0.29) was comparable to the<br />
number of body-orienting movements when they were<br />
foraging with non-calf individuals (M = 0.33, SD = 0.58),<br />
and both were less than the number of body-orienting<br />
movements of those mothers foraging with their calves<br />
present (M = 1.94, SD = 0.42).<br />
Additional analyses<br />
The differences observed for chase latencies and number of<br />
body-orienting movements could not be attributed to prey<br />
type, as there were no significant differences in the foraging<br />
behaviors between prey species. Prey species were identified<br />
for 14 observations. When foraging with calves present,<br />
mothers were observed foraging for Snakefish (family<br />
Synodontidae), n = 5, flounder (family Bothidae), n = 2,<br />
and razorfish (family Clinidae), n = 1. The mean chase<br />
latencies, M Snakefish = 19.0 (SD = 11.03), M Flounder = 20.5<br />
(SD = 14.85), MRazorfish = 25.0 (SD = 0.0), and number<br />
of body-orienting movements, MSnakefish = 0.6 (SD =<br />
0.89), M Flounder = 25.0 (SD = 2.12), M Razorfish = 3.0<br />
(SD = 0.0), were comparable for all three species of prey<br />
when foraging with calves present. Both snakefish, n = 9,<br />
and flounder, n = 4, were observed as prey types when the<br />
females were observed foraging without calves present. The<br />
mean chase latencies, MSnakefish = 2.22 (SD = 1.09),<br />
M Flounder = 3.83 (SD = 3.57), and number of body-orienting<br />
movements, MSnakefish = 0.5 (SD = 0.76), MFlounder =<br />
0.5 (SD = 0.58), were comparable for both species when<br />
foraging without calves present, and both were lower than<br />
when foraging with calves present.<br />
Additionally, individual foraging events of dolphins not<br />
included in the study observed foraging for either snakefish<br />
or flounder were collected and analyzed. Ten foraging<br />
events each for catches of snakefish and flounder using<br />
both dolphins in the present study and dolphins not used in<br />
the study were randomly selected, and their corresponding<br />
chase latencies and number of body-orienting movements<br />
were compared between the two types of prey using an<br />
independent samples Student’s t-test. There were no significant<br />
differences found between the two types of prey for<br />
either chase latencies, (M flounder = 4.60, SD flounder = 3.86,<br />
Msnakefish = 2.80, SDsnakefish = 1.62) t18 =-1.36,<br />
P = 0.19, or number of body orienting movements for<br />
dolphins foraging without calves present, (M flounder = 0.50,<br />
SDflounder = 0.71, Msnakefish = 0.37, SDsnakefish = 0.48)<br />
t18 =-0.49, P = 0.63. This supports the conclusion that<br />
the observed differences in this study for mothers foraging<br />
with calves present would not likely be due to the type of<br />
prey, as chase latencies and the number of body-orienting<br />
123<br />
movements do not normally vary significantly between<br />
snakefish and flounder, the two main types of prey observed<br />
being caught by mother dolphins in this study.<br />
To assess possible effects of age of calf and age of<br />
mother on the dependent measures, correlations with each<br />
dependent measure were computed separately with the age<br />
of the calf and the age of the mother for the foraging events<br />
when the calf was present. Correlations between age of the<br />
calf and chase latency, r13 = 0.37, P = 0.10, and number<br />
of body-orienting movements, r13 =-0.30, P = 0.15,<br />
were both nonsignificant. Correlations between mother’s<br />
age and number of body-orienting movements were significant,<br />
r13 =-0.47, P = 0.046, with older mothers<br />
making fewer body-orienting movements than younger<br />
mothers. The correlation between mother’s age and chase<br />
latencies was not significant, r13 = 0.01, P = 0.48. When<br />
comparing the foraging behaviors to the age of the mothers<br />
when the calves were not present, correlations with both<br />
chase latencies r23 =-0.07, P = 0.73, and number of<br />
body-orienting movements, r 23 =-0.12, P = 0.58, were<br />
nonsignificant. The foraging behaviors of eight of the nine<br />
mothers without calves were compared between those<br />
females observed as juveniles and those observed as adults,<br />
with three mothers being observed foraging as juveniles<br />
and five mothers observed foraging as adults. 1 There were<br />
no significant differences between chase latencies, t (6) =-<br />
0.10, P = 0.92, or number of body-orienting movements,<br />
t(6) =-0.12, P = 0.91, of females observed foraging<br />
when juveniles or adults.<br />
Additionally, of the nine mothers observed in this study,<br />
three of the mothers, Little Gash, Mugsy, and PR2, were<br />
observed foraging both with and without their calves<br />
present during the same year, and each of those three<br />
mothers being observed foraging both with and without<br />
calves during the same field encounter. The mean chase<br />
latencies for these three mothers are compared in Table 2.<br />
The chase latencies were much longer foraging with calves<br />
than foraging alone for the mothers at the same age and<br />
comparable to the mean chase latencies of foraging alone<br />
for the mothers at an earlier age.<br />
The position of the calf during the chase period was<br />
indicated for each of the foraging events (N = 14) in which<br />
the calf was present. Although some foraging events<br />
involved multiple calf positions, the observation position<br />
1 One mother, Little Gash, was observed foraging without her calf<br />
present both as an adult and as a juvenile. When group means were<br />
compared including Little Gash there were no significant differences<br />
between adults and juveniles for either chase latencies, t(8) = 0.05,<br />
P = 0.96, or body-orienting movements, t (8) = 0.41, P = 0.69. Both<br />
the mean chase latencies, M A = 3.67 and M J = 6.89, and mean<br />
number of body-orienting movements, M A = 0.5 and M J = 1.0, for<br />
Little Gash were comparable as an adult and as a juvenile.
Anim Cogn (2009) 12:43<strong>–</strong>53 49<br />
Table 2 Maternal age<br />
comparison for mean chase<br />
latency and number of bodyorienting<br />
movements<br />
was found to be the most common, taking place in 11<br />
(79%) of the foraging events. Head-under-head position<br />
was present in one of the foraging events, Echelon was<br />
present in two foraging events, and Infant position was<br />
seen in one foraging event. Only Infant, Echelon, and<br />
Observation positions were seen solely in an individual<br />
foraging event, with one, one, and nine occurrences,<br />
respectively.<br />
For the foraging events collected in which the ingestion<br />
by the mother was known (as seen by video or first-hand<br />
account), there were five events, one event each for five of<br />
the nine mothers, in which the prey was not eaten. Each of<br />
these foraging events occurred when the mothers’ calves<br />
were present (36% of the total events with calves). In<br />
addition, the calves were allowed to pursue the prey in each<br />
of these events and were confirmed to have eaten the prey<br />
in three of the foraging events, despite the fact that they<br />
were still nursing and not dependent upon fish for food.<br />
Four of the nine mothers were observed eating the prey in<br />
seven of the fourteen events in which calves were present.<br />
In two of the events in which calves were present it was not<br />
known whether the prey was eaten.<br />
Discussion<br />
Encounter dates<br />
by mother<br />
The nine mother dolphins observed in this study displayed<br />
significantly longer chase latencies and made significantly<br />
more body-orienting movements when foraging in the<br />
presence of their calves than when foraging alone. The<br />
chase latency data are particularly impressive. Mean chase<br />
latencies were eight times longer for the female dolphins<br />
when foraging with their calves (22.24 s) than without<br />
them (2.74 s). As illustrated in Fig. 3, the distributions<br />
were non-overlapping, with every mother having a longer<br />
latency when foraging with her calf than when foraging<br />
without her calf.<br />
Although differences were not as robust for the bodyorienting<br />
movements, the overall number of body-orienting<br />
Mean chase latency<br />
with calf (n)<br />
Mean chase latency<br />
without calf (n)<br />
Mean BOM<br />
with calf<br />
Mean BOM<br />
without calf<br />
Little Gash<br />
30 May 2000 38.33 (1) 3.67 (2) 0.00 (1) 0.50 (2)<br />
27 May 1991<br />
Mugsy<br />
<strong>–</strong> 4.00 (2) <strong>–</strong> 1.50 (2)<br />
29 August 1993 34.00 (1) 2.00 (1) 5.00 (1) 0.00<br />
13 May 2000 <strong>–</strong> 4.00 (1) <strong>–</strong> 0.00<br />
11 August 2000<br />
PR2<br />
12.00 (1) <strong>–</strong> 0.00 <strong>–</strong><br />
14 June 1994 19.67 (1) 4.00 (1) 2.00 0.00<br />
movements was significantly greater when the mothers<br />
were foraging with their calves than when foraging alone,<br />
with six of nine mothers displaying this pattern. Previous<br />
research has shown that dolphins are capable of understanding<br />
the human gesture of pointing (Herman et al.<br />
1999; Pack and Herman 2004), and of producing spontaneous<br />
referential gestures in artificial experimental<br />
contexts (Xitco et al. 2001). Although these body-orienting<br />
movements were not demonstrated by all sampled mothers,<br />
the present results may be evidence of referential gesturing<br />
in a more ecologically valid context.<br />
In addition to the elongated chase and presumed referential<br />
body-orienting movements in the direction of the<br />
prey, the altered foraging behavior in the presence of the<br />
calf appeared more exaggerated when compared to the<br />
mothers foraging alone or with another non-calf dolphin.<br />
Mothers seemed to toy with their prey, making it more like<br />
play behavior and less like the typical foraging behavior of<br />
mothers observed foraging without their calves present,<br />
similar to descriptions of possible teaching of predatory<br />
behavior from other species such as cats and killer whales<br />
(Caro and Hauser 1992; Guinet and Bouvier 1995). Some<br />
of the mothers were observed letting the prey swim away<br />
and then digging them back out of the sand, sometimes<br />
multiple times, after initially digging the fish out of the<br />
sand, and also make jawing motions in the direction of the<br />
prey. Mothers also allowed the seemingly attentive calves<br />
to participate in the chase, and calves were observed eating<br />
the prey in three of the events. Although 4 of the 9 mothers<br />
were observed eating the prey in 7 of the 14 events, all of<br />
the mothers were observed allowing the calves to participate<br />
and made little to no effort in these events to consume<br />
the prey, only seemingly facilitating the calves’ experience<br />
in chasing the prey. Despite the altered foraging behavior,<br />
mothers were never observed losing the prey. In all events<br />
in which ingestion of the prey was observed, either the<br />
mother or the calf ate the prey, or the mother made no<br />
further attempts to ingest the prey and left the prey to the<br />
calf to chase. This altered foraging behavior can be<br />
123
50 Anim Cogn (2009) 12:43<strong>–</strong>53<br />
observed in the online supplemental movies S1 and S2<br />
which exemplify the markedly different foraging behavior<br />
of an adult female foraging alone (S1) with the foraging<br />
behavior of a mother foraging in the presence of her calf<br />
(S2).<br />
There was a potential cost of the alteration of the<br />
mothers’ foraging behavior in their calves’ presence. The<br />
elongated chase time allowed more opportunity for prey to<br />
escape, although this did not occur in any of the events<br />
observed, as well as taking time away from catching<br />
additional fish or other activities. Instead of the typical grab<br />
and ingestion, these mothers toyed with their prey as their<br />
calves watched, and, as mentioned above, sometimes even<br />
avoided ingestion in these altered foraging events. Often in<br />
teaching, the model will provide an exaggerated or elongated<br />
version of the typical behavior in front of the naïve<br />
observer, much as the mother chimpanzees observed in<br />
Boesch’s (1991) work. These exaggerated foraging<br />
behaviors may provide a window of opportunity for the<br />
calves to observe, and possibly learn from, the example<br />
provided by their mothers, and thus be worthy of the extra<br />
time and energy put forth by the mothers at a cost to<br />
themselves in order to help ensure the success of their<br />
offspring.<br />
An alternative explanation for the altered foraging<br />
behavior may be that the mothers were distracted by their<br />
calves, resulting in longer chase latencies and exaggerated<br />
movements to compensate for their divided attention.<br />
However, we do not believe this to be the cause of the<br />
altered foraging behavior for a few reasons. First, three of<br />
the nine mothers observed in the present study demonstrated<br />
the ability of dolphin mothers to forage without<br />
distraction even when their calves were in the nearby<br />
vicinity, but not directly observing the mothers. These<br />
three mothers, Little Gash, Mugsy, and PR2, were<br />
observed foraging both with and without their calves<br />
present during the same field encounter. These mothers<br />
only altered their behavior when the calves were<br />
observing the mother, but not during the fish catches in<br />
which the calves were within the vicinity and not<br />
observing. Table 2 compares the means for those events,<br />
in which the mean chase latencies were much longer<br />
foraging with calves than foraging alone during the same<br />
field encounter and comparable to the mean chase latencies<br />
of foraging alone during separate field encounters.<br />
These data suggest that the calves’ being merely nearby is<br />
not a distraction for the mothers, but instead the altered<br />
foraging behavior only occurred when the calves were<br />
directly observing the mothers’ behavior. If the calves<br />
were a distraction for the mothers, the altered foraging<br />
behavior should have occurred whenever the calves were<br />
within the vicinity, regardless of whether the calves were<br />
directly observing the mothers.<br />
123<br />
Second, Atlantic spotted dolphins participate in alloparenting,<br />
which permits mothers to frequently forage<br />
separately from their calves (Herzing 1996). If the mother’s<br />
nutritional needs are being met in the absence of their calf<br />
with the help of alloparenting, this would enable more time<br />
and energy on the part of the mother for altered foraging<br />
events for teaching when the calf is present. Additionally, it<br />
would not be advantageous for the mother to forage with<br />
the calf present solely for her nutritional purposes if there is<br />
an alloparent available. In this study, there were potential<br />
alloparents, juvenile or adult female dolphins, available in<br />
all 14 of the calf-present events that would have permitted<br />
the mothers to forage without distraction, which is often<br />
observed in Atlantic spotted dolphins (Herzing 1996),<br />
including the events for the previously mentioned three<br />
mothers observed foraging with and without calves during<br />
the same field encounter.<br />
Third, young dolphin calves are very precocious and the<br />
calves in the observed foraging events appeared attentive<br />
and interactive. Some of the evidence for the calves’<br />
attentiveness to the mothers’ foraging behavior was<br />
observed through the calves’ positions relative to the<br />
mothers, which was indicated during the chase period for<br />
each of the foraging events in which the calf was present.<br />
Although previous research (Mann and Smuts 1999) has<br />
shown that dolphin calves in the wild, and in captivity<br />
spend the majority of their time in infant or echelon position<br />
until the time of weaning, calves were most commonly<br />
found during the chase period in the observation position<br />
relative to their mothers in which the calf was potentially<br />
exposed to both visual and acoustic information and in<br />
which calves appeared attentive to both the mother and the<br />
prey 2 (Figure 2).<br />
Additionally, the calves were allowed to pursue the prey<br />
in five of the foraging events and were confirmed to have<br />
eaten the prey in three of the events, despite the fact that<br />
they were still nursing and not dependent upon fish for<br />
food. It is important to note that the only time when the fish<br />
was not ingested was when the mothers were foraging with<br />
their calves as part of their altered foraging behavior. The<br />
prey was always consumed when foraging alone or in the<br />
presence of a non-calf dolphin. This is potentially a costly<br />
behavior for the mother as she is depriving herself of calories<br />
needed to nourish her still nursing calf by allowing<br />
2 It is important to note that research in captivity (Pack and Herman<br />
1995) has demonstrated the ability of dolphins to perceive and<br />
recognize objects through either vision or echolocation. In addition,<br />
their perceptions are readily shared or integrated across the senses,<br />
regardless of which modality the dolphin originally perceived the<br />
external stimuli. However, the sounds emitted from the dolphins were<br />
not measured in the present study, but should be looked at in future<br />
research in order to determine what sensory information the calf is<br />
receiving.
Anim Cogn (2009) 12:43<strong>–</strong>53 51<br />
the prey to escape or allowing the calf to eat the prey, as<br />
well as the energy to play with the prey rather than just<br />
eating it quickly and efficiently herself. It would be more<br />
efficient and less costly for the mothers to have simply<br />
caught and consumed the fish, or perhaps to have left the<br />
calf in the care of an alloparent, as opposed to this altered<br />
foraging behavior observed in these events.<br />
Furthermore, the chase latencies and foraging behaviors<br />
were drastically different with their calves present, not just<br />
with the longer latencies but also with the presumed referential<br />
behaviors toward the prey objects and allowing the<br />
still nursing calf to participate, which would be more<br />
indicative of social learning. If the calves were merely a<br />
distraction, the mothers could have either immediately<br />
consumed the prey or disciplined the calf, as previously<br />
observed by mothers in the population for undesirable<br />
behaviors (Miles and Herzing 2003), rather than allowing<br />
the calf to participate. Therefore, because the calves<br />
appeared attentive, interacted with the prey and the mother,<br />
and appeared to need little care during the event, they were<br />
not likely a distraction to the mother.<br />
Reciprocally, the mothers may have been altering their<br />
foraging behavior because their calves were attentive, so<br />
that a calf’s attention may have stimulated the altered<br />
maternal foraging behavior. Further analysis into the<br />
calves’ behavior is warranted and may also elucidate the<br />
exact learning mechanism at play on behalf of the calf.<br />
For the majority of the mothers, there was a difference<br />
in maternal age between the events observed foraging<br />
without their calves compared to the events foraging with<br />
their calves. It is not likely that the age differences, or<br />
resulting level of experience would have resulted in the<br />
observed differences in chase latencies and number of<br />
body-orienting movements. Rather, an older or more<br />
experienced dolphin should be expected to have quicker<br />
chase latencies and fewer body-orienting movements due<br />
to increased efficiency, not the longer latencies and<br />
increased number of body-orienting movements as<br />
observed here. Additionally, of the nine mothers observed<br />
in this study, three of the mothers, Little Gash, Mugsy, and<br />
PR2, were observed foraging both with and without their<br />
calves present during the same year, and each of these<br />
three mothers were observed foraging both with and<br />
without calves during the same field encounter. For these<br />
three mothers, mean chase latencies were much longer<br />
foraging with calves than foraging alone at the same age<br />
and comparable to the mean chase latencies of foraging<br />
alone at an earlier age (Table 2). This suggests that there<br />
was not a difference in foraging behavior due to difference<br />
in age or experience, but rather due to the presence of their<br />
calves.<br />
There were also no significant differences for chase<br />
latencies or number of body-orienting movements for the<br />
prey species observed in this study: thus the altered foraging<br />
behaviors were also not likely due to the difficulty of<br />
catching a specific type of prey.<br />
The three episodes in which a target dolphin was<br />
observed foraging with a non-calf individual, juvenile in<br />
age or older, produced chase latencies and number of bodyorienting<br />
movements comparable to when these dolphins<br />
foraged alone, and both substantially different from the<br />
levels of behavior observed when these dolphins foraged<br />
with their calves present. This suggests that the change in<br />
behavior was not merely a social phenomenon seen in the<br />
presence of other individuals, but instead reflects teaching<br />
behavior targeted at a naïve observer. However, the naïve<br />
observers in this study were presumably the calves of the<br />
observed mothers. Future research is needed to explore if<br />
these same mothers or other alloparents also exhibit similar<br />
altered foraging behavior when in the presence of other<br />
naïve observers that are not their offspring.<br />
This altered foraging behavior on behalf of the mothers<br />
may be a valuable social learning mechanism for Atlantic<br />
spotted dolphins. The mothers in the study clearly altered<br />
their typical foraging behavior in the presence of their<br />
calves at a potential cost to themselves due to the exaggerated<br />
behavior, as well as sometimes foregoing ingestion<br />
of the prey. However, as per Caro and Hauser’s definition<br />
of teaching, the observer must benefit from the model’s<br />
altered behavior, in this case through more rapid or skillful<br />
acquisition of foraging behavior. In dolphin society this<br />
skill would be essential, as mothers give birth approximately<br />
every 3 years, at which time the older calf would<br />
become weaned and rely more on independently caught<br />
fish. It is advantageous for the mother to ensure the success<br />
of her offspring, presumably by investing her time and<br />
energy into the calf’s foraging capabilities before the calf<br />
is weaned. Weaning is a gradual process in spotted dolphins,<br />
and although the calves were not dependent upon<br />
fish for food, consuming the fish may have been an<br />
important part of the development of the calves’ foraging<br />
behavior, perhaps because it reinforced the social-learning<br />
process.<br />
Future research is warranted to explore the development<br />
and skillfulness of the young calf’s pre-weaning foraging<br />
capabilities in order to examine the full effect of the<br />
mother’s presumed teaching efforts. If this is true teaching<br />
behavior, the calf will derive some benefit from observing<br />
the mother’s altered foraging behavior. Additionally, a<br />
comparison is needed to compare the calves of teaching<br />
and non-teaching or less attentive mothers. However,<br />
because all of the mothers observed in this study demonstrated<br />
the altered foraging behavior, it may be difficult to<br />
assess the consequences of naturally occurring individual<br />
differences in the mothers’ foraging behavior on the calves.<br />
Although the full benefit for the calf is unknown, it seems<br />
123
52 Anim Cogn (2009) 12:43<strong>–</strong>53<br />
that the calf is indeed the target of this altered behavior,<br />
which was observed only in the presence of the calf.<br />
Additionally, further research is needed into the calf’s<br />
behavior and attention to the mother. Future research can<br />
hopefully clarify if the calf is indeed attentive to the<br />
mother’s potential teaching behavior to gain something<br />
from the experience, as well as clarify what mechanism<br />
the calves may be using to learn from the mothers, such<br />
as imitation, stimulus enhancement, or local enhancement.<br />
Data from this study, such as the calves being<br />
predominantly in the observation position relative to the<br />
mother during the events, some of the calves chasing and<br />
ingesting the fish, and the calves’ interactions with both<br />
the mothers and prey objects, may be evidence that the<br />
calves were attentive to the mothers’ altered foraging<br />
behavior and support the argument for teaching. Future<br />
data from the calf behavior will also hopefully<br />
strengthen our argument that the altered maternal foraging<br />
behavior may be an example of teaching. The data<br />
from the calf behavior are currently being collected and<br />
analyzed as part of a separate ongoing project and will<br />
be available for future publication. Despite this drawback,<br />
we believe that this study detailing the altered<br />
foraging behavior of the mothers is a significant finding<br />
in the area of animal cognition, even without data on the<br />
calf behavior.<br />
Despite previous research that has shown that dolphins<br />
pass mirror self-recognition tests (Reiss and Marino 2001),<br />
understand referential pointing (Herman et al. 1999), and<br />
spontaneously use referential gesturing in captive situations<br />
(Xitco et al. 2001), further evidence is needed before<br />
attributing theory of mind to dolphins, which some<br />
researchers argue is required for true teaching (Tomasello<br />
et al. 1993). Regardless of the lack of conclusive evidence<br />
supporting theory of mind in dolphins, the perspectivetaking<br />
abilities by dolphins supported by previous research<br />
(Herman et al. 1999; Xitco et al. 2001) might be sufficient<br />
for the presumed teaching behavior shown here. Although<br />
the cognitive abilities behind the clear alteration of foraging<br />
behavior of the mother dolphins in the presence of their<br />
calves are yet to be determined, the observed teaching<br />
behaviors in dolphins are nonetheless remarkable. Mother<br />
dolphins provide opportunities for calves to observe foraging<br />
behaviors, and sometimes even provide opportunities<br />
for calves to practice foraging. Teaching, then, may be an<br />
important way in which aspects of cetacean social learning<br />
and possibly culture are transmitted from one generation to<br />
the next.<br />
Acknowledgments We would like to thank Stan Kuczaj and Jesse<br />
Bering for their helpful comments on earlier drafts of this manuscript,<br />
and John Bender, Miley Fishero, Megan Rothrock, Melissa Ingui,<br />
Wisline Shepherd, Sheryl Spencer, and the Wild Dolphin Project for<br />
their assistance on the project.<br />
123<br />
References<br />
Bering JM (2001) Theistic percepts in other species: can chimpanzees<br />
represent the minds of non-natural agents? J Cogn Cult 1:107<strong>–</strong><br />
137<br />
Bering JM, Povinelli DJ (2003) Comparing cognitive development.<br />
In: Maestripieri D (ed) Primate psychology. Harvard University<br />
Press, Cambridge, pp 205<strong>–</strong>233<br />
Bjorklund DF, Bering JM (2003) Big brains, slow development, and<br />
social complexity: the developmental and evolutionary origins of<br />
social cognition. In: Brüne M, Ribbert H, Schiefenhövel W (eds)<br />
The social brain: evolutionary aspects of development and<br />
pathology. Wiley, New York, pp 133<strong>–</strong>151<br />
Bjorklund DF, Yunger JL, Bering JM, Ragan P (2002) The<br />
generalization of deferred imitation in enculturated chimpanzees<br />
(Pan troglodytes). Anim Cogn 5:49<strong>–</strong>58<br />
Boesch C (1991) Teaching in wild chimpanzees. Anim Behav<br />
41:530<strong>–</strong>532<br />
Boesch C (1993) Toward a new image of culture in chimpanzees.<br />
Behav Brain Sci 16:514<strong>–</strong>515<br />
Boesch C, Tomasello M (1998) Chimpanzee and human cultures.<br />
Curr Anthropol 39:591<strong>–</strong>614<br />
Caro TM, Hauser MD (1992) Is there teaching in nonhuman animals?<br />
Q Rev Biol 67:151<strong>–</strong><strong>17</strong>4<br />
Greenfield P, Maynard A, Boehm C, Schmidtling EY (2000) Cultural<br />
apprenticeship and cultural change: tool learning and imitation in<br />
chimpanzees and humans. In: Parker ST, Langer J, McKinney<br />
ML (eds) Biology, brains, and behavior: the evolution of human<br />
development. School of American Research Press, Santa Fe, pp<br />
237<strong>–</strong>277<br />
Guinet C, Bouvier J (1995) Development of intentional stranding<br />
hunting techniques in killer whale (Orcinus orca) calves at<br />
Crozet Archipelago. Canadian Journal of Zoology 73:27<strong>–</strong>33<br />
Herman LM (2002) Vocal, social, and self-imitation by bottlenosed<br />
dolphins. In: Nehaniv C, Dautenhahn K (eds) Imitation in<br />
animals and artifacts. MIT Press, Cambridge, pp 63<strong>–</strong>108<br />
Herman LM, Abichandani SL, Elhajj AN, Herman EYK, Sanchez JL,<br />
Pack AA (1999) Dolphins (Tursiops truncatus) comprehend the<br />
referential character of the human pointing gesture. J Comp<br />
Psychol 113:347<strong>–</strong>364<br />
Herzing DL (1996) Vocalizations and associated underwater behavior<br />
of free-ranging Atlantic spotted dolphins, Stenella frontalis, and<br />
bottlenose dolphins, Tursiops truncatus. Aquat Mamm 22:61<strong>–</strong>79<br />
Herzing DL (1997) The life history of free-ranging Atlantic spotted<br />
dolphins (Stenella frontalis): age classes, color phases an female<br />
reproduction. Mar Mamm Sci 13:576<strong>–</strong>595<br />
Herzing DL (2005) Transmission mechanisms of social learning in<br />
dolphins: underwater observations of free-ranging dolphins in<br />
the Bahamas. In: Delfour F (ed) Autour de L’ethologie et de la<br />
cognition animale. Presses Universitaires de Lyon, Lyon, pp<br />
185<strong>–</strong>194<br />
Herzing DL, Brunnick BJ (1997) Coefficients of association of<br />
reproductively active female Atlantic spotted dolphins, (Stenella<br />
frontalis). Aquat Mamm 23:155<strong>–</strong>162<br />
Kruetzen M, Mann J, Heithaus M, Connor R, Bejder L, Sherwin WB<br />
(2005) Cultural transmission of tool use in bottlenose dolphins.<br />
Proc Natl Acad Sci USA 105:8939<strong>–</strong>8943<br />
Kuczaj SAII, Paulos RD, Ramos JA (2005) Imitation in apes, children<br />
and dolphins: implications for the ontogeny and phylogeny of<br />
symbolic representation. In: Namy L (ed) Symbol use and<br />
symbolic development. MIT Press, Cambridge, pp 221<strong>–</strong>243<br />
Mann J, Smuts BB (1999) Behavioral development in wild bottlenose<br />
dolphin newborns (Tursiops sp.). Behaviour 136:529<strong>–</strong>566<br />
Marino L, Connor RC, Fordyce RE, Herman LM, Hof PR, Lefebvre<br />
L, Lusseau D, McCowan B, Nimchinsky EA, Pack AA, Rendell
Anim Cogn (2009) 12:43<strong>–</strong>53 53<br />
L, Reidenberg JS, Reiss D, Uhen MD, Van der Gucht E,<br />
Whitehead H (2007) Cetaceans have complex brains for<br />
complex cognition. PLoS Biol 5:966<strong>–</strong>972<br />
Miles JA, Herzing DL (2003) Underwater analysis of the behavioural<br />
development of free-ranging Atlantic spotted dolphin (Stenella<br />
frontalis) calves (birth to 4 years of age). Aquat Mamm 29:363<strong>–</strong><br />
377<br />
Pack AA, Herman LM (1995) Sensory integration in the bottlenosed<br />
dolphin: immediate recognition of complex shapes across the<br />
senses of echolocation and vision. J Acoust Soc Am 98:722<strong>–</strong>733<br />
Pack AA, Herman LM (2004) Dolphins (Tursiops truncatus) comprehend<br />
the referent of both static and dynamic human gazing and<br />
pointing in an object choice task. J Comp Psychol 118:160<strong>–</strong><strong>17</strong>1<br />
Reiss D, Marino L (2001) Self-recognition in the bottlenose dolphin:<br />
a case of cognitive convergence. Proc Natl Acad Sci USA<br />
98:5937<strong>–</strong>5942<br />
Rendell L, Whitehead H (2001) Culture in whales and dolphins.<br />
Behav Brain Sci 24:309<strong>–</strong>382<br />
Spininelli LHP, Jesus AH, Nascimento LF, Yamamoto ME (2006)<br />
Prey-transfer in the marine tucuxi dolphin, Sotalia fluviatilis, on<br />
the Brazilian coast. JMBA2, Biodiversity Records (published<br />
online)<br />
Thornton A, McAuliffe K (2006) Teaching in wild meerkats. Science<br />
313:227<strong>–</strong>229<br />
Tomasello M (1996) Do apes ape? In: Heyes C, Galef B (eds) Social<br />
learning in animals: the role of culture. Academic Press, San<br />
Diego, pp 319<strong>–</strong>346<br />
Tomasello M (2000) Culture and cognitive development. Curr Dir<br />
Psychol Sci 9:37<strong>–</strong>40<br />
Tomasello M, Kruger AC, Ratner HH (1993) Cultural learning.<br />
Behav Brain Sci 16:495<strong>–</strong>552<br />
Whiten A (2005) The second inheritance system of chimpanzees and<br />
humans. Nature 437:52<strong>–</strong>55<br />
Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V,<br />
Sugiyama Y, Tutin CEG, Wrangham RW, Boesch C (1999)<br />
Cultures in chimpanzees. Nature 399:682<strong>–</strong>685<br />
Xitco MJ, Gory JD, Kuczaj SAII (2001) Spontaneous pointing by<br />
bottlenose dolphins (Tursiops truncatus). Anim Cogn 4:115<strong>–</strong>123<br />
123
Aquatic Mammals 2009, 35(1), 62-71, DOI 10.1578/AM.35.1.2009.62<br />
Characterization of Resting Holes and Their Use by the<br />
Antillean Manatee (Trichechus manatus manatus)<br />
in the Drowned Cayes, Belize<br />
Marie-Lys C. Bacchus, 1, 2 Stephen G. Dunbar, 1, 2 3, 4<br />
and Caryn Self-Sullivan<br />
1 Department of Earth and Biological Sciences, Loma Linda University, Loma Linda, CA 92350, USA;<br />
E-mail: mlbacchus@gmail.com<br />
2 Marine Research Group, Loma Linda University, Loma Linda, CA 92350, USA<br />
3 Texas A&M University, Department of Wildlife and Fisheries, College Station, TX 77843, USA<br />
4 <strong>Sirenian</strong> <strong>International</strong>, Inc., 200 Stonewall Drive, Fredericksburg, VA 22401, USA<br />
Abstract<br />
In the Drowned Cayes area of Belize, manatees<br />
(Trichechus manatus manatus) are commonly<br />
observed resting in depressions in the substrate,<br />
locally referred to as manatee resting holes. To<br />
understand why manatees prefer locations with<br />
resting holes, the physical and environmental<br />
attributes of the depressions were characterized<br />
and diurnal and nocturnal use by manatees at four<br />
resting hole sites were documented over two summers.<br />
Twelve resting hole sites were compared<br />
with 20 non-resting hole sites in the Drowned<br />
Cayes, using water depth, substrate type, vegetation,<br />
water velocity, salinity, and water temperature.<br />
Four resting holes were chosen for repeated<br />
diurnal and nocturnal observations, during which<br />
sea and weather conditions were recorded in addition<br />
to the presence/absence of manatees. Resting<br />
holes were significantly deeper and had slower<br />
surface water velocity than areas without resting<br />
holes. A total of 168 point scans were conducted<br />
over 55 d, resulting in 39 manatee sightings over<br />
two summers. There was a significant difference<br />
in the number of sightings between research years<br />
and between day and night scans. Given the large<br />
number of resting holes in the Drowned Cayes,<br />
many of which are in sheltered areas with slow<br />
currents, it is possible that manatees select these<br />
spots based on the tranquility of the water and<br />
environment. The combination of slow currents,<br />
protection from waves, low numbers of boats, and<br />
nearby seagrass beds would make these ideal resting<br />
areas. These findings have implications for the<br />
conservation of important manatee habitat.<br />
Key Words: Antillean manatee, Trichechus<br />
manatus manatus, Belize, resting holes, nocturnal,<br />
diurnal, conservation, habitat, sightings<br />
Introduction<br />
The West Indian manatee (Trichechus manatus)<br />
is one of only four extant species in the Order<br />
Sirenia, which is the only group of herbivorous<br />
marine mammals. The species is subdivided into<br />
Florida (T. m. latirostris) and Antillean (T. m.<br />
manatus) manatees (Domning & Hayek, 1986).<br />
The 2007 IUCN Red List categorized the species<br />
as “vulnerable” and both subspecies as “endangered”<br />
(Deutsch et al., 2007).<br />
West Indian manatees inhabit rivers, lakes,<br />
lagoons, and coastal marine environments from<br />
southeastern North America to northeastern<br />
South America, including Central America and<br />
the Caribbean (Lefebvre et al., 2001; Deutsch<br />
et al., 2007). Within Central America, aerial<br />
surveys have indicated that the cayes east of<br />
Belize City are important manatee habitat (Auil,<br />
2004). Manatees osmoregulate and thermoregulate<br />
behaviorally by moving between activity<br />
centers (Deutsch et al., 2007). They can survive<br />
in marine environments for extended periods of<br />
time; however, a dependence on periodic access to<br />
fresh water is presumed (Ortiz et al., 1998, 1999;<br />
Deutsch et al., 2003). Thus, when found in marine<br />
or coastal environments, they tend to be located<br />
near freshwater sources (Powell et al., 1981;<br />
Powell & Rathbun, 1984; Rathbun et al., 1990;<br />
Olivera-Gomez & Mellink, 2005). Proximity to<br />
seagrass beds is another important characteristic in<br />
determining manatee habitat preference (Hartman,<br />
1979; Powell et al., 1981; Deutsch et al., 2003). In<br />
Chetumal Bay, Mexico, manatee movements were<br />
most strongly associated with food distribution<br />
(Axis-Arroyo et al., 1998), and in Florida, aerial<br />
surveys indicated a strong association between<br />
location of manatees and distribution of seagrass<br />
beds. Work by Kinnaird (1985) and Provancha<br />
& Hall (1991), showed manatee density was<br />
positively correlated with seagrass abundance
(Halodule wrightii and Syringodium filiforme) in<br />
the northern Banana River, Florida. Some of the<br />
most important manatee food items around the<br />
cayes of Belize are turtle grass (Thalassia testudinum),<br />
manatee grass (S. filiforme), and shoal grass<br />
(H. wrightii) (LaCommare et al., 2008). There<br />
is also evidence that resting areas for cows and<br />
calves, as well as travel corridors between activity<br />
centers, may be critical habitats for manatees<br />
(Packard & Wetterqvist, 1986; Deutsch et al.,<br />
2007). Major threats to survival of the species<br />
include habitat degradation and loss, illegal hunting,<br />
boat strikes, entanglement in fishing gear,<br />
entrapment in water control structures, pollution,<br />
disease, and human disturbance (Deutsch et al.,<br />
2007).<br />
Conservation of this species is dependent on<br />
conservation of habitat that meets the requirements<br />
described above. Compared to osmoregulation,<br />
thermoregulation, and foraging requirements,<br />
little is known about manatee resting<br />
behavior and the importance of resting habitat<br />
for conservation of the species. Hartman (1979)<br />
and Bengtson (1981) described Florida manatees<br />
as arrhythmic, spending most of their time feeding,<br />
resting, idling, cruising, and socializing with<br />
no correlation of activity with time of day. The<br />
hypothesis that Florida manatees lack a daily<br />
activity pattern is supported by two lines of evidence:<br />
(1) similar nocturnal and diurnal behaviors<br />
(Hartman, 1971) and (2) lack of a pineal body<br />
(Ralph et al., 1985; Wally Welker, pers. comm.).<br />
Hartman (1979) also found that Florida manatees<br />
devoted considerable time to resting behavior<br />
with no strict preference for resting sites, which<br />
included limestone shelves, oyster bars, and vegetation.<br />
However, more recent studies indicated<br />
that manatees have a diurnal or nocturnal preference<br />
for resting in different areas of Florida. For<br />
example, during winter along the Atlantic coast,<br />
manatees rested in canals during the day and foraged<br />
at night (Deutsch et al., 2003). In Sarasota,<br />
they will bottom-rest in deeper areas during the<br />
winter when it is colder, and then will switch to<br />
surface resting when temperatures increase (Sheri<br />
W. Barton, pers. comm.). Along the Gulf Coast<br />
during winter months, manatees use the warm<br />
water springs in Crystal River to rest at night and<br />
often travel downriver to feed on Ruppia during<br />
the day (Reep & Bonde, 2006). Similar evidence<br />
of resting during certain times has been found for<br />
other manatee species. Observations of a solitary<br />
Amazonian manatee (T. inunguis) suggested an<br />
activity pattern with the individual sleeping for the<br />
first half of the night (Mukhametov et al., 1992). In<br />
areas with high instances of hunting, there is evidence<br />
that manatees (including the West African<br />
manatee, T. senegalensis) may have shifted to<br />
Resting Holes of the Antillean Manatee in Belize 63<br />
greater nocturnal activity and a more diurnal resting<br />
pattern (Rathbun et al., 1983; Reynolds et al.,<br />
1995; Powell & Kouadio, 2006).<br />
Manatees also appear to follow patterns in<br />
both resting times and resting sites. During low<br />
tides and the dry season in Venezuela and Costa<br />
Rica, they often take refuge in holes or channels<br />
in the rivers that range from 6 to 12 m in depth<br />
(O’Shea et al., 1988; Smethurst & Nietschmann,<br />
1999). Anecdotal observations suggest that manatees<br />
in Nicaragua rest in quiet, sheltered, deep<br />
water during most of the day, leaving to feed only<br />
during the night, early morning, and late afternoon<br />
(Jiménez, 2002). In Belize, resting behavior<br />
has been correlated with the presence of a “resting<br />
hole” (CSS & K. LaCommare, unpub. data),<br />
which has also been identified as a significant<br />
factor for predicting the presence of manatees at<br />
54 permanent sampling points in the Drowned<br />
Cayes during a daytime study (LaCommare et al.,<br />
2008). The objectives of this study were to determine<br />
what factors influence the use of previously<br />
identified resting holes by manatees and whether<br />
manatees use the resting holes more often during<br />
the day or at night.<br />
Materials and Methods<br />
Study Area<br />
The study was conducted over the summers of<br />
2005 and 2006 in the Drowned Cayes of Belize<br />
(located about 15 km east of Belize City at N <strong>17</strong>°<br />
28.0', W 88° 04.5'; Figure 1), which are a labyrinth<br />
of mangrove islands approximately 13 km<br />
long separated by lagoons, coves, and bogues<br />
(channels that separate the mangrove cayes both<br />
within and between ranges) (Ford, 1991). The<br />
ecosystem is marine with an average salinity of<br />
35 ppt, although a few low salinity sites have been<br />
identified (CSS, unpub. data). Extensive seagrass<br />
beds surround these islands, making them excellent<br />
foraging habitats for manatees.<br />
Characterization of Resting Holes<br />
For the purpose of this study, a resting hole is<br />
defined as an area where manatees have consistently<br />
been observed in a resting state of behavior.<br />
Twelve resting holes in the Drowned Cayes<br />
were chosen as a subset of 54 permanent scanpoints<br />
where manatees have been studied since<br />
2001 (LaCommare et al., 2008). The holes were<br />
characterized by depth using a HondexTM digital<br />
depth sounder (Forestry Suppliers), substrate<br />
(mud or sand), and vegetation (seagrass, algae,<br />
none). Water temperature (using a Taylor ® classic<br />
instant read pocket digital thermometer), salinity<br />
(using a WP-84 conductivity-TDS-temperature<br />
meter from TPS Pty), and water velocity (using
64 Bacchus et al.<br />
Figure 1. Map of Belize (inset) and the Drowned Cayes in relation to Belize City<br />
a General Oceanics mechanical flowmeter model<br />
2030R from Forestry Suppliers; speed range<br />
10 cm/s to 7.9 m/s) were measured at three points<br />
in the water column—surface, mid-water, and<br />
bottom—above each resting hole. Latitude and<br />
longitude were recorded above the deepest point<br />
of each resting hole with a GPS (Garmin ® Global<br />
Positioning System 12 Personal Navigator ® ). For<br />
comparison, these variables were measured at 20<br />
non-resting hole scan-points.
Diurnal and Nocturnal Scans<br />
Four of the 12 resting holes were selected for systematic<br />
diurnal and nocturnal scans during the summers<br />
of 2005 and 2006: B1Co, 30 CCo, 31 CLa,<br />
and 50 GiCr. Two 30-min scans focused around<br />
a fixed survey point were conducted at each site<br />
daily, once between 1000 and 1500 h, and again<br />
between 1900 and 0000 h (sunset occurred ~1900<br />
h). This survey method is referred to as a point<br />
scan (Self-Sullivan et al., 2003; LaCommare et al.,<br />
2008). The order in which sites were scanned was<br />
randomly chosen by the researcher each day. An<br />
8-m fiberglass boat with an outboard engine was<br />
used with observers standing in the boat during<br />
the scans. As the scan-point was approached, the<br />
boat engine was turned off, a long pole used to<br />
pull the boat to the scan-point, and the boat tied<br />
to the grounded pole. During each point scan, the<br />
research team continuously searched for manatees<br />
by performing 180° visual scans from different<br />
points on the boat and listening for breaths (for<br />
detailed point scan methods, see Bacchus, 2007;<br />
Self-Sullivan, 2008). Nocturnal scans were performed<br />
in the same manner as diurnal scans, with<br />
the addition of spotlights to help sight manatees.<br />
Data collected during each point scan included<br />
manatee sighting (a sighting was defined as detection<br />
of one or more manatees during the 30-min<br />
scan), sea and weather conditions, tidal state, sea<br />
surface temperature (SST), salinity, and air temperature.<br />
Data Analyses<br />
All statistics were run in SPSS, Version 13.0,<br />
with α = 0.05. Summary statistics were run for<br />
resting hole characterizations and data collected<br />
during the diurnal and nocturnal scans. The variable<br />
“depth” was adjusted to account for tides<br />
using predictions from a Belize City tide chart<br />
and visually verified via water level marks on the<br />
mangrove roots during each scan. The variable<br />
“water velocity” was calculated from 5-min rotation<br />
counts using the formula provided by General<br />
Oceanics:<br />
Resting Holes of the Antillean Manatee in Belize 65<br />
(1)<br />
where 26,873 is the rotor constant, and 999,999<br />
is the maximum number of rotations divided<br />
by 5 min and multiplied by 60 s to obtain cm/s.<br />
T-tests were used to test for differences in depth<br />
and water velocity between areas with and areas<br />
without resting holes. Water velocity was ranked to<br />
adjust for unequal variances to meet assumptions<br />
of normality. T-tests were also run to analyze SST<br />
and salinity between the two years, and between<br />
day and night. Log likelihood ratio tests (G 2 ) were<br />
run to test for variance in the number of manatee<br />
sightings between the two years and between day<br />
and night. A step-wise, logistic regression was<br />
used to detect any associations between specific<br />
habitat factors and manatee sightings. Factors<br />
included location, tide, salinity, SST, day/night,<br />
and year.<br />
Results<br />
Characterization of Resting Holes<br />
There was a significant difference in water depth<br />
between resting holes and scan-points without<br />
resting holes (t(15) = 4.541, p < 0.001) with a mean<br />
difference of 1.5 ± 0.32 m. Mean water depth for<br />
resting holes was 3.5 ± 0.30 m (n = 12; range: 2.0<br />
to 5.2 m) compared with non-resting hole areas<br />
with a mean depth of 2.0 ± 0.12 m (n = 20; range:<br />
1.4 to 3.3 m) (Table 1).<br />
Mean surface water velocity for resting holes<br />
was 0.89 ± 0.51 cm/s (n = 10; range: 0 to 5.20<br />
cm/s). Non-resting hole areas had a mean surface<br />
water velocity of 4.26 ± 1.14 cm/s (n = 20; range:<br />
0.01 to <strong>17</strong>.12 cm/s) (Table 1). These represent a<br />
significant difference in surface water velocity<br />
between resting holes and areas without resting<br />
holes (t(28) = -2.880, p = 0.008). All of the holes<br />
characterized had little or no bottom vegetation.<br />
The substrate in and around the resting holes<br />
tended to be a silty mixture of mud and sand,<br />
which was easily compressed and resuspended.<br />
Sparse to abundant vegetation was found proximal<br />
to resting holes, usually consisting of shoal<br />
grass (Halodule sp.), macro algae (Halimeda and<br />
Penicillus sp.), and/or turtle grass. One of the resting<br />
holes also had a visible “manatee highway” to<br />
and from the hole where manatees swim or paddle<br />
along the bottom with their forelimbs and/or torso<br />
in contact with the seafloor. This deeper, narrow<br />
channel started at the entrance of the small cove<br />
containing the resting hole near the mangroves<br />
and followed a straight path to the deepest part of<br />
the resting hole.<br />
Diurnal and Nocturnal Scans<br />
A total of 168 scans were conducted at four resting<br />
hole sites during the summer of 2005 (n = 81)<br />
and the summer of 2006 (n = 87). Manatees were<br />
sighted during 39 scans—26 times in 2005 and 13<br />
times in 2006 (Figure 2)—representing a significant<br />
difference in the number of sightings at these<br />
four sites between years (G 2 (1) = 7.01, p < 0.01).<br />
Table 2 shows that SST and salinity were significantly<br />
different for these four points between<br />
years (SST: t(158) = 3.292, p = 0.001; salinity: t(166) =<br />
20.629, p < 0.001). There was no significant difference<br />
in air temperature between years.
66 Bacchus et al.<br />
Table 1. Descriptive statistics of environmental characteristics of areas without and with resting holes; means are reported<br />
with ± 1 SE.<br />
There were 107 day scans and 61 night scans<br />
over both summers, with 34 day sightings<br />
and 5 night sightings (Figure 3), resulting in a<br />
significant difference in the number of manatee<br />
sightings between day and night (G 2 (1) = 13.68, p <<br />
0.001). Table 3 shows that both sea surface and air<br />
temperatures were significantly different between<br />
day and night, although surface salinity was not<br />
significantly different between day and night.<br />
A forward, step-wise logistic regression was conducted<br />
to determine which independent variables<br />
Non-resting hole sites Resting hole sites<br />
N Min Max Mean ± SE N Min Max Mean ± SE Significance<br />
Depth (m) 20 1.4 3.3 2.0 ± 0.12 12 2.0 5.2 3.5 ± 0.30
Figure 3. The number of scans (white bars) and sightings<br />
(black bars) during diurnal (left) and nocturnal (right) scans<br />
for each location<br />
Characterization of Resting Holes<br />
Sites with resting holes were located in dead end<br />
bogues, lagoons, and coves close to the mangroves<br />
with little current and calm sea state. Water<br />
depth was greater and surface velocity was lower<br />
at resting holes than at scan-points without resting<br />
holes. However, comparison of water velocity<br />
between resting holes could not be made because<br />
measurements within resting holes were all less<br />
than 10 cm/s, which was the lower threshold for<br />
the flowmeter. Slow current and calm waters may<br />
enable manatees to rest without exerting energy<br />
to hold their position. These results support previous<br />
studies that reported manatees are found in<br />
areas sheltered from high currents, surf, and wind<br />
(Hartman, 1979; Powell et al., 1981; Lefebvre<br />
et al., 2001; Jiménez, 2005; Olivera-Gomez &<br />
Mellink, 2005).<br />
Although there was no seagrass growing in<br />
the resting holes themselves, many seagrass beds<br />
were nearby. This would facilitate a short distance<br />
of travel for manatees between resting sites and<br />
feeding sites in calm waters. Given their large<br />
body mass and dependence on a low-energy, lowprotein<br />
diet, manatees must spend a large proportion<br />
of their time feeding to meet metabolic<br />
requirements. Manatees tend to feed from 5 to<br />
8 h/d, consuming about 7% of their body weight in<br />
wet vegetation (Hartman, 1979; Bengtson, 1983;<br />
Resting Holes of the Antillean Manatee in Belize 67<br />
Etheridge et al., 1985). Manatees may choose foraging<br />
strategies that allow them to maximize food<br />
intake while minimizing energy output. Similar<br />
studies in the Drowned Cayes indicated that sightings<br />
of manatees were most probable near resting<br />
holes and seagrass beds (LaCommare et al., 2008).<br />
The more familiar an animal is with locations of<br />
essential resources, the more efficient it will be in<br />
energy acquisition since this reduces searching<br />
time (Deutsch et al., 2003).<br />
Manatees were frequently encountered at three<br />
of the four previously identified resting holes<br />
within the Drowned Cayes. All four holes were<br />
located in protected coves or lagoons with minimal<br />
current, making it likely that manatees choose<br />
these areas because of the quiet conditions and<br />
create, or at least maintain, resting holes at these<br />
sites. Although previously categorized as a resting<br />
hole (LaCommare et al., 2008), the scan-point<br />
B1Co was likely not being used as a resting site in<br />
the summers of 2005 and 2006; it had a uniform<br />
1.5 m depth, with patches of shoal grass and no<br />
depression. However, there were many feeding<br />
scars and forelimb marks where manatees had dug<br />
into the mud to eat seagrass roots and rhizomes.<br />
Only one manatee was observed resting at this<br />
site in the summer of 2005, and no manatees were<br />
observed resting in the cove during the summer<br />
of 2006. This may indicate that manatee use of<br />
particular resting holes changes over time.<br />
Although manatees frequently used known<br />
resting holes, there was no evidence that manatees<br />
intentionally excavated or maintained the holes.<br />
Our current hypothesis is that they make use of<br />
natural depressions at these sites. The substrate<br />
in the holes is primarily uncompacted silt with<br />
high interstitial space, which is easily stirred-up<br />
as manatees rise off the bottom to surface for<br />
breaths. By resting regularly in the same place,<br />
the presence and vertical movement of a manatee<br />
during breathing cycles could easily deepen and<br />
maintain the depression.<br />
However, we do not yet know if individual<br />
manatees return to preferred holes exclusively<br />
or if resting hole use is indiscriminate. Also, it is<br />
unknown how manatees learn about the presence<br />
Table 3. Descriptive statistics for sea surface temperature, surface salinity, and air temperature during day and night scans<br />
of four locations<br />
Day Night<br />
N Min Max Mean ± SE N Min Max Mean ± SE Significance<br />
Sea surface temperature (°C) 106 27.2 34.3 30.9 ± 0.12 61 28.6 31.8 30.4 ± 0.10 0.001<br />
Surface salinity (ppt) 107 13.0 39.0 34.0 ± 0.30 61 28.0 37.0 34.0 ± 0.30 NS<br />
Air temperature (°C) 107 26.6 33.4 31.5 ± 0.12 61 27.4 32.0 29.4 ± 0.08
68 Bacchus et al.<br />
of these specific resting holes. In Florida, individual<br />
manatees show strong site fidelity to warmseason<br />
and winter ranges where manatee mothercalf<br />
pairs have been tracked remotely over time<br />
(Deutsch et al., 2003). Use of specific sites, initially<br />
with their mothers and then as independent<br />
subadults, has demonstrated natal philopatry to<br />
specific warm-season and winter ranges as well<br />
as to migratory patterns and travel routes among<br />
sites. Similarly, Antillean manatees may learn of<br />
specific resting holes as dependent calves and<br />
return to these same areas as adults.<br />
Diurnal and Nocturnal Use of Resting Holes<br />
There was a significant difference in the number<br />
of sightings of manatees between day and night.<br />
Although spotlights were used to increase the<br />
probability of detection during night scans, we<br />
acknowledge the increased potential for observer<br />
error during nocturnal scans. However, this potential<br />
was further reduced by increased vigilance<br />
during night scans, both visually and aurally. To<br />
increase successful detection and maintain consistency<br />
in effort, the same researchers and assistants<br />
were used for all night scans during this study.<br />
Despite concerns over detectability, these data<br />
suggest that manatees used resting holes more<br />
frequently during the day than at night in the<br />
Drowned Cayes. These results are consistent with<br />
observations in Florida where manatees leave<br />
the warm water refuge of Blue Spring in the late<br />
winter afternoons to feed all night, returning to the<br />
refuge in the morning (Bengtson, 1981; Rathbun<br />
et al., 1990). However, the animals abandoned<br />
the diurnal resting behavior in Blue Spring when<br />
water temperatures warmed in the spring. If use<br />
of resting sites is driven more by resources than<br />
a daily activity pattern, perhaps manatees use the<br />
resting holes in the current study area during the<br />
day when human activity on the water is high,<br />
then leave at night to feed in nearby seagrass beds<br />
or travel to freshwater sites. Interviews with local<br />
people indicated a similar pattern in Nicaragua<br />
where manatees rest in deep, quiet waters during<br />
the day and move to shallow waters to feed during<br />
late afternoon, night, and early morning, presumably<br />
to avoid hunting predation (Jiménez, 2002).<br />
Another possible reason for different resting<br />
hole uses during the day and night is the need for<br />
fresh water. Behavioral, ecological, and physiological<br />
evidence suggest that manatees require<br />
fresh or low salinity water for osmoregulation<br />
when living in a marine environment and feeding<br />
exclusively on seagrasses (Hartman, 1979; Powell<br />
& Rathbun, 1984; Ortiz et al., 1998, 1999). A lack<br />
of fresh water for extended periods leads to dehydration,<br />
although manatees may be able to osmoregulate<br />
via oxidation of fat for short periods of<br />
time (Ortiz et al., 1999). In Puerto Rico, where<br />
manatees are primarily observed in marine habitats,<br />
Powell et al. (1981) noted that 85.8% of<br />
manatee sightings were within 5 km of freshwater<br />
sources. Those authors also reported anecdotes<br />
by local fisherman indicating that manatees often<br />
visited nearby rivers to drink water. In aerial surveys<br />
in Belize, a higher number of manatees have<br />
been observed in near-shore habitat (e.g., estuary,<br />
lagoon, and river) than in cay habitat, with one<br />
of the highest numbers recorded from the Belize<br />
River (Auil, 2004). Also, more manatees were<br />
sighted in near-shore habitat during the dry season<br />
(November to May), and more manatees were<br />
sighted in the cay habitat during the wet season<br />
(May/June to November) (Auil, 2004).<br />
Similar to the situation in Ten Thousand Islands,<br />
Florida (Butler et al., 2003; J. P. Reid, pers.<br />
comm.), manatees in the Drowned Cayes travel to<br />
the Belize River, the closest source of substantial<br />
fresh water (about 12 to 15 km away) during the dry<br />
season (Auil, 2004). Manatees may also be traveling<br />
to the Belize River on a shorter time frame.<br />
Since boat traffic, both on the river and between<br />
the river and the cayes, is much higher during the<br />
day, manatees that travel from the Drowned Cayes<br />
to the river at night may reduce the probability<br />
of being struck by boats. This hypothesis is supported<br />
by studies in Florida and Costa Rica, where<br />
manatees have been known to avoid areas of high<br />
boat traffic, changing their behavioral state and<br />
moving out of a geographical area in response to<br />
approaching vessels (Buckingham et al., 1999;<br />
Smethurst & Nietschmann, 1999; Miksis-Olds,<br />
2006).<br />
There was a significant difference in sightings<br />
between the two summers with twice as<br />
many manatees seen in the summer of 2005 than<br />
in the summer of 2006. The reasons for this are<br />
unclear, but longer-term data for all 54 scanpoints<br />
detected no significant difference in the<br />
probability of encountering manatees over the<br />
period of 2001 to 2004, even as a result of season<br />
(Self-Sullivan, 2008). There was a significant<br />
difference in water temperature and salinity for<br />
the four scan locations between these two summers,<br />
with 2005 being warmer and more saline.<br />
However, the difference in temperature (0.5° C),<br />
although significantly different between years, is<br />
relatively small and probably not significant to<br />
manatees given what is known about their range.<br />
A similar situation occurred with day and night<br />
SST with only a 0.5° C difference. Although manatees<br />
can sense water temperature and salinity differences<br />
(Deutsch et al., 2003), the differences in<br />
the temperature readings in this study seem small<br />
enough to insignificantly affect manatee habitat<br />
choices. Salinity was 36 ppt in 2005 and 32
in 2006. Although this difference is fairly small,<br />
this may indicate a slightly more rainy summer<br />
season, which might have some impact on the<br />
probability of manatee presence at the four resting<br />
holes studied. Even though manatees were<br />
not seen as often in the summer of 2006, this does<br />
not necessarily mean that they used the Drowned<br />
Cayes habitat less frequently but rather that they<br />
were not detected as often at the particular resting<br />
holes being studied.<br />
West Indian manatee populations must deal<br />
with environmental changes on a daily (e.g.,<br />
tides, currents, weather, human activity levels),<br />
seasonal (e.g., rainfall, temperature), and generational<br />
basis as human development continues to<br />
dramatically alter their habitats throughout the<br />
Caribbean (Deutsch et al., 2007). Although little<br />
is known about habitat use in the population over<br />
long periods of time, changes in habitat use by<br />
manatees have been observed at Swallow Caye<br />
in the Drowned Cayes of Belize, presumably as<br />
a result of increased human disturbance. When<br />
tour operators began and then ended a "swimwith"<br />
program at Swallow Caye, manatee use of<br />
the resting hole decreased during the program but<br />
increased again when the program was discontinued<br />
(Self-Sullivan, 2008). For generations of<br />
manatees, the undeveloped Drowned Cayes have<br />
provided a safe haven within a short distance of<br />
essential fresh water from the Belize River.<br />
Between 2001 and 2004, there was an exponential<br />
increase in cruise ship tourism in Belize.<br />
In 2001, there were 48 cruise ships with 48,116<br />
passengers; in 2002, there were 200 ships with<br />
319,690 passengers, representing a 584% growth;<br />
in 2003, there were 315 ships with 575,196 passengers,<br />
an 80% growth from 2002; and in 2004,<br />
there were 406 ships with 851,436 passengers,<br />
which is a 49% growth from 2003 (Belize Tourism<br />
Board, 2006). Although cruise ship numbers have<br />
leveled-off since 2005, the impact of cruise ship<br />
tourism remains above the 2003 level with a continued<br />
increase in development and human disturbance<br />
of manatee habitat in the Drowned Cayes<br />
area. Manatee mortality rate near Belize City is<br />
relatively high (Auil, 2004). This might be due to<br />
the increase in boat traffic as passengers are ferried<br />
from the ships by tender boats and are taken<br />
on tours to the reef, around the Drowned Cayes,<br />
and to small sandy cayes in the area (Self-Sullivan,<br />
2008). Also, there has been an increase in development<br />
of some of the other cayes in the Drowned<br />
Cayes, causing increases in boat traffic in these<br />
areas with the potential to cause siltation that may<br />
destroy the seagrass beds that the manatees feed<br />
on.<br />
The implications of our findings should be taken<br />
into consideration by conservation managers who<br />
Resting Holes of the Antillean Manatee in Belize 69<br />
are concerned with development in the Drowned<br />
Cayes. Conservation strategies designed to protect<br />
manatees should include preservation of<br />
low-energy, secluded areas where manatees can<br />
continue to maintain resting holes near seagrass<br />
beds. Results of the current study not only increase<br />
understanding of manatee resting habitats but provide<br />
necessary data for decisionmakers and wildlife<br />
managers to establish limits of acceptable change<br />
within the Drowned Cayes area and other strategies<br />
for the conservation of manatee habitat in Belize.<br />
Acknowledgments<br />
This work was supported by a Marine Research<br />
Group (LLU) grant to M-LB, funding from Dr.<br />
Robert Ford in the Department of Social Work and<br />
Social Ecology (LLU), the Earthwatch Institute,<br />
and the Hugh Parkey Foundation for Marine<br />
Awareness and Education. We thank our Belizean<br />
field assistants and boat drivers, Gilroy Robinson<br />
and Dorian Alvarez, as well as field assistants<br />
Leigh Bird, Shauna King, Haydée Domínguez,<br />
Karen Petkau, Ryan Roland, Austin Bacchus,<br />
Coralie Lallemand, and the volunteers from the<br />
Earthwatch Teams 2, 3, and 4 in 2005 and 2006.<br />
We are also grateful to multiple colleagues at<br />
Loma Linda University; Dr. William Hayes and<br />
Dr. Zia Nisani who helped with statistics; and<br />
Daniel Gonzalez who helped in the field. Thanks<br />
to Rick Ware for providing the Belize City tide<br />
tables. Our research was carried out in compliance<br />
with the laws of the United States and Belize under<br />
MNREI Forest Department Permits, Ref. Nos.<br />
CD/60/3/05(29) and CD/60/3/06(32), and LLU<br />
IACUC Protocol #86006. This is contribution<br />
Number 8 of the Marine Research Group (LLU).<br />
Literature Cited<br />
Auil, N. E. (2004). Abundance and distribution trends of<br />
the West Indian manatee in the coastal zone of Belize:<br />
Implications for conservation. Master’s thesis, Texas<br />
A&M University, College Station. 83 pp.<br />
Axis-Arroyo, J., Morales-Vela, B., Torruco-Gomez, D., &<br />
Vega-Cendejas, M. E. (1998). Variables asociadas con el<br />
uso de habitat del manatí del Caribe (Trichechus manatus),<br />
en Quintana Roo, México (Mammalia). Revista de<br />
Biologia Tropical, 46, 791-803.<br />
Bacchus, M-L. C. (2007). Characterization of resting holes<br />
and use by the Antillean manatee (Trichechus manatus<br />
manatus). Master’s thesis, Loma Linda University, CA.<br />
84 pp.<br />
Belize Tourism Board. (2006). Cruise arrivals by month.<br />
Retrieved December 2006 from www.belizetourism.org.<br />
Bengtson, J. L. (1981). Ecology of manatees (Trichechus<br />
manatus) in the St. Johns River, Florida. Ph.D. dissertation,<br />
University of Minnesota, Minneapolis. 126 pp.
