The quantitative study of marked individuals in ecology, evolution ...

EURING Technical Meeting 2003: 

The quantitative study of marked individuals in 

ecology, evolution and conservation biology 

hosted by 

Max Planck Research Centre for Ornithology 

Vogelwarte Radolfzell 

06 – 11 October 2003 in Radolfzell 

scientific committee 

Juan Carlos Senar (chair) 

Mike Conroy, André Dhondt (co-chairs) 

Neil Arnason, Stephen Baillie, Franz Bairlein, 

Charles Brown, Ken Burnham, Eammanuelle Cam, 

Jean Clobert, Mike Conroy, Evan Cooch, 

Paul Doherty, Charles Francis, Jim Hines, Darryl 

MacKenzie, Jean-Dominique Lebreton, Danny Lee, 

Jim Nichols, Ken Pollock, Carl Schwarz, 

David Thomson, Gary White 

supported by: head of local organizing team 

Wolfgang Fiedler

Radolfzell (Germany) 

Contents 

welcome ............................................................................................................. 2 

program ............................................................................................................. 3 

oral sessions 

short courses .......................................................................................... 5 

evolutionary biology & life histories ......................................................... 6 

random effects ..................................................................................... 11 

multi-state models ................................................................................ 13 

standardizing terminology (roundtable) ................................................. 16 

methodological advances ..................................................................... 17 

computing & software ........................................................................... 21 

population dynamics & monitoring applied to decision making ............. 24 

dispersal & migration ............................................................................ 27 

analysis using large-scale ringing data ................................................. 32 

abundance estimation and conservation biology .................................. 36 

population dynamics ............................................................................. 39 

Honor Speaker (James Nichols) ........................................................... 42 

poster abstracts ................................................................................................ 43 

list of participants ............................................................................................. 57 

1

Technical Meeting 2003 

Welcome 

Dear EURING 2003 participants, 

Since many years now EURING (The European Union for Bird Ringing) is active in promoting 

the development of bird ringing as a tool for science and wildlife conservation. Among 

the most important results EURING has obtained, the so called ‘technical meetings’ rank 

surely high. Following the idea put forward by the past President Pertti Saurola, it is now since 

1986 that EURING could attract top scientists to discuss and develop statistical tools for 

the analysis of mark-recapture data. 

These meetings are fairly unique venues, where biostatisticians and ornithologists interact 

and exchange their experiences to set up priorities for the next steps to be undertaken for the 

best use of data derived from ringing and marking of our birds. 

The truly international profile gained by the EURING technical meetings is also confirmed by 

what has now become a tradition of holding them alternatively in Europe and the U.S. These 

meetings have also invariably produced volumes of proceedings of the highest scientific level 

and contents, which is a further reason of their success. And beyond the proceedings, new 

software has been produced also thanks to the analytical opportunities offered by unique 

data sets originating from ringing. 

Presently much of this software is free, allowing scientists from all over the world to benefit 

from these analytical tools produced by famous biostatisticians who deserve all our appreciation 

for their positive availability to support research on marked individuals. 

Increasingly during the years the need for a detailed knowledge of the demographic 

processes affecting bird populations has become an important component of conservation 

policies. The study of marked birds contributes to applied aspects of the management, harvesting 

and conservation of their populations, which belong to the international community 

and represent the subject of international directives. 

The contribution that also the EURING technical meetings have offered to the uderstanding 

of the mechinisms governing bird populations have influenced the design of some of the 

most important ringing projects coordinated by EURING and devoted to bird monitoring 

across Europe, like our Constant Effort Site project. 

This meeting is kindly hosted by the good friends of Vogelwarte Radolfzell, an Istitute which 

has offered so much to the understanding of the adaptive roots of bird migration and to longterm 

monitoring of migratory birds. I’m sure this will be a most interesting conference, and I 

wish all participants a stimulating working week, full of contacts and new ideas. 

Fernando Spina 

EURING Chairman 

2


EURING Technical Meeting 2003 

06 – 11 October 2003 in Radolfzell 

PROGRAM 

All sessions are held in the small hall (“Kleiner Saal”) at the conference centre 

“Tagungs und Kulturzentrum Milchwerk” (TKM). Posters are displayed in the foyer. 

Monday, October 6 

08:45 AM - 05:15 PM short courses (see separate schedule p. 5) 

06:00 PM - 07:00 PM dinner (only for those who registered for the short course) 

Tuesday, October 7 

08:00 AM - 08:15 AM Opening and Welcome 

08:15 AM - 10:10 AM evolutionary biology & life histories 

10:10 AM - 10:25 AM break 

10:25 AM - 11:45 AM evolutionary biology & life histories (continued) 

11:45 AM - 01:15 PM break (lunch) 

01:15 PM - 03:35 PM random effects 

03:35 PM - 03:50 PM break 

03:50 PM - 06:10 PM multi-state models 

06:10 PM - 07:30 PM reak (dinner) 

07:30 PM - 09:30 PM roundtable: standardizing terminology 

Wednesday, October 8 

08:15 AM - 10:10 AM methodological advances 

10:10 AM - 10:30 AM break 

10:30 AM - 11:45 AM methodological advances (continued) 

11:45 AM - 01:15 PM break (lunch) 

01:15 PM - 03:35 PM computing & software 

03:35 PM - 03:50 PM break 

03:50 PM - 06:10 PM population dynamics & monitoring applied to decision 

making 

06:10 PM - 07:30 PM break (dinner) 

07:30 PM - 09:30 PM poster session 

3

Technical Meeting 2003 

Thursday, October 9 

Departure of busses from „Messeplatz“ (100 m from TKM) – see city map 

08:00 AM - 03:30 PM field trip 1: Wetland „Wollmatinger Ried“, Vogelwarte Radolfzell 

07:00 AM - 08:00 PM field trip 2: alps 

Friday, October 10 

08:15 AM - 11:45 AM dispersal & migration 

10:10 AM - 10:25 AM break 

08:15 AM - 11:45 AM dispersal & migration (continued) 

11:45 AM - 01-15 PM break (lunch) 

01:15 PM - 03:35 PM analysis using large-scale ringing data 

03:35 PM - 03:50 PM break 

03:50 PM - 06:10 PM abundance estimation & conservation biology 

06:15 PM - 10:00 PM conference dinner, discussion about next meeting etc. 

Saturday, October 11 

08:15 AM - 10:35 AM population dynamics 

10:35 AM - 11:00 AM break 

11:00 AM - 12:00 AM Honor Speaker (Jim Nichols) 

12:00 AM - 13:30 AM lunch, poster removal, departure 

4


Short courses (organizer: Evan Cooch) 

For the EURING 2003 short-courses, we will focus on 3 topics: (i) parameter counting & redundancy, 

(ii) goodness-of-fit testing, and (ii) Bayesian inference. All three of these topics are 

of considerable interest: the first 2 for very practical reasons, regardless of the approach you 

take in your analyses, and the latter reflecting a growing interest in the use of Bayesian methods 

for complex problems. While many attending EURING 2003 will have some level of 

familiarity with one or more of these topics, the design and intent of the short-courses is to 

ensure some common level of understanding, as much as possible, in advance of the formal 

papers later in the week (many of which make use of, or rely upon, one or more of these 

subjects). 

08:45 AM - 09:00 AM Opening and Welcome Evan Cooch 

09:00 AM - 10:00 AM Goodness of Fit testing Roger Pradel & Darryl Mackenzie 

10:00 AM - 10:15 AM break 

10:15 AM - 11:15 AM Parameter redundancy Olivier Giminez, Anne Viallefont 

& counting 

& Ted Catchpole 

11:15 AM - 11:30 AM Introduction to Bayesian Steve Brooks, Ruth King 

short-course 

& Byron Morgan 

11:30 AM - 12:30 PM Introduction to Bayesian Steve Brooks, Ruth King 

methods 


12:30 PM - 01:30 PM break (lunch) 

01:30 PM - 02:00 PM Introduction to WinBUGS Steve Brooks, Ruth King 


02:30 PM - 03:00 PM Case studies I & II Steve Brooks, Ruth King 


03:00 PM - 03:30 PM break 

03:30 PM - 04:00 PM Case study III Steve Brooks, Ruth King 


04:00 PM - 05:00 PM Introduction to Bayesian Steve Brooks, Ruth King 

model selection 


05:00 PM - 05:15 PM Summary overview Steve Brooks, Ruth King 


06:00 PM - 07:00 PM dinner 

5

Evolutionary biology & life histories 

Evolutionary biology & life histories (chairs: David Thomson & Charles Brown) 

Plenary Address 08:15 AM - 08:55 AM 

Using Mark-Recapture to Study Social Behavior: A Case Study of Cliff Swallows 

Charles R. Brown 

Mark-recapture methods are being increasingly applied to problems in ecology, evolution, 

and behavior. These methods and the analytical tools associated with them 

can provide important insight into issues that field biologists have grappled with for 

decades, such as estimating population size, survival probabilities as a component of 

fitness, dispersal, and recruitment and the likelihood of breeding. But effective use of 

mark-recapture methods usually requires a long-term data set, with animals followed 

over multiple capture periods, and a relatively large number of marked individuals for 

estimation of multiple parameters and the effects upon them. These constraints often 

limit the application of these methods, especially in behavioral ecology. Using a 22- 

year field project on colonially nesting cliff swallows (Petrochelidon pyrrhonota) that 

has included over 148,000 marked individuals, I illustrate the insights into social behavior 

and coloniality that have been made possible by mark-recapture. Specifically, 

I will explore how annual survival probability has increased our understanding of fitness 

associated with different clutch sizes, laying dates, and alternative reproductive 

strategies; how estimating the presence of transient birds from mark-recapture has 

been applied to studies of between-group parasite transmission; and how daily survival 

probability during the breeding season can integrate the disparate costs and benefits 

of coloniality. Future work for both swallows and other species should emphasize 

multistate analyses so that between-year survival and movement as a function of 

a bird’s group size in different years can be estimated. 

Individual Papers 

88:55 AM - 9:20 AM 

Effects of Spring conditions on breeding propensity of Greater Snow Goose females 

Eric Reed, Gilles Gauthier & Jean-Francois Giroux 

Breeding propensity, the probability that a sexually mature adult will breed in a given 

year, is an important determinant of annual productivity. It is also one of the least 

known demographic parameters in vertebrates. Although recent methodological 

advances in mark-recapture methods have allowed the estimation of breeding propensity 

for some bird species, little is known about factors responsible for temporal 

variation in this parameter. We studied the relationship between breeding propensity 

and conditions encountered on the spring staging areas and the breeding grounds in 

Greater Snow Geese (Chen caerulescens atlantica), a long distance migrant that 

breeds in the High Arctic. We used an Robust Design approach where classic capture 

histories were combined with auxiliary information from radio-marked birds and nest 

survey data to estimate breeding propensity over a 7-year period. More specifically, 

mark-recapture data was used to estimate probabilities of capture if alive and present 

in the superpopulation, and information from radio-marked birds was used to estimate 

probabilities of capture given presence in the sampled area. We also had to correct 

our estimates for nest success, because failed nesters emigrated from the sampled 

area, and incomplete fidelity to brood rearing areas (i.e. sampling area). We tested 

6

EURING 2003 Radolfzell 

the effects of spring snow cover and a spring conservation hunt on breeding propensity 

using a weighted least squares approach. We also used an empirical variancecomponents 

approach and determined that true temporal variation in breeding propensity 

was considerable (mean breeding propensity: 0.574 [95% CI considering only 

process variation: 0.13 to 1]). Spring snow cover was negatively related to breeding 

propensity (βsnow = -2.05 ± 0.96 SE) and tended to be reduced in years with a spring 

hunt (β = -0.78 ± 0.35). Nest densities on the breeding colony and fall young/adult ratio 

were good indices of breeding propensity, with nest densities being slightly more 

precise. These results suggest that environmental conditions and disturbance encountered 

during the pre-breeding period can have a significant impact on productivity 

of Arctic-nesting birds. 

09:20 - 09:45 AM 

Earlier recruitment or earlier death? On assumptions of homogeneous survival rates 

in capture-recapture models to estimate recruitment 

Emmanuelle Cam, Evan Cooch & Jean-Yves Monnat 

Realized patterns of age of recruitment observed in the breeding segment of populations 

are governed by the product of two demographic components: [survival probability 

from birth to age ]*[transition probability from state prebreeder to state first-time 

breeder at age ]. To again insight into selective pressures shaping age of recruitment, 

one may address temporal variation in age of first breeding and covariation with population 

size or social and environmental factors. This involves comparison among 

"groups" (e.g., cohorts) of individuals encountering different environmental conditions 

when they reach a given age as prebreeders. 

However, measures of recruitment based exclusively on data from the breeding segment 

of the population ignore the size of the pool of prebreeders (as opposed to transition 

probability). However, in a large number of species with deferred breeding, individuals 

are not encountered prior to breeding. Approaches to estimating age-specific 

recruitment probability in the absence of data from prebreeders using mark-recapture 

have been developed (Pradel and Lebreton 1999, Schwarz and Arnasson 2001, Williams 

et al. 2002). Unless the survival component of realized age-specific recruitment 

rates is known to be identical among the groups compared, inferences about the 

"cause" of variation in realized age of first breeding among groups are difficult: such 

differences may in fact result from differences in survival probability before individuals 

make the transition between states. Inferences about differences in realized recruitment 

rates among groups exclusively based on data from the breeding segment of 

the population reflect differences in recruitment probability under the assumption that 

there is no difference in prebreeder survival among these groups. 

We assessed the consequences of violations of this assumption on our perception of 

age-specific realized recruitment rates using numerical simulations; the scenarios 

considered correspond to various biological hypotheses about group-specific variation 

in survival and transition probabilities. Data were simulated under regular multistate 

models (Nichols and Kendall 1995) and truncated individual histories were then analyzed 

using the reverse-time approach (Pradel 1996) and the approach developed by 

Schwarz and Arnasson (2001). Depending on the scenario, realized recruitment was 

delayed or advanced compared to the underlying pattern of age-specific transition 

probabilities. We also addressed age-specific recruitment probability in a long-lived 

seabird species, the kittiwake (Rissa tridactyla) using a data set including data from 

prebreeders. We compared results from the reverse-time approach and the multistate 

approach, and showed that transition probability directly estimated or derived from 

7


measures of recruitment based on the breeding segment of the population are different. 

09:45 AM - 10:10 AM 

Predictors of reproductive costs in Soay sheep: a multistate capture-recapture analysis 

Giacomo Tavecchia, T. Coulson, B.J.T. Morgan, J. Pemberton, J. Pilkington 

& T.H. Clutton-Brock 

We investigate factors influencing the shape of the trade-off between survival and reproduction 

in female Soay sheep (Ovis aries). Multistate capture-recapture models 

are used to incorporate the state-specific recapture probability and to model the cost 

function directly from data on marked individuals. The cost of reproduction is identified 

as a quadratic function of the mother's age, being greatest for females breeding 

at the end of their first year and when more than 7 years old. Furthermore, the cost is 

only present during severe environmental conditions, when wet and cold winters 

coincide with high population density. Population density and winter severity negatively 

influence the probability of successfully breeding in the first year of life. Finally, 

the recapture probability depends on the breeding state: breeding females are virtually 

always recaptured while non-breeding individuals are not. The significant influence 

of the interaction between age and time on the phenotypic trade-off function 

might be responsible for maintaining cohort differences in demographic parameters. 