70 Bacchus et al.<br />
Bengtson, J. L. (1983). Estimating food consumption of<br />
free-ranging manatees in Florida. Journal of Wildlife<br />
Management, 47, 1186-1192.<br />
Buckingham, C. A., Lefebvre, L. W., Schaefer, J. M., &<br />
Kochman, H. I. (1999). Manatee response to boating<br />
activity in a thermal refuge. Wildlife Society Bulletin,<br />
27, 514-522.<br />
Butler, S. M., Reid, J. P., & Stith, B. M. (2003). Detailed<br />
movements and habitat use patterns of radio tagged<br />
manatees in the western Everglades. Poster presentation<br />
at the 15th Biennial Conference on the Biology of<br />
Marine Mammals. Greensboro, NC.<br />
Deutsch, C. J., Self-Sullivan, C., & Mignucci-Giannoni,<br />
A. A. (2007). Trichechus manatus. In 2007 IUCN<br />
red list of threatened species. Gland, Switzerland:<br />
<strong>International</strong> Union for Conservation of Nature and<br />
Natural Resources.<br />
Deutsch, C. J., Reid, J. P., Bonde, R. K., Easton, D. E.,<br />
Kochman, H. I., & O’Shea, T. J. (2003). Seasonal movements,<br />
migratory behavior, and site fidelity of West<br />
Indian manatees along the Atlantic coast of the United<br />
States. Wildlife Monographs, 151, 1-77.<br />
Domning, D. P., & Hayek, L. C. (1986). Interspecific<br />
and intraspecific morphological variation in manatees<br />
(Sirenia: Trichechus). Marine Mammal Science, 2,<br />
87-144.<br />
Etheridge, K., Rathbun, G. B., Powell, J. A., & Kochman,<br />
H. I. (1985). Consumption of aquatic plants by the West<br />
Indian manatee. Journal of Aquatic Plant Management,<br />
23, 21-25.<br />
Ford, R. E. (1991). Toponymic generics, environment and<br />
culture history in pre-independence Belize. Names, 39,<br />
1-25.<br />
Hartman, D. S. (1971). Behavior and ecology of the Florida<br />
manatee, Trichechus manatus latirostris (Harlan), at<br />
Crystal River, Citrus County. Ph.D. dissertation, Cornell<br />
University, Ithaca, NY. 285 pp.<br />
Hartman, D. S. (1979). Ecology and behavior of the manatee<br />
(Trichechus manatus) in Florida. American Society<br />
of Mammalogists, Special Publication 5.<br />
Jiménez, I. (2002). Heavy poaching in prime habitat:<br />
The conservation status of the West Indian manatee in<br />
Nicaragua. Oryx, 36, 272-278.<br />
Jiménez, I. (2005). Development of predictive models to<br />
explain the distribution of the West Indian manatee<br />
Trichechus manatus in tropical watercourses. Biological<br />
Conservation, 125, 491-503.<br />
Kinnaird, M. F. (1985). Aerial census of manatees in northeastern<br />
Florida. Biological Conservation, 32, 59-79.<br />
LaCommare, K. S., Self-Sullivan, C., & Brault, S. (2008).<br />
Distribution and habitat use of Antillean manatees<br />
(Trichechus manatus manatus) in the Drowned Cayes<br />
area of Belize, Central America. Aquatic Mammals,<br />
34(1), 35-43.<br />
Lefebvre, L. W., Marmontel, M., Reid, J. P., Rathbun, G. B.,<br />
& Domning, D. P. (2001). Status and biogeography of<br />
the West Indian manatee. In C. A. Wood & F. E. Sergile<br />
(Eds.), Biogeography of the West Indies: Patterns and<br />
perspectives (2nd ed.) (pp. 425-474). Boca Raton, FL:<br />
CRC Press.<br />
Miksis-Olds, J. L. (2006). Manatee response to environmental<br />
noise. Ph.D. dissertation, University of Rhode<br />
Island, Narragansett. 228 pp.<br />
Mukhametov, L. M., Lyamin, O. I., Chetyrbok, I. S.,<br />
Vassilyev, A. A., & Diaz, R. P. (1992). Sleep in an<br />
Amazonian manatee, Trichechus inunguis. Experientia,<br />
48, 4<strong>17</strong>-419.<br />
Olivera-Gomez, L. D., & Mellink, E. (2005). Distribution<br />
of the Antillean manatee (Trichechus manatus manatus)<br />
as a function of habitat characteristics, in Bahia de<br />
Chetumal, Mexico. Biological Conservation, 121, 127-<br />
133.<br />
Ortiz, R. M., Worthy, G. A. J., & Byers, F. M. (1999).<br />
Estimation of water turnover rates of captive West Indian<br />
manatees (Trichechus manatus) held in fresh and salt<br />
water. Journal of Experimental Biology, 202, 33-38.<br />
Ortiz, R. M., Worthy, G. A. J., & MacKenzie, D. S. (1998).<br />
Osmoregulation in wild and captive West Indian manatees<br />
(Trichechus manatus). Physiological Zoology, 71,<br />
449-457.<br />
O’Shea, T. J., Correa-Viana, M., Ludlow, M. E., & Robinson,<br />
J. G. (1988). Distribution, status, and traditional significance<br />
of the West Indian manatee Trichechus manatus in<br />
Venezuela. Biological Conservation, 46, 281-301.<br />
Packard, J. M., & Wetterqvist, O. F. (1986). Evaluation of<br />
manatee habitat systems on the northwestern Florida<br />
coast. Coastal Zone Management Journal, 14, 279-310.<br />
Powell, J. A., & Kouadio, A. (2006). Trichechus senegalensis.<br />
In 2007 IUCN red list of threatened species. Gland,<br />
Switzerland: <strong>International</strong> Union for Conservation of<br />
Nature and Natural Resources.<br />
Powell, J. A., & Rathbun, G. B. (1984). Distribution and<br />
abundance of manatees along the northern coast of the<br />
gulf of Mexico. Northeast Gulf Science, 7, 1-28.<br />
Powell, J. A., Belitsky, D. W., & Rathbun, G. B. (1981).<br />
Status of the West Indian manatee (Trichechus manatus)<br />
in Puerto Rico. Journal of Mammalogy, 62, 642-646.<br />
Provancha, J. A., & Hall, C. R. (1991). Observations of<br />
associations between seagrass beds and manatees in east<br />
central Florida. Florida Scientist, 54, 87-98.<br />
Ralph, C. L., Young, S., Gettinger, R., & O’Shea, T. J.<br />
(1985). Does the manatee have a pineal body? Acta<br />
Zoologica, 66, 55-60.<br />
Rathbun, G. B., Powell, J. A., & Cruz, G. (1983). Status<br />
of the West Indian manatee in Honduras. Biological<br />
Conservation, 26, 301-308.<br />
Rathbun, G. B., Reid, J. P., & Carowan, G. (1990).<br />
Distribution and movement patterns of manatees<br />
(Trichechus manatus) in northwestern peninsular<br />
Florida. Florida Marine Research Publications, 48,<br />
1-33.<br />
Reep, R. L., & Bonde, R. K. (2006). The Florida manatee:<br />
Biology and conservation. Gainesville: University Press<br />
of Florida. 189 pp.<br />
Reynolds, J. E., Szelistowski, W. A., & Leon, M. A. (1995).<br />
Status and conservation of manatees Trichechus manatus
manatus in Costa Rica. Biological Conservation, 71,<br />
193-196.<br />
Self-Sullivan, C. (2008). Conservation of Antillean manatees<br />
in the Drowned Cayes area of Belize. Ph.D. dissertation,<br />
Texas A&M University, College Station. 129 pp.<br />
Self-Sullivan, C., Smith, G. W., Packard, J. M., &<br />
LaCommare, K. S. (2003). Seasonal occurrence of male<br />
Antillean manatees (Trichechus manatus manatus) on<br />
the Belize Barrier Reef. Aquatic Mammals, 29(3), 342-<br />
354.<br />
Smethurst, D., & Nietschmann, B. (1999). The distribution<br />
of manatees (Trichechus manatus) in the coastal<br />
waterways of Tortuguero, Costa Rica. Biological<br />
Conservation, 89, 267-274.<br />
Resting Holes of the Antillean Manatee in Belize 71
J Coast Conserv (2011) 15:573<strong>–</strong>583<br />
DOI 10.1007/s11852-011-0146-3<br />
Aerial surveys of manatees (Trichechus manatus) in Lee<br />
County, Florida, provide insights regarding manatee<br />
abundance and real time information for managers<br />
and enforcement officers<br />
Deirdre J. Semeyn & Carolyn C. Cush &<br />
Kerri M. Scolardi & Jennifer Hebert &<br />
Justin D. McBride & Denis Grealish & John E. Reynolds<br />
Received: 8 October 2010 /Accepted: 7 January 2011 /Published online: 28 January 2011<br />
# Springer Science+Business Media B.V. 2011<br />
Abstract Conservation and management of the endangered<br />
Florida manatee is often centered on reducing mortality<br />
caused by watercraft collisions. Lee County, Florida, has<br />
led the state in watercraft-related mortality for eight of the<br />
last 10 years. This county is of particular concern as it<br />
contains important habitat for manatees, including extensive<br />
feeding grounds and an artificial warm-water refuge<br />
where more than 900 manatees have been recorded on a<br />
single day. Distributional aerial surveys were conducted<br />
from April 2007 through April 2009 over Lee County<br />
waters. Surveys yielded higher numbers of manatees than<br />
D. J. Semeyn<br />
5507 S. Bernie St.,<br />
Tampa, FL 33611, USA<br />
C. C. Cush : K. M. Scolardi : J. E. Reynolds (*)<br />
Mote Marine Laboratory,<br />
1600 Ken Thompson Parkway,<br />
Sarasota, FL 34236, USA<br />
e-mail: reynolds@mote.org<br />
J. Hebert<br />
2740 Chariton Street,<br />
Oakton, VA 22124, USA<br />
J. D. McBride<br />
Lee County Division of Natural Resources Marine Program,<br />
1500 Monroe Street,<br />
Fort Myers, FL 33902, USA<br />
D. Grealish<br />
Florida Fish and Wildlife Conservation Commission,<br />
Division of Law Enforcement,<br />
Edwards Drive,<br />
Fort Myers, FL 33901, USA<br />
previously observed in this area. Using GIS methodology,<br />
kernel density analysis illustrated seasonal changes in<br />
distribution patterns and highlighted areas where manatees<br />
were most densely clustered. For example, during summer<br />
months, manatees were widely distributed throughout the<br />
survey area, with high-density areas associated with<br />
seagrass beds. During winter months, manatees were<br />
densely clustered at warm-water sites and over feeding<br />
grounds within close distance of these sites. These seasonal<br />
distribution patterns coincide well with speed zone designations.<br />
Counts and distributions of manatees were made<br />
available, almost immediately if necessary, to local marine<br />
law enforcement in an attempt to focus resources toward<br />
reducing manatee-watercraft collisions. Future studies<br />
should implement similar communication strategies to<br />
improve conservation efforts.<br />
Keywords Florida manatee . Aerial survey . Kernel density<br />
analysis . Lee County . Conservation . Management<br />
Introduction<br />
Over the past 40 years, aerial surveys have been used to<br />
monitor and assess number and distribution of Florida<br />
manatees, Trichechus manatus latirostris (e.g., Craig and<br />
Reynolds 2004; Wright et al. 2002). Aerial survey data are<br />
one of the primary sources of information used to create<br />
state and federal manatee sanctuaries and boat speed<br />
regulatory zones, as well as individual county manatee<br />
management plans (Laist and Shaw 2006).<br />
Listed as endangered under the Endangered Species Act of<br />
1973, the Florida manatee continues to face anthropogenic
574 D. J. Semeyn et al.<br />
and natural hazards that threaten the long-term viability of the<br />
species. Human related threats contributing to manatee<br />
mortality include watercraft collisions and coastal development<br />
leading to habitat loss and degradation (e.g. effects on<br />
warm-water availability). Naturally occurring events such as<br />
harmful algal blooms, intense coastal storms, and prolonged<br />
exposure to cold temperatures are also major causes of<br />
mortality in manatees (Ackerman et al. 1995; Langtimmand<br />
Beck 2003). The subpopulation of manatees in southwestern<br />
Florida is affected by all of these anthropogenic and natural<br />
factors, leading to the highest annual mortality rate (Florida<br />
Fish and Wildlife Conservation Commission 2010) and<br />
lowest estimated adult survival rates for manatees in the<br />
state (Langtimm et al. 2004).<br />
Lee County is often the focus of scrutiny with regard to<br />
management of manatees in southwestern Florida. This<br />
county contains critical habitat for manatees with over 590<br />
miles of natural shoreline (USA Cities 2009), extensive<br />
seagrass beds, and three important warm-water refuges:<br />
Florida Power & Light Company power plant (FPL) on the<br />
Orange River, the dredged canals of Matlacha Isles, and 10<br />
Mile Canal off of Estero Bay. It is also home to nearly<br />
600,000 people, and 50,464 total registered vessels, 48,853<br />
of which are recreational (United States Census Bureau<br />
2009; Florida Fish and Wildlife Conservation Commission<br />
2008). Consequently, manatee and human use of waterways<br />
overlap, creating increased potential for watercraft related<br />
collisions and mortality.<br />
The Florida Fish and Wildlife Conservation Commission<br />
(FWC) and U.S. Fish and Wildlife Service (USFWS) have<br />
developed and implemented boat speed and boat access<br />
regulations throughout the county in an attempt to reduce<br />
watercraft mortality. Seven types of regulatory zones exist<br />
throughout Lee County, although it is mainly comprised of three:<br />
1) Slow speed all year;<br />
2) Slow speed April 1-November 15; 25 mph rest of year;<br />
and<br />
3) 25 mph within marked channel.<br />
Even with these regulatory zones, Lee County surpasses<br />
almost all other counties in terms of annual human-related<br />
mortality of manatees, with an average of 15 per year over the<br />
last 10 years (Florida Fish and Wildlife Conservation<br />
Commission 2010). The Potential Biological Removal level<br />
(PBR) determined for the statewide population of manatees<br />
is 12 individuals (U.S. Fish and Wildlife Service 2009). 1<br />
Thus, human-related mortality in this one county exceeds the<br />
current (i.e. 2010) maximum number of manatees that may<br />
1 NOTE that this figure was calculated prior to the extremely high<br />
aerial survey counts of January 2010. As a consequence of there being<br />
more manatees than was previously believed, the PBR value of 12<br />
manatees is likely to be adjusted upward.<br />
be removed from the entire statewide population by such<br />
causes each year (U.S. Fish and Wildlife Service 2009).<br />
The original objective of this study was to conduct aerial<br />
surveys in order to provide current distributional data to Lee<br />
County managers for use in reviews of permits for new or<br />
modified marine facilities or other developments. This county<br />
is one of the 13 “key counties” in Florida, which are mandated<br />
by the State to develop and implement a Manatee Protection<br />
Plan (MPP). A major component of each MMP is the Marine<br />
Facility Siting Element (MFSE). The MFSE component of the<br />
Lee County MPP creates a slip-to-shoreline ratio for existing<br />
and proposed projects, based upon each project’s projected<br />
risk to manatees. The MFSE utilizes a series of input<br />
components including manatee mortality and distribution<br />
data. Whereas mortality data are updated each calendar year,<br />
until the completion of this project, Lee County resource<br />
managers had used distribution data collected in 1997<strong>–</strong>1998<br />
to evaluate marine facilities proposals. As the study progressed,<br />
we realized that managers could utilize survey data<br />
immediately to enhance, and later to evaluate, two other<br />
features of the MPP: law enforcement coordination and sitespecific<br />
boat speed zones (mentioned above).<br />
This study was not designed to provide a statistically robust<br />
estimate of manatee abundance in Lee County. However, it<br />
does provide an approach for monitoring spatial and temporal<br />
changes in distribution and relative abundance of manatees in<br />
order to formulate effective and proactive management and<br />
enforcement for this highly protected endangered species.<br />
Methods<br />
Aerial survey<br />
Aerial surveys were selected as an efficient and cost effective<br />
method to collect information regarding number and distribution<br />
of manatees (Reynolds et al. 2010). Lee County<br />
encompasses 803.6 square miles, and is criss-crossed with<br />
rivers, creeks, islands and canals (United States Census<br />
Bureau 2009). To prevent observer fatigue and ensure the<br />
entire county was surveyed in optimal light and weather<br />
conditions, the survey area was divided into northern and<br />
southern sections. From April 2007 through April 2009, two<br />
simultaneous surveys were flown twice each month; one<br />
covered waters of the northern county, and the other surveyed<br />
waters of the southern portion. Each route followed a<br />
racetrack pattern along shorelines and over open water. With<br />
the exception of the Gulf of Mexico, the entirety of Lee<br />
County’s waterways was surveyed as well as five miles<br />
beyond the county boundary where a water body continued<br />
into Charlotte, Collier, or Hendry Counties. Two high-winged<br />
aircraft (Cessna <strong>17</strong>2 or 182) were used. Each plane flew at an<br />
altitude of 250 m and a speed of 80<strong>–</strong>90 knots.
Aerial surveys of manatees (Trichechus manatus) in Lee County, Florida 575<br />
A Garmin GPSmap 76 recorded the survey path in each<br />
aircraft. Observers recorded the number of manatees (adults<br />
and calves), location, habitat and behavior for each sighting.<br />
Manatees within close proximity of each other and displaying<br />
similar behavior were grouped together as one sighting. Wind<br />
speed and direction were recorded based on local airport<br />
conditions. Other environmental conditions such as water<br />
clarity, percent cloud cover, and water surface conditions were<br />
estimated and recorded by the surveyor. To ensure that water<br />
clarity and visibility were acceptable for efficient spotting of<br />
manatees, surveys were terminated when sustained winds<br />
exceeded 15 knots or surface conditions scored four or higher<br />
on the Beaufort Wind Scale.<br />
Spatial analysis<br />
Kernel density estimation calculates home range and animal<br />
density by estimating the probability of locating an<br />
individual at a specific place and time (Horne and Garton<br />
2006; Worton 1995). The method creates a “kernel” or<br />
probability density over each point in a dataset (Seaman<br />
and Powell 1996). The density estimate is calculated by<br />
averaging the densities of all kernels that overlap each point.<br />
Kernel density acts as a decay-function with high estimates in<br />
locations with numerous points (sightings), and low densities<br />
in locations with few points (Seaman and Powell 1996).<br />
Using ArcGIS 9.3 (ESRI), survey data were exported as<br />
point shapefiles, and the Spatial Analyst Extension was used to<br />
calculate kernel densities for the total number of manatees per<br />
time period in the survey area. To reflect seasonal shifts,<br />
consideration of manatee distribution and habitat use is often<br />
divided into two periods: winter (November 15<strong>–</strong>March 15) and<br />
non-winter (March 16<strong>–</strong>November 14). These two seasonal<br />
designations were used to create two new point shape files<br />
from the survey point dataset. In addition, kernel densities were<br />
generated for sightings in the Central region (Fig. 1) forwinter<br />
(as usually defined, above) and three other time periods: postwinter<br />
(March 16<strong>–</strong>April 30), summer (May 1<strong>–</strong>September 30),<br />
and pre-winter (October 1<strong>–</strong>November 14). The Central region<br />
is of particular importance because 1) its seagrass beds are the<br />
closest foraging grounds to the primary warm-water refuge at<br />
the FPL power plant (Fig. 2) and 2) most manatees that use<br />
this refuge must travel through the Central region. This region<br />
was analyzed separately and according to shorter time periods<br />
in order to identify fine-scale temporal and spatial changes in<br />
the distribution of manatees in this area.<br />
An output cell size of 90 and a search radius of 2500m 2<br />
were applied to both analyses to generate density estimates<br />
for each survey point. The distribution of data was best<br />
represented on maps by using quantile classifications of<br />
numeric data. This method creates classes that may be close<br />
together in terms of the values, but provides a scale that<br />
groups the majority of the data most appropriately given the<br />
density distribution of manatees at survey points. Color<br />
intensity was used to depict areas of high and low density.<br />
Management<br />
When large numbers or notable shifts in distribution of<br />
manatees occurred, counts and observations were reported to<br />
local and state managers and shared with law enforcement via<br />
the Lee County Marine Law Enforcement Task Force<br />
(LCMLETF). This group was formed in 2003 and includes<br />
members from the Lee County Sheriff’s Office, Fort Myers<br />
Police Department, Cape Coral Police Department, Sanibel<br />
Police Department, U.S. Coast Guard Station Fort Myers<br />
Beach, U.S. Coast Guard Cutter Marlin, USFWS Division of<br />
Law Enforcement, USFWS/J.N. “Ding” Darling National<br />
Wildlife Refuge, FWC Division of Law Enforcement Fort<br />
Myers Field Office, Florida Department of Environmental<br />
Protection Division of Law Enforcement, U. S. Power<br />
Squadrons, and U.S. Army Corps of Engineers. The group’s<br />
Mission Statement specifically addresses resource protection:<br />
“The agencies of the Lee County Marine Law<br />
Enforcement Task Force are committed to providing<br />
the highest quality of marine law enforcement to<br />
protect the users of Lee County’s waterways, safeguard<br />
property, and conserve/protect marine life<br />
along with its environment.” (Lee County Marine<br />
Law Enforcement Task Force 2004)<br />
This well-organized group meets monthly to coordinate<br />
and maximize efforts to enforce boating safety and marine<br />
wildlife regulations. The goal of providing survey information<br />
to this task force was to allow the leaders of state and<br />
local law enforcement to deploy officers in ways that<br />
optimized manatee protection, while also ensuring that<br />
other enforcement obligations were handled as well.<br />
Current speed zone regulations for the county were<br />
acquired (Fig. 3) and compared with the kernel density maps<br />
to evaluate the zonal boundaries, both year-round and<br />
seasonal, and the potential need for regulatory improvement.<br />
Results<br />
Survey counts<br />
A total of 34 flights with both the north and south surveys<br />
completed was considered for both spatial and numerical<br />
analysis; all incomplete flights (n=3) were removed from<br />
the dataset for this analysis. Each point (n=4,000)<br />
corresponded to a sighting of 1<strong>–</strong>102 manatees. The<br />
majority of sightings were single animals or small groups<br />
(mode=1, mean=3). Total counts among survey dates for<br />
northern and southern surveys combined ranged from 135
576 D. J. Semeyn et al.<br />
Fig. 1 Reference map of study<br />
area showing the locations of<br />
warm-water refuges at FPL<br />
Fort Myers, Matlacha Isles and<br />
Ten Mile Canal. The spatial<br />
extent of the Central region is<br />
indicated by the black rectangle.<br />
manatees to 626 manatees (Table 1). The mean number of<br />
manatees counted per flight was 345 (± 23). An independent<br />
T-test showed that mean counts were significantly<br />
different (P
Aerial surveys of manatees (Trichechus manatus) in Lee County, Florida 577<br />
Fig. 2 Map illustrating locations of seagrass beds throughout the entire<br />
study area (source: South Florida Water Management District 2003)<br />
(Fig. 5a, b). This movement coincides with a significant<br />
increase in mean number of manatees sighted per survey<br />
(p=0.00102). From pre-winter to winter, manatees<br />
became even more densely clustered (Fig. 5c), but the<br />
increase in the mean number of manatees per survey was<br />
not significantly different (p=0.582). During post-winter<br />
there were generally fewer high-density areas and fewer<br />
manatees sighted than there were for winter, indicating a<br />
dispersal of manatees as the weather warmed (Fig. 5d).<br />
Management<br />
With regard to our study, the LCMLETF provided a stable<br />
infrastructure for quick dissemination of individual flight<br />
results to enhance conservation efforts with the near real-time<br />
information gathered. The information we provided regarding<br />
distribution, density, and behavior of manatees was reported to<br />
resource managers and LCMLETF coordinators, and was<br />
subsequently used to position marine enforcement units in the<br />
most critical areas to provide additional protection in needed<br />
areas.<br />
A comparison between the Kernel density analyses<br />
(Figs. 4 and 5) and the county’s site-specific boat speed<br />
zones (Fig. 3) show that, with a few exceptions, areas with<br />
high densities of manatees year-round overlap well with<br />
year-round speed zones, and areas with high densities<br />
during “non-winter” months overlap remarkably well with<br />
the seasonally (Apr 1<strong>–</strong>Nov 15) designated zones. There<br />
appear to be relatively few areas where high densities of<br />
manatees are found year-round in seasonally designated<br />
zones and vice versa. While it would be in the interest of<br />
local and state officials to identify these areas where<br />
enforcement zones could be added or re-assessed, it is not<br />
within the scope of this paper to do so.<br />
Discussion<br />
Aerial survey counts<br />
The results reaffirm that Lee County supports large<br />
numbers of manatees, particularly during fall and winter<br />
months, and further emphasize the need for careful<br />
management of this particular population. These results<br />
can be compared to results from statewide synoptic surveys<br />
which also indicate that a large proportion of the state’s<br />
manatees is found in Lee County during winter. In the<br />
January 2010 synoptic survey, biologists counted 5,067<br />
manatees statewide, with 877 (approximately <strong>17</strong>%) in Lee<br />
County (H. Edwards FWC pers. comm.). This count was<br />
approximately 200 manatees more than the highest count in<br />
this study; however synoptic surveys are flown on days<br />
where sighting conditions are optimal. In February of the<br />
same year, after a prolonged period of cold weather, 905<br />
manatees were counted at the Orange River FPL plant alone<br />
(Reynolds, unpub. data).<br />
Our surveys produced much higher counts of manatees<br />
than did previous studies conducted in the area. The most<br />
recent study prior to ours was completed by FWC from<br />
1994 to 1998. For reference, the previous study reported an<br />
average of 147.6 manatees counted per survey (Fish and<br />
Wildlife Commission unpublished), whereas the mean<br />
count in our study was 345 manatees per survey. However,<br />
due to differences in flight path and overall survey effort,<br />
comparison between the 1994<strong>–</strong>1998 counts and ours is not<br />
straightforward. Regardless, both studies showed that<br />
manatees were present in Lee County in much higher<br />
numbers during winter than non-winter months. Also, the<br />
highest counts from both studies were recorded on surveys<br />
where large aggregations of manatees were present at<br />
warm-water sites, indicating that animals traveled to Lee<br />
County specifically for this resource and/or that manatees<br />
aggregated at the plant are easier to count than smaller<br />
groups of widely dispersed manatees. Survey data from the<br />
Central region also showed that there were more manatees<br />
present during winter than non-winter. However, over
578 D. J. Semeyn et al.<br />
Fig. 3 Reference map of major<br />
speed zones within Lee County<br />
(source: Florida Fish and<br />
Wildlife Conservation<br />
Commission- Fish and Wildlife<br />
Research Institute)<br />
shorter time periods, analyses support the observation that<br />
the number of manatees in this region during pre-winter<br />
was not significantly different from winter numbers. The<br />
aggregation of manatees at a warm-water refuge or nearby<br />
seagrass beds provides an opportunity for protection of<br />
manatees during pre-winter and winter that is more difficult<br />
to achieve when the animals are dispersed during nonwinter<br />
months.<br />
Spatial analysis and management implications<br />
The figures depicting manatee densities provide a clear<br />
comparison of spatial and temporal changes in numbers and<br />
habitat use in Lee County. This information has important<br />
management implications. For example, two key months,<br />
October and November, mark the influx of tourists and<br />
seasonal residents to southwestern Florida. Density analyses<br />
showed that during these months, the distribution of manatees<br />
was highly concentrated within San Carlos Bay and Matlacha<br />
Pass.Thispre-winterdistributionismorelikethatofthe<br />
winter than the summer, suggesting that manatees may be<br />
positioning or “staging” themselves for quick access to<br />
essential warm-water habitats well before cold weather<br />
arrives. Bengtson (1981) noted a similar strategy where<br />
manatees moved closer to Blue Spring, Florida in the fall to<br />
be nearer to a warm water refuge. These large concentrations,<br />
however, may indicate more than just positioning<br />
for imminent movement to warm water. Given the fact that<br />
San Carlos Bay contains abundant seagrass beds, it is<br />
possible that manatees use this pre-winter time to feed in<br />
preparation for bouts of fasting during potentially prolonged<br />
residential periods within warm-water refuge (Rommel et al.<br />
2003). In any case, the coincidental arrival of expanded<br />
numbers of both people and manatees to specific waterways<br />
presents a significant challenge for enforcement, and<br />
limitations in the size of enforcement staff underscore the
Aerial surveys of manatees (Trichechus manatus) in Lee County, Florida 579<br />
Table 1 Total number of manatees observed on each of 34 flights,<br />
and total number of manatees observed in the Central region<br />
Survey date Survey total Central region total<br />
4/2/2007 158 24<br />
4/24/2007 <strong>17</strong>9 44<br />
5/<strong>17</strong>/2007 216 70<br />
5/29/2007 235 52<br />
6/8/2007 203 18<br />
6/28/2007 207 56<br />
7/2/2007 216 78<br />
7/9/2007 220 96<br />
8/13/2007 242 90<br />
8/20/2007 135 37<br />
9/7/2007 234 84<br />
10/8/2007 203 90<br />
11/6/2007 398 180<br />
11/20/2007 499 211<br />
12/3/2007 504 233<br />
12/18/2007 396 163<br />
1/7/2008 416 89<br />
2/19/2008 465 193<br />
4/<strong>17</strong>/2008 353 1<strong>17</strong><br />
5/13/2008 414 104<br />
5/27/2008 349 105<br />
6/6/2008 357 125<br />
7/18/2008 244 106<br />
8/28/2008 288 96<br />
10/8/2008 357 187<br />
11/5/2008 497 294<br />
12/8/2008 420 151<br />
12/18/2008 485 283<br />
1/9/2009 518 188<br />
2/27/2009 626 328<br />
3/9/2009 610 211<br />
3/30/2009 411 127<br />
4/10/2009 261 91<br />
4/29/2009 397 113<br />
Total 1<strong>17</strong>13 4434<br />
Table 2 Lee County survey study area: total number of surveys,<br />
manatees sighted, and the mean number of manatees sighted ± one<br />
standard error per survey for each specified time period<br />
Winter Non-winter<br />
# of Surveys 11 25<br />
Total # Manatees 4939 6774<br />
Mean #/Survey 494±24 282±20<br />
Table 3 Central Region: total number of surveys, manatees sighted ± one<br />
standard error, and the mean number of manatees sighted per survey for<br />
each specified time period<br />
Pre-winter Winter Post-winter Summer<br />
# of Surveys 4 10 6 14<br />
Total # of Manatees 751 2050 516 2384<br />
Mean #/Survey 188±42 205±21 86±<strong>17</strong> 80±8<br />
need to operate as efficiently as possible. Therefore, the<br />
identification of important seasonal habitats where management<br />
and enforcement efforts should be focused is an<br />
essential outcome of this study.<br />
Spatial data for the Central region also show an<br />
interesting post-winter pattern. At this time, manatees focus<br />
on feeding, but their distributions are more similar to<br />
summer patterns, which are more widely distributed. When<br />
cold weather ceases, manatees appear to disperse quickly,<br />
returning to forage in areas other than those within San<br />
Carlos Bay and Matlacha Pass. This rapid dispersal pattern<br />
has previously been noted as a contributing factor to the<br />
mass mortality of manatees and other marine animals in<br />
southwest Florida during a 1996 red-tide event (Landsberg<br />
and Steidinger 1998). This paper concluded that manatees are<br />
at high risk from February through April should a “perfect<br />
storm” of environmental factors combine to generate persistent<br />
red-tide blooms in the region. Exposure to the neurotoxins<br />
produced by the bloom-forming dinoflagellate, Karenia<br />
brevis, causes disorientation and severe health problems in<br />
manatees (O’Shea et al. 1991; Bossart et al. 1998), making<br />
them particularly vulnerable to boat strikes. While little can<br />
apparently be done to prevent manatees from encountering<br />
these blooms, increased protection against human-related<br />
threats during such events should be a priority for managers.<br />
Manatee travel patterns within seasons are also of concern<br />
to resource managers. For example, surveyors observed that<br />
on the coldest winter days most manatees were found at<br />
warm-water sites, whereas on warmer days between cold<br />
fronts many of them were found in San Carlos Bay,<br />
presumably foraging. In order to make this transition from<br />
one critical resource to another, manatees must travel a<br />
minimum of 38 km up or down the Caloosahatchee River.<br />
This is a concern for managers since the primary travel<br />
corridor for manatees coincides with the primary channel for<br />
boat traffic. Even more troublesome, a 1998 study assessing<br />
the effectiveness of speed zones at multiple sites within this<br />
river yielded an overall compliance of just 58% (Gorzelany<br />
2004). Although this study considered only a subset of Lee<br />
County waters, without evidence to the contrary it can be<br />
assumed that compliance throughout the rest of the county is<br />
similar to that of the Caloosahatchee. On a more positive<br />
note, Gorzelany (2004) also reported a significant increase in
580 D. J. Semeyn et al.<br />
Fig. 4 Kernel density analysis of distribution of manatee sightings within the entire study area for (a) non-winter (16 March<strong>–</strong>14 November) and<br />
(b) winter (15 November<strong>–</strong>15 March)<br />
compliance levels when law enforcement was present. The<br />
results from his boater compliance study imply that our<br />
communication efforts with the LCMLETF not only efficiently<br />
directed limited enforcement staff to areas of potential<br />
watercraft incidents, but also indirectly increased the<br />
likelihood of boaters obeying the posted regulations.<br />
Management in action<br />
Watercraft-related mortality within Lee County did not<br />
decrease over the course of this study. In their evaluation of<br />
the impact of boat speed restrictions on watercraft-related<br />
deaths, Laist and Shaw (2006) listed four reasons for a lack<br />
of decline in deaths state-wide after implementation of speed<br />
zones in 13 key counties: (1) manatees are unable to avoid<br />
slow-moving vessels, and therefore speed restrictions offer<br />
little protection (2) low boater compliance occurs within<br />
speed zones (3) current speed zones are not substantial<br />
enough to offer protection (4) increasing numbers of both<br />
boaters and manatees result in increase numbers of collisions,<br />
effectively outpacing the speed zone reductions of<br />
these collisions. The Laist and Shaw study revealed that<br />
implementing slow speed rules within two densely populated<br />
manatee areas substantially decreased watercraft-related<br />
deaths. Based on their findings and those from our study,<br />
which indicated that the federal and state speed zones are<br />
spatially and seasonally substantiated in Lee County, the first<br />
and third reasons for mortality not decreasing seem unlikely.<br />
We conclude that in the case of Lee County, the absence of a<br />
decline in watercraft-related mortality is most likely explained<br />
by an increase in numbers of both manatees and boaters and/or,<br />
as discussed above, low boater compliance.<br />
One of the major benefits realized from the original<br />
objective of the study was the incorporation of new spatial<br />
data into the evaluation process for proposed marine facilities.<br />
The new spatial data from our study provide managers with<br />
up-to-date information reflecting the distribution of manatees<br />
in Lee County, which is important for the siting of marine<br />
facilities. In addition, prompt, near real-time reporting resulted<br />
in excellent communication and partnership among biologists,<br />
managers and law enforcement. These efforts helped to<br />
enhance both awareness of manatee presence and protection<br />
needs and the enforcement of posted manatee zones, which<br />
will hopefully result in reduced mortality rates in the<br />
upcoming years. Ultimately, the law enforcement officers are<br />
among the most important end users of the results of our<br />
research, as they have more daily contact with the boating<br />
public than biologists or decision makers. In fiscal year 2008,<br />
four Lee County marine law enforcement agencies conducted<br />
2,665 non-citation or educational contacts (Lee County<br />
Department of Natural Resources, unpublished data). This<br />
demonstrates the tremendous potential educational impact<br />
that law enforcement has on the boating public. To take<br />
advantage of this opportunity, it is the responsibility of
Aerial surveys of manatees (Trichechus manatus) in Lee County, Florida 581<br />
Fig. 5 Kernel density analysis of the distribution of manatee sightings within the Central region for four time periods: (a) summer (01 May<strong>–</strong>30<br />
September); (b) pre-winter (01 October<strong>–</strong>14 November); (c) winter (15 November<strong>–</strong>15 March); and (d) post-winter (16 March<strong>–</strong>30 April)
582 D. J. Semeyn et al.<br />
biologists and other researchers to provide law enforcement<br />
with the most timely, accurate, and understandable<br />
information possible.<br />
Open dialogue among biologists, resource managers, and<br />
law enforcement officers is critical to reaching a balance<br />
between ensuring human access to waterways and resources<br />
contained therein and providing protection for vulnerable<br />
natural resources. This coordination is especially important in<br />
areas such as Lee County where there is high use of key<br />
habitats by both people and manatees, and where tension<br />
exists surrounding regulations aimed at protecting a species or<br />
habitat. Open dialogue fosters partnerships, mutual respect,<br />
and an understanding that is crucial for manatee conservation<br />
and protection. Lee County and the LCMLETF is a model for<br />
this type of partnership.<br />
Conclusions<br />
With minimal cost and time, this project contributed to<br />
conservation and management of manatees in Lee County.<br />
It provided:<br />
□ aerial survey data that clarified both the larger-thanexpected<br />
numbers of manatees present in Lee County<br />
waters and the habitats they prefer seasonally;<br />
□ a spatial and temporal monitoring tool for long term<br />
decision making with regard to balancing the development<br />
of coastal waterways and protection of key areas<br />
for manatees;<br />
□ real-time information on manatee numbers and locations<br />
to enhance enforcement actions;<br />
□ validation of current state regulatory zones set within<br />
the county.<br />
Possibly the largest impact was in the simplest aspect of<br />
the project: communication. Through building partnerships<br />
and interagency cooperation, studies such as this one can<br />
hope to promote conservation at a much higher level. Not<br />
only did it provide managers with information to make<br />
sound decisions when considering regulations or permits in<br />
a dynamic environment, but it allowed law enforcement to<br />
maximize limited resources and focus on the most critical<br />
areas.<br />
Resource managers benefit from the knowledge of<br />
when and where to be most vigilant when navigating the<br />
waterways of Lee County. This information should be<br />
relayed to the public through a variety of media outlets<br />
in several different mediums. Future research in any of<br />
the 13 key counties should examine distribution of<br />
manatees compared to existing regulatory zones and<br />
focus on communication between agencies and with the<br />
public to further conservation efforts of this endangered<br />
species.<br />
Acknowledgements We thank the pilots and observers that conducted<br />
the surveys including Lew Lawrence, Dr. James Powell, Greg Baker,<br />
Jorge Neumann, and Kat Frisch. We are especially grateful to the staff of<br />
Florida Power & Light Company, particularly Winifred Perkins and Jodie<br />
Gless, for their patience and support throughout the course of the study as<br />
they coordinated our surveys with FPL security staff. Special appreciation<br />
also goes to the air traffic controllers at Fort Myers Page Field and<br />
Southwest Florida <strong>International</strong> Airport. Thanks also go to the Florida<br />
Fish and Wildlife Conservation Commission Law Enforcement officers<br />
and the Lee County Marine Law Enforcement Task Force for their<br />
interest and cooperation. The authors would also like to thank Jay<br />
Sprinkel for his advice and review of the statistical analysis, Kimberly<br />
Miller, Alexis Levengood, and Adrien Miller for their help editing and<br />
formatting the manuscript, and Holly Edwards for providing information<br />
on present and previous studies conducted by the FWC in Lee County.<br />
Funding for this project was provided by the Lee County Department of<br />
Natural Resources, and airplane fuel was kindly donated by Dolphin<br />
Aviation of Sarasota, Florida.<br />
References<br />
Ackerman BB, Wright SD, Bonde RK, O’Dell DK, Banowetz DJ<br />
(1995) Trends and patterns in mortality of manatees in Florida,<br />
1974<strong>–</strong>1992. In: O’Shea TJ, Ackerman BB, Percival HF (eds)<br />
Population biology of the Florida manatee. U.S. Department of<br />
the Interior, National Biological Service, Information and<br />
Technology Report 1, pp 223<strong>–</strong>258<br />
Bengtson JL (1981) Ecology of manatees (Trichechus manatus) in the St.<br />
John’s River,Florida.Master’s Thesis, University of Minnesota<br />
Bossart GD, Baden DG, Ewing RY, Roberts B, Wright SD (1998)<br />
Brevetoxicosis in manatees (Trichechus manatus latirostris) from<br />
the 1996 epizootic: gross, histological, and immunohistochemical<br />
features. Environ Toxicol Pathol 26:276<strong>–</strong>282<br />
Craig BA, Reynolds JE III (2004) Determination of manatee<br />
population trends along the Atlantic coast of Florida using a<br />
Bayesian approach with temperature-adjusted aerial survey data.<br />
Mar Mamm Sci 20:386<strong>–</strong>400<br />
Florida Fish and Wildlife Conservation Commission, Division of Law<br />
Enforcement (2008) Boating Accidents Statistical Report pp 23<br />
http://www.myfwc.com/docs/Safety/2008_Boating_Statbook.pdf.<br />
Accessed December 2009<br />
Florida Fish and Wildlife Conservation Commission, Fish and<br />
Wildlife Research Institute (2010) Manatee Mortality Database<br />
1974<strong>–</strong>2009. http://research.myfwc.com/features/category_sub.<br />
asp?id=2241. Accessed January 2010<br />
Gorzelany JF (2004) Evaluation of boater compliance with manatee speed<br />
zones along the Gulf coast of Florida. Coast Manage 32:215<strong>–</strong>226<br />
Horne JS, Garton EO (2006) Likelihood cross-validation versus least<br />
squares cross-validation for choosing the smoothing parameter in<br />
kernel home-range analysis. J Wildl Manage 70:641<strong>–</strong>648<br />
Laist DW, Shaw C (2006) Preliminary evidence that boat speed restrictions<br />
reduce deaths of Florida manatees. Mar Mamm Sci 22:472<strong>–</strong>479<br />
Landsberg JH, Steidinger KA (1998) A historical review of Gymnodinium<br />
breve red tides implicated in mass mortalities of the manatee<br />
(Trichechus manatus latirostris) in Florida, USA. In: Reguera B,<br />
Blanco J, Fernández ML, Wyatt T (eds) Harmful algae. Xunta de<br />
Galicia and IOC of UNESCO Publishers, p 97<strong>–</strong>100<br />
Langtimm CA, Beck CA (2003) Lower survival probabilities for adult<br />
Florida manatees in years with intense coastal storms. Ecol Appl<br />
13:257<strong>–</strong>268<br />
Langtimm CA, Beck CA, Edwards HH, Fick-Child KJ, Ackerman<br />
BB, Barton SL, Hartley WC (2004) Survival estimates for<br />
Florida manatees from the photo-identification of individuals.<br />
Mar Mamm Sci 20:438<strong>–</strong>463
Aerial surveys of manatees (Trichechus manatus) in Lee County, Florida 583<br />
Lee County Marine Law Enforcement Task Force (2004) http://www.<br />
marinetaskforce.com. Accessed January 2010<br />
O’Shea TJ, Rathbun GB, Bonde RK, Buergelt CD, Odell DK (1991)<br />
An epizootic of Florida manatees associated with a dinoflagellate<br />
bloom. Mar Mamm Sci 7:165<strong>–</strong><strong>17</strong>9<br />
Reynolds JE III, Morales-Vela B, Lawler I, Edwards H (2010) Utility<br />
and design of aerial surveys for sirenians. In: Hines E, Reynolds<br />
JE III, Mignucci-Giannoni AA, Aragones LV, Marmontel M<br />
(eds) <strong>Sirenian</strong> conservation: issues and strategies in developing<br />
countries. University Press of Florida, Gainesville<br />
Rommel SA, Reynolds JE III, Lynch HA (2003) Adaptations of the<br />
herbivorous marine mammals In: Mannetje L, Ramírez-Avilás L,<br />
Sandoval-Castro C, Ku-Vera JC (eds) Proceedings of an <strong>International</strong><br />
Symposium on the Nutrition of Herbivores pp 287<strong>–</strong>306<br />
Seaman DE, Powell RA (1996) An evaluation of the accuracy of<br />
kernel density estimators for home range analysis. Ecology<br />
77:2075<strong>–</strong>2085<br />
United States Census Bureau (2009) Lee County Florida. http://quickfacts.<br />
census.gov/qfd/states/12/12071.html. Accessed November 2009<br />
United States Fish and Wildlife Service (2009) Department of the<br />
Interior. Marine Mammal Protection Act, Stock Assessment<br />
Report. http://www.fws.gov/northflorida/Manatee/SARS/<br />
20091230_frn_NOA_Manatee_Final_SARs_74_fr_69136.pdf.<br />
Accessed January 2010<br />
USA Cities Online (2009) Lee County Florida. http://www.<br />
usacitiesonline.com/flleecounty.htm Accessed December 2009<br />
Worton BJ (1995) Using Monte Carlo simulation to evaluate<br />
kernel-based home range estimators. J Wildl Manage 59:794<strong>–</strong><br />
800<br />
Wright IE, Reynolds JE III, Ackerman BB, Ward LI, Weigle BL,<br />
Szelistowski WA (2002) Trends in manatee (Trichechus<br />
manatus latirostris) counts and habitat use in Tampa Bay,<br />
1987<strong>–</strong>1994: implications for conservation. Mar Mamm Sci<br />
18:259<strong>–</strong>274
Intraspecific and geographic variation of West Indian manatee<br />
(Trichechus manatus spp.) vocalizations (L)<br />
Douglas P. Nowacek<br />
Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 and<br />
Sensory Biology and Behavior Program, Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota,<br />
Florida 34236<br />
Brandon M. Casper<br />
College of Marine Science, University of South Florida, 140 Seventh Avenue South, St. Petersburg,<br />
Florida 33701<br />
Randall S. Wells and Stephanie M. Nowacek<br />
Chicago Zoological Society, c/o Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota,<br />
Florida 34236<br />
David A. Mann a)<br />
College of Marine Science, University of South Florida, 140 Seventh Avenue South, St. Petersburg,<br />
Florida 33701 and Sensory Biology and Behavior Program, Mote Marine Laboratory,<br />
1600 Ken Thompson Parkway, Sarasota, Florida 34236<br />
Received 25 October 2003; revised 11 April 2003; accepted 14 April 2003<br />
Recordings of manatee Trichechus manatus spp. vocalizations were made in Florida and Belize to<br />
quantify both intraspecific and geographic variation. Manatee vocalizations were relatively<br />
stereotypical in that they were short tonal harmonic complexes with small frequency modulations at<br />
the beginning and end. Vocalizations ranged from almost pure tones to broader-band tones that had<br />
a raspy quality. The loudest frequency was typically the second or third harmonic, with average<br />
received levels of the peak frequency of about 100 dB re 1 Pa. Signal parameters measured from<br />
these calls showed the manatees from Belize and Florida have overlapping distributions of sound<br />
duration, peak frequency, harmonic spacing, and signal intensity, indicating no obvious<br />
distinguishing characteristics between these isolated populations. © 2003 Acoustical Society of<br />
America. DOI: 10.1121/1.1582862<br />
PACS numbers: 43.80.Ka, 43.30.Sf, 43.80.Ev WA<br />
I. INTRODUCTION<br />
Florida manatees, Trichechus manatus latirostris, are endangered,<br />
and many animals are killed or injured each year<br />
by boat strikes. At least 25% of annual documented manatee<br />
deaths are due to collisions with vessels Marine Mammal<br />
Commission, 2002. Despite the fact that behavioral evidence<br />
indicates that manatees do have the ability to detect<br />
and respond to approaching vessels Nowacek et al., 2000;<br />
Weigle et al., 1994, the animals continue to be hit. In an<br />
effort to contribute to solutions that might reduce manatee<br />
morbidity and mortality from boat strikes, we set out to build<br />
a device to alert boaters of the presence of manatees based<br />
on passive detection of their sounds.<br />
Most of the boat strike mitigation effort has been directed<br />
toward testing the hearing capabilities of manatees<br />
Gerstein et al., 1999; Ketten et al., 1992 and their behavioral<br />
response in the presence of boats Nowacek et al.,<br />
2000; Weigle et al., 1994 in hopes of understanding if and<br />
how they respond to oncoming vessels. Manatee sound production<br />
has been documented in only four papers Evans and<br />
Herald, 1970; Schevill and Watkins, 1965; Sonoda and Take-<br />
a<br />
Author to whom correspondence should be addressed. Electronic mail:<br />
dmann@marine.usf.edu<br />
mura, 1973; Sousa-Lima et al., 2002, including very little<br />
discussion of variability inherent in the animals’ vocal repertoire.<br />
These reports describe several types of sounds, but the<br />
primary vocalization recorded was a tonal sound often having<br />
several harmonics with the second or third harmonic often<br />
stronger than the fundamental frequency; intraspecific<br />
variation was not reported. The fundamental frequency<br />
ranged from 2.5 to 5 kHz for the West Indian manatee,<br />
Trichechus manatus spp., and 2.6 to 5.9 kHz for the Amazonian<br />
manatee, Trichechus inunguis.<br />
Vocalizations may be the most easily detected sounds<br />
produced by manatees compared to chewing and flatulence<br />
because published reports show them to have the highest<br />
signal-to-noise ratio and the vocalizations occur in frequency<br />
bands most dissimilar to the natural background noise of the<br />
manatees’ environment. To determine the amount of stereotypy<br />
in vocalizations we recorded sounds from Antillean<br />
manatees Trichechus manatus manatus in Southern Lagoon,<br />
Belize Lat/Lon: <strong>17</strong>° 12N, 88° 20W) and Florida<br />
manatees in Crystal River, FL Lat/Lon: 28° 53 N, 82°<br />
35W). These two groups of manatees are subspecies of the<br />
West Indian manatee, so comparing the structure of their<br />
vocalizations could add to our understanding of the similarities<br />
and differences between the two subspecies.<br />
66 J. Acoust. Soc. Am. 114 (1), July 2003 0001-4966/2003/114(1)/66/4/$19.00 © 2003 Acoustical Society of America
II. METHODS<br />
A. Belize recordings<br />
In Belize we recorded sounds on a digital archival tag,<br />
the DTAG Johnson and Tyack, 2003, which simultaneously<br />
records the attitude pitch and roll, heading, and depth of the<br />
animal with sounds produced or received by the animal. The<br />
hydrophone signal was digitized continuously at 32 kHz. Selected<br />
manatees were encircled by a seine net in shallow<br />
water and brought aboard the capture vessel for health assessment,<br />
measurements, and tagging. DTAGs were attached<br />
to three individual manatees via temporary attachments using<br />
standard manatee peduncle belts. Floats tethered to the belts<br />
carried VHF transmitters for direct radio-tracking, and in<br />
some cases a GPS receiver. Sonic tags incorporated into the<br />
FIG. 2. Power spectrum of a typical manatee vocalization. Plot shows a<br />
2048-point FFT of a 2048-point segment from the middle of the signal<br />
shown in Fig. 1a. * indicates the fundamental frequency.<br />
belts permitted underwater tracking. The tags were recovered<br />
either when the belts released via a corrodible link or animals<br />
were recaptured for recovery of the DTAG and removal of<br />
other tags and belts. Attachments lasted approximately 24 h<br />
with acoustic data recorded for 4 or 8 h depending on the tag<br />
settings used.<br />
B. Crystal River, FL recordings<br />
Recordings in Crystal River, FL were made using a hydrophone<br />
HTI 96-min; sensitivity 164 dBV/Pa, 20 Hz to<br />
32 kHz connected to a data acquisition system Tucker-<br />
Davis Technologies, RP2.1 24-bit ADC and laptop computer.<br />
Signals were acquired continuously for approximately<br />
2 h at 48 828.125 Hz sample rate from an anchored boat with<br />
the hydrophone on the bottom at 1-m depth. Recordings<br />
were made approximately 20 m away from a group of about<br />
50 manatees that were in the spring. Individuals would occasionally<br />
move out of the spring into the spring run to<br />
within 2moftheboat.<br />
C. Data analysis<br />
FIG. 1. Spectrograms of representative<br />
manatee vocalizations. a Typical<br />
tonal harmonic vocalization Crystal<br />
River, FL, b tonal vocalization transitioning<br />
to less tonal Crystal River,<br />
FL, c broader-band harmonic vocalization<br />
Crystal River, FL, and d<br />
typical tonal harmonic vocalization<br />
Southern Lagoon, Belize. The acoustic<br />
tag used to record manatees in Belize<br />
sampled at F’s32 kHz. The<br />
scale bar shows relative sound levels<br />
in decibels.<br />
Data were analyzed with MATLAB Mathworks, Inc..<br />
Peak frequency frequency with the most energy and the<br />
level of the peak frequency were determined by performing a<br />
1024-point FFT on a 1024-point segment of the middle of<br />
each call. This minimized smearing of frequencies due to<br />
changes in frequencies during the call. Harmonic spacing<br />
which is equivalent to the fundamental frequency was determined<br />
by performing an autocorrelation of the FFT, and<br />
measuring the first peak after lag zero. Duration was measured<br />
by MATLAB with an automatic detection algorithm,<br />
and then verified by hand from the spectrogram. While this<br />
may lead to a small uncertainty in measuring the time, it was<br />
advantageous for signals with low signal-to-noise ratios.<br />
J. Acoust. Soc. Am., Vol. 114, No. 1, July 2003 Nowacek et al.: Letters to the Editor<br />
67
III. RESULTS<br />
Manatees in Florida and Belize produced vocalizations<br />
that were harmonic complexes with small frequency modulations<br />
at the beginning and end Fig. 1. These ranged from<br />
almost pure tones to broader-band tones that have a noisy<br />
quality. The loudest frequency was typically the second or<br />
FIG. 3. Distribution of manatee vocalization<br />
parameters from a Crystal<br />
River, FL and b Southern Lagoon,<br />
Belize data from 11 March 2002.<br />
Panels show peak frequency, level of<br />
peak frequency, fundamental frequency<br />
harmonic spacing, and duration<br />
from top to bottom. We show only<br />
the Belize data from 11 March because<br />
this animal was in a similar behavioral<br />
situation to those in Crystal River. The<br />
animal was with a group of manatees,<br />
so these samples include sounds from<br />
other manatees as well as from the<br />
tagged animal. We felt that comparing<br />
sounds produced in similar situations<br />
provided the most accurate comparison<br />
for our study.<br />
third harmonic Figs. 2 and 3. Signal parameters measured<br />
from these calls show the manatees from Belize and Florida<br />
have overlapping distributions of sound duration, peak frequency,<br />
harmonic spacing, and signal intensity Table I and<br />
Fig. 3. Statistical comparisons between these signals were<br />
not possible because it is not known which individuals pro-<br />
68 J. Acoust. Soc. Am., Vol. 114, No. 1, July 2003 Nowacek et al.: Letters to the Editor
duced which sounds. Treating each call as a separate replicate<br />
would lead to pseudoreplication. The rate of vocalization<br />
in Crystal River, accounting for the number of animals<br />
present, was 1.29 vocalizations per minute, and in Belize it<br />
ranged from 0.09 to 0.75 per minute for the three animals<br />
tagged. When alone, the Belize animals were often silent for<br />
periods of 10 min. The rates of vocalization we measured<br />
were similar to earlier reports Bengtson and Fitzgerald,<br />
1985.<br />
IV. DISCUSSION<br />
TABLE I. Basic statistics of sounds recorded from manatees in Southern Lagoon, Belize and Crystal River, FL.<br />
The mean for each data set is shown with standard deviation in parentheses.<br />
Belize<br />
9 March<br />
2002<br />
There are no obvious differences in the vocalizations<br />
produced by manatees from Florida and Belize. For the parameters<br />
that were characterized sound duration, peak frequency,<br />
and harmonic spacing manatees from Florida and<br />
Belize had overlapping distributions. It is possible, however,<br />
that there are differences that we did not characterize.<br />
The finding that the second or third harmonic of the<br />
vocalization is usually most intense could be due either to<br />
how the sound is produced, or to propagation effects where<br />
the lower frequencies do not propagate as well in shallow<br />
water, or a combination of the two Rogers and Cox, 1988.<br />
Given that previous research on captive animals also found<br />
the fundamental to be less intense Evans and Herald, 1970;<br />
Schevill and Watkins, 1965, our results are probably best<br />
explained by the production system of the animals. Indeed,<br />
in all of these recordings we do not know the distance to the<br />
sound-producing manatee, only a range to the manatees that<br />
were in the area. The data from Belize include both the animal<br />
wearing the DTAG as well as animals vocalizing nearby.<br />
Still the received levels can be taken to show the range of<br />
levels that might be produced by manatees, and they are<br />
likely within 6<strong>–</strong>15 dB of source levels given the range of<br />
distances over which the recordings were made assuming<br />
losses of 3 dB per doubling of distance.<br />
The motivation for this study was to determine the range<br />
of natural variability in manatee sounds, so that a passive<br />
acoustic detection device can be developed to warn boaters<br />
of the presence of manatees. These data show that West Indian<br />
manatee sounds are relatively stereotypical, even be-<br />
Belize<br />
10 March<br />
2002<br />
Belize<br />
11 March<br />
2002<br />
Crystal<br />
River<br />
n 26 105 208 218<br />
Peak frequency Hz 3180 727 7080 2207 5560 2559 5223 1937<br />
Peak level<br />
dB re 1 Pa<br />
97.1 4.3 92.5 6.6 100.0 4.7 103.6 6.8<br />
Fundamental<br />
frequency Hz<br />
3180 728 4380 1618 3630 1620 2867 1059<br />
Duration s 0.032 0.0<strong>17</strong> 0.161 0.10 0.2<strong>17</strong> 0.098 0.228 0.074<br />
tween subspecies, in that they are short tonal harmonic complexes,<br />
and lend themselves readily to this application.<br />
ACKNOWLEDGMENTS<br />
We thank John E. Reynolds III, Buddy Powell, Robert<br />
Bonde, Mesha Gough, and Kevin Andrewyn for their assistance<br />
in the field and/or their consultation on this manuscript<br />
and Mark Johnson and Alex Shorter for their assistance preparing<br />
and deploying the DTAGs. This project was supported<br />
by the Florida Fish and Wildlife Conservation Commission,<br />
Wildlife Trust, the Chicago Board of Trade, and the Chicago<br />
Zoological Society.<br />
Bengtson, J. L., and Fitzgerald, S. M. 1985. ‘‘Potential role of vocalizations<br />
in West Indian manatees,’’ J. Mammal. 664, 816<strong>–</strong>819.<br />
Evans, W. E., and Herald, E. S. 1970. ‘‘Underwater calls of a captive<br />
Amazon manatee, Trichechus inunguis,’’ J. Mammal. 51, 820<strong>–</strong>823.<br />
Gerstein, E. R., Gerstein, L., Forsythe, S. E., and Blue, J. E. 1999. ‘‘The<br />
underwater audiogram of the West Indian manatee Trichechus manatus,’’<br />
J. Acoust. Soc. Am. 105, 3575<strong>–</strong>3583.<br />
Johnson, M. P., and Tyack, P. L. 2003. ‘‘A digital acoustic recording tag for<br />
measuring the response of wild marine mammals to sound,’’ IEEE J.<br />
Ocean. Eng. 281, 3<strong>–</strong>12.<br />
Ketten, D. R., Odell, D. K., and Domning, D. P. 1992. Marine Mammal<br />
Sensory Systems, edited by J. Thomas Plenum, New York, pp. 77<strong>–</strong>95.<br />
Nowacek, S. M., Wells, R. S., Nowacek, D. P., Owen, E. C. G., Speakman,<br />
T. R., and Flamm, R. O. 2000. ‘‘Manatee behavioral responses to vessel<br />
approaches,’’ Report No. FWC-00127, 2000.<br />
Rogers, P. H., and Cox, M. 1988. Sensory Biology of Aquatic Animals,<br />
edited by J. Atema, A. N. Popper, and R. R. Fay Springer-Verlag, New<br />
York.<br />
Schevill, W. E., and Watkins, W. A. 1965. ‘‘Underwater calls of Trichechus<br />
Manatee,’’ Nature London 2054969, 373<strong>–</strong>374.<br />
Sonoda, S., and Takemura, A. 1973. ‘‘Underwater sounds of the manatees,<br />
Trichechus manatus and T. inunguis Trichechidae,’’ Report of the Institute<br />
for Breeding Research, Tokyo University of Agriculture, Vol. 4, pp.<br />
19<strong>–</strong>24.<br />
Sousa-Lima, R. S., Paglia, A. P., and Da Fonseca, G. A. B. 2002. ‘‘Signature<br />
information and individual recognition in the isolation calls of Amazonian<br />
manatees, Trichechus inunguis Mammalia: Sirenia,’’ Anim. Behav.<br />
63, 301<strong>–</strong>310.<br />
U. S. Marine Mammal Commission 2002. ‘‘Annual Report of the U.S.<br />
Marine Mammal Commission,’’ U.S. Marine Mammal Commission, 4340<br />
East West Highway, Suite 905, Bethesda, MD.<br />
Weigle, B. L., Wright, I. E., and Huff, J. A. 1994. ‘‘Responses of manatees<br />
to an approaching boat: a pilot study,’’ presented at the First international<br />
manatee and dugong research conference, Gainesville, FL.<br />
J. Acoust. Soc. Am., Vol. 114, No. 1, July 2003 Nowacek et al.: Letters to the Editor<br />
69
Genetica (2011) 139:833<strong>–</strong>842<br />
DOI 10.1007/s10709-011-9583-z<br />
Evidence of two genetic clusters of manatees with low genetic<br />
diversity in Mexico and implications for their conservation<br />
Coralie Nourisson • Benjamín Morales-Vela • Janneth Padilla-Saldívar •<br />
Kimberly Pause Tucker • AnnMarie Clark • Leon David Olivera-Gómez •<br />
Robert Bonde • Peter McGuire<br />
Received: 25 February 2011 / Accepted: 18 May 2011 / Published online: <strong>17</strong> June 2011<br />
Ó Springer Science+Business Media B.V. 2011<br />
Abstract The Antillean manatee (Trichechus manatus<br />
manatus) occupies the tropical coastal waters of the<br />
Greater Antilles and Caribbean, extending from Mexico<br />
along Central and South America to Brazil. Historically,<br />
manatees were abundant in Mexico, but hunting during the<br />
pre-Columbian period, the Spanish colonization and<br />
throughout the history of Mexico, has resulted in the significantly<br />
reduced population occupying Mexico today.<br />
The genetic structure, using microsatellites, shows the<br />
presence of two populations in Mexico: the Gulf of Mexico<br />
(GMx) and Chetumal Bay (ChB) on the Caribbean coast,<br />
with a zone of admixture in between. Both populations<br />
show low genetic diversity (GMx: NA = 2.69; HE = 0.41<br />
C. Nourisson (&) B. Morales-Vela J. Padilla-Saldívar<br />
El Colegio de la Frontera Sur, Av. Centenario Km 5.5,<br />
77000 Chetumal, Quintana Roo, Mexico<br />
e-mail: coralie.nourisson@gmail.com<br />
K. P. Tucker<br />
Department of Natural Sciences, College of Coastal Georgia,<br />
3700 Altama Avenue, Brunswick, GA 31520, USA<br />
A. Clark<br />
ICBR Genetic Analysis Laboratory, University of Florida,<br />
2033 Mowry Road, Gainesville, FL 32610, USA<br />
L. D. Olivera-Gómez<br />
Division Academica de Ciencias Biologicas, Universidad Juarez<br />
Autonoma de Tabasco, Km. 0.5 carretera Villahermosa-<br />
Cárdenas, 86039 Villahermosa, Tabasco, Mexico<br />
R. Bonde<br />
US Geological Survey, 2201 NW 40th Terrace, Gainesville,<br />
FL 32605, USA<br />
P. McGuire<br />
University of Florida, 2015 SW 16th Ave, Gainesville,<br />
FL 32610, USA<br />
and ChB: NA = 3.0; HE = 0.46). The lower genetic<br />
diversity found in the GMx, the largest manatee population<br />
in Mexico, is probably due to a combination of a founder<br />
effect, as this is the northern range of the sub-species of T.<br />
m. manatus, and a bottleneck event. The greater genetic<br />
diversity observed along the Caribbean coast, which also<br />
has the smallest estimated number of individuals, is possibly<br />
due to manatees that come from the GMx and Belize.<br />
There is evidence to support limited or unidirectional gene<br />
flow between these two important areas. The analyses<br />
presented here also suggest minimal evidence of a handful<br />
of individual migrants possibly between Florida and<br />
Mexico. To address management issues we suggest considering<br />
two distinct genetic populations in Mexico, one<br />
along the Caribbean coast and one in the riverine systems<br />
connected to the GMx.<br />
Keywords Antillean manatee Microsatellite<br />
Conservation genetic Genetic structure<br />
Introduction<br />
The National Manatee Conservation Program, implemented<br />
in 2001, divided Mexico’s manatee (Trichechus<br />
manatus manatus) population into three management units<br />
(MU) (SEMARNAT 2001) (Fig. 1). These MUs were<br />
based on the geographic distribution of the manatees, the<br />
different regional threats, local research capacities, regional<br />
socio-cultural factors, and similarities in habitat characteristics<br />
including water quality and availability that has<br />
led to a protection plan for the region (SEMARNAT 2001).<br />
An understanding of the manatee population structure and<br />
diversity in Mexico, based on population genetics, will<br />
provide a valuable tool to improve the scientific<br />
123
834 Genetica (2011) 139:833<strong>–</strong>842<br />
Fig. 1 Sampling sites, with<br />
existing manatee management<br />
units recognized in Mexico.<br />
Dots represent specific sampling<br />
sites<br />
information lending support for management actions, such<br />
as the new National Manatee Conservation Action Plan for<br />
Mexico (PACE manati). That plan is expected to be<br />
adopted this year (2011). In Mexico, manatees have been<br />
protected since 1922 (Diario Oficial 01/20/1922), listed as<br />
a threatened species in 1991 under Federal Law (SEDUE<br />
1991) and since 1994 has been classified as a species at risk<br />
of extinction by the Mexican Government (SEMARNAT<br />
2002). At the international level, Antillean manatees were<br />
classified as vulnerable by the <strong>International</strong> Union for<br />
Conservation of Nature (IUCN) Red List from 1982 to<br />
2007 and endangered since 2008 (Deutsch et al. 2008).<br />
West Indian manatees have also been listed as an endangered<br />
species by the Convention on <strong>International</strong> Trade in<br />
Endangered Species (CITES 2009) since 1975.<br />
The Antillean manatee occupies the tropical coastal<br />
waters of the Greater Antilles and Caribbean, extending from<br />