The use of multi-state capture-recapture analysis has allowed us to draw broader 

conclusions than earlier work, which was based on conditional methods of analysis 

10:10 AM -10:25 AM - break 

10:25 AM - 10:45 AM 

Breeding cycles and reproductive investment: the case of the Leatherback sea turtle 

Philippe Rivalan, R. Pradel, R. Choquet, J.-P. Briane, M. Girondot & A.C. Prevot-Julliard 

An important demographic parameter is the trade-off between current and future reproduction. 

A particular case is the relation between the number of years without reproduction 

and reproductive investment in years with reproduction.. We studied this 

trade-off in a long-lived species with intermittent breeding, the leatherback sea turtle, 

in French Guiana. Leatherback sea turtles lay several clutches within a reproductive 

year but reproduce every two or more years. Female leatherback sea turtles were 

captured, marked and recaptured when they lay eggs on nesting beach, from 1990 to 

2002 (60,000 CMR events divided in two parts: (1) females marked from 1990 to 

1994 with monel tags and subject to high tag-loss rate; (2) females marked from 1995 

to 2002 with permanent pit-tags). We modeled intermittent breeding and reproductive 

investment as following: 

(1) We estimated the number of years without reproduction by using multistate models 

incorporating several non-observable states using the software M-Surge (Choquet 

et al. 2003). We included tag-loss probabilities in transition matrices for females 

marked from 1990 to 1994. 

(2) We assessed investment in reproduction (i.e., number of clutches) by estimating 

the duration of presence of females on the beach. This duration is equivalent to a 

stopover duration of migrating birds. We estimated this annual stopover duration of 

females on the beach using the software SODA (Schaub et al. 2001). 

8


We found two breeding strategies which have the same fitness, characterized by the 

number of years without reproduction (2 or 3) and the number of clutches within reproductive 

years. However, such strategies are not fixed for a given female. 

10:45 AM - 11:05 AM 

Modelling senility and dispersal of Red deer 

Ted Catchpole, T.N. Coulson, Y. Fan, & B.J.T. Morgan 

Red deer (Cervus elaphus) on the island of Rum have been closely studied for many 

years. In particular, this has resulted in an extensive mark-recapture-recovery (mrr) 

data set. Models for these data need an elaborate age-structure for survival, because 

deer react differently to environmental factors at different stages of their lives. An additional 

interesting feature of the data is that animals may leave the study area. Thus 

we are interested not only in modelling deer survival, but also deer dispersal. Our 

modeling approach is a classical statistical one. The two sexes need to be considered 

separately. Within each sex, we use standard likelihood tools and information criteria 

to identify age-classes, separately for survival and dispersal. Within each age-class, 

the relevant probability does not vary over the ages desribed by that class, with the 

exception of the oldest age-class for survival. This then allows us to undertake logistic 

regressions of the relevant age-class probabilities on a mixture of environmental and 

individual covariates. Senility, a gradual decline in survival probability with age, is described 

by means of a logistic regression on age within the oldest survival ageclasses 

for males and females. The final models that we select for males and for females 

have an attractive simplicity. They demostrate clearly the differences that exist 

between males and females, and may be used for predicting future behaviour, and 

the effects of alternative management policies. 

The work of this paper is a natural extension of the models and model-selection 

procedures that have been developed for Soay sheep (Ovis aries) in Catchpole et al 

(2000) and Coulson et al (2001). The present work builds on the modelling of male 

red-deer alone by Catchpole et al (2002), and the modelling of both sexes in Fan et al 

(2002). 

11:05-11:25 AM 

Assessing senescence patterns in populations of large mammals 

Jean-Michel Gaillard & Anne Viallefont 

Theoretical models such as Gompertz and Weibull models are commonly used to 

study senescence in survival for humans (Olshansky & Carnes 1997) and laboratory 

or captive animals (Ricklefs 2000) for which a complete follow up of individuals is 

available. For wild populations of vertebrates, senescence in survival has more commonly 

been assessed by fitting simple linear or quadratic relationships between survival 

and age (e. g., McDonald et al. 1996, Newton and Rothery 1997, Nichols et al. 

1997, Loison et al. 1999). By using appropriate constraints on survival parameters in 

Capture-Mark-Recapture (CMR) models, we propose a first analysis of the suitability 

of Gompertz and Weibull models for describing aging-related mortality in free-ranging 

populations of ungulates. We first show how to handle Gompertz and Weibull models 

in the context of CMR analyses. Then we perform a comparative analysis of senescence 

patterns in both sexes of two ungulate species highly contrasted according 

to the intensity of sexual selection. Evolutionary implications of our results are 

discussed. 

9


11:25 AM - 11:45 AM 

Breeding site-fidelity and survival estimation of a migratory songbird: implications for 

conservation, life-history and study designs 

Duane Diefenbach, M.R. Marshall, L. Wood, & R. Cooper 

We used data from a 5-year banding study of 423 Prothonotary Warblers (Protonotaria 

citrea) from the White River National Wildlife Refuge, Arkansas, USA to estimate 

annual apparent survival rates of adults. Also, we documented inter-annual changes 

in the placement of breeding territories to estimate emigration rates and distances 

moved. Recognizing that the apparent survival estimates included both survival and 

permanent emigration, and possibly temporary emigration, we simulated these movement 

patterns over a range of true survival rates (0.3 - 0.9) to estimate the bias introduced 

by emigration if apparent survival estimates were used as true survival rate 

estimates. The simulations suggested that the observed warbler movements resulted 

in apparent survival estimates, compared to true survival rates, being 25% and 34% 

less (percent relative bias) for males and females, respectively. Therefore, our observed 

apparent survival rates of 0.52 and 0.39 for male and female warblers, respectively, 

may represent true survival rates of 0.69 and 0.60. Differential emigration rates 

confound inferences regarding differences in apparent survival between sexes and 

among species. Moreover, the bias introduced by using apparent survival rates for 

true survival rates can have profound effects on the predictions of population persistence 

through time, source/sink dynamics, and other aspects of life-history theory. 

For instance, we demonstrate with a stochastic population persistence model that if 

apparent survival rate estimates are used as true survival estimates, the underestimate 

of true survival because of emigration results in the misclassification of a local 

"source" population as a "sink". 

We investigated two study designs and analysis approaches that may result in apparent 

survival estimates that are closer to true survival estimates. The first involved a 

smaller "core" area where all marking of birds takes place and progressively larger 

"resighting" areas surrounding the core where researchers search for marked birds. 

We demonstrate that as the resighting areas get progressively larger, and therefore 

incorporate more "emigrants," apparent survival estimates begin to approximate true 

survival rates. Second, we investigated using a Robust Design data collection and 

analysis approach to estimate emigration rates directly. Both approaches are limited 

by logistical difficulties of data collection, but provide less biased estimates of survival. 

10


Random effects (chairs: Jean Clobert & Ken Burnham) 

Plenary Address - 01:15 PM -01:55 PM 

Random Effects: Present and future 

Carl Schwarz 

Random effects models have been typically used in capture-recapture settings as an 

intermediate model between fully time dependent and constant over time models. We 

review the approaches to fitting these models and discuss the advantages and disadvantages 

of each. Then extensions of random effect models to incorporate autocorrelation 

over time, to model classical design-of-experiments random effects, and to 

model other phenomena will be discussed. 


01:55 PM - 02:20 PM 

Evaluation of some Bayesian MCMC random effects inference applicable to bird ringing 

data 

Ken Burnham & Gary White 

The study design used to evaluate a method of moments random effects analysis for 

CJS data with per-occasion survival probability as a beta-distributed random variable 

(Burnham and White, 2002, Journal of Applied Statistics 29:245-264) is herein repeated 

to evaluate a fully Bayesian MCMC random effects analysis. The design has 5 

factors: occasions (t=7, 15, 23, 31), new releases per occasion (u=25, 100, 400), 

capture probabilities (p=0.4, 0.6, 0.8; no time effects), expected survival probability 

(E(S)=0.4, 0.6, 0.8; no time effects), and process variation, var(S)=σ2 (s=0, 0.025, 

0.05, 0.1). At each of these 432 design points we generated 500 independent (Monte 

Carlo) data sets. Then MCMC inference was used, based on 1,000 nearly independent 

samples from the Bayesian posterior for uniform marginal priors on p, E(S) and 

σ2 (0 to 0.25). We look at properties of point and interval inference on per-occasion 

S, E(S) and σ2 based on the posterior sample. Some issues considered: coverage of 

equal tailed versus shortest 95% credibility intervals, and point estimation, especially 

of σ2 (mean versus mode). The simulations are done, basic results are known. Comparisons 

to Burnham and White (2002) will be made; the Bayesian approach did well. 

02:20 PM - 2:45 PM 

Testing the additive versus compensatory mortality hypothesis using a random effects 

model 

Michael Schaub & Jean-Dominique Lebreton 

We propose a new way to test whether a specific cause of mortality (say cause A) is 

additive to the remaining mortality or whether it is compensated for by other forms of 

mortality. In contrary to the existing tests, our approach does not require an independent 

estimate of the "intensity" that caused mortality due to a specific reason (e.g. 

hunting effort). It is enough to have ring-recovery data, where the cause of death of 

the recovered animals is known. First we estimate the mortality probability due to 

11

Random effects 

cause A and due to all other causes of death with a time-dependent multi-state model 

where the states are "alive", "newly dead because of A" and "newly dead because of 

another reason than A." Second, we test the two opposing hypotheses by using random 

effects methodology. If the "additive mortality hypothesis" would be true, the correlation 

between the two mortalities over time is zero. If there would be some form of 

compensation, there is a negative correlation between the two sources of mortality 

over time. To extract this temporal correlation from the overall correlation we used 

random effects model. We illustrate the method with a case study of White Storks, 

where the two opposing sources of mortalities are natural and electrocution mortality. 

It appears that the two sources of mortality are partially compensatory in the adults, 

but additive in the first year birds. 

02:45 PM - 3:10 PM 

Random effects and individual differences: looking for trees in the 'life history' forest 

Emmanuelle Cam, Bill Link, Evan Cooch, Jean-Yves Monnat, & Etienne Danchin 

We investigated the influence of age on survival and breeding rates in a long-lived 

species Rissa tridactyla using models with individual random effects permitting variation 

and covariation in fitness components among individuals. Differences in survival 

or breeding probabilities among individuals are substantial, and there was positive 

covariation between survival and breeding probability; birds that were more likely to 

survive were also more likely to breed, given that they survived. The pattern of agerelated 

variation in these rates detected at the individual level differed from that observed 

at the population level. Our results provided confirmation of what has been 

suggested by other investigators: within-cohort phenotypic selection can mask senescence. 

Although this phenomenon has been extensively studied in humans and captive 

animals, conclusive evidence of the discrepancy between population-level and 

individuallevel patterns of age-related variation in life-history traits is extremely rare in 

wild animal populations. Evolutionary studies of the influence of age on life-history 

traits should use approaches differentiating population level from the genuine influence 

of age: only the latter is relevant to theories of life-history evolution. The development 

of models permitting access to individual variation in fitness is a promising 

advance for the study of senescence and evolutionary processes. 

03:10 PM - 3:45 PM 

Nonidentifiability of population size from capture-recapture data with heterogeneous 

detection probabilities 

Bill Link 

Heterogeneity in detection probabilities has long been recognized as problematic in 

mark-recapture studies, and numerous models developed to accommodate its effects. 

Individual heterogeneity is especially problematic, in that reasonable alternative 

models may predict essentially identical observations from populations of substantially 

different sizes. Thus even with very large samples, the analyst will not be able to 

distinguish among reasonable models of heterogeneity, even though these yield quite 

distinct inferences about population size. I illustrate the problem using the simple 

closed population model 

12


Multi-state models (chairs: Emmanuelle Cam & Neil Arnason) 

Plenary Address - 03:50 PM - 04:30 PM 

Coping with unobservable and misclassified states in multi-state capture-recapture 

studies 

Bill Kendall 

Multi-state capture-recapture models have proven to be very useful in the study of animal 

population and meta-population dynamics. Principal parameters of interest include 

state-specific survival and transition probabilities. Depending on the study, transition 

probabilities can refer to movement between sub-populations (e.g., breeding colonies) 

or transitions between life history stages (e.g., size classes, breeding states). Inherent 

to traditional multi-state models are the assumptions that in each capture period each 

animal in the population or meta-population is observable (i.e., subject to detection), 

and the state of each captured animal is known with certainty. In fact violations of these 

assumptions are not uncommon. Examples of temporary movement of marked animals 

out of the study area can include roaming within a home range, movement underground 

(e.g., transition into torpor), and transition from breeding to non-breeding state 

where the study areas consist of breeding colonies. Misclassification of the state of an 

animal can occur when a cue that determines state is missed or misinterpreted. Breeders 

can be misclassified as non-breeders when young are present but not observed. 

Diseased animals can be misclassified as healthy because symptoms are absent or 

not observed. In other cases (e.g., sexually monomorphic bird species) sex can remain 

undetermined for several or all encounters because sex-specific behavioral cues are 

not observed. 

I will outline the impact of unobservable and misclassified states on estimators for survival 

and transition, derived from models that ignore these phenomena. I will then describe 

approaches for coping with biases associated with unobservable and misclassified 

states. These approaches will include both design (to minimize the nuisance aspect 

of these phenomena) and modeling (to adjust for these phenomena) elements. 

The latter will often require the collection of extra information, including capture data 

from subsamples (i.e., the robust design), telemetry, band recoveries, or ancillary observations. 


04:30 PM - 04:50 PM 

Costs of first reproduction in a long-lived bird: effects of environmental and individual 

covariates 

Christophe Barbraud & Henri Weimerskirch 

How animals balance their investment in young against their own chances to survive 

and reproduce in the future? This life-history trade-off, referred to as the cost of reproduction, 

holds a central place in life-history theory. Long-lived birds are good candidates 

to be used as models to detect these costs. They should be more restrictive than 

short-lived birds in the degree to which they exhibit increased effort, because even a 

small reduction in adult survival would reduce the number of subsequent breeding attempts, 

thereby greatly lowering lifetime reproductive success. However, at least two 

13

Multi-state models 

factors are likely to confound the measurement of this trade-off in the wild. First, there 

could be differences in the amount of energy individuals acquire and allocate to various 

functions. In that case we might expect that some individuals would perform well in 

both reproduction and survival, whereas low quality individuals would die sooner. Second, 

there could be variations in resource availability affecting energy acquisition and 

allocation. Theoretical models examining the optimal phenotypic balance between reproduction 

and adult survival under variable breeding conditions have recently investigated 

the second issue. However, very little is known on the influence of individual 

quality on the costs of reproduction. Here, we use a capture-recapture dataset on blue 

petrels to investigate the costs of first reproduction. The use of multi-state models with 

three states (non-breeder, first time breeder, and experienced breeder) allowed us to 

show that first time breeders have a lower probability of breeding the following year 

than experienced breeders, and that in some years, first time breeders have a lower 

survival probability that experienced and non-breeders. These results suggest that first 

time breeding may act as a filter, selecting good quality individuals. Using environmental 

and individual covariates, we further show that the costs of first reproduction are 

particularly acute in years when environmental conditions are poor, and that individual 

body condition affects both survival and breeding probabilities. 