Mexico along Central and South America to Brazil (Husar<br />
1978; Lefebvre et al. 1989, 2001). The distribution of manatees<br />
in Mexico includes the coastal and wetland systems of<br />
the Gulf of Mexico (GMx) and the Caribbean coast (Fig. 1).<br />
The majority of manatees inhabit the wetland systems of<br />
Veracruz, Tabasco, Chiapas and Campeche states along the<br />
GMx, where a conservative estimate of population size is<br />
between 500 and 1,500 manatees (Olivera-Gómez 2006). It<br />
is estimated that approximately 200 to 250 manatees reside<br />
in Quintana Roo region (Morales-Vela and Padilla-Saldívar<br />
123<br />
2011) and much fewer reside within the Yucatan state<br />
(Morales-Vela et al. 2003). In Quintana Roo, most manatees<br />
occur in Chetumal Bay (ChB) with an estimate of about 100<br />
to 150 manatees (Morales-Vela and Padilla-Saldivar 2009a).<br />
In Ascencion Bay (AB) the estimate is smaller, with<br />
approximately 25 manatees counted during a single aerial<br />
survey in July of 2009 (Landero-Figueroa 2010). ChB and<br />
AB are located approximately 335 km apart (shore distance).<br />
North of the Yucatan Peninsula no estimates have<br />
been provided and manatee abundance is presumed to be<br />
very low (Morales-Vela et al. 2003).<br />
Previous genetic studies on the West Indian manatee<br />
suggest that their mitochondrial DNA (mtDNA) variability<br />
ranges between one (Florida) to as many as eight haplotypes<br />
(Colombia) per country (Garcia-Rodriguez et al.<br />
1998; Vianna et al. 2006). Genetic studies of manatees in<br />
Mexico, from Tabasco to Quintana Roo indicate differences<br />
among the populations. The biggest difference is<br />
between the GMx, where only one haplotype (J) is found,<br />
and the Caribbean coast where three haplotypes (J, A, A4)<br />
have been identified (Medrano-González et al. 1997;<br />
Castañeda-Sortibrán unpublished data). However, the small<br />
sample size from the GMx (n = 4) at the time of this<br />
mtDNA study was not sufficient to provide significant<br />
results.<br />
Historically, manatees were abundant in Mexico from<br />
Tamaulipas to the Yucatan Peninsula, but hunting for food
Genetica (2011) 139:833<strong>–</strong>842 835<br />
during the pre-Columbian period by Mayans (McKillop<br />
1985), the Spanish colonization and throughout the history<br />
of Mexico (Durand 1983), has resulted in the significantly<br />
reduced population occupying Mexico today. Even as<br />
recently as 1983, it was reported that manatees were occasionally<br />
harpooned in the region north of Quintana Roo as a<br />
source of food (Gallo-Reynoso 1983). Direct hunting is rare<br />
now in the GMx, but manatees caught in nets are still<br />
sometimes killed for food, with children reporting that they<br />
had eaten fresh manatee meat recently (Olivera-Gómez<br />
2006). Fishermen have also declared that the manatee<br />
population decline in the Northern and Western coasts of<br />
the Yucatan Peninsula was caused by a number of factors,<br />
including hunting for local consumption, entanglement as a<br />
result of higher net-fishing activities in rivers, and habitat<br />
destruction and coastal construction due to human population<br />
growth and hurricane impacts (Morales-Vela et al.<br />
2003). The last documented manatee hunting activities on<br />
the Northeast coast of the Yucatan Peninsula occurred in the<br />
1960<strong>–</strong>1970s, but since then opportunistic poaching has<br />
continued into the 1990s (Morales-Vela et al. 2003). In the<br />
GMx other factors such as natural or artificial closing of<br />
drainages from freshwater ecosystems have restricted<br />
manatee access to vegetation, increased water temperature,<br />
and have resulted in increases in pollution, sedimentation<br />
and eutrophication due to agricultural and poultry runoff<br />
(Olivera-Gómez 2006).<br />
The high level of anthropogenic and habitat destruction<br />
pressure on the manatee population may have caused a<br />
population bottleneck, which is known to reduce genetic<br />
diversity. Typically, habitat loss and degradation increase<br />
the rate of fragmentation resulting in the isolation of small<br />
populations and leads to an increased probability for<br />
inbreeding and unstable demographics (Frankham et al.<br />
2002).<br />
Herein, we present a fine-scale population structure of<br />
manatees in Mexico using analysis of microsatellite DNA<br />
markers. The genetic diversity and population structure<br />
were compared to a geographically close conspecific, the<br />
Florida manatee (Trichechus manatus latirostris), which<br />
has a larger estimated population size. The results of this<br />
microsatellite DNA marker study were compared with<br />
those obtained previously from an unpublished Mexican<br />
mtDNA study.<br />
Methods<br />
Microsatellite DNA amplification and fragment<br />
analysis<br />
We analyzed 94 samples from different regions of Mexico<br />
including the Peninsula of Yucatan MU (ChB: 51, AB: 15),<br />
the central GMx MU (Tabasco: 15, Chiapas: 5 and Campeche:<br />
1) and the north of the GMx MU (Veracruz: 7)<br />
(Fig. 1). Blood or skin tissue from the tail was collected<br />
from wild manatees captured for health assessment and<br />
radio tagging studies. Skin tissue was collected from carcasses<br />
recovered by manatee research projects throughout<br />
Mexico. Blood from captive manatees was also utilized for<br />
this study because their original rescue location was<br />
known. To resolve population structure, we included the<br />
genotypes of 95 individuals from Florida based on the<br />
estimated population size in each of the four recognized<br />
MU’s in that state (Pause 2007).<br />
Blood and tissue samples were preserved with lysis or<br />
tissue buffer respectively (lysis buffer: 100 mM Tris<strong>–</strong>HCl,<br />
100 mM EDTA, 10 mM NaCl, 1.0% SDS (White and<br />
Densmore 1992); SED tissue buffer: saturated NaCl;<br />
250 mM EDTA pH 7.5; 20% DMSO (Amos and Hoelzel<br />
1991; Proebstel et al. 1993)). DNA extractions, amplifications<br />
and fragment analysis were performed at the University<br />
of Florida, ICBR Genetic Analysis Laboratory in<br />
Gainesville, Florida, USA and at the US Geological Survey,<br />
Southeast Ecological Science Center Conservation<br />
Genetics Laboratory in Gainesville, FL, USA.<br />
DNA extractions were carried out using either a standard<br />
phenol<strong>–</strong>chloroform protocol (Hillis et al. 1990) or the<br />
DNeasy tissue extraction kit (QIAGEN, Valencia, CA, USA).<br />
Polymerase chain reaction (PCR) amplifications were<br />
performed for each sample for each of 13 microsatellite<br />
loci previously designed for manatees: TmaA02, TmaE02,<br />
TmaE08, TmaE11, TmaE26, TmaF14, TmaM79 (Garcia-<br />
Rodriguez et al. 2000), TmaSC13, TmaE7, TmaH13,<br />
TmaE14, TmaK01, TmaJ02 (Pause et al. 2007). Amplifications<br />
were performed in Biometra UNOII, UNO-<br />
Thermoblock, T-Gradient thermocyclers (BiometraÒ,<br />
Göttengen, Germany) or a PTC-200 (MJ Research, Waltham,<br />
MA) thermocycler using the following conditions:<br />
95°C for 5 min, 35 cycles of 95°C for 30 s, with the specific<br />
annealing temperature as listed in the original publication<br />
for each primer, with the exception of TmaM79 =<br />
54°C; TmaA02 = 56°C; TmaE02, TmaE11, TmaE26 and<br />
TmaF14 = 58°C; TmaE08 = 60°C for 30 s, 72°C for<br />
30 s, and a final extension at 72°C for 10 min. Amplifications<br />
were performed in a total volume of 15 lL, with<br />
10 ng target DNA, 19 Sigma PCR Buffer (10 mM Tris<strong>–</strong><br />
HCl, pH 8.3, 50 mM KCl, 0.001% gelatin), MgCl2 as<br />
indicated in the original publications, 0.2 mM each dNTP,<br />
0.04 units of Sigma JumpStart Taq polymerase (Sigma<strong>–</strong><br />
Aldrich, St. Louis, MO, USA), 0.25 lM each primer, and<br />
bovine serum albumin (BSA), where indicated (Garcia-<br />
Rodriguez et al. 2000; Pause et al. 2007). For fragment<br />
analysis, the forward primers were labeled with the fluorescent<br />
dyes HEX or 6-FAM for processing and visualization<br />
on an ABI 3730xl Automated DNA Analyzer.<br />
123
836 Genetica (2011) 139:833<strong>–</strong>842<br />
Fragment data from the PCR products were collected from<br />
the ABI 3730xl and analyzed using GENEMARKER 1.5<br />
(SoftGenetics 2008) to determine allele sizes. Allele sizes<br />
were standardized using previously analyzed Florida samples<br />
as the baseline.<br />
Data analysis<br />
CONVERT (Glaubitz 2004) was used to convert the data into<br />
different input file formats. We used STRUCTURE, version<br />
2.3.1, (Pritchard et al. 2000) to identify possible subpopulation<br />
designations, without an a priori assignment of the<br />
overall population structure in Mexico. The data from<br />
Mexico and Florida were analyzed together. The admixture<br />
model was used and the number of populations (K) was set<br />
from 1 to 10 with a burn-in period of 100,000 iterations,<br />
followed by 1,000,000 Monte Carlo Markov Chain iterations.<br />
Five independent analyses were simulated for each<br />
value of K. The value of K with the lowest posterior<br />
probability was identified as the optimum number of subpopulations,<br />
as recommended by the STRUCTURE manual.<br />
GENECLASS2 (Piry 2004) and WHICHRUN version 4.1 (Banks<br />
and Eichert 2000) were used to test individuals attributed to<br />
a different population than the geographic location<br />
assigned by STRUCTURE. GENALEX 6.2 (Peakall and Smouse<br />
2006) was used to calculate genetic distance between<br />
individuals which were used on a Principal Coordinate<br />
Analysis (PCA).<br />
Estimates of the effective population size were made by<br />
NEESTIMATOR using the Linkage Disequilibrium algorithm<br />
(Peel et al. 2004). GENALEX 6.2, GENEPOP 3.4 (Raymond<br />
and Rousset 1995; Rousset 2008) and ARLEQUIN 3.1 (Excoffier<br />
et al. 2005) were used to compare and determine the<br />
genetic diversity and genetic differentiation including<br />
allelic richness N A and N E, heterozygosity observed (HO)<br />
and expected (HE), estimate of population subdivision FST<br />
and RST and inbreeding coefficients (FIS). ARLEQUIN was<br />
used to check for deviation from Hardy<strong>–</strong>Weinberg equilibrium.<br />
GENEPOP webversion was used to examine for<br />
Linkage Disequilibrium. The presence of null alleles was<br />
analyzed with MICRO-CHECKER (Oosterhout et al. 2004). The<br />
presence of a potential bottleneck was estimated using<br />
BOTTLENECK examining the heterozygosity excess (Cornuet<br />
and Luikart 1996).<br />
Results<br />
Results from the three different software packages (GEN-<br />
ALEX, GENEPOP, and ARLEQUIN) were compared and similar<br />
results were observed for genetic diversity and population<br />
structure as had been previously published. Therefore, only<br />
the results from GENALEX are presented in Tables 1 and 3.<br />
123<br />
Population structure<br />
Results for the STRUCTURE analysis identified that k = 3<br />
was the appropriate number of clusters (Fig. 2) based on<br />
the manual’s recommendation of the lowest posterior<br />
probability LnP(D). Most individuals were assigned to a<br />
cluster with Q [ 80%. In all simulations, Florida manatees<br />
were assigned to a separate population cluster (92.8%<br />
assignment) and the GMx and ChB populations were<br />
assigned to a cluster that corresponded to their geography<br />
(89.0 and 84.6% respectively). Samples collected from AB<br />
were not as clearly delineated, and were clustered with<br />
either the GMx (56.1%) or ChB (41.6%).<br />
The STRUCTURE analysis suggests that there were some<br />
individuals that may have had mixed ancestry based on this<br />
analysis. Interestingly, some of the AB individuals with a<br />
high percentage of ancestry corresponding to the GMx<br />
cluster do not share the GMx’s unique haplotype. Instead,<br />
these individuals’ haplotypes corresponded to the AB/ChB<br />
regions. Four manatees sampled from ChB show high<br />
percent ancestry with the Florida cluster (between 59% and<br />
76%). One of them was a female manatee carcass sampled<br />
in ChB which shared a 59% ancestry with the Florida<br />
population. This manatee died of natural causes (birthing<br />
complications) and her calf was not recovered. Another<br />
was a female juvenile who shared 75.6% ancestry with<br />
Florida but her haplotype did not correspond with the<br />
Florida haplotype (Castañeda-Sortibrán pers. comm). One<br />
male manatee, who was rescued as a calf in March 2002<br />
and is now in captivity in Veracruz, also shares a high<br />
percent ancestry with Florida samples (51%). One female<br />
manatee from Veracruz shares ancestry with the GMx<br />
samples (44%) and ChB region samples (47%). Two<br />
manatees from Florida share a high percent ancestry (45<br />
and 60%) with the GMx individuals. Results from GENE-<br />
CLASS2 and WHICHRUN analysis corroborate these assignments<br />
to populations other than their geographic location<br />
(Table 1). The Veracruz samples have lower ancestry<br />
values and cannot be totally attributed to Florida or the<br />
GMx as they appear to be mixtures of both populations on<br />
GENECLASS2 and WHICHRUN analysis (Table 1).<br />
The pairwise FST value between ChB and AB is low but<br />
significant (FST = 0.047), and RST is not significant<br />
(RST = 0.016) (Table 2). All pairwise FST and RST values<br />
are presented in Table 2. Significant levels of subdivision<br />
were observed between the GMx and all other regions with<br />
FST and RST, and among all regions with FST. All values<br />
were significant when calculated between the three regions:<br />
the Caribbean coast (including ChB and AB), the GMx and<br />
Florida (data not show) but most of RST values were not<br />
significant when calculated by separating the Caribbean<br />
coast into two areas: ChB and AB and compared with the<br />
two other geographic regions: the GMx and Florida
Genetica (2011) 139:833<strong>–</strong>842 837<br />
Table 1 Population assignments of some individuals using STRUCTURE, GENECLASS2 and WHICHRUN<br />
Individual Sampling<br />
population<br />
STRUCTURE<br />
assignment<br />
(Table 2). A higher level of differentiation (via FST) was<br />
observed between ChB and the GMx than between ChB<br />
and Florida. This does not correspond to the STRUCTURE<br />
analysis, which consistently grouped Florida into its own<br />
cluster, separate from ChB. A handful of individuals had<br />
microsatellite genotypes that clustered with Florida, but<br />
overall, the group is distinct from Florida. This can also be<br />
STRUCTURE Q value GENECLASS2 results WHICHRUN<br />
BCh GMx FL Most Score Other Score Other Score<br />
Attribution<br />
probable for most population for population pop3<br />
population probable<br />
population<br />
(pop2) pop2 (pop3)<br />
BCH40 BCH BCh 0.98 0.01 0.02 BCH 100 FL 0 GMx 0 BCh<br />
BCH13 BCH BCh 0.97 0.02 0.01 BCH 100 GMx 0 FL 0 BCh<br />
BCH46 BCH FL 0.15 0.09 0.76 FL 81.1 BCH 15.44 GMx 3.46 FL<br />
BCH47 BCH FL 0.23 0.02 0.75 FL 85.32 BCH 14.35 GMx 0.33 FL<br />
BCH02 BCH FL 0.22 0.02 0.76 FL 85.16 BCH 14.69 GMx 0.14 FL<br />
N2 BCH FL 0.31 0.1 0.59 FL 76.96 BCH 15.55 GMx 7.49 FL<br />
TOO2 GMx GMx 0.01 0.98 0.01 GMx 99.96 BCH 0.05 FL 0 GMx<br />
V2 GMx GMx 0.01 0.98 0.01 GMx 99.73 BCH 0.27 FL 0 GMx<br />
V3 GMx BCh 0.47 0.43 0.1 GMx 90.13 BCH 9.86 FL 0.01 GMx<br />
V1 GMx FL 0.04 0.46 0.51 GMx 51.33 FL 47.97 BCH 0.7 GMx<br />
TM1007 FL FL 0.01 0.01 0.98 FL 100 GMx 0 BCH 0 FL<br />
TM772 FL FL 0.01 0.01 0.98 FL 100 BCH 0 GMx 0 FL<br />
TM775 FL FL 0.06 0.<strong>17</strong> 0.77 FL 79 GMx 18.55 BCH 2.45 FL<br />
TM925 FL FL 0.28 0.04 0.69 FL 76.06 BCH 23.35 GMx 0.59 FL<br />
TM523 FL FL 0.<strong>17</strong> 0.24 0.6 FL 78.25 GMx 15.36 BCH 6.39 GMx (similar<br />
probability<br />
for each<br />
population)<br />
TM636 FL FL 0.02 0.24 0.73 FL 86.96 GMx 12.14 BCH 0.9 FL<br />
TM993 FL GMx 0.05 0.64 0.31 GMx 57.76 FL 41.08 BCH 1.16 GMx<br />
TM641 FL FL 0.1 0.42 0.48 FL 63.68 GMx 27.06 BCH 9.26 FL<br />
Values generated by STRUCTURE in italics are individuals that have higher ancestry percentage associations when compared to other populations,<br />
relative to where the sample was collected. Underlined results indicate higher association to a population other than their original sampling<br />
location population. Results not in italics indicate two samples assigned to their geographic population for comparison<br />
Fig. 2 Proportions of ancestry for individuals were assessed without<br />
a priori information using Bayesian clustering via STRUCTURE. This<br />
graphic represents the best fit of the data, where three population<br />
clusters are clearly distinguished (K = 3). Manatees from Florida<br />
consistently grouped separately from the manatees from Mexico.<br />
Manatees from the Chetumal Bay (ChB) cluster together and those<br />
from the Gulf of Mexico (GMx) river systems clustered together.<br />
Samples from Ascencion Bay (AB) suggest it may be a mixing zone<br />
for the Mexican manatee populations<br />
observed in the PCA from genetic distance that shows<br />
separate clusters for Florida and Mexico (Fig. 3). The null<br />
hypothesis tested for heterozygosity excess using the<br />
Wilcoxon test, with BOTTLENECK software, provided<br />
(P = 0.0004 and P = 0.07) probabilities under Two-Phase<br />
mutation (TPM) and Step-wise mutation (SMM) respectively,<br />
with a normal L-shaped distribution for ChB<br />
123
838 Genetica (2011) 139:833<strong>–</strong>842<br />
Table 2 Pairwise F ST and R ST values comparing the Mexican manatee<br />
populations defined as Chetumal Bay (ChB), Ascencion Bay<br />
(AB) and the Gulf of Mexico (GMx), as well as the Florida manatee<br />
population, generated by GenAlEx 6.2<br />
ChB AB GMx Florida<br />
ChB <strong>–</strong> 0.016 0.114* 0.025<br />
AB 0.047* <strong>–</strong> 0.087* 0.014<br />
GMx 0.131* 0.088* <strong>–</strong> 0.089*<br />
Florida 0.096* 0.094* 0.106* <strong>–</strong><br />
* Significant values<br />
The FST values are presented below the diagonal, and RST values are<br />
above diagonal<br />
Fig. 3 PCA study from genetic distance using GENALEX. On the left<br />
is the Mexican population with individuals from the Caribbean and<br />
the Gulf of Mexico coasts. An admixture zone is apparent between<br />
these two areas. The population on the right corresponds to the<br />
Florida population<br />
population. For the GMx and AB populations the Wilcoxon<br />
test indicates a probability of 0.40 (TPM) and 0.<strong>17</strong> (SMM)<br />
with a shifted mode distribution for the GMx and 0.046<br />
(TPM) and 0.10 (SMM) with a shifted mode distribution<br />
for AB. For Florida, the Wilcoxon test shows a probability<br />
of 0.02 (TPM) and 0.55 (SMM) with a normal L-shaped<br />
distribution.<br />
Genetic diversity<br />
No evidence of null alleles was identified in either the<br />
Mexico or Florida populations. After 78 comparisons and a<br />
sequential Bonferroni correction, no linkage disequilibrium<br />
was observed (overall a = 0.05). All loci were in Hardy<strong>–</strong><br />
Weinberg equilibrium (HWE) after a sequential Bonferroni<br />
correction for all Mexican populations but TmaE02 was<br />
not within HWE for Florida. The error rate was determined<br />
by re-genotyping 12% of the samples. No errors were<br />
detected. Estimates of the mean number of alleles, effective<br />
number of alleles, and heterozygosity expected and<br />
observed, FIS, estimated population size and effective size<br />
for each determinate population (ChB, AB, the GMx and<br />
Florida) are presented in Table 3. Private alleles were<br />
found in Florida (n = 16), ChB (n = 2) and AB (n = 4).<br />
Discussion<br />
The GMx, with extended riverine and lagoon systems in<br />
Tabasco, Veracruz, Chiapas and Campeche, is a very<br />
favourable habitat for manatees. It has abundant vegetation,<br />
fresh water and areas where manatees can evade<br />
hunters. The Caribbean coast also represents very suitable<br />
habitat, especially ChB and AB, which are protected areas.<br />
ChB is the largest marine protected area at the state level in<br />
Mexico and is shared with Belize. AB is part of the Biosphere<br />
Reserve of Sian Ka’án, one of the largest Federal<br />
marine reserves in Mexico.<br />
Table 3 Diversity statistics over the 13 microsatellite loci examined for manatees from Mexico and Florida<br />
Population ChB AB GMx Florida<br />
Number of samples 51 15 28 95<br />
NA 3.00 ± 0.32 3.00 ± 0.32 2.62 ± 0.24 3.62 ± 0.48<br />
NE 1.99 ± 0.<strong>17</strong> 2.00 ± 0.22 1.84 ± 0.<strong>17</strong> 2.04 ± 0.16<br />
HE 0.46 ± 0.04 0.43 ± 0.05 0.41 ± 0.05 0.47 ± 0.04<br />
HO 0.47 ± 0.05 0.45 ± 0.07 0.44 ± 0.05 0.47 ± 0.04<br />
FIS -0.059 -0.035 -0.056 -0.006<br />
LD Abs Abs Abs Abs<br />
Null alleles Abs Abs Abs Abs<br />
Estimated population 100<strong>–</strong>150 23 (aerial survey) 500<strong>–</strong>1,500 5,000<br />
Effective population size 32.7 (23.9<strong>–</strong>47.4) 4.9 (4.0<strong>–</strong>6.2) 27.6 (18.0<strong>–</strong>49.4) 424.3 (208.3<strong>–</strong>6,816.5)<br />
The mean number of alleles (N A), effective number of alleles (N E), observed and expected heterozygosity (H O and H E, respectively), inbreeding<br />
coefficient (F IS), linkage disequilibrium after sequential Bonferroni correction (LD), null alleles, estimated population size, and effective<br />
population size for each population examined: Chetumal Bay (ChB), Ascencion Bay (AB), the Gulf of Mexico (GMx) and Florida for<br />
comparison<br />
123
Genetica (2011) 139:833<strong>–</strong>842 839<br />
Manatees are known to travel long distances along<br />
coastlines with suitable habitat (e.g. a Florida manatee was<br />
observed as far north as Rhode Island), as well as infrequent<br />
travel across open ocean (Alvarez-Alemán et al.<br />
2010; Deutsch et al. 2003; Fertl et al. 2005; Reep and<br />
Bonde 2006). Using existing tools, genetic analyses of<br />
Florida manatees have suggested little population structure<br />
throughout Florida (McClenaghan and O’Shea 1988; Garcia-Rodriguez<br />
et al. 1998; Pause pers. com.). Although<br />
manatees are capable of travelling along the shoreline<br />
between the GMx and the Caribbean coast, seasonal<br />
migration in response to cold weather is not necessary for<br />
survival in Mexico as it is in Florida.<br />
Radio tagging studies in ChB illustrated movement by<br />
five males and one female along the shoreline between<br />
discrete manatee habitats in ChB and Belize (Morales-Vela<br />
et al. 2007). One of these males was also observed participating<br />
in a mating group in Belize (Auil-Gomez pers.<br />
comm). The movement pattern from ChB to Belize was very<br />
similar as manatees made directed, continuous moves along<br />
the shoreline between discrete manatee habitats (Morales-<br />
Vela et al. 2007). None of the 19 manatees tagged with<br />
Argos-linked GPS radio tags, either from ChB or AB<br />
showed movements north of the initial tagging site (Morales-Vela<br />
and Padilla-Saldívar 2009b). The natural morphological<br />
features of ChB, oriented in the south of Mexico,<br />
easily connect ChB with both the Caribbean Sea and coastal<br />
Belize. An extensive coral reef barrier forms a natural<br />
marine corridor that may promote easier movement of the<br />
manatees between ChB to the south along the coast of<br />
Belize, than to the north of the Yucatan Peninsula. The<br />
Caribbean group is represented by ChB and AB, with a<br />
significant but low FST between these two locations. AB<br />
appears as a mixture of the southern portion of the Caribbean<br />
coast (ChB) and the GMx manatees, which is why the F ST<br />
between ChB and AB was significant but low (Fig. 2). It<br />
seems that migrants from the GMx breed in AB and do not<br />
continue south towards ChB. Mothers and calves are often<br />
observed in AB, which hosts suitable and abundant habitat<br />
for manatees. The mixture found in AB with the presence of<br />
individuals with both ancestral roots was not observed in the<br />
GMx region. The higher genetic diversity found in the<br />
Caribbean coast when compared to the GMx are consistent<br />
with results from the mtDNA data. Analyses of mtDNA<br />
analysis identified three haplotypes along the Caribbean<br />
coast, one common to Florida, another in common with<br />
Belize and the last shared a haplotype unique to the GMx<br />
(Castañeda-Sortibrán unpublished data). The Caribbean<br />
Cluster is geographically closer to Florida (approximately<br />
800 km) than to the GMx (approximately 1,150 km).<br />
A high level of genetic diversity in a population is<br />
thought to increase the probability of surviving a catastrophic<br />
event (Frankham et al. 2002). In this study, the<br />
mean number of observed alleles was very low and corresponds<br />
to less than what is predicted for wildlife populations<br />
that have been hunted or have been significantly fragmented<br />
(DiBattista 2008). However, these data describing low<br />
variability are consistent with other West Indian manatee<br />
populations throughout their range (Hunter et al. 2010,<br />
Pause pers. comm.). The lower genetic diversity, based on<br />
the number of alleles and heterozygosity found in the GMx,<br />
is probably due to a combination of a founder effect and a<br />
bottleneck event. This is likely due to manatees occupying<br />
the northern range of the subspecies and the extensive historical<br />
exploitation experienced during and since the pre-<br />
Columbian period up to the 1960<strong>–</strong>1970s. That depletion of<br />
manatee stocks resulted in the reduced population numbers<br />
observed today (Durand 1983; McKillop 1985; Medrano-<br />
González et al. 1997). The population with the greatest<br />
genetic diversity observed along the Caribbean coast also<br />
has the smallest estimated number of individuals (Table 3).<br />
This is possibly due to manatees that come from the GMx<br />
and Belize. Some breeding contribution of manatees from<br />
Belize is suggested by recent radio tagging data (Morales-<br />
Vela and Padilla-Saldívar 2009b) and by the presence of<br />
shared mtDNA haplotypes (Vianna et al. 2006). The proximity<br />
of the large Belize manatee population, only<br />
200<strong>–</strong>300 km south of ChB, may explain the dispersal route<br />
of the Caribbean coast manatees. Having a large population<br />
close by with an easy access route, makes it easier for the<br />
manatees to travel south than to try to make the longer<br />
journey to the GMx. There is evidence from radio tagging<br />
studies where some male manatees move from ChB to south<br />
Belize and return (Morales-Vela et al. 2007). However<br />
mtDNA shows different matrilineal trends in haplotype<br />
dispersal patterns in Belize and Mexico (Vianna et al. 2006)<br />
indicating that ChB and the Belize populations are not one<br />
well mixed population but have persisted independently<br />
over time. The genetic diversity along the Caribbean coast<br />
may have been recently influenced by potential migration<br />
from the growing manatee population in Florida as suggested<br />
by the few individuals that share remote ancestry<br />
with the Florida population.<br />
The Belize population shows a stronger separation from<br />
Florida (FST = 0.141) (Hunter et al. 2010) than the Mexican<br />
populations, with FST = 0.096 between the Mexican<br />
Caribbean coast and Florida and F ST = 0.106 between the<br />
GMx population and Florida. Belize and the Caribbean<br />
coast show a different pattern of haplotype distribution<br />
with two haplotypes in common and one private haplotype.<br />
The Florida haplotype, which is absent in Belize, yet is<br />
present in Mexico, confirms the closer relationship between<br />
Florida and Mexico than between Florida and Belize as<br />
confirmed by microsatellite data.<br />
Current collaborative studies utilizing samples from<br />
Mexico, Belize, Puerto Rico, Florida and other Caribbean<br />
123
840 Genetica (2011) 139:833<strong>–</strong>842<br />
countries will be completed in 2011<strong>–</strong><strong>2012</strong> to further assess<br />
the phylogeography and dispersal patterns of West Indian<br />
manatees. The effective population size is lower than the<br />
estimated population size for manatees in Mexico, which<br />
may be a reflection of the reduction in population size due<br />
to prior hunting pressure. The low Ne of the manatee<br />
population in the GMx may be due to a founder effect<br />
resulting in low genetic diversity, recent bottleneck events,<br />
and migratory limitations due to climate change and glacial<br />
periods. This could also be an artefact of the small sample<br />
size for the GMx. More samples are urgently needed from<br />
this area.<br />
Manatees in the GMx population tend to stay within the<br />
inland river systems, not in the coastal waters, and the<br />
group is likely fragmented by urban and agricultural<br />
development. An area of 140 km in the northeastern region<br />
of the Yucatan Peninsula is unsuitable habitat for manatees<br />
due to urban development, and the lack of vegetation and<br />
fresh water sources, resulting in a likely barrier to gene<br />
flow between the GMx and Caribbean coast (Medrano-<br />
González et al. 1997; Morales-Vela et al. 2003). Close to<br />
this zone, there are also significant coastal fishing activities<br />
and poaching (Morales-Vela et al. 2003). Examination of<br />
oceanographic structures shows a differentiation between<br />
the GMx and the Yucatan Peninsula (Laura Carrillo, pers.<br />
comm.; Merino 1997). This barrier may explain the low<br />
gene flow between the GMx and the Caribbean coast.<br />
Nevertheless, a group of six large manatees were observed<br />
in 2006 along the northeast coast of the Yucatan Peninsula<br />
(Reyes-Mendoza and Morales-Vela 2007). It is not known<br />
if these manatees were transient or resident.<br />
The analyses presented here suggest minimal evidence<br />
of a handful of individual migrants possibly between<br />
Florida and Mexico. One Florida manatee and six potential<br />
second generation migrants from Florida were detected in<br />
Mexico, which raises the question of whether migration or<br />
breeding occurs between the two subspecies located in<br />
Florida and Mexico. Two manatees from Florida share a<br />
high percent ancestry (45 and 60%) with the GMx individuals<br />
suggesting individuals may be second generation<br />
migrants. One of these manatees was a female with a<br />
known record of Florida maternal lineage from photoidentification.<br />
She was born in the Naples, FL area in 1995<br />
and captured in Port of the Islands, FL in February 2001.<br />
The other was the carcass of a male calf found in August<br />
1997 in the Banana River on the east coast of Florida<br />
(Beck, USGS, pers. comm.). One possibility is that<br />
migrants could be from Cuban waters; however, no genetic<br />
data are currently available from manatees in Cuba. Major<br />
currents go from Mexico to Florida and travel close to<br />
Cuba. Events such as tropical storms and hurricanes can<br />
change the currents and can separate manatees from the<br />
coast, taking them offshore and subject to drift (Langtimm<br />
123<br />
and Beck 2003). In that way, currents may also allow<br />
manatees to travel to Mexico from Florida. Manatees are<br />
able to survive short open water travels, and adults have no<br />
major natural predators so they will be able to survive<br />
incidental open sea travel. For example a known long-term<br />
resident female from the Florida manatee photo-identification<br />
catalog was sighted in Cuba in 2007 (Alvarez-Alemán<br />
et al. 2010). This manatee successfully travelled<br />
across open sea illustrating potential for population<br />
expansion. Reports of Florida manatees in the northern<br />
GMx indicate that they also can travel as far west as the<br />
southern Texas coast following the shoreline (Laguna<br />
Madre and Rio Grande) (Fertl et al. 2005). Historical distribution<br />
of manatees in Tamaulipas, Mexico which is<br />
north of the GMx shows a possible link that may exist<br />
between the GMx and Florida for manatees to travel<br />
(Lazcano-Barrero and Packard 1989). This leads to the<br />
question of successful reproduction between the two subspecies<br />
(T. m. manatus and T. m. latirostris) and the possible<br />
positive (increased variability, heterosis) or negative<br />
consequences (outbreeding depression) of breeding.<br />
Conclusion/conservation application<br />
The genetic structure of manatees in Mexico indicates two<br />
clusters with gene flow from the GMx to the Caribbean<br />
coast with no migration from the Caribbean back towards<br />
the GMx. This movement pattern could not be detected<br />
using mitochondrial DNA as there is only one haplotype<br />
(J) present in the GMx which is also present in the<br />
Caribbean. The three haplotypes found in the Caribbean are<br />
also found in the GMx, Florida and Belize. The individuals<br />
from AB appear as a mixture of the populations from ChB<br />
and the GMx. Additional evidence should be collected to<br />
determine the connectivity of other Central American<br />
populations of T. m. manatus, as well as the connectivity<br />
between the two subspecies.<br />
To address management issues, we suggest considering<br />
two distinct genetic populations in Mexico, one along the<br />
Caribbean coast and one in the riverine systems connected<br />
to the GMx. There is evidence to support unidirectional or<br />
limited gene flow between these two important areas.<br />
Special attention for conservation in the AB population<br />
would result in a healthier genetic stock, as this region is a<br />
source for potential mixture between the two primary<br />
genetic clusters.<br />
Based on this information it is important to maintain the<br />
natural migration routes of manatees from the GMx to the<br />
Caribbean area. This can be accomplished by conserving<br />
suitable manatee habitat along the north and east coasts of<br />
the Yucatan Peninsula, enforcing protection, reducing<br />
poaching and helping local initiatives that increase public
Genetica (2011) 139:833<strong>–</strong>842 841<br />
education. Ongoing tourism and urban development projects<br />
along those coasts are a major issue of concern.<br />
All of the captive manatees in Mexico were from the<br />
GMx region where they were originally rescued. However,<br />
most of them are in facilities located along the Caribbean<br />
coast. In the event of their release, returning them to the<br />
GMx would help maintain distinct populations. Also, if<br />
these facilities continue breeding manatees in captivity, a<br />
program that minimizes inbreeding and prevents inbred<br />
individuals or manatees with parents from different genetic<br />
populations from being released into the wild population<br />
must be considered. Genetic tools to assess lineages are<br />
currently being evaluated (Nourisson 2011).<br />
Acknowledgments This study is part of the doctoral dissertation<br />
research of the first author (CN), who was funded by a grant from the<br />
Secretaria de Relaciones Exteriores de México and by the Ministère<br />
des Affaires Etrangères Française—EGIDE. Financial support for<br />
training in Florida was provided by the Marine Mammal Commission.<br />
The project was funded by SEMARNAT/CONACYT (Project<br />
2002-C01-1128) and Dolphin Discovery for manatee captures and<br />
genetic analysis, as well as the US Geological Survey-Sirenia Project<br />
for genetic analysis. Mexican Government Research Permits and<br />
sample collection authorizations were obtained from DGVS # 03144;<br />
04513; 03670 and 03675. Samples were collected under scientific<br />
collection permits NUM/SGPA/DGVS/03144, NUM/SGPA/DGVS/<br />
04513, NUM/SGPA/DGVS/03670/06, SGPA/DGVS/04060/06;<br />
SGPA/DGVS/01103/07, SEDUM/SSMA/DGPPE/0191/2004 and<br />
RBSK OFICIO CUN 037/04. Samples were transported to the<br />
research laboratories in the US under CITES export permits<br />
MX22523, MX24463, MX28578, MX26485, MX38416, MX34645,<br />
MX31591 and CITES import permits 06US808447/9 and<br />
07US808447/9. Samples were processed in Florida under authority of<br />
USFWS wildlife research permit MA79<strong>17</strong>21 issued to the USGS<br />
Sirenia Project. All sample collection was done under approval of<br />
IACUC standards. Samples from the Gulf of Mexico area came from<br />
various projects and facilities: Dolphin Discovery, Veracruz Aquarium,<br />
Xcaret, ViaDelphi and Universidad Juárez Autónoma de<br />
Tabasco. We thank Dr. Margaret Hunter of the USGS Southeast<br />
Ecological Science Center for advice and help in the laboratory with<br />
various software packages. Any use of trade, product, or firm names is<br />
for descriptive purposes only and does not imply endorsement by the<br />
US Government.<br />
References<br />
Alvarez-Alemán A, Beck CA, Powell JA (2010) First report of a<br />
Florida manatee (Trichechus manatus latirostris) in Cuba. Aquat<br />
Mamm 36(2):148<strong>–</strong>153<br />
Amos B, Hoelzel AR (1991) Long-term preservation of whale skin<br />
for DNA analysis. Rep Int Whal Comm Special Issue 13:99<strong>–</strong>103<br />
Banks MA, Eichert W (2000) WHICHRUN (Version 3.2): a computer<br />
program for population assignment of individuals based on<br />
multilocus genotype data. J Hered 91:87<strong>–</strong>89<br />
CITES (2009) Appendices I, II and III. In: The CITES Appendices.<br />
Convention on <strong>International</strong> Trade in Endangered Species of<br />
Wild Fauna and Flora. http://www.cites.org/eng/app/appendices.<br />
shtml. Accessed 29 Oct 2009<br />
Cornuet JM, Luikart G (1996) Description and power analysis of two<br />
tests for inferring recent population bottlenecks from allele<br />
frequency data. Genetics 144:2001<strong>–</strong>2014<br />
Deutsch CJ, Reid JP, Bonde RK, Easton DE, Kochman HI, O’Shea TJ<br />
(2003) Seasonal movements, migratory behavior and site fidelity<br />
of West Indian manatees along the Atlantic Coast of the United<br />
States. Wildlife Monographs 151:1<strong>–</strong>77<br />
Deutsch CJ, Self-Sullivan C, Mignucci-Giannoni A (2008) Trichechus<br />
manatus. In: The IUCN Red List of Threatened Species<br />
Version 2009.1. <strong>International</strong> Union for Conservation of Nature<br />
and Natural Resources. http://www.iucnredlist.org/details/22<br />
103/0. Accessed 29 Oct 2009<br />
DiBattista JD (2008) Patterns of genetic variation in anthropogenically<br />
impacted populations. Conserv Genet 9(1):141<strong>–</strong>156<br />
Durand J (1983) Ocaso de sirenas, esplendor de manatíes, 2nd edn.<br />
Fondo de cultura económica, México<br />
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an<br />
integrated software package for population genetics data analysis.<br />
Evol Bioinform Online 1:47<strong>–</strong>50<br />
Fertl DA, Schiro J, Regan GT, Beck CA, Adimey N, Price-May L,<br />
Amos A, Worthy GAJ, Crossland R (2005) Manatee occurrence<br />
in the northern Gulf of Mexico, West of Florida. Gulf Caribb Res<br />
<strong>17</strong>:69<strong>–</strong>94<br />
Frankham R, Ballou JD, Briscoe DA (2002) Introduction to conservation<br />
genetics. Cambridge University Press, Cambridge<br />
Gallo-Reynoso JP (1983) Notas sobre la distribución del manati<br />
(Trichechus manatus) en las costas de Quintana Roo. An Inst<br />
Biol Univ Nac Auton Mex 53(1982), Ser Zool (1):443<strong>–</strong>448<br />
Garcia-Rodriguez AI, Bowen BW, Domning D, Mignucci-Giannoni<br />
AA, Marmontel M, Montoya-Ospina RA, Morales-Vela B,<br />
Rudin M, Bonde RK, McGuire PM (1998) Phylogeography of<br />
the West Indian manatee (Trichechus manatus): how many<br />
populations and how many taxa? Mol Ecol 7:1137<strong>–</strong>1149<br />
Garcia-Rodriguez AI, Moraga-Amador D, Farmerie W, McGuire P,<br />
King TL (2000) Isolation and characterization of microsatellite<br />
DNA markers in the Florida manatee (Trichechus manatus<br />
latirostris) and their application in selected <strong>Sirenian</strong> species. Mol<br />
Ecol 9:2161<strong>–</strong>2163<br />
Glaubitz JC (2004) CONVERT: a user-friendly program to reformat<br />
diploid genotypic data for commonly used population genetic<br />
software packages. Mol Ecol Notes 4:309<strong>–</strong>310<br />
Hillis DM, Larson A, Davis SK, Zimmer EA (1990) Nucleic acids III:<br />
sequencing. In: Hillis DM, Moritz C (eds) Mol Syst. Sinauer<br />
Associates, Sunderland, pp 318<strong>–</strong>370<br />
Hunter ME, Auil-Gomez NE, Tucker KP, Bonde RK, Powell J,<br />
McGuire PM (2010) Low genetic variation and evidence of<br />
limited dispersal in the regionally important Belize manatee.<br />
Anim Conserv 13:592<strong>–</strong>602<br />
Husar LS (1978) Trichechus manatus. Mamm Sp 93:1<strong>–</strong>5<br />
Landero-Figueroa MM (2010) Distribución potencial del manatí<br />
(Trichechus manatus manatus) en Bahía de la Ascensión,<br />
Quintana Roo. Master dissertation. CINVESTAV-Merida,<br />
Mexico<br />
Langtimm CA, Beck CA (2003) Lower survival probabilities for adult<br />
Florida manatees in years with intense coastal storms. Ecol Appl<br />
13:257<strong>–</strong>268<br />
Lazcano-Barrero MA, Packard JM (1989) The ocurrence of manatees<br />
(Trichechus manatus) in Tamaulipas, Mexico. Mar Mamm Sci<br />
5(2):202<strong>–</strong>205<br />
Lefebvre LW, Marmontel M, Reid JP, Rathbun GB, Domning DP<br />
(2001) Status and biogeography of the West Indian manatee. In:<br />
Woods CA, Sergile FE (eds) Biogeography of the West Indies:<br />
patterns and perspectives, 2nd edn. CRC Press, Boca Raton,<br />
pp 425<strong>–</strong>474<br />
Lefebvre LW, O’Shea TJ, Rathbun GB, Best RC (1989) Distribution,<br />
status, and biogeography of the West Indian manatee. Biogeography<br />
of the West Indies:567<strong>–</strong>610<br />
McClenaghan LR, O’Shea TJ (1988) Genetic variability in the Florida<br />
manatee (Trichechus manatus). J Mamm 69:481<strong>–</strong>488<br />
123
842 Genetica (2011) 139:833<strong>–</strong>842<br />
McKillop HI (1985) Prehistoric exploitation of the manatee in the<br />
Maya and circum-Caribbean areas. World Archaeol 16:337<strong>–</strong>353<br />
Medrano-González L, Morales-Vela B, García-Rodríguez A, Robles-<br />
Saavedra R, Baker S (1997) Análisis preliminar de la variación<br />
del DNA mitocondrial y del complejo mayor de histocompatibilidad<br />
en la laguna de Catazajá, Chiapas, y en la bahía de<br />
Chetumal, Quintana Roo. In: Morales-Vela B, Medrano-González<br />
L (eds) Variación genética del manatí (Trichechus<br />
manatus), en el sureste de México y monitoreo con radiotransmisores<br />
en Quintana Roo. Informe final. El Colegio de la<br />
Frontera Sur, Comisión Nacional para el Conocimiento y Uso de<br />
la Biodiversidad (H164), Chetumal, pp 12<strong>–</strong>38. http://www.cona<br />
bio.gob.mx/institucion/proyectos/resultados/InfH164.pdf.pdf. Accessed<br />
29 Oct 2009<br />
Merino M (1997) Upwelling on the Yucatan Shelf: hydrographic<br />
evidence. J Mar Syst 13:101<strong>–</strong>121<br />
Morales-Vela B, Padilla-Saldivar JA (2009a) Aspectos biológicos de<br />
los manatíes en el sur de Quintana Roo. In: Espinoza-Ávalos J,<br />
Islebe GA, Hernández-Arana HA (eds) El sistema ecológico de la<br />
bahía de Chetumal/Corozal: costa occidental del Mar Caribe. El<br />
Colegio de la Frontera Sur. Chetumal, Quintana Roo, pp 115<strong>–</strong>123.<br />
http://w2.ecosur-qroo.mx/cna/julio/libbahia.pdf. Accessed 20 Feb<br />
2010<br />
Morales-Vela B, Padilla-Saldívar JA (2009b) Demografía, ecología y<br />
salud de la población de manatíes (Trichechus manatus manatus)<br />
en Quintana Roo, y su variación y representación genética en<br />
México. Informe Técnico Final Proyecto SEMARNAT/CONA-<br />
CYT 2002-C01-1128. El Colegio de la Frontera Sur, Chetumal,<br />
259 p<br />
Morales-Vela B, Padilla-Saldívar JA (2011) El manatí. La sirena del<br />
Caribe. In: Pozo C, Armijo-Canto N, Calmé S (eds) Riqueza<br />
biológica de Quintana Roo. Un análisis para su conservación.<br />
Tomo 1. ECOSUR, CONABIO, Gob. Edo. Q. R., PPD, México,<br />
pp 248<strong>–</strong>255<br />
Morales-Vela B, Padilla-Saldívar JA, Mignucci-Giannoni AA (2003)<br />
Status of the manatee (Trichechus manatus) along the northern<br />
and western coasts of the Yucatán Peninsula, México. Caribb J<br />
Sci 39(1):42<strong>–</strong>49<br />
Morales-Vela B, Padilla-Saldívar JA, Reid J, Butler S (2007) First<br />
records of long-distance manatee movements between Mexico<br />
and Belize. Sirenews 47:12<strong>–</strong>14<br />
Nourisson C (2011) Estructura genética de los manatíes en México.<br />
Tesis de doctorado, El Colegio de la Frontera Sur, Chetumal,<br />
Quintana Roo, México. 121pp<br />
Olivera-Gómez LD (2006) Estado actual del manatí (Trichechus<br />
manatus) en humedales del sur del Golfo de México. In: Primer<br />
Simposio para la Biología y Conservación del manatí antillano<br />
en Mesoamérica. Loma Linda University. http://resweb.llu.edu/<br />
rford/research/Manatees/ProgramayResumenes.pdf. Accessed 30<br />
Oct 2009<br />
Oosterhout CV, Hutchinson WF, Wills DPM, Shipley P (2004)<br />
MICRO-CHECKER: software for identifying and correcting<br />
genotyping errors in microsatellite data. Mol Ecol Notes 4:<br />
535<strong>–</strong>538<br />
Pause KC (2007) Conservation genetics of the Florida manatee,<br />
Trichechus manatus latirostris. PhD dissertation. University of<br />
Florida, USA<br />
Pause KC, Nourisson C, Clark A, Kellogg ME, Bonde RK, McGuire<br />
PM (2007) Polymorphic microsatellite DNA markers for the<br />
123<br />
Florida manatee (Trichechus manatus latirostris). Mol Ecol<br />
Notes 7:1073<strong>–</strong>1076<br />
Peakall R, Smouse PE (2006) Genalex 6: genetic analysis in Excel.<br />
Population genetic software for teaching and research. Mol Ecol<br />
Notes 6(1):288<strong>–</strong>295<br />
Peel D, Ovenden JR, Peel SL (2004) NeEstimator: software for<br />
estimating effective population size. In: Fisheries. Queensland<br />
Government, Department of Primary Industries and Fisheries.<br />
http://www.dpi.qld.gov.au/cps/rde/dpi/hs.xsl/28_6908_ENA_<br />
HTML.htm. Accessed 3 Nov 2009<br />
Piry S (2004) GeneClass2: A software for genetic assignment and<br />
first-generation migrant detection. J Her 95:536<strong>–</strong>539<br />
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population<br />
structure using multilocus genotype data. Genetics 155:945<strong>–</strong>959<br />
Proebstel DS, Evans RP, Shiozawa DK, Williams RN (1993)<br />
Preservation of nonfrozen tissue samples from a salmonine fish<br />
Brachymystax lenok (Pallas) for DNA analysis. J Ichthyol 9:9<strong>–</strong><strong>17</strong><br />
Raymond M, Rousset F (1995) GENEPOP (version 1.2): population<br />
genetics software for exact tests and ecumenicism. J Hered<br />
86:248<strong>–</strong>249<br />
Reep RL, Bonde R (2006) The Florida manatee. The University Press<br />
of Florida, Gainesville, FL<br />
Reyes-Mendoza O, Morales-Vela B (2007) New observations of<br />
manatees off the northern coast of Quintana Roo, Mexico.<br />
Sirenews 47:14<br />
Rousset F (2008) Genepop’007: a complete reimplementation of the<br />
Genepop software for Windows and Linux. Mol Ecol Resour<br />
8:103<strong>–</strong>106<br />
SEDUE (1991) Acuerdo por el que se establecen los criterios<br />
ecológicos CT-CERN-001-91 que determinan las especies raras,<br />
amenazadas, en peligro de extinción o sujetas a protección<br />
especial y sus endemismos, de la flora y la fauna terrestres y<br />
acuáticas en la República Mexicana. In: Diario Oficial de la<br />
Federación, <strong>17</strong> de mayo de 1991. Gobierno Constitucional de los<br />
Estados Unidos Mexicanos. México, pp 7<strong>–</strong>36<br />
SEMARNAT (2001) Proyecto de conservación, recuperación y<br />
manejo del manatí Trichechus manatus en México. Ser PREP<br />
num 11. Secretaría de Medio Ambiente y Recursos Naturales,<br />
México<br />
SEMARNAT (2002) Norma Oficial Mexicana NOM-059-SEMAR-<br />
NAT-2001 Protección Ambiental—Especies nativas de México<br />
de Flora y Fauna Silvestres—Categorías de Riesgo y Especificaciones<br />
para su Inclusión, Exclusión o Cambio—Lista de<br />
Especies en Riesgo. In: Diario Oficial de la Federación, 6 de<br />
marzo de 2002, Segunda Sección. Gobierno Constitucional de<br />
los Estados Unidos Mexicanos. México, pp 1<strong>–</strong>80<br />
SoftGenetics (2008) GeneMarker, the biologist friendly software. In<br />
SoftGenetics, software powertools for genetic analysis. http://www.<br />
softgenetics.com/GeneMarker.html. Accessed 30 Oct 2009<br />
Vianna JA, Bonde RK, Caballero S, Giraldo JP, Lima RP, Clark A,<br />
Marmontel M, Morales-Vela B, de Souza MJ, Parr L, Rodríguez-<br />
López M, Mignucci-Giannoni AA, Powell JA, Santos FR (2006)<br />
Phylogeography, phylogeny and hybridization in trichechid<br />
sirenians: implications for manatee conservation. Mol Ecol<br />
15:433<strong>–</strong>447<br />
White PS, Densmore LD (1992) Mitochondrial DNA isolation. In:<br />
Hoezel AR (ed) Molecular genetic analysis of populations: a<br />
practical approach. IRL Press, Oxford University Press, New<br />
York, pp 29<strong>–</strong>58
Fish Sci (2011) 77:795<strong>–</strong>798<br />
DOI 10.1007/s12562-011-0388-x<br />
ORIGINAL ARTICLE Biology<br />
The differences in behavioral responses to a net obstacle<br />
between day and night in captive manatees; does entanglement<br />
happen at night?<br />
Mumi Kikuchi • Miwa Suzuki • Keiichi Ueda •<br />
Hirokazu Miyahara • Senzo Uchida<br />
Received: 23 March 2011 / Accepted: 27 June 2011 / Published online: 23 July 2011<br />
Ó The Japanese Society of Fisheries Science 2011<br />
Abstract Entanglement in fishing gear occurs in endangered<br />
manatees and may result in serious injury or death.<br />
Such incidents may happen more frequently at night when<br />
the animal’s visual sense is limited. In this study, we<br />
examined the differences in behavioral response of captive<br />
manatees to a net obstacle during light (day) and dark<br />
(night) periods. We used a plastic net as the obstacle, and<br />
video-recorded the manatees’ behavior. The experiments<br />
showed that captive manatees avoided the obstacle during<br />
the day more frequently than at night, which suggests that<br />
the manatees can perceive the obstacle more readily during<br />
light periods. However, there was no difference in the<br />
frequency of bumping or actively touching the obstacle<br />
between light and dark periods. The results suggest that the<br />
manatees can recognize the net obstacle even at night by<br />
purposely touching it, but they avoid it less frequently, and<br />
M. Kikuchi (&)<br />
Laboratory of Fisheries Biology, Graduate School<br />
of Agricultural and Life Science, University of Tokyo,<br />
1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan<br />
e-mail: mumi@cocoa.plala.or.jp<br />
M. Suzuki<br />
Department of Marine Science and Resources, College<br />
of Bioresource Sciences, Nihon University, 1866 Kameino,<br />
Fujisawa, Kanagawa 252-0880, Japan<br />
e-mail: miwa@brs.nihon-u.ac.jp<br />
K. Ueda H. Miyahara S. Uchida<br />
Okinawa Churaumi Aquarium, 424 Ishikawa, Motobu,<br />
Okinawa 905-0206, Japan<br />
e-mail: k_ueda@kaiyouhaku.or.jp<br />
H. Miyahara<br />
e-mail: h_miyahara@kaiyouhaku.or.jp<br />
S. Uchida<br />
e-mail: s_uchida@kaiyouhaku.or.jp<br />
that entanglement during light periods may occur during<br />
accidental bumping, rather than from a failure to recognize<br />
it altogether.<br />
Keywords Obstacle Behavior Trichechus manatus<br />
Introduction<br />
The greatest threat to many marine mammals is incidental<br />
entanglement and mortality in a fishing net, or bycatch [1].<br />
The three species of manatees are aquatic mammals of the<br />
order Sirenia, and are known as generalist aquatic herbivores<br />
with no apparent natural predators. However, there<br />
are conservation concerns for all three species because of<br />
collisions with boats, entanglement in fishing nets, hunting<br />
and habitat degradation. In Florida, collisions of the manatee<br />
Trichechus manatus latirostris with watercraft are the<br />
primary cause of human-contact-related manatee mortality,<br />
but other preventable deaths result from entrapment in<br />
water control structures and entanglement in fishing gear<br />
[2]. The Antillean manatee Trichechus manatus manatus is<br />
primarily threatened by entanglement in fishing nets [3, 4].<br />
The intensity of boat traffic throughout its range is relatively<br />
low in the areas used by Antillean manatees, and<br />
may not be a significant threat [3]. In the West African<br />
manatee Trichechus senegalensis, subsistence hunters have<br />
heavily exploited this species [5]; however, at present,<br />
incidental entanglement seems to be the most significant<br />
threat [6]. Even when manatees survive such encounters,<br />
entanglement may still result in serious injury, the loss of<br />
part of a pectoral flipper, and even death later on [5, 7, 8].<br />
A high degree of sensitivity for brightness discrimination<br />
has been observed in West Indian manatees, which is<br />
comparable to some piscivorous predators, such as fur seals<br />
123
796 Fish Sci (2011) 77:795<strong>–</strong>798<br />
[9]. Furthermore, manatees have dichromatic color vision<br />
that is most sensitive in the blue and green region of the<br />
spectrum [10, 11]. Color vision greatly enhances pattern<br />
recognition and may compensate for some deficiencies in<br />
resolution, at least in comparison to cetaceans with<br />
monochromatic vision [12]. Hartman [7] concluded that<br />
vision is the preferred means of exploring their surroundings<br />
in manatees. Therefore, we might safely assume that<br />
manatee entanglement happens more frequently at night<br />
when their visual sense may be more limited. In order to<br />
test this assumption, any possible difference in cognitive<br />
performance needs to be assessed between light and dark<br />
periods. In this study, we examined the difference in<br />
behavioral responses of captive manatees to a net obstacle<br />
between day and night periods, in order to better understand<br />
the apparent causes of manatee entanglement.<br />
Materials and methods<br />
Experiments were conducted with four captive Antillean<br />
manatees Trichechus manatus manatus from June to September<br />
2004 at the Okinawa Churaumi Aquarium, Japan<br />
(Table 1). The manatees were kept in two indoor pools<br />
with the sexes separated. The first pool, containing two<br />
males, was 8.0 m long, 4.9 m wide and 3.0 m deep; the<br />
second pool, containing two females, was 8.0 m long,<br />
10.8 m wide and 3.0 m deep. The indoor facility has large<br />
windows that allow natural sunlight into the pools.<br />
Experiments were carried out randomly in the morning<br />
(0730 hours) and at night (1900<strong>–</strong>2000 hours). All experiments<br />
were conducted outside of regular business hours to<br />
minimize potential human interference. A total of eight and<br />
nine independent trials were conducted during the daytime<br />
and nighttime, respectively. In each trial, the subject’s<br />
behaviors were recorded for one hour with a video camera<br />
(FV40/FV300, Canon, Japan). During nighttime trials, a<br />
night-vision scope (WV-BP50, Panasonic, Japan) was also<br />
used to observe the subject’s behavior. Illumination<br />
intensity was recorded near the surface of the water using<br />
a digital luxmeter (LX-1332, Custom Co., Japan) at the<br />
beginning and end of each trial.<br />
Table 1 Name, standard body length (SL), body mass (BM), age<br />
(estimated for all except Yuma), and sex of each manatee used in the<br />
current study (data from 2004)<br />
ID Name SL (m) BM (kg) Age (year) Sex<br />
1 Yukatan 2.7 302 26 Male<br />
2 Ryu 2.7 3<strong>17</strong> 14 Male<br />
3 Maya 2.8 412 <strong>17</strong> Female<br />
4 Yuma No data No data 3 a<br />
Female<br />
a Yuma was born at the aquarium<br />
123<br />
Fig. 1 Photo of the black plastic net obstacle used in this study. The<br />
obstacle was constructed from PVC pipe and plastic netting, with<br />
weights attached to the bottom<br />
In order to examine the manatee’s behavioral response<br />
to the obstacle, a plastic net with no danger of entanglement<br />
was used. The obstacle (150 9 160 cm, black) was<br />
constructed from PVC tubing and plastic netting (1.5 9<br />
1.5 cm mesh size, 0.2 cm width, Takiron Co., Ltd.), with<br />
weights attached to the lower part of the obstacle to submerge<br />
it perpendicularly in the water (Fig. 1). The obstacle<br />
was placed in the pool such that the manatees had sufficient<br />
space to pass around it with no interference. The position<br />
of the obstacle was changed for each trial. To analyze the<br />
behavioral response to the obstacle, each individual’s<br />
behavior was classified into three categories: avoiding,<br />
bumping, or active touching with their snout or pectoral fin.<br />
The number of times these behaviors were observed and<br />
the duration of active touching were recorded. In addition,<br />
the number and durations of events of inactive behavior at<br />
the surface or on the bottom were calculated in order to<br />
compare with their active behaviors between day and night.<br />
To confirm the difference in illumination intensity<br />
between day and night, a generalized linear mixed model<br />
(GLMM) was analyzed with gamma errors, with each trial<br />
considered a random factor, and day<strong>–</strong>night differences<br />
treated as explanatory variables. To test the effect of the<br />
day<strong>–</strong>night difference on the number and duration of each<br />
recorded behavior, a GLMM was used with Poisson and<br />
gamma errors, respectively, and each individual manatee
Fish Sci (2011) 77:795<strong>–</strong>798 797<br />
was included as a random factor. The most parsimonious<br />
model was selected based on the Akaike information criterion<br />
(AIC). For all statistical analyses, we used the<br />
software R, and the GLMM was analyzed using the<br />
package lme4 and the function lmer.<br />
Results<br />
The GLMM for illumination intensity including the day<strong>–</strong><br />
night difference as an explanatory variable was the best<br />
model, with the lowest AIC. The model had an AIC value<br />
that was 578.2 lower than the second best model, which<br />
included random effects only. This suggests that illumination<br />
intensity clearly differed between day and night<br />
(GLMM: day<strong>–</strong>night, estimated value during the day:<br />
1161.0 ± 63.5 SE, night: 1.1 ± 63.5 SE).<br />
The GLMM for the number of events of avoiding<br />
behavior that included the day<strong>–</strong>night difference as an<br />
explanatory variable was the best model, with the lowest<br />
AIC (Table 2). This model had an AIC value that was 31.1<br />
lower than the second best model, which included random<br />
effects only. These results suggest that the number of<br />
events of avoiding behavior differed between the day and<br />
night: these subjects showed a greater number of events of<br />
obstacle avoidance behavior during the day than at night<br />
(GLMM: day<strong>–</strong>night, estimated value during the day:<br />
1.7 ± 3.5 SE, at night: 0.8 ± 1.1 SE).<br />
The model with the lowest AIC value with respect to<br />
the number of events of bumping or active touching<br />
included random effects only (Table 2). The model had an<br />
AIC value of 0.8 and 1.7 lower than the second best<br />
model, which included the day<strong>–</strong>night difference. The<br />
GLMM for the number of events of bumping (GLMM:<br />
random effects only, 0.9 ± 1.3 SE) and active touching<br />
(GLMM: random effects only, 7.2 ± 2.1 SE) revealed<br />
that it was not related to the day<strong>–</strong>night difference. The<br />
GLMM for the duration of active touching, including<br />
random effects only, was the best model with the lowest<br />
AIC. The AIC value was 2.0 lower than the second best<br />
model, which included the day<strong>–</strong>night difference. These<br />
results suggest that the duration of active touching was not<br />
different between day and night (GLMM: random effects<br />
only, 49.8 ± 12.8 SE).<br />
The GLMM for the number or duration of events of<br />
inactive behavior, including random effects only, was the<br />
best model with the lowest AIC. The model had an AIC<br />
value of 0.7 and 0.8 lower than the second best model,<br />
which included day<strong>–</strong>night. These results indicate that the<br />
number (GLMM: random effects only, 1.3 ± 1.5 SE) and<br />
duration (GLMM: random effects only, 104.4 ± 10.7 SE)<br />
of events of inactive behavior were not related to the day<strong>–</strong><br />
night difference.<br />
Discussion<br />
In this study, the number and duration of events of<br />
inactive behavior did not differ between day and night.<br />
Therefore, we concluded that our subjects’ activities did<br />
not differ between day and night, and that it did not affect<br />
their behavioral responses to the obstacle during the<br />
experiments.<br />
Focusing on the differences in behavioral responses to a<br />
net obstacle between day and night, we found that captive<br />
manatees show a greater number of avoidance behaviors<br />
during the day than at night. These results suggest that<br />
manatees can recognize the obstacle more easily during<br />
light periods. However, there was no difference in the<br />
number of bumping events between light and dark periods.<br />
Therefore, accidental bumping may be expected to occur<br />
regardless of the level of illumination and manatee’s higher<br />
degree of associated recognition.<br />
All sirenian species have sparse hairs on their bodies and<br />
orofacial that are of the sinus type, and typically tactile in<br />
function [13<strong>–</strong>18]. In fact, it has been specifically reported<br />
that a manatee’s orofacial hairs are used for tactile exploration<br />
and discrimination [13<strong>–</strong>16, 18]. In the current study,<br />
the number and the duration of active touching behavioral<br />
events did not differ between day and night. These results<br />
indicate that manatees heavily rely on their tactile senses to<br />
recognize an obstacle, even during light periods that<br />
facilitate purely visual recognition. In field studies, when<br />
wild Antillean manatees are caught by encircling them with<br />
Table 2 The GLMM models for the number of events of each behavioral response to the obstacle (avoidance, active touching, and bumping of<br />
the obstacle) along with AIC and delta AIC (DAIC) values<br />
Model The number of events of each behavioral response to the obstacle<br />
Avoidance Active touching Bumping<br />
AIC DAIC AIC DAIC AIC DAIC<br />
Random effects only 200.7 31.1 180.8 0.0 48.8 0.0<br />
Day<strong>–</strong>night 169.6 0.0 182.5 1.7 49.6 0.8<br />
Favoured models (bold type) were evaluated on the basis of the lowest AIC<br />
123
798 Fish Sci (2011) 77:795<strong>–</strong>798<br />
nets, the animals often explore the ‘‘weak areas’’ of the net<br />
via tactile contact in order to escape (Olivera-Gomez LD,<br />
2011, pers. comm.). They tend to descend to the bottom of<br />
the net and pull it up with its snout. There are many reports<br />
of these escape behaviors by local fishermen. Wild Antillean<br />
manatees can escape from the net in this way, and this<br />
more often occurs at deeper depths, where the net is more<br />
vertical (Olivera-Gomez LD, 2011, pers. comm.). In shallow<br />
depths, where the net forms a bag, manatees enter the<br />
fold of the net and are more easily caught on it (Olivera-<br />
Gomez LD, 2011, pers. comm.). In addition, entanglement<br />
often occurs with larger mesh size nets, which are used to<br />
catch large fishes (Olivera-Gomez LD, 2011, pers. comm.).<br />
Therefore, we further recommend examining the position<br />
of the net relative to the depth and the effect of the mesh<br />
size in order to clarify the causes of manatee entanglement.<br />
Acknowledgments We would like to thank all of the staff at the<br />
Okinawa Churaumi Aquarium for their assistance in conducting these<br />
experiments. We would like to thank K. Asahina at Nihon University,<br />
H. Kato at Tokyo University of Marine Science and Technology, L.<br />
D. Olivera-Gomez at the Autonomous University Juarez of Tabasco,<br />
and D. Gonzalez-Socoloske at Duke University for their discussions<br />
and insightful comments on previous drafts. We also thank J.<br />
A. Mobley and D. Gonzalez-Socoloske for correcting the English<br />
manuscript. The present study was supported by the Fisheries<br />
Research Agency, National Research Institute of Far Seas Fisheries.<br />
References<br />
1. Grech A, Marsh H, Coles R (2008) A spatial assessment of the<br />
risk to a mobile marine mammal from bycatch. Aquat Conserv<br />
Mar Freshw Ecosys 18:1127<strong>–</strong>1139<br />
2. Reep RL, Bonde RK (2006) The Florida manatee: biology and<br />
conservation. University Press of Florida, Gainesville<br />
3. de Thoisy B, Spiegelberger T, Rousseau S, Talvy G, Vogel I, Vie<br />
JC (2003) Distribution, habitat, and conservation status of the<br />
West Indian manatee Trichechus manatus in French Guiana.<br />
Oryx 37:431<strong>–</strong>436<br />
4. Castelblanco-Martinez DN, Bermudez-Romero AL, Gomez-<br />
Camelo IV, Rosas FCW, Trujillo F, Zerda-Ordonez E (2009)<br />
Seasonality of habitat use, mortality and reproduction of the<br />
123<br />
vulnerable Antillean manatee Trichechus manatus manatus in the<br />
Orinoco River, Colombia: implications for conservation. Oryx<br />
43:235<strong>–</strong>242<br />
5. Reeves RR, Tuboku-Metzger D, Kapindi RA (1988) Distribution<br />
and exploitation of manatees in Sierra Leone West Africa. Oryx<br />
22:75<strong>–</strong>84<br />
6. Silva MA, Araujo A (2001) Distribution and current status of<br />
the West African manatee (Trichechus senegalensis) in Guinea-<br />
Bissau. Mar Mammal Sci <strong>17</strong>:418<strong>–</strong>424<br />
7. Hartman DS (1979) Ecology and behavior of the manatee<br />
(Trichechus manatus) (Spec Publ 5). Am Soc Mammals,<br />
Lawrence<br />
8. Beck CA, Barros NB (1991) The impact of debris on the Florida<br />
manatee. Mar Pollut Bull 22:508<strong>–</strong>510<br />
9. Griebel U, Schmid A (1997) Brightness discrimination ability in<br />
the West Indian manatee (Trichechus manatus). J Exp Biol<br />
200:1587<strong>–</strong>1592<br />
10. Cohen JL, Tucker GS, Odell DK (1982) The photoreceptors of<br />
the West Indian manatee. J Morphol <strong>17</strong>3:197<strong>–</strong>202<br />
11. Griebel U, Schmid A (1996) Color vision in the manatee<br />
(Trichechus manatus). Vision Res 36:2747<strong>–</strong>2757<br />
12. Bauer GB, Colbert DE, Gaspard JCIII, Littlefield B (2003)<br />
Underwater visual acuity of Florida manatees (Trichechus manatus<br />
latirostris). Int J Comp Sociol Psychol 16:130<strong>–</strong>142<br />
13. Marshall CD, Clark LA, Reep RL (1998) The muscular hydrostat<br />
of the Florida manatee (Trichechus manatus latirostris): a functional<br />
morphological model of perioral bristle use. Mar Mamm<br />
Sci 14:290<strong>–</strong>303<br />
14. Reep RL, Marshall CD, Stoll ML, Whitaker DM (1998) Distribution<br />
and innervation of facial bristles and hairs in the Florida<br />
manatee (Trichechus manatus latirostris). Mar Mamm Sci<br />
14:257<strong>–</strong>273<br />
15. Bachteler D, Dehnhardt G (1999) Active touch performance in<br />
the Antillean manatee: evidence for a functional differentiation of<br />
facial tactile hairs. Zool 102:61<strong>–</strong>69<br />
16. Reep RL, Stoll ML, Marshall CD, Homer BL, Samuelson DA<br />
(2001) Microanatomy of facial vibrissae in the Florida manatee:<br />
the basis for specialized sensory function and oripulation. Brain<br />
Behav Evol 58:1<strong>–</strong>14<br />
<strong>17</strong>. Reep RL, Marshall CD, Stoll ML (2002) Tactile hairs on the<br />
postcranial body in Florida manatees: a mammalian lateral line?<br />
Brain Behav Evol 59:141<strong>–</strong>154<br />
18. Marshall CD, Maeda H, Iwata M, Furuta M, Asano S, Rosas F,<br />
Reep RL (2003) Orofacial morphology and feeding behaviour of<br />
the dugong, Amazonian, West African and Antillean manatees<br />
(Mammalia: Sirenia): functional morphology of the muscular<strong>–</strong><br />
vibrissal complex. J Zool 259:245<strong>–</strong>260
Low genetic variation and evidence of limited dispersal in<br />
the regionally important Belize manatee<br />
M. E. Hunter 1,2 , N. E. Auil-Gomez 3,4 , K. P. Tucker 5 , R. K. Bonde 1,2 , J. Powell 4 & P. M. McGuire 2<br />
1 Sirenia Project, Southeast Ecological Science Center, U.S. Geological Survey, Gainesville, FL, USA<br />
2 Department of Physiological Sciences, College of Veterinary Medicine, University of Florida,Gainesville, FL, USA<br />
3 Wildlife Trust, Belize City, Belize<br />
4 Sea to Shore Alliance, St Petersburg, FL, USA<br />
5 College of Marine Science, University of South Florida, St. Petersburg, FL, USA<br />
Keywords<br />
conservation genetics; low diversity;<br />
microsatellite; mitochondria; marine<br />
mammal; West Indian manatee.<br />
Correspondence<br />
Margaret E. Hunter, Sirenia Project,<br />
Southeast Ecological Science Center, US<br />
Geological Survey, 2201 NW 40th Terrace,<br />
Gainesville, FL 32605, USA. Tel: +1 352<br />
264 3484; Fax: +1 352 378 4956<br />
Email: mkellogg@usgs.gov<br />
Received 26 October 2009; accepted 13 June<br />
2010<br />
doi:10.1111/j.1469-<strong>17</strong>95.2010.00383.x<br />
Introduction<br />
Abstract<br />
The West Indian manatee Trichechus manatus is a threatened<br />
aquatic mammal found throughout the south-eastern United<br />
States, Central and South America and the Caribbean. The<br />
Florida manatee Trichechus manatus latirostris and Antillean<br />
manatee Trichechus manatus manatus are the two recognized<br />
subspecies of the West Indian manatee. A low reproductive<br />
rate, environmental impacts and direct threats from the human<br />
population have historically limited manatee population<br />
growth. In 1982, the <strong>International</strong> Union for Conservation of<br />
Nature classified all West Indian manatee populations as<br />
vulnerable to extinction (IUCN, 2007).<br />
During the <strong>17</strong>th through 19th centuries, the Spaniards in<br />
Belize and throughout the region severely exploited the<br />
592<br />
Animal Conservation. Print ISSN 1367-9430<br />
The Antillean subspecies of the West Indian manatee Trichechus manatus is found<br />
throughout Central and South America and the Caribbean. Because of severe<br />
hunting pressure during the <strong>17</strong>th through 19th centuries, only small populations of<br />
the once widespread aquatic mammal remain. Fortunately, protections in Belize<br />
reduced hunting in the 1930s and allowed the country’s manatee population to<br />
become the largest breeding population in the Wider Caribbean. However,<br />
increasing and emerging anthropogenic threats such as coastal development,<br />
pollution, watercraft collision and net entanglement represent challenges to this<br />
ecologically important population. To inform conservation and management<br />
decisions, a comprehensive molecular investigation of the genetic diversity,<br />
relatedness and population structure of the Belize manatee population was<br />
conducted using mitochondrial and microsatellite DNA. Compared with other<br />
mammal populations, a low degree of genetic diversity was detected (HE=0.455;<br />
N A=3.4), corresponding to the small population size and long-term exploitation.<br />
Manatees from the Belize City Cayes and Southern Lagoon system were genetically<br />
different, with microsatellite and mitochondrial FST values of 0.029 and<br />
0.078, respectively (P 0.05). This, along with the distinct habitats and threats,<br />
indicates that separate protection of these two groups would best preserve the<br />
region’s diversity. The Belize population and Florida subspecies appear to be<br />
unrelated with microsatellite and mitochondrial F ST values of 0.141 and 0.63,<br />
respectively (P 0.001), supporting the subspecies designations and suggesting<br />
low vagility throughout the northern Caribbean habitat. Further monitoring and<br />
protection may allow an increase in the Belize manatee genetic diversity and<br />
population size. A large and expanding Belize population could potentially assist<br />
in the recovery of other threatened or functionally extinct Central American<br />
Antillean manatee populations.<br />
Antillean subspecies for sustenance (Lefebvre et al., 2001).<br />
By 1936, the population decline was so severe in Belize that<br />
Manatee Protection Ordinances were introduced to preserve<br />
the population (McCarthy, 1986). Today, the Belize manatee<br />
is listed as endangered by the Belize Wildlife Protection<br />
Act of 1981, Part II, Section 3(a) (Auil, 1998, 2004). A<br />
Manatee Recovery Plan was also proposed requesting<br />
information on habitat use and movement patterns to aid<br />
the development of conservation policies for the protection<br />
of the Belize manatee (Auil, 1998).<br />
While studying Caribbean manatee populations, O’Shea<br />
& Salisbury (1991) concluded that ‘Belize remains one of the<br />
last strongholds for the species in this part of the world.’<br />
From 1977 to 1991, the Belize manatee population size<br />
appeared stable, however, an overall negative trend in<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
M. E. Hunter et al.<br />
numbers was detected from 1997 to 2002 (Auil, 2004).<br />
Although today Belize is considered the largest breeding<br />
Antillean population, it is still under pressure from anthropogenic<br />
threats and the census size of 1000 individuals is<br />
well below that recommended for the long-term genetic<br />
sustainability of a population (Frankham, Ballou & Briscoe,<br />
2002; Quintana-Rizzo & Reynolds III, 2007).<br />
Little is known of the abundance or distribution of<br />
manatees in the other Central American populations,<br />
although they are considered elusive and thought to be rare<br />
(Quintana-Rizzo & Reynolds III, 2007). However, supplementation<br />
from Belize or Mexico manatees in concert with<br />
coordinated regional efforts could potentially assist in the<br />
recovery of the Guatemala populations and those further<br />
the south. The large Mexico population is also protected<br />
Reduced Belize manatee dispersal and genetic variation<br />
Figure 1 Geographic map of Belize with the<br />
three study sites inset, Belize City Cayes<br />
(north), Northern & Southern Lagoons (center),<br />
and Placencia Lagoon (south).<br />
and manatees from Mexico have been documented traveling<br />
to Belize and even participating in a mating herd (Auil et al.,<br />
2007; Morales-Vela et al., 2007). Quintana-Rizzo & Reynolds<br />
III (2007) suggest that safeguarding the Belize and<br />
Mexico populations as a resource of regional importance<br />
should be a high conservation priority.<br />
Long-term exploitation and small population sizes<br />
can lower genetic diversity and decrease fitness-related<br />
traits, such as fecundity and survival (Hoelzel et al., 1993;<br />
Frankham et al., 2002; Dixon et al., 2007). Genetic variation<br />
often acts in concert with demographic stochasticity in<br />
small populations. Loss of diversity can ultimately lead to<br />
an ‘extinction vortex,’ or a cyclic reduction in population<br />
size, resulting in loss of the population (Gilpin & Soulé,<br />
1986).<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London 593
Reduced Belize manatee dispersal and genetic variation M. E. Hunter et al.<br />
A previous mitochondrial DNA (mtDNA) study identified<br />
three haplotypes (A03, A04 and J01) in 43 animals in Belize<br />
(Vianna et al., 2006). The resultant haplotype parameters<br />
included 28 polymorphic sites and nucleotide substitutions<br />
with h=0.558 and p=0.037. The distribution and genetic<br />
structure of the mtDNA haplotypes was not addressed, but<br />
could provide valuable information on the population structure,<br />
especially coupled with biparentally inherited nuclear<br />
data. Presented here are mitochondrial and nuclear microsatellite<br />
DNA studies addressing the genetic diversity,<br />
relatedness and population structure in Belize manatees.<br />
The Antillean Belize population is also compared with<br />
T. m. latirostris to provide information on the extent of West<br />
Indian manatee migration and relatedness.<br />
Materials and methods<br />
Sample collection and DNA extraction<br />
Manatee epidermis tissue and/or white blood cells were<br />
collected from recovered carcasses or during wild manatee<br />
health assessments in Belize from 1997 to 2007. Genetic<br />
samples were collected from a total of 118 unique animals<br />
in 213 health assessments (50% female and 50% male; Auil<br />
et al., 2007) and 12 carcass samples. Genomic DNA was<br />
isolated using Qiagen’s DNeasy blood and tissue kits<br />
(Valencia, CA, USA). Within the Florida dataset, 96 genotypes<br />
were randomly chosen, proportionally representing<br />
the four demographically defined management units<br />
(K.P. Tucker et al., unpubl. data).<br />
Study locations<br />
The Belize City Cayes (BCC), including the near shore and<br />
outlying cayes, are utilized by the tourist industry and experience<br />
heavy boat traffic and recreational activities from Belize<br />
City, the largest port in the country (Fig. 1). Manatee tours are<br />
largely conducted around Swallow Caye within the BCC.<br />
The Southern Lagoon system (SLS) consists of the Southern<br />
and Northern Lagoons and has the highest recorded<br />
number of manatees, with 55 individuals sighted in a single<br />
aerial survey (O’Shea & Salisbury, 1991). While the population<br />
is not heavily affected by watercraft, it is potentially<br />
impacted by salinity changes due to increased rainfall,<br />
influencing the abundance of aquatic vegetation, the primary<br />
food source for manatees.<br />
Placencia Lagoon (PL) is located 75 miles south of the<br />
SLS (Fig. 1). Placencia Village is a popular tourist destination,<br />
with a large amount of coastal development and<br />
agricultural runoff severely altering the environment and<br />
aquatic vegetation. Nitrification from shrimp farm effluent<br />
causes extensive algae blooms that can compromise the<br />
quality and growth of seagrass (Thornton, Shanahan &<br />
Shanahan, 2003). Nitrification also provides a food source<br />
for a marine snail, thought to be an intermediate host of a<br />
respiratory fluke Pulmonicola cochleotrema (Blair, 2005),<br />
which is known to parasitize manatee lungs (Beck &<br />
Forrester, 1988). Manatees in PL present at necropsy with<br />
594<br />
large loads of P. cochleotrema and copious mucoid discharge<br />
that could hinder respiration (Auil et al., 2007).<br />
The carcass samples utilized here were recovered from<br />
Corazol (n=1), Stan Creek (n=1), the Belize River mouth<br />
in Belize City (an area of high watercraft activity), or south of<br />
the river, possibly caught in the ocean current (n=10). The<br />
northern most manatee reserve in Belize is a section of<br />
Chetumal Bay, which continues into Mexico (Fig. 1). Samples<br />
collected in Chetumal Bay are not included in this study.<br />
Laboratory procedures<br />
Mitochondrial DNA analysis<br />
Primers CR-4 and CR-5 from García-Rodríguez et al.<br />
(1998) amplified a 410 bp portion of the mtDNA control<br />
region displacement loop for 113 individuals; 101 live<br />
captures and 12 carcasses. All PCR amplifications were<br />
carried out on a PTC-200 thermal cycler (MJ Research,<br />
Waltham, MA, USA). The PCR reaction conditions followed<br />
Kellogg (2008). Representatives from each haplotype<br />
and any ambiguous sequences were sequenced in both<br />
directions to ensure the accuracy of nucleotide designations.<br />
Microsatellite DNA analysis<br />
A panel of 16 polymorphic microsatellite primers (García-<br />
Rodríguez et al., 2000; Pause et al., 2007) was PCR amplified<br />
using 118 individuals. The study included 88 individuals<br />
from the SLS system (12 Northern Lagoon and 76 Southern<br />
Lagoon), 21 from BCC, two from PL and seven amplifiable<br />
carcass samples. PCR conditions followed Kellogg (2008;<br />
Table 1). MgCl2 concentrations were 3 mM, except for<br />
TmaH13, TmaKb60 and TmaSC5, which required 2 mM.<br />
Fragment analysis was performed on an Applied Biosystems<br />
(Foster City, CA, USA) ABI 3730 Genetic Analyzer. All<br />
individuals amplified at 14 or more loci. The Florida data<br />
were kindly provided for use in this paper (K.P. Tucker<br />
et al., unpubl. data).<br />
Statistical analysis<br />
Mitochondrial DNA analysis<br />
The degree of differentiation, F ST and F ST, betweentheBCC<br />
and SLS and between Belize and Florida was calculated using<br />
ARLEQUIN 3.1 (Excoffier, Laval & Schneider, 2005). Comparisons<br />
with PL were not conducted due to the small sample size.<br />
Estimates of sequence divergence used the Kimura twoparameter<br />
genetic distance model (Kimura, 1980; Jin & Nei,<br />
1990). Lastly, Tajima’s D of selective neutrality, genetic<br />
diversity (h), nucleotide diversity (p), number of polymorphic<br />
sites (S) and number of nucleotide substitutions (NS) were<br />
calculated (Nei, 1987; Tajima, 1993).<br />
Microsatellite DNA analysis<br />
The level of polymorphism was estimated by the observed<br />
(HO) and expected heterozygosity (HE) and the average<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
M. E. Hunter et al.<br />
Table 1 Characteristics and PCR conditions for the 16 microsatellite loci implemented in the Belize manatee samples<br />
Locus name Tm ( 1C) BSA NA HO HE PIC NE TmaA02 56 2 0.310 0.349 0.534 1.537<br />
TmaE1 60 + 4 0.615 0.619 1.121 2.622<br />
TmaE02 55 2 0.265 0.300 0.477 1.429<br />
TmaE7 62 + 4 0.299 0.323 0.627 1.477<br />
TmaE08 58 3 0.701 0.657 1.084 2.9<strong>17</strong><br />
TmaE11 56 5 0.761 0.757 1.488 4.123<br />
TmaE14 62 + 4 0.564 0.573 1.035 2.340<br />
TmaE26 60 4 0.381 0.341 0.622 1.518<br />
TmaF14 58 2 0.390 0.403 0.593 1.675<br />
TmaH13 58 + 2 0.359 0.355 0.540 1.550<br />
TmaJ02 60 2 0.504 0.482 0.675 1.932<br />
TmaK01 54 2 0.530 0.390 0.578 1.638<br />
TmaKb60 56 5 0.416 0.483 0.782 1.935<br />
TmaM79 56 + 3 0.627 0.593 0.974 2.454<br />
TmaSC5 58 3 0.410 0.366 0.663 1.577<br />
TmaSC13 58 2 0.162 0.230 0.391 1.298<br />
Mean 3.1 0.456 0.451 0.762 2.001<br />
The optimized annealing temperature (T m), BSA requirement of 0.4 mg mL 1 , number of alleles (N A), the observed and expected heterozygosity<br />
(H O and H E), polymorphic information content (PIC), and effective number of alleles (N E).<br />
number of alleles per locus (NA) using GenAlEx 6.2 (Peakall<br />
& Smouse, 2006). Departures from linkage disequilibrium<br />
and Hardy<strong>–</strong>Weinberg equilibrium (HWE) were tested (dememorization<br />
10 000, batches 100, iterations per batch<br />
5000) in GENEPOP 4.0 (Raymond & Rousset, 1995) and<br />
Bonferroni’s corrections were applied for multiple comparisons.<br />
To assess overall genetic differentiation at the population<br />
level, GenAlEx 6.2 calculated F ST using the infinite<br />
alleles model and RST using the stepwise mutation model<br />
through an AMOVA. Comparisons included SLS, BCC and<br />
recovered carcasses, and Belize and Florida.<br />
GENECAP (Wilberg & Dreher, 2004) calculated the unbiased<br />
probability of identity (PID), which is the probability that two<br />
individuals drawn at random from a population will have the<br />
same genotype at the assessed loci (Paetkau & Strobeck, 1994)<br />
and P(ID)sib, a related more conservative statistic for calculating<br />
PID among siblings (Evett & Weir, 1998). The program<br />
additionally searched for duplicate genotypes. GenAlEx 6.2<br />
was used to estimate the shadow effect and P(ID)observed to<br />
determine whether unbiased P(ID) or P(ID)sib is the more<br />
accurate probability (Mills et al., 2000; Waits, Luikart &<br />
Taberlet, 2001) following Hunter et al. (2010).<br />
The relationship between genetic and geographic distance<br />
was evaluated to test for isolation by distance within Belize<br />
and for Belize and Florida based on 10 000 randomizations<br />
in GenAlEx 6.2. BOTTLENECK 1.2.02 evaluated heterozygote<br />
excess under the two-phased model using the Wilcoxon signranked<br />
test (Piry, Luikart & Cornuet, 1999). The M-ratio<br />
test used M_P_Val.ex to measure the proportion of unoccupied<br />
allelic states, which is lowered during a population<br />
reduction (Garza & Williamson, 2001). Input values included<br />
a mutation rate of 5 10 4 and y=0.274; and the<br />
recommended values of Dg=3.5 and ps=0.9. Datasets<br />
using seven or more loci and the appropriate mutation<br />
model can be assumed to have experienced a reduction in<br />
Reduced Belize manatee dispersal and genetic variation<br />
population size with an Mo0.68. The coefficient of genetic<br />
relatedness, r xy (Queller & Goodnight, 1989), was used to<br />
test the genetic variability within Belize using 95% confidence<br />
intervals in GenAlEx 6.2. Intra-populational relatedness<br />
should be significantly higher in populations that have<br />
undergone bottlenecks or founding events.<br />
To test the distribution of rxy between all pairs of<br />
individuals, KINGROUP 2 simulated expected unrelated and<br />
full-sib pairwise relatedness (10 000 pairs) from observed<br />
allele frequencies and the rxy values were plotted to compare<br />
the distribution (Konovalov, Manning & Manning, 2004).<br />
A deviation from random expectation could result from<br />
nonrandom mating among related individuals and/or overrepresentation<br />
from a few families, suggesting an excess of<br />
inbred individuals in the population.<br />
The program STRUCTURE 2.3.2 was used to identify the<br />
genetic subdivision within Belize and the genetic relationship<br />
and ancestral source populations of Belize and Florida<br />
manatees (Pritchard, Stephens & Donnelly, 2000). The most<br />
probable number of populations, K, was determined by<br />
evaluating the likelihood of the posterior probability [L(K);<br />
Pritchard, Wen & Falush, 2007] and by calculating DK, an<br />
ad hoc quantity related to the change in posterior probabilities<br />
between runs of different K values (Evanno, Regnaut &<br />
Goudet, 2005) in STRUCTURE HARVESTER (Earl, 2009). Simulations<br />
were conducted using the admixture model of<br />
K=1<strong>–</strong>10 with 10 000 repetitions of MCMC, following the<br />
burn-in period of 10 000 iterations with and without a priori<br />
‘population’ information using the admixture model. The<br />
LOCPRIOR setting was used to detect cryptic structure by<br />
providing priors for the Bayesian assignment process based<br />
on the sample location. The LOCPRIOR model allows structure<br />
to be detected with lower levels of divergence and is not<br />
biased towards detecting structure when it is not present<br />
(Hubisz et al., 2009). Individual assignment success was<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London 595
Reduced Belize manatee dispersal and genetic variation M. E. Hunter et al.<br />
recorded as the likelihood of assignment (Q). Individuals<br />
were assigned to a cluster with QZ60%.<br />
Results<br />
Mitochondrial DNA analysis<br />
The Belize samples contained three haplotypes (A03, A04 and<br />
J01). The BCC samples were comprised of the A04 (n=4) and<br />
J01 (n=15) haplotypes. The SLS individuals contained the A04<br />
(n=52), J01 (n=22) and A03 (n=6) haplotypes. A04 (n=1)<br />
and J01 (n=1) were identified in the PL samples. Recovered<br />
carcasses contained the A04 (n=6) and J01 (n=6)haplotypes.<br />
MtDNA sequence divergence estimates were h=0.534 0.025<br />
and p=0.031 0.015. Within Belize, Tajima’s D was large and<br />
positive (D=4.118) consistent with a population bottleneck,<br />
however it was not significant (Po1.000); therefore the<br />
null hypothesis of selective neutrality cannot be rejected.<br />
Genetic differentiation estimates between BCC and SLS<br />
were FST=0.078 (Po0.045 0.015) and FST= 0.036<br />
(Po0.712 0.052). Twenty-eight variable sites and nucleotide<br />
substitutions (6.8%) were identified in the three haplotypes.<br />
Within Florida, one haplotype was observed (A01, p=0.000;<br />
unpubl. data; Garc´ ıa-Rodríguez et al., 1998). Belize and Florida<br />
mtDNA sequence divergence estimates were FST=0.626 (Po0.00001 0.0) and FST=0.302(Po0.00001 0.0).<br />
Microsatellite DNA analysis<br />
The 16 nuclear microsatellite markers had lower levels of<br />
variation [HE=0.453 (0.230<strong>–</strong>0.757); HO=0.456 (0.162<strong>–</strong><br />
0.761), N A=3.1 (2<strong>–</strong>5); Table 1] than the Florida population<br />
over 14 loci (K.P. Tucker et al., unpubl. data). In Belize,<br />
TmaKb60 and TmaSC13 had evidence of null alleles due to a<br />
heterozyogote deficiency. Deviation from HWE was found<br />
for TmaKb60 and TmaK01, even after a sequential Bonferroni<br />
adjustment (Po0.001). After 120 comparisons, linkage<br />
disequilibrium was observed between TmaE1 and TmaE14<br />
(Po0.05). FIS was 0.012 overall, suggesting slight inbreeding<br />
in the population. The error rate was determined by<br />
re-genotyping 11% of the individuals; no errors or duplicate<br />
genotypes were detected.<br />
Genetic differentiation between the BCC, SLS and recovered<br />
carcass samples was significant (Po0.05), although low<br />
(Table 2). The majority of private alleles were detected in<br />
SLS (n=5) as compared with BCC (n=1). The genetic<br />
distances showed a positive relationship with geographical<br />
distances between the BCC and SLS [r=0.20, (Po0.001)]<br />
and Belize and Florida [(r=0.48, (Po0.001)]. Pairwise F ST<br />
and R ST values for Belize and Florida were 0.141 and 0.082<br />
(Po0.001), respectively, supporting the subspecies distinction.<br />
Private alleles were detected for Florida (n=<strong>17</strong>) and<br />
Belize (n=1).<br />
The loci produced an unbiased P(ID) estimate of<br />
4.55E 08 and a P(ID)sib estimate of 3.60E 04, in which<br />
unrelated individuals could be identified in a population<br />
size of 4.7E+03 and 53, respectively, with a shadow<br />
effect of 1.00. Calculations in GenAlEx determined that<br />
P (ID)observed was nearly identical to that of unbiased P (ID),<br />
demonstrating that relatedness in this population is unlikely<br />
to affect match probability, because the population is<br />
o4.7E+03.<br />
The BCC and SLS Mantel test was positive and significant<br />
(Po0.001). All mutation model heterozygosity excess tests<br />
were significant (P 0.05), with low heterozygosity deficiency<br />
to excess ratios, SMM (4:12), IAM (2:14) and TPM (3:13).<br />
However, the mode shift test was L-shaped with no distortion<br />
of allele frequencies. The M-ratio was 0.54, indicative<br />
of a recent reduction in population size. The Belize intrapopulational<br />
pairwise relatedness was rxy=0.184 (95%<br />
CI=0.<strong>17</strong>4<strong>–</strong>0.185), and fell above of the 95% CI ( 0.022<strong>–</strong><br />
0.015) for permutations assuming random mating.<br />
The empirical distribution of pairwise relatedness values<br />
was low (mean=0.002, range=0.852<strong>–</strong>0.815), but not significantly<br />
different from the simulated distribution under<br />
random mating (mean=0.144; Mann<strong>–</strong>Whitney U-test,<br />
Table 2 Genetic differentiation and haplotype proportions of Belize manatees in the Southern Lagoon System (SLS), Belize City Cayes (BCC), and<br />
recovered carcasses (CAR)<br />
Geographic region<br />
(a) Genetic differentiation<br />
SLS BCC CAR<br />
SLS <strong>–</strong> 0.031 0.045<br />
BCC 0.041 <strong>–</strong> 0.025<br />
CAR 0.215 0.033 <strong>–</strong><br />
(b) Haplotype proportions<br />
F ST values (above diagonal) and R ST values (below diagonal) generated from a survey of 16 microsatellite loci. Statistically significant values are in<br />
italics (P 0.01).<br />
Pie charts indicating the haplotype proportions of the A03 (black), A04 (dark gray), and J01 (light gray) in the SLS (n=80), BCC (n=19) and CAR<br />
(n=12) samples.<br />
596<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
M. E. Hunter et al.<br />
P40.05) or the full-sib distribution (Mann<strong>–</strong>Whitney U-test,<br />
P40.05; Fig. 2). However, the full-sib pairwise relatedness<br />
indicated a full-sib detection threshold of 0.21 with 19.9% of<br />
the pairs potentially representing full-sibs under this criterion,<br />
suggesting a strong representation of a few families in<br />
the population.<br />
The Bayesian and resulting Evanno et al. (2005) analyses<br />
strongly supported two genetic clusters (P 1; Fig. 3a and c).<br />
Belize (QAVE=98.5%) and Florida (QAVE=98.6%) had no<br />
admixture and strong assignment of individuals to their<br />
respective populations. When Belize was analyzed alone<br />
without a priori ‘population’ information, one population<br />
was supported (P 1). Alternatively, under the LOCPRIOR<br />
model, K=2 was supported (P 1; Fig. 3b) and a single<br />
genetic cluster or more than two were rejected (P 0, K=1,<br />
3<strong>–</strong>10). DK=2 also had weak support as the largest modal<br />
Figure 2 Distributions of pairwise relatedness values r xy (Queller &<br />
Goodnight, 1989) for the Belize manatee population (gray bars).<br />
Simulated unrelated (solid line) and full-sib (dashed line) curves are<br />
based on 10 000 iterations.<br />
value (Fig. 3d). The two clusters corresponded closely with<br />
the sampled geographic regions, the BCC (95% correct<br />
assignment, QAVE=91.6%) and SLS (86% correct assignment,<br />
QAVE=81.0%).<br />
Discussion<br />
Genetic diversity<br />
Reduced Belize manatee dispersal and genetic variation<br />
Trichechus manatus manatus is believed to have inhabited<br />
the majority of the coastal habitat of Central and South<br />
America historically, although no specific census size estimates<br />
are available (Husar, 1978; IUCN, 2007). A few<br />
thousand manatees were estimated in the 18th century in<br />
Belize (Auil, 1998). Today, the subspecies is severely reduced,<br />
rare, or absent in many areas where manatees were<br />
previously identified as abundant (Husar, 1978). Current<br />
population estimates place the Antillean manatee subspecies<br />
at o2500 mature individuals (IUCN, 2007) and Belize at<br />
approximately 1000 individuals after recovering from a<br />
severe decline (Quintana-Rizzo & Reynolds III, 2007). The<br />
low levels of haplotype diversity, microsatellite heterozygosity<br />
and allelic variation found in the Belize manatee<br />
population are characteristic of small, isolated populations<br />
enduring bottlenecks or long-term persecution (Jamieson,<br />
Wallis & Briskie, 2006).<br />
Belize manatees have lower levels of diversity than other<br />
endangered or bottlenecked populations (Gebremedhin<br />
et al., 2009), including a koala population on French Island,<br />
Australia, which was colonized by three animals (HE=0.48,<br />
N A=3.8, 15 loci; Cristescu et al., 2009), the Wanglang giant<br />
panda (HE=0.68, NA=6.2, 13 loci; He et al., 2008), and the<br />
East African cheetah (HE=0.47, NA=3.55, 82 loci;<br />
Figure 3 Plot of mean and standard deviation of the posterior probabilities, L(K), among STRUCTURE 2.3.3 runs for each value of K (1<strong>–</strong>10) for (a)<br />
Florida and Belize and (b) Belize alone. Plot of DK versus K for (c) Florida and Belize and (d) Belize alone.<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London 597
Reduced Belize manatee dispersal and genetic variation M. E. Hunter et al.<br />
Driscoll et al., 2002). Examples of species with lower<br />
diversity than the Belize manatee include the Mediterranean<br />
monk seal (HE=0.41, NA=2.3, 15 loci; Paster et al., 2004)<br />
and the critically endangered Amur leopard (H E=0.37,<br />
NA=2.6, 25 loci; Uphyrkina et al., 2002).<br />
The Belize population was also found to have reduced<br />
variation compared with other marine mammal populations.<br />
Trichechus manatus latirostris currently has greater diversity<br />
and a quickly recovering large population, possibly owing to<br />
a shorter and/or less intense bottleneck period. Hunting of<br />
other marine mammal populations, such as the large cetaceans,<br />
occurred for a similar duration and intensity as for the<br />
Antillean manatee, and also resulted in the commercial<br />
extinction of many populations in the 1800s or 1900s.<br />
Populations with intense hunting and smaller historical<br />
sizes, such as the North Atlantic right whale (NARW), appear<br />
to have lost much of their diversity. The NARW was hunted<br />
for an extended period (11th<strong>–</strong>20th century) and was reduced<br />
from 12 000 individuals to the low hundreds, where it<br />
remains today (Waldick et al., 2002). The NARW genetic<br />
diversity reflects the extended and severe bottleneck the<br />
population has endured (NA=4.1, HE=0.46, 11 polymorphic<br />
loci). Alternatively, the related South Atlantic right<br />
whale population experienced less intense whaling that allowed<br />
the population to remain in the thousands and has<br />
higher genetic diversity (Waldick et al., 2002). Genetic studies<br />
have found large diversity, pre-whaling population estimates<br />
and recovered population sizes for fin (Berube et al., 1998),<br />
minke (Pastene, Goto & Kanda, 2009) and humpbacks whales<br />
(Garrigue et al., 2004), and suggest that the cessation of<br />
hunting in the early 20th century, allowed these populations<br />
to remain above 1000 individuals (Roman & Palumbi, 2003).<br />
Though it is reported as once being abundant in the<br />
Caribbean, T. m. manatus historical population sizes were<br />
not likely as large as some whale populations (Husar, 1977,<br />
1978). Further, a protracted period of severe hunting reduced<br />
the regional populations to o1000 individuals and in<br />
598<br />
Figure 4 Plot of posterior probability of assignment<br />
for individuals (vertical lines) based<br />
on Bayesian analysis of variation at 16 microsatellite<br />
loci. Individuals are grouped by population,<br />
(a) Belize and Florida and (b) the Belize<br />
City Cayes and Southern Lagoon System. Red<br />
indicates genetic cluster 1 and green indicates<br />
genetic cluster 2 in each plot.<br />
some cases to effective extinction (O’Shea & Salisbury, 1991;<br />
UNEP/CEP, 1995). This population contraction could have<br />
reduced the genetic diversity in Belize, which is below even<br />
that of the critically endangered NARW. However, lower<br />
dispersal and immigration capabilities in manatees may<br />
limit the exchange and development of inherent diversity in<br />
the taxa.<br />
The detected heterozygosity excess, mean pairwise values,<br />
and recent reduction in the population size suggests that a<br />
bottleneck/founder event, inbreeding, genetic drift and/or a<br />
reproductive skew has increased the relatedness among<br />
Belize manatees. Because diversity is considered necessary<br />
for adaptation to diseases and environmental changes,<br />
erosion of the currently low variation could negatively affect<br />
the population in the future (Reusch & Wood, 2007).<br />
Populations characterized by low diversity and past inbreeding<br />
are at risk for further losses and those not intensely<br />
monitored can become inbred with little warning (Frankham,<br />
1995; Bijlsma, Bundgaard & Boerema, 2000). Alternatively,<br />
inbreeding accumulating gradually in populations<br />
can provide the opportunity for natural selection to remove<br />
deleterious alleles. To date, there are no indications of<br />
decreased fitness due to inbreeding depression in the Belize<br />
population. However, further reduction of population size<br />
could increase inbreeding above the ‘inbreeding threshold,’<br />
resulting in loss of fitness and risk of extinction (Frankham,<br />
1995). Further monitoring could help to detect inbreeding<br />
depression or declines in population size (Schwartz, Luikart<br />
& Waples, 2007).<br />
The immigration of genetically distinct individuals can<br />
substantially reduce the deleterious effects of inbreeding<br />
(Frankham et al., 2002). However, the enhancement of<br />
diversity through this mechanism is limited since many<br />
manatee populations are small, isolated and movement is<br />
restricted by their dependence on fresh water. Manatees<br />
prefer near-shore habitat, have limited coastal ranges and<br />
infrequently move long distances, highlighting the need for<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
M. E. Hunter et al.<br />
detection and management of genetically isolated populations<br />
(Lefebvre et al., 2001; Vianna et al., 2006; Quintana-<br />
Rizzo & Reynolds III, 2007).<br />
Genetic population structure<br />
Genetic differentiation, private alleles and a strong discrepancy<br />
in haplotype proportions indicate restricted admixture<br />
between the BCC and SLS regions (Table 2). A<br />
significant Mantel test suggests a potential barrier to gene<br />
flow and isolation by distance. The BCC and SLS had<br />
disparate proportions of haplotypes, J01 (79%) and A04<br />
(65%), respectively, and greater mtDNA FST values than<br />
microsatellite FST values, indicating female philopatry and<br />
male-biased gene flow. These data also suggest a larger<br />
number of individuals moving from SLS to BCC. Limiting<br />
human activity in the Bar River and Belize River corridors<br />
might encourage movement and breeding between the SLS<br />
and BCC, leading to greater genetic diversity.<br />
Mitochondrial DNA haplotypes have been analyzed<br />
throughout the manatee range and the A03 haplotype has<br />
been found exclusively in the SLS (presented here; Vianna<br />
et al., 2006), suggesting that A03 females remain isolated,<br />
although further analyses of samples from the region are<br />
needed. Reduced movement has also been observed in manatees<br />
tracked using VHF, GPS or UHF transmitters. The<br />
majority of manatees tagged had strong site fidelity, many<br />
remaining within a 15 mile radius of the tagging location (Auil<br />
et al., 2007). The distinct habitats and lack of movement and<br />
admixture indicate that protection and preservation of the<br />
population as at least two genetic populations would best<br />
protect the discrete diversity in each region.<br />
The increasing incidence of boat strike near Belize City<br />
could quickly erode the BCC population size and diversity<br />
(O’Shea & Salisbury, 1991; Auil, 1998, 2007). The recovered<br />
carcasses from the Belize River mouth or to the south were<br />
not statistically different from the BCC samples. The identified<br />
cause of death in the carcasses was boat strike or<br />
undetermined due to decomposition, corresponding to the<br />
high boat traffic near the mouth of the Belize River.<br />
Phylogeographic relationships<br />
Recently, known Florida manatees have been documented<br />
traveling great distances from Florida, including, Cuba, the<br />
Bahamas and Rhode Island (Reid, 2000; Deutsch et al.,<br />
2003; Alvarez-Alemán, Powell & Beck, 2007). Florida<br />
manatees could potentially travel to Mexico along the Gulf<br />
of Mexico coast or by way of the Yucatán Peninsula. In fact,<br />
Mexico and Florida manatees share the A01 mitochondrial<br />
haplotype, possibly relating the two populations. However,<br />
although gene flow between Florida and Belize through<br />
Mexico is possible, the nuclear and mitochondrial DNA<br />
data do not support admixture, indicating possible barriers<br />
to gene flow (Domning & Hayek, 1986).<br />
Radio-tracking technologies have documented manatees<br />
traveling between Mexico and Belize (Auil et al., 2007;<br />
Morales-Vela et al., 2007). Although these movements have<br />
been tracked and the populations share two haplotypes, each<br />
country also contains a private haplotype (A01 and A03,<br />
respectively; Vianna et al., 2006), suggesting a paucity of<br />
female gene flow and incomplete genetic mixing. The limited<br />
relatedness of the three West Indian populations illustrates the<br />
historically low vagility and genetic exchange of the species.<br />
Future directions<br />
Quantifying the degree of migration and diversity in the<br />
Wider Caribbean using genetic tools is recommended to<br />
clarify the current and future role Belize may play in<br />
maintaining the Antillean subspecies in the region. Additionally,<br />
a more thorough examination of genetic connectedness<br />
among West Indian manatee populations<br />
throughout the range will assist in determining breeding<br />
populations and appropriate units of management for conservation.<br />
Further nuclear studies are anticipated to provide<br />
more detail on the genetic relationships. Identification of<br />
isolated or fragmented populations is important as lack of<br />
gene flow has significant deleterious effects on inbreeding,<br />
fitness and population sustainability (Frankham, 1995;<br />
Frankham et al., 2002). The growth and expansion of the<br />
Belize manatee population could also assist in the repopulation<br />
of small or functionally extinct Antillean manatee<br />
populations throughout the Wider Caribbean region<br />
(O’Shea & Salisbury, 1991; Bonde & Potter, 1995).<br />
Acknowledgments<br />
This work was conducted under the USFWS Wildlife Research<br />
Permit MA79<strong>17</strong>21, issued to the USGS, Sirenia Project.<br />
Funding for this project was provided by the USGS, and<br />
the University of Florida College of Veterinary Medicine<br />
Aquatic Animal Health Program. We would like to thank<br />
the Belize Manatee Project that provided access to genetic<br />
samples, logistical support and permission to collaboratively<br />
work under their Belize Forestry Department scientific collection<br />
and research permits CD/72/02 and CD/60/3/03 with<br />
subsequent amendments issued to N. Auil-Gomez and J.A.<br />
Powell. Additionally, Ginger Clark with the UF ICBR and<br />
Sean McCann provided critical technical assistance and<br />
Martha Sudholt and Susana Caballero provided helpful<br />
editorial recommendations. Any use of trade, product, or firm<br />
names is for descriptive purposes only and does not imply<br />
endorsement by the US Government.<br />
References<br />
Reduced Belize manatee dispersal and genetic variation<br />
Alvarez-Alemán, A., Powell, J.A. & Beck, C.A. (2007). First<br />
report of a Florida manatee documented on the North<br />
Coast of Cuba. Sirenews 47, 9<strong>–</strong>10.<br />
Auil, N.E. (1998). Belize Manatee Recovery Plan. UNDP/<br />
GEF Coastal Zone Management Project. Kingston: Caribbean<br />
Environment Programme.<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London 599
Reduced Belize manatee dispersal and genetic variation M. E. Hunter et al.<br />
Auil, N.E. (2004) Abundance and distribution trends of the West<br />
Indian manatee in the coastal zone of Belize: Implications for<br />
conservation. Master’s thesis, Texas A&M University.<br />
Auil, N.E., Powell, J.A., Bonde, R.K., Andrewin, K. &<br />
Galves, J. (2007). Belize Conservation Programme 10 year<br />
summary to Liz Claiborne Art Ortenberg Foundation. New<br />
York: Wildlife Trust.<br />
Beck, C. & Forrester, D.J. 1988. Helminths of the Florida<br />
manatee, Trichechus manatus latirostris, with a discussion<br />
and summary of the parasites of sirenians. J. Parasitol. 74,<br />
628<strong>–</strong>637.<br />
Berube, M., Aguilar, A., Dendanto, D., Larsen, F., Di Sciara,<br />
G.N., Sears, R., Sigurjonsson, J., Urban-R, J. & Palsboll,<br />
P.J. (1998). Population genetic structure of North Atlantic,<br />
Mediterranean Sea and Sea of Cortez fin whales, Balaenoptera<br />
physalus (Linnaeus <strong>17</strong>58): analysis of mitochondrial<br />
and nuclear loci. Mol. Ecol. 7, 585<strong>–</strong>599.<br />
Bijlsma, R., Bundgaard, J. & Boerema, A.C. (2000). Does<br />
inbreeding affect the extinction risk of small populations?<br />
Predictions from Drosophila. J. Evol. Biol. 13, 502<strong>–</strong>514.<br />
Blair, D. 2005. Family Opisthotrematidae Poche, 1926. In<br />
Keys to Trematoda: 401<strong>–</strong>406. Jones, A., Bray, R.A. &<br />
Gibson, D.I. (Eds). London: CABI <strong>International</strong> and The<br />
Natural History Museum.<br />
Bonde, R.K. & Potter, C.W. (1995). Manatee butchering sites<br />
in Port Honduras. Gainesville: USGS, Florida Integrated<br />
Science Center, Sirenia Project.<br />
Cristescu, R., Cahill, V., Sherwin, W.B., Handasyde, K.,<br />
Carlyon, K., Whisson, D., Herbert, C.A., Carlsson, B.L.J.,<br />
Wilton, A.N. & Cooper, D.W. (2009). Inbreeding and<br />
testicular abnormalities in a bottlenecked population of<br />
koalas (Phascolarctos cinereus). Wildl. Res. 36, 299<strong>–</strong>308.<br />
Deutsch, C.J., Reid, J.P., Bonde, R.K., Easton, D.E., Kochman,<br />
H.I. & O’Shea, T.J. (2003). Seasonal movements,<br />
migratory behavior, and site fidelity of West Indian manatees<br />
along the Atlantic Coast of the United States. Wildl.<br />
Monogr. 151, 1<strong>–</strong>77.<br />
Dixon, J.D., Oli, M.K., Wooten, M.C., Eason, T.H.,<br />
McCown, J.W. & Cunningham, M.W. (2007). Genetic<br />
consequences of habitat fragmentation and loss: the case of<br />
the Florida black bear (Ursus americanus floridanus). Conserv.<br />
Genet. 8, 455<strong>–</strong>464.<br />
Domning, D.P. & Hayek, L.C. (1986). Interspecific and<br />
intraspecific morphological variation in manatees (Sirenia:<br />
Trichechus). Mar. Mamm. Sci. 2, 87<strong>–</strong>144.<br />
Driscoll, C.A., Menotti-Raymond, M., Nelson, G., Goldstein,<br />
D. & O’Brien, S.J. (2002). Genomic microsatellites as<br />
evolutionary chronometers: a test in wild cats. Genome Res.<br />
12, 414<strong>–</strong>423.<br />
Earl, D. (2009). Structure Harvester v0.3, Available at http://<br />
users.soe.ucsc.edu/ dearl/software/struct_harvest<br />
(accessed 28 June 2010).<br />
Evanno, G., Regnaut, S. & Goudet, J. (2005). Detecting the<br />
number of clusters of individuals using the software<br />
STRUCTURE: a simulation study. Mol. Ecol. 14,<br />
2611<strong>–</strong>2620.<br />
600<br />
Evett, I.W. & Weir, B.S. (1998). Interpreting DNA evidence:<br />
statistical genetics for forensic scientists. Sunderland: Sinauer<br />
Associates Inc.<br />
Excoffier, L., Laval, G. & Schneider, S. (2005). Arlequin ver<br />
3.0: an integrated software package for population genetics<br />
data analysis. Evol. Bioinform. Online 1, 47<strong>–</strong>50.<br />
Frankham, R. (1995). Inbreeding and extinction <strong>–</strong> a threshold<br />
effect. Conserv. Biol. 9, 792<strong>–</strong>799.<br />
Frankham, R., Ballou, J.D. & Briscoe, D.A. (2002). Introduction<br />
to conservation genetics. Cambridge: Cambridge<br />
University Press.<br />
Garc ıa-Rodr ´ ıguez, ´ A.I., Bowen, B.W., Domning, D.P.,<br />
Mignucci-Giannoni, A.A., Marmontel, M., Montoya-Ospina,<br />
R.A., Morales-Vela, B., Rudin, M., Bonde, R.K. &<br />
McGuire, P.M. (1998). Phylogeography of the West Indian<br />
manatee (Trichechus manatus): how many populations and<br />
how many taxa? Mol. Ecol. 7, 1137<strong>–</strong>1149.<br />
Garc ıa-Rodr ´ ıguez, ´ A.I., Moraga-Amador, D., Farmerie,<br />
W.G., McGuire, P.M. & King, T.L. (2000). Isolation and<br />
characterization of microsatellite DNA markers in the<br />
Florida manatee (Trichechus manatus latirostris) and their<br />
application in selected sirenian species. Mol. Ecol. 9,<br />
2161<strong>–</strong>2163.<br />
Garrigue, C., Dodemont, R., Steel, D. & Baker, S. (2004).<br />
Organismal and ‘gametic’ capture-recapture using microsatellite<br />
genotyping confirm low abundance and reproductive<br />
autonomy of humpback whales on the wintering<br />
grounds of New Caledonia. Mar. Ecol. Prog. Ser. 274,<br />
251<strong>–</strong>262.<br />
Garza, J.C. & Williamson, E.G. (2001). Detection of reduction<br />
in population size using data from microsatellite loci.<br />
Mol. Ecol. 10, 305<strong>–</strong>318.<br />
Gebremedhin, B., Ficetola, G.F., Naderi, S., Rezaei, H.R.,<br />
Maudet, C., Rioux, D., Luikart, G., Flagstad, Ø., Thuiller,<br />
W. & Taberlet, P. (2009). Combining genetic and ecological<br />
data to assess the conservation status of the endangered<br />
Ethiopian walia ibex. Anim. Conserv. 12, 89<strong>–</strong>100.<br />
Gilpin, M.E. & Soul e, ´ M.E. (1986). Minimum viable populations:<br />
processes of extinction. In Conservation biology: the<br />
science of scarcity and diversity: 19<strong>–</strong>34. Soulé, M.E. (Ed.).<br />
Sunderland: Sinauer Associates.<br />
He, W., Lin, L., Shen, F.J., Zhang, W.P., Zhang, Z.H., King,<br />
E. & Yue, B.S. (2008). Genetic diversities of the giant<br />
panda (Ailuropoda melanoleuca) in Wanglang and Baoxing<br />
nature reserves. Conserv. Genet. 9, 1541<strong>–</strong>1546.<br />
Hoelzel, A.R., Halley, J., Obrien, S.J., Campagna, C., Arnbom,<br />
T., Leboeuf, B., Ralls, K. & Dover, G.A. (1993).<br />
Elephant seal genetic variation and the use of simulationmodels<br />
to investigate historical population bottlenecks.<br />
J. Hered. 84, 443<strong>–</strong>449.<br />
Hubisz, M.J., Falush, D., Stephens, M. & Pritchard, J.K.<br />
(2009). Inferring weak population structure with the assistance<br />
of sample group information. Mol. Ecol. Resour. 9,<br />
1322<strong>–</strong>1332.<br />
Hunter, M.K., Broderick, D., Ovenden, J.R., Tucker, K.P.,<br />
Bonde, R.K., McGuire, P.M. & Lanyon, J.M. (2010).<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
M. E. Hunter et al.<br />
Characterization of highly informative cross-species microsatellite<br />
panels for the Australian dugong (Dugong<br />
dugon) and Florida manatee (Trichechus manatus latirostris)<br />
including five novel primers. Mol. Ecol. Resour. 10,<br />
368<strong>–</strong>377.<br />
Husar, S. (1977) The West Indian manatee (Trichechus manatus).<br />
Wildlife Research Report 7, United States Department<br />
of the Interior. Washington, 22 pp.<br />
Husar, S. (1978). Mammal species: Trichechus manatus. Am.<br />
Soc. Mammal. 93, 1<strong>–</strong>5.<br />
IUCN. (2007). 2008 IUCN Red List of Threatened Species.<br />
Gland.<br />
Jamieson, I.G., Wallis, G.P. & Briskie, J.V. (2006). Inbreeding<br />
and endangered species management: is New Zealand<br />
out of step with the rest of the world? Conserv. Biol. 20,<br />
38<strong>–</strong>47.<br />
Jin, L. & Nei, M. (1990). Limitations of the evolutionary<br />
parsimony method of phylogenetic analysis. Mol. Biol.<br />
Evol. 7, 82<strong>–</strong>102.<br />
Kellogg, M.E. (2008) <strong>Sirenian</strong> conservation genetics and Florida<br />
manatee (Trichechus manatus latirostris) cytogenetics.<br />
Doctoral dissertation, University of Florida.<br />
Kimura, M. (1980). A simple method for estimating evolutionary<br />
rate of base substitution through comparative<br />
studies of nucleotide sequences. J. Mol. Evol. 16, 111<strong>–</strong>120.<br />
Konovalov, D.A., Manning, C. & Henshaw, M.T. (2004).<br />
KINGROUP: a program for pedigree relationship reconstruction<br />
and kin group assignments using genetic markers.<br />
Mol. Ecol. Notes 4, 779<strong>–</strong>782.<br />
Lefebvre, L.W., Marmontel, M., Reid, J.P., Rathbun, G.B. &<br />
Domning, D.P. (2001). Status and biogeography of the<br />
West Indian manatee. In Biogeography of the West Indies:<br />
patterns and perspectives. Second edn. 425<strong>–</strong>474. Woods,<br />
C.A. & Sergile, F.E. (Eds). Boca Raton: CRC Press.<br />
McCarthy, T.J. (1986). The gentle giants of Belize. Part II:<br />
distribution of manatees. Belize Audubon Soc. Bull. 18, 1<strong>–</strong>4.<br />
Mills, L., Citta, J., Lair, K., Schwartz, M. & Tallmon, D.<br />
(2000). Estimating animal abundance using noninvasive<br />
DNA sampling: promise and pitfalls. Ecol. Appl. 10,<br />
283<strong>–</strong>294.<br />
Morales-Vela, B., Padilla-Sald ıvar, ´ J.A., Reid, J. & Butler, S.<br />
(2007). First records of long-distance manatee movements<br />
between Mexico and Belize. Sirenews 47, 12<strong>–</strong>14.<br />
Nei, M. (1987). Genetic distance and molecular phylogeny. In<br />
Population genetics and fishery management: 193<strong>–</strong>223. Ryman,<br />
N. & Utter, F. (Eds). Seattle: University of Washington<br />
Press.<br />
O’Shea, T.J. & Salisbury, J.S. (1991). Belize <strong>–</strong> a last stronghold<br />
for the manatees in the Caribbean. Oryx 25, 156<strong>–</strong>164.<br />
Paetkau, D. & Strobeck, C. (1994). Microsatellite analysis of<br />
genetic variation in black bear populations. Mol. Ecol. 3,<br />
489<strong>–</strong>495.<br />
Pastene, L., Goto, M. & Kanda, N. (2009) Genetic analysis on<br />
stock structure in the Antarctic minke whales from the<br />
JARPA research area based on mitochondrial DNA and<br />
microsatellites. <strong>International</strong> Whaling Commission SC/<br />
Reduced Belize manatee dispersal and genetic variation<br />
D06/J9, p. 22. Report presented to the IWC Scientific<br />
Commitee Workshop.<br />
Paster, T., Garza, J., Allen, P., Amos, W. & Aguilar, A.<br />
(2004). Low genetic variability in the highly endangered<br />
Mediterranean monk seal. J. Hered. 95, 291<strong>–</strong>300.<br />
Pause, K.C., Nourisson, C., Clark, A., Kellogg, M.E., Bonde,<br />
R.K. & McGuire, P.M. (2007). Polymorphic microsatellite<br />
DNA markers for the Florida manatee (Trichechus manatus<br />
latirostris). Mol. Ecol. Notes 7, 1073<strong>–</strong>1076.<br />
Peakall, R. & Smouse, P.E. (2006). GENALEX 6: genetic<br />
analysis in Excel. Population genetic software for teaching<br />
and research. Mol. Ecol. Notes 6, 288<strong>–</strong>295.<br />
Piry, S., Luikart, G. & Cornuet, J.M. (1999). BOTTLE-<br />
NECK: a computer program for detecting recent reductions<br />
in the effective population size using allele frequency<br />
data. J. Hered. 90, 502<strong>–</strong>503.<br />
Pritchard, J.K., Stephens, M. & Donnelly, P. (2000). Inference<br />
of population structure using multilocus genotype<br />
data. Genetics 155, 945<strong>–</strong>959.<br />
Pritchard, J.K., Wen, X. & Falush, D. (2007) Documentation<br />
for Structure Software:Version 2.3. Available at http://<br />
pritch.bsd.uchicago.edu/software. html (accessed 28 June<br />
2010).<br />
Queller, D.C. & Goodnight, G.K. (1989). Estimating relatedness<br />
using genetic markers. Evolution 43, 258<strong>–</strong>275.<br />
Quintana-Rizzo, E. & Reynolds, J. III (2007) Regional management<br />
plan for the West Indian manatee (Trichechus<br />
manatus). CEP Technical Report. Caribbean Environment<br />
Programme, United Nations Environment Programme,<br />
Gosier, Guadeloupe, France.<br />
Raymond, M. & Rousset, F. (1995). GENEPOP (version 1.2):<br />
population genetics software for exact tests and ecumenicism.<br />
J. Hered. 86, 248<strong>–</strong>249.<br />
Reid, J.P. (2000). Florida manatee now resident in the<br />
Bahamas. Sire News 33, 7<strong>–</strong>8.<br />
Reusch, T.B.H. & Wood, T.E. (2007). Molecular ecology of<br />
global change. Mol. Ecol. 16, 3973<strong>–</strong>3992.<br />
Roman, J. & Palumbi, S.R. (2003). Whales before whaling in<br />
the North Atlantic. Science 301, 508<strong>–</strong>510.<br />
Schwartz, M.K., Luikart, G. & Waples, R.S. (2007). Genetic<br />
monitoring as a promising tool for conservation and<br />
management. Trends Ecol. Evol. 22, 25<strong>–</strong>33.<br />
Tajima, F. (1993). Measurement of DNA polymorphism. In<br />
Mechanisms of molecular evolution. Introduction to molecular<br />
paleopopulation biology: 37<strong>–</strong>59.Takahata, N. & Clark,<br />
A.G. (Eds). Tokyo/Sunderland: Japan Scientific Societies<br />
Press/Sinauer Associates Inc.<br />
Thornton, C., Shanahan, M. & Williams, J. (2003). From<br />
wetlands to wastelands: impacts of shrimp farming. London:<br />
Environmental Justice Foundation.<br />
UNEP/CEP. (1995) United Nations Environmental Programme<br />
- Caribbean Environment Programme. Technical<br />
Report No. 35. Regional Management Plan for the West<br />
Indian Manatee, Trichechus manatus.<br />
Uphyrkina, O., Miquelle, D., Quigley, H., Driscoll, C. &<br />
O’Brien, S.J. (2002). Conservation genetics of the Far<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London 601
Reduced Belize manatee dispersal and genetic variation M. E. Hunter et al.<br />
Eastern leopard (Panthera pardus orientalis). J. Hered. 93,<br />
303<strong>–</strong>311.<br />
Vianna, J.A., Bonde, R.K., Caballero, S., Giraldo, J.P., Lima,<br />
R.P., Clark, A., Marmontel, M., Morales-Vela, B., De<br />
Souza, M.J., Parr, L., Rodriguez-Lopez, M.A., Mignucci-<br />
Giannoni, A.A., Powell, J.A. & Santos, F.R. (2006).<br />
Phylogeography, phylogeny and hybridization in trichechid<br />
sirenians: implications for manatee conservation. Mol.<br />
Ecol. 15, 433<strong>–</strong>447.<br />
Waits, L.P., Luikart, G. & Taberlet, P. (2001). Estimating the<br />
probability of identity among genotypes in natural<br />
602<br />
populations: cautions and guidelines. Mol. Ecol. 10,<br />
249<strong>–</strong>256.<br />
Waldick, R.C., Kraus, S.S., Brown, M. & White, B.N. (2002).<br />
Evaluating the effects of historic bottleneck events:<br />
an assessment of microsatellite variability in the<br />
endangered, North Atlantic right whale. Mol. Ecol. 11,<br />
2241<strong>–</strong>2250.<br />
Wilberg, M.J. & Dreher, B.P. (2004). GENECAP: a program<br />
for analysis of multilocus genotype data for non-invasive<br />
sampling and capture-recapture population estimation.<br />
Mol. Ecol. Notes 4, 783<strong>–</strong>785.<br />
Animal Conservation 13 (2010) 592<strong>–</strong>602 c 2010 The Authors. Animal Conservation c 2010 The Zoological Society of London
Tourism and wildlife habituation: Reduced population fitness<br />
or cessation of impact?<br />
J.E.S. Higham *, E.J. Shelton<br />
Department of Tourism, School of Business, University of Otago, P.O. Box 56, Dunedin, New Zealand<br />
article info<br />
Article history:<br />
Received <strong>17</strong> August 2010<br />
Accepted 15 December 2010<br />
Keywords:<br />
Tourism<br />
Wildlife<br />
Habituation<br />
Stimulus control<br />
Tolerance<br />
Sensitisation<br />
Management<br />
1. Introduction<br />
abstract<br />
The concept of wildlife habituation traditionally has been treated<br />
uncritically within the field of nature-based tourism (Edington &<br />
Edington, 1986; Shelton & Higham, 2007). Typically the concept of<br />
habituation is accepted as a unitary phenomenon that is a negative<br />
possible consequence of tourist interactions with animals in the wild<br />
(Higginbottom, 2004; Newsome et al., 2005; Shackley, 1996). This<br />
global and stable behavioural descriptor, habituation, is an unhelpful<br />
way to formulate most observed lack-of-wildlife-response to<br />
human interactions. A more fine-grained behavioural and temporal<br />
approach to understanding wildlife habituation may be of considerable<br />
relevance to the optimal management of tourist interactions<br />
with wild animals.<br />
This paper attempts two points of departure from established<br />
discourses on wildlife habituation. The first challenges the contrast<br />
between the use of habituation in zoological studies (e.g. to mitigate<br />
researcher impacts), and the treatment of habituation in tourism<br />
management as entirely undesirable (Seddon, Ellenberg, & van<br />
Heezik, in press). Second, it asks if in some instances habituation<br />
* Corresponding author.<br />
E-mail addresses: james.higham@otago.ac.nz, jhigham@business.otago.ac.nz<br />
(J.E.S. Higham), eric.shelton@otago.ac.nz (E.J. Shelton).<br />
0261-5<strong>17</strong>7/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.<br />
doi:10.1016/j.tourman.2010.12.006<br />
Tourism Management 32 (2011) 1290e1298<br />
Contents lists available at ScienceDirect<br />
Tourism Management<br />
journal homepage: www.elsevier.com/locate/tourman<br />
Habituation typically is viewed as a negative consequence of human interactions with wildlife<br />
(Higginbottom, 2004; Newsome, Dowling, & Moore, 2005; Shackley, 1996). While animal habituation<br />
commonly is used in the laboratory and field-based zoology studies, attempts to consider deliberate<br />
habituation specifically in a tourism management context (Shelton, Higham, & Seddon, 2004) has been<br />
received unsympathetically by biological scientists and wildlife managers on the grounds that habituation,<br />
by definition, is undesirable. This paper puts forward the case that the global and stable behavioural<br />
descriptor, habituation, is not the most useful way to formulate most observed lack-of-wildlife-response to<br />
visitor approach and observation. It presents an applied behaviour analysis of wildlife habituation that is<br />
situated within learning theory. This analysis differentiates between avoidance/approach behaviours,<br />
tolerance, habituation and sensitisation. This provides a formulative framework for humanewildlife<br />
interactions, that is then considered specifically in terms of tourism businesses seeking to provide<br />
sustainable visitor interactions with wild animals. A tourism management model derived from this critique<br />
of habituation is presented and discussed.<br />
Ó 2011 Elsevier Ltd. All rights reserved.<br />
may be deployed as a deliberate tourism management strategy in<br />
pursuit of cessation of visitor impacts upon focal animals (Nisbet,<br />
2000). These discussions draw upon three sources; thousands of<br />
unstructured personal observations on the part of the authors<br />
derived from employment as wildlife tour guides in New Zealand;<br />
extensive in situ personal observations of humanewildlife encounters<br />
in New Zealand including New Zealand’s sub-Antarctic Islands,<br />
and Ross Sea Dependency (Antarctica); and key empirical literature<br />
(e.g. Bejder, Samuels, Whitehead, Finn, & Allen, 2009; Ellenberg,<br />
Mattern, Seddon, & Jorquera, 2006; Johns, 1996; Knight, 2009;<br />
Romero & Wikelsi, 2002). The discussions that follow draw upon<br />
person observations unless indicated otherwise through reference<br />
citation.<br />
2. Ecotourism and wildlife encounters<br />
Wildlife viewing, once the domain of dedicated enthusiasts, or<br />
‘specialists’ (Duffus & Dearden, 1990), has moved into the mainstream<br />
of commercial tourism (Knight, 2009). With this has come<br />
a proliferation and diversification of opportunities to encounter<br />
wildlife (Higham, Lusseau, & Hendry, 2008). This course of development<br />
has occurred despite an inescapable tension. Knight (2009:<br />
p. 167) identifies a fundamental contradiction in wildlife viewing in<br />
that “wild animals are generally human-averse; they avoid humans
and respond to human encounters by fleeing and retreating to<br />
cover”. A gradual reduction of such avoidance commonly is labelled<br />
habituation. However humanewildlife encounters represent<br />
a complexity of interaction stimuli that render the unitary term<br />
habituation problematic (Bejder et al., 2009). With this point in<br />
mind, it is surprising that wildlife habituation continues to be<br />
treated uncritically within the context of nature-based tourism and<br />
sustainable tourism management.<br />
Successful commercial wildlife viewing requires that visitors are<br />
concentrated in well-defined locations where interactions with<br />
wild animals are predictable and constant (Whittaker, 1997).<br />
Usually, viewing wildlife takes place where sightings are most<br />
consistent, focal animals can be viewed in abundance or, where<br />
spectacular behaviours may be predictably observed and experienced<br />
(e.g. courtship and socialising behaviours). The critical nature<br />
of these sites, in terms of site ecology and wildlife behaviours, raises<br />
two important points; (1) That the behavioural state of wild animals<br />
varies significantly over time (e.g. across times of day, through<br />
different stages of the breeding cycle and across life course) and (2)<br />
that animal responses to external stimuli (e.g. the presence of<br />
tourists) is likely to vary over time, as influenced by these temporal<br />
determinants.<br />
These points raise the challenge of understanding and<br />
responding to the potential adverse impacts of humanewildlife<br />
interactions, not only for the sake of the focal animals, but also from<br />
an experiential standpoint with the aim of (1) reducing avoidance,<br />
flight or retreat responses and (2) mitigating adverse impacts on<br />
animals that may otherwise discontinue critical, site-specific<br />
behaviours e.g. by instigating site abandonment. Furthermore,<br />
increasingly tourists seek to be assured that their mere presence at<br />
wildlife viewing sites (and their associated behaviours) are not to<br />
the detriment of the animals that they seek to experience first hand,<br />
either in terms of the welfare of individual animals or wider population<br />
fitness (Higham & Lusseau, 2004; Muloin, 1998).<br />
Such concerns on the part of tourists (and site managers) are well<br />
founded. Knight (2009) critiques the dichotomy between wildlife<br />
viewing and wildlife hunting as non-consumptive and consumptive<br />
respectively, highlighting that viewing and hunting wildlife have<br />
some fundamental similarities. First, both engage in locating and<br />
identifying target wild animals, which are generally “wary of human<br />
presence and reluctant to expose themselves to human eyes”<br />
(Knight, 2009: p. 169). Such wariness may be expressed through<br />
anti-predatory adaptations, both physical (e.g. camouflage colouration)<br />
and behavioural (e.g. concealment) to avoid detection. The<br />
directed, intensive and sustained tourist gaze offers a second<br />
parallel with hunting (and predation more generally), which also is<br />
likely to trigger alarm and anti-predatory responses in individual<br />
animals or groups of animals.<br />
Tourist satisfaction is commonly associated with close-up,<br />
unconstrained and prolonged interactions with wild animals, the<br />
experience of critical behaviours (e.g. hunting, feeding, socialising<br />
and courtship) and, in some cases, immediate proximity extending<br />
to touch (e.g. Muloin, 1998). Although there are some exceptions to<br />
this rule (Orams, 2000), the importance of managing ‘humane<br />
wildlife viewing interactions’ (as opposed to ‘non-consumptive<br />
viewing’ e given the consumptive nature of pursuit, intensive gaze<br />
and proximal interaction) is evident. Knight (2009: p. <strong>17</strong>3) proposes<br />
that “there appear to be three main ways in which wild animals can<br />
be made available for human viewing: capture and confinement;<br />
habituation and attraction.”<br />
The first, capture and confinement, brings with it behavioural<br />
diminution associated with wild animals being physically removed<br />
from their natural ecosystem. Also, in many cases, the capture and<br />
confinement of ‘charismatic’ wild animals is illegal. The last, attraction,<br />
through interventions such as provisioning (i.e. feeding) is<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298 1291<br />
equally unpalatable due to its association with manipulated wildlife<br />
distributions, compromised feeding (e.g. predatoreprey) relationships,<br />
increased probability of aggressive behaviour (both towards<br />
animals and people) and consequential diminished behaviours<br />
(Orams, 1995). Attraction of animals through supplemental feeding,<br />
while affording close and enduring viewing opportunities, also<br />
compromises the notion of wild, creating indistinct dividing lines<br />
between animals that are wild and those that are confined, tamed or<br />
domesticated (Knight, 2009). Of the three approaches to making<br />
wild animals viewable, this leaves habituation.<br />
Habituation has been defined as “a decrease in the strength of<br />
a response after repeated presentations of a stimulus that elicits that<br />
response” (Mazur, 2006). As such, habituation typically is viewed as<br />
a negative consequence of human interactions with wildlife due to<br />
the likely consequential reduction of population fitness arising from<br />
reduced danger flight response. However, habituation has been<br />
actively applied by zoological scientists in their field studies (i.e. in<br />
the wild), having been pioneered by primatologists such as Jane<br />
Goodall (Tanzania) working with chimpanzees (Pan troglodytes) and<br />
Dian Fossey (Rwanda) with gorillas (Gorilla gorilla) (Knight, 2009).<br />
Interestingly, Fossey’s habituation of Gorillas made possible the<br />
development of Gorilla tourism as an alternative to Gorilla poaching<br />
(Shackley, 1995). Yet habituation has never been treated critically in<br />
the specific context of sustainable commercial wildlife viewing<br />
(Nisbet, 2000).<br />
The relationship between tourists and wild animals is extremely<br />
complex (Bejder et al., 2009). Duffus and Dearden (1990) contend<br />
that this complexity arises from the interplay of three key components<br />
of the wildlife experience; site users, focal wildlife species<br />
(both individual animals and local animal populations) and the wider<br />
ecology of the viewing site. Given these components, considerations<br />
of habituation in ecotourism can be contemplated only in terms of<br />
site users (e.g. visitor management, guiding practice), the focal<br />
wildlife population (i.e. species-level variables and characteristics of<br />
individual animals) and the wider ecology of the focal species (e.g.<br />
spatio-temporal ecology) (Higham et al., 2008). Duffus and Dearden<br />
(1990) highlight also the temporal dimension in so far as tourist<br />
interactions with wildlife animals vary within diurnal, seasonal and<br />
life course timeframes, a point that has been inadequately addressed<br />
in academic discussions of habituation (Bejder et al., 2009).<br />
Following Duffus and Dearden (1990) then, some species of wild<br />
animals, and some individual animals, demonstrate a well-established<br />
response of inquisitiveness to humans or signs of human<br />
activity. Others are intensely private, remain concealed from<br />
onlookers, and flee readily from human interaction. In terms of<br />
inquisitiveness and active interaction, gulls are a ubiquitous example<br />
but rarely are gulls the focal species of interest to ecotourism operators<br />
and their visitors (Shelton & Higham, 2007). Useful approaches<br />
to the task of analysing the behaviour of species that are of tourist<br />
interest should, however, prove to be of considerable value.<br />
Applied behaviour analysis, based on the notion of stimulus<br />
control and situated within learning theory, is well suited to provide<br />
a formulative framework for those humanewildlife interactions that<br />
are of interest to commercial tourism operators. In this respect, the<br />
term habituation has been described as any situation where wildlife<br />
come to tolerate the presence of humans without any obvious signs<br />
of physiological or behavioural response (Shackley, 1996). Under this<br />
definition, habituation may be considered a mitigation or cessation<br />
of impact. The use of the term habituation in tourism and recreation<br />
has been applied also when animals approach people, or scavenge<br />
for food (Newsome et al., 2005). Park managers in North America<br />
attempt to educate the public not to engage in behaviours that will<br />
merge habituation and approach-for-food, particularly in the case of<br />
deer (The National Park Service, 2006a, 2006b) and bears (Davis,<br />
Wellwood, & Ciarniello, 2002).