04:50 PM - 5:10 PM 

Density dependence in North American ducks 

Steve.P. Brooks and Lara E. Jamieson 

The existence or otherwise of density dependence within a population can have important 

implications for the management of that population. Here, we use estimates of 

abundance obtained from annual aerial counts on the major breeding grounds of a variety 

of North American duck species and use a state space model to separate the observation 

and ecological system processes. This state space approach allows us to 

impose a density dependence structure upon the true underlying population rather than 

on the estimates and we demonstrate the improved robustness of this procedure for 

detecting density dependence in the population. We also show how the inclusion of time-varying 

covariates such as the number of May ponds provides additional descriptive 

power within the model and that their omission may sometimes lead to erroneous 

conclusions as to the presence of density dependence. We adopt a Bayesian approach 

to model fitting, using Markov chain Monte Carlo (MCMC) methods and use a reversible 

jump MCMC scheme to calculate posterior model probabilities which assign probabilities 

to the presence of density dependence within the population, for example. We 

show how these probabilities can be used either to discriminate between models or to 

provide model-averaged predictions which fully account for both parameter and model 

uncertainty. 

05:10 PM - 05:30 PM 

Demographic estimation methods for plants in the presence of dormancy 

Marc Kery & Kathy B. Gregg 

Demographic analysis seems straightforward in plants due to their sessile nature. Problems 

arise when there is an unobservable dormant state that stays belowground for one or more 

growing seasons. Conventional analysis methods make strong assumptions about the duration 

of dormancy and obtain estimates of demographic parameters, which, however, will be 

biased to an unknown degree. In contrast, we use capture-recapture (CR) methods to obtain 

unbiased estimates of the fraction of a population that is in the dormant state, and of survival 

and transition rates between life-states. As an illustration, we analyze a 10-year data set on 

14


Cleistes bifaria, a terrestrial orchid with frequent dormancy, using both single- and multi-state 

CR models as well as five conventional methods for comparison. During the study period, 

35% of ramets were dormant at least once, for between 1 and 4 (mean 1.4) years. Capturerecapture 

models estimated ramet survival rate at 0.86 (SE ~0.01), ranging 0.77-0.94 

(SE

Notation and terminology 

Notation and terminology (organizer: David Thomson) 

Roundtable Discussion 

David Thomson 

This will be a discussion session about notation and terminology. EURING technical 

meetings typically involve collaboration between scientists from diverse backgrounds 

along the spectrum from field biologists to theoretical statisticians. The success of this 

inter-disciplinary collaboration largely depends on the quality of the communication 

and comprehension. In an effort to enhance this, we aim to highlight and discuss ambiguities 

which have arisen in terminology and notation and we aim through discussion 

to strive towards standards and consensus. 

Participants were encouraged to report ambiguities which they would like to see 

discussed. Also views on particular notation and terminology which should or should 

not be used, were gathered. These contributions will form the basis for an agenda, 

and through interactive discussion on each topic we will try to work towards standards 

which will be acceptable to all. 

16


Methodological advances (chairs: Jean-Dominique Lebreton & Ken Pollock) 

Plenary Address - 08:15 AM - 08:55 AM 

Principles and interest of GOF tests for multistate models 

Roger Pradel, Olivier Gimenez & Jean-Dominique Lebreton 

The assessment of the fit of multistate models has until now been available only as the 

comparison of the expected and observed values in the m-array provided by program 

MSSURVIV. This test is partial and omnibus. An alternative approach based on contingency 

tables has been proposed recently for the JMV model, a generalization of the 

Arnason-Schwarz model (Brownie et al. 1993). Although this test is for just one model, 

we examine whether it could nonetheless provide a good insight into the data. It has 

two main components: Test 3G which compares the future of animals released simultaneously 

and in the same state according to their past capture histories, and Test M 

which compares the pattern of recaptures of animals previously released and not captured 

at a given occasion to that of animals released at this occasion. A likely departure 

from the assumptions of both the JMV and the AS models is memory of past locations. 

A suitable partitioning of Test 3G leads to a subcomponent Test WBWA which can be 

used to detect memory. We examine which statistics to employ to test efficiently for 

memory. The AS model is often biologically more relevant than the JMV model. To assess 

the fit of the AS model one can supplement the GOF test to the JMV model with 

the likelihood ratio test between JMV and AS. Here we examine an alternative solution: 

replacing Test M with a comparison of observed and expected values in the m-array 

using a parametric bootstrap procedure. 


08:55 AM - 09:20 AM 

Open capture-recapture models with heterogeneity: III. The robust design 

Shirley Pledger, Kenneth H. Pollock, & James L. Norris 

Finite mixture methods may be used in capture-recapture studies to allow for heterogeneity 

of survival. Following the methods of Norris and Pollock (1996), Pledger (2000) and Pledger 

and Schwarz (2002), animals are assumed to belong to one of finitely many groups, each of 

which has its own survival rates and capture rates. The group to which a specific animal belongs 

is not known, so its survival and capture probabilities are random vectors from a finite 

mixture. The promising results from the papers above have led to a series of three proposed 

papers by Pledger, Pollock and Norris. The first, "Open Capture-Recapture Models with Heterogeneity: 

I. Cormack-Jolly-Seber Model" is almost ready for submission to Biometrics.It 

deals with analyses which condition on first capture, allowing the modelling of heterogeneous 

capture and survival probabilities. The second, "Open Capture-Recapture Models with Heterogeneity: 

II. Jolly-Seber Model" deals with abundance estimates and their sensitivity to the 

presence of heterogeneity. The paper is partly written, with the computer programs functioning. 

This will be submitted to Biometrics in 2003. This third in the series, "Open Capture- 

Recapture Models with Heterogeneity: III. The robust design", will summarise and tie together 

the series of three papers, and give details and examples of these methods with the robust 

design sampling scheme. This paper complements and completes the development of 

likelihood-based models in Kendall, Pollock and Brownie (1995). 

17

Methodological advances 

09:20 AM - 09:45 AM 

Multiple Species Capture-Recapture and Removal Models 

Kenneth H. Pollock, James L. Norris, & Shirley Pledger 

A large body of literature exists for estimating species richness. These models allow 

for uncertain detection of each species (Boulinier et al. 1998a). Statistical methods 

adapted from capture-recapture and removal sampling, originally used to estimate individual 

species abundance can be applied here. (See for example, Burnham and 

Overton 1979; Nichols and Conroy 1996; Boulinier et al. 1998a). These methods 

have begun gradually to be used in the ecological literature, but progress has been 

slow, considering that the methodology has been around since 1979. In this paper, 

we shall begin with a detailed description of the estimation of species richness in the 

literature and then develop general removal and capture-recapture sampling models 

where multiple species are removed or marked and recaptured and the objective is to 

estimate species richness and the species abundance curve. 

09:45 AM - 10:10 AM 

Time Varying, Continuous Covariates in the Cormack-Jolly-Seber Model 

Simon Bonner & Carl Schwarz 

One area of research into capture-recapture methodology has focused on techniques 

to study the impact of different covariates on the population parameters. Straightforward 

methods have been developed for cases where the covariates are either discrete, 

constant over time, or apply to the population as a whole, but the problem has 

not been solved for the case of continuous, time dependent covariates unique to each 

individual (e.g. body mass). 

Because it is impossible to measure such variables on occasions where an individual 

was not captured, estimation can be considered a missing data problem. In this project, 

a model was constructed using a diffusion process to describe changes in the 

covariate. Logistic functions were used to link the covariate to capture and survival 

rates and incorporate the data into the Cormack-Jolly-Seber model. Two methods of 

parameter estimation were developed based on techniques commonly used to handle 

missing data: namely the EM-algorithm and Gibb's sampling. These methods were 

both applied to simulated and real data sets, and comparison of the results is provided. 

In short, though both methods gave similar results the Gibb's sampling technique 

was much easier to implement, more efficient to compute, and yielded estimates of 

the standard errors more readily than the EM-algorithm. 

10:10 AM -10:30 AM - break 

10:30 AM - 10:55 AM 

Application of Bayesian Statistical Inference to Capture-Recapture Data, using 

WinBUGS 

Howard Stauffer 

I will describe Bayesian statistical analysis modeling solutions for capture-recapture 

data. The solutions are based upon Monte Carlo Markov Chain (MCMC) iteration 

methods available with the Bayesian software WinBUGS. This analysis approach offers 

an alternative to the maximum likelihood estimation approach available with the 

frequentist capture-recapture software MARK. The capture-recapture analysis is illu- 

18


strated with data from a hen clam pollution experiment and compared with results 

from previous analysis by Anderson, Burnham, and White, using MARK. For noninformative 

priors, the Bayesian and frequentist statistical results are comparable. For 

informative priors, however, results may differ. Bayesian inference offers a sequential 

approach to analysis, based upon multiple datasets obtained from population monitoring, 

providing periodic reassessments of parameters along with estimates of risk. 

Such reassessments are useful for adaptive management decision-making. Bayesian 

models can also be compared, analogous to frequentist models, using informationtheoretic 

methods based upon AIC and DIC weights, for their relative competitiveness 

at fitting population sample data, and model averaging techniques can be applied to 

provide robust estimates of parameters. 

10:55 AM - 11:20 AM 

Evaluation of ultrastructure and random effects band recovery models for estimating 

relationships between survival and harvest rates in exploited populations 

Dave Otis & Gary White 

The functional relationship between vital rates and harvest rates of exploited populations 

is a fundamental interest of population biologists. Despite the development by 

many authors of a collection of density- dependent population models and functional 

representations of the relationship between annual survival and harvest rates, statistical 

analysis techniques for empirical investigation of these phenomena are extremely 

limited. In the case of band recovery data from harvested species, standard practice 

has been to incorporate ultrastructure functions of the form Si = S0 ( 1 - b*Ki) or Si = 

S0 ( 1 - Ki)b into band recovery models and use the estimated parameter b as a index 

for the relative evidence for additive or compensatory harvest mortality. Satisfactory 

performance of this approach has been inconsistent, and limited Monte Carlo simulations 

of the statistical performance of the estimator have revealed some problematic 

distributional and bias properties. Furthermore, the sensitivity of the estimator for detecting 

annual survival rate and harvest rate relationships is unknown. 

An alternative approach to the use of fixed effect ultrastructure models is possible if 

we consider annual harvest rates and survival rates as random effects. We envision 

an underlying process covariation between these 2 rates, which represents the parameter 

of interest, that is randomly perturbed by a collection of additional biotic and 

abiotic factors. The perturbation could be temporal, as in the case of released banded 

cohorts from single population for a series of years. Alternatively, if banding is done in 

multiple populations, we might consider survival and harvest as fixed effects in a given 

population, and assume random spatial perturbation in the parameters. 

We construct underlying models that incorporate specified additive, linear compensatory, 

and nonlinear compensatory functional relationships between harvest and 

natural mortality in a seasonally exploited population, and use Monte Carlo simulation 

to generate annual samples of band recovery data. These datasets are then analyzed 

using fixed effect ultrastructure models and random effects models. Parameter estimation 

is accomplished by using customized SAS code in PROC NLMIXED. Summary 

statistics of performance the alternative techniques are presented and compared, 

design considerations are discussed, and recommendations are made for further development. 

19

Methodological advances 

11:20 AM - 11:45 AM 

Integrated analysis of wildlife population dynamics 

B. Morgan, P. Besbeas, L. Thomas, S. Buckland, J. Harwood, C. Duck, & P. Pomeroy 

We describe two related methods for the combined analysis of mark-recapture and 

census data, using data for British grey seals as our motivating example. The first 

method is based on maximum likelihood: the likelihood for the mark-recapture data is 

combined with a likelihood from Kalman filter theory applied to a state-space model 

for the census data. We show that this approach can allow estimation of state parameters 

(such as fecundity) that would not be estimable using either dataset separately. 

The Kalman filter works well for normal linear models, but more complex models 

require more complex analysis methods. As an example, we demonstrate the use of 

a Bayesian nonlinear particle filter called sequential importance sampling (SIS) to fit 

nonlinear models including density dependence and movement between breeding 

colonies. In principle, these approaches could be extended to provide an integrated 

analysis of many diverse types of data. 

20


Computing & software (chairs: Jim Hines & Gary White) 


The "Mother of All Models" 

Richard Barker 

With an increasing proliferation of mark-recapture models and studies collecting 

mark-recapture data, software and analysis methods are being continually revised. 

We consider the construction of the likelihood for a general model that incorporates 

all the features of the recently developed models: it is a multistate robust-design 

mark-recapture model that includes dead recoveries and resightings of marked animals 

and is parameterised in terms of state-specific recruitment, survival and capture 

probabilities, state-specific abundances, and state-specific recovery and resighting 

probabilities. The construction that we outline is based on a factorisation of the likelihood 

function with each factor corresponding to a different component of the data. 

Such a construction would allow the likelihood function for a mark-recapture analysis 

to be customized according to the components that are actually present in the dataset. 


01:55 PM - 02:20 PM 

Computing issues concerning hierarchical models 

Ken Burnham 

The subject of capture-recapture (CR) data analysis is incomplete without welldeveloped, 

user capable software (such as MARK) that can fit models incorporating 

random effects as well as fixed effects. The MCMC (i.e., Bayesian) model fitting approach 

provides a sound theory for fitting all CR models. Why not just use it? Answer: 

lack of software suitable for all potential users, and the relative slowness of MCMC 

analysis (there are reasons why speed still matters). Pragmatically I maintain that likelihood 

analysis, including profile likelihood intervals, as needed, is eminently adequate 

for CR data analysis when models have all fixed effects: the inferential results 

are virtually the same as those from MCMC with vague priors; but the analyses are 

much faster. This talk considers what we might be able to do to obtain CR random 

effects inferences, say incorporated into MARK, that are suitable approximations to 

MCMC inferences at a fraction of the computing time. 

The technical issue is simple. If all effects are considered fixed, numerical likelihood 

inference only requires function evaluations over a K dimensional space. Exact inference 

for a set of K random parameters requires integration of an even more complex 

function (than the likelihood) over a K dimensional space. Such integration is computationally 

more demanding than function maximization. MCMC analysis has complexity 

at the same level as the integration approach. The random effects method in 

MARK can be reformulated and extended. So doing has value, even though I accede 

to the MCMC approach as the gold standard for rigor and generality. 

The talk will note several possible approximate analyses approaches for some types 

of random effects (along with fixed effects, so technically mixed models). Under any 

inference computations we will augment the fixed effects likelihood with a model for 

21

Computing & software 

the "randomly" varying parameters, a model which includes fixed hyperparameters. 

Focus is usually on survival probabilities. In some sense the resultant analysis is either 

a type of penalized likelihood for inference about random effects, or a smoothed 

likelihood for inference about fixed effects, including hyperparameters. Thus, the issue 

can be thought of more as a computing issue than as a modeling issue: the same 

model underlies all inference methods for random effects. 

02:20 PM - 02:45 PM 

Estimation of nest success rates and comparing methods avaialbe in SAS GLM, SAS 

NLMIXED, and MARK 

Jay Rotella, Steve Dinsmore & Terry Shaffer 

Estimating nesting success and evaluating factors potentially related to nest survival 

are key aspects of many studies of avian populations. A strong interest in nest survival 

has led to a rich literature detailing a variety of estimation methods for this vital 

rate. In recent years, modeling approaches have undergone especially rapid development. 