1292<br />
To be as useful as possible, habituation must be distinguished<br />
from foraging or increased approach behaviour, even though the<br />
three phenomena regularly occur together (Davis et al., 2002). The<br />
starting point for such an undertaking is a critical behavioural<br />
analysis of wildlife habituation. In response to this need, a critical<br />
discussion of animal responses to human presence in presented, in<br />
order to achieve a more fine-grained understanding of habituation<br />
in all of its complexity. This critique provides a platform from which<br />
then to consider habituation within the context of sustainable<br />
wildlife tourism management.<br />
3. A behavioural formulation of wildlife habituation<br />
A behavioural formulation of wildlife habituation should begin<br />
by acknowledging the poor fit between observable animal behaviour<br />
and internal state (Ellenberg et al., 2006). In other words, the<br />
apparent tolerance of some wildlife species to approach and<br />
observation may not necessarily mean that these wild animals are<br />
not being impacted. Bejder et al. (2009) explain that wildlife tolerance<br />
of human stimuli may arise from various factors including:<br />
(1) Displacement: e.g. less tolerant individual animals may be<br />
displaced, resulting in a bias towards more tolerant animals<br />
that remain at a given site.<br />
(2) Physiology: e.g. reduced responsiveness to human stimuli due<br />
to physiological impairment.<br />
(3) Ecology: e.g. lack of suitable adjacent habitat to which animals<br />
may otherwise relocate.<br />
This poor fit has been demonstrated in the case of penguins<br />
(Ellenberg et al., 2006) where artificial eggs placed under incubating<br />
birds record elevated heart rate in birds that outwardly appeared to<br />
be unaffected when approached (Nimon, Schroter, & Stonehouse,<br />
1995). An understanding of habituation must acknowledge both<br />
the presence and absence of specific behaviours, and that impacts of<br />
significance at both the individual and species level may arise from<br />
both behavioural classes. The presence or absence of behaviours<br />
may vary at different scales of analysis, from individual animals to<br />
Table 1<br />
A critical behavioural formulation of the unitary term habituation.<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298<br />
entire populations (Bejder et al., 2006; Lusseau, 2003; Shelton &<br />
Higham, 2007), and is dynamic over time. Following Bejder et al.<br />
(2009), a behavioural formulation of habituation in the specific<br />
context of wildlife-based tourism may be presented in four parts; 1.<br />
Avoidance/approach behaviours, 2. Tolerance, 3. Habituation and 4.<br />
Sensitisation (Table 1).<br />
3.1. Avoidance/approach behaviour<br />
An important starting point for any critical discussion of habituation<br />
is an understanding of the complexity of avoidance and<br />
approach behaviours. The avoidance behaviours of cetaceans when<br />
engaged in interactions with tourist vessels have been particularly<br />
well researched (Baker, Perry, & Vequist, 1988; Corkeron, 1995;<br />
Lusseau, 2003; Salden, 1988; ). Although responses to boat traffic<br />
vary between species, some show signs of active avoidance, ranging<br />
from altered movement patterns (Bejder, Dawson, & Harraway,1999;<br />
Campagna, Rivarola, Greene, & Tagliorette, 1995; Edds & MacFarlane,<br />
1987; Nowacek, Wells, & Solow, 2001; Salvado, Kleiber, & Dizon,<br />
1992), increases in dive intervals (Baker & Herman, 1989; Baker<br />
et al., 1988; Blane, 1990; Janik & Thompson, 1996; MacGibbon,<br />
1991) and increases in swimming speed (Blane & Jaakson, 1995;<br />
Williams, Trites, & Bain, 2002). Avoidance behaviours in Bottlenose<br />
dolphins (Tursiops truncatus) have been shown to vary at an individual<br />
level (Lusseau, 2003), where male animals show a greater<br />
propensity to avoid boat traffic than females with calves (see Higham<br />
& Lusseau, 2004).<br />
These studies present clear evidence of animal avoidance<br />
behaviours, not only as a result of the presence of boats, but also<br />
arising from the manoeuvring of boats, including sudden changes<br />
in vessel speed and rapid approaches (Constantine, 2001; Gordon,<br />
Leaper, Hartley, & Chappell, 1992; MacGibbon, 1991). In 2006, the<br />
<strong>International</strong> Whaling Commission reached agreement that “there<br />
is compelling evidence that the fitness of individual odontocetes<br />
repeatedly exposed to whale-watching vessel traffic can be<br />
compromised and that this can lead to population-level effects”<br />
(IWC, 2006). This consensus has been reached in light of recent<br />
studies that suggest avoidance responses from cetaceans when<br />
Concept Variables Cases<br />
1. Avoidance/approach Avoidance response (e.g. pre-emptive avoidance) Male Bottlenose dolphins (Tursiops truncatus)<br />
behaviour<br />
Response behaviours (e.g. generalised vigilance, concealment,<br />
discontinuation of critical behaviours)<br />
Male Bottlenose dolphins (Tursiops truncatus)<br />
Stimulus control (i.e. group and individual behaviours) Adele penguins (Pygoscelie adeliae); Little penguins<br />
(Eudyptula minor); Bottlenose dolphins (Tursiops truncatus)<br />
Exploratory approach behaviour Elephant seal (Mirounga leonine)<br />
Opportunistic approach behaviour Buller’s mollymawk (Diomedea bulleri); Giant petrel<br />
(Macronectes giganteus)<br />
Discriminative stimulus Fantail/piwakawaka (Rhipidura fuliginosa); Southern Royal albatross<br />
(Diomedea epomophora)<br />
2. Tolerance Intrinsic tolerance Little penguins (Eudyptula minor); Spotted shags<br />
(Stictocarbo punctatus); Stewart Island shags<br />
(Leucocarbo chalconotus); Little shags (Phalacrocorax melanoleucos)<br />
Selective tolerance New Zealand sea lion (Phocarctos hookeri)<br />
Seasonal tolerance New Zealand wood pigeon/kereru (Hemiphaga novaseelandiae)<br />
3. Habituation Approach habituation New Zealand dabchick (Poliocephalus rufopectus)<br />
Spatio-temporally bound habituation Colonial breeding female New Zealand fur seal (Arctocephalus forsteri).<br />
Socially acquired absence of avoidance behaviour New Zealand fur seal (Arctocephalus forsteri) neonates<br />
Socially acquired habituation Female New Zealand fur seal (Arctocephalus forsteri)<br />
(e.g. new colony recruits)<br />
Reduced physiological arousal Northern Royal albatross (Diomedea epomophora sanfordi)<br />
Reduced avoidance behaviour Yellow-eyed penguin (Megadyptes antipodes)<br />
4. Sensitisation Sensitisation Stoats(Mustela erminea) (i.e. in the laboratory)<br />
Paradoxical sensitisation Galapagos marine iguana (Mblyrhynchus cristatus)
tourist vessels are present, and that those responses disrupt the<br />
behavioural budgets of affected animals. Repeated disturbance can<br />
then lead to displacement from preferred habitat and reduced<br />
fitness at the population level (Bejder et al., 2006) as opposed to<br />
any evidence of habituation.<br />
3.1.1. Stimulus control<br />
The presence or absence of avoidance/approach behaviours<br />
might usefully begin with the treatment of stimulus control. This<br />
term may be used to describe naturally-occurring wild animal<br />
behaviours that are resistant to modification. Adele penguins<br />
(Pygoscelie adeliae) launching from an Antarctic ice floe engage in<br />
grouping and rushing behaviours, presumably to reduce the likelihood<br />
of fatal attack by predators such as Leopard seals (Hydrurga<br />
leptonyx). The same stimulus control applies to Little penguins<br />
(Eudyptula minor) at tourist sites such as the Philip Island Penguin<br />
Parade (Victoria, Australia) and Oamaru Blue Penguin Colony (New<br />
Zealand). In this instance proximity to the landing site and<br />
grouping induces a rushing response which is unchanged despite<br />
millions of unchallenged landings over many generations of birds<br />
(Houston & Russell, 2000). This failure to habituate to coastline<br />
proximity has no obvious evolutionary cost to the species. The<br />
lighting, public address systems and grandstands full of visitors at<br />
these sites do not deter the birds from landing where they have, in<br />
recent history, always landed. This lack of avoidance response has<br />
not changed over time and is uniquely characteristic of this species<br />
of penguin which does not demonstrate habituation of approach or<br />
avoidance behaviour.<br />
Another relevant case involves cetaceans. Lusseau’s (2003)<br />
study of Bottlenose dolphins (Tursiops truncatus) in Doubtful<br />
Sound (Fiordland, New Zealand) found that these marine mammals<br />
entertained visitors on tour boats with their bowriding and leaping<br />
out of the water, such that it might be assumed that dolphins<br />
engage in this behaviour by choice. Power boats, though, act as<br />
a stimulus for dolphins not to rest; that is, bowriding and leaping<br />
are more likely to be under stimulus control and therefore not the<br />
result of choice. There are clear potential negative impacts of<br />
decreased resting on this species, including a disrupted energy<br />
budget and likely compromised reproductive success (Lusseau &<br />
Higham, 2004).<br />
Repeated exposure to tourist vessels does not seem to reduce<br />
the strength of the dolphins’ approach responses and this situation<br />
clearly demonstrates the difficulties inherent in referring to stable<br />
or increasing approach behaviour as habituation. Lusseau and<br />
Higham (2004) suggest the spatial segregation of non-permitted<br />
boats and Bottlenose dolphins through multi-level zoning. An<br />
additional management strategy could, in this case, include<br />
inducing habituation via stimulus mitigation; that is, reducing the<br />
dolphin’s responsiveness to the presence of tourist vessels through<br />
the close management of key stimuli (e.g. boat speed and noise) so<br />
as to counter the stimulus control that these animals appear to act<br />
under.<br />
3.1.2. Exploratory approach behaviour<br />
A further manifestation of approach/avoidance behaviour, with<br />
an apparent temporal (i.e. life course) dimension may be termed<br />
exploratory approach behaviour. Elephant seals (Mirounga leonine)<br />
on sub-Antarctic Campbell Island may be exposed fleetingly to two<br />
or three groups of visitors per year at North West Bay (and not at all<br />
elsewhere on Campbell Island). Juvenile Elephant seals will<br />
approach and investigate tour parties at close range. The lack of<br />
similar interest on the part of adult Elephant seals very probably<br />
reflects a decreasing exploratory behaviour as a consequence of<br />
maturation rather than through habituation (i.e. repeated exposure<br />
to humans) where tourist encounters are so uncommon.<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298 1293<br />
This change in responsiveness raises important questions<br />
around the naturally-occurring contingencies and physiological<br />
processes that operate to reduce exploratory behaviour during<br />
maturation from neonate through juvenile to adult. Repeated nonconsequential<br />
exposure to a range of human behaviours should be<br />
situated somewhere in this process. On mainland New Zealand,<br />
where tourist encounters are more common, adult Elephant seals<br />
will rest ashore, selecting where possible confined and inaccessible<br />
resting sites, and demonstrating complete indifference to approach,<br />
based on a confidence of remaining unmolested. This indifference<br />
may be understood as reduced exploratory approach behaviour<br />
rather than habituation.<br />
3.1.3. Opportunistic approach behaviour<br />
The term opportunistic approach behaviour may be used to<br />
address approach stimuli linked to an individual’s energy budget.<br />
Offshore boat-based tourist encounters with some seabird species<br />
such as Buller’s mollymawk (Diomedea cauta eremita) may result in<br />
close but fleeting observations of birds derived from what may be<br />
descried as ‘investigatory fly past’. Opportunistic approach behaviour<br />
may be a generalisation from the birds’ habit of following fishing<br />
boats whose crews routinely engage in gutting offshore. Giant<br />
petrels (Macronectes giganteus) also will engage in this behaviour.<br />
Distinct from provisioning and scavenging, these birds approach in<br />
response to a specific stimulus. Despite daily exposure to the tourist<br />
boat, and never having been fed, this exploratory approach behaviour<br />
persists over time. Due no doubt to a failure to distinguish<br />
between tourist vessels and fishing boats, and with so little energy<br />
required to complete an investigation, these birds, as an optimal<br />
feeding strategy, casually and routinely approach every boat.<br />
3.1.4. Discriminative stimuli<br />
Related to this, approach behaviours may be further differentiated<br />
from habituation via an understanding of discriminative<br />
stimuli. The behaviour of fantails/piwakawaka (Rhipidura fuliginosa)<br />
is typically misleading. Visitors often encounter the situation where<br />
they are walking along a bush track and suddenly numerous<br />
fantails appear and flutter tantalisingly close-by. Many may incorrectly<br />
assume that it is their presence that has attracted the birds<br />
which are attracted to human presence. Rather, the birds are<br />
attracted to small flies that have risen from the leaf litter as a result<br />
of human disturbance. The attraction to visitors in the bush is<br />
indirect. People in this case act as a discriminative stimulus for<br />
a likely disturbed forest floor, and consequently available prey.<br />
Similarly, Southern Royal albatrosses treat private boats as<br />
discriminative stimuli for fish being scaled and gutted and land in<br />
their wakes. These examples only serve to further confirm the poor<br />
fit between observable behaviour and internal state in terms of<br />
understanding animal behaviours in the wild.<br />
3.2. Tolerance<br />
3.2.1. Intrinsic tolerance<br />
Within investigations of humanewildlife interactions the<br />
apparent tolerance of wild animals to human approach has<br />
remained almost entirely undifferentiated from habituation. The<br />
exception is Bejder et al. (2009: p. 181) who define tolerance as the<br />
“intensity of disturbance that an individual. tolerates without<br />
responding in a defined way”. Tolerance may readily be confused<br />
with the complete absence of tourist impact. Indeed Higham<br />
(1998:529), in reference to viewing Northern Royal albatrosses<br />
(Diomedea epomophora sanfordi), warned that “tolerance should not<br />
disguise the fact that serious impacts may still take place”. The<br />
manifestations of disturbance in the case of these outwardly<br />
tolerant birds have been revealed only by long-term monitoring and
1294<br />
analysis (Higham et al., 2008). A more critical treatment of tolerance<br />
allows, in the first instance, for consideration of intrinsic tolerance,<br />
which varies both between individual animals and species.<br />
Many species of seabirds will demonstrate intrinsic tolerance to<br />
boat approach, seemingly oblivious to engine vibration and noise<br />
(both mechanical and human). Little penguins (Eudyptula minor)<br />
demonstrate intrinsic tolerance of boat engine vibration and noise,<br />
which indicates that they clearly are able to ignore human presence<br />
across settings (i.e. coastal landing zones and open water). Various<br />
species of New Zealand shag (cormorant) also demonstrate<br />
intrinsic tolerance to human approach (e.g. Spotted shags (Stictocarbo<br />
punctatus), Stewart Island shags (Leucocarbo chalconotus) and<br />
Little shags (Phalacrocorax melanoleucos)) (Shelton & Higham,<br />
2007). However tolerance inevitably is behaviour (stimulus)<br />
specific. Supporting Knight’s (2009) treatment of the tourist gaze is<br />
the fact that an intense and prolonged stare is sufficient to cause<br />
some individual birds (e.g. Spotted shags) to abandon their perches.<br />
3.2.2. Selective tolerance<br />
Tolerance, like habituation, has a spatio-temporal component.<br />
Solitary adult male New Zealand sea lions (Phocarctos hookeri)<br />
demonstrate various responses to human approach which may be<br />
termed selective tolerance (or intolerance). Some adult male New<br />
Zealand sea lions will, when in the water in the autumn months,<br />
respond intolerantly (e.g. aggressive approach, charging and biting)<br />
to the passing kayaks of a commercial operator. Repeated exposure<br />
to kayakers, at this time of year, does not decrease the likelihood of<br />
aggressive behaviour. At other times of the day when hauled out on<br />
a nearby beach the same individual animals will tolerate close<br />
approaches by visitors (Shelton & Higham, 2007). The sharp<br />
contrasts between agonistic and indifferent responses (which<br />
clearly vary on a spatio-diurnal basis) are almost certainly linked to<br />
specific spatio-ecological (i.e. territorial dominance) and temporal<br />
(i.e. reproduction) elements.<br />
3.2.3. Seasonal tolerance<br />
A variation of selective tolerance is seasonal tolerance. The New<br />
Zealand wood pigeon/kereru (Hemiphaga novaseelandiae) frequents<br />
some urban areas in New Zealand and can appear to be indifferent<br />
to everyday human activity. Closer analysis reveals seasonal variations<br />
in the species’ tolerance of human stimuli depending on food<br />
availability (Shelton & Higham, 2007). As the season progresses,<br />
and food becomes more abundant, kereru increasingly favour<br />
branches higher up in trees, and seasonal responses to human<br />
activity appear to change.<br />
In this case, it is clearly necessary to elucidate precisely what<br />
role people are playing as discriminative stimuli. People minding<br />
their own business, presumably measured by their unmodified<br />
patterns of movement, uninterrupted chatter and non-birddirected<br />
gaze, act as discriminative stimuli for approach tolerance.<br />
A sudden silence, ceasing walking, or bird-directed gaze act as<br />
stimuli for flight. The paradox for the commercial wildlife tourism<br />
operator who intends that visitors experience pigeons close-up is<br />
that the best advice to be given is to ignore them. Natural settings<br />
do not offer the frequency of close contact that would be required<br />
to attempt deliberately to manipulate the birds’ existing responses.<br />
These two examples illustrate a more general concern with how<br />
humans can best interact with multiple species of birds in urban<br />
areas (Blumstein, Fernandez-Juricic, Zollner, & Garity, 2005).<br />
3.3. Habituation<br />
3.3.1. Approach behaviour<br />
Bejder et al. (2009: p. 181) define habituation as the “relative<br />
persistent waning of a response as a result of repeated stimulation<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298<br />
which is not followed by any kind of reinforcement”. Shackley<br />
(1996) observes that wildlife may habituate to the presence of<br />
humans, showing no obvious signs of physiological or behavioural<br />
response. If so, such a lack of response may fruitfully be explored<br />
within the context of sustainable wildlife tourism and cessation of<br />
impact. However, habituation can be assumed to have important<br />
spatio-temporal dimensions. New Zealand dabchick (Poliocephalus<br />
rufopectus) may become habituated to human approach following<br />
multiple exposures (Bright, Reynolds, Innes, & Waas, 2003). Similarly<br />
diurnal or seasonal habituation may take place in some species<br />
of wild animals under certain circumstances. Non-breeding male<br />
New Zealand fur seals (Arctocephalus forsteri) may be placid on<br />
land, where they demonstrate a decreased gross behaviour<br />
response to human approach. The same animals may be aggressive<br />
to swimmers in the water, which acts against describing any individual<br />
animals as being habituated. Nearby breeding colonies of<br />
female New Zealand fur seals that are exposed to regular stimuli<br />
may become unresponsive to approach.<br />
3.3.2. Socially acquired habituation<br />
Furthermore, there is evidence to suggest that the absence of<br />
response behaviours may be socially acquired or learnt (Bejder<br />
et al., 2009) by wild animals. Early exposure, in the presence of<br />
an unresponsive mother, may result in seal pups that have never<br />
perceived tourist approach as a stimulus for flight. Seal pups do<br />
vary in their post-weaning exploratory audacity and some, having<br />
moved around the site once in the presence of humans with no illeffects,<br />
begin to include indifference in their behavioural repertoire.<br />
Each uneventful visit further reduces the discriminative value of<br />
humans. Those adult females who had been born at the colony<br />
could conceivably retain their neonate indifference but those<br />
recruited from other colonies, where avoidance response to less<br />
frequent human approach takes place, may demonstrate socially<br />
acquired habituation through their peers modelling indifference.<br />
Such complexities of seal habituation to human approach,<br />
including the confounding effect of probable previous exposure,<br />
has been investigated by van Polanen Petel (2005) who suggests<br />
a species-specific approach is required in order to manage<br />
humaneseal interactions optimally.<br />
3.3.3. Reduced physiological arousal<br />
Habituation in the form of complete indifference may be<br />
usefully differentiated from reduced physiological arousal, in which<br />
behavioural responses remain evident, but with reduced energy<br />
expenditure and arousal. At Taiaroa Head (Dunedin, New Zealand)<br />
Northern Royal albatross (Diomedea epomophora sanfordi) are<br />
closely monitored (Higham, 1998) and tourists are confined to<br />
a soundproof hide glazed with reflective glass. Throughout the<br />
breeding season all individual birds in the colony are exposed to<br />
close contact and attention from conservation staff, including<br />
weekly handling during weighing and measuring. These birds do<br />
not abandon their nests but do continue the defencive use of their<br />
beaks, as do Northern Royal albatrosses on the remote Campbell<br />
Island. Despite having been handled periodically over the years<br />
since their hatching these birds did not seem to have habituated to<br />
this particular human behaviour. However, although there had<br />
been no apparent change in the gross motor component of the<br />
birds’ defence rituals, there may well have been reductions in<br />
human-induced physiological arousal. Thus, the defence behaviour<br />
persists, but with reduced determination or persistence.<br />
3.3.4. Reduced avoidance behaviour<br />
A fourth manifestation of habituation may be termed reduced<br />
avoidance behaviour. This arises in cases where approach/avoidance<br />
is at the discretion of individual animals. While it is conceivable
that approach behaviour may be habituated the more likely case in<br />
wild animal populations is that avoidance behaviour may decrease<br />
over time. This appears to be the case in Yellow-eyed penguin<br />
(Megadyptes antipodes) colonies that are exposed to increasing<br />
tourist interactions, where established members of the colony may<br />
become less timid over time, but there may be a simultaneous<br />
reduction in the recruitment of prospecting birds. The impact of<br />
humans on this most phlegmatic but inquisitive of penguins is the<br />
subject of ongoing study (Seddon, Smith, Dunlop, & Mathieu, 2004).<br />
This work is predicated on the fact that measuring penguin habituation<br />
using only approach/avoidance behaviour is inappropriate<br />
since physiological arousal fluctuates independently of observable<br />
behaviour (Ellenberg & Mattern, 2004) and the relationship varies<br />
both between individual penguins of the same species and different<br />
penguin species (Ellenberg et al., 2006).<br />
3.4. Sensitisation<br />
Finally, it is important to attempt to differentiate between<br />
sensitisation and habituation. Usually, sensitisation refers to an<br />
increase in the strength of a response upon repeated or ongoing<br />
exposure to a stimulus that has significant consequences (Bejder<br />
et al., 2009). However, it is noteworthy that repeated exposure to<br />
tourists appeared to lower the production of corticosterone,<br />
commonly a stress hormone, in Galapagos marine iguana (Romero &<br />
Wikelsi, 2002). Lower corticosterone production in response to<br />
human interaction, compared with animals who have no human<br />
interaction, seems to demonstrate sensitisation, the opposite of<br />
habituation, but in this case the lowering of corticosterone production,<br />
although demonstrating sensitisation, may legitimately be<br />
labelled paradoxical sensitisation. Of course one study is inconclusive<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298 1295<br />
but it does raise an intriguing possibility. If robust, this finding<br />
implies that chronic sub-optimal physiological processes may be<br />
a naturally-occurring phenomenon in some species of wild animals.<br />
4. Wildlife habituation and sustainable nature-based<br />
tourism: a management model<br />
This discussion addresses the proposition that ecotourism<br />
operators who are committed to providing a sustainable naturebased<br />
product will be interested to know how their activities may<br />
be influencing the environments in which they do business. The<br />
inherent tensions between business practice and sustainability in<br />
nature-based tourism have been well described (Fennell, 2003;<br />
McKercher, 1993a, 1993b, 1998; Newsome et al., 2005). The<br />
preceding discussions allow for the development of a management<br />
model for sustainable wildlife tourism interactions (Fig. 1) based on<br />
a critical behavioural formulation of habituation.<br />
Following Duffus and Dearden (1990) Fig. 1 is constituted<br />
principally by three key elements; site users, focal animals (individuals<br />
or local populations) and the wider ecology of the wild<br />
animals engaged with human approach and observation. Such<br />
a representation recognises that any humanewildlife interactions<br />
are spatially bound; that is, human interactions with individual<br />
wild animals can take place in a variety of settings that vary across<br />
a range of habitats where different behaviours (e.g. migration,<br />
feeding, resting, socialising, breeding, raising neonates etc.) take<br />
place. Animal responses to human approach and interaction inevitably<br />
vary between settings.<br />
The model recognises also the temporal variation that exists in<br />
terms both of site users (e.g. timing and duration of visit, and<br />
seasonal context), and focal animals (e.g. diurnal, seasonal and life<br />
Fig. 1. Management model for sustainable wildlife tourism interactions based on a critical behavioural formulation of the term habituation.
1296<br />
course), which influence the outcomes of humaneanimal interactions.<br />
Avoidance/approach behaviours, tolerance, habituation and<br />
sensitisation can fruitfully be considered within this context.<br />
A critical understanding of spatio-temporal ecology and the interactions<br />
of tourists and focal animals can then be applied in terms of<br />
tourism management planning and consideration of conscious<br />
management interventions (Fig. 1). This would require a critical<br />
understanding of the unitary term habituation, which is attempted<br />
in this article. It would also necessitate a mixed (biological and<br />
social) science platform of research to inform an integrated and<br />
adaptive management approach (Higham, Bejder, & Lusseau, 2009).<br />
5. Discussion<br />
As nature-based products are offered in more and more remote<br />
places (i.e. previously unexploited ecosystems) it becomes<br />
increasingly likely that any given individual animal will encounter<br />
humans, engaged in particular behaviours, at some stage of its life.<br />
The animal may approach the human, as neonates and juveniles of<br />
many species do. If, on repeated exposure to these particular<br />
human behaviours, this approach behaviour decreases, and is not<br />
replaced by escape/avoidance behaviour, then the animal may<br />
accurately be described as being habituated to the human behaviours<br />
to which it has been exposed, in a particular spatio-temporal<br />
setting.<br />
Any habituation/sensitisation process is dynamic, and may<br />
reflect the hormonal state of the individual animal at the time of<br />
exposure to the particular human behaviour. If the individual<br />
animal demonstrated indifference to the human behaviour right<br />
from the start then habituation is not the best term to use to<br />
describe the situation; tolerance or resilience may be more useful.<br />
Similarly, animals may respond to human activity through<br />
increases in physiological processes, commonly grouped under the<br />
term arousal. Decreasing physiological arousal upon repeated<br />
exposure indicates habituation, increased arousal indicates sensitisation.<br />
The unusual situation where an initially high level of<br />
arousal may be reduced by exposure to human activity may be<br />
tentatively labelled paradoxical sensitisation. It is possible also for<br />
wildlife to decrease one response to human presence, for example<br />
fleeing, but to increase another response. Chimpanzees’ activity<br />
rates in Kibale Forest, Uganda, habituate to the presence of<br />
increasing numbers of observers but vocalisation increases (Johns,<br />
1996).<br />
It is useful at this point to address two issues. Firstly, why is it<br />
important to specify to what human behaviours, in what settings<br />
and at what times do individual animals habituate, and to avoid the<br />
simpler descriptor; habituation to humans? Habituation to<br />
humans, by definition, involves decreased vigilance in response to<br />
human presence, decreased communication of alarm upon coming<br />
across humans and/or decreased avoidance of, or fleeing from,<br />
humans. The argument then is that habituated animals are made<br />
vulnerable in that habituation to one potential predator, Homo<br />
sapiens will induce a generalised reduction of predator-preparedness<br />
(Beale & Monaghan, 2004).<br />
We argue that this concern evaporates when habituation is<br />
more usefully formulated. If habituation is related to spatially and<br />
temporally constrained specific human behaviours, rather than<br />
simply to human presence, then it is reasonable to hypothesise that<br />
individual animals will retain predator-preparedness outside of<br />
these parameters. Discriminating between human behaviours is<br />
consistent with laboratory-based enquiry that has demonstrated<br />
that many species have the ability to discriminate even between<br />
individual human beings performing the same behaviours (Davis,<br />
2002). Predator-preparedness training in birds, in common with<br />
exploratory behaviours (Huber, Rechberger, & Taborsky, 2001), can<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298<br />
occur through social learning. For example adult Sandhill cranes<br />
teach predator-preparedness by modelling vigilance and predatorresponse<br />
even to unrelated captive-reared conspecific chicks<br />
(Heatley, 1995). There is debate about how possible habituation to<br />
human activity acquired while in captivity, for example in a rehabilitation<br />
facility, should be managed. A fine-grained analysis is<br />
available addressing the issue of how much deliberate training-forrelease<br />
should take place, what such training should comprise, and<br />
exactly how humans should be involved in its delivery (Bauer,<br />
2005).<br />
It is reasonable to suppose that habituation can be socially learnt<br />
also in natural settings, and be passed on in a sophisticated rather<br />
than generalised way. Different species teach their offspring<br />
different sensitivities. For cliff-dwelling birds visited by tourists,<br />
size-of-party has been reported to have species-specific effects<br />
(Beale & Monaghan, 2005). It is not unreasonable to suggest that<br />
certain species may teach their young not to respond to a particular<br />
wildlife guide, acting within certain behavioural limits. In such<br />
cases it is critical that individual animals are the unit of analysis in<br />
rigorous longitudinal studies (Bejder et al., 2009).<br />
The second issue that needs to be addressed is one of<br />
management. If, as suggested here, some or all of the behaviours of<br />
many wild species can be brought under stimulus control, then<br />
what happens to the wild in wildlife? The application of stimulus<br />
control techniques to the management of human behaviour is<br />
widely reported in the Applied Behaviour Analysis literature. Using<br />
the nuanced approach to habituation outlined here it is conceivable<br />
that wildlife managers and visitor managers could co-operate to<br />
apply behavioural technologies to structure almost completely the<br />
nature of humanewildlife interactions site-by-site, species-byspecies,<br />
and in some cases to the level of individual animals, taking<br />
into account time of day and time of year. This level of management,<br />
although common in zoos and other captive settings, if<br />
applied to natural settings would challenge directly the motivation<br />
for visiting such sites in the first place. Nonetheless, in certain cases<br />
the conservation outcomes available using such an approach may<br />
be justified.<br />
6. Conclusion<br />
This article highlights the wide variability of wild animal<br />
responses to human approach and interaction. In doing so, it begs<br />
the question; which individual animals or groups of animals<br />
respond in what fashion to what kind of tourist behaviour in what<br />
contexts? Clearly individual (or groups of) animals that respond<br />
passively in one setting, may in other settings not be so sanguine.<br />
For example, where a particular set of contingencies operate to<br />
induce decreasing gross behavioural responsiveness, on land, to<br />
a fairly narrow set of visitor behaviours, in a group of seals who<br />
shared non-breeding status, evidence of spatially and temporally<br />
bounded habituation may exist. The responsiveness of physiological<br />
processes within individual seals, the role of developmental<br />
maturation and the influence of vicarious or social learning through<br />
the observation of calm conspecifics remain largely unexamined,<br />
although available for formulation.<br />
Clearly, a fine-grained behavioural approach is required to<br />
overcome the poor fit between observable wildlife behaviour in the<br />
presence of humans and individuals’ internal states, and to<br />
accommodate the fact that habituation is spatially and temporally<br />
bounded. Given that such a poor fit is unhelpful to individual,<br />
species or whole-of-ecosystem management, it seems desirable<br />
that ecotourism operators actively address likely habituation or<br />
other conditioning to humans as part of their business plan<br />
or concession application. As such, wildlife managers and visitor<br />
managers could co-operate to apply behavioural technologies
to structure almost completely the nature of humanewildlife<br />
interactions. A tourism management model that accommodates<br />
a nuanced understanding of wildlife habituation is presented,<br />
building upon a formulative framework for humanewildlife interactions.<br />
The development of empirical case studies to inform and<br />
monitor tour operator best practices is a valuable avenue of future<br />
research.<br />
Knight (2009) presents three approaches to making wildlife<br />
viewable; capture and confinement, habituation, and attraction.<br />
While of the three, habituation is the least likely to be associated<br />
with diminished behaviours, the question of diminished wildness<br />
remains unanswered. Exploring this question will, no doubt, reveal<br />
a trade off between using habituation as a tourism management<br />
tool and the potential evolutionary costs to focal animal populations,<br />
not only in terms of physiological and behavioural<br />
responses, but also in terms of exposure to disease (Bejder et al.,<br />
2009). One provocative approach, integrated with the field of<br />
neurobiology, will be designer wild species, specifically habituated<br />
for viewing. Alternatively, Knight (2009) observes that learnt<br />
animal behaviours can be unlearnt, giving rise to the notion of the<br />
reversal of habituation. AsKnight (2009: p. 180) comments, “In an<br />
age when our visual appetite for wildlife has never been greater,<br />
there may be good reasons for keeping this appetite in check and<br />
acting to make viewable wild animals unviewable again”. Before<br />
further contemplating the reversal of habituation, a more critical<br />
understanding of habituation itself is required. This amplifies the<br />
need for “studies (that) adopt a long-term experimental design<br />
involving sequential sampling of the same individuals at different<br />
levels of exposure to a disturbance” (Bejder et al., 2009: p. 181). In<br />
the meantime, these thoughts pose intriguing questions that are<br />
available for scholars to debate and research.<br />
References<br />
Baker, C. S., & Herman, L. M. (1989). Behavioral responses of summering humpback<br />
whales to vessel traffic: Experimental and opportunistic observations. Final Report<br />
to the National Park Service. Anchorage, Alaska: Alaska Regional Office.<br />
Baker, C. S., Perry, A., & Vequist, G. (1988). Humpback whales of Glacier Bay, Alaska.<br />
Whalewatcher Fall, 13e<strong>17</strong>.<br />
Bauer, G. (2005). Research training for releasable animals. Conservation Biology,<br />
19(6), <strong>17</strong>79e<strong>17</strong>89.<br />
Beale, C., & Monaghan, P. (2004). Human disturbance: people as predation-free<br />
predators? Journal of Applied Ecology, 41, 335e343.<br />
Beale, C., & Monaghan, P. (2005). Modeling the effects of limiting number of visitors<br />
on failure rates of seabird nests. Conservation Biology, 19(6), 2015e2019.<br />
Bejder, L., Dawson, S. M., & Harraway, J. A. (1999). Responses by Hector’s dolphins to<br />
boats and swimmers in Porpoise Bay, New Zealand. Marine Mammal Science,<br />
15(3), 738e750.<br />
Bejder, L., Samuels, A., Whitehead, H., Finn, H., & Allen, S. (2009). Impact assessment<br />
research: use and misuse of habituation, sensitisation and tolerance in<br />
describing wildlife responses to anthropogenic stimuli. Marine Ecology Progress<br />
Series, 395, <strong>17</strong>7e185.<br />
Bejder, L., Samuels, A., Whitehead, H., Gales, N., Mann, J., Conner, R., et al. (2006).<br />
Decline in relative abundance of bottlenose dolphins exposed to long-term<br />
disturbance. Conservation Biology, 20(6), <strong>17</strong>91e<strong>17</strong>98.<br />
Blane, J. M. (1990). Avoidance and interactive behaviour of the Saint Lawrence<br />
beluga whale (Delphinapterus leucas) in response to recreational boating. M.A.<br />
thesis, Canada: University of Toronto.<br />
Blane, J. M., & Jaakson, R. (1995). The impact of ecotourism boats on the Saint<br />
Lawrence beluga whales. Environmental Conservation, 21(3), 267e269.<br />
Blumstein, D., Fernandez-Juricic, E., Zollner, P., & Garity, S. (2005). Inter-specific<br />
variation in avian responses to human disturbance. Journal of Applied Ecology,<br />
42(5), 943e953.<br />
Bright, A., Reynolds, G., Innes, J., & Waas, J. (2003). Effects of motorised boat passes<br />
on the time budget of New Zealand dabchick, Poliocephalus rufopectus. Wildlife<br />
Research, 30, 237e244.<br />
Campagna, C., Rivarola, M. M., Greene, D., & Tagliorette, A. (1995). Watching southern<br />
right whales in Patagonia. Unpublished report to UNEP, Nairobi.<br />
Constantine, R. (2001). Increased avoidance of swimmers by wild bottlenose<br />
dolphins (Tursiops truncatus) due to long-term exposure to swim-with-dolphin<br />
tourism. Marine Mammal Science, <strong>17</strong>(4), 689e702.<br />
Corkeron, P. J. (1995). Humpback whales (Megaptera novaeangliae) in Hervey Bay,<br />
Queensland: behaviour and responses to whale-watching vessels. Canadian<br />
Journal of Zoology, 73, 1290e1299.<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298 1297<br />
Davis, H. (2002). Prediction and preparation: Pavlovian implications of research<br />
animals discriminating among humans. Institute for Laboratory Animal Research,<br />
43(1), 19e26.<br />
Davis, H., Wellwood, D., & Ciarniello, L. (2002). “Bear smart” community program:<br />
Background report. Victoria B.C.: British Columbia Ministry of Water, Land and<br />
Air Protection.<br />
Duffus, D. A., & Dearden, P. (1990). Non-consumptive wildlife-oriented recreation:<br />
a conceptual framework. Biological Conservation, 53(3), 213e231.<br />
Edds, P. L., & MacFarlane, J. A. F. (1987). Occurrence and general behaviour of<br />
balaenopterid cetaceans summering in the Saint Lawrence Estuary, Canada.<br />
Canadian Journal of Zoology, 65, 1363e1376.<br />
Edington, J. M., & Edington, M. A. (1986). Ecology, recreation and tourism. Cambridge:<br />
Cambridge University Press.<br />
Ellenberg, U., & Mattern, T. (2004). The most timorous of all? Impact of human<br />
disturbance on Humboldt penguins. New Zealand Journal of Zoology, 31,<br />
120e121.<br />
Ellenberg, U., Mattern, T., Seddon, P. J., & Jorquera, G. L. (2006). Physiological and<br />
reproductive consequences of human disturbance in Humboldt penguins: the<br />
need for species-specific visitor management. Biological Conservation, 133(1),<br />
95e106.<br />
Fennell, D. (2003). Ecotourism: An introduction (2nd ed.). London: Routledge.<br />
Gordon, J., Leaper, R., Hartley, F. G., & Chappell, O. (1992). Effects of whale watching<br />
vessels on the surface and underwater acoustic behaviour of sperm whales off<br />
Kaikoura, New Zealand. InScience and Research Series, Vol. 52. Wellington, New<br />
Zealand: Department of Conservation.<br />
Heatley, J. (1995). Antipredator conditioning in Mississippi Sandhill cranes (Grus<br />
canadensis pulla). Unpublished Master of Science, Texas A&M.<br />
Higginbottom, K. (Ed.). (2004). Wildlife tourism: Impacts, management and planning.<br />
Altona (Victoria, Australia): CRC for Sustainable Tourism.<br />
Higham, J. E. S. (1998). Tourists and albatrosses: the dynamics of tourism at the<br />
Northern Royal Albatross Colony, Taiaroa Head, New Zealand. Tourism<br />
Management, 19, 521e531.<br />
Higham, J. E. S., Bejder, L., & Lusseau, D. (2009). An integrated and adaptive<br />
management model to address the long-term sustainability of tourist interactions<br />
with cetaceans. Environmental Conservation, 35(4), 294e302.<br />
Higham, J. E. S., & Lusseau, D. (2004). Ecological impacts and management of tourist<br />
engagements with cetaceans. In R. Buckley (Ed.), Environmental impacts of<br />
ecotourism (pp. <strong>17</strong>3e188). Wallingford: CAB <strong>International</strong>.<br />
Higham, J. E. S., Lusseau, D., & Hendry, W. (2008). The viewing platforms from which<br />
animals are observed in the wild: a discussion of emerging research directions.<br />
Journal of Ecotourism, 7(2/3), 132e141, Special Issue on Australian Wildlife<br />
Tourism.<br />
Houston, D. M., & Russell, J. J. (2000). The impact of tourism on blue penguins<br />
(Eudyptula minor). Unpublished conference paper. The 4th international<br />
penguin conference, La Serena, Chile. September 4e8th 2000.<br />
Huber, L., Rechberger, S., & Taborsky, M. (2001). Social learning affects object<br />
exploration and manipulation in keas, Nestor notabilis. Animal Behaviour, 62,<br />
945e954.<br />
IWC. (2006). <strong>International</strong> whaling commission’s 58th annual meeting in St. Kitts and<br />
Nevis 2006. Retrieved 25 August, 2007, from. http://iwcoffice.org/_documents/<br />
commission/IWC58docs/iwc58docs.htm.<br />
Janik, V. M., & Thompson, P. M. (1996). Changes in surfacing patterns of<br />
bottlenose dolphins in response to boat traffic. Marine Mammal Science, 12,<br />
597e602.<br />
Johns, B. (1996). Responses of chimpanzees to habituation and tourism in the Kibale<br />
Forest, Uganda. Biological Conservation, 78, 257e262.<br />
Knight, J. (2009). Making wildlife viewable: habituation and attraction. Society and<br />
Animals, <strong>17</strong>, 167e184.<br />
Lusseau, D. (2003). The effects of tour boats on the behavior of bottlenose dolphins:<br />
using Markov chains to model anthropogenic impacts. Conservation Biology,<br />
<strong>17</strong>(6), <strong>17</strong>85e<strong>17</strong>93.<br />
Lusseau, D., & Higham, J. E. S. (2004). Managing the impacts of dolphin-based<br />
tourism through the definition of critical habitats: the case of bottlenose<br />
dolphins (Tursiops spp.) in Doubtful Sound, New Zealand. Tourism Management,<br />
25(5), 657e667.<br />
MacGibbon, J. (1991). Responses of sperm whales (Physeter macrocephalus) to<br />
commercial whale watching boats off the coast of Kaikoura. Unpublished report to<br />
the Department of Conservation. University of Canterbury, Christchurch, New<br />
Zealand.<br />
McKercher, B. (1993a). The unrecognised threat to tourism: can tourism survive<br />
“sustainability”? Tourism Management, 14(2), 131e136.<br />
McKercher, B. (1993b). Some fundamental truths about tourism: understanding<br />
tourism’s social and environmental impacts. Journal of Sustainable Tourism, 1(1),<br />
6e16.<br />
McKercher, B. (1998). The business of nature-based tourism. Melbourne: Hospitality<br />
Press.<br />
Mazur, J. (2006). Learing and behaviour (6th ed.). Upper Saddle River, NJ: Pearson<br />
Prentice Hall.<br />
Muloin, S. (1998). Wildlife tourism: the psychological benefits of whale watching.<br />
Pacific Tourism Review, 2, 199e212.<br />
National Park Service. (2006a). Deer and people at Fire Island National Seashore.<br />
Retrieved 27 March 2008, from. http://www.nps.gov.fiis/deerpeople/deer.html.<br />
National Park Service. (2006b). Deer and people at Fire Island National Seashore.<br />
Retrieved 22nd June, 2006, from. http://www.nps.gov/fiis/deerpeople/deer.<br />
html.
1298<br />
Newsome, D., Dowling, R., & Moore, S. (2005). Wildlife tourism. Clevedon: Channel View.<br />
Nimon, A. J., Schroter, R. C., & Stonehouse, B. (1995). Heart rate of disturbed<br />
penguins. Nature, 374(6521), 415.<br />
Nisbet, I. C. T. (2000). Disturbance, habituation, and management of waterbird<br />
colonies. Waterbirds, 23(2), 312e332.<br />
Nowacek, S. M., Wells, R. S., & Solow, A. R. (2001). Short-term effects of boat traffic<br />
on bottlenose dolphins, Tursiops truncatus, in Sarasota Bay, Florida. Marine<br />
Mammal Science, <strong>17</strong>(4), 673e688.<br />
Orams, M. (1995). Development and management of a feeding program for wild<br />
bottlenose dolphins at Tangalooma, Australia. Aquatic Mammals, 21(2), 137e147.<br />
Orams, M. (2000). Tourists getting close to whales, is it what whale-watching is all<br />
about? Tourism Management, 21(6), 561e569.<br />
van Polanen Petel, T. (2005). Effects of human activity on Weddell seals (Leptonychotes<br />
weddellii) in Antarctica. Retrieved 23 June, 2006, from. http://www.crctourism.<br />
com.au/CRCServer/documents/education/TamaraCompletion_report.doc.<br />
Romero, L., & Wikelsi, M. (2002). Exposure to tourism reduces stress-induced corticosterone<br />
levels in Galapagos marine iguanas. Biological Conservation, 108, 371e374.<br />
Salden, D. R. (1988). Humpback whales encounter rates offshore of Maui, Hawaii.<br />
Journal of Wildlife Management, 52(2), 301e304.<br />
Salvado, C. A. M., Kleiber, P., & Dizon, A. E. (1992). Optimal course by dolphins for<br />
detection avoidance. Fishery Bulletin, 90, 4<strong>17</strong>e420.<br />
J.E.S. Higham, E.J. Shelton / Tourism Management 32 (2011) 1290e1298<br />
Seddon, P., Ellenberg, U., & van Heezik, Y. (in press). Yellow-eyed penguin. In D.<br />
Boersma, & P. Garcia Borboroglu (Eds). Penguins. Washington, USA: University of<br />
Washington Press.<br />
Seddon, P., Smith, A., Dunlop, E., & Mathieu, R. (2004). Tourist visitor attitudes,<br />
activities and impacts at a yellow-eyed penguin breeding site on the Otago<br />
Peninsula, Dunedin, New Zealand. New Zealand Journal of Zoology, 31,<br />
119e120.<br />
Shackley, M. (1995). The future of gorilla tourism in Rwanda. Journal of Sustainable<br />
Tourism, 3(2), 61e72.<br />
Shackley, M. (1996). Wildlife tourism. London: Thomson Learning.<br />
Shelton, E. J., & Higham, J. E. S. (2007). Ecotourism and wildlife habituation. In<br />
J. E. S. Higham (Ed.), Critical issues in ecotourism: Understanding a complex<br />
tourism phenomenon (pp. 270e286). Oxford: Butterworth Heinemann.<br />
Shelton, E. J., Higham, J. E. S., & Seddon, P. (2004). Habituation, penguin research<br />
and ecotourism: some thoughts from left field. New Zealand Journal of Zoology,<br />
31(4), 119.<br />
Whittaker, D. (1997). Capacity norms on bear viewing platforms. Human Dimensions<br />
of Wildlife, 2(2), 37e49.<br />
Williams, R., Trites, A. W., & Bain, D. (2002). Behavioural responses of killer whales<br />
(Orcinus Orca) to whale-watching boats: opportunistic observations and<br />
experimental approaches. Journal of Zoology, 256, 255e270.
MARINE MAMMAL SCIENCE, 27(3): 606<strong>–</strong>621 (July 2011)<br />
C○ 2011 by the Society for Marine Mammalogy<br />
DOI: 10.1111/j.<strong>17</strong>48-7692.2010.00435.x<br />
Using distance sampling techniques to estimate<br />
bottlenose dolphin (Tursiops truncatus) abundance<br />
at Turneffe Atoll, Belize<br />
DOROTHY M. DICK 1<br />
ELLEN M. HINES<br />
Department of Geography and Human Environmental Studies,<br />
San Francisco State University,<br />
1600 Holloway Avenue,<br />
HSS Room 279,<br />
San Francisco, California 94132, U.S.A.<br />
E-mail: doridick14@gmail.com<br />
ABSTRACT<br />
Reliable abundance estimates are critical for management and conservation of<br />
coastal small cetaceans. This is particularly important in developing countries<br />
where coastal human populations are increasing, the impacts of anthropogenic<br />
activities are often unknown, and the resources necessary to assess coastal cetaceans<br />
are limited. We adapted ship-based line transect methods to small-boat surveys<br />
to estimate the abundance of bottlenose dolphins (Tursiops truncatus) at Turneffe<br />
Atoll, Belize. Using a systematic survey design with random start and uniform<br />
coverage, 34 dolphin clusters were sighted during small-boat line transect surveys<br />
conducted in 2005<strong>–</strong>2006. Distance sampling methods estimated abundance at 216<br />
individuals (CV = 27.7%, 95% CI = 126<strong>–</strong>370). Due to species rarity in the<br />
Atoll, small sample size, and potential violations in line transect assumptions, the<br />
estimate should be considered preliminary. Nevertheless, it provides up-to-date<br />
information on the status of a regional population in an area under increasing threat<br />
of habitat loss and prey depletion via uncontrolled development and unsustainable<br />
fishing. This information will be useful as Belize develops a new conservation<br />
initiative to create a comprehensive and resilient marine protected area system.<br />
Our study illustrates the application of distance sampling methods to small-boat<br />
surveys to obtain abundance estimates of coastal cetaceans in a region lacking<br />
resources.<br />
Key words: survey design, small-boat surveys, distance sampling, line transects,<br />
Distance 5.0, abundance, bottlenose dolphin, Tursiops truncatus, Turneffe Atoll,<br />
Belize.<br />
1 Also affiliated with: Oceanic Society, Fort Mason Quarters 35, San Francisco, California 94123,<br />
U.S.A. Current address: Department of Geosciences, Geography Program, Oregon State University, 104<br />
Wilkinson Hall, Corvallis, Oregon 97331, U.S.A.<br />
606
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 607<br />
Although the common bottlenose dolphin (Tursiops truncatus) is well studied and<br />
widespread across various marine habitats in temperate and tropical waters (Wells<br />
and Scott 1999), its worldwide status is unknown. In the IUCN 2008 Red List of<br />
Threatened Species update, T. truncatus was classified as Species of Least Concern and<br />
listed habitat destruction and degradation, disturbance and harassment, prey depletion,<br />
pollution, and direct/indirect takes as specific threats of concern (Hammond<br />
et al. 2008). While these threats apply to the species as a whole, any one may impact<br />
certain regional populations more than others as seen in the Mediterranean (Bearzi<br />
and Fortuna 2006, Bearzi et al. 2008), Moray Firth, Scotland (Wilson et al. 1997, Sini<br />
et al. 2005), and Peru (Van Waerebeek et al. 1997). Declines in regional populations<br />
of apex predators, such as bottlenose dolphins, could have far reaching effects on<br />
the community structure of an ecosystem (Currey et al. 2009), as seen in western<br />
Alaska when increased killer whale predation on sea otters lead to a dramatic rise<br />
in sea urchins and subsequent decreases in kelp forests (Estes et al. 1998). For these<br />
reasons, coastal bottlenose dolphin populations should be assessed on a regional scale<br />
when evaluating status, managing threats, and implementing conservation measures<br />
(Reeves et al. 2003, Currey et al. 2009).<br />
The effects of anthropogenic activities on coastal bottlenose dolphin populations<br />
are of particular concern as human populations continue to grow along coastlines,<br />
especially in developing countries (Aragones et al. 1997, Dawson et al. 2008). The<br />
extent of these activities, however, is generally unknown and the data needed to<br />
substantiate the impacts are rarely available. Without adequate knowledge about<br />
the status and life history of these populations future management actions are limited.<br />
Consequently, there is a strong need to conduct population assessment studies<br />
on coastal marine mammal populations in underdeveloped countries (Vidal 1993,<br />
Aragones et al. 1997, Hines et al. 2005, Dawson et al. 2008).<br />
Line transect surveys using distance sampling protocols are a common method<br />
used to assess marine mammal populations (Buckland et al. 2001). Surveys of this<br />
type typically use large ships or fixed-wing aircraft, sophisticated equipment (high<br />
powered binoculars, navigational tools, e.g., radar and electronic charts), and may traverse<br />
sizeable ocean expanses (Barlow 1988, 1995; Barlow et al. 1988; Calambokidis<br />
and Barlow 2004; Forcada et al. 2004; Mullin and Fulling 2004; Zerbini et al. 2007).<br />
As a result they are prohibitively expensive and inaccessible to developing countries<br />
with limited budgets and expertise and shallow coastal regions (Dawson et al. 2008).<br />
Adapting ship-based line transect methods to small-boat surveys (Vidal et al. 1997;<br />
Dawson et al. 2004; Williams and Thomas 2007, 2009) may be the best option to<br />
estimate coastal marine mammal populations in less affluent nations.<br />
Turneffe Atoll (Fig. 1), located in Belize, Central America, is the largest most<br />
biologically diverse of the nation’s three atolls (Stoddart 1962) and the only one<br />
without long-term ecological protection. Turneffe Atoll provides year-round habitat<br />
to a small population of coastal bottlenose dolphins (Grigg and Markowitz 1997,<br />
Campbell et al. 2002). Photo-identification studies from the 1990s estimated a<br />
population of less than 90 individuals (Campbell et al. 2002). More than a decade<br />
later there have been no new abundance estimates. Protected from import/export,<br />
wildlife trade, and hunting by Belize’s 1981 Wildlife Protection Act, threats from<br />
human induced mortality are currently minimal. However, unsustainable fishing<br />
(overfishing and illegal fishing) and rapid coastal development (mangrove clearing,<br />
dredging, and overdevelopment) have been identified as severe threats to the ecological<br />
integrity of Turneffe Atoll (World Resources Institute 2005, Granek 2006).<br />
There is a high probability that the small dolphin population could be threatened by
608 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
Figure 1. The location of Turneffe Atoll, Belize, the survey design, and the bottlenose<br />
dolphin sighting locations during small-boat surveys conducted in 2005<strong>–</strong>2006. The survey<br />
was created with the automated survey design function in Program Distance 5.0 release 2<br />
using two substrata, the lagoon area and the western area between the mangrove cays and the<br />
fringing reef.<br />
habitat degradation, prey depletion, vessel traffic, and pollution as the atoll’s human<br />
population increases (Wells and Scott 1999). Obtaining an up-to-date abundance<br />
estimate for the population before further anthropogenic impacts occur is imperative<br />
to provide the baseline information necessary to guide future management and<br />
conservation actions.
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 609<br />
We sought to (1) develop and implement a repeatable and economically feasible<br />
systematic survey for use on bottlenose dolphins at Turneffe Atoll, (2) provide the<br />
first non-mark-recapture quantitative estimates of dolphin abundance in the study<br />
area, and (3) develop survey methods potentially applicable for other small cetacean<br />
coastal surveys in underdeveloped countries. This project is part of a long-term<br />
social and behavioral ecology research program of bottlenose dolphins conducted at<br />
Turneffe Atoll.<br />
MATERIALS AND METHODS<br />
Study Area<br />
Located 50 km east of mainland Belize and 18 km beyond the Mesoamerican Barrier<br />
Reef in the western Caribbean Sea, Turneffe Atoll is roughly 50 km long, 16 km<br />
wide at its widest point (average width 8<strong>–</strong>10 km), and encompasses approximately<br />
531 km2 (Fig. 1) (Stoddart 1962). The atoll contains >200 mangrove cays (islands),<br />
incorporates three shallow (
610 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
Survey Procedures/Field Work<br />
Using the GPS unit to guide the boat along each transect line, surveys were<br />
conducted from either an 8.2 m skiff with two 85 hp outboard motors or a 6.4 m<br />
boat equipped with one 150 hp outboard motor. Boat speed was kept constant<br />
between 12 and 15 km/h and surveys were only conducted in Beaufort sea states of<br />
≤3 and wave heights
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 611<br />
one outlier at around 200 m, allowing better model fit (Buckland et al. 2001).<br />
Distance 5.0 release 2 (Thomas et al. 2010) was used to estimate detection probability<br />
using Conventional Distance Sampling (CDC) and Multiple Covariate Distance<br />
Sampling (MCDS) (Buckland et al. 2001, 2004). Unlike CDC, which assumes the<br />
detection of an object is the sole function of its distance from the line, MCDS allows<br />
for the inclusion of additional variables that are likely to impact the detection<br />
probability. For cetacean surveys, Beaufort sea state is a common covariate (Palka<br />
1996, Barlow et al. 2001) and was used here. Following Buckland et al. (2001),<br />
several standard detection function models in CDS and MCDS were fitted to the<br />
data. The Kolmogorov<strong>–</strong>Smirnov goodness-of-fit test for each model was used to examine<br />
the absolute fit of the model (Buckland et al. 2004). Estimates from the model<br />
with the lowest Akaike’s Information Criterion (AIC) were selected (Buckland et al.<br />
2001).<br />
Results from a size-biased regression (the default method in Distance) indicated<br />
weak evidence of dependence between cluster size and distance (rs = 0.19, P =<br />
0.14). Mean sample cluster size (¯s ) was considered an unbiased estimator of the<br />
estimated mean cluster size of the population ( Ê (s )) (or Ê (s ) = ¯s ) (Buckland et al.<br />
2001). Bottlenose dolphin abundance ( ˆ N) within the survey area was estimated using<br />
(Buckland et al. 2001, Marques and Buckland 2003):<br />
ˆN = A ·<br />
n · ¯s<br />
c · 2L · ,<br />
where ˆ N is the estimated abundance, A the size of the study area, n the number of<br />
clusters seen, ¯s the mean sample cluster size, c the constant (or multiplier) indicating<br />
the number of times each line was surveyed, L the total length of transect line, and<br />
the effective strip half-width.<br />
The coefficient of variations (CV) for n, , and ¯s were each calculated individually.<br />
Empirical estimation of CV(n), as recommended by Buckland et al. (2001), plus an<br />
adjustment for multiple visits to transect lines followed:<br />
where var(n) =<br />
CV(n) =<br />
TL<br />
var(n)<br />
n 2<br />
k<br />
ti · li · (n i/(ti · li) − n/TL) 2<br />
i=1<br />
k − 1<br />
TL is the total line length traveled, represented by, TL = k<br />
i=1 ti · li, ti the number<br />
of times a line was traveled, li the length of transect line i, ni the number<br />
of detections on line i, k the number of transect lines. CV() and CV(¯s ) were<br />
each calculated by dividing the standard error of the estimator by itself. Individual<br />
CVs were used to compute the abundance estimate CV using (Buckland et al.<br />
2001):<br />
CV( ˆ N) = [CV(n)] 2 + [CV()] 2 + [CV(¯s )] 2 .<br />
,
612 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
RESULTS<br />
Survey Design<br />
The total planned survey effort was 220.54 km across 310.25 km2 . Due to shallow<br />
areas inaccessible to the boat, the realized survey effort was <strong>17</strong>5.21 km across<br />
298.5 km2 (about 80% of planned) for a set of transect lines. Six survey sets were<br />
completed, totaling 1,051.26 km of effort over 471.7 h during the rainy seasons of<br />
2005 (November<strong>–</strong>December) and 2006 (June<strong>–</strong>August and October<strong>–</strong>December) and<br />
during a two-week period (March<strong>–</strong>April) in the 2006 dry season.<br />
Abundance Estimation<br />
Thirty-four dolphin clusters, totaling 97 animals, were sighted on-effort (Fig. 1).<br />
Small sample size (n = 33 after truncation) precluded stratification by season; ungrouped<br />
perpendicular distances were pooled across all survey sets for analysis. Cluster<br />
size ranged from 1 to 12, with a mean cluster size of 2.61 (CV = <strong>17</strong>%, 95% CI =<br />
1.85<strong>–</strong>3.68). The majority of sightings (84.5%) were of clusters ≤3 dolphins.<br />
Model results and the best fitting plotted detection function are presented in<br />
Table 1 and Figure 2. Estimated abundance and density for the study area was<br />
216 dolphins (CV = 27.7%, 95% CI = 126<strong>–</strong>370) and 0.749 dolphins/km2 (CV =<br />
27.7%, 95% CI = 0.437<strong>–</strong>1.282), respectively. A detection probability of 0.50 (CV =<br />
15%, 95% CI = 0.372<strong>–</strong>0.681) occurred within the study area.<br />
DISCUSSION<br />
Our study produced a new abundance estimate for a species of conservation concern<br />
in a region where such estimates are scarce and without governmental financial<br />
support. Only two prior studies from the 1990s, both using photo-identification,<br />
Figure 2. Histogram of perpendicular sighting distances truncated at 90 m and the fitted<br />
detection function for the best fitting model, MCDS half-normal model (no adjustment<br />
parameters) with sea state as a covariate. One sighting at 200 m was truncated.
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 613<br />
Table 1. Model results for each of the tested models. Data were ungrouped, pooled across all surveys sets, and truncated at 90 m (n = 33).<br />
No. of parameters<br />
Model<br />
(Key + Adjustment) Key Adjustments AIC Ns<br />
ˆ CV(%) Dˆ CV(%) Ds<br />
ˆ CV(%) P CV(%) K-S p<br />
CDS<br />
Uniform + cosine 0 1 287.75 220 25.6 0.763 25.6 0.306 21.2 0.57 10.0 0.195<br />
Uniform + simple polynomial 0 1 288.04 204 25.0 0.707 25.0 0.272 20.5 0.64 9.0 0.153<br />
Half-normal + cosine 1 0 287.35 227 27.0 0.785 27.0 0.3<strong>17</strong> 22.8 0.55 13.0 0.210<br />
Half-normal + hermite 1 0 287.35 227 27.0 0.785 27.0 0.3<strong>17</strong> 22.8 0.55 13.0 0.210<br />
MCDS with Beaufort sea state as a covariate<br />
Half-normal + cosine 1 0 285.05 216 27.7 0.749 27.7 0.347 23.8 0.50 15.0 0.246<br />
Half-normal + hermite 1 0 285.05 216 27.7 0.749 27.7 0.347 23.8 0.50 15.0 0.246<br />
Hazard-rate + cosine 2 0 291.05 216 28.0 0.749 28.0 0.347 24.1 0.52 15.0 0.243<br />
Note: AIC = Akaike Information Criterion; ˆ Ns = abundance estimate for survey area; ˆ D = estimated density of dolphins/km 2 ; ˆ Ds = estimated<br />
density of clusters/km 2 ; P = detection probability; CV = coefficient of variation; K-S p = Kolmogorov<strong>–</strong>Smirnov goodness-of-fit p-value.
614 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
report abundance estimates for regional bottlenose dolphin populations in Belize:<br />
Campbell et al. (2002) estimated 82<strong>–</strong>86 individuals at Turneffe Atoll and Kerr<br />
et al. (2005) reported 122 individuals in the Drowned Cayes, located 16 km west<br />
of Turneffe within the Mesoamerican Barrier Reef. The different dolphin abundance<br />
estimates for Turneffe is not surprising, primarily because the two methods measure<br />
slightly different things and are therefore, not directly comparable. Mark-recapture<br />
sampling estimates the abundance of the overall biological population whether or<br />
not all individuals are present at a moment in time, while line transect sampling<br />
estimates the size of the population within the study area during the survey interval<br />
(Calambokidis and Barlow 2004). Nevertheless, the abundance estimates for<br />
this region are small and suggest the population cannot withstand high levels of<br />
mortality.<br />
To our knowledge there are no reports of hunting, fisheries bycatch, or boat strikes<br />
of dolphins at Turneffe, and Campbell et al. (2002) reported an absence of crescentshaped<br />
scars indicative of shark predation. Moreover, bottlenose dolphins are fully<br />
protected under Belize’s 1981 Wildlife Protection Act. This would suggest threats to<br />
the dolphins at Turneffe are currently minimal. However, mounting evidence indicates<br />
that continued unsustainable fishing (overfishing and illegal fishing) and rapid<br />
development are impacting the ecological integrity of the Atoll (World Resources<br />
Institute 2005, Granek 2006). Within Belize, fisheries are primarily artisanal; ongoing<br />
commercial fisheries heavily and unsustainably exploit Caribbean spiny lobster<br />
and queen conch, and a small-scale fishery targets snapper, grouper, and other fish<br />
species (Gillet 2003, FAO 2005). Coblentz (1997) demonstrated artisanal fisheries<br />
to be unsustainable and can quickly alter reef communities. Furthermore, gill nets,<br />
recognized as a leading cause of cetacean mortality worldwide (Read et al. 2006),<br />
are one of several techniques used in this fishery, often by illegal Guatemalan and<br />
Honduran fishers (Gillet 2003, FAO 2005, Perez 2009). The threat of entanglement<br />
in gill nets is cause for concern to bottlenose dolphin populations in Turneffe and<br />
the rest of Belize.<br />
Removal of mangroves and dredging of seagrass beds for private and commercial<br />
development are increasing at Turneffe. As important nursery areas for reef fish<br />
(e.g., snappers, grunts, and parrotfish), elimination of either habitat type can have<br />
significant impacts to adjacent reef fish communities and may lead to cascading<br />
effects at higher trophic levels (Nagelkerken et al. 2002, Mumby et al. 2004, Manson<br />
et al. 2005). Although quantitative diet studies for Turneffe dolphins do not yet<br />
exist, if unsustainable fisheries and development are left unchecked, changes to<br />
the atoll’s ecosystem could potentially lead to prey depletion. This is especially<br />
concerning because females with neonatal and older, but presumed nursing, calves<br />
are sighted year-round suggesting the atoll may be an important calving area (Grigg<br />
and Markowitz 1997, Campbell et al. 2002). Prey depletion could have negative<br />
ramifications on reproductive success since lactating females have much higher energy<br />
requirements than non nursing individuals (Oftedal 1984, Cheal and Gales 1991).<br />
On a larger scale, site fidelity patterns indicate a high proportion of dolphins<br />
are transient and use a much larger area than Turneffe (Campbell et al. 2002).<br />
Belize is a small western Caribbean country and the dolphins observed at Turneffe<br />
could move beyond the country’s borders. Extensive gill net use occurs in Mexican,<br />
Honduran, and Guatemalan artisanal fisheries and bottlenose dolphins have been<br />
historically found entangled or dead in these nets (Vidal et al. 1994, FAO 2000).<br />
To better understand the impacts of these threats on Turneffe dolphins, surveys<br />
should continue at the atoll in conjunction with studies that focus on dolphin
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 615<br />
movement beyond the atoll, including a Belize-wide bottlenose dolphin population<br />
assessment, and the development of a method for monitoring and reporting fisheries<br />
interactions.<br />
Potential Violations of Line Transect Assumptions and Considerations for Future Surveys<br />
Survey design—Although it is possible to design a survey based on the amount<br />
of effort required to attain a certain level of precision (Buckland et al. 2001), often<br />
practical considerations will dictate the survey design, as was the case in this study.<br />
This is the first rigorous systematic survey conducted at Turneffe and its use is<br />
expected to continue. The choice of an equally spaced zig-zag design, with a few<br />
minor adjustments (see further), was considered successful.<br />
One particular issue arose during field surveys that required additional consideration.<br />
The full length of most transects between the mangrove shoreline and the<br />
fringing reef on the western side and several areas within the lagoons could not be<br />
surveyed; coral heads and shallow sand bars made these areas inaccessible. Instead,<br />
the areas were scanned for dolphins from the closest point that could be reached<br />
by the boat. Thomas et al. (2007) reported a similar experience and suggested this<br />
problem may be avoided by using high-resolution maps during the survey design<br />
process; however, such items are not available for this region. The issue was resolved<br />
by excluding the unsurveyed areas from the total line length traveled during analysis<br />
(Thomas et al. 2007). Because unsurveyed areas often occurred at the apexes of the<br />
zigzags, exclusion of these areas had the added benefit of further improving even<br />
coverage probability (Dawson et al. 2008).<br />
Abundance estimation—Visibility bias or incomplete detection at distance zero<br />
(g(0) < 1) can be problematic in marine mammal surveys and cause negatively<br />
biased estimates. This bias is classified in two ways: (1) animals may not be available<br />
to be seen by observers (availability bias) because they are not at the water’s surface<br />
where they can be seen (e.g., during diving), and (2) animals may potentially be<br />
visible to an observer but are not detected (perception bias) because of factors such as<br />
environmental conditions (e.g., sea state) (Marsh and Sinclair 1989). Consequently,<br />
species with longer dive times or found in smaller clusters are more likely to be<br />
missed by observers, thereby impacting the assumption g(0) = 1 (Dawson et al.<br />
2008). Bottlenose dolphins have relatively short dive durations in shallow areas. For<br />
example, along Florida’s Gulf Coast in areas with depths ranging between 1 and 5 m,<br />
average dive times for this species were recorded to be 20<strong>–</strong>25 s off Sanibel Island<br />
(Shane 1990), 30<strong>–</strong>40 s in Sarasota Bay (Irvine et al. 1981), and 28.5 s in Tampa Bay<br />
(Mate et al. 1995). Similar environmental conditions, including shallow waters, sand<br />
flats, and seagrass beds, occur at Turneffe Atoll; therefore dive times are likely to be<br />
comparable to those recorded in Florida. This short dive duration makes it unlikely<br />
that dolphins missed detection on the transect line because of being underwater. In<br />
addition, despite the small mean observed cluster size, almost 85% of the sightings<br />
were of clusters with ≤3 dolphins indicating smaller group sizes were not being<br />
missed. Introduction of significant bias to the abundance and density estimates from<br />
the assumption g(0) = 1 is, therefore, not expected.<br />
Responsive animal movement either toward or away from the survey vessel prior<br />
to detection will introduce bias (Turncock and Quinn 1991, Palka and Hammond<br />
2001). Bottlenose dolphins are known to bow-ride, and in at least 15% of the<br />
sightings, dolphins were first detected as they were quickly approaching the boat
616 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
and just prior to bow-riding. Attraction to vessels is seen in other small cetacean<br />
species including Dall’s porpoise (Phocoenoides dalli, Turncock and Quinn 1991),<br />
Pacific white-sided dolphins (Lagenorhynchus obliquidens, Williams and Thomas 2007),<br />
and white-beaked dolphins (Lagenorhynchus albirostris, Palka and Hammond 2001).<br />
Lemon et al. (2006) noted behavioral changes by Indo-Pacific bottlenose dolphins<br />
(T. aduncus) to an approaching boat at 100 m. Calculated distances from this study<br />
found all but one sighting occurred at ≤90 m, suggesting that the presence of the<br />
boat may have already attracted dolphins to the boat prior to detection. During future<br />
surveys, every effort should be made to ensure dolphins are sighted as far ahead of<br />
the vessel as possible. Correction factors, to account for attractive movement and<br />
positively biased estimates, should be developed through the collection of animal<br />
orientation data (Palka and Hammond 2001).<br />
Measurement accuracy also influences estimates and can be problematic regardless<br />
of survey platform type, although it can be exacerbated in small-boat surveys. Estimation<br />
of distances and angles is common (Vidal et al. 1997, Hammond et al. 2002);<br />
however, it may lead to measurement rounding and biased estimates. For this reason,<br />
estimation is not recommended unless observers are well trained and continually<br />
tested throughout the survey or correction factors are developed (Buckland et al.<br />
2001, Dawson et al. 2008). Lerczak and Hobbs (1998) and Buckland et al. (2001)<br />
describe several acceptable methods for acquiring measurements using tools such as<br />
angle boards and binoculars with reticles and/or compasses. An angle board was used<br />
to measure sighting angles and it helped to avoid angle rounding. A nonstandard<br />
method was used to determine sighting distances due to site-specific limitations,<br />
primarily that the horizon was rarely visible due to the enclosed nature of the atoll.<br />
A clinometer was chosen, as it is self-leveling and does not require the horizon as a<br />
reference point. However, declination angles
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 6<strong>17</strong><br />
other small cetacean abundance studies (e.g., 27.9% short-beaked common dolphin<br />
(Delphinus delphis), 48.1% bottlenose dolphin (Barlow 1995); 15.7% Hector’s dolphin<br />
(Cephalorhynchus hectori) (Dawson et al. 2004); 29.2% Dall’s porpoise, 35.3% Pacific<br />
white-sided dolphin (Williams and Thomas 2007), the abundance (216 dolphins,<br />
CV = 27.7%) and density (0.749 dolphins/km 2 ,CV= 27.7%) reported here for<br />
Turneffe Atoll are the best estimates given the conditions and should be considered<br />
preliminary. The development of correction factors to help account for potential<br />
line transect analysis violations is strongly recommended. Future surveys would also<br />
benefit from a higher observation platform and the consistent use of binoculars;<br />
however, this should only be done after careful thought as protocol modification after<br />
six survey sets will impact the ability to detect trends.<br />
Being able to obtain reliable abundance estimates and identify critical habitats<br />
are vital to the creation and success of marine protected areas for cetaceans (Hoyt<br />
2005). Our results provide an up-to-date population assessment for the area and<br />
should serve as a baseline as Belize moves forward with the development of a new<br />
conservation initiative to create a comprehensive and resilient marine protected area<br />
system. The year-round presence of dolphins, including mom/calf pairs (Grigg and<br />
Markowitz 1997, Campbell et al. 2002), suggests this area may be an important<br />
calving/nursery site and should be a priority for protection. Moreover, the high<br />
frequency of sightings near mangrove shorelines and atoll openings, 2 suggests these<br />
areas are important habitat features to the dolphins and should remain undeveloped.<br />
This and other recent studies (Vidal et al. 1997; Dawson et al. 2004, 2008; Williams<br />
and Thomas 2009) have shown that line transect surveys can be successfully modified<br />
for small-boats, provided the survey design is well planned and field methods are<br />
designed to address the main assumptions of line transect analysis. As coastal human<br />
populations continue to grow and the threats to small cetacean species increase, the<br />
reduction in cost and the improved accessibility to shallow areas through the use<br />
of small-boats will provide more opportunities for underdeveloped nations to assess<br />
their coastal marine mammal populations.<br />
ACKNOWLEDGMENTS<br />
We thank L. Thomas for providing helpful guidance during the survey design phase,<br />
S. G. Allen, B. Holzman, and H. Edwards for their reviews of earlier drafts of this paper,<br />
M. Christman, C. Keller, and P. Kubilis for their statistical advice, and R. Williams and an<br />
anonymous reviewer for comments on the manuscript. We are grateful to the many Oceanic<br />
Society and Elderhostel volunteers who helped in the field. Financial and logistical support<br />
was provided by Oceanic Society Expeditions. D. Dick received financial support from San<br />
Francisco State University’s Department of Geography and the College of Behavioral and<br />
Social Sciences Graduate Stipend. All research was approved by the Office for the Protection<br />
of Human and Animal Subjects at SFSU and conducted under a research permit issued to the<br />
Oceanic Society from the Ministry of Natural Resources Forest Department, Belize.<br />
LITERATURE CITED<br />
Aragones, L. V., T. A. Jefferson and H. Marsh. 1997. Marine mammal survey techniques<br />
applicable in developing countries. Asian Marine Biology 14:15<strong>–</strong>39.<br />
2 1 Unpublished data provided by Dorothy Dick, 5227 /2 27th Avenue S, Gulfport, Florida 33707,<br />
July 2009.
618 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
Barlow, J. 1988. Harbor porpoise, Phocoena phocoena, abundance estimation for California,<br />
Oregon, and Washington: I. ship surveys. Fishery Bulletin 86:4<strong>17</strong><strong>–</strong>432.<br />
Barlow, J. 1995. The abundance of cetaceans in California waters. Part 1: Ship surveys in<br />
summer and fall of 1991. Fishery Bulletin 93:1<strong>–</strong>14.<br />
Barlow, J., C. W. Oliver, T. D. Jackson and B. L. Taylor. 1988. Harbor porpoise, Phocoena phocoena,<br />
abundance estimation for California, Oregon, and Washington: II. aerial surveys.<br />
Fishery Bulletin 86:433<strong>–</strong>444.<br />
Barlow, J., T. Gerrodette and J. Forcada. 2001. Factors affecting perpendicular sighting<br />
distances on shipboard line-transect surveys for cetaceans. Journal of Cetacean Research<br />
and Management 3:201<strong>–</strong>212.<br />
Bearzi, G., and C. M. Fortuna. 2006. Common bottlenose dolphin Tursiops truncatus (Mediterranean<br />
subpopulation). Pages 64<strong>–</strong>73 in R. R. Reeves and G. N. di Sciara, eds. The status<br />
and distribution of cetaceans in the Black Sea and Mediterranean Sea. IUCN Centre for<br />
Mediterranean Cooperation, Malaga, Spain.<br />
Bearzi, G., S. Agazzi, S. Bonizzoni, M. Costa and A. Azzellino. 2008. Dolphins in a bottle:<br />
Abundance, residency patterns and conservation of bottlenose dolphins Tursiops truncatus<br />
in the semi-closed eutrophic Amvrakikos Gulf, Greece. Aquatic Conservation: Marine<br />
and Freshwater Ecosystems 18:130<strong>–</strong>146.<br />
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers and L. Thomas.<br />
2001. Introduction to distance sampling: Estimating abundance of biological populations.<br />
Oxford University Press, Oxford, U.K.<br />
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers and L. Thomas.<br />
2004. Advanced distance sampling: Estimating abundance of biological populations.<br />
Oxford University Press, Oxford, U.K.<br />
Calambokidis, J., and J. Barlow. 2004. Abundance of blue and humpback whales in the<br />
eastern North Pacific estimated by capture-recapture and line-transect methods. Marine<br />
Mammal Science 20:63<strong>–</strong>85.<br />
Campbell, G. S., B. A. Bilgre and R. H. Defran. 2002. Bottlenose dolphins (Tursiops truncatus)<br />
in Turneffe Atoll, Belize: Occurrence, site fidelity, group size, and abundance. Aquatic<br />
Mammals 28:<strong>17</strong>0<strong>–</strong>180.<br />
Cheal, A. J., and N. J. Gales. 1991. Body mass and food intake in captive, breeding bottlenose<br />
dolphins, Tursiops truncatus. Zoo Biology 10:451<strong>–</strong>456.<br />
Coblentz, B. E. 1997. Consumption of coral reef fish suggests non-sustainable extraction.<br />
Conservation Biology 11:559<strong>–</strong>561.<br />
Currey, R. J. C., S. M. Dawson and E. Slooten. 2009. An approach for regional threat<br />
assessment under IUCN Red List criteria that is robust to uncertainty: The Fiordland<br />
bottlenose dolphins are critically endangered. Biological Conservation 142:1570<strong>–</strong>1579.<br />
Dawson, S., E. Slooten, S. Dufresne and D. Clement. 2004. Small-boat surveys for coastal<br />
dolphins: Line-transect surveys for Hector’s dolphins (Cephalorhynchus hectori). Fisheries<br />
Bulletin 201:441<strong>–</strong>451.<br />
Dawson, S., P. Wade, E. Slooten and J. Barlow. 2008. Design and field methods for sighting<br />
surveys of cetaceans in coastal and riverine habitats. Mammal Review 38:19<strong>–</strong>49.<br />
ESRI. 2006. ArcInfo 9.2, 1999<strong>–</strong>2006. Environmental Systems Research Institute, Redlands,<br />
CA.<br />
Estes, J. A., M. T. Tinker, T. M. Williams and D. F. Doak. 1998. Killer whale predation on<br />
sea otters linking oceanic and nearshore ecosystems. Science 282:473<strong>–</strong>476.<br />
FAO. 2000. FAO fishery country profile: The Republic of Guatemala. Available at http://<br />
www.fao.org.fi/oldsite/FCP/en/GTM/profile.htm (accessed 14 March 2010).<br />
FAO. 2005. FAO fishery country profile: Belize. Available at http://www.fao.org/fi/<br />
oldsite/FCP/en/BLZ/profile.htm (accessed 14 March 2010).<br />
Forcada, J., M. Gazo, A. Aguilar, J. Gonzalvo and M. Fernández-Contreras. 2004. Bottlenose<br />
dolphin abundance in the NW Mediterranean: Addressing heterogeneity in distribution.<br />
Marine Ecology Progress Series 275:275<strong>–</strong>287.<br />
Gillet, V. 2003. The fisheries of Belize. Fisheries Centre Research Reports 11:141<strong>–</strong>147.
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 619<br />
Gischler, E. 2003. Holocene lagoonal development in the isolated carbonate platforms off<br />
Belize. Sedimentary Geology 159:113<strong>–</strong>132.<br />
Granek, E. F. 2006. Linkages between mangroves forests and coral reeds: quantifying disturbance<br />
effects and energy flow between systems. Ph.D. dissertation, Oregon State<br />
University, Corvalis, OR. 192 pp.<br />
Grigg, E., and H. Markowitz. 1997. Habitat use by bottlenose dolphins (Tursiops truncatus)<br />
at Turneffe Atoll, Belize. Aquatic Mammals 23:163<strong>–</strong><strong>17</strong>0.<br />
Hammond, P. S., P. Berggren, H. Benke, et al. 2002. Abundance of harbour porpoise and other<br />
cetaceans in the North Sea and adjacent waters. Journal of Applied Ecology 39:361<strong>–</strong>376.<br />
Hammond, P. S., G. Bearzi, A. Bjørge, et al. 2008. Tursiops truncatus. IUCN Red List of<br />
Threatened Species. Version 2009.2. Available at http://www.iucnredlist.org (accessed<br />
15 February 2010).<br />
Hines, E. M., K. Adulyanukosol and D. A. Duffus. 2005. Dugong (Dugong dugon) abundance<br />
along the Andaman Coast of Thailand. Marine Mammal Science 21:536<strong>–</strong>549.<br />
Hoyt, E. 2005. Marine protected areas for whales, dolphins and porpoises: A world handbook<br />
for cetacean habitat conservation. Earthscan, London, U.K.<br />
Irvine, A. B., M. D. Scott, R. S. Wells and J. H. Kaufmann. 1981. Movements and activities<br />
of the Atlantic bottlenose dolphin, Tursiops truncatus, near Sarasota, Florida. Fishery<br />
Bulletin 79:671<strong>–</strong>688.<br />
Kerr, K. A., R. H. Defran and G. S. Campbell. 2005. Bottlenose dolphins (Tursiops truncatus)in<br />
the Drowned Cayes, Belize: Group size, site fidelity, and abundance. Caribbean Journal<br />
of Science 41:<strong>17</strong>2<strong>–</strong><strong>17</strong>7.<br />
Lemon, M., T. P. Lynch, D. H. Cato and R. G. Harcourt. 2006. Response of traveling<br />
bottlenose dolphins (Tursiops aduncus) to experimental approaches by a powerboat in<br />
Jervis Bay, New South Wales, Australia. Biological Conservation. 127:363<strong>–</strong>372.<br />
Lerczak, J. A., and R. C. Hobbs. 1998. Calculating sightings distances from angular readings<br />
during shipboard, aerial, and shore-based marine mammal surveys. Marine Mammal<br />
Science 14:590<strong>–</strong>599.<br />
Manson, F. J., N. R. Loneragan, G. A. Skilleter and S. R. Phinn. 2005. An evaluation of<br />
the evidence for linkages between mangroves and fisheries: A synthesis of the literature<br />
and identification of research directions. Oceanography and Marine Biology: An Annual<br />
Review 43:485<strong>–</strong>515.<br />
Marques, F. F. C., and S. T. Buckland. 2003. Incorporating covariates into standard line<br />
transect analysis. Biometrics 59:924<strong>–</strong>935.<br />
Marsh, H., and D. F. Sinclair. 1989. Correcting for visibility bias in strip transect aerial<br />
surveys of aquatic fauna. Journal of Wildlife Management 53:10<strong>17</strong><strong>–</strong>1024.<br />
Mate, B. R., K. A. Rossbach, S. L. Nieukirk, R. S. Wells, A. B. Irvine, M. D. Scott and A. J.<br />
Read. 1995. Satellite-monitored movements and dive behavior of a bottlenose dolphin<br />
(Tursiops truncatus) in Tampa Bay, Florida. Marine Mammal Science 11:452<strong>–</strong>463.<br />
Minnesota Department of Natural Resources. 2008. DNR Garmin, version 5.1.1. The Minnesota<br />
Department of Natural Resources, St. Paul, MN. Available at http://www.<br />
dnr.state.mn.us/mis/gis/tools/arcview/extensions/DNRGarmin/DNRGarmin.html (accessed<br />
18 October 2008).<br />
Mullin, K. D., and G. L. Fulling. 2004. Abundance of cetaceans in the oceanic northern Gulf<br />
of Mexico, 1996<strong>–</strong>2001. Marine Mammal Science 20:787<strong>–</strong>807.<br />
Mumby, P. J., A. J. Edwards, J. E. Arias-González, et al.. 2004. Mangroves enhance the<br />
biomass of coral reef fish communities in the Caribbean. Nature 427:533<strong>–</strong>536.<br />
Nagelkerken, I., C. M. Roberts, G. Van Der Velde, M. Dorenbosch, M. C. van Riel, E.<br />
Cocheret de la Morinière and P. H. Nienhuis. 2002. How important for mangroves and<br />
seagrass beds for coral-reef fish? The nursery hypothesis tested on an island scale. Marine<br />
Ecology Progress Series 244:299<strong>–</strong>305.<br />
NASA Landsat Program. 2004. Landsat ETM+ Scene p018r048_7t20001108, Orthorectified,<br />
Geocover. EarthSat. Rockville, MD. Acquisition date 11/8/2000. Source for this data<br />
set Global land Cover Facility, University of Maryland, College Park, MD. Available at<br />
http://www.landcover.org (accessed 24 May 2005).
620 MARINE MAMMAL SCIENCE, VOL. 27, NO. 3, 2011<br />
Oftedal, O. T. 1984. Milk composition, milk yield and energy output at peak lactation:<br />
A comparative review. Pages 33<strong>–</strong>85 in M. Peaker, R. G. Vernon and C. H. Knight,<br />
eds. Physiological strategies in lactation. Symposia of the Zoological Society of London,<br />
number 51. Academic Press, London, U.K.<br />
Palka, D. 1996. Effects of Beaufort sea state on the sightability of harbor porpoises in the<br />
Gulf of Maine. Report of the <strong>International</strong> Whaling Commission 46:575<strong>–</strong>582.<br />
Palka, D. L., and P. S. Hammond. 2001. Accounting for responsive movement in line transect<br />
estimates of abundance. Canadian Journal of Fisheries and Aquatic Science 58:777<strong>–</strong>787.<br />
Perez, A. 2009. Fisheries management at the tri-national border between Belize, Guatemala,<br />
and Honduras. Marine Policy. 33:195<strong>–</strong>200.<br />
Platt, S. G., T. R. Rainwater, B. W. Miller and C. M. Miller. 2000. Notes on the mammals<br />
of Turneffe Atoll, Belize. Caribbean Journal of Science 36:166<strong>–</strong>168.<br />
Read, A. J., P. Drinker and S. Northridge. 2006. Bycatch of marine mammals in U.S. and<br />
global fisheries. Conservation Biology 20:163<strong>–</strong>169.<br />
Reeves, R. R., B. D. Smith, E. A. Crespo and G. N. di Sciara (compilers). 2003. Dolphins,<br />
whales and porpoises: 2002<strong>–</strong>2010 Conservation Action Plan for the World’s Cetaceans.<br />
IUCN/SSC Cetacean Specialist Group. IUCN, Gland, Switzerland.<br />
Shane, S. H. 1990. Behavior and ecology of the bottlenose dolphin at Sanibel Island, Florida.<br />
Pages 245<strong>–</strong>265 in S. Leatherwood and R. R. Reeves, eds. The bottlenose dolphin.<br />
Academic Press, San Diego, CA.<br />
Sini, M. I., S. J. Canning, K. A. Stockin and G. J. Pierce. 2005. Bottlenose dolphins around<br />
Aberdeen harbour, north-east Scotland: A short study of habitat utilization and the<br />
potential effects of boat traffic. Journal of the Marine Biological Association of the<br />
United Kingdom 85:1547<strong>–</strong>1554.<br />
Stoddart, D. R. 1962. Three Caribbean atolls: Turneffe Islands, Lighthouse Reef, and Glover’s<br />
Reef, British Honduras. Atoll Research Bulletin 87:1<strong>–</strong>151.<br />
Strindberg, S., and S. T. Buckland. 2004. Zigzag survey designs in line transect sampling.<br />
Journal of Agricultural, Biological, and Environmental Statistics 9:443<strong>–</strong>461.<br />
Thomas, L., S. T. Buckland, K. P. Burnham, D. R. Anderson, J. L. Laake, D. L. Borchers<br />
and S. Strindberg. 2002. Distance sampling. Encyclopedia of Environmetrics 1:544<strong>–</strong><br />
553.<br />
Thomas, L., R. Williams and D. Sandilands. 2007. Designing line transect surveys for complex<br />
survey regions. Journal of Cetacean Research and Management 9:1<strong>–</strong>11.<br />
Thomas, L., S. T. Buckland, E. A. Rexstad, J. L. Laake, S. Strindberg, S. Hedley, J. R. B.<br />
Bishop, T. A. Marques and K. P. Burnham. 2010. Distance software: Design and analysis<br />
of distance sampling surveys for estimating population size. Journal of Applied Ecology<br />
47:5<strong>–</strong>14.<br />
Turncock, B. J., and T. J. Quinn II. 1991. The effect of responsive movement on abundance<br />
estimation using line transect sampling. Biometrics 47:701<strong>–</strong>715.<br />
Van Waerebeek, K., M. F. Van Bressem, F. Felix, J. Alfaro-Shigueto, A. Garcia-Godos, L.<br />
Chávez-Lisambart, K. Ontón, D. Montes and R. Bello. 1997. Mortality of dolphins<br />
and porpoises in coastal fisheries off Peru and southern Ecuador in 1994. Biological<br />
Conservation 81:43<strong>–</strong>49.<br />
Vidal, O. 1993. Aquatic mammal conservation in Latin America: Problems and perspectives.<br />
Conservation Biology 7:788<strong>–</strong>795.<br />
Vidal, O., K. Van Waerebeek and L. T. Findley. 1994. Cetaceans and gillnet fisheries in<br />
Mexico, Central America and the wider Caribbean: A preliminary review. Report of the<br />
<strong>International</strong> Whaling Commission (Special Issue 15): 221<strong>–</strong>233.<br />
Vidal, O., J. Barlow, L. A. Hurtado, J. Torre, P. Cendón and Z. Ojeda. 1997. Distribution<br />
and abundance of the Amazon River dolphin (Inia geoffrensis) and the tucuxi (Sotalia<br />
fluviatilis) in the Upper Amazon River. Marine Mammal Science 13:427<strong>–</strong>445.<br />
Wells, R. S., and M. D. Scott. 1999. Bottlenose dolphin Tursiops truncatus (Montagu, 1821).<br />
Pages 137<strong>–</strong>182 in S. H. Ridgway and Sir R. Harrison, eds. Handbook of marine<br />
mammals. Volume 6. The second book of dolphins and the porpoises. Academic Press,<br />
San Diego, CA.
DICK AND HINES: BOTTLENOSE DOLPHIN ABUNDANCE 621<br />
Wilson, B., P. M. Thompson and P. S. Hammond. 1997. Habitat use by bottlenose dolphins:<br />
Seasonal distribution and stratified movement patterns in the Moray Firth, Scotland.<br />
Journal of Applied Ecology 34:1365<strong>–</strong>1374.<br />
Williams, R., and L. Thomas. 2007. Distribution and abundance of marine mammals in<br />
the coastal waters of British Columbia, Canada. Journal of Cetacean Research and<br />
Management 9:15<strong>–</strong>28.<br />
Williams, R., and L. Thomas. 2009. Cost-effective abundance estimation of rare animals: Testing<br />
performance of small-boat surveys for killer whales in British Columbia. Biological<br />
Conservation 142:1542<strong>–</strong>1547.<br />
Williams, R., R. Leaper, A. N. Zerbini and P. S. Hammond. 2007. Methods for investigating<br />
measurement error in cetacean line-transect surveys. Journal of the Marine Biological<br />
Association of the United Kingdom 87:313<strong>–</strong>320.<br />
World Resources Institute. 2005. Belize coastal threats atlas. 20 pp. Available at<br />
http://pdf.wri.org/belize_threat_atlas.pdf (accessed 14 March 2010).<br />
Zerbini, A. N., J. M. Waite, J. W. Durban, R. Leduc, M. E. Dahlheim and P. R. Wade. 2007.<br />
Estimating abundance of killer whales in the nearshore waters of the Gulf of Alaska and<br />
Aleutian Islands using line-transect sampling. Marine Biology 150:1033<strong>–</strong>1045.<br />
Received: 3 August 2009<br />
Accepted: 16 July 2010
Journal of Zoology<br />
The lesser of two evils: seasonal migrations of Amazonian<br />
manatees in the Western Amazon<br />
E. M. Arraut 1,2 , M. Marmontel 3 , J. E. Mantovani 4 , E. M.L.M. Novo 1 , D. W. Macdonald 2 & R. E. Kenward 5<br />
1 Earth Observation General Coordination, Remote Sensing Division, National Institute for Space Research, São José dos Campos, São Paulo, SP,<br />
Brazil<br />
2 Wildlife Conservation Research Unit, Zoology Department, The Recanati-Kaplan Centre, University of Oxford, Tubney, Oxfordshire, UK<br />
3 Mamirauá Sustainable Development Institute, Tefé, AM, Brazil<br />
4 National Institute for Space Research, Natal Regional Center, Natal, RN, Brazil<br />
5 Centre for Ecology and Hydrology, Wallingford, Oxfordshire, UK<br />
Keywords<br />
Trichechus inunguis; habitat selection;<br />
remote sensing; geographical information<br />
system; Sirenia; migration.<br />
Correspondence<br />
Eduardo Moraes Arraut, Earth Observation<br />
General Coordination, Remote Sensing<br />
Division, National Institute for Space<br />
Research, Avenida dos Astronautas, 1 758,<br />
Jardim da Granja, PO Box 515, CEP 12227-<br />
101, São José dos Campos, SP, Brazil. Tel:<br />
55 12 3945 6486; Fax: 55 12 3945 6488<br />
Email: arraut@dsr.inpe.br or<br />
arraut@gmail.com<br />
Editor: Andrew Kitchener<br />
Received 9 March 2009; revised 28 August<br />
2009; accepted 13 September 2009<br />
doi:10.1111/j.1469-7998.2009.00655.x<br />
Introduction<br />
Abstract<br />
Migrationisanadaptationtoenvironmentsinwhichhabitat<br />
quality in different regions changes asynchronously in space<br />
and/or time (Dingle & Drake, 2007). This implies that habitat<br />
quality in the destination will be better than that at the origin,<br />
but not necessarily that it must be good. Here, we show that<br />
Amazonian manatees that live in the region of the Mamirauá<br />
and Amana˜ Sustainable Development Reserves (RDSM and<br />
RDSA, respectively) are subject to challenging habitat conditions<br />
during part of the year, and that they migrate into an<br />
area that is their best option under difficult circumstances.<br />
Journal of Zoology. Print ISSN 0952-8369<br />
We investigated the paradox of why Amazonian manatees Trichechus inunguis<br />
undergo seasonal migrations to a habitat where they apparently fast. Ten males<br />
were tracked using VHF telemetry between 1994 and 2006 in the Mamirauá and<br />
Amana˜ Sustainable Development Reserves, constituting the only long-term<br />
dataset on Amazonian manatee movements in the wild. Their habitat was<br />
characterized by analysing aquatic space and macrophyte coverage dynamics<br />
associated with the annual flood-pulse cycle of the River Solimo˜es. Habitat<br />
information came from fieldwork, two hydrographs, a three-dimensional model<br />
of the water bodies and classifications of Landsat-TM/ETM + images. We show<br />
that during high-water season (mid-May to end-June), males stay in várzea lakes in<br />
association with macrophytes, which they select. We then show that, during lowwater<br />
(October<strong>–</strong>November), the drastic reduction in aquatic space in the várzea<br />
leads to the risk of their habitat drying out and increases the manatees’<br />
vulnerability to predators such as caimans, jaguars and humans. This explains<br />
why males migrate to Ria Amana˜. Based on data on illegal hunting, we argue that<br />
this habitat variability influences females to migrate too. We then use published<br />
knowledge of the environment’s dynamics to argue that when water levels are high,<br />
the habitats that can support the largest manatee populations are the várzeas of<br />
white-water rivers, and we conjecture that rias are the species’ main low-water<br />
refuges throughout Western Amazonia. Finally, we warn that the species may be<br />
at greater risk than previously thought, because migration and low-water levels<br />
make manatees particularly vulnerable to hunters. Moreover, because the flooding<br />
regime of Amazonian rivers is strongly related to large-scale climatic phenomena,<br />
there might be a perilous connection between climate change and the future<br />
prospects for the species. Our experience reveals that the success of research and<br />
conservation of wild Amazonian manatees depends on close working relationships<br />
with local inhabitants.<br />
We studied the influence of seasonal habitat variation on<br />
the migration of the Amazonian manatee Trichechus inunguis,<br />
which is the only member of the order Sirenia that lives<br />
exclusively in freshwater (Bertram & Bertram, 1973). Its<br />
distribution spans the Amazon basin, from Ecuador (Timm,<br />
Albuja & Clauson, 1986) and Peru (Reeves et al., 1996) to<br />
the Atlantic coast of Brazil (Best, 1984). Phylogenetic<br />
studies suggest that manatees from the mid-Solimo˜es, midand<br />
low-Amazonas form a single panmictic population<br />
(Cantanhede et al., 2005).<br />
Amazonian manatees are herbivores that in the RDSM<br />
and RDSA have been reported to feed on 63 species of<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 247
Seasonal migration of Amazonian manatees E. M. Arraut et al.<br />
aquatic macrophytes (annual freshwater plants) (Guterres &<br />
Marmontel, 2008). Adults reach up to 3 m in length and<br />
450 kg in weight (Caldwell & Caldwell, 1985), and consume<br />
about 8% of their body weight in aquatic macrophytes per<br />
day (Rosas, 1994). Because of strong habitat seasonality (see<br />
‘Materials and methods’), Amazonian manatees face an<br />
annual period of food shortage, which may last 7 months,<br />
during which they apparently fast (Best, 1983).<br />
The first tracking of an Amazonian manatee was carried<br />
out by Montgomery, Best & Yamakoshi (1981). A juvenile<br />
male was captured in the wild, kept in captivity for<br />
20 months (approximately half its life) and then released<br />
during the rising-water period in a várzea (a floodplain of a<br />
river with nutrient-rich and high-silt content water) near<br />
Manaus (a different region from where it had been captured).<br />
The manatee was tracked for 20 days, during which it<br />
spent most of its time feeding on aquatic macrophytes, and<br />
moved at a similar rate by day and by night.<br />
The Amazonian manatee’s migration was first reported<br />
by Marmontel et al. (2002), who tracked five adult males<br />
during a period of four and a half years. This showed that<br />
individuals migrated each year between várzea lakes, where<br />
they spent the high-water period, and Ria Amanã (a ria is<br />
long narrow lake formed by the partial submergence of a<br />
river valley), where they spent the low-water period (see<br />
‘Materials and methods’). Here, we extend the initial sample<br />
to 10 radio-tagged manatees to show (1) new migratory<br />
routes and (2) that the migration pattern remains consistent.<br />
We also indicate associations between Amazonian manatee<br />
248<br />
migratory movements and (3) availability of preferred food<br />
and (4) aquatic space reduction and predator aggregation.<br />
On that basis, we propose that the migratory behaviour,<br />
which is paradoxical insofar as animals travel to a place<br />
where, seemingly, they cannot eat, is a balance of feeding<br />
and predator avoidance. Finally, in the light of our results,<br />
we discuss the next steps for the species’ conservation.<br />
Materials and methods<br />
Study site<br />
The study region comprises just over 1 million hectares and<br />
lies within RDSM and RDSA, mid-River Solimo˜es region,<br />
Amazonas, Brazil (Fig. 1). The region was chosen because<br />
there were previous data on radio-tracked manatees (Marmontel<br />
et al., 2002), and the RDSM infrastructure facilitated<br />
fieldwork and the development of relationships with<br />
local communities. Its 120 km longitudinal extent<br />
(W65103 0 43.60 00 <strong>–</strong>W63151 0 49.80 00 ) encompassed all locations<br />
of the tracked manatees, and its 70 km latitudinal extent<br />
(S02114 0 39.34 00 <strong>–</strong>S03137 0 45.74) is bounded by várzea and by<br />
the Ria Amanã.<br />
In the várzea study region, water levels fluctuate annually<br />
over a range of 16 m, while in Ria Amana˜ annual variation is<br />
about 10 m (Fig. 2a). This variation in water level is caused<br />
by a flood pulse in the River Solimo˜es, and this pulse is the<br />
principal determinant of the extent and depth of flooding<br />
(Fig. 2b) and of the water’s physical chemistry (Junk, Bayley<br />
Figure 1 Study region showing in: small quadrant,<br />
location in relation to South America; large<br />
quadrant, band 4 of Landsat-TM image overlaid<br />
by migration routes of male Amazonian manatees<br />
Trichechus inunguis. Thin white lines are<br />
routes reported by Marmontel et al., (2002),<br />
thicker white lines are routes of manatees<br />
tracked later; thicker lines end at Lake Castanho<br />
so as not to mask previously detected<br />
routes, but all routes lead to or depart from Ria<br />
Amanã. Also shown are location of gauges<br />
(white squares) and of local human communities<br />
(white crosses).<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London
E. M. Arraut et al.<br />
& Sparks, 1989). As a result of this large variation in water<br />
level, and because of the irregular geomorphology of many<br />
lakes and channels, water depth is the principal determinant<br />
of the connectivity between water bodies. The differences of<br />
geomorphology and of water level variation explain, for<br />
example, why although Lake Mamirauá and Ria Amana˜<br />
have similar depths during high-water, the former might<br />
become isolated during low-water while the latter remains<br />
deeper, connected to adjacent water bodies and with more<br />
aquatic space (Arraut, 2008). Based on this habitat seasonality,<br />
we distinguish between periods of lowering-water<br />
(July<strong>–</strong>August), low-water (September<strong>–</strong>November), risingwater<br />
(December<strong>–</strong>April) and high-water (May<strong>–</strong>June), defined<br />
on the basis of the water level and its speed of vertical<br />
change (Arraut, 2008).<br />
Manatee tracking<br />
Seasonal migration of Amazonian manatees<br />
Figure 2 (a) Water-level variation as a result of<br />
River Solimões’s flood pulse measured by<br />
gauges positioned in the channel of Lake<br />
Mamirauá and in Ria Amanã (see Fig. 1). The<br />
gauges are not inter-calibrated, so hydrographs<br />
are only comparable with respect to time and<br />
water-level variation. (b) Habitats vary depending<br />
on flooding duration and flood pulse phase.<br />
When water levels are high, lakes, rivers<br />
and flooded forest are all passable for the<br />
manatee. When levels are low, lakes may loose<br />
connectivity.<br />
From 1994 to 2006 (excepting 2004), 10 males were radiotracked<br />
with an Advanced Telemetry Systems (ATS) transmitter<br />
(Advanced Telemetry Systems Inc., Isanti, MN,<br />
USA) on the 164 MHz band using a three-element Yagi<br />
antenna. They were caught under government permit by<br />
research teams, sometimes aided by local hunters, mostly<br />
using a wide mesh net. Attempts were made to locate each<br />
individual once a day. Although hunting of Brazilian native<br />
fauna is illegal (Castello-Branco & Gomes, 1967), hunters<br />
eventually killed six out of our 10 tracked Amazonian<br />
manatees. We also captured three females, but we lost track<br />
of two on the following day and one was found dead 2 days<br />
later (Table 1). On some occasions, individuals were tracked<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 249
Seasonal migration of Amazonian manatees E. M. Arraut et al.<br />
Table 1 Information about tracked Amazonian manatees<br />
ID Gender<br />
Capture<br />
Location Water season<br />
but no GPS coordinate was associated to the location (Table<br />
1); in these instances, the geographical location is recorded<br />
as the name of the place (lake or part of lake) where the<br />
individuals were active and, for analysis, subsequently given<br />
coordinates at the centre of that location. As a result, for<br />
this subsample of the data, all locations of all manatees that<br />
had been recorded as being from that place were allocated<br />
the same coordinates. Thus, for example, individual seven<br />
was located 168 times, and these sites are allocated, for<br />
analysis, to only 12 GPS positions.<br />
Estimation of manatee locations<br />
We triangulated using the Best bi-angulation algorithm of<br />
the software LOCATION OF A SIGNAL (LOAS) 4.0b (ESS, 2006).<br />
Magnetic bearing corrections were calculated using the<br />
software DECLINAc¸a˜O MAGNÉTICA 2.0 (LEEE, 2003). Field<br />
experiments in which manatee tracking conditions were<br />
simulated at distances up to 2 km gave a mean positional<br />
error of 140 m (Arraut, 2008); we used a tracking resolution<br />
of 300 m, which was much smaller than either the range size<br />
or the dimensions of the habitat fragments.<br />
Migration detection and home-range<br />
calculations<br />
We first separated ranging from migratory movements using<br />
the dispersal detector algorithm in the software RANGES8<br />
(Kenward et al., 2008). We detected that all the eight<br />
manatees that we tracked for more than one flood-pulse<br />
period had migrated. Then, because in the scale of our<br />
analysis the habitat showed annual seasonality, (e.g. macro-<br />
Tracking period<br />
Seasonal<br />
ranges<br />
N_locs used for<br />
range estimate<br />
1 M L. Mamirauá High 4 years A 36 1289<br />
B 18 518<br />
2 M L. Mamirauá High 3 years 4 months A <strong>17</strong>0 794<br />
B 150 1952<br />
3 M L. Mamirauá High 4 years 4 months A 58 727<br />
B 50 1031<br />
4 M R. Amanã Lowering 1 year 4 month A 92 394<br />
B 61 2664<br />
5 M R. Amanã Low 5 months A 9 3474<br />
B 6 234<br />
6 M P. do Castanho Low 3 years A 85 308<br />
7 M L. Mamirauá High 2 years 2 months A 156 250<br />
8 M R. Amanã Lowering 3 months B 55 1186<br />
9 M R. Amanã Lowering 3 months B 43 21<strong>17</strong><br />
10 M R. Amanã Low 6 months B 19 748<br />
11 F P. do Castanho Lowering Not tracked <strong>–</strong> <strong>–</strong> <strong>–</strong><br />
12 F P. do Castanho Lowering Not tracked <strong>–</strong> <strong>–</strong> <strong>–</strong><br />
13 F R. Amanã Low Not tracked <strong>–</strong> <strong>–</strong> <strong>–</strong><br />
phyte cover and flooding were similar in high-water periods<br />
of different years), we pooled locations of each individual in<br />
the same geographic area and flood-pulse period. Home<br />
ranges were estimated using 95% kernels with fixed-smoothing<br />
parameter (Kenward et al., 2008). This model was<br />
especially appropriate because of variability in positional<br />
accuracy and owing to the small number of locations in<br />
some ranges (mean=59, range=6<strong>–</strong><strong>17</strong>0, n=15): kernels<br />
with fixed smoothing can give maximum range area estimates<br />
with as few as 12<strong>–</strong>15 locations (Kenward, 2001) (all<br />
but one animal had at least 12 locations); fixed smoothing<br />
also generates better overall surface estimates than adaptive<br />
smoothing (Seaman & Powell, 1996). Five individuals had<br />
two seasonal home ranges, whereas for the five other<br />
individuals, we could only calculate one home range.<br />
Habitat characterization<br />
95% kernel<br />
range size (ha)<br />
Number of locations used to calculate each of the seasonal home ranges is shown. Letter ‘A’ identifies high-water ranges in várzea lakes and<br />
letter ‘B’ identifies lowering-water ranges in River Japurá and low-water ranges in Ria Amanã (see ‘Habitat Analysis’ section).<br />
L., lake; P., paraná; M, male; F, female; ID, individual identification.<br />
250<br />
To analyse the spatial<strong>–</strong>temporal corollaries of manatee<br />
movement we considered: (1) water-level variation, measured<br />
daily from gauges with 1 cm interval positioned at<br />
Lake Mamirauá (since 1992) and Ria Amana˜ (since 2001);<br />
(2) the bottom topography and depths of water bodies,<br />
based on a bathymetric model from the field survey and<br />
geographical information system modelling; (3) the growth<br />
cycle and spatial distribution of macrophytes combined with<br />
the flooding dynamics in the area, from classifications of 11<br />
Landsat-TM and ETM+ images acquired between 1995<br />
and 2005 (three of the high-water, four of the loweringwater,<br />
two of the low-water and two of the rising-water).<br />
The total of 5 months of fieldwork was distributed throughout<br />
all four flood-pulse periods.<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London
E. M. Arraut et al.<br />
Frequent cloud cover over the Amazon limited the availability<br />
of good-quality satellite images. Previous studies,<br />
however, suggest that on a broad scale, the environment<br />
shows annual seasonality with regard to the flooding extent,<br />
water physical chemistry and macrophyte cover dynamics<br />
(Junk et al., 1989; Junk & Piedade, 1993; Barbosa, 2005;<br />
Arraut, 2008). Thus, when images from the same dates as<br />
the tracking data were not available, we used images from a<br />
different year but same flood pulse period.<br />
In the classifications, we distinguished six habitat<br />
classes: (1) macrophytes on water; (2) flooded forest; (3)<br />
open-water; (4) macrophytes on non-flooded land; (5) nonflooded<br />
forest; (6) non-flooded land. The first three occurred<br />
during high-water and the last three, plus open-water,<br />
during the low-water. Classification assessment was<br />
based on a confusion matrix (Congalton & Green, 2008),<br />
which indicated accuracies above 90%. Further details of<br />
the habitat characterization process are given in Arraut<br />
(2008).<br />
Habitat analysis<br />
Habitat analysis addressed three questions:<br />
Question 1: Selective use of habitat<br />
Are males selective in their use of habitats, and specifically<br />
of aquatic macrophytes?<br />
Question 2: Proportion of forage within home ranges<br />
during high-water<br />
During the high-water, was there proportionally more<br />
forage in manatee home ranges within várzea lakes than if<br />
they had remained in River Japur a´ or Ria Amana˜?<br />
Question 3: Seasonal reduction in flooded area within<br />
home ranges<br />
Did the proportional seasonal reduction in the aquatic<br />
area within home ranges differ between manatees in várzea<br />
lakes and those in River Japurá or Ria Amana˜?<br />
Selective use of habitat (question 1) was assessed using<br />
compositional analysis (CA) on log ratios of used and<br />
available habitats (Aebischer, Robertson & Kenward,<br />
1993). To answer questions 2 and 3, we separated the 15<br />
home ranges occupied by the 10 manatees into two<br />
categories: ranges in várzea lakes during the high-water<br />
(category A) and ranges in River Japurá during the lowering-water<br />
or in Ria Amana˜ during the low-water<br />
(category B). Ranges in River Japur a´ and Ria Amana˜ were<br />
grouped into one category because the objective was to<br />
discover, with quantitative evidence, why male manatees<br />
remained in várzea lakes during high-water, and why they<br />
did not remain there during low-water (i.e. why they<br />
migrated); this grouping also increased sample size for<br />
statistical analyses.<br />
We carried out analyses of variance (ANOVA) by means<br />
of general linear models (GLM) using MINITAB v15.1.1<br />
(Minitab, 2007). For question 2, the response variable in<br />
the GLM was the ‘proportion of the home range covered by<br />
aquatic macrophytes’ (arcsine root transformed), and for<br />
question 3, it was the ‘proportional seasonal reduction in the<br />
flooded area’ (arcsine root transformed). Although there<br />
were too few ranges to test for inclusion of multiple<br />
predictor variables, we included individual identification<br />
(ID) and location number in each home range (N_locs) in<br />
the GLMs with category variable A/B to exclude the<br />
possibility that relationships with the response variable<br />
depended on differences between individuals and sampling<br />
characteristics.<br />
Results<br />
Seasonal habitat use and migration<br />
Males remained in várzea lakes from the middle of the rising<br />
to the beginning of the lowering-water period (approximately<br />
from February to the beginning of August), when<br />
flooding facilitated unconstrained access to all lakes, rivers<br />
and flooded forest. During the lowering-water, they left<br />
várzea lakes and migrated to Ria Amanã, where they stayed<br />
during the low-water. Then, during the rising-water, males<br />
migrated back to várzea lakes, thus closing the annual<br />
migratory cycle. The three females captured were either at<br />
(during low-water) or migrating to (during lowering-water)<br />
Ria Amana˜, thus, moving in accordance with what was<br />
observed for the radio-tracked males.<br />
Habitat analysis<br />
Selective use of habitat<br />
Seasonal migration of Amazonian manatees<br />
CA revealed aquatic macrophytes as the only preferred<br />
habitat class during high-water, and ranked the three<br />
habitat classes: (2) macrophytes, (1) open-water and (0)<br />
flooded forest [F (2, 5 d.f.)=7.56, P=0.04].<br />
Proportion of forage within home ranges during<br />
high-water<br />
During high-water, home ranges in várzea lakes (category<br />
A) encompassed almost sevenfold the area of aquatic<br />
macrophytes that occurred within home ranges in River<br />
Japurá or Ria Amana˜ (A: mean=0.27, SE=0.13; B:<br />
Figure 3 Macrophyte proportion in the home ranges of individuals<br />
was greater when in várzea lakes (^) than when in River JapuráorRia<br />
Amanã (’). ID, individual identification.<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 251
Seasonal migration of Amazonian manatees E. M. Arraut et al.<br />
mean=0.04, SE=0.04) (Fig. 3). In the GLM, the category<br />
A/B variable explained the most variance in the response<br />
variable, and continued to do so with the inclusion of ID<br />
and N_locs (Table 2). Model residuals showed no trend,<br />
indicating that the model was appropriate. Thus, the proportion<br />
of the home range classified as macrophyte habitat<br />
was greatest for males while they were in várzea lakes.<br />
Seasonal reduction in flooded area within home<br />
ranges<br />
Reduction in the flooded area in várzea lakes was over 4.5<br />
times greater than in River Japur a´ or Ria Amana˜ (A:<br />
mean=0.98, SE=0.01; B: mean=0.21, SE=0.07) (Fig. 4).<br />
In the GLM, variable A/B explained most of the variance in<br />
the response variable, and continued to do so with inclusion of<br />
ID and N_locs (Table 3). Analysis of full model residuals<br />
showed no particular trend, indicating the model was appropriate.<br />
Areas within high-water ranges practically dried out<br />
during low-water, while areas where manatees had low-water<br />
ranges suffered much smaller reduction in the aquatic space.<br />
After leaving Mamirauá while the water was lowering in<br />
1996, individuals one, two and three remained for a few days<br />
in an area of River Japur a. ´ They then disappeared during<br />
the low-water, only to return in the next rising-water. On the<br />
other hand, individuals captured at Paraná do Castanho<br />
during the lowering-water were moving in the direction of<br />
Table 2 General linear model results: F statistic and P values for<br />
response variable ‘macrophyte proportion in the home ranges’ for full<br />
and reduced models<br />
Explanatory variables A/B N_locs ID<br />
Variable alone 38.48 a<br />
0.51 d<br />
0.73 d<br />
A/B with each <strong>–</strong> 31.20 b<br />
67.50 b<br />
A/B with both <strong>–</strong> 121.36 c<br />
a Po0.001.<br />
b P=0.001.<br />
c P=0.002.<br />
d P not significant.<br />
ID, individual identification.<br />
Figure 4 Proportion reduction in flooded area (from high- to lowwater)<br />
was much greater for ranges in várzea lakes (^) than for those<br />
in River Japurá or Ria Amanã (’). ID, individual identification.<br />
252<br />
Table 3 General linear model results: F statistic and P values of the A/<br />
B variable ‘proportion reduction in flooded area’ for the full and<br />
reduced models<br />
Explanatory variables A/B N_locs ID<br />
Variable alone 227.37 a<br />
0.93 c<br />
0.40 c<br />
A/B with each <strong>–</strong> 238.94 a<br />
165.14 a<br />
A/B with both <strong>–</strong> 138.70 b<br />
a Po0.001.<br />
b P=0.001.<br />
c P not significant.<br />
ID, individual identification.<br />
Ria Amana˜, and all those captured and/or tracked during<br />
low-water were within Ria Amana˜ (n=9). This suggests<br />
that manatees aggregate in Ria Amana˜ during low-water.<br />
Discussion<br />
Várzea species are adapted to the seasonality of the environment<br />
(Junk et al., 1989). In the RDSM region, river dolphins<br />
Inia geoffrensis spend the high-water in várzea lakes and in<br />
flooded forest, but during low-water, they are in the main<br />
channels of Rivers Solimo˜es and Japur a ´ (Martin & da Silva,<br />
2004). Male Amazonian manatees also lived in várzea lakes<br />
during high-water, but in the low-water migrated to Ria<br />
Amana˜.<br />
The várzea lakes used during high-water offered forage in<br />
greater abundance [Results (1) and (2); Fig. 3, Table 2] and<br />
diversity (Junk et al., 1989; Arraut, 2008; Guterres &<br />
Marmontel, 2008), and less water current than rivers and<br />
paran as ´ (channels that connect rivers). When the water level<br />
dropped and the flooded area significantly contracted,<br />
restricting their home ranges [Result (3), Fig. 4, Table 3],<br />
male manatees migrated to Ria Amana˜ where they spent the<br />
low-water season. There, apart from more space, they<br />
encounter reduced risk from predators like: caimans Melanosuchus<br />
niger and Caiman crocodilus (Nunes Pereira, 1947),<br />
jaguars Panthera onca (Bertram & Bertram, 1973) and<br />
humans Homo sapiens sapiens (Domning, 1982a; Marmontel,<br />
2008). During the low-water period, caimans aggregate<br />
in whichever water bodies remain in the várzea (E. M.<br />
Arraut & M. Marmontel, pers. obs.), jaguars frequent the<br />
water margins (Ramalho, 2006) and humans come to gather<br />
the fish that aggregate there. Conversely, local hunters assert<br />
that the manatees in Ria Amana˜ are more difficult to kill<br />
(Calvimontes, 2009).<br />
Further information from interviews with local hunters<br />
indicates that female manatees have seasonal movement<br />
patterns similar to those of males. Sixty-four manatees were<br />
killed in Lake Castanho, Ria Amana˜ and the migration<br />
route between them during 2002<strong>–</strong>2004 (Calvimontes, 2009).<br />
These manatees were killed in Lake Castanho only between<br />
the rising- and the mid-lowering-water seasons and in Ria<br />
Amana˜, only between the lowering- and the beginning of the<br />
rising-water seasons. Of the 35 that had their sex determined,<br />
there were 13 males and 22 females. All four<br />
manatees killed in upper-Amanã at the beginning of the<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London
E. M. Arraut et al.<br />
rising-water were females, including two that were calves<br />
(Calvimontes, 2009). From estimated birth dates of 24<br />
neonate Amazonian manatees, Best (1982) suggested that<br />