Despite these advances, most recent field studies still employ Mayfield's adhoc 

method (1961) or, in some cases, the maximum-likelihood estimator of Johnson 

(1979) and Bart and Robson (1982). Such methods allow for analyses of stratified 

data but do not allow for more complex and realistic models of survival data that include 

covariates that vary by individual, nest age, time, etc. and that may be continuous 

rather than categorical. Methods that allow researchers to rigorously assess 

the importance of a variety of biological factors that might affect nest survival can now 

be readily implemented in Program MARK and in SAS's Proc GENMOD and Proc 

NLMIXED. In this paper, we first describe the likelihood used for these models and 

then consider the question of what the effective sample size is for computation of 

AICc. Next, we consider the advantages and disadvantages of these different programs 

in terms of ease of data input and model construction; utility/flexibility of generated 

estimates and predictions; ease of model selection; and ability to estimate variance 

components. Finally, we discuss improvements that would, if they became 

available, promote a better general understanding of nest survival. 

2:45-3:10 PM 

M-SURGE: an integrated software for multistate recapture models 

Remi Choquet, Anne-Marie Reboulet, Roger Pradel, Olivier Gimenez & 

Jean-Dominique Lebreton 

M-SURGE (along with its companion program U_CARE) has been written specifically 

to handle multistate capture-recapture models (Lebreton and Pradel 2002) with the 

ultimate concern to alleviate their inherent difficulties (model specification, quality of 

convergence, flexibility of parameterization, assessment of fit…). In its domain, 

MSURGE covers a broader range of models than a general program like MARK 

(White and Burnham 1999), while being more user-friendly than a research program 

like MS-SURVIV (Hines 1994). 

Among the main features of MSURGE is a wide class of models and a variety of parameterizations: 

• M-SURGE integrates conditional models with probability of recapture depending 

on the current state (Arnason-Schwarz type models) but also on the current and 

previous state (Jolly-MoVement type models; (Brownie, Hines et al. 1993)). In 

both cases, age and/or time-dependence and multiple groups can be considered. 

22


• Combined Survival-Transition probabilities can be represented as such or decomposed 

into transition and survival probabilities (Hestbeck, Nichols et al. 1991). 

• Among the transition probabilities with the same state of departure, the one to be 

computed as 1 minus the others can be freely picked by the user. User-friendliness is 

enhanced by the easiness with which constrained models are built, using a language, 

interpreted by a generator of design matrices called GEMACO. This language is alike 

those in general statistical software such as SAS or GLIM, i.e., a formula such as t+g 

generates a model with additive effects of time and group, thus avoiding tedious and 

error-prone matrix manipulations using an editor or a spread-sheet. Examples of various 

types of multistate models are developed and presented. 

You can download M-Surge freely from ftp.cefe.cnrs-mop.fr/biom/Soft-CR. 

03:10 PM - 03:45 PM 

DENSITY: software for fitting spatial detection functions to data from passive 

sampling 

Murray Efford & Deanna Dawson 

Rigorous sampling of bird populations to estimate density raises the problem of incomplete 

detection (e.g., Pollock et al. 2002). Varying detectability is widely acknowledged, 

but variation in its spatial component (how detection declines with distance) is 

considered less often. Active methods (double sampling and distance sampling) require 

an observer to determine the instantaneous distance between the sampling 

point and each animal. Passive methods (mistnets or traps) rely on animals moving to 

the detector: instantaneous locations are unknown, and movement is an important 

unmeasured component of detectability. 

New methods have been developed to fit spatial detection functions to capturerecapture 

data from passive detectors. The methods are computer-intensive and depend 

on specialised software ('DENSITY'), available for download at 

www.landcareresearch.co.nz. DENSITY provides a graphical interface for the analysis 

of closed-population capture-recapture data from arrays of passive detectors. Its 

simulation capability enables users to perform power analysis of different sampling 

designs before going into the field. 

We demonstrate the use of DENSITY to estimate bird population density from mist 

netting data. Netting was conducted over 1992 to 1996 on a forest-pasture ecotone in 

Mexico. In each of 16 netting sessions, six local arrays of 20 nets were run for 2-3 

consecutive days. Despite the large number of captures in total, within-session recaptures 

were rare for most species. This restricted application of the method to a few 

common species (e.g. Sporophila torqueola) and to species aggregates (e.g. 'winter 

residents'). Although the use of the method for mist netting data was experimental, it 

may lead through simulation in DENSITY to improved design of mist net arrays where 

density estimation is a study goal. 

The spatially explicit framework of DENSITY opens up new possibilities and we expect 

the software to evolve. An exciting prospect is the direct fitting of simple density 

surfaces during estimation, given suitable spatial covariates. 

23

Population dynamics and monitoring applied to decision making 


(Chairs: Mike Conroy & Danny Lee) 


Ecosystem Management via Interacting Models of Political and Ecological Processes 

Timothy C. Haas 

The decision to implement environmental protection options is a political one. Depending 

on the political mechanisms operating, a country may or may not heed the most 

persuasive scientific analysis of an ecosystem's future health. A predictive understanding 

of the political processes that result in ecosystem management decisions may 

help guide the formulation of ecosystem management policy. To this end, this article 

develops a stochastic, temporal model of how political processes influence and are 

influenced by ecosystem processes. This model is realized in a system of interacting 

influence diagrams that model the decision making of a country's president, environmental 

protection agency, legislature, and rural inhabitants. These within-country decisions 

interact with models of international environmental protection organizations 

and the ecosystem enclosed by the country. As an example, this modeling framework 

is used to represent the decisions made to manage the cheetah population in the 

countries of Kenya, Tanzania, and Uganda. The model is fitted to both political decision 

and ecological data from these countries. This estimated model is then used to 

predict which management decisions will be the most politically acceptable to these 

countries. Finally, cheetah extinction probabilities are computed under these decisions. 

All software for building such a model of other managed ecosystems is freely 

available from www.uwm.edu/~haas/ems-cheetah/. 


04:30 PM - 04:55 PM 

A Bayesian integrated population dynamics model to analyze data for the eastern Pacific 

Ocean spotted dolphin 

Mark Maunder & Simon Hoyle 

Restrictions on fishing operations have been increasing in an effort to protect at-risk 

species taken as bycatch. With the increasing popularity of the precautionary approach, 

these restrictions are often conservative, because there is insufficient information 

about the effects of bycatch on many protected species. Informed decisionmaking 

requires quantitative analyses of both the effects of bycatch on these species 

and the effect of regulations on the fisheries. Uncertainty pervades management of 

protected species and, to be consistent with the precautionary approach, the uncertainty 

in analyses of these species must be described if the appropriate decisions are 

to be made. Bayesian analysis is an ideal framework for considering uncertainty during 

the decision-making process. Bayesian analysis also allows expert judgment or 

information from other populations or species to be included in the analysis if appropriate. 

Integrated analysis attempts to include all relevant data for a population into 

one analysis by combining analyses, sharing parameters, and simultaneously estimating 

all parameters, using a combined objective function. Integrated analysis ensures 

that model assumptions and parameter estimates are consistent throughout the analysis, 

that uncertainty is propagated through the analysis, and that the correlations 

24


among parameters are preserved. We combine Bayesian analysis and integrated 

analysis to develop a population dynamics model for the eastern Pacific Ocean (EPO) 

spotted dolphin. The model is developed to include the various types of data that are 

available for this population. Informative priors are included for several model parameters. 

Forward projections are used to investigate different management options. 

04:55 PM - 05:20 PM 

Application of Bayesian decision making and MCMC to the conservation of a harvested 

species 

Chris Fonnesbeck & Mike Conroy 

When endeavoring to make informed decisions, conservation biologists must frequently 

contend with disparate sources of data and competing hypotheses about the 

likely impacts of proposed decisions on the resource's status. Frequently, statistical 

analyses, modeling (e.g., for population projection) and optimization or simulation to 

investigate candidate alternative decisions, are conducted as separate exercises. For 

example, a population model might be constructed, whose parameters are then estimated 

from data (e.g., ringing studies, population surveys); finally, the parameterized 

model might then be used to investigate alternative candidate decisions, via simulation, 

optimization, or both. This approach, while effective, does not take full advantage 

of the integration of data and model components for prediction and updating; we propose 

a Bayesian context to provide this integration. 

In the case of American black ducks (Anas rubripes) managers are simultaneously 

faced with trying to extract a sustainable harvest from the species, while maintaining 

individual stocks above acceptable thresholds. The problem is complicated by spatial 

heterogeneity in the growth rates and carrying capacity of black ducks stocks, movement 

between stocks, regional differences in the intensity of harvest pressure, and 

heterogeneity in the degree of competition from a close congener, mallards (Anas 

platyrynchos) among stocks. We have constructed a population life cycle model that 

takes these components into account and simultaneously performs parameter estimation 

and population prediction in a Bayesian framework. Ringing data are used to 

develop posterior predictive distributions for harvest mortality rates, given as input 

decisions about harvest regulations. Population surveys of black ducks and mallards 

are used to obtain stock-specific estimates of population size for both species, for inputs 

into the population life-cycle model. These estimates are combined with the posterior 

distributions for harvest mortality, to obtain posterior predictive distributions of 

future population status for candidate sets of regional harvest regulations, under alternative 

biological hypotheses for black duck population dynamics. These distributions 

are then used both for the exploration of optimal harvest policies and for sequential 

updating of model posteriors, via comparison of predictive distributions to future 

survey estimates of stock-specific abundance. Our approach illustrates advantages of 

MCMC for integrating disparate data sources into a common predictive framework, for 

use in conservation decision making. 

05:20 PM - 05:45 PM 

Decision models for the optimal management of biodiversity trust fund 

Martin Drechler & Frank Wätzold 

The conservation of species generally requires long-lasting commitments over many 

years or decades. Even though technically, management plans can be designed for 

such long timeframes their practical implementation is constrained by the future 

25


available financial budget which may vary in time. A possibility to deal with the problem 

of variable future budgets is the employment of trust funds. At any point in time 

the decision maker can pay a certain proportion of the currently available money into 

the fund (with the remaining money being spent for conservation) or alternatively, 

draw a certain amount from the fund to add to the currently available budget. The optimal 

decision depends on various ecological and economic factors and state variables 

which may vary in time. We present two types of conservation problems: the 

management of an endangered population and the selection of nature reserves which 

are described by mathematical models in a general manner. Using a stochastic dynamic 

programming approach we derive analytical solutions for these two decision 

problems and deduce general guidelines for the efficient use of trust funds in the conservation 

of biodiversity. 

5:45-6:10 PM 

Costs of population measurement uncertainty in index-based monitoring 

Clint Moore & Bill Kendall 

Managers of wildlife populations commonly rely on indirect measures of the population 

in making decisions regarding conservation, harvest, or control. The main appeal 

in the use of such measures, or "indices," is their low material expense compared to 

methods that directly measure the population. Implicit in the use of indices is the assumption 

that they proportionately reflect population size. However, this assumption 

is rarely affirmed in practice, and decisions based on indices may or may not be the 

same as those that would be made if population status were known. Therefore, if the 

relationship between population size and its indirect measurement is unknown, then 

management based on indices incurs expected costs beyond the directly measurable 

costs of the monitoring program itself. Here, we analyze the making of optimal silvicultural 

decisions at the Piedmont National Wildlife Refuge (USA) for the joint benefit 

of two bird populations: the endangered red-cockaded woodpecker (Picoides borealis) 

and a shrub-nesting neotropical migrant, the wood thrush (Hylocichla mustelina). 

Response of the wood thrush population to management actions is largely unknown, 

therefore the degree to which management for the woodpecker conflicts with management 

for the wood thrush is uncertain. In our dynamic optimization model, we specifically 

address this form of structural uncertainty. We also address uncertainty in the 

relationship of wood thrush population status to an indirect measurement of abundance. 

In the model, state variables available to the manager at any decision opportunity are: 

(1) amounts of forest in each of three seral classes, (2) the indirect measure of wood 

thrush population size in each seral class, and (3) class-specific estimates of wood 

thrush population growth rate (8) obtained from values of the indirect population measures 

at successive decision opportunities. If the indirect measure of population size 

is a true index, then estimates of lambda should be unbiased, and decisions based on 

such indices would be exactly the same as those that would be made had the corresponding 

population abundances been known; that is, management based on such 

indices is optimal. Otherwise, management based on an incorrect belief in the strict 

proportionality of the index is suboptimal. Through analysis of the expected value of 

information, we calculate the expected cost of uncertainty in the relationship between 

the monitoring index and population size, and we do so in currency units of woodpekker 

habitat and wood thrush population growth. Thus, financial savings achieved by 

foregoing surveys that yield unbiased estimates of abundance may be balanced 

against expected resource costs incurred under simpler monitoring programs in which 

the relationship between the index and population size is not established. 

26


Dispersal & migration (Chairs: Carl Schwarz & Franz Bairlein) 


Quantifying variation in migratory strategies using ring-recoveries 

G.M. Siriwardena, C.V. Wernham & S.R. Baillie 

Bird populations have traditionally been labelled as “migrant” or “resident” on the basis 

of field observations and qualitative interpretations of patterns of ring-recoveries. 

However, even such a non-systematic approach has identified many intermediate 

species where only part of the population migrates (partial migrants) or where different 

components of the population migrate to different extents (differential migrants). 

A method that would allow a quantitative definition of migratory tendency to be derived 

for many species would facilitate investigations into the ecological causes and 

life-history consequences of migratory behaviour. Species or populations could then 

be placed objectively into the continuum between true residency and an obligate, 

long-distance migratory habit. 

We present a novel method for the analysis of ring-recovery data sets that produces 

just such a quantitative index of migratory tendency for British birds, developed as 

part of the BTO's Migration Atlas project. The method uses distributions of ringing-torecovery 

distances to classify individual species' patterns of movement relative to 

those of other species. The areas between species' cumulative distance distributions 

are treated as inter-species dissimilarities and a one-dimensional map is then constructed 

using multi-dimensional scaling. We have used the method in example analyses 

to show how it could be used to investigate the factors that affect the migratory 

strategies that species adopt, such as diet, territoriality and distribution, and in studies 

of their consequences for demographic parameters such as annual survival and the 

timing of breeding. We have also conducted initial analyses to show how temporal 

changes in the indices could reveal otherwise unmeasured population consequences 

of environmental change and thus have an important application in conservation science. 

Finally, we discuss how our approach to producing indices of migratory tendency 

could be enhanced to reduce the bias that follows from spatial or temporal variation 

in reporting rates and how they could be made more broadly valuable by incorporating 

other data sets and recovery data from other countries. 


08:55 AM - 09:20 AM 

How do geometric constraints influence migration patterns? 

K. Thorup & C. Rahbek 

Null models exclusively invoking geometric constraints have recently been demonstrated 

to provide new insight as to what explains geographic patterns of species 

richness and range size distribution. Analyses of migration patterns have traditionally 

been conducted in the absence of appropriate simulations and analytical models. Here 

we present a null model exclusively invoking geometric constraints and a more 

advanced analytical model incorporating spread along a migration direction that allow 

investigation of the influence of physiographical and physiological boundaries for terrestrial 

taxa, with ocean and sea as geometric constraints, in relation to observed 

patterns of migration. Our models take into account the low recovery probability of terrestrial 

taxa over sea. The null model was not found to explain any of the directional 

27

Dispersal & migration 

variation in the ring-recoveries, but when comparing the distribution of data modeled 

using a simple clock-and-compass model with distributions of ring-recoveries, geometric 

constraints were found to explain up to 22% of the variation in ring-recoveries. 

However, the assumed directional concentrations per step used in the model were 

much higher than expected, and the qualitative fit of the model was rather poor even 

when non-terrestrial sites of recoveries were excluded. 

09:20 AM - 09:45 AM 

Estimating dispersal in birds: taming the unknown 

A.J. van Noordwijk 

A fundamental problem in measuring dispersal is that the observations made are restricted 

by the observations that could possibly be made. This quickly leads to a situation 

where the distribution of observers in time and space determines the observations 

more than the behaviour of the species studied. Although we will never know 

where individuals lost out of sight went to, or where immigrants came from, it is possible 

to obtain estimates of dispersal distances that have been corrected for distance 

specific observation probabilities (Thomson, van Noordwijk & Hagemeijer, J. Appl 

Stat in press). Here I want to explore a complementary approach, we can use the frequency 

of immigrants of unknown origin as an estimate of the unknown part of a 'dispersal 

as a function of distance' curve. Getting a number for how much we don't know 

allows us to get a better perspective on what we do know. In nearly all data sets, we 

will have a short range of distances in which nearly all potential recruits were marked, 

a further range of distances in which a proportion of all potential recruits were marked 

and a furthest range of distances in which no potential recruits were marked. Where 

we can get an estimate of the proportion marked in the middle range, from population 

densities or distribution of suitable habitat, we can derive estimates for the numbers 

of immigrants coming from these distances. Although we cannot know where the rest 

of the immigrants came from, we do have an estimate of "the total area under the 

curve" of the furthest distances. I will apply these ideas to data on natal dispersal in 

Swallows, Blue Tits and Great Tits. This approach will allow us to compare dispersal 

curves between groups (e.g. sexes), but also curves for the same species among 

areas differing in the distribution of suitable habitat (differing in fragmentation). 

09:45 AM - 10:10 AM 

Multistate modeling of movement and philopatry of breeding ross's geese 

K. Drake & R. Alisauskas 

We estimated rates of breeding philopatry and dispersal within the Queen Maud Gulf 

metapopulation of Ross's Geese (Chen rossii) using multistate modeling of neckband 

observations at five breeding colonies, 1999-2002. Philopatry was female-biased, but 

probability of philopatry varied among colonies. Despite probabilities of =0.62 for 

breeding philopatry to a given colony by either sex, this study demonstrates that annual 

movement among breeding colonies was substantial, and greater than previously 

assumed. We provide behavioral evidence (consistent with genetic studies 

which suggested little or no phylogeographic structure in the closely-related lesser 

snow goose, Chen caerulescens) for extensive interconnectedness among breeding 

populations. High likelihood of annual movements among colonies emphasize (1) the 

influence of dispersal on changes in breeding distribution, and (2) provides insight 

about the contributions of emigration and immigration to colony-specific population 

growth rates. Although movement was the primary parameter of interest, estimates of 

male survival from multistate modeling differed from survival estimates from band re- 

28


covery models, whereas female survival estimates were consistent with those from 

band recovery models. We suspect the difference between sexes in survival estimated 

from multistate models was due to higher rates of neckband loss in males compared 

to females, as has been found in numerous other studies, and/or higher rates of 

permanent emigration by males from sampled areas. The importance of violating model 

assumptions depends upon the biological parameter of interest, and whether inference 

is restricted to the sample population or extended to the entire metapopulation. 

10:10 AM - 10:25 AM break 

10:25 AM - 10:45 AM 

Modelling Dolphin Behaviour using Multi-State Models with Time-Varying Covariates 

S. Brooks & R. King 

Effective management is the key to the protection of many endangered species. 

Identification of the primary factors affecting survival will often lead to the introduction 

of strategies to improve survival rates. In this talk, we consider a small population of 

Hector's dolphins located of the coast of New Zealand and the impact that the establishment 

of a seasonal sanctuary has on their survival and migration rates. We use a 

multi-state model to describe the movement of the dolphins around the habitat and 

adopt a Bayesian approach using reversible jump Markov chain Monte to distinguish 

between a wide range of competing models. We also examine the impact of the inclusion 

of time-varying covariates such as catch-effort information and demonstrate 

the added value that these data provide in terms of both model discrimination and parameter 

estimation. In particular, we find a whole class of models that provide a far 

better fit to the data (and therefore better prediction and ultimately better management) 

than those adopted in previous analyses. 

10:45 AM - 11:05 AM 

Spatial distribution of breeding Pied Flycatchers Ficedula hypoleuca in respect to 

their natal sites 

L. Sokolov, N. Chernetsov, V. Kosarev, D. Leoke, M. Markovets, A. Tsvey, A. Shapoval 

Research project on the study of philopatry and dispersal of Pied Flycatchers Ficedula 

hypoleuca has been underway on the Courish Spit on the Baltic since 1980. A 

total of 8006 nestlings were ringed in the Russian part of the Courish Spit. In subsequent 

years, 512 individuals (6,4%) were recaptured in the whole study plot, 300 males 

and 212 females. The aim of this study was to test two alternative hypotheses 

proposed to explain the distribution of breeding Pied Flycatchers in respect to their 

natal site. The breeding area imprinting hypothesis (Löhrl 1959, Berndt & Winkel 

1979, Sokolov et al. 1984) assumes that in spring, migrants try to return to a relatively 

small area, with a radius of one to several km. The other hypothesis forwarded by Vysotsky 

(2001) suggests that all first-year birds and many adults in spring arrive randomly 

to an area much larger than the initial study area. In 2000, some 800 nestboxes 

were added so that their overall number reached 1340 and the length of the 

study plot reached 44 km. To study natal dispersal, we compared the observed distribution 

of natal dispersal distances with the distribution derived from the model assuming 

random arrival to the study area. Significantly more Pied Flycatchers were recaptured 

within several km from the natal site that at larger distance, and this was not 

a function of the control effort. Therefore, it is concluded that the home area imprinted 

29

Dispersal & migration 

by juvenile Pied Flycatchers has a diameter of several kilometres and not several dozens 

of kilometres. 

11:05 AM - 11:25 AM 

Timing and routes of migration based on ring recoveries and randomization methods 

H. Lokki & P. Saurola 

In the first EURING Technical Meeting we recommended randomization tests for 

comparing wintering areas of different species or cohorts of birds on the basis of ring 

recoveries. In this contribution we will use randomization methods when trying to 

answer the following important questions of migration studies: Do the timing and/or 

routes of migration of different species, sexes or age classes differ from each other? 

Can anything statistically sound be inferred from ring recoveries? As a rule, the data 

consists of non-normally distributed recovery records (ti, x(ti), y(ti)) where x(ti) are 

longitudes, y(ti) latitudes of birds recorded in N dates ti, i = 1, 2, … , N. We assume 

that these records constitute a representative sample of places where the birds have 

been found during their lives. At first, in order to sum up the information included in 

the recovery locations in an "average route", we create a time-window and slide it 

through the set of recoveries. In each time-window we sample the records with replacement 

and compute the average of the records in the sample. By repeating this 

procedure many times we obtain an empirical distribution of the average. Then we 

combine these distributions computed for each time-window to get the average route 

of the birds. For comparing two routes, we compute the difference of the averages of 

the records of the two groups in each time-window. Then we compute the sum of the 

differences. In order to find out whether or not the observed sum of differences is statistically 

significant we compute the empirical distribution of the sum of differences as 

follows. In each time-window we randomize many times the inclusion of records into 

the two groups. Then we compute differences of the averages of the randomized 

groups and accordingly the sums of the differences. The ranking of the observed sum 

in the empirical distribution of the sums of differences determines whether or not the 

two routes are similar geographically and in respect of timing. Finally we illustrate the 

methods with data sets extracted from the Finnish recovery database. 

11:25-11:45 AM 

Multistate modeling of brood amalgamation in white-winged scoters (Melanitta fusca 

deglandi) 

Joshua J. Traylor, Ray T. Alisauskas, and F. Patrick Kehoe 

There is a proclivity for female North American waterfowl to lose or abandon offspring 

after hatch. Often these abandoned ducklings are subsequently brooded by foster 

hens giving rise to the phenomena of post-hatch brood amalgamation (PHBA). The 

potential fitness implications that arise from this behavior has brought about considerable 

debate on physiological or ecological motivations for PHBA of young, and 

potential costs and benefits of ducklings brooded in amalgamated broods. Few studies 

have estimated the effects of PHBA on offspring survival and have not followed 

individually marked offspring: most studies have been restricted to studying PHBA 

based on maternal characteristics because only broods have been individually color 

marked. Here, we explore relationships between probabilities of movements from 

maternal to amalgamated broods by using a population of individually marked whitewinged 

scoters (i.e., females and offspring) (Melanitta fusca deglandi) and pertinent 

ecological covariates utilizing multistate modeling. We tested hypotheses about movement 

probabilities and 1) hatch date, 2) brood size, 3) female size and condition, 4) 

30


duckling size and condition, and 5) weather. We defined ducklings as belonging to 

one of 3 social strata: (A) natal mother with no foster ducklings, (B) foster mother and 

conspecific non-siblings, (C) natal mother with conspecific non-siblings and limited 

our study of movement probabilities to two weeks after hatch because this period is 

when most mortality and movement occurred. Thus, we estimated probabilities of 

survival, recapture, and movement for each stratum focusing on covariates for movement. 

We evaluate hypotheses suggested to motivate PHBA (e.g., accidental mixing, 

brood size and brood success, and energetic stress hypotheses) and evaluate 

the influence on duckling survival. Because the importance of predation in motivating 

abandonment or adoption of ducklings remains equivocal, we discuss how duckling 

traits, in a population located near high gull densities, may provide insights into the 

relative importance of predation to PHBA. 

31

Analysis using large-scale ringing data 

Analysis using large-scale ringing data (chairs: Paul Doherty & Stephen Baillie) 


Spatial Point Process Models for Waterfowl Band Recoveries 

J. Andrew Royle 

Estimation of the recovery distribution of bands is an important problem in waterfowl 

management. The recovery distribution is fundamental to characterizing the distribution 

of harvest, derivation of harvest, and geographic variation in reporting rate. The 

classical view of this problem is to stratify the recovery domain into a small number of 

discrete strata (harvest areas). Conventional multinomial band recovery models (e.g., 

"Brownie models") can be used to estimate the stratum-specific recovery probabilities. 

In order to estimate recovery probabilities, a small number of strata are required to 

ensure that sufficient recoveries occur in each stratum. This stratification is subjective 

and limits the ability to characterize variation in recovery rates. The spatial indexing of 

the multinomial cell probabilities suggests an alternative view of this problem, one in 

which the assumed multinomial model contains a very large number of cell probabilities 

that are spatially correlated. This treatment of the problem is equivalent to viewing 

the locations of band recoveries as a spatial point process. Efforts are then 

focused on estimating the spatially varying intensity function describing the relative 

occurrence probability of recoveries. Restatement of the problem in terms of estimating 

a point process intensity function avoids the need for stratification and provides 

for harvest distribution estimates at arbitrary scales. In this talk I give a brief introduction 

to spatial point process models and elaborate on their use for modeling waterfowl 

band recovery data with the goal of characterizing the harvest distribution and geographic 

variation in reporting rate. Data from a recent reward banding initiative are 

analyzed. 


01:55 PM - 02:20 PM 

Population dynamics of the White Stork in the Netherlands: assessing life-history and 

behavioural traits using data collected at large spatial scales 

Blandine Doligez, David L. Thomson & Arie van Noordwijk 

Following the decline of its populations all over Europe after 1945, the White Stork 

Ciconia ciconia has been the object of several successful reintroduction programs. As 

a consequence of the development of these programs, White Stork populations have 

been monitored over large spatial scales. Despite these intense efforts, very few reliable 

estimates of life-history traits of the White Stork are however currently available, 

in particular very little is known about their variation with age. Such general knowledge 

however constitutes a prerequisite for investigating the consequences of conservation 

measures in terms of population biology. Ultimately, this knowledge is required 

to assess population dynamics expected under different scenarios, allowing adequate 

further conservation measures to be taken. 

In the Netherlands, the reintroduction of the White Stork consisted of a captive breeding 

program coupled with intensive supplementary feeding of captive and free-flying 

birds throughout the year. Identification of breeding adults and ringing of young in the 

32


growing population has been performed over the last 20 years throughout the country 

by many volunteers. Using this large-scale and long-term ringing and resighting data 

set, we constructed capture-recapture models to investigate the variation of lifehistory 

traits underlying the dynamics of this population, and the consequences of the 

reintroduction program on these traits. 

In a first step, we fully describe the effects of age, time, cohort and trap-dependence 

on White stork survival and resighting rates, thus providing precise estimates of these 

traits. A gradual increase in both survival and resighting rates with age is described 

for the first time in this long-lived species. Interestingly, survival rate decreases with 

time, paralleling the increase in the proportion of free-flying individuals in the population. 

These results also allow us to account for time and age effects in investigating 

the consequences of supplementary feeding on life-history traits, while keeping capture-recapture 

models tractable in these further analyses. 

The exact causes of the White Stork population decline were not clear, but mortality 

on migration is thought to have increased (exacerbated by the proliferation of highvoltage 

power lines across Europe). Adverse weather on the wintering grounds could 

also have depressed survival. Supplementary feeding may then have ameliorated 

survival either by giving the birds a better body condition prior to migration, or by encouraging 

them to stay behind, surviving the European winter on artificial food supplements. 

In a second step, we thus used detailed data on the location of breeding 

attempts with respect to the food sources over the country (food was provided at the 

breeding stations) to assess the consequences of supplementary feeding on survival, 

resighting rate and migratory behaviour. More specifically, we investigate whether 

survival rate to the next year, resighting rate and migrating probability depend on food 

availability during the breeding season. Food availability was considered to decrease 

with increasing distance from the nest to the nearest breeding station. Further, we 

analyse potential differences in survival rate according to individual migratory status 

(i.e. migratory vs. resident individuals). This status is defined using resighting data 

collected all over Europe and Africa. Preliminary analyses suggest that survival does 

not differ according to food availability, but the probability of migrating decreases with 

increasing food availability. 

Overall, our results emphasize the importance of the large scale population monitoring 

in obtaining life-history trait estimates that will prove useful for managers to make 

efficient decisions for future strategies of conservation for the White stork in the 

Netherlands. Building an integrated demographic population model with the precise 

estimates obtained via our analyses, based on data collected at large spatial scales, 

will indeed eventually allow reliable predictions on the long-term population dynamics 

to be made under different conservation scenarios. 