calves are normally born during the period of rising-water.<br />
As gestation is 12<strong>–</strong>14 months (Rosas, 1994), mating is<br />
expected to occur during low-water or the start of risingwater.<br />
Therefore, Ria Amana˜ may be both a mating and<br />
calving ground for Amazonian manatees.<br />
Do manatees that live in the várzeas of other<br />
white-water rivers move similarly as those in<br />
the mid-Solimões region?<br />
In the Amazon, immense white-water rivers such as the<br />
Solimo˜es, Amazonas, Puru´s and Madeira form várzeas and<br />
have rias annexed to their várzeas. They are also typified by<br />
flood pulses, and seasonal macrophytes (Junk & Furch,<br />
1993) (Fig. 5). The habitats available to manatees in these<br />
other regions of the Amazon, thus, seem similar to that<br />
which we have described in our study region.<br />
Evidence for the occurrence of manatees in these areas<br />
Acuna ˜ (1641), who reported them as abundant enough to be<br />
easily seen throughout the Rivers Solimo˜es and Amazonas,<br />
and Domning (1982a), who analysed the records of manatees<br />
hunted (legally at the time) that were documented in<br />
Manaus between <strong>17</strong>85 and 1983. Based on such information,<br />
as well as on how the large-scale movements of<br />
manatees in our study region coincide closely with the flood<br />
pulse, we conjecture that the habitats that can sustain the<br />
largest populations of T. inunguis during the high-water are<br />
the várzeas of large white-water rivers.<br />
On the other hand, during low-water, when most of the<br />
várzea dries out, rias remain the main places where aquatic<br />
space (Hess et al., 2003) with minimal current can be found.<br />
We thus conjecture that rias are the main low-water refuges<br />
of manatees that live in the floodplains of these major whitewater<br />
rivers; these floodplains comprise the great majority<br />
of the manatees’ known geographical range.<br />
Evolutionary ecology of the Amazonian<br />
manatees’ migration<br />
Seasonal migrations have been well documented in other<br />
aquatic mammals, and the ecological processes that drive<br />
River Japurá<br />
Ria Tefé<br />
Ria Amanã<br />
River Solimões<br />
River Negro<br />
River Amazonas<br />
migration often seem to be related to habitat variability. The<br />
Florida manatee Trichechus manatus latirostris (Deutsch<br />
et al., 2003) and the dugong Dugong dugon (Sheppard et al.,<br />
2006) migrate to warmer waters in autumn, and back to<br />
their summer areas in spring. In these two species, the<br />
evolution of migration can be explained as a means of<br />
optimizing forage when it is available and avoiding harsh<br />
environmental conditions during the rest of the year. We<br />
have demonstrated comparable seasonal movements for<br />
Amazonian manatees, although in their case the ‘seasons’<br />
are determined by variation in water level and the limiting<br />
factors are space and predation.<br />
Trichechus inunguis is thought to be the most derived<br />
species of manatee, having evolved from ancestral trichechids<br />
that colonized the Western Amazon in the Pliocene<br />
(2<strong>–</strong>6 million years ago) (Domning, 1982b). Geological evidence<br />
suggests that the várzea of River Solimo˜es achieved its<br />
present form in the Holocene (o10 000 years), and that<br />
terraces on its sides were formed a bit earlier, in the last<br />
47 300 years (Rossetti, Toledo & G oes, ´ 2005). Because rias<br />
are formed by the excavation of such terraces by ancient<br />
rivers, they must be younger than the terraces. Thus, the use<br />
of Ria Amana˜ as a breeding and calving ground, and<br />
migration to it, must be a relatively recent phenomenon in<br />
the species’ evolutionary history and, if our conjecture is<br />
correct, this also applies to the use of other rias. The<br />
geological history of the Western Amazon thus indicates<br />
that Amazonian manatees have changed their spatial behaviour<br />
in response to changes in the habitat that occurred<br />
within the last few tens of thousands of years.<br />
The fact that manatees living from the mouth of the River<br />
Amazon and those occurring throughout Amazonia to the<br />
west are part of a single panmictic population (Cantanhede<br />
et al., 2005) indicates that any differences in the movements<br />
of individuals in these widely separated areas are probably a<br />
result of behavioural plasticity in response to local habitat<br />
conditions. This raises the question: what are the movement<br />
patterns of the manatees living near the mouth of River<br />
Amazon (Domning, 1981), where water level varies daily<br />
(ANA, 2009) because of the Atlantic tides and where there<br />
are no rias? How do these differences in the spatial<strong>–</strong>temporal<br />
dynamics of their habitats affect manatee reproductive<br />
ecology? For example, if manatees find forage and living<br />
space in the same area throughout the year, we would not<br />
River Tapajós<br />
N<br />
1320 km<br />
Seasonal migration of Amazonian manatees<br />
Figure 5 Rias (shown by arrows), which are<br />
numerous and of various sizes, occur throughout<br />
the margins of Rivers Solimões and Amazonas<br />
and are conjectured to be the Amazonian<br />
manatee’s main low-water refuges. Study-region<br />
limits indicated by dashed-line polygon.<br />
Landsat-TM mosaic of the central Amazon<br />
várzea. Source: Shimabukuro, Novo & Mertes<br />
(2002).<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 253
Seasonal migration of Amazonian manatees E. M. Arraut et al.<br />
expect them to migrate. Answers to these questions await<br />
further research.<br />
Conservation implications<br />
Migrating Amazonian manatees pass through narrow channels<br />
where hunters can wait within harpoon range. Because<br />
human settlements are numerous and widespread (Fig. 1),<br />
and local inhabitants relish manatee meat, and are aware of<br />
the timing of their migration, manatees have to pass through<br />
perilous bottlenecks to arrive at the relative safety of Ria<br />
Amana˜. Then, in the beginning of the rising-water, water<br />
levels rise sufficiently to flood the few beaches where macrophytes<br />
have been growing close to the mouth of Ria Amana˜.<br />
The manatees aggregate at these beaches, hungry after<br />
several months of fasting and again become vulnerable to<br />
hunters. Hence, the migration to, and prolonged occupation<br />
of, Ria Amana˜ present clear risks to contemporary manatees.<br />
Nevertheless, the fact that manatees migrate to Ria<br />
Amana˜, and that, as argued above, this migratory behaviour<br />
is probably not genetically determined, suggests that staying<br />
in várzea lakes during the low-water has been (at least until<br />
recently) even more perilous. The lowering- and low-water<br />
seasons are, thus, periods when Amazonian manatees living<br />
in the mid-Solimo˜es region have to choose between the<br />
lesser of two evils.<br />
Any climate changes that lead to more intense droughts,<br />
and thus to a reduction in the aquatic space during the lowwater<br />
period, would increase the exposure of Amazonian<br />
manatees to hunters. A connection between climate change<br />
and an increase in the frequency of droughts in the Amazon<br />
is predicted under all scenarios of all 23 IPCC climatic<br />
models (Malhi et al., 2008), and relationships between<br />
large-scale climatic phenomena, such as the El Nino ˜ Southern<br />
Oscillation and the Tropical North Atlantic Ocean<br />
Temperatures, and the levels of major Amazonian rivers,<br />
were determined by Richey, Nobre & Deser (1989), Schongart<br />
& Junk (2007) and Marengo et al. (2008a,b). A situation<br />
that might become more frequent is exemplified by the<br />
drought of 2005, estimated to be the most intense in the last<br />
40 years, when the Amazon River and its major tributaries<br />
were left with only a fraction of their normal volume<br />
(Marengo et al., 2008a,b). During this drought, enormous<br />
quantities of fish died, clogging rivers and poisoning the<br />
water, and people living in small communities had to walk<br />
several kilometres to find water and food (Giles, 2006).<br />
Amazonian manatees also suffered. Local people informed<br />
us of the killings of at least five manatees in Ria Tefe´, and<br />
colleagues working in River Puru´s informed us of at least 12<br />
that were killed in Ria Jari (B. Marioni, E. V. Mullen & F.<br />
Rossoni, pers. comm.).<br />
In order to protect the manatees better now, and to<br />
safeguard them in the face of climate change, it is important<br />
to research their status in other regions, and to complement<br />
such research with education. Our experience indicates that to<br />
be effective, any conservation programme needs to learn from,<br />
educate and bring benefits to local populations, because local<br />
people are necessary actors in conserving the manatees and<br />
254<br />
their environment. Our conjecture that rias are the manatee’s<br />
main low-water refuge can serve as a starting point in the<br />
search for manatees in other regions of the Amazon.<br />
Acknowledgements<br />
We are grateful to the Coordenac¸ão de Aperfeiçoamento de<br />
Pessoal de Nível Superior (CAPES) for funding Arraut’s<br />
PhD stipend and expenses, to the Rede de Modelagem<br />
Ambiental da Amazoˆnia (GEOMA) Network and to the<br />
Graduate program in Remote Sensing at INPE (Brazil). We<br />
are thankful to Antonio Pinto de Oliveira for his help during<br />
fieldwork, and to all the local hunters and trainees who<br />
helped with data collection. We thank Dr Paul Johnson for<br />
his suggestions during data analyses, and Drs Marion<br />
Valeix, Jorgelina Marino, Thomas Merckx, Laura Fasola<br />
and Ruth Kanski for fruitful comments on an earlier draft.<br />
We also thank Dr Dilce Rossetti, Dr Thiago Silva and two<br />
anonymous referees for their suggestions.<br />
References<br />
Acuna, ˜ C. (1641). Novo descobrimento do Grande Rio Amazonas/Critóbal<br />
de Acuña, 1641. Rio de Janeiro: Agir.<br />
Aebischer, N.J., Robertson, P.A. & Kenward, R.E. (1993).<br />
Compositional analysis of habitat use from animal radiotracking<br />
data. Ecology 74, 1313<strong>–</strong>1325.<br />
ANA. (2009). Hidroweb <strong>–</strong> Hydrological Information System,<br />
Brazilian National Agency of the Waters. Available at<br />
http://hidroweb.ana.gov.br/ (accessed 10 February 2009).<br />
Arraut, E. (2008). Migraça˜o do Peixe-boi Amazoˆnico: uma<br />
abordagem por sensoriamento remoto, radiotelemetria e<br />
geoprocessamento. PhD thesis, Instituto Nacional de Pesquisas<br />
Espaciais (National Institute for Space Research),<br />
Sa˜o Jose´ dos Campos.<br />
Barbosa, C.C.F. (2005). Sensoriamento Remoto da dinaˆmica<br />
da circulaça˜o daágua do sistema planície de Curuai/Rio<br />
Amazonas. Sa˜oJose´ dos Campos: Instituto Nacional de<br />
Pesquisas Espaciais.<br />
Bertram, G.C.L. & Bertram, C.K.R. (1973). The modern<br />
Sirenia: their distribution and status. Biol. J. Linn. Soc. 5,<br />
297<strong>–</strong>338.<br />
Best, R.C. (1982). Seasonal breeding in the Amazonian<br />
manatee, Trichechus inunguis (Mammalia, Sirenia). Biotropica<br />
14, 76<strong>–</strong>78.<br />
Best, R.C. (1983). Apparent dry-season fasting in Amazonian<br />
Manatees (Mammalia: Sirenia). Biotropica 15, 61<strong>–</strong>64.<br />
Best, R.C. (1984). The aquatic mammals and reptiles of the<br />
Amazon. In The Amazon limnology and landscape ecology<br />
of a mighty tropical river and its basins: 371<strong>–</strong>412. Sioly, H.<br />
(Ed.). Dordrecht: Dr Junk W. Publishers.<br />
Caldwell, D.K. & Caldwell, M.C. (1985). Manatees Trichechus<br />
manatus (Linnaeus, <strong>17</strong>58); Trichechus senegalensis<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London
E. M. Arraut et al.<br />
(Link, <strong>17</strong>95) and Trichechus inunguis (Natterer, 1883). In<br />
Handbook of marine mammals: 33<strong>–</strong>66. Ridgway, S.H. &<br />
Harisson, R. (Eds). London: Academic Press.<br />
Calvimontes, J. (2009). Etnoconocimiento, uso y conservación<br />
del manatí amazónico Trichechus inunguis en la Reserva de<br />
Desarrollo Sostenible Amana˜, Brasil. Lima: Universidad<br />
Nacional Agraria La Molina.<br />
Cantanhede, A., Da Silva, V., Farias, I., Hrbek, T., Lazzarini,<br />
S. & Alves-Gomes, J. (2005). Phylogeography and population<br />
genetics of the endangered Amazonian manatee,<br />
Trichechus inunguis Natterer, 1883 (Mammalia, Sirenia).<br />
Mol. Ecol. 14, 401<strong>–</strong>413.<br />
Castello-Branco, H. & Gomes, S.F. (1967). Lei de proteça˜oda<br />
fauna silvestre (Law for the protection of native fauna).<br />
Brasília: C. C. Presideˆncia da Repu´blica, Subchefia de<br />
Assuntos Jur ıdicos. ´<br />
Congalton, R.G. & Green, K. (2008). Assessing the<br />
accuracy of remotely sensed data: principles and practices,<br />
2nd edn. Boca Raton: CRC Press, Taylor & Francis<br />
Group.<br />
Deutsch, C., Reid, J., Bonde, R., Easton, D., Kochman, H. &<br />
O’Shea, T. (2003). Seasonal movements, migratory behavior,<br />
and site fidelity of West Indian manatees along the<br />
Atlantic coast of the United States. Wildl. Monogr. 151,<br />
1<strong>–</strong>77.<br />
Dingle, H. & Drake, V. (2007). What is migration? Bioscience<br />
57, 113<strong>–</strong>121.<br />
Domning, D. (1981). Distribution and status of manatees<br />
Trichechus spp. near the mouth of the Amazon River,<br />
Brazil. Biol. Conserv. 19, 85<strong>–</strong>97.<br />
Domning, D. (1982a). Commercial exploitation of manatees<br />
Trichechus in Brazil c. <strong>17</strong>85<strong>–</strong>1973. Biol. Conserv. 22,<br />
101<strong>–</strong>126.<br />
Domning, D. (1982b). Evolution of manatees: a speculative<br />
history. J. Paleontol. 56, 599<strong>–</strong>619.<br />
ESS. (2006). Location of a Signal (LOAS). Heygmagas:<br />
Ecological Software Solutions LLC.<br />
Giles, J. (2006). The outlook for Amazonia is dry. In Nature<br />
news: 726<strong>–</strong>727. Campbell, P. & Lincoln, T. (Eds). London:<br />
Nature Publishing Group.<br />
Guterres, M. & Marmontel, M. (2008). Anatomia e Morfologia<br />
de plantas aquáticas da Amazoˆnia utlizadas como potencial<br />
alimento pelo peixe-boi amazoˆnico. Bele´m: Instituto de<br />
Desenvolvimento Sustentável Mamirau a. ´<br />
Hess, L.L., Melack, J.M., Novo, E., Barbosa, C.C.F. &<br />
Gastil, M. (2003). Dual-season mapping of wetland inundation<br />
and vegetation for the central Amazon basin.<br />
Remote Sens. Environ. 87, 404<strong>–</strong>428.<br />
Junk, W.J., Bayley, P.B. & Sparks, R.E. (1989). The flood<br />
pulse concept in river<strong>–</strong>floodplain systems. In Proceedings of<br />
the <strong>International</strong> Large River Symposium (LARS):<br />
110<strong>–</strong>127. Dodge, D.P. (Ed.). Toronto: Canadian Special<br />
Publication of Fisheries and Aquatic Sciences.<br />
Junk, W.J. & Furch, K. (1993). A general review of tropical<br />
South American floodplains. Wetl. Ecol. Mgmt. 2,<br />
231<strong>–</strong>238.<br />
Seasonal migration of Amazonian manatees<br />
Junk, W.J. & Piedade, M.T.F. (1993). Biomass and<br />
primary-production of herbaceous plant communities<br />
in the Amazon floodplain. Hydrobiologia 263,<br />
155<strong>–</strong>162.<br />
Kenward, R.E. (2001). A manual for wildlife radio tagging.<br />
London: Academic Press.<br />
Kenward, R.E., Walls, S.S., South, A.B. & Casey, N. (2008).<br />
Ranges8 for the analysis of tracking and location data.<br />
Wareham: Anatrack Ltd.<br />
LEEE. (2003). Declinaça˜o Magne´tica 2.0. Santa Catarina:<br />
Laborat orio ´ de Eficieˆncia Energética em Edificac¸o˜es.<br />
Malhi, Y., Roberts, J.T., Betts, R.A., Killeen, T.J., Li, W. &<br />
Nobre, C.A. (2008). Climate change, deforestation, and the<br />
fate of the Amazon. Science 319, 169<strong>–</strong><strong>17</strong>2.<br />
Marengo, J.A., Nobre, C.A., Tomasella, J., Cardoso, M.F. &<br />
Oyama, M.D. (2008a). Hydro-climatic and ecological<br />
behaviour of the drought of Amazonia in 2005. Philos.<br />
Trans. Roy. Soc. Lond. Ser. B Biol. Sci. 363, <strong>17</strong>73<strong>–</strong><strong>17</strong>78.<br />
Marengo, J.A., Nobre, C.A., Tomasella, J., Oyama, M.D.,<br />
Oliveira, G.S., Oliveira, R., Alves, L.M. & Brown, I.F.<br />
(2008b). The drought of Amazonia in 2005. J. Clim. 21,<br />
495<strong>–</strong>516.<br />
Marmontel, M. (2008). Trichechus inunguis. Available at<br />
http://www.iucnredlist.org (accessed 19 February 2009).<br />
Marmontel, M., Guterres, M., Meirelles, A.C., Calvimontes,<br />
J. & Rosas, F.C.W. (2002). Lago Amana˜: destino estival de<br />
manaties amazoˆnicos en la Amazonia occidental<br />
brasilena.: ˜ 40. Reunion de Trabajo de Especialistas en<br />
Mam ıferos ´ Acuáticos de Ame´rica del Sur (RT), 10<br />
Congresso Sociedad Latinoamericana de Especialistas<br />
en Mam ıferos ´ Acuáticos (SOLAMAC), Vald ıvia, ´<br />
Chile.<br />
Martin, A.R. & da Silva, V.M.F. (2004). River dolphins and<br />
flooded forest: seasonal habitat use and sexual segregation<br />
of botos (Inia geoffrensis) in an extreme cetacean environment.<br />
J Zool. (Lond.) 263, 295<strong>–</strong>305.<br />
Minitab. (2007). Minitab. State College: Minitab Inc.<br />
Montgomery, G.G., Best, R.C. & Yamakoshi, M. (1981). A<br />
radio-tracking study of the Amazonian manatee<br />
Trichechus inuguis (Mammalia: Sirenia). Biotropica 13,<br />
81<strong>–</strong>85.<br />
Nunes Pereira, M. (1947). O peixe-boi da Amazoˆnia. Manaus:<br />
Imprensa Official.<br />
Ramalho, E.E. (2006). Uso do hábitat e dieta da Onça-pintada<br />
(Panthera onca) em uma área de várzea, Reserva de<br />
Desenvolvimento Sustentável Mamirauá, Amazoˆnia Central,<br />
Brasil. Masters thesis, Instituto Nacional de Pesquisas da<br />
Amazoˆnia (National Institute for Research in Amazonia),<br />
Manaus.<br />
Reeves, R.R., Leatherwood, S., Jefferson, T.A. & Curry, B.E.<br />
(1996). Amazonian manatees, Trichechus inunguis, in Peru:<br />
distribution, exploitation, and conservation status. Intercieˆncia<br />
21, 246<strong>–</strong>254.<br />
Richey, J.E., Nobre, C. & Deser, C. (1989). Amazon river<br />
discharge and climate variability: 1903 to 1985. Science<br />
246, 101<strong>–</strong>103.<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 255
Seasonal migration of Amazonian manatees E. M. Arraut et al.<br />
Rosas, F.C.W. (1994). Biology, conservation and status of the<br />
Amazonian manatee Trichechus inunguis. Mammal. Rev.<br />
24, 49<strong>–</strong>59.<br />
Rossetti, D.F., Toledo, P.M. & Góes, A.M. (2005). New<br />
geological framework for Western Amazonia (Brazil) and<br />
implications for biogeography and evolution. Quat. Res.<br />
63, 78<strong>–</strong>89.<br />
Schongart, J. & Junk, W.J. (2007). Forecasting the flood-pulse in<br />
Central Amazonia by ENSO-indices. J. Hydrol. 335, 124<strong>–</strong>132.<br />
Seaman, D.E. & Powell, R.A. (1996). An evaluation of the<br />
accuracy of kernel density estimators for home range<br />
analysis. Ecology. 77, 2075<strong>–</strong>2085.<br />
256<br />
Sheppard, J., Preen, A., Marsh, H., Lawler, I., Whiting, S. &<br />
Jones, R. (2006). Movement heterogeneity of dugongs,<br />
Dugong dugon (Müller), over large spatial scales. J. Exp.<br />
Mar. Biol. Ecol. 334, 64<strong>–</strong>83.<br />
Shimabukuro, Y., Novo, E.M.L.M. & Mertes, L.K. (2002).<br />
Amazonas river mainstream floodplain Landsat TM digital<br />
mosaic. Int. J. Remote Sens. 23, 57<strong>–</strong>69.<br />
Timm, R., Albuja, V. & Clauson, B. (1986). Ecology, distribution,<br />
harvest, and conservation of the Amazonian<br />
manatee Trichechus inunguis in Ecuador. Biotropica 18,<br />
150<strong>–</strong>156.<br />
Journal of Zoology 280 (2010) 247<strong>–</strong>256 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London
The Journal of Experimental Biology 212, 2349-2355<br />
Published by The Company of Biologists 2009<br />
doi:10.1242/jeb.027565<br />
Carbon and nitrogen stable isotope turnover rates and diet<strong>–</strong>tissue discrimination in<br />
Florida manatees (Trichechus manatus latirostris)<br />
Christy D. Alves-Stanley1 and Graham A. J. Worthy1,2, *<br />
1Physiological Ecology and Bioenergetics Lab, Department of Biology, University of Central Florida, 4000 Central Florida Boulevard,<br />
Orlando, FL 32816, USA and 2Hubbs-SeaWorld Research Institute, 6295 Sea Harbor Drive, Orlando, FL 32821, USA<br />
INTRODUCTION<br />
The isotopic composition of consumer tissues reflects that of local<br />
food webs and can be used to predict diet composition, the trophic<br />
level at which the consumer is feeding, and even habitat use and<br />
migratory patterns (e.g. Deniro and Epstein, 1978; Deniro and<br />
Epstein, 1981; Fry, 1981; Peterson and Fry, 1987) (reviewed by<br />
Hobson, 1999). Two stable isotope ratios commonly analyzed in<br />
feeding ecology studies are those of carbon ( 13 C/ 12 C) and nitrogen<br />
( 15 N/ 14 N). Carbon isotope ratios indicate the likely source of<br />
primary production and have been used to differentiate between C3<br />
and C4 plants, terrestrial and marine ecosystems, deep forest and<br />
open habitat consumers, and benthic and pelagic aquatic systems<br />
(e.g. Cloern et al., 2002; Cerling et al., 2004; Hall-Aspland et al.,<br />
2005). Nitrogen isotope ratios exhibit a predictable, step-wise<br />
enrichment between trophic levels and also have been shown to<br />
differ between terrestrial and marine ecosystems (e.g. Hobson and<br />
Welch, 1992).<br />
Stable isotope analysis is especially advantageous when<br />
investigating the feeding ecology and habitat use of marine mammals<br />
where it is often impossible to directly observe feeding or migratory<br />
behavior. Tissue samples, such as skin or blubber, may be analyzed<br />
for stable isotope ratios without sacrificing the animal and this<br />
approach has been successfully applied to a variety of marine<br />
mammal species including mysticetes (e.g. Lee et al., 2005),<br />
odontocetes (e.g. Walker et al., 1999), pinnipeds (e.g. Hobson and<br />
Welch, 1992; Kurle and Worthy, 2002; Newsome et al., 2006),<br />
sirenians (e.g. Ames et al., 1996; MacFadden et al., 2004; Yamamuro<br />
et al., 2004; Reich and Worthy, 2006), sea otters (Clementz and<br />
Koch, 2001) and polar bears (e.g. Ramsay and Hobson, 1991).<br />
In order to accurately interpret isotopic results, it is imperative<br />
to determine both isotopic discrimination (the difference in isotopic<br />
*Author for correspondence (e-mail: gworthy@mail.ucf.edu)<br />
Accepted 3 May 2009<br />
SUMMARY<br />
The Florida manatee (Trichechus manatus latirostris) is a herbivorous marine mammal that occupies freshwater, estuarine and<br />
marine habitats. Despite being considered endangered, relatively little is known about its feeding ecology. The present study<br />
expands on previous work on manatee feeding ecology by providing critical baseline parameters for accurate isotopic data<br />
interpretation. Stable carbon and nitrogen isotope ratios were examined over a period of more than 1 year in the epidermis of<br />
rescued Florida manatees that were transitioning from a diet of aquatic forage to terrestrial forage (lettuce). The mean half-life for<br />
13 C turnover was 53 and 59 days for skin from manatees rescued from coastal and riverine regions, respectively. The mean halflife<br />
for 15 N turnover was 27 and 58 days, respectively. Because of these slow turnover rates, carbon and nitrogen stable isotope<br />
analysis in manatee epidermis is useful in summarizing average dietary intake over a long period of time rather than assessing<br />
recent diet. In addition to turnover rate, a diet<strong>–</strong>tissue discrimination value of 2.8‰ for 13 C was calculated for long-term captive<br />
manatees on a lettuce diet. Determining both turnover rate and diet<strong>–</strong>tissue discrimination is essential in order to accurately<br />
interpret stable isotope data.<br />
Key words: turnover, stable isotope, Florida manatee, diet<strong>–</strong>tissue discrimination, 13 C, 15 N, Trichechus manatus, feeding ecology.<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
2349<br />
ratios between consumer tissue and diet) and turnover rate (the time<br />
it takes for the isotope to be assimilated into the consumer’s tissue)<br />
of the sampled tissue. Diet<strong>–</strong>tissue discrimination may be difficult<br />
to determine for animals feeding on multiple, isotopically distinct<br />
prey items for which the proportions contributing to the diet are<br />
unknown. However, controlled captive studies on a variety of taxa<br />
have allowed for more precise measurements (e.g. Roth and Hobson,<br />
2000; Cherel et al., 2005; Logan et al., 2006; Seminoff et al., 2006).<br />
In addition to diet<strong>–</strong>tissue discrimination, turnover rates in tissues<br />
must be determined in order to assess whether the isotope signature<br />
of the tissue represents the most recent diet or the long-term diet.<br />
An effective method to determine turnover rate is to experimentally<br />
switch an animal from one known diet to another isotopically distinct<br />
diet. Turnover rates of stable isotopes have been calculated using<br />
this method for mammals (e.g. Tieszen et al., 1983), birds (e.g.<br />
Hobson and Clark, 1992a), fish (e.g. Bosley et al., 2002) and<br />
invertebrates (e.g. Olive et al., 2003). These studies have shown<br />
that tissues with higher metabolic activity (e.g. blood, liver) have<br />
faster turnover rates than less active tissue (e.g. bone).<br />
The endangered Florida manatee (Trichechus manatus latirostris<br />
L.) is known to feed on a variety of aquatic plants in fresh, estuarine<br />
and marine habitats (e.g. Campbell and Irvine, 1977; Hartman, 1979;<br />
Best, 1981), each of which has a distinct isotopic signature (e.g.<br />
Fry and Sherr, 1984; Reich and Worthy, 2006) (C.D.A.-S. and<br />
G.A.J.W., in preparation). Little is known about fine-scale manatee<br />
feeding ecology and habitat use because manatees often occupy<br />
shallow, turbid water. In addition, manatee population counts and<br />
trends remain unclear (Lefebvre et al., 1995; US Fish and Wildlife<br />
Service, 2001). Consequently, it has become increasingly important<br />
to understand manatee feeding ecology and its relation to habitat<br />
use in order to improve conservation efforts.
2350<br />
C. D. Alves-Stanley and G. A. J. Worthy<br />
The present study used skin samples from Florida manatees<br />
transitioning between two isotopically distinct diets (aquatic to<br />
terrestrial) to determine turnover rates and diet<strong>–</strong>tissue discrimination<br />
in epidermis tissue. Manatees were part of the rehabilitation program<br />
at SeaWorld of Florida, and were in need of captive care for reasons<br />
including physical trauma, nutritional stress and/or cold stress. The<br />
overall objectives of the present study were to determine 13 C and<br />
15 N turnover rates in epidermis tissue and to calculate diet<strong>–</strong>tissue<br />
discrimination values for carbon and nitrogen stable isotopes in<br />
manatee skin.<br />
MATERIALS AND METHODS<br />
Sample collection<br />
Manatees held long-term at SeaWorld of Florida (Orlando, FL, USA)<br />
were fed a diet consisting primarily of romaine lettuce (>90% by<br />
mass) with minimal amounts of other terrestrial vegetation (<strong>17</strong>6 Subadult<strong>–</strong>adult 512<br />
Charlotte F 1136 330 Adult >365<br />
Primo F 494 277 Adult >365<br />
Rita F >900 >275 Adult >365<br />
Sarah F 1136 325 Adult >365<br />
Stubby F 823 252* Subadult<strong>–</strong>adult >365<br />
SWF Tm 0110 F 367 255 Subadult 1231<br />
SWF Tm 0302 F Unknown ><strong>17</strong>6 Subadult<strong>–</strong>adult 512<br />
SWF Tm 0338 F Unknown ><strong>17</strong>6 Subadult<strong>–</strong>adult 429<br />
*Missing large portion of paddle.<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
at various time intervals as they transitioned from wild forage to a<br />
captive diet. Carbon and nitrogen stable isotope turnover rates in skin<br />
were calculated for manatees rescued from two habitats in Florida:<br />
‘coastal’ (Naples on the Gulf coast and Cape Canaveral on the central<br />
east coast) and ‘riverine’ or fresh water (St Johns River near<br />
Jacksonville; Fig.1). The terms ‘coastal’ and ‘riverine’ represent rescue<br />
locations only, and are not intended to describe overall habitat use.<br />
Biopsies of epidermal tissue were collected from the trailing edge<br />
of the paddle using either a scalpel or ronguers (samples were<br />
approximately 10mm5mm, full depth of the paddle). Partially<br />
sloughed epidermis was collected directly off individual manatees<br />
(approximately 2cm2cm) if biopsies were not available. Body mass<br />
and sex were determined, and length measurements taken, where body<br />
length was measured as the straight distance from snout to paddle<br />
(O’Shea et al., 1985). Manatees were categorized into three age classes<br />
based on body length measurements (adults >275cm, subadults/late<br />
juveniles <strong>17</strong>6<strong>–</strong>275cm, and calves
Fig. 1. Rescue locations of manatees in Florida.<br />
time (days) since diet switch. Turnover rate was expressed in terms<br />
of half-life, the time it takes for the isotopic composition of the<br />
tissue to reach a midpoint between the initial and final values:<br />
X = (ln0.5) / c , (3)<br />
In order to better fit turnover data to the exponential model, an<br />
‘anchor point’ based on the mean stable isotope ratio<br />
(δ 13 C=<strong>–</strong>24.4±0.6‰, δ 15 N=2.7±0.5‰, means ± s.e.) of skin samples<br />
from nine long-term captive manatees at SeaWorld Of Florida was<br />
set at 600 days. These animals had been fed a diet of mainly lettuce<br />
for multiple years. The position of the anchor point at 600 days was<br />
chosen for several reasons: it was beyond the maximum sampling<br />
time for all rescued manatees (no manatee was sampled later than<br />
418 days), plots generally reached an asymptote at or before this<br />
point, and positions greater than 600 days did not alter results.<br />
Goodness of fit was first expressed by calculating the coefficient<br />
of determination (R 2 ) using the anchor point as part of the data set.<br />
To further illustrate fit, data for skin from each rescued manatee<br />
were paired with each individual data point contributing to the mean<br />
anchor point and minimum and maximum R 2 values were computed.<br />
All statistical analyses were judged to be significant at P
2352<br />
C. D. Alves-Stanley and G. A. J. Worthy<br />
δ 13 C (‰)<br />
δ 15 N (‰)<br />
<strong>–</strong>10<br />
<strong>–</strong>12<br />
<strong>–</strong>14<br />
<strong>–</strong>16<br />
<strong>–</strong>18<br />
<strong>–</strong>20<br />
<strong>–</strong>22<br />
<strong>–</strong>24<br />
<strong>–</strong>26<br />
<strong>–</strong>28<br />
11<br />
10<br />
9<br />
8<br />
7<br />
6<br />
5<br />
4<br />
3<br />
2<br />
0 100 200 300 400 500 600<br />
1<br />
0 100 200 300 400 500 600<br />
manatees were also fitted to the exponential decay model (Fig.2).<br />
Carbon turnover half-lives in the skin of riverine manatees ranged<br />
from 39 to 72 days with a mean of 59±18 days; however, a halflife<br />
was not calculated for manatee 0341 because the equation<br />
showed little to no change in signature over time (Fig.2; Table4).<br />
Skin samples from these riverine manatees were also enriched<br />
in 15 N relative to that of captive manatees (mean enrichment=<br />
6.3±1.6‰) and nitrogen half-lives ranged from 21 to 115 days with<br />
a mean of 58±42 days (Fig.2; Table4). These turnover times were<br />
not significantly different from carbon half-lives (paired t-test,<br />
t=0.13, d.f.=2, P=0.91). MANOVA results indicated there were no<br />
significant differences in stable carbon or nitrogen isotope half-lives<br />
between manatees rescued from riverine vs coastal regions (F-test:<br />
F2,4=0.58, P=0.60).<br />
DISCUSSION<br />
Carbon enrichment values calculated in the present study<br />
(2.8±0.9‰) were similar to values previously reported. Ames and<br />
colleagues (Ames et al., 1996) found sloughed skin from captive<br />
manatees to be enriched in 13 C by an average of 4.1‰ compared<br />
with lettuce. Reich and Worthy (Reich and Worthy, 2006) assumed<br />
a carbon enrichment value of 3.0‰ in manatee skin when applying<br />
an isotope mixing model (Phillips and Gregg, 2001) to diet<br />
interpretation of free-ranging manatees. The only other known study<br />
on diet<strong>–</strong>tissue discrimination in any mammalian skin is that of<br />
Hobson and colleagues (Hobson et al., 1996) in which seal skin<br />
was found to be enriched in 13 C relative to diet by 2.8‰. In the<br />
present study, nitrogen enrichment could not be determined because<br />
Coastal manatees Riverine manatees<br />
Animal ID Half-life<br />
<strong>–</strong>10<br />
Animal ID Half-life<br />
0301 63 days <strong>–</strong>12<br />
0334 72 days<br />
0318 60 days<br />
0322 45 days<br />
<strong>–</strong>14<br />
0340 39 days<br />
0341 N/A<br />
0431 42 days <strong>–</strong>16<br />
0501 67 days<br />
Animal ID Half-life<br />
0301 36 days<br />
0318 33 days<br />
0322 14 days<br />
0431 23 days<br />
<strong>–</strong>18<br />
<strong>–</strong>20<br />
<strong>–</strong>22<br />
<strong>–</strong>24<br />
<strong>–</strong>26<br />
<strong>–</strong>28<br />
11<br />
10<br />
9<br />
8<br />
7<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0 100 200 300 400 500 600<br />
Days since diet change<br />
0 100 200 300 400 500 600<br />
Animal ID Half-life<br />
0334 115 days<br />
0340 34 days<br />
0341 62 days<br />
0501 21 days<br />
Fig. 2. 13 C and 15 N turnover in epidermis from manatees rescued near Naples and Cape Canaveral (coastal manatees) and from the St Johns River (riverine<br />
manatees) in Florida. Mean (±95% CI) stable isotope ratios for skin from long-term captive manatees () were used as ‘anchor points’ set at 600 days to<br />
better fit the models. To illustrate diet<strong>–</strong>tissue discrimination, mean (±95% CI) stable isotope ratios for the main diet items fed in captivity (romaine lettuce and<br />
spinach) are indicated by a horizontal solid black line and horizontal dashed lines.<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
of the high degree of variability of nitrogen signatures in the lettuce<br />
diet. Typically, diet<strong>–</strong>tissue discrimination values for nitrogen are in<br />
the range 2<strong>–</strong>5‰ (Peterson and Fry, 1987; Kelly, 2000).<br />
Manatees rescued from coastal regions were ideal subjects for<br />
carbon isotope turnover calculations because carbon signatures in<br />
their skin differed dramatically from those of captive manatees.<br />
Interpreting δ 13 C values in the skin of riverine manatees was<br />
problematic because of variability in values at the time of rescue<br />
and the similarity of δ 13 C values between the skin of rescued<br />
manatees and that of long-term captive manatees. The half-lives in<br />
skin from riverine manatees were not significantly different from<br />
those of skin from coastal manatees; however, the carbon turnover<br />
data for manatees rescued from riverine regions did not fit the<br />
exponential decay models as closely as those for coastal manatees.<br />
The carbon half-life calculated for manatee epidermis was very<br />
slow compared with previous turnover studies on other species.<br />
Stable isotope turnover rates can differ based on the particular<br />
isotope, tissue and/or taxon analyzed, diet, physiological state,<br />
feeding rate and/or growth rate of the animal (e.g. Fry and Arnold,<br />
1982; Bosley et al., 2002; Hobson and Bairlein, 2003; Olive et al.,<br />
2003). Additionally, some studies removed lipids from samples<br />
while others did not. Therefore, direct comparisons between studies<br />
are difficult. Dalerum and Angerbjorn (Dalerum and Angerbjorn,<br />
2005) cautioned that comparisons of turnover rates should be made<br />
between the same tissues to avoid these complications. Additionally,<br />
turnover rates in tissues should be compared between animals of<br />
similar body size as metabolic rates have an effect on isotope<br />
turnover (Sponheimer et al., 2006).
The present study is the first to calculate stable isotope turnover<br />
rates in the skin of any marine mammal species. Isotope turnover<br />
rates that have been reported for large terrestrial mammals including<br />
bears (Hilderbrand et al., 1996), alpacas (Sponheimer et al., 2006),<br />
and domestic cattle and horses (Schwertl et al., 2003; Ayliffe et al.,<br />
2004) were determined using blood, muscle and liver, and hair,<br />
respectively. As no appropriate comparison between turnover rates<br />
in the skin of large mammals was possible, the results from this<br />
study will be cautiously compared with others. Previous studies on<br />
stable isotope ratios in hair were omitted from this comparison as<br />
hair is a metabolically inert tissue in which the isotopic composition<br />
represents the period of growth.<br />
The only reported carbon half-lives in animal tissue that are<br />
greater than those of manatee epidermis are those of alpaca muscle<br />
[<strong>17</strong>9 days (Sponheimer et al., 2006)], bat wing membrane and whole<br />
blood [102<strong>–</strong>134 days (Voigt et al., 2003)], and quail bone collagen<br />
[<strong>17</strong>3 days (Hobson and Clark, 1992a)]. In the alpaca study, muscle<br />
tissues were not lipid extracted so direct comparisons may be<br />
problematic. Voigt and colleagues (Voigt et al., 2003) suggested<br />
the slow turnover rate in bat wing membrane was largely due to the<br />
tissue being composed primarily of collagen and elastin, which are<br />
known to have slow turnover rates in bone. Additionally, Voigt and<br />
colleagues (Voigt et al., 2003) attributed the slow turnover rate in<br />
bat blood to their long-lived erythrocytes. Finally, the long half-life<br />
in quail bone aligns with collagen being a less metabolically active<br />
tissue. Epidermal tissue is composed of keratin in the epithelial<br />
lamina and collagen and elastin in the basal lamina. Manatee<br />
epidermis has been described as thick and possesses the<br />
characteristic of hyperkeratosis (Sokolov, 1982; Graham et al.,<br />
2003). It is possible that a slow replacement of keratin in manatee<br />
Manatee stable isotope turnover<br />
Table 4. Exponential decay equations and half-lives representing stable isotope turnover in epidermis sampled from rehabilitated Florida<br />
manatees<br />
R 2 range<br />
Animal Half-life δ13C at<br />
Carbon turnover ID Equation R2 (days) day 0 (‰) Min. Max.<br />
Coastal manatees 0301 y=<strong>–</strong>24.4+14.2e <strong>–</strong>0.01094x 1.00 63 <strong>–</strong>10.2 1.00 1.00<br />
0318 y=<strong>–</strong>24.8+15.1e <strong>–</strong>0.01158x 0.97 60 <strong>–</strong>9.7 0.96 0.97<br />
0322 y=<strong>–</strong>24.5+14.3e <strong>–</strong>0.01550x 0.83 45 <strong>–</strong>10.2 0.79 0.84<br />
0431 y=<strong>–</strong>24.1+11.6e <strong>–</strong>0.01657x 0.97 42 <strong>–</strong>12.5 0.97 0.97<br />
Mean 53 <strong>–</strong>10.7<br />
Riverine manatees 0334 y=<strong>–</strong>24.6+8.2e <strong>–</strong>0.00963x 0.95 72 <strong>–</strong>16.4 0.90 0.96<br />
0340 y=<strong>–</strong>24.7+0.9e <strong>–</strong>0.0<strong>17</strong>87x 0.36 39 <strong>–</strong>23.8 0.05 0.51<br />
0341 y=<strong>–</strong>23.8+1.9e <strong>–</strong>20920x 0.50 n.d. <strong>–</strong>21.9 0.43 0.51<br />
0501 y=<strong>–</strong>24.5+6.2e <strong>–</strong>0.01028x 0.70 67 <strong>–</strong>18.3 0.59 0.75<br />
Mean 59 <strong>–</strong>20.1<br />
R 2 range<br />
Animal Half-life δ15N at<br />
Nitrogen turnover ID Equation R2 (days) day 0 (‰) Min. Max.<br />
Coastal manatees 0301 y=2.3+3.4e <strong>–</strong>0.01918x 0.96 36 5.7 0.83 1.00<br />
0318 y=2.4+2.4e <strong>–</strong>0.02125x 0.57 33 4.8 0.46 0.62<br />
0322 y=2.3+3.6e <strong>–</strong>0.04898x 0.56 14 5.9 0.29 0.68<br />
0431 y=2.5+4.7e <strong>–</strong>0.03038x 0.97 23 7.2 0.91 0.97<br />
Mean 27 5.9<br />
Riverine manatees 0334 y=2.6+4.9e <strong>–</strong>0.00601x 1.00 115 7.5 0.98 1.00<br />
0340 y=2.7+6.4e <strong>–</strong>0.02055x 0.92 34 9.1 0.90 0.92<br />
0341 y=2.3+5.9e <strong>–</strong>0.01125x 0.85 62 8.2 0.80 0.87<br />
0501 y=2.0+9.4e <strong>–</strong>0.03273x 0.95 21 11.4 0.92 0.97<br />
n.d., not determined (see text).<br />
Mean 58 9.1<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
2353<br />
epidermis and the presence of collagen and elastin in the basal lamina<br />
also contributed to the slow isotope turnover rate in the skin.<br />
Another potentially significant factor impacting on turnover rate<br />
is the manatees’ overall slow metabolism. Metabolic rates in adult<br />
Florida manatees have been shown to be lower than those predicted<br />
based on body size [15<strong>–</strong>40% of predicted values (Irvine, 1983;<br />
Worthy et al., 2000)]. It is also possible that food passage time<br />
impacts on turnover rate [as discussed by Post (Post, 2002)].<br />
Manatees use hindgut fermentation and have a passage rate through<br />
the digestive tract of 146<strong>–</strong>147h (Lomolino and Ewel, 1984; Larkin<br />
et al., 2007). This rate is consistent with that of the dugong, another<br />
sirenian [145<strong>–</strong>166h (Lanyon and Marsh, 1995)], but much slower<br />
than those of other large hindgut fermenters such as elephants<br />
[21<strong>–</strong>46h (Rees, 1982)], horses [26<strong>–</strong>27h (Rosenfeld et al., 2006)]<br />
and rhinoceros [61h (Clauss et al., 2005)]. Manatees sampled in<br />
the present study were rescued for reasons including cold stress,<br />
entanglement and watercraft injuries. Because of their physical<br />
condition, their intake rates may have been slower than those of<br />
manatees not in need of rehabilitation.<br />
The gruel mixture fed to the manatees for the first few weeks of<br />
rehabilitation was enriched in 13 C compared with lettuce and<br />
spinach. It is presumed that the initial supplementation of the diet<br />
with gruel would have had minimal to no impact on the carbon<br />
turnover rate as turnover was already very slow. There were no<br />
significant differences in δ 13 C values between biopsy and sloughed<br />
skin samples, so the differing sample types had no effect on carbon<br />
turnover rate. Carbon isotope ratios of manatee fecal material did<br />
not differ from those of the main diet items, so even if manatees<br />
were engaging in coprophagy it would not have had any effect on<br />
carbon turnover rate in the skin. This result is another indication
2354<br />
C. D. Alves-Stanley and G. A. J. Worthy<br />
that stable isotope analysis of fecal material has great potential in<br />
assessing short-term, recent dietary choices.<br />
Phillips and Koch (Phillips and Koch, 2002) suggested<br />
incorporating carbon and nitrogen concentration analysis to aid in<br />
stable isotope dietary reconstruction, especially when there are great<br />
differences in C and N concentration between diet items. We were<br />
unable to include a concentration analysis in this study; however,<br />
it may be a useful addition to future analyses.<br />
Coastal manatees were poor subjects for nitrogen turnover<br />
calculations because of the high variability in their initial δ 15 N values<br />
at the time of rescue, the similarity of δ 15 N values between the skin<br />
of rescued animals and that of captive animals, and the high<br />
variability in δ 15 N values of captive diet items. Consequently,<br />
calculated nitrogen half-lives were inconsistent. Riverine manatees<br />
were better subjects because there was a greater difference in<br />
nitrogen signatures of skin between rescued and long-term captive<br />
animals. Even then, nitrogen half-lives for skin from riverine<br />
manatees were still variable, most likely because of the variability<br />
in nitrogen signatures of romaine lettuce and spinach fed in captivity.<br />
The lettuce and spinach in the captive manatee diet often originated<br />
from different agricultural producers and it is possible that different<br />
fertilization techniques were used resulting in high variability in<br />
δ 15 N values (see Georgi et al., 2005).<br />
There are very few studies on nitrogen turnover in other species.<br />
Nitrogen half-lives in mammalian tissues have been calculated for<br />
blood plasma and cells in black bears [3 and 22 days, respectively<br />
(Hilderbrand et al., 1996)] and avian whole blood [10.0<strong>–</strong>14.4 days<br />
(Bearhop et al., 2002; Hobson and Bairlein, 2003; Ogden et al.,<br />
2004)] and plasma [0.5<strong>–</strong>1.7 days (Pearson et al., 2003)]. Nitrogen<br />
turnover in manatee skin was relatively slow compared with the<br />
results of these studies, most likely owing to the low metabolic rate<br />
of manatees, epidermal tissue composition and/or food passage rate<br />
as previously discussed in terms of carbon turnover.<br />
There was no compounding effect of the gruel supplement on<br />
nitrogen turnover rate as the signature of the gruel was not<br />
significantly different from that of romaine lettuce and spinach.<br />
Likewise, coprophagy would have had no effect on nitrogen<br />
turnover rate in manatee skin as the δ 15 N values of manatee fecal<br />
material did not differ from those of the main diet items. However,<br />
sloughed skin samples were significantly enriched in 15 N compared<br />
with biopsy samples. Though δ 15 N values were adjusted to account<br />
for this enrichment, variability in nitrogen turnover rates and lack<br />
of fit indicate that sample type may have contributed to the difficulty<br />
in calculating more precise half-lives. At the present time, it is<br />
unclear why sloughed samples differed in δ 15 N values, but not δ 13 C<br />
values, from biopsy samples. It is possible that materials were<br />
redistributed within the skin while sloughing, and sloughed samples<br />
also contained less epidermal depth than biopsies. Both of these<br />
factors could have contributed to differing δ 15 N values between<br />
sample types.<br />
Cerling and colleagues (Cerling et al., 2007) and Ayliffe and<br />
colleagues (Ayliffe et al., 2004) have suggested that some stable<br />
isotope turnover data may be better fitted to a multiple-pool model<br />
than to the traditional single-pool, exponential decay model used<br />
in the present study. Proper application requires knowledge of each<br />
pool in vivo (Sponheimer et al., 2006), which at this time is unknown<br />
for manatees. Consequently, we did not incorporate a multiple-pool<br />
analysis. Single-pool analyses are still useful for intraspecific and<br />
interspecific comparisons of stable isotope turnover (Sponheimer<br />
et al., 2006).<br />
When proportions of food sources contributing to a mixed diet<br />
are unknown, mixing models are often used to aid in estimating these<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
proportions (e.g. Phillips and Gregg, 2001; Newsome et al., 2004;<br />
Reich and Worthy, 2006). If a change in diet occurs, the resulting<br />
signature may not be representative of the current diet but, in fact,<br />
may be some intermediate value between two distinct diets. While<br />
this possibility is true for all stable isotope analyses, the impact is<br />
minimized in tissues with high turnover rates because the time frame<br />
is short. Free-ranging manatees are known to switch diet sources<br />
(Best, 1981; Lefebvre et al., 2000), and the very slow turnover rates<br />
for carbon and nitrogen stable isotopes in epidermis tissue mean that<br />
unless the manatee has been feeding on the same diet for an extended<br />
period of time, the skin signature will always be in some transitional<br />
state. Slow turnover rates in manatee skin especially complicate<br />
estimation of the proportions of freshwater, estuarine and marine<br />
sources in the diet because δ 13 C values for estuarine vegetation are<br />
intermediate between those of freshwater vegetation and seagrasses<br />
(Reich and Worthy, 2006) (C.D.A.-S. and G.A.J.W., in preparation).<br />
The incorporation of nitrogen stable isotope analysis can aid in further<br />
separation of these three diet sources as nitrogen signatures for<br />
freshwater and estuarine vegetation differ from those of seagrasses<br />
(C.D.A.-S. and G.A.J.W., in preparation).<br />
Computing a precise diet<strong>–</strong>tissue discrimination value is essential<br />
when interpreting isotopic results. Discrimination values for carbon<br />
have previously been calculated for manatee skin on a captive diet<br />
(Ames et al., 1996) and discrimination values for carbon and nitrogen<br />
have been estimated in the skin of free-ranging manatees on<br />
possible diets of freshwater, estuarine and/or marine vegetation<br />
(Reich and Worthy, 2006). It is unknown whether diet<strong>–</strong>tissue<br />
discrimination in manatee skin differs between diet types as has<br />
been shown in other studies (e.g. Hobson and Clark, 1992b). The<br />
results of the present study are the most extensive thus far.<br />
Carbon and nitrogen stable isotope analysis of manatee epidermal<br />
tissue is difficult, if not impossible, to use when assessing shortterm<br />
or recent changes in diet and habitat use because of slow<br />
turnover rates. This technique would potentially have more direct<br />
application in summarizing average dietary intake over longer<br />
periods of time. In order to accurately interpret isotopic analyses,<br />
determining diet<strong>–</strong>tissue discrimination factors and turnover rates in<br />
the tissue is essential. The difficulty with most studies is that isotope<br />
discrimination and turnover are best calculated under controlled<br />
situations in captivity. Mixing model results for tissues with slow<br />
turnover rates should be interpreted with caution, especially in<br />
species that may be switching between diets in which an intermediate<br />
isotope ratio may be mistakenly described as indicating a single<br />
diet source instead of a mixture of sources.<br />
We thank R. Bonde, M. Ross and B. Chittick for assistance with manatee skin<br />
sample collections and we are grateful for funding from the University of Central<br />
Florida (UCF) Department of Biology and Graduate Studies to C.D.A.-S. and a<br />
Provosts Research Enhancement Award to G.A.J.W. Manatee research was<br />
carried out under UCF-Institutional Animal Care and Use Committee protocol 02-<br />
09W and US Fish and Wildlife Service (USFWS) permit number MA056326<br />
issued to G.A.J.W. and USFWS permit number MA79<strong>17</strong>21 issued to R. Bonde.<br />
Plant samples were collected under Florida Department of Environmental<br />
Protection plant collection permit number <strong>17</strong>53 issued to G.A.J.W. We thank J.<br />
Roth and J. Weishampel for assistance with editing and J. Fauth for assistance<br />
with statistical analyses. Additional thanks to T. Maddox, R. Harris, R. Runnels, T.<br />
Doyle, K. Fuhr, J. Greenawalt, L. Hoopes, A. Stephens, M. DiPiazza, N. Browning<br />
and J. Stanley for field and lab assistance.<br />
References<br />
Ames, A. L., VanVleet, E. S. and Sackett, W. M. (1996). The use of stable carbon<br />
isotope analysis for determining the dietary habits of the Florida manatee,<br />
Trichechus manatus latirostris. Mar. Mamm. Sci. 12, 555-563.<br />
Ayliffe, L. K., Cerling, T. E., Robinson, T., West, A. G., Sponheimer, M., Passey,<br />
B. H., Hammer, J., Roeder, B., Dearing, M. D. and Ehleringer, J. R. (2004).<br />
Turnover of carbon isotopes in tail hair and breath CO2 of horses fed an isotopically<br />
varied diet. Oecologia 139, 11-22.
Bearhop, S., Waldron, S., Votier, S. C. and Furness, R. W. (2002). Factors that<br />
influence assimilation rates and fractionation of nitrogen and carbon stable isotopes<br />
in avian blood and feathers. Physiol. Biochem. Zool. 75, 451-458.<br />
Best, R. C. (1981). Foods and feeding habits of wild and captive Sirenia. Mamm. Rev.<br />
11, 3-29.<br />
Bosley, K. L., Witting, D. A., Chambers, R. C. and Wainright, S. C. (2002).<br />
Estimating turnover rates of carbon and nitrogen in recently metamorphosed winter<br />
flounder Pseudopleuronectes americanus with stable isotopes. Mar. Ecol. Prog. Ser.<br />
236, 233-240.<br />
Campbell, H. W. and Irvine, A. B. (1977). Feeding ecology of West Indian manatee<br />
Trichechus manatus Linnaeus. Aquaculture 12, 249-251.<br />
Cerling, T. E., Hart, J. A. and Hart, T. B. (2004). Stable isotope ecology in the Ituri<br />
Forest. Oecologia 138, 5-12.<br />
Cerling, T. E., Ayliffe, L. K., Dearing, M. D., Ehleringer, J. R., Passey, B. H.,<br />
Podlesak, D. W. A., T. and West, A. G. (2007). Determining biological tissue turnover<br />
using stable isotopes: the reaction progress variable. Oecologia 151, <strong>17</strong>5-189.<br />
Cherel, Y., Hobson, K. A. and Hassani, S. (2005). Isotopic discrimination between<br />
food and blood and feathers of captive penguins: implications for dietary studies in<br />
the wild. Physiol. Biochem. Zool. 78, 106-115.<br />
Clauss, M., Polster, C., Kienzle, E., Wiesner, H., Baumgartner, K., von Houwald,<br />
F., Ortmann, S., Streich, W. J. and Dierenfeld, E. S. (2005). Studies on digestive<br />
physiology and feed digestibilities in captive Indian rhinoceros (Rhinoceros<br />
unicornis). J. Anim. Physiol. Anim. Nutr. 89, 229-237.<br />
Clementz, M. T. and Koch, P. L. (2001). Differentiating aquatic mammal habitat and<br />
foraging ecology with stable isotopes in tooth enamel. Oecologia 129, 461-472.<br />
Cloern, J. E., Canuel, E. A. and Harris, D. (2002). Stable carbon and nitrogen<br />
isotope composition of aquatic and terrestrial plants of the San Francisco Bay<br />
estuarine system. Limnol. Oceanogr. 47, 713-729.<br />
Craig, H. (1957). Isotopic standards for hydrogen and oxygen and correlation factors<br />
for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 42,<br />
495-506.<br />
Dalerum, F. and Angerbjorn, A. (2005). Resolving temporal variation in vertebrate<br />
diets using naturally occurring stable isotopes. Oecologia 144, 647-658.<br />
Deniro, M. J. and Epstein, S. (1978). Influence of diet on distribution of carbon<br />
isotopes in animals. Geochim. Cosmochim. Acta 42, 495-506.<br />
Deniro, M. J. and Epstein, S. (1981). Influence of diet on the distribution of nitrogen<br />
isotopes in animals. Geochim. Cosmochim. Acta 45, 341-351.<br />
Fry, B. (1981). Natural stable carbon isotope tag traces Texas shrimp migrations. Fish.<br />
Bull. 79, 337-345.<br />
Fry, B. and Arnold, C. (1982). Rapid C-13/C-12 turnover during growth of brown<br />
shrimp (Penaeus aztecus). Oecologia 54, 200-204.<br />
Fry, B. and Sherr, E. B. (1984). δ 13 C measurements as indicators of carbon flow in<br />
marine and freshwater ecosystems. Contrib. Mar. Sci. 27, 13-47.<br />
Georgi, M., Voerkelius, S., Rossmann, A., Grassmann, J. and Schnitzler, W. H.<br />
(2005). Multielement isotope ratios of vegetables from integrated and organic<br />
production. Plant Soil 275, 93-100.<br />
Graham, A. R., Samuelson, D. A., Isaza, R. and Lewis, P. A. (2003). Histological<br />
comparison of manatee and elephant integument. In 15th Biennial Conference on<br />
the Biology of Marine Mammals, pp. 62-63. Greensboro, NC: Society for Marine<br />
Mammalogy.<br />
Hall-Aspland, S. A., Rogers, T. L. and Canfield, R. B. (2005). Stable carbon and<br />
nitrogen isotope analysis reveals seasonal variation in the diet of leopard seals. Mar.<br />
Ecol. Prog. Ser. 305, 249-259.<br />
Hartman, D. S. (1979). Ecology and behavior of the manatee (Trichechus manatus) in<br />
Florida: The American Society of Mammalogists Special Publication No. 5.<br />
Hilderbrand, G. V., Farley, S. D., Robbins, C. T., Hanley, T. A., Titus, K. and<br />
Servheen, C. (1996). Use of stable isotopes to determine diets of living and extinct<br />
bears. Can. J. Zool. 74, 2080-2088.<br />
Hobson, K. A. (1999). Tracing origins and migration of wildlife using stable isotopes: a<br />
review. Oecologia 120, 314-326.<br />
Hobson, K. A. and Bairlein, F. (2003). Isotopic fractionation and turnover in captive<br />
garden warblers (Sylvia borin): implications for delineating dietary and migratory<br />
associations in wild passerines. Can. J. Zool. 81, 1630-1635.<br />
Hobson, K. A. and Clark, R. G. (1992a). Assessing avian diets using stable isotopes<br />
1. Turnover of 13 C in tissues. Condor 94, 181-188.<br />
Hobson, K. A. and Clark, R. G. (1992b). Assessing avian diets using stable isotopes<br />
2. Factors influencing diet-tissue fractionation. Condor 94, 189-197.<br />
Hobson, K. A. and Welch, H. E. (1992). Determination of trophic relationships within<br />
a high Arctic marine food web using δ 13 C and δ 15 N analysis. Mar. Ecol. Prog. Ser.<br />
84, 9-18.<br />
Hobson, K. A., Schell, D. M., Renouf, D. and Noseworthy, E. (1996). Stable carbon<br />
and nitrogen isotopic fractionation between diet and tissues of captive seals:<br />
implications for dietary reconstructions involving marine mammals. Can. J. Fish<br />
Aquat. Sci. 53, 528-533.<br />
Irvine, A. B. (1983). Manatee metabolism and its influence on distribution in Florida.<br />
Biol. Conserv. 25, 315-334.<br />
Kelly, J. F. (2000). Stable isotopes of carbon and nitrogen in the study of avian and<br />
mammalian trophic ecology. Can. J. Zool. 78, 1-27.<br />
Kurle, C. M. and Worthy, G. A. J. (2002). Stable nitrogen and carbon isotope ratios in<br />
multiple tissues of the northern fur seal Callorhinus ursinus: implications for dietary<br />
and migratory reconstructions. Mar. Ecol. Prog. Ser. 236, 289-300.<br />
Lanyon, J. M. and Marsh, H. (1995). Digesta passage times in the dugong. Aust. J.<br />
Zool. 43, 119-127.<br />
Larkin, I. L. V., Fowler, V. F. and Reep, R. L. (2007). Digesta passage rates in the<br />
Florida manatee (Trichechus manatus latirostris). Zoo Biol. 26, 503-515.<br />
Lee, S. H., Schell, D. M., McDonald, T. L. and Richardson, W. J. (2005). Regional<br />
and seasonal feeding by bowhead whales Balaena mysticetus as indicated by stable<br />
isotope ratios. Mar. Ecol. Prog. Ser. 285, 271-287.<br />
THE JOURNAL OF EXPERIMENTAL BIOLOGY<br />
Manatee stable isotope turnover<br />
2355<br />
Lefebvre, L. W., Ackerman, B. B., Portier, K. M. and Pollock, K. H. (1995). Aerial<br />
survey as a technique for estimating trends in manatee population size: problems<br />
and prospects. In Population biology of the Florida Manatee (ed. T. J. O’Shea, B. B.<br />
Ackerman and H. F. Percival), pp. 63-74. Washington, DC: Information and<br />
Technology Report 1, National Biological Service.<br />
Lefebvre, L. W., Reid, J. P., Kenworthy, W. J. and Powell, J. A. (2000).<br />
Characterizing manatee habitat use and seagrass grazing in Florida and Puerto<br />
Rico: implications for conservation and management. Pac. Conserv. Biol. 5, 289-<br />
298.<br />
Logan, J., Haas, H., Deegan, L. and Gaines, E. (2006). Turnover rates of nitrogen<br />
stable isotopes in the salt marsh mummichog, Fundulus heteroclitus, following a<br />
laboratory diet switch. Oecologia 147, 391-395.<br />
Lomolino, M. W. and Ewel, K. C. (1984). Digestive efficiencies of the West Indian<br />
manatee (Trichechus manatus). Florida Sci. 47, <strong>17</strong>6-<strong>17</strong>9.<br />
MacFadden, B. J., Higgins, P., Clementz, M. T. and Jones, D. S. (2004). Diets,<br />
habitat preferences, and niche differentiation of Cenozoic sirenians from Florida:<br />
evidence from stable isotopes. Paleobiology 30, 297-324.<br />
Newsome, S. D., Phillips, D. L., Culleton, B. J., Guilderson, T. P. and Koch, P. L.<br />
(2004). Dietary reconstruction of an early to middle Holocene human population from<br />
the central California coast: insights from advanced stable isotope mixing models. J.<br />
Archaeol. Sci. 31, 1101-1115.<br />
Newsome, S. D., Koch, P. L., Etnier, M. A. and Aurioles-Gambao, D. (2006). Using<br />
carbon and nitrogen isotope values to investigate maternal strategies in northeast<br />
Pacific otariids. Mar. Mamm. Sci. 22, 556-572.<br />
O’Shea, T. J., Beck, C. A., Bonde, R. K., Kochman, H. I. and Odell, D. K. (1985).<br />
An analysis of manatee mortality patterns in Florida, 1976-81. J. Wildl. Manage. 49,<br />
1-11.<br />
Ogden, L. J. E., Hobson, K. A. and Lank, D. B. (2004). Blood isotopic (d 13 C and<br />
d 15 N) turnover and diet-tissue fractionation factors in captive Dunlin (Calidris alpina<br />
pacifica). Auk 121, <strong>17</strong>0-<strong>17</strong>7.<br />
Olive, P. J. W., Pinnegar, J. K., Polunin, N. V. C., Richards, G. and Welch, R.<br />
(2003). Isotope trophic-step fractionation: a dynamic equilibrium model. J. Anim.<br />
Ecol. 72, 608-6<strong>17</strong>.<br />
Pearson, S. F., Levey, D. J., Greenberg, C. H. and del Rio, C. M. (2003). Effects of<br />
elemental composition on the incorporation of dietary nitrogen and carbon isotopic<br />
signatures in an omnivorous songbird. Oecologia 135, 516-523.<br />
Peterson, B. J. and Fry, B. (1987). Stable isotopes in ecosystem studies. Annu. Rev.<br />
Ecol. Syst. 18, 293-320.<br />
Phillips, D. L. and Gregg, J. W. (2001). Uncertainty in source partitioning using stable<br />
isotopes. Oecologia 127, <strong>17</strong>1-<strong>17</strong>9.<br />
Phillips, D. L. and Koch, P. L. (2002). Incorporating concentration dependence in<br />
stable isotope mixing models. Oecologia 130, 114-125.<br />
Post, D. M. (2002). Using stable isotopes to estimate trophic position: models,<br />
methods, and assumptions. Ecology 83, 703-718.<br />
Ramsay, M. A. and Hobson, K. A. (1991). Polar bears make little use of terrestrial<br />
food webs: evidence from stable-carbon isotope analysis. Oecologia 86, 598-600.<br />
Rau, G. H., Ainley, D. G., Bengtson, J. L., Torres, J. J. and Hopkins, T. L. (1992).<br />
15 N/ 14 N and 13 C/ 12 C in Weddell Sea birds, seals, and fish: implications for diet and<br />
trophic structure. Mar. Ecol. Prog. Ser. 84, 1-8.<br />
Rees, R. A. (1982). Gross assimilation efficiency and food passage time in the African<br />
elephant. Afr. J. Ecol. 20, 193-198.<br />
Reich, K. J. and Worthy, G. A. J. (2006). An isotopic assessment of the feeding<br />
habits of free-ranging manatees. Mar. Ecol. Prog. Ser. 322, 303-309.<br />
Rosenfeld, I., Austbo, D. and Volden, H. (2006). Models for estimating digesta<br />
passage kinetics in the gastrointestinal tract of the horse. J. Anim. Sci. 83, 3321-<br />
3328.<br />
Roth, J. D. and Hobson, K. A. (2000). Stable carbon and nitrogen isotopic<br />
fractionation between diet and tissue of captive red fox: implications for dietary<br />
reconstruction. Can. J. Zool. 78, 848-852.<br />
Schwertl, M., Auerswald, K. and Schnyder, H. (2003). Reconstruction of the isotopic<br />
history of animal diets by hair segmental analysis. Rapid Commun. Mass Spectrom.<br />
<strong>17</strong>, 1312-1318.<br />
Seminoff, J. A., Jones, T. T., Eguchi, T., Jones, D. R. and Dutton, P. H. (2006).<br />
Stable isotope discrimination (δ 13 C and δ 15 N) between soft tissues of the green sea<br />
turtle Chelonia mydas and its diet. Mar. Ecol. Prog. Ser. 308, 271-278.<br />
Sokolov, V. E. (1982). Mammal Skin. Berkeley, CA: University of California Press.<br />
Sponheimer, M., Robinson, T. F., Cerling, T. E., Tegland, L., Roeder, B. L., Ayliffe,<br />
L., Dearing, M. D. and Ehleringer, J. R. (2006). Turnover of stable carbon isotopes<br />
in the muscle, liver, and breath CO2 of alpacas (Lama pacos). Rapid Commun.<br />
Mass Spectrom. 20, 1395-1399.<br />
Tieszen, L. L., Boutton, T. W., Tesdahl, K. G. and Slade, N. A. (1983). Fractionation<br />
and turnover of stable carbon isotopes in animal tissues: implications for δ 13 C<br />
analysis of diet. Oecologia 57, 32-37.<br />
US Fish and Wildlife Service (2001). Florida Manatee Recovery Plan (Trichechus<br />
manatus latirostris), 3rd revision. Atlanta, GA: US Fish and Wildlife Service.<br />
Voigt, C. C., Matt, F., Michener, R. and Kunz, T. H. (2003). Low turnover rates of<br />
carbon isotopes in tissues of two nectar-feeding bat species. J. Exp. Biol. 206, 1419-<br />
1427.<br />
Walker, J. L., Potter, C. W. and Macko, S. A. (1999). The diets of modern and<br />
historic bottlenose dolphin populations reflected through stable isotopes. Mar.<br />
Mamm. Sci. 15, 335-350.<br />
Worthy, G. A. J., Miculka, T. A. and Wright, S. D. (2000). Manatee response to cold:<br />
how cold is too cold? In Florida Manatees and Warm Water: Proceedings of the<br />
Warm Water Workshop (ed. W. Perkins), pp. 1-6. Jacksonville, FL: US Fish and<br />
Wildlife Service.<br />
Yamamuro, M., Aketa, K. and Uchida, S. (2004). Carbon and nitrogen stable isotope<br />
ratios of the tissues and gut contents of a dugong from the temperate coast of<br />
Japan. Mammal Study 29, <strong>17</strong>9-183.