02:20 PM - 02:45 PM 

Estimating correlates of survival rates from nationally coordinated ringing data on 

owls in Finland 

Pertti Saurola & Charles Francis 

Since 1974, bird ringers in Finland have been encouraged to ring both nestlings and 

adults of many species of birds of prey, especially owls. This coordinated effort, involving 

several hundred ringers, now results in more than 30,000 potential nest sites for 

owls being checked annually, and has led to over 200,000 owls being ringed in Finland 

up until 2002. Many of these owls are subsequently recaptured as breeding 

adults, by the same or different ringers, while others are recovered dead by the general 

public. All of the ringing and encounter data, as well as many biometric data are 

centrally computerized, allowing for large-scale analyses of geographic and temporal 

33

Analysis using large-scale ringing data 

variation in such parameters as survival and dispersal rates. These can then be related 

to external covariates, such as winter snow depth and prey abundance, as well as 

individual covariates such as biometrics. In this presentation, we illustrate both the 

potential and the challenges of working with these types of data, using data from 

9000 recaptures and recoveries from over 30,000 Tawny Owls ringed as nestlings or 

adults. Average survival rates of this species increased from 30% over the first year 

post-fledging, to 60% in the second year, and 74% in subsequent years, but varied 

among years in response to winter severity and vole abundance. Survival rates were 

lowest after winters with deep snow accumulations, and in years when vole populations 

crashed. Similar variation occurred in dispersal, with greater dispersal distances 

in years when voles crashed just after the breeding season. Combining data from 

many different contributors to the national scheme, in addition to greatly increasing 

the sample size, allows one to address questions that could not be addressed using 

data from a single study area. For example, it is possible to examine the impact of 

dispersal on survival rate estimates, and the relative influence of survival and dispersal 

on population dynamics of this species. It is also possible to test for geographic 

variation, such as north-south clines in demographic parameters. However, combining 

data from many sources also presents some challenges, such as dealing with 

geographic variation in the intensity of coverage. 

02:45 PM - 03:10 PM 

Population dynamic and temporal variation in recruitment and survival of 14 common 

passerines 

Romain Julliard 

At large spatial scale, variation of breeding abundance of a given species results from 

variation of survival of established adults and variation of recruitment of new individuals. 

Which of these two parameters is the most variable and the best predictor of 

variation of abundance is a central issue of population dynamic (Saether et al. Science 

2002). Yet the problem looks simple, an important pitfalls is the difficulty to obtain 

independent estimates of the different parameters. 

In many countries, large scale breeding bird monitoring based on counts are established 

to monitor population abundance at the scale of the country. They are often 

coupled with standardized mist-netting scheme set up for the specific purpose of monitoring 

demographic parameters for the same populations. From these capturerecapture 

data, variation of survival and recruitment (using Pradel's approach) may be 

estimated, independently of population size variation estimated from count survey. In 

addition, captures of young individuals are used to estimate variation of reproductive 

success (using the young adult ratio as an index). Because survival and recruitment 

are not independently estimated, their temporal variation may not be compared directly. 

Rather, the amount of temporal variation as well as the synchronization of 

temporal variation across sites, and how temporal variation were correlated with 

changes in abundance and productivity index were compared between survival and 

recruitment for a given species. This was done for 14 species with high enough recapture 

rate to estimate demographic parameters. The data comes from the French 

common bird monitoring scheme settled in 1989, and included capture-recapture data 

from 20 to 40 sites per year for 13 years (about 35,000 captures). 

34


03:10 PM - 03:35 PM 

Population dynamics of small mammals at three spatial scales: a 5-years study of 122 

trapping grids 

Nigel Yoccoz & Rolf Ims 

Small mammals are known to exhibit a large variety of dynamics, both in time (multiannual 

cycles vs. seasonal variation only) and space (regional synchrony, travelling 

waves). Small mammals have therefore been the focus of a large number of studies, 

that used mostly trapping indices. These studies do not therefore take into account 

differences in trappability that may indeed be confounded with the phenomenon we 

want to explain (for example, that trappability is lower in the increase than in the decrease 

phase of the cycle). In this paper, we use our own study investigating population 

dynamics and demography of small mammals at three spatial scales (0.1, 10 and 

100 kms) to address some methodological and practical issues. The study is based 

on an unusually large sampling effort (122 grids trapped twice a year in 5 years, in 

spring and fall, each trapping session being 4 days long), with more than 10,000 individuals 

captured of three different species. The main methodological issue we 

address is: what is the consequence of using a too simple/complex model for capture 

probabilities on the estimates of spatial patterns in population densities, growth rates 

and demography, and how should we model these probabilities in practice. We explore 

in particular the use of Bayesian models that include hierarchical random effects. 

35

Abundance estimation & conservation biology 

Abundance estimation & conservation biology 

(chairs: Jim Nichols & Darryl MacKenzie) 


Hierarchical mark-recapture models: A framework for inference about demographic 

processes 

Bill Link & Richard Barker 

The development of sophisticated mark-recapture models over the last four decades has 

provided fundamental tools for the study of wildlife populations, allowing reliable inference 

about population sizes and demographic rates based on clearly formulated models for the 

sampling processes. Mark-recapture models are now routinely described by large numbers 

of parameters. These large models provide the next challenge to wildlife modelers: the extraction 

of signal from noise in collections of parameters. 

Pattern among parameters can be described by strong, deterministic relations (as in ultrastructural 

models) but is more flexibly and credibly modeled using weaker, stochastic relations. 

Trend in survival rates is not likely to be manifest by a sequence of values falling precisely 

on a given parametric curve; rather, if we could somehow know the true values, we 

might anticipate a regression relation between parameters and explanatory variables, in 

which true value = signal plus noise. 

Hierarchical models provide the appropriate framework for inference about collections of related 

parameters. Instead of regarding parameters as fixed but unknown quantities, we regard 

them as realizations of stochastic processes governed by hyperparameters. Inference 

about demographic processes is based on investigation of these hyperparameters. 

We describe analysis of capture-recapture data from an open population based on hierarchical 

extensions of the Cormack-Jolly-Seber model. In addition to recaptures of marked animals, 

we model first captures of animals and losses on capture, allowing estimation of survival 

and birth rates. We present analyses for covariation between survival and birth rates. For 

example, years that are poor for survival may also be poor years for reproduction. In this 

case we would expect the survival parameters and birth rate parameters to be related. 

The hierarchical approach that we describe provides the framework needed for exploiting 

structural relationships between parameters, improving individual estimates by considering 

the parameters in context of groups of related parameters. In addition it provides a mechanism 

for summarizing a large number of parameters in terms of hyperparameters that characterize 

higher-order distributions. 

We believe that hierarchical models should represent an entirely natural mode of thinking for 

biologists. Were it possible for biologists to dispense with mark-recapture estimation and instead 

obtain exact values of parameters they would be unlikely to be satisfied with the parameters 

as a summary of the population processes that generated them. Instead we would 

expect them to treat the parameter values as data in some type of analysis. 

Standard methods for analyzing mark-recapture data are poorly suited to hierarchical modeling. 

Analysis of deviance based on restricted models, as described by Lebreton et al. (1992), 

can be used to explore deterministic relationships between parameters. However, methods 

for fitting random effects models that allow for stochastic higher-order relationships between 

parameters are relatively crude. For example, the variance components procedure incorporated 

in program MARK uses method of moments, a procedure that can result in inadmissible 

estimates and that relies on asymptotic sampling properties for inference. We describe 

flat-prior Bayesian analyses using Markov chain Monte Carlo as a useful solution to these 

problems. 

36



04:30 PM - 04:55 PM 

Components of population growth rate for white-winged scoters in Saskatchewan 

Ray T. Alisauskas, Joshua J. Traylor, Cindy J. Swoboda, & F. Patrick Kehoe 

There have been considerable declines in breeding range and abundance of White-winged 

scoters (Melanitta fusca deglandi) in northwestern North America. Consequently, we began 

studying population biology of scoters in 2000 at Redberry, Saskatchewan, Canada, where 

long-term had been done previously. Krementz et al (1997) estimated apparent survival rates 

for this population using capture-recapture of birds from 1975-1985. Since then, there 

have been developments in the direct estimation of population growth rate using reversetime, 

capture-recapture of animals. We used the same capture histories used by Krementz 

et al (1997) to directly estimate survival, seniority and capture probabilities. Seniority probability 

is a parameter useful for understanding the proportion of population growth rate composed 

of survival, i.e., seniority = survival/population growth. Thus, if seniority and survival probabilities 

are estimable, then population growth rate, lambda, can be estimated by substitution. 

Values of seniority approaching 1.0 suggest that there is very little recruitment contributing 

to population growth rate. We used Program Mark to compare 9 candidate models for 

estimating survival, seniority and capture probabilities; the 9 models were combinations of 

different time constraints (time-specific, time trend and time-invariant) imposed on each of 

the three latent parameters. We then used model-averaging to derive a annual estimates of 

each of the three latent parameters. Estimates of population size, derived using estimates of 

capture probability, declined over the course of the study. Calculation of population growth 

rate suggested that this population was in decline (lambda = 0.78±0.02 SE) over the 11 years 

of previous investigation; because survival probability was constant, and seniority was 

very close to one, the results further suggest that population declines were the result of virtually 

no local recruitment (neither through immigration nor local production of young). These 

historical findings provide context for current local population dynamics of White-winged 

scoters at Redberry Lake studied more recently (since 2000); they also have conservation 

implications for scoters over their range in North America. 

04:55 PM - 05:20 PM 

Statistical Aspects of using genetic markers for individual identification in capturerecapture 

studies 

Paul Lukacs & Ken Burnham 

We describe analysis issues in which genetic markers differ from bird ringing in applicability 

for use in capture-recapture studies. Identification of individual birds based on genetic 

markers, often mircosatellites, involves some uncertainty. The amount of uncertainty depends 

on many factors including the number of loci used in the genetic analysis, number of 

alleles per locus, allele frequencies, degree of relatedness of individuals, etc. The uncertainty 

can be quantified based on the probability of two genotypes being identical (matching) given 

they belong to different individuals. Given the match probabilities, all possible capture histories 

and their probability of occurrence can be determined. We present a likelihood based 

method for the analysis of capture-recapture data using genetic markers as individual identification 

which incorporates potential uncertainty in identification. The capture histories and 

their associated probabilities of occurrence are then used in the capture-recapture analysis 

to estimate parameters such as population size or survival. This method also applies to situations 

where physical marks of birds are only partially read. 

37

Abundance estimation and conservation biology 

05:20 PM - 05:45 PM 

Occupancy as a surrogate for abundance estimation 

Darryl MacKenzie & Jim Nichols 

In many monitoring programmes it may be prohibitively expensive to estimate the actual 

abundance of a bird species in a defined area, particularly at large spatial scales, or where 

birds occur at very low densities. Often it may be appropriate to consider the proportion of 

area occupied by the species as an alternative state variable. However, as with abundance 

estimation, issues of detectability must be taken into account in order to make accurate inferences: 

the non-detection of the species does not imply the species in genuinely absent. Here 

we review some recent modelling developments that permit unbiased estimation of the 

proportion of area occupied, colonization and local extinction probabilities; allow for unequal 

sampling effort; and enable covariate information on sampling locations to be incorporated. 

We also describe how these models could be extended to incorporate information from marked 

individuals, which would enable finer questions of population dynamics (such as turnover 

rate of nest sites by specific breeding pairs) to be addressed. We believe these models may 

be very applicable to a wide range of bird species, and may also be useful for investigating 

questions about habitat quality. For example, the species is more likely to have higher local 

extinction probabilities, or higher turnover rates of specific breeding pairs, in poor quality habitats. 

05:45 PM - 06:10 PM 

Modeling population dynamics with ringing, age ratio, and breeding population sample 

surveys 

Mark Otto 

We modeled mid-continent North American mallard population dynamics while accounting 

for the sampling error in each of the demographic time series. We used survey sample estimates 

and covariances from all vital statistics: hunting and non-hunting survival from the 

USGS Bird Banding Laboratory's ringing data, reproduction indexed by the ratio of juveniles 

to adults in the USFWS Harvest Survey Section's parts collection survey, and breeding population 

data from the USFWS May aerial waterfowl breeding ground survey. Modeling all 

surveys with their sampling error simultaneously was a multivariate approach in which all the 

component demograpic series were treated equally to obtain the best estimates of the population 

process over time. 

The model was a two-level hierarchical mixed effects model. The secondary level used a 

Leslie-matrix population model to combine the series and nonstationary signal extraction to 

obtain true value. The true values were compromises between the population dynamics and 

the survey estimates. The moethod change less precise survey estimates to minimize the 

differernces between the breeding population true values and their predictions, given past 

data. 

The primary hierarchical level consisted of random effects, the survey sample estimates and 

their covariances. The ringing data estimates had a multinomial sampling distribution where 

the hunting and non-hunting survival rates varied over time. The reproduction and breeding 

population estimates had normal sampling distributions with known and fixed variance estimates. 

Having data from all vital statistics exposed inconsistencies among the series and allowed 

us to estimate biases. 

38


Population dynamics (chairs: Evan Cooch & André Dhondt) 


Beyond survival estimation: mark-recapture, matrix population models, and population 

dynamics 

Hal Caswell 

Survival probability is of interest primarily as a component of population dynamics. Only 

when survival estimates are included in a demographic model can their population implications 

be calculated. Survival describes the transition between living and dead. 

Biologically important as this transition is, it is only of many transitions in the life cycle. 

Others include transitions between immature and mature, unmated and mated, larva 

and adult, small and large, and location x and location y. The demographic consequences 

of these transitions can be captured by matrix population models, and such models 

provide a natural link connecting multi-stage mark-recapture methods and population 

dynamics. 

I will explore some of those connections in this talk, with examples taken from an ongoing 

analysis of the endangered North Atlantic right whale (Eubalaena glacialis). Formulating 

problems in terms of a matrix population model provides an easy way to compute 

the likelihood of capture histories. It extends the list of demographic parameters 

for which mark-recapture methods provide maximum likelihood estimates to include 

population growth rate, the sensitivity and elasticity of population growth rate, the net 

reproductive rate, generation time, and measures of transient dynamics. In the future, 

multi-stage mark-recapture methods, linked to matrix population models, will become 

an increasingly important part of demography. 


08:55 AM - 09:15 AM 

Examining competing hypotheses for mechanisms of Cassin's Auklet population trajectory 

Nadav Nur, Derek Lee & Bill Sydeman 

Cassins Auklet (Ptycoramphus aleuticus) numbers are considered to be at their lowest 

point in history throughout their range, and breeding populations are currently in decline 

at Triangle Island, British Columbia, and Southeast Farallon Island (SEFI), California, 

the only large colonies with long term data available. Using data from 21 years of 

mark-recapture, breeding performance, diet, chick growth rates, and oceanographic 

conditions at a Cassin's Auklet colony on SEFI, we examined interannual and decadal 

scales of variation in the population's vital rates and how they can explain population 

trends for this isolated population. Thanks to the depth and breadth of our data on this 

population, we were able to examine variation on multiple time scales, and with the aim 

of uncovering processes at work within this population, both ultimate factors (e.g., related 

to climate change and physical forcing of the marine environment) and proximal 

factors (reflected in indices of prey availability, indices of predation, and variation in 

chick growth rates). 

39

Population dynamics 

09:15 AM - 09:35 AM 

Constrained influence of cyclic environmental variability on population dynamics of Antarctic 

predators. 

Stephanie Jenouvrier, Cristophe Barbraud, Bernard Cazelles & Henri Weimerskirch 

Ecological and population processes are affected by climatic fluctuations, and top predators 

such as seabirds can provide an integrative view on the consequences of environmental variability 

on ecosystems. Over the Southern Ocean cyclic fluctuations of physical environment 

such as pression, sea surface temperature or sea ice are linked to large scale climatic oscillations 

like the southern oscillation. White and Peterson (1996) described also an Antarctic 

Circumpolar Wave (ACW) that propagates eastward coupled warm anomalies of sea surface 

temperature and sea ice extent with a period of 4-5 years. In this study we examine the cyclic 

dynamics of several seabirds population (southern fulmar, emperor penguin and snow petrel) 

in Antarctica, to evaluate the impact of environmental variability on the demographic parameters 

and to model the dynamics of populations. 

Population dynamics of southern fulmar, emperor penguin and snow petrel show cyclic fluctuations 

with a period of 2-4 years. By using wavelet analysis we pointed out that southern 

fulmar showed a period of 2-4 years after the eighties, whereas snow petrel and emperor 

penguins showed a similar period before the eighties. A regime shift occurred during the 

eighties and appear to synchronise the populations of these 3 species, which were not synchronised 

during the greater part of the forty years study. 

Long term monitoring of marked individuals of the 3 species enabled us to estimate demographics 

parameters such as breeding success or survival with capture-recapture analysis. 

We studied the influence of several environmental parameters (sea surface temperature, sea 

ice) on these life history traits and showed that species have constrated responses to climatic 

variations. For example, the survival rate of emperor penguins and southern fulmars increases 

when sea ice extent increases, while the survival rate of snow petrels decreases. 

Using these long- term series on demographic parameters together with their relationships 

with climatic factors, we model the population dynamics with Leslie matrices integrating the 

cyclicity in the variation of the demographics parameters. These modelled population trajectories 

were then compared with the observed cyclicity in the population size surveys. 

09:35 AM - 09:55 AM 

Estimating the effects of fluctuating prey on demographic parameters of tawny owls 

Charles Francis & Pertti Saurola 

Breeding populations and demographic parameters of many species of owls have been 

shown to vary in response to fluctuating prey populations. For example, breeding success 

and emigration patterns of several European owl species fluctuate in response to the 3-4 year 

cycle of many rodents (Saurola 1997), while breeding propensity, clutch size, nesting success, 

movement patterns and survival rates of Great Horned Owls vary in relation to the ~10- 

year cycle in Snowshoe Hare abundance (Houston & Francis 1995). However, little is known 

of the relative importance of these different factors on the population dynamics of owls. 

Since 1974, over 30,000 individuals of both young and breeding adult Tawny Owls have been 

ringed in Finland, generating over 9,000 live recaptures and dead recoveries. In addition, 

many territories are monitored each year, providing information on breeding propensity, 

clutch size, and nesting success. We used mark-recapture-recovery models to estimate variation 

in survival rates in relation to lemming cycles, as well as the severity of winter weather. 

By combining recapture and recovery data, we can estimate the relative importance of emigration 

and mortality on variation in apparent survival rates. We then use these estimates, 

together with productivity and nesting propensity data in a stage-based matrix model, to de- 

40


termine the relative influence of variation in these different demographic parameters on variation 

in population size of the species. Finally, we compare the model predictions with direct 

estimates of changes in population size derived from censuses and mark-recapture analyses. 

09:55 AM - 10:15 AM 

Population models in greater snow geese: a comparison of different approaches 

Gilles Gauthier & Jean-Dominique Lebreton 

Population model is a powerful tool to guide decision making when managing wildlife populations. 

We will compare different modeling approaches that we used to evaluate the effect of increased 

harvest on the population growth of Greater Snow Geese. For this population, we benefit 

of two unique datasets. On one hand, fecundity and survival data come from a long-term 

capture-recapture study conducted since 1990 at the breeding colony of Bylot Island in the 

Canadian Arctic. On the other hand, accurate estimates of the total size of the population come 

from an annual spring photo inventory conducted since 1970 in southern Quebec. Harvest 

data can be obtained from the national hunter surveys. In a first approach, we included environmental 

stochasticity in a matrix projection model by simulating good, average and bad years 

to account for the large inter-annual variation in fecundity and first-year survival, a common 

feature of birds nesting in the Arctic. However, caution must be used with this approach because 

different stochastic growth rates can be obtained according to the model formulation used 

(post-breeding vs pre-breeding census) when covariance among matrix elements is present, 

as this was the case here. A second approach that we developed is based on the functional 

relationships between generation time and elasticity on one hand, and harvest rate and survival 

on the other hand. Generation time was obtained from the mean transition matrix based on 

the observed proportion of good, average and bad years between 1985-98. The model assumes 

that hunting mortality is additive to natural mortality, for which we have good evidence. 

This yielded a simple formula that can predict changes in lambda as a function of changes in 

harvest. A third, and potentially more robust approach, consists in combining different sources 

of information in the same model, that is demographic data (i.e. transition matrix) and census 

data (i.e. annual survey). The Kalman Filter is a technique that precisely allows that. The 

advantage of this approach is that it attempts to minimize both uncertainties associated with 

the survey and demographic parameters based on the variance of each estimate. We will use 

the case of the greater snow goose to illustrate this newest approach. 

10:15 AM - 10:35 AM 

A Bayesian Approach to Combining Animal Abundance and Demographic Data 

Steven Brooks, Ruth King & Byron Morgan 

In studies of wild animals, one frequently encounters both census and mark-recapture recovery 

data. In this talk we consider a Bayesian analysis of joint ring-recovery and census data using 

a state-space model allowing for the fact that not all members of the population are directly observable. 

We then impose a Leslie-matrix-based model on the true population counts describing 

the natural birth-death and age transition processes. 

We use Markov chain Monte Carlo (MCMC) methods to perform the analysis, removing the 

need to use the Kalman filter and thereby allowing us to avoid the need for the potentially restrictive 

normality assumptions commonly assumed for analyses of this sort. We illustrate our 

approach on two important British bird species, the lapwing and the heron. In both cases, we 

introduce additional time-varying covariates such as the number of frost days each year in order 

to better explain the annual variability in the population. We use reversible jump MCMC 

to discriminate between alternative models describing the underlying population dynamics. 

41

Honour speaker 

Honour speaker 

11:00 AM - 12:00 AM 

Quantitative Methods for the Study and Management of Avian Populations: Evolution, 

Current State and Prognosis 

James D. Nichols 

The tradition of the EURING meetings has been to focus on quantitative methods for 

use in the study and management of bird populations. The EURING meetings have 

emphasized estimation methods, an important component of a scientific process that 

also includes development of models of ecological processes and, in the case of management 

and conservation, development of decision-theoretic methods. I will 

discuss the evolution, current state and prognosis for 3 classes of estimation methods 

(band-recovery analysis, capture-recapture analysis, and distance sampling), as well 

as for two more general issues involving estimation methodology, model selection 

and dissemination of new data-analytic methods to biological users. In order to emphasize 

the role of estimation methods as components of larger processes requiring 

use of other classes of methods, I also discuss the evolution of process modeling and 

decision-theoretic methods used in the study and management of avian populations. 

The EURING meetings and their attendees have played an important role in the development 

of quantitative methods. I illustrate this point by claiming that the rapid 

evolution over the past few decades of each of the discussed topics can be traced 

back to seminal work by a EURING contributor who retired recently. In fact the sensitivity 

of the growth rate of methodological knowledge to the contributions of individual 

researchers has been extremely large for this person. Retrospective considerations of 

the work of this individual lead to a posteriori hypotheses about factors associated 

with "high-sensitivity" investigators and teams. Although experimental tests of these 

hypotheses are unlikely, some inference may be possible based on predictions and 

eventual contributions of investigators using different approaches to the conduct of 

science and management. 

42


Poster abstracts (organizer: Charles Francis) 

Population dynamics of radio-marked individuals: A case study of wild turkey in the 

Virginias 

Russell Alpizar-Jara, Elizabeth N. Brooks, Kenneth H. Pollock, David E. Steffen, James C. 

Pack & Gary W. Norman 

Wild turkey (Meleagris gallopavo silvestris) hunting is a popular sport in the United 

States, with broad regional and national significance. A 2-sex Leslie-type matrix 

model was developed to understand the effect of hunting on the dynamics of wild turkey 

populations in Virginia and West Virginia (Alpizar-Jara et al. 2001). A unique aspect 

of this model is that it incorporates recruitment and survival parameters derived 

from a large-scale radio telemetry study. About 1543 hens were radio-tagged and 

monitored weekly during a period of 5 years to determine causes of mortality and to 

estimate survival and recruitment rates (Pack et al. 1999, Norman et al. 2001). The 

model provides projections of population size and harvest for Spring-Summer and 

Fall-Winter periods, as well as information about population growth rate and age/sex 

structure. Deterministic and stochastic versions of the model are available online 

(www4.stat.ncsu.edu/~pollock/turkey/turkeymod.html) to facilitate evaluation of important 

hypotheses concerning turkey populations with similar structural and biological 

features (Brooks et al. 2002). This should enable managers to address questions 

related to setting seasonal harvest levels, to make comparisons among various 

strategies for rehabilitating declining populations and to determine appropriate harvest 

regulations. We illustrate how the model can be used to explore a series of 

hunting strategies and to evaluate the consequences of process variation in the dynamics 

of the population. Of particular interest is the impact of cause-specific mortality 

(illegal, natural and hunting) and recruitment components in the population growth 

rate. Our analyses show that fall hunting has the strongest negative effect on the 

population growth rate, and that the proportion of males in the population was more 

sensitive to fall rather than spring hunting. 

Senescence in the great tit 

Eduardo J. Belda & Markku Orell 

Theories of ageing propose that ageing evolves as the necessary cost of processes 

increasing early reproductive success or because of weaker selection against lateacting 

mutations. They predict that the rate of ageing should increase with the increase 

in the intensity of age and condition independent (i.e. extrinsic) mortality. In 

this study we used 488 individual capturerecapture histories from a 20-year study of 

the great tit (Parus major) breeding at Oulu, northern Finland (65o N). We compared 

our age-specific survival pattern with results from a great tit population in southern 

England (52o N). Our observation of lower survival probability from age 1 to age 2 in 

northern Finland than in southern England suggests higher extrinsic mortality in the 

North. We found that in Northern Finland, the survival probability improved at age 1, 

remained constant at age 2 and age 3 and declined at age 4. For great tits in southern 

England McCleery et al. (1996) found senescence in the survival probability but 

the effect appeared at an older age (5 years). Thus, our results support the prediction 

from evolutionary theories that the rate of ageing should increase with the intensity of 

the risk of extrinsic mortality. 

43

Poster abstracts 

Daily survival probabilities of adult and juvenile cliff swallows vary with colony size 

and ectoparasite load 

Charles R. Brown & Mary Bomberger Brown 

One potential benefit of colonial nesting in birds is increased survival of breeding 

adults and/or fledged juveniles due to improved predator avoidance or better food 

finding in groups. The cliff swallow (Petrochelidon pyrrhonota) of western North 

America breeds in colonies ranging from a few pairs to sometimes thousands at a 

single site, and a number of separate costs and benefits of coloniality have been 

identified in this species. Yet it is unknown how these interact to affect fitness components 

such as survival of birds in different sized groups. Using mark-recapture data 

from 213 different colonies from 1983-2002 and a total sample size of 129,995 adults 

and 19,584 juveniles, we estimated within-season daily survival probabilities of adult 

and recently fledged juvenile cliff swallows at breeding colonies ranging in size from 

15 to 3000 active nests at our study site in southwestern Nebraska, USA. Because 

transient swallows were present at some colonies, we used “age”-dependent models 

to estimate only the survival of resident birds. Some colonies had been fumigated to 

remove ectoparasitic bugs, and these allowed us to investigate the effect of ectoparasitism 

on daily survival. Across all years, withinseason daily survival probabilities increased 

as colony size increased for both adults and juveniles in both fumigated and 

non-fumigated colonies, although the pattern was statistically stronger for fumigated 

colonies, and there was annual variation in the pattern among nonfumigated colonies. 

Birds in the parasite-free colonies had significantly higher daily survival probabilities, 

on average, than those occupying naturally infested colonies. The results reveal a 

previously unknown advantage of colonial nesting in this species, document another 

apparent cost of ectoparasitism, and represent the first measurement of daily adult 

survival during the breeding season in relation to colony size for any bird. 

Applying band recovery models to an evaluation of the demographic impacts of exceptional 

conservation measures 

Anna M. Calvert & Gilles Gauthier 

In response to extremely rapid population growth in recent years, unusual conservation 

measures were implemented for greater snow geese in spring 1999 with the intention 

of reducing adult survival and ultimately stabilising abundance. These included 

the liberalisation of existing sport hunt regulations in autumn and winter and, 

most significantly, the initiation of a spring hunting season on staging grounds in 

Québec. Long-term banding and recovery data covering the periods before and after 

regulation changes provided us with an opportunity to evaluate the effects of these 

exceptional conservation measures, serving as a kind of large-scale natural experiment. 

We use band recovery models in MARK based on a comprehensive dataset of 

hunter-recovered birds (approximately 4,000 recoveries of 42,000 geese banded from 

1990- 2001) to determine how the changes in harvest regulations may have differentially 

affected age groups and sexes, and to compare the magnitude of these effects 

to base-line temporal variability in survival and recovery rates. Models incorporating 

annual harvest rate (determined independently from hunter surveys) as an external 

covariate are also tested in order to determine whether harvest trends interpreted 

from survey data accurately reflect demographic impacts. We use a simplified reporting 

rate estimate based on the recovery rate of radiocollared birds (assumed to be 

100%) and on literature values to calculate annual hunting mortality rates from recovery 

rates for each component of the population, for comparison with changes in sport 

harvest regulations. As expected, we found an increase in recovery rates for all agesex 

groups after the initiation of conservation measures, but did not note any correspondingdecline 

in survival rates, though this may be due to the low power of detec- 

44


tion implicit in the use of band-recovery data. Given migratory route changes observed 

in the population during the period of abundance increase, we also explore 

different methodological options available for an evaluation of the variation in survival 

and recovery rates due to kill-location and season, and discuss the additional information 

but also the biases that arise from these analyses. 

The Possible Effects of Contaminants on the Survival, Breeding Dispersal, and Natal 

Dispersal of Red-tailed Tropicbirds Nesting on Johnston Atoll 

Paul F. Doherty, Jr., E. A. Schreiber & Gary A. Schenk 

Annual survival and dispersal rates of adult and juvenile red-tailed tropicbirds were 

examined in connection with exposure to heavy metals. The incineration of a U.S. 

stockpile of chemical weapons stored at Johnston Atoll, in the central Pacific Ocean, 

exposed tropicbirds nesting in the vicinity of the plant to increased levels of human 

disturbance, smoke stack emissions and potential leaks. Birds nesting in this site 

(downwind of the plant) were compared to those nesting in a reference site (upwind of 

the plant) with less human disturbance, and no exposure to smoke stack emissions or 

other potential incineration emissions. We did not find any effect of the weapons incineration 

on survival of adults or juveniles between the two sites. Adult breeding dispersal 

rates did not differ between the sites but we did find differences in the 

agespecific natal dispersal rates. Birds fledged from downwind areas were less likely 

to return to their natal area to nest and more likely to immigrate to the upwind area 

than vice-versa. This asymmetry in natal dispersal rates could be caused by many 

factors, but we believed it is most likely due to differing vegetation densities and disturbance 

regimes. These results have implications for vegetation management in relation 

to tropicbird nest success and population size. 