Trend Detection in a Boat-based Method for Monitoring <strong>Sirenian</strong>s: Antillean Manatee<br />
Case Study<br />
Katherine S. LaCommare a* , Solange Brault a , Caryn Self-Sullivan b , Ellen M. Hines c<br />
a University of Massachusetts, Boston 100 Morrissey Blvd. Boston, MA 02125<br />
katie.lacommare@umb.edu, solange.brault@umb.edu<br />
b <strong>Sirenian</strong> <strong>International</strong>, Inc., 200 Stonewall Dr. Fredericksburg, VA 22401 caryns@sirenian.org<br />
c Department of Geography and Human Environmental Studies, San Francisco State University,<br />
1600 Holloway Ave. San Francisco, CA 94132 ehines@sfsu.edu<br />
Abstract. Accurate monitoring is a critical step in evaluating the conservation and management<br />
needs of endangered species. We evaluated a low cost, effective survey method for monitoring<br />
West Indian manatees (Trichechus manatus manatus) in Belize, Central America. The<br />
objectives for this paper are (1) to evaluate a count-based population index derived from a boatbased<br />
survey method, (2) to examine trends in manatee abundance in the Drowned Cayes area,<br />
(3) to conduct a power analysis to explore our ability to detect a trend and the ramifications of<br />
survey structure on trend detection. We used a generalized linear model to evaluate the impact<br />
of environmental conditions on sighting probability and to determine whether the number of<br />
manatees observed per 20-minute scan changed from 2001 - 2007. We used simulations to<br />
determine statistical power - the ability to detect potential declines of 10%, 25% or 50% over 15<br />
years and for various sampling regimes. The number of manatees sighted per scan was not<br />
affected by sighting conditions. There was no change in the mean number of manatees sighted<br />
per scan from 2001 to the 2007. Our ability to detect a trend ranged from 9% to 100%<br />
depending on the level of decline, scan duration, number of points surveyed and number of<br />
surveys. This survey protocol is a practical and repeatable way to examine population trends of<br />
sirenians in similar habitats around the world.<br />
Keywords: Monitoring, trend detection, power analysis, West Indian manatees, Trichechus sp.,<br />
sirenians, boat surveys.<br />
*Corresponding author: Tel. (248) 756-3985. E-mail addresses: kslacommare@gmail.com.<br />
Abbreviations: IUCN- <strong>International</strong> Union for the Conservation of Nature; SPAW- Specially Protected<br />
Areas and Wildlife; CITES- Convention on <strong>International</strong> Trade of Endangered Species, MMPA <strong>–</strong> Marine<br />
Mammal Protection Act<br />
1
1 <br />
2 <br />
3 <br />
4 <br />
5 <br />
6 <br />
7 <br />
8 <br />
9 <br />
10 <br />
11 <br />
12 <br />
13 <br />
14 <br />
15 <br />
16 <br />
<strong>17</strong> <br />
18 <br />
19 <br />
20 <br />
21 <br />
22 <br />
1 Introduction<br />
All four extant species in the order Sirenia are vulnerable to extinction due to small<br />
population sizes, population declines, fragmentation and continued exploitation (Deutsch et<br />
al. 2010; Lefebvre et al. 2001; Marsh and Lefebvre 1994). They are listed as vulnerable on<br />
the IUCN Red List (Deutsch et al. 2010), protected by CITES (CITES 2010), the SPAW<br />
protocol (CEP 2010), the Memorandum of Understanding on the Conservation and<br />
Management of Dugongs (CMS 2010), the US Endangered Species Act (USFWS 2001),<br />
and other national and international wildlife protection laws (see Quintana-Rizzo and<br />
Reynolds 2008). For many sirenian populations, little information exists on their status<br />
(e.g. Hines et al. 2005; Lefebvre et al. 2001; Marsh and Lefebvre 1994; Quintana-Rizzo and<br />
Reynolds 2008). Monitoring <strong>–</strong> a preliminary step in determining conservation status,<br />
effects of exploitation and improving management decisions (Gibbs 2000; Gibbs et al.<br />
1998; Marsh and Trenham 2008; Martin et al. 2006) <strong>–</strong> is the process of gathering data to<br />
draw inferences about population abundance changes over time (Yoccoz et al. 2001).<br />
Monitoring is listed as a key objective in most manatee (Trichechus spp) and dugong<br />
(Dugong dugon) conservation and management plans (eg. Auil 1998; CEP 1995; USFWS<br />
2001).<br />
As others have pointed out, “the need for high quality data is clear, the means for<br />
getting it is not (Dawson et al. 2008, p. 20; USFWS 2001),” especially for small<br />
populations of marine mammals that live in complex coastal and riverine habitats <strong>–</strong><br />
conditions that constrain survey designs (see Dawson et al. 2004; Dawson et al. 2008; Dick<br />
and Hines 2011; Hines et al. 2005; Williams and Thomas 2009). A well designed<br />
2
23 <br />
24 <br />
25 <br />
26 <br />
27 <br />
28 <br />
29 <br />
30 <br />
31 <br />
32 <br />
33 <br />
34 <br />
35 <br />
36 <br />
37 <br />
38 <br />
39 <br />
40 <br />
41 <br />
42 <br />
43 <br />
44 <br />
45 <br />
monitoring program requires two key components: a reliable population index and<br />
powerful statistical analysis. Indices are a proxy for population abundance (Caughley<br />
1977; Gibbs 2000; Kindberga et al. 2009) and are based on the premise that systematic<br />
surveys will detect the same proportion of the population over time. Thus, changes in the<br />
number of animals detected reflect changes in population size (Gibbs 2000). To be a good<br />
surrogate to population size, an index must have a positive, linear relationship with actual<br />
abundance (Gibbs 2000; Gibbs et al. 1998; Williams and Thomas 2009), and a constant<br />
detection probability over habitat, sighting conditions and time (Anderson 2001; Gibbs<br />
2000; Gibbs et al. 1998; Thompson 2004; Williams and Thomas 2009).<br />
Detecting trends in abundance depends on statistical power (Gerrodette 1987; Gibbs<br />
2000; Gibbs et al. 1998; Taylor and Gerrodette 1993) and an effective monitoring program<br />
must generate data that can be statistically analyzed to detect trends (Gibbs 2000; Gibbs et<br />
al. 1998). Both the accuracy of the population index and sampling structure <strong>–</strong> number of<br />
plots, survey frequency, scan duration, number of years <strong>–</strong> will affect the ability to detect<br />
changes in population abundance. Power analysis, through simulation, is the process used<br />
to determine the statistical power of an index and sampling regime (Gerrodette 1987; Gibbs<br />
2000; Gibbs et al. 1998; Taylor and Gerrodette 1993).<br />
Most sirenian populations are found in developing nations where monitoring funds<br />
are scarce. To be valuable, monitoring methods must be sound, repeatable and inexpensive<br />
(Aragones et al. in press; Aragones et al. 1997; Dick and Hines 2011; Hines et al. 2005;<br />
Williams and Thomas 2009). This paper evaluates a low cost and repeatable boat-based<br />
method for monitoring sirenians. We used this method on West Indian manatees<br />
(Trichechus manatus manatus) in the Drowned Cayes area of Belize, Central America and<br />
3
46 <br />
47 <br />
48 <br />
49 <br />
50 <br />
51 <br />
52 <br />
53 <br />
54 <br />
55 <br />
56 <br />
57 <br />
58 <br />
59 <br />
60 <br />
61 <br />
62 <br />
63 <br />
64 <br />
65 <br />
66 <br />
examined trends in manatee abundance from 2001-2007. Through simulations, we<br />
conducted a power analysis to determine trend detection ability and the ramifications of<br />
sampling structure on trend detection.<br />
2. Materials and Methods<br />
2.1 Study area<br />
The Drowned Cayes, including Swallow Caye, are a string of mangrove islands -14 km<br />
long by 4 km wide - along the central coast of Belize, 10-15 km east of Belize City and 3-5<br />
km west of the Belize Barrier reef (Figure 1). These islands are surrounded by shallow<br />
seagrass beds. The depth of the study area is less than 1 meter in some places and is never<br />
greater than 6 meters. The study area also encompasses two points along the Belize Barrier<br />
Reef. Prior to this study, aerial and boat surveys and DNA analyses documented that this<br />
area has an important population of manatees (Auil 2004; Bengston and Magor 1979;<br />
Hunter et al. 2010; LaCommare et al. 2008, Morales-Vela et al. 2000; O'Shea and Salisbury<br />
1991, Self-Sullivan 2008).<br />
This study area is typical of manatee coastal habitat making it appropriate for<br />
testing our survey protocol. It comprises: shallow water (1-3m deep), proximity to deeper<br />
channels (3-6 m) for escape from danger (boats), freshwater sources, aquatic vegetation<br />
(mostly submerged), warm water (<strong>17</strong>°C minimum temperature), shelter from storms,<br />
waves, strong winds and currents; and travel corridors between habitat areas (Moraes-<br />
Arraut et al. in press).<br />
4
67 <br />
68 <br />
69 <br />
70 <br />
71 <br />
72 <br />
73 <br />
74 <br />
75 <br />
76 <br />
77 <br />
78 <br />
79 <br />
80 <br />
81 <br />
82 <br />
83 <br />
84 <br />
85 <br />
86 <br />
87 <br />
88 <br />
2.2 Survey design<br />
We devised a point-based survey design to monitor manatee occurrence by counting the<br />
number of manatees sighted during a 20-minute period at designated locations throughout<br />
the Drowned Cayes area (Figure 1). We marked 54 permanent points throughout the study<br />
area with a global positioning system (GPS) and used a small (8m) fiberglass boat to survey<br />
the points on a regular basis. To minimize boat and engine-noise disturbance to the<br />
manatees, we maneuvered the boat with a 8-meter pole when we were within 100 meters of<br />
the exact point location. During this time, 3 - 13 observers started scanning for manatees.<br />
At least three experienced observers were on the boat at all times. Experienced observers<br />
included one of the two of the authors (KSL and CSS), a field assistant that worked with<br />
the authors for the entire study and student interns that spent 3 <strong>–</strong> 6 months at the field site.<br />
Student interns were trained to spot manatees <strong>–</strong> nose, tail, back - and signs of manatees -<br />
shadows under the water, tail pressure wave, disturbed silt, floating grass -by the authors<br />
and field assistant. The number of inexperienced, but trained observers (volunteers),<br />
ranged from 0 to 10. Inexperienced observers were trained to spot manatees and manatee<br />
signs by the authors, field assistants and interns at the beginning of each two-week session.<br />
Volunteers were introduced to manatee viewing first via a PowerPoint lecture and then<br />
asked to practice spotting and viewing live manatees by visiting Swallow Caye Wildlife<br />
Sanctuary <strong>–</strong> a local manatee sanctuary and viewing area. Once anchored in position with<br />
the pole, observers searched for manatees in a 360 o circle around the boat. Having at least<br />
three experienced manatee spotters on the boat at all times ensured that experienced<br />
observers could cover the entire 360 degree circle and not rely on volunteers to detect<br />
5
89 <br />
90 <br />
91 <br />
92 <br />
93 <br />
94 <br />
95 <br />
96 <br />
97 <br />
98 <br />
99 <br />
100 <br />
101 <br />
102 <br />
103 <br />
104 <br />
105 <br />
106 <br />
107 <br />
108 <br />
109 <br />
110 <br />
manatees. A manatee was counted if a tail, back, nose or entire manatee was spotted. If a<br />
manatee was spotted by an inexperienced observer, it was only counted if it was confirmed<br />
by an experienced observer. From 2001-2007, we conducted these point scans 4 - 8 times<br />
per year during two-week survey sessions and sampled the points without replacement for<br />
the two-week period. Since not all points could be surveyed in one day, we divided the<br />
study area into 8 zones with 4 <strong>–</strong> 6 points each. Each day, we randomly chose the survey<br />
zone and starting point and then surveyed 4 <strong>–</strong> 5 points.<br />
2.3 Index and detection probability<br />
Our population index was the number of manatees sighted per 20-minutes per point. A<br />
valid population index must have a constant, linear relationship with manatee abundance.<br />
Two conditions can violate this assumption. When using call indices (indices that rely on<br />
animal calls such as frog calls or bird songs) or presence/absence data (Gibbs 2000; Gibbs<br />
et al. 1998), high population densities have an asymptotic relationship between the index<br />
and abundance. Alternatively, low population densities can have a threshold below which<br />
animals will not be detected (Gibbs 2000; Gibbs et al. 1998).<br />
We were not using presence/absence data or call indices, and since we were<br />
evaluating population declines rather than population increases, we were more concerned<br />
with the problem of a bottom threshold than population saturation. This point-based<br />
method has been successfully used to locate and count small and sparse populations of<br />
manatees in: Lake Volta, Ghana; Estero Hondo, Dominican Republic; and Lake Ossa,<br />
Cameroon (Self-Sullivan, personal communication; Dominquez, personal communication).<br />
This method can detect manatees at low population densities. To further evaluate this<br />
6
111 <br />
112 <br />
113 <br />
114 <br />
115 <br />
116 <br />
1<strong>17</strong> <br />
118 <br />
119 <br />
120 <br />
121 <br />
122 <br />
123 <br />
124 <br />
125 <br />
126 <br />
127 <br />
128 <br />
129 <br />
130 <br />
131 <br />
132 <br />
133 <br />
assumption, we divided our locations into high-sighting probability sites (>30% probability<br />
of sighting a manatee per scan) and low-sighting probability sites (< 30% probability of<br />
sighting a manatee per scan) (LaCommare et al. 2008). We then examined cumulative<br />
histograms of time to first sighting for both categories. We expected that each histogram<br />
would have a similar asymptotic curve indicating that our ability to detect manatees was<br />
similar for high versus low-sighting probability locations.<br />
The relationship of the index to the actual number of animals counted is also a<br />
function of detection probability (Anderson 2001; Gibbs 2000; Gibbs et al. 1998;<br />
Thompson 2004; Williams and Thomas 2009) <strong>–</strong> the ability to detect an animal when it is<br />
present (MacKenzie and Royle 2005). Three classes of variables affect this: observers,<br />
environment and species behavior. For each point scan, we counted the number of<br />
manatees and recorded variables that might influence detection probabilities. We used a<br />
generalized linear model (GLM) with a negative binomial distribution and a log-link<br />
function to determine whether the number of manatees sighted was influenced by sighting<br />
conditions. This distribution was the most appropriate for our over-dispersed (variance<br />
greater than the mean) Poisson-distributed response variable, number of manatees (Agresti<br />
1996; Quinn and Keough 2002). Our predictor variables were: environmental conditions<br />
(sun glare - yes/no, precipitation -dry/light rain, cloud cover -clear/scattered clouds/partly<br />
cloudy/overcast, Beaufort sea state -0/1/2/3, swell height -in 0.15m increments; water<br />
clarity and time of day), and manatee behavior (disturbed/feeding/resting/social/travel/<br />
undetermined - Table 1). Water clarity was a measure of the horizontal distance between<br />
an underwater observer and Secchi disk held 0.5 meters under the water. Water clarity was<br />
measured in this fashion because the vertical Secchi depth was usually greater than or equal<br />
7
134 <br />
135 <br />
136 <br />
137 <br />
138 <br />
139 <br />
140 <br />
141 <br />
142 <br />
143 <br />
144 <br />
145 <br />
146 <br />
147 <br />
148 <br />
149 <br />
150 <br />
151 <br />
152 <br />
153 <br />
154 <br />
to the depth. We did not take seasonality into account because previous analyses indicated<br />
that there is no difference in seasonal sighting probability within the Cayes (LaCommare et<br />
al. 2005; Self-Sullivan 2008).<br />
To test whether detection probability was influenced by number and experience of<br />
volunteers we used a generalized linear model (GLM) with a negative binomial distribution<br />
and a log-link function to determine whether the number of manatees sighted was<br />
influenced by number of volunteers and year (Table 2).<br />
2.4 Trends of manatee counts in the Drowned Cayes<br />
Forty-two survey periods were conducted over 79 months from 2001-2007. We used a<br />
GLM with negative binomial distribution and a log-link function to determine whether<br />
there was a change in the number of manatees per 20-minute scan from the first survey<br />
period in 2001 to the last survey period in 2007.<br />
2.5 Power analysis of trend data<br />
We used simulations, following Taylor and Gerrodette (1993) and Gibbs (1998, 2000), to<br />
determine the statistical power of our ability to detect a trend. We fit a negative binomial<br />
distribution to our manatee count data from our 20-minute scans for the whole 2001 - 2007<br />
period. Using the parameters from this distribution, we generated a random array of the<br />
number of manatees for 28 points for each of 6 survey periods per year. The potential<br />
number of manatees sighted was truncated at 5 since we only had 1 sighting with more than<br />
5 manatees in 7 years. We chose this sampling regime because this most closely matched<br />
our annual survey structure from 2001 <strong>–</strong> 2007. We repeated this process for 15 “years”.<br />
8
155 <br />
156 <br />
157 <br />
158 <br />
159 <br />
160 <br />
161 <br />
162 <br />
163 <br />
164 <br />
165 <br />
166 <br />
167 <br />
168 <br />
169 <br />
<strong>17</strong>0 <br />
<strong>17</strong>1 <br />
<strong>17</strong>2 <br />
<strong>17</strong>3 <br />
<strong>17</strong>4 <br />
<strong>17</strong>5 <br />
Then we examined whether we could detect a decline in the number of manatees over that<br />
15 year period. We generated our random array, added a linear trend, and ran our GLM<br />
1000 times to determine our ability to detect each of three declines in the number of<br />
manatees at the end of these 15 years: slight (10%), moderate (25%) and precipitous (50%).<br />
These declines are equivalent to a 0.7%, 1.7% and 3.6% annual decrease in the number of<br />
manatees sighted per 20-minute scan. We considered that a negative trend was detected if<br />
the slope of the model in each simulation run was negative and significantly different from<br />
zero. Following Taylor et al. (2007), we chose a 15 year period for our power analysis.<br />
2.6 Power analysis of sampling design<br />
Survey structure affects variability in detection probability and therefore statistical power<br />
(Gibbs 1998, 2000). Number of plots, sampling frequency, sampling interval and scan<br />
duration interact to influence statistical power and need to be considered when designing a<br />
monitoring protocol. These factors also influence the cost and time of monitoring.<br />
Therefore, determining the most efficient experimental design can be critical to carrying<br />
out an effective monitoring program. In the field, we sampled approximately 28 of our 54<br />
points during two-week survey periods, 4-8 times per year. Each point was scanned for a<br />
minimum of 20 minutes. A fraction of the point scans lasted 30 minutes, but this decreased<br />
the number of points that could be sampled in any one day and during a two-week survey<br />
period. How does a 30-minute scan duration with fewer samples impact our ability to<br />
detect a trend? Using data that we collected from our 30-minute scans, we generated new<br />
negative binomial distribution parameters and a new random array of the number of<br />
9
<strong>17</strong>6 <br />
<strong>17</strong>7 <br />
<strong>17</strong>8 <br />
<strong>17</strong>9 <br />
180 <br />
181 <br />
182 <br />
183 <br />
184 <br />
185 <br />
186 <br />
187 <br />
188 <br />
189 <br />
190 <br />
191 <br />
192 <br />
193 <br />
194 <br />
195 <br />
196 <br />
manatees for 21 sampling points per two-week survey period for 6 survey periods per year<br />
and repeated our power analysis as described above.<br />
In the Drowned Cayes, 28 out of 54 point scan locations have a greater than 30%<br />
probability of sighting a manatee (LaCommare et. al, 2008). Would restricting our survey<br />
protocol to locations with a higher sighting probability yield a greater ability to detect a<br />
trend? We generated new negative binomial distribution parameters using our manatee<br />
count data from 2001 - 2007 from locations with a greater than 30% probability of sighting<br />
a manatee (LaCommare et. al 2008). Using these parameters, we generated a new random<br />
array of the number of manatees for 28 sampling points for 6 two-week survey periods.<br />
We repeated our power analysis as described above. From that distribution, we also created<br />
an array of the number of manatees for 8 two-week survey periods with both 28 and 21<br />
points and for 8 two-week survey periods, 28 points with just a 20-minute scan duration.<br />
3. Results<br />
3.1 Index and encounter probability<br />
Histograms of time-to-first sighting indicate that for locations that had a greater than 30%<br />
probability of sighting a manatee, 90% of all manatees are sighted within the first 20<br />
minutes of a 30-minute scan (Figure 2). For locations that had a less than 30% probability<br />
of sighting a manatee, it took slightly longer to spot the first manatee and 90% of all<br />
manatees were sighted within 24 rather than 20 minutes (Figure 2).<br />
The number of manatees sighted per scan was not affected by sighting conditions<br />
(overall Likelihood Ratio Chi-Square = 12.578, df = 18, p = 0.816, n= 131, Table 1). None<br />
10
197 <br />
198 <br />
199 <br />
200 <br />
201 <br />
202 <br />
203 <br />
204 <br />
205 <br />
206 <br />
207 <br />
208 <br />
209 <br />
210 <br />
211 <br />
212 <br />
213 <br />
214 <br />
215 <br />
216 <br />
2<strong>17</strong> <br />
of the individual factors or covariates had a significant influence on the number of<br />
manatees sighted per scan (Table 1).<br />
There was a negative effect between number of volunteers and number of manatees<br />
spotted (Wald Chi-Square = 7.343 df =1, p < 0.007, Table 2). The greater number of<br />
volunteers present on the boat, the fewer number of manatees spotted.<br />
3.2 Trends of manatee counts in the Drowned Cayes<br />
We conducted 960 20-minute scans during 42 survey periods from 2001 - 2007. The<br />
number of manatees sighted per scan ranged from 0 to 6. Overall mean number of<br />
manatees sighted per scan was 0.54 (0.89 SD) (Figure 3). Mean number of manatees per<br />
survey period ranged from 0.21 to 1.08 manatees per scan. Yearly means per scans ranged<br />
from 0.33 manatees in 2002 to 0.71 manatees in 2001 (Figure 4). There was no detectable<br />
change in the mean number of manatees sighted per scan from the first survey period in<br />
2001 to the last survey period in 2007 (Likelihood Ratio Chi-Square = 1.56, df = 1, p =<br />
0.212, B = -0.003, Figure 4).<br />
3.3 Power Analysis<br />
Using a 20-minute scan duration, 28 points per survey period, and 6 periods, we had 4%<br />
power to detect a 10% decline in number of manatees over 15 years within 7 years <strong>–</strong> our<br />
study duration, 9% power to detect a 25% decline and 22% power to detect a 50% decline.<br />
Extending this study duration to 15 years would give us a 11% power to detect a 10%<br />
decline, 51% power to detect a 25% decline and 100% power to detect a 50% decline<br />
(Table 3; Figure 5a, b, c).<br />
11
218 <br />
219 <br />
220 <br />
221 <br />
222 <br />
223 <br />
224 <br />
225 <br />
226 <br />
227 <br />
228 <br />
229 <br />
230 <br />
231 <br />
232 <br />
233 <br />
234 <br />
235 <br />
236 <br />
237 <br />
238 <br />
3.4 Power Analysis of sampling design<br />
The highest power, i.e. 16%, 71% and 100% after 15 years, is achieved for all three decline<br />
rates when 28 points with a high (>30%) probability of sighting a manatee are surveyed 8<br />
times per year (Table 3). This survey design has a 90% power to detect a 50% decline by<br />
the 11 th year (Figure 5c). The design with the least power, i.e. 8%, 34% and 99% after 15<br />
years, surveys just 21 points, regardless of the probability of sighting a manatee, 6 times a<br />
year (Table 3). Overall, survey protocol changes had little impact on our ability to detect<br />
slight declines over 15 years, but did impact our ability to detect moderate and precipitous<br />
declines (Figure 5a, b, c) over that time horizon. Increasing the number of scan points from<br />
21 to 28 had about the same effect as increasing the scan duration from 20 to 30 minutes<br />
(about a 10% increase in power to detect moderate declines). Increasing the number of<br />
survey periods from 6 to 8 resulted in an increase in power from 57% to 71% in the<br />
moderate decline scenario (Table 3). Differences in power across designs are not observed<br />
before 10 years of survey in the case of slight declines, 8 years for moderate declines, and 6<br />
years for precipitous declines (Figure 5a, b, c). All designs show the same basic result:<br />
only precipitous declines can be detected with a 100% certainty on a 15 year time horizon<br />
(Figure 5a, b, c).<br />
4. Discussion<br />
Using 30-minute point scans from a small boat platform is an effective and repeatable<br />
method for monitoring manatees in typical manatee habitats around the world. Several<br />
authors debate the validity of population indices in comparison to population estimates - the<br />
12
239 <br />
240 <br />
241 <br />
242 <br />
243 <br />
244 <br />
245 <br />
246 <br />
247 <br />
248 <br />
249 <br />
250 <br />
251 <br />
252 <br />
253 <br />
254 <br />
255 <br />
256 <br />
257 <br />
258 <br />
259 <br />
260 <br />
261 <br />
former characterized as estimating relative abundance, the latter absolute abundance (see<br />
Thompson 2004; Williams and Thomas 2009). These authors caution against the use of<br />
indices in favor of more robust estimating procedures such as distance sampling or mark-<br />
recapture procedures (e.g. Anderson 2001; Williams and Thomas 2009). Yet, as long as the<br />
population index is linearly related to actual abundance <strong>–</strong> that is it doesn’t have an<br />
asymptote or bottom threshold, and detection probabilities are constant over space and<br />
time, count-based indices are valid (Skalski et al. 1983; Thompson 2004; Williams and<br />
Thomas 2009).<br />
Species behavior <strong>–</strong> long dive times, short surfacing bouts - and severely limited<br />
resources hamper the ability to conduct robust monitoring procedures on Antillean<br />
manatees and other sirenians in developing nations (Aragones et al. in press; Aragones et<br />
al. 1997; Dawson et al. 2008; Dick and Hines 2011). Distance sampling requires the ability<br />
to consistently measure the distance from the observer to the animal using a range finder,<br />
reticulated binoculars or spotting scope (Buckland et al. 2001). <strong>Sirenian</strong>s often surface<br />
very briefly and only expose their nose to the surface when breathing. In our study, even<br />
the best trained observers captured distances via range finder only 10% of the time<br />
(LaCommare and Self-Sullivan, personal observation). Mark-recapture methods are<br />
equally challenging. It is estimated that 50% of manatees in the study area are scarred by<br />
boats (Self-Sullivan 2008). Although possible (Self-Sullivan 2008; Auil and Powell,<br />
personal communication), above-water photo identification methods are difficult because<br />
manatees in Belize and other tropical habitats do not seasonally aggregate at warm water<br />
sites, nor do they tend to float at the surface, exposing their scarred backs. Therefore,<br />
photo-identification via underwater video capturing techniques is more appropriate.<br />
13
262 <br />
263 <br />
264 <br />
265 <br />
266 <br />
267 <br />
268 <br />
269 <br />
270 <br />
271 <br />
272 <br />
273 <br />
274 <br />
275 <br />
276 <br />
277 <br />
278 <br />
279 <br />
280 <br />
281 <br />
282 <br />
283 <br />
284 <br />
However, this involves considerable time and effort by the observer and does not always<br />
yield individual identification due to lack of natural markings and/or poor visibility in many<br />
habitats (LaCommare and Self-Sullivan, personal observation). Preliminary analysis<br />
indicates that population abundance estimates via mark-recapture techniques might be<br />
possible (Self-Sullivan 2008, Auil & Powell, Personal communication). Distance sampling<br />
and photo-identification requires considerable observer training. Equipment and observer<br />
training may price monitoring programs out of reach of most wildlife and conservation<br />
agencies in developing nations (Aragones et al. in press; Aragones et al. 1997).<br />
Our index does not appear to have a bottom threshold below which manatees are not<br />
detected, but 30-minutes is a better scan duration than 20-minutes. The histograms of time-<br />
to-first-sighting indicate that for locations with a greater than 30% probability of sighting a<br />
manatee, roughly 90% of all manatees are sighted within the first 20 minutes of a 30-<br />
minute scan. But, for locations with a less than 30% probability of sighting a manatee, it<br />
takes longer to spot the first manatee and it takes up 24 minutes before 90% of all manatees<br />
are sighted.<br />
Based on our analysis of possible influences on detection probability, the<br />
relationship between our index and population abundance is constant over conditions and<br />
time because environmental conditions and manatee behavior do not appear to significantly<br />
influence the number of manatees sighted. This is a surprising result. Visibility bias, bias<br />
from the result of observers missing animals due to observer inconsistencies and<br />
environmental conditions, is significant in aerial surveys for manatees (See Packard et al.<br />
1985 and Wright et al. 2002). Sighting conditions such as water clarity, turbidity, sea state<br />
and surface glare have all been shown to affect visibility (i.e. Reynolds and Wilcox 1994)<br />
14
285 <br />
286 <br />
287 <br />
288 <br />
289 <br />
290 <br />
291 <br />
292 <br />
293 <br />
294 <br />
295 <br />
296 <br />
297 <br />
298 <br />
299 <br />
300 <br />
301 <br />
302 <br />
303 <br />
304 <br />
305 <br />
306 <br />
307 <br />
in aerial surveys. There are two principal reasons why these factors may not affect<br />
visibility of manatees from boats. From a boat, observers search for manatees across the<br />
horizontal plane of the water looking for signs of the manatees’ body (a nose, tail or back)<br />
breaking through the surface of the water. Water clarity isn’t going to affect this to the<br />
same extent as aerial surveys in which observers are searching for manatees within the<br />
water column. In extremely turbid water, like that found in “black water” lake and river<br />
systems of the Amazon, lower vantage points allow better visibility of an animal breaking<br />
the surface of the water than do higher vantage points (Aragones et al. in press). Sea state<br />
may not impact visibility because when surface choppiness is slight <strong>–</strong> sea state of zero or<br />
one <strong>–</strong> any disturbance to the surface of the water is noticeable and therefore manatees will<br />
be spotted. Because manatees need to rise their nose above the water to breathe, when<br />
there is a sea state of two or three <strong>–</strong> choppy conditions <strong>–</strong> manatees will raise their entire<br />
head out of the water (Personal observation, KSL and CSS) which facilitates the observer’s<br />
ability to spot manatees and negates the impact of sea state on visibility. Surveys were not<br />
conducted in a sea state higher than three. It is not clear why surface glare does not impact<br />
visibility. There are three possible explanations. Surveys are conducted between 9:30 am<br />
and 4:30 pm when surface glare, due to a sun low on the horizon, is at a minimum.<br />
Multiple trained observers searching for manatees may reduce the negative effects of sun<br />
glare. And, manatees usually surface multiple times during a 30-minute search duration<br />
which increases the possibility of spotting a manatee even during difficult observation<br />
conditions.<br />
The biggest concern with our study was that number of volunteers had a negative<br />
effect on our ability to spot manatees. The greater the number of volunteers the fewer<br />
15
308 <br />
309 <br />
310 <br />
311 <br />
312 <br />
313 <br />
314 <br />
315 <br />
316 <br />
3<strong>17</strong> <br />
318 <br />
319 <br />
320 <br />
321 <br />
322 <br />
323 <br />
324 <br />
325 <br />
326 <br />
327 <br />
328 <br />
329 <br />
330 <br />
number of manatees spotted. Volunteers are easily distracted by each other during long<br />
scan periods and survey days and do not have the same level of focus as trained observers.<br />
This can easily be minimized in the future by limiting the number of volunteers present<br />
during surveys.<br />
We do not know the relationship of our index to actual abundance. It is an index of<br />
relative abundance that can be used to detect trends. Index validation, by comparing the<br />
number of manatees counted via boat surveys to the number counted during concurrent,<br />
localized aerial surveys, could provide an indication of the relationship between boat counts<br />
and aerial survey counts which could help determine bias and thus the relationship of our<br />
index to actual abundance. Aerial surveys were conducted along the entire Belizean<br />
coastline, including the Drowned Cayes, from 1997 <strong>–</strong> 2002 (Auil, 2004). These surveys<br />
overlapped with our boat surveys in 2000 - 2002, but were only conducted once during the<br />
wet and once during the dry season and did not directly correspond to our survey effort and<br />
unfortunately, are not useful for evaluating bias of our method. It is interesting to note that<br />
the trends of Auil’s (2004) index of relative abundance from 2000-2002 mirrors our trends<br />
across those same years.<br />
There was a very slight negative trend in the data, but this trend was not significant.<br />
Our analysis indicates that we would have only had a 4%, 9% and 22% power to detect<br />
slight, moderate or precipitous declines after 7 years <strong>–</strong> our study duration. Even slight<br />
negative trends can be important for small populations which can be impacted by stochastic<br />
events or behavioral biology that can drive small populations to extinction (Meffe and<br />
Carroll 1997) or prevent these populations from recovering (Meffe and Carroll 1997). The<br />
Antillean subspecies of the West Indian Manatee is currently listed as endangered on the<br />
16
331 <br />
332 <br />
333 <br />
334 <br />
335 <br />
336 <br />
337 <br />
338 <br />
339 <br />
340 <br />
341 <br />
342 <br />
343 <br />
344 <br />
345 <br />
346 <br />
347 <br />
348 <br />
349 <br />
350 <br />
351 <br />
352 <br />
353 <br />
IUCN Red List because it is a small and declining population without ongoing, effective<br />
conservation actions (Self Sullivan et al. 2011). These data provide confirmation of the<br />
appropriateness of the current IUCN status and highlight the need to improve our ability to<br />
establish the magnitude of decline.<br />
Extending our study to 15 years considerably improves our ability to detect a<br />
population trend. Over a study duration of 15 years, we would still have a limited ability to<br />
detect slight and moderate declines, but our index and protocol can detect precipitous<br />
declines within the benchmarks established in the Marine Mammal Protection Act, the<br />
IUCN Red List (Taylor et al 2007) and other publications. Taylor et al. (2007) defined a<br />
precipitous decline in marine mammal abundance as a 50 percent decline after 15 years<br />
because a 50% decline over this period would result in a stock being classified as<br />
“depleted” under the MMPA. Under the IUCN Red List guidelines, a decline of this<br />
magnitude could result in a status designation anywhere between “vulnerable” to<br />
“endangered” depending on initial population size, cause, certainty and the reversibility of<br />
the decline (IUCN 2010). IUCN Red List guidelines establish level of endangerment <strong>–</strong><br />
nearly threatened to critically endangered <strong>–</strong> based on declines of 10 <strong>–</strong> 90% over 10 years or<br />
3 generations (whichever is longer), depending on initial population size, cause, certainty<br />
and the reversibility of the decline (IUCN 2010). Bart et al. (2004) recommended a<br />
standard for landbird surveys of an 80% power to detect a 50% decline over 20 years.<br />
Similarly, Hatch (2003) recommended a standard for counts of colonial seabirds of a 90%<br />
power to detect a 50% decline over 10 years. Using this sampling regime over a 15 year<br />
time interval, we would have a power of 11%, 51% and 100% to detect slight, moderate or<br />
precipitous declines. All of our survey protocols had 100% power to detect 50% declines<br />
<strong>17</strong>
354 <br />
355 <br />
356 <br />
357 <br />
358 <br />
359 <br />
360 <br />
361 <br />
362 <br />
363 <br />
364 <br />
365 <br />
366 <br />
367 <br />
368 <br />
369 <br />
370 <br />
371 <br />
372 <br />
373 <br />
374 <br />
375 <br />
over 15 years making it a valuable tool for determining population status based on the<br />
benchmarks discussed above. Taylor et al. (2007) found that most marine mammal<br />
monitoring programs had a low power (0-50%) to detect precipitous declines in abundance<br />
within this time frame.<br />
We can improve the power of our survey protocol by surveying at least 28 points, 8<br />
times per year, increasing our scan duration to 30-minutes and choosing locations with a<br />
30% sighting probability. Under this regime, we have 100% power to detect a 50% decline<br />
over a 15 year period. We have a power of 90% to detect a 50% decline after 10 years <strong>–</strong><br />
exceeding the MMPA benchmark. The power of this sampling regime also meets the<br />
IUCN guidelines for establishing whether a population should be listed as critically<br />
endangered, endangered or vulnerable (IUCN, 2010). We would still have a limited ability<br />
to detect slight (16% power) declines, but a reasonable ability (71% power) to detect<br />
moderate declines. The IUCN benchmarks for establishing the vulnerability status of small<br />
populations for whom the causes and reversibility of declines are uncertain range from 10 <strong>–</strong><br />
30 % declines over 3 generations (60 years for manatees, (Deutsch et al. 2008)).<br />
To ensure the greatest success in a monitoring program, pilot studies are highly<br />
recommended (Aragones et al. in press; Buckland et al. 2001; Gibbs 2000; Gibbs et al.<br />
1998; Scheiner and Gurevitch 2001; Thompson 2004). Local knowledge can be utilized in<br />
pilot studies to locate high use areas in a short period of time <strong>–</strong> e.g. 1 year (Self-Sullivan<br />
and LaCommare, personal observation). Including local knowledge in research programs<br />
also has the effect of improving success in corollary conservation programs - e.g. poaching<br />
reduction programs (Aragones et al. in press; Kendall 2009). In addition to being a<br />
18
376 <br />
377 <br />
378 <br />
379 <br />
380 <br />
381 <br />
382 <br />
383 <br />
384 <br />
385 <br />
386 <br />
387 <br />
388 <br />
389 <br />
390 <br />
391 <br />
392 <br />
393 <br />
394 <br />
395 <br />
396 <br />
397 <br />
398 <br />
399 <br />
400 <br />
401 <br />
practical and repeatable way to examine population trends, boat surveys can be used to<br />
assess spatial distribution and habitat use (LaCommare et al. 2008).<br />
4. Conclusion<br />
Using a small-boat platform for systematic surveys is a sound and repeatable<br />
method for monitoring sirenian populations in developing nations where constraints on<br />
fiscal resources are severe. We recommend pilot studies and using local knowledge to<br />
locate high use areas and then using a 30-minute scan duration to conduct point scans. Our<br />
analysis indicates that surveys should be conducted at least 8 times per year and at least 21<br />
points should be included in the sampling protocol. For our study area, we estimate this<br />
monitoring protocol would require approximately 37 dedicated days on the water per year<br />
(4-5 scans per day, 3 days per month) with the only expenses being boat, fuel, and three<br />
well-trained observers. Alternatively, this protocol would be well suited for the eco-<br />
tourism industry where well trained local tour guide/researchers could conduct one or two<br />
scans during each tour, which are randomly distributed among 21 sites with a >30%<br />
sighting probability.<br />
Acknowledgements. We thank Earthwatch Institute for substantial financial support and<br />
Earthwatch Volunteers for logistical support and data collection. We thank Conservation<br />
Action Fund for financial support that specifically supported capacity building among local<br />
field assistants and tour guides. We are ever grateful to Mr. Sidney Turton and Spanish<br />
Bay Resort, who provided logistical and in kind support from 2001-2004, and to Ms.<br />
Teresa Parkey and The Hugh Parkey Foundation for Marine Awareness and Education,<br />
who provided logistical and in kind support in 2005 - 2007. Supplemental financial support<br />
was provided by two individual Earthwatch volunteers (Mr. Greenwalt, amd Mr Burtt),<br />
Project Aware, Virtual Explorers and the Lerner-Gray Marine Research Fund (AMNH).<br />
We are grateful to the numerous interns and field assistants who were vital to data<br />
collection and field logistics. One author (CSS) was also supported by an NSF Graduate<br />
19
402 <br />
403 <br />
404 <br />
405 <br />
406 <br />
Fellowship from 2001-2003. We are especially thankful for the help and guidance of our<br />
colleague, friend, primary field assistant and source of traditional knowledge, Mr. Gilroy<br />
Robinson. Finally, we thank two anonymous reviewers whose comments greatly improved<br />
this paper.<br />
20
406 <br />
407 <br />
408 <br />
409 <br />
410 <br />
411 <br />
412 <br />
413 <br />
414 <br />
415 <br />
416 <br />
4<strong>17</strong> <br />
418 <br />
419 <br />
420 <br />
421 <br />
422 <br />
423 <br />
424 <br />
425 <br />
426 <br />
427 <br />
428 <br />
429 <br />
430 <br />
431 <br />
432 <br />
433 <br />
434 <br />
435 <br />
6. Literature Cited<br />
Agresti, A., 1996. An Introduction to Categorical Data Analysis. Wiley, New York.<br />
Anderson, D.R., 2001. The need to get the basics right in wildlife field studies. Wildlife<br />
Society Bulletin 29, 1294-1297.<br />
Aragones, L., LaCommare, K.S., Kendall, S., Castelblanco-Martinez, D.N., Gonzalez-<br />
Socoloske, D., in press. Boat and land-based surveys for sirenians, In Hines, E., Reynolds,<br />
J., Mignucci-Giannoni, A, Aragones, L., and M. Marmontel (eds). <strong>Sirenian</strong> Conservation:<br />
Issues and Strategies in Developing Countries. University Press of Florida.<br />
Aragones, L.V., Jefferson, T.A., Marsh, H., 1997. Marine mammal survey techniques in<br />
developing countries. Asian Marine Biology 14, 15-39.<br />
Auil, N., 1998. Belize Manatee Recovery Plan, In UNDP/GEF Coastal Zone Management<br />
Project No. BZE/92/G31. United Nations Environment Program, Caribbean Environment<br />
Program, Kingston, Jamaica.<br />
Auil, N.E., 2004. Abundance and Distribution Trends of the West Indian Manatee in the<br />
Coastal Zone of Belize: Implication for Conservation, In Department of Fisheries and<br />
Wildlife. p. 83. Texas A&M University, College Station, TX.<br />
Bart, J., Burnham, K.P., Dunn, E.H., Francis, C.M., Ralph, C.J., 2004. Goals and strategies<br />
for estimating trends in landbird abundance. The Journal of Wildlife Management 68, 611-<br />
626.<br />
Bengston, J.L., Magor, D., 1979. A Survey of manatees in Belize. Journal of Mammalogy<br />
60, 230-232.<br />
Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., Borchers, D.L., Thomas, L.,<br />
2001. Introduction to Distance Sampling: Estimating abundance of biological populations.<br />
Oxford University Press, New York.<br />
Caughley, G., 1977. Analysis of Vertebrate Populations. Wiley-Interscience Publication.<br />
CEP, 1995. Regional Management Plan for the West Indian Manatee (Trichechus manatus<br />
manatus), In Technical Report No. 35. CEP UNEP, Kingston, Jamaica.<br />
CEP, 2010. Protocol Concerning Specially Protected Areas and Wildlife to the Convention<br />
for the Protection and Development of the Marine Environment of the Wider Caribbean<br />
21
436 <br />
437 <br />
438 <br />
439 <br />
440 <br />
441 <br />
442 <br />
443 <br />
444 <br />
445 <br />
446 <br />
447 <br />
448 <br />
449 <br />
450 <br />
451 <br />
452 <br />
453 <br />
454 <br />
455 <br />
456 <br />
457 <br />
458 <br />
459 <br />
460 <br />
461 <br />
462 <br />
463 <br />
Region. http://www.cep.unep.org/cartagena-convention/spaw-protocol/spaw-protocolen.pdf.<br />
Last Accessed: April 15, 2010<br />
CITES, 2010. Convention on <strong>International</strong> Trade in Endangered Species of Wild Fauna and<br />
Flora. http://www.cites.org/eng/disc/species.shtml. Last Accessed: April 15, 2010<br />
CMS, 2010. The Memorandum of Understanding on the conservation and management of<br />
dugongs (Dugong dugon) and their habitats throughout their range, ed. C.o.M. Species.<br />
http://www.cms.int/species/dugong/dugong_mou.htm. Last Accessed: April 15, 2010<br />
Dawson, S., Slooten, E., DuFresne, S., Wade, P., Clement, D., 2004. Small-boat surveys for<br />
coastal dolphins: line-transect surveys for Hector's dolphins (Cephalorhynchus hectori).<br />
Fishery Bulletin 102, 441-451.<br />
Dawson, S., Wade, P., Slooten, E., Barlow, J., 2008. Design and field methods for sighting<br />
surveys of cetaceans in coastal and riverine habitats. Mammal Review 38, 19-49.<br />
Deutsch, C.J., Self-Sullivan, C., Mignucci-Giannoni, A., 2008. Trichechus manatus<br />
manatus, In IUCN Red List of Threatened Species. IUCN 2010 Version 2010.1.<br />
. Last Accessed: 15 April 2010.<br />
Dick, D.M. and E.M. Hines. 2011. Development and implementation of distance sampling <br />
techniques to determine bottlenose dolphin (Tursiops truncatus) abundance at Turneffe <br />
Atoll, Belize. Marine Mammal Science: 27(3): 606‐621. <br />
Gerrodette, T., 1987. A power analysis for detecting trends. Ecology 68, 1364-1372.<br />
Gibbs, J.P., 2000. Monitoring Populations, In: L. Boitani and T. Fuller (eds) Research<br />
Techniques in Animal Ecology: Controversies and Consequences. pp. 212-252. Columbia<br />
University Press, New York.<br />
Gibbs, J.P., Droege, S., Eagle, P., 1998. Monitoring populations of plants and animals.<br />
BioScience 48, 935-940.<br />
Hatch, S.A., 2003. Statistical power for detecting trends with applications to seabird<br />
monitoring. Biological Conservation 111, 3<strong>17</strong>-329.<br />
Hines, E.M., Adulyanuskol, K., Duffus, D.A., 2005. Dugong abundance along the Adaman<br />
coast of Thailand. Marine Mammal Science 21, 536-549.<br />
22
464 <br />
465 <br />
466 <br />
467 <br />
468 <br />
469 <br />
470 <br />
471 <br />
472 <br />
473 <br />
474 <br />
475 <br />
476 <br />
477 <br />
478 <br />
479 <br />
480 <br />
481 <br />
482 <br />
483 <br />
484 <br />
485 <br />
486 <br />
487 <br />
488 <br />
489 <br />
490 <br />
491 <br />
492 <br />
493 <br />
Hunter, M.E., Auil-Gomez, N.E., Tucker, K.P., Bonde, R.K., Powell, J., McGuire, P.M.,<br />
2010. Low genetic variation and evidence of limited dispersal in the regionally important<br />
Belize manatee. Animal Conservation 13 592<strong>–</strong>602.<br />
Kendall, S., 2009. Making Science Matter, In <strong>International</strong> Marine Conservation Congress<br />
Annual Convention. Washington, DC.<br />
Kindberga, J., Ericssona, G., Swensonb, J.E., 2009. Monitoring rare or elusive large<br />
mammals using effort-corrected voluntary observers. Biological Conservation 142, 159-<br />
165.<br />
LaCommare, K.S., Self-Sullivan, C., Brault, S., 2005. Distribution, Seasonal Occurrence<br />
and Habitat Use of Antillean Manatees in the Drowned Cayes Area of Belize, Central<br />
America, In 16th Biennial Conference on the Biology of Marine Mammals. San Diego, CA.<br />
LaCommare, K.S., Self-Sullivan, C., Brault, S., 2008. Distribution and Habitat Use of<br />
Antillean Manatees (Trichechus manatus manatus) in the Drowned Cays Area of Belize,<br />
Central America. Aquatic Mammals 34, 35-43.<br />
Lefebvre, L.W., Marmontel, M., Reid, J.P., Rathbun, G.B., Domning, D.P., 2001. Status<br />
and biogeography of the West Indian manatee, In: C.A. Woods and F.E. Sergile (eds)<br />
Biogeography of the West Indies: Patterns and Perspectives. pp. 425-474. CRC Press,<br />
Boca Raton.<br />
MacKenzie, D.I., Royle, J.A., 2005. Designing occupancy studies: general advice and<br />
allocating survey effort. Journal of Applied Ecology 42, 1105-1114.<br />
Marsh, D.M., Trenham, P.C., 2008. Current trends in plant and animal population<br />
monitoring. Conservation Biology 22, 647.<br />
Marsh, H., Lefebvre, L.W., 1994. <strong>Sirenian</strong> status and conservation efforts. Aquatic<br />
Mammals 20, 155-155.<br />
Martin, J., Kitchens, W., Hines, J., 2006. Importance of well-designed monitoring programs<br />
for the conservation of endangered species: case study of the snail kite conservation<br />
Biology 21, 472-481.<br />
Meffe, G. K, Carroll, C. R. 1997. Principles of Conservation Biology. 2 nd Ed. Sinauer and<br />
Associates, Sunderland, MA.<br />
23
494 <br />
495 <br />
496 <br />
497 <br />
498 <br />
499 <br />
500 <br />
501 <br />
502 <br />
503 <br />
504 <br />
505 <br />
506 <br />
507 <br />
508 <br />
509 <br />
510 <br />
511 <br />
512 <br />
513 <br />
514 <br />
515 <br />
516 <br />
5<strong>17</strong> <br />
518 <br />
519 <br />
520 <br />
521 <br />
522 <br />
523 <br />
Moraes-Arraut, E., Ortega-Argueta, A., Olivera-Gómez, D. L. in press. Delineating and<br />
Assessing Habitats for <strong>Sirenian</strong>s In Hines, E., Reynolds, J., Mignucci-Giannoni, A,<br />
Aragones, L., and M. Marmontel (eds). <strong>Sirenian</strong> Conservation: Issues and Strategies in<br />
Developing Countries. University Press of Florida.<br />
Morales-Vela, B., Olivera-Gomez, D., III, J.E.R., Rathbun, G.B., 2000. Distribution and<br />
habitat use by manatees (Trichechus manatus manatus) in Belize and Chetumal Bay,<br />
Mexico. Biological Conservation 95, 67-75.<br />
O'Shea, T.J., Salisbury, C.A.L., 1991. Belize - a last stronghold for manatees in the<br />
Caribbean. Oryx 25, 156-164.<br />
Packard, J. M., Summers R. C., Barnes, L. B. (1985) Variation in visibility bias during<br />
aerial surveys. Journal of Wildlife Management 49: 347-351.<br />
Quinn, G.P., Keough, M.J., 2002. Experimental Design and Data Analysis for Biologists.<br />
Cambridge University Press, Cambridge.<br />
Quintana-Rizzo, E., Reynolds, J.E., 2008. Regional management plan for the West Indian<br />
Manatee (Trichechus manatus), In Caribbean Environment Programme. United Nations<br />
Environment Program, Gosier, Guadeloupe, France.<br />
Reynolds, J.E. III, Wilcox, J. R. 1994. Observations of Florida manatees (Trichechus<br />
manatus latirostris) around selected power plants in winter. Marine Mammal Science<br />
10:163-<strong>17</strong>7.<br />
Scheiner, S.M., Gurevitch, J. eds., 2001. Design and Analysis of Ecological Experiments,<br />
2nd ed. Oxford University Press, Oxford.<br />
Self-Sullivan, C., 2008. Conservation of Antillean Manatees in the Drowned Cays Area of<br />
Belize, In Fisheries and Wildlife. p. 143. Texas A&M.<br />
Self-Sullivan, C. & Mignucci-Giannoni, A. 2008. Trichechus manatus ssp. manatus. In:<br />
IUCN 2011. IUCN Red List of Threatened Species. Version 2011.1.<br />
. Downloaded on 09 November 2011.<br />
Skalski, J.R., Robson, D.S., Simmons, M.A., 1983. Comparative census procedures using<br />
single mark-recapture methods. Ecology 64, 752-760.<br />
Taylor, B.L., Gerrodette, T., 1993. The uses of statistical power in conservation biology:<br />
the vaquita and northern spotted owl. Conservation Biology 7, 489-500.<br />
24
524 <br />
525 <br />
526 <br />
527 <br />
528 <br />
529 <br />
530 <br />
531 <br />
532 <br />
533 <br />
534 <br />
535 <br />
536 <br />
537 <br />
538 <br />
539 <br />
Taylor, B.L., Martinez, M., Gerrodette, T., Barlow, J., Hrovat, Y.N., 2007. Lessons from<br />
monitoring trends in abundance of marine mammals. Marine Mammal Science 23, 157-<br />
<strong>17</strong>5.<br />
Thompson, W.L. ed., 2004. Sampling rare or elusive species. Island Press, Washington.<br />
USFWS, 2001. Florida Manatee Recovery Plan (Trichechus manatus latirostris), Thrid<br />
Revision, ed. U.S.F.W. Service, p. 138 pp. U.S. Fish and Wildlife Service.<br />
Williams, R., Thomas, L., 2009. Cost-effective abundance estimation of rare animals:<br />
Testing performance of small-boat surveys for killer whales in British Columbia. Biological<br />
Conservation 142, 1542-1547.<br />
Wright, I.E., Reynolds J.E. III, Ackerman B. B., Ward, L.I., Weigle, B. L., Szelistowski,<br />
W. A. 2002. Trends in manatee (Trichechus manatus latirostris) counts and habitat use in<br />
Tampa Bay, 1987-1994: implications for conservation. Marine Mammal Science 18:259-<br />
274.<br />
Yoccoz, N.G., Nichols, J.D., Boulinier, T., 2001. Monitoring of biological diversity in<br />
space and time. Trends in Ecology & Evolution 16, 446-453.<br />
<br />
25
Table 1. Results from the GLM relating number of manatees to variables that may influence detection probability (n = 113).<br />
Dependent variable is number of manatees. Independent variables are number of volunteers, total number of observers,<br />
precipitation, cloud cover, sea state, swell height and behavior.<br />
Variables in the Model<br />
df Significance<br />
Wald Chisquare<br />
n = 131<br />
Precipitation (subcategories: dry/light rain)* 0.426 1 NS<br />
Cloud cover (subcategories: clear/scattered/partly cloudy/overcast)* 0.267 3 NS<br />
<br />
<br />
<br />
<br />
<br />
Dependent<br />
variable<br />
Number of manatees<br />
<br />
<br />
Independent variables<br />
Environment<br />
variables<br />
<br />
<br />
<br />
Sea state (subcategories: Beaufort scale <strong>–</strong> 0/1/2/3)* 0.<strong>17</strong>4 3 NS<br />
Glare (subcategores: yes/no)* 0.002 1 NS<br />
Swell Height 0.001 1 NS<br />
Secchi Distance 0.481 1 NS<br />
Sighting Time 0.604 1 NS<br />
Manatee Behavior*<br />
Behavior (subcategories: disturbed, feed, mill, rest, social, travel, other) 2.916 7 NS<br />
*No significant subcategories<br />
<br />
<br />
26
Variables in the Model<br />
df Significance<br />
Wald Chisquare<br />
n = 597<br />
Dependent variable<br />
Number of manatees<br />
Independent variables<br />
Number of volunteers 7.343 1 0.007<br />
Number of volunteers * Year<br />
2001 1.620 7 NS<br />
2002 0.013 1 NS<br />
2003 1.094 1 NS<br />
2004 0.616 1 NS<br />
2005 0.243 1 NS<br />
2006 0.338 1 NS<br />
2007 * * *<br />
Table 2. Generalized linear model (GLM) with a negative binomial distribution and a log-link function to determine<br />
whether the number of manatees sighted was influenced by number of volunteers and year<br />
<br />
<br />
<br />
<br />
27
28
Table 3. Simulation results: Percentage of declines detected (in bold italics) for six different sampling regimens and three different<br />
levels of decline, assuming a perfectly consistent relationship between population index and actual abundance (1000 runs for each<br />
simulation). Scan duration indicates whether the negative binomial distribution parameters were generated from 20 or 30-minute<br />
scans. Number of points indicates the number of points used to create the array of manatee sightings for the simulation models<br />
Parentheses indicates which points were used to generate the negative binomial distribution parameters. > 30% means points with<br />
greater than 30% sighting probability. Number of two-week survey periods indicates the number of survey periods used in the<br />
Sampling Regime<br />
Scan Interval 20-minute scan 30-minute scan<br />
6 8 6 6 8 8<br />
21<br />
(>30 %)<br />
28<br />
(>30 %)<br />
28<br />
(>30 %)<br />
21 points<br />
(all points)<br />
28 points<br />
(>30% )<br />
28 points<br />
(all points)<br />
Num. of two-week survey<br />
periods<br />
Num. of points<br />
(Points used to generate<br />
negative binomial<br />
distribution parameters.)<br />
Results<br />
10% 11 14 9 12 16 13<br />
25% 51 65 33 57 71 60<br />
50% 100 100 99 100 100 100<br />
simulation. We ran our simulation for three potential declines in the number of manatees after 15 years: 10%, 25% and 50%.<br />
These declines are equivalent to a 0.7%, 1.7% and 3.6% decrease in the number of manatees sighted per scan per year.<br />
<br />
<br />
<br />
29
Figure 1. Map of the Drowned Cayes area and scan points<br />
30
a.<br />
b.<br />
c.<br />
Number of sightings<br />
Number of sightings<br />
Number of sightings<br />
100<br />
80<br />
60<br />
40<br />
20<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
25<br />
20<br />
15<br />
10<br />
5<br />
0<br />
0<br />
1 2 5 8 9 14 <strong>17</strong> 20 23 26 29<br />
Figure 3. Histogram of number of manatees counted per 20-minute point scan (mean =<br />
0.54, SD = 0.891, n = 960) with negative binomial distribution (solid line).<br />
32
2001 2002 2003 2004 2005 2006 2007<br />
Figure 4. No significant trend in mean number of manatees sighted per 20-minute scan<br />
was detected from 2001-2007 in the Drowned Cayes area of Belize (Likelihood Ratio Chi-<br />
Square = 1.56, df = 1, B = -0.003, p = 0.212 (NS), n = 960). Error bars: +/- 1 SE. Bottom<br />
axis is number of the survey period. Number of manatees per 20-minute scan was the<br />
dependent variable. Independent variable was the time over which the forty-two survey<br />
periods were conducted.<br />
33
a.<br />
b.<br />
c.<br />
Power<br />
Power<br />
Power<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
20 min 6 teams 28 pts<br />
30 min 6 teams 21 pts<br />
30 min 6 teams 28 pts<br />
20 min 8 teams 28 pts<br />
30 min 8 teams 28 pts<br />
<br />
30 min 6 teams 21 pts<br />
2 4 6 8 10 12 14 15<br />
2 4 6 8 10 12 14 15<br />
2 4 6 8 10 12 14 15<br />
Number of years<br />
34<br />
> 30 %<br />
sighting<br />
probability
Figure 5. Power to detect 10% (a), 25% (b), and 50% (c) declines in the number of<br />
manatees over 15 years.<br />
35