Assessing and Correcting for Effects of Bias on Manatee Aerial Survey Counts at the 

TECO Big Bend Power plant in Tampa Bay, Florida 

Holly H. Edwards, Bruce B. Ackerman, John E. Reynolds, James A. Powell & Kenneth H. 

Pollock 

In winter, Florida manatees (Trichechus manatus latirostris) aggregate in the warmwater 

outflows of power plants and natural springs seeking refuge from colder ambient 

waters temperatures. Aerial surveys that cover manatee habitat in small, welldefined 

areas (usually at winter aggregation sites) are useful for obtaining minimum 

population estimates and for assessing effects of cold weather on manatees. However, 

obtaining accurate population estimates from these surveys are problematic because 

imperfect detection is not taken into account. To identify how environmental 

conditions affect manatee aerial detection probability, an intensive study was conducted 

during winters 1999-2000 through 2002-2003 at the Tampa Electric Company’s 

(TECO) Big Bend power plant discharge canal in Tampa Bay, Florida. Flights 

were flown on mostly consecutive days (4-10 days) following the passage of cold 

fronts. To estimate components of detection probability we marked manatees by attaching 

colored flags to their tail-stocks. Marked animals helped: (1) determine the 

overall probability of detecting manatees from the air via the use of marked animals 

and aerial observers (p); (2) determine the probability of detecting an animal, given 

that it is available to the observer using ground (p2) and aerial observers (p1); (3) 

determine the probability, from information in 1 and 2 EURING 2003 – Poster Session 

4 above, of a manatee being available for observation, (pa). Overall probability of 

detecting manatees for 6 surveys ranged from p = 0.41 to 0.83 and the probability of 

detection, given available, for air and ground observers ranged from p1 = 0.61 to 0.75 

and p2 = 0.60 to 0.76 respectively. The probability of a manatee being available to the 

45


observer was pa = 0.46 to 0.76. We will use these detection probabilities to develop 

better population estimates for manatees. 

Predicting the effect of shellfish fishery on mortality in Oystercatchers Haematopus 

ostralegus 

B.J. Ens & C. Rappoldt 

In the Netherlands, mechanized fishing for shellfish occurs in several intertidal areas, 

whose prime function is that of being a nature reserve. The fishery is hotly debated, 

since conservationists are convinced that declines observed in Oystercatcher populations 

are due to the fishermen taking away the food of the birds. To address this dispute, 

we developed a model that predicts how the Oystercatchers will deplete the 

stock of shellfish in the course of the winter. For the Oosterschelde, the first area for 

which we tested the model, the predicted amount of shellfish left at the end of winter 

coincided closely with the stock estimate of the fishery institute. The model also calculates 

a measure which we call food stress and which indicates the difficulty the 

birds experience in meeting their daily energy needs. The model does not distinguish 

between individuals, because we wanted to stay as close to the available measurements 

as possible. However, the food stress in a given year can be related to the 

mortality of the birds. We estimated mortality using ringing recoveries and found a 

close correlation between mortality and food stress. The higher the food stress during 

a given winter, the higher the overwinter mortality of the birds. The highest stress and 

overwinter mortality occurred during very cold winters when the intertidal mudflats 

froze over and the birds could not reach their food. We were also able to estimate the 

extra food stress caused by the fishery. It turns out that initially the policy of food reservation 

for the birds, which forbids shellfish fishery when shellfish stocks are low, did 

not provide sufficient guarantees for the birds. Recently, minimum stocks were set 

higher and preliminary calculations suggest that the new levels do provide sufficient 

guarantees for the birds. 

Estimating Overdispersion using a Conditional Bootstrap 

David Fletcher & Richard Barker 

For large samples, the deviance of a mark-recapture model has a chi-squared distribution 

(when the model is correct). This result has lead to the standard measure of 

overdispersion known as c-hat, which is the deviance divided by its degrees of freedom. 

In order to avoid using this large-sample approximation, Program Mark now has 

a parametric bootstrap procedure for estimating c-hat. This procedure involves generating 

a large number of bootstrap samples and estimating c-hat by dividing the observed 

deviance by the mean of the deviances from these samples. We suggest that 

this procedure should be made conditional, in the sense that we only consider bootstrap 

samples with the same minimal sufficient statistics as the original data. We present 

the results of a simulation study comparing the conditional and unconditional 

bootstrap procedures, focussing in particular on the coverage rates of confidence intervals 

for survival rates obtained using the corresponding value of c-hat. 

What is the significance of the F parameter in combined models of live and dead encounters: 

site fidelity, colour-ring retention or correction for heterogeneity? 

Morten Frederiksen 

In the framework of the combined model of live and dead encounters of marked animals 

developed by Burnham (1993), the parameter denoted F is defined as an esti- 

46


mate of fidelity to the study area. It is obvious that if live encounters concern colourmarked 

individuals observed at a distance, whereas dead encounters occur irrespective 

of the presence of a colour-ring, F will estimate the compound probability of retaining 

the colour marker and remaining within the study area. However, in many 

studies of this type, e.g. of geese, resighting probability will not be homogeneous 

within the study area, resulting in sometimes pronounced resighting heterogeneity. If 

some animals show fidelity to locations where they are unlikely to be observed, this 

heterogeneity takes the form of “trap-dependence”, which in pure live-encounter 

studies can cause negative bias in estimated survival. How does resighting heterogeneity 

affect survival estimates under Burnham’s combined model? Through a simulation 

study I showed that as long as dead encounters occur independently of how 

likely an individual is to be observed alive, Burnham’s model provides unbiased estimates 

of survival. Even in simulations with no permanent emigration from the study 

area, F is estimated as less than 1 if resightings are heterogeneous. This means that 

whenever resighting heterogeneity occurs in a study population, the F parameter becomes 

very difficult to interpret and should probably most often be regarded as a correction 

factor which allows unbiased estimation of survival. 

Estimating true age-dependence in survival from long-lived birds only resighted as 

adults 

Morten Frederiksen, Michael P. Harris & Sarah Wanless 

In many long-lived birds with delayed recruitment, such as most seabirds, most information 

on survival comes from adult breeders. Pre-breeders are often more or less 

unobservable, and even though a proportion of marked adults may have been 

marked as chicks and therefore be of known age, the estimation of age-specific survival 

is complicated by the absence of observations during the first years of life. As a 

result, many published studies have used “time since first observation as breeder” as 

a surrogate for true age, making tests for e.g. senescent declines in survival less 

powerful. New developments in MARK now allow modelling of agedependence in 

survival through an automatically updated individual covariate, thus obviating the 

need for modelling the capture history between ringing as chick and first observation 

as breeder. We used this powerful approach to model age-dependence in survival of 

black-legged kittiwakes Rissa tridactyla at a North Sea colony. Although only 69 

breeders were of known age (out of a total of 471 colour-ringed adults), there was 

strong evidence for a quadratic relationship between true age and survival, with an 

initial increase followed by a pronounced decline in old age. In contrast, an analysis of 

the complete data set using “time since first observation as breeder” as a surrogate 

showed a linear decline in survival with this parameter. We believe that this simple 

but powerful approach could be implemented for many long-lived species and could 

provide improved estimates of how survival changes with age, a central theme in life 

history theory. 

Population dynamics of Starlings Sturnus vulgaris breeding in Britain: an integratedanalysis 

S.N. Freeman, R.A.Robinson, J.A.Clark, B.M.Griffin, S.Y.Adams & S.R.Baillie 

We describe a population model fitted to census data on the Starling at UK Common 

Birds Census (CBC) plots in 1965-2000. The demographic features of the model 

(survival and components of productivity) are calculated beforehand from national 

ring-recoveries and nest record data. In the most general form of the model, all parameters 

are assumed to vary between years, and survival rates of first-year and 

older birds are permitted to differ. We fit the model not to a single population index 

47


derived from the CBC data, as has been done in earlier Integrated Population Analyses 

of complete sets of CBC data, but to the site-based CBC counts themselves, thus 

better accounting for sampling covariances of annual abundance indices. Counts are 

assumed Poisson distributed, with expected values determined by the population 

model. Parameters are estimated by maximum likelihood and the method is readily 

applied via any Generalized Linear Modelling software. By removing temporal fluctuations 

in demographic model parameters in turn, we attempt to identify which of 

them have the most profound influence upon changes in abundance for this species 

of conservation concern. It is shown that changes in mortality rates of first-year birds 

have probably been the major cause of the substantial national decline witnessed 

over this period. A regional population model however – the first constructed using 

CBC data for any species – suggest a greater role for adult mortality in the largely arable 

Eastern region, the species’ UK stronghold. Population trends were similar in 

Northern and Western Britain, where farming is largely pastoral, and in each case 

changes in first-year survival remain the best candidate to explain the population 

changes. This difference might reflect differences in habitat availability and juvenile 

dispersal. 

Generalized Least Square Estimators for Arnason-Schwarz Capture-Recapture Models 

Olivier Gimenez, Rémi Choquet & Jean-Dominique Lebreton 

In the analysis of multistate capture-recapture data (Williams et al., 2002, section 

17.3; Lebreton and Pradel, 2003), Maximum Likelihood Estimates (Maximum Likelihood 

Estimates/Estimators, MLEs) are obtained by minimizing the deviance. Most 

often, iterative algorithms (Newton or Quasi-Newton methods; e.g., Dennis and 

Schnabel, 1983, sections 5 and 6) have to be used. Unfortunately, the success of this 

type of algorithm in reaching MLEs is frequently hampered by the presence of local 

minima and/or saddle points (Lebreton and Pradel, 2003). We propose to provide 

classical optimization algorithms with initial values that are “good” in the sense that 

they belong to the attraction region of the global minimum, in order to speed up and 

ensure the convergence towards the global minimum (Dennis and Schnabel, 1983). 

To this aim, we considered first a saturated parameterization of the JMV capturerecapture 

model (Brownie et al., 1993), a simple generalization of the Arnason- 

Schwarz model, inspired by the ring-recovery models (Schwarz et al., 1993). Using 

the Generalized Least Squares theory (e.g., Rao, 1973, section 4i.4), we derived explicit 

estimates (Generalized Least Squares Estimates/Estimators, GLSEs) for the Arnason-Schwarz 

and other related models derived from our saturated model by linear 

constraints, that are asymptotically equivalent to MLEs EURING 2003 – Poster Session 

7 (Gimenez, 2003). To illustrate our approach, GLSEs and MLEs were derived 

on a real example about the reproductive status of Sooty shearwaters (Scofield et al., 

2001). We then performed a general evaluation of the relative merits of GLSEs in two 

steps: 1) based on relative bias, relative efficiency and mean square error, we performed 

a sequence of simulations in order to compare GLSEs and MLEs for finite 

sample size data sets; 2) based on six data sets from the literature, we compared the 

quality and the speed of convergence when initial values are a) the GLSEs for the 

model considered, b) default initial values in MSURGE (Choquet et al., 2003) and 

MARK (White and Burnham, 1999) and c) random numbers distributed uniformly over 

the acceptable interval for each parameter. Although our approach - when it is used 

to get point estimates - seems to be overcome by the MLE, particularly in terms of 

precision, we advocate the use of GLSE as initial values since, regarding the examples 

we deal with, its direct computation speeds up and guarantees the convergence 

to the global minimum of the deviance. 

48


Estimating albatross survival: dealing with unobservable states 

Christine M. Hunter & Hal Caswell 

Survival estimation for intermittently-breeding seabirds has previously been problematic 

because mark-recapture methods could not account for the unobservability of 

non-breeding adults. New methods using multi-state mark-recapture frameworks 

have been proposed by Fujiwara and Caswell (2002) and Kendall and Nichols (2002) 

to deal with unobservable states. The estimability of relevant parameters for these 

models requires either extra information (e.g. by use of Pollock’s robust design) or 

model constraints (e.g. time constancy) and has been investigated for a number of 

simple model structures. Here we examine a series of more complicated models that 

differ in the distribution of inter-breeding intervals and that distinguish between successful 

and failed breeders. For each model we determine sets of assumptions sufficient 

to make the relevant parameters estimable. Estimability of parameters for these 

models is of interest for many albatross species because the length of time they remain 

in the unobservable non-breeding state depends on the outcome of the last 

breeding attempt. We will identify a series of models applicable to estimating adult 

survival probability and inter-breeding intervals for Wandering Albatross (Diomedea 

exulans). 

Estimating the chick survival of colour ringed gulls 

Risto Juvaste & Jari Valkama 

We present a method to estimate the chick survival of individually colour ringed gulls. 

It is based on large scale colour ringing of Herring Gulls (Larus argentatus) (HG) and 

Lesser Black-backed Gulls (Larus fuscus) (LBBG) in Finland during 1993-2000. Altogether 

7,636 chicks of Herring Gulls and 6,045 chicks of Lesser Black-backed Gulls 

were ringed with individually coded colour rings (cr). By 14 February 2003, there were 

42,646 sightings of 4,342 HG individuals (57%) and 21,827 sightings of 1,669 LBBG 

individuals (27%) in the database of Helsinki Ringing Centre. Because most of the 

birds were aged during ringing by measuring the wing length, the estimate for survival 

after ringing can be calculated from classified EURING 2003 – Poster Session 8 percentages 

of total sightings. The method has been tested with HG and LBBG populations 

in Finland. It was found that due to the high percentage of cr-sightings, the survival 

of HG chicks can be estimated even in small populations. An example is a cohort 

of 178 chicks ringed near Lappeenranta (SE Finland), wings 55 mm-335 mm, 

median 200 mm. Altogether there are 370 sightings of 72 individuals from these birds. 

By simple analysis of cr-sightings based on linear regression we estimated that about 

125 chicks fledged. The survival of chicks during the last 3- 4 weeks before fledging 

was about 50%. By use of population analysis (Popan5/Jolly- Seber/full/birth) of the 

sighting data (57 sightings/37 gulls) from a nearby dump it was estimated that 104 of 

the fledged birds (83%) came to the dump. Estimate is however rough. Based on the 

yearly sightings we estimate that 24 birds (SD=5.2) were still alive in the year 2001 

(Popan5/Jolly-Seber/dead only). In a similar analysis of LBBG-chicks ringed at Finnish 

lakes during years 1993-2000 (n=3,457) there was an unexpected decrease in 

survival in the group of nearly fledged chicks. The average survival decreased nearly 

10% from the group 250-300 mm to the group 300-380 mm. The percentages of 

sightings in these groups were 54% and 45% (χ2=5,9 df=1 P=0.015 n=940). The reason 

for this decrease, which is against the expected trend and found in all year cohorts, 

will be studied by analysing the sub-cohorts, using both resightings and recoveries 

of metal rings. 

49


Estimating population sizes of Herring Gulls at Joensuu dump by mark-resighting 

data 

Risto Juvaste 

The populations of Herring Gulls (Larus argentatus) at the rubbish dump of Joensuu 

were studied by population analysis based on mark-resighting data of 5,624 daysightings 

of 657 colour-ringed Herring Gull individuals in the year 2000. During the 

breeding season the total weekly numbers of gull individuals visiting the dump were 

almost ten times higher than the maximum numbers of gulls seen present at any one 

time. The peak estimate at the end of April was over 12,000 individuals in a week 

(Popan5/Jolly-Seber/full). The counts of the birds present at the dump showed maximum 

numbers in October, when 2,000 Herring Gulls were seen at a time. However, 

the estimate for that week was only about 3,000 individuals. The total number of Herring 

Gulls visiting the Joensuu dump during the year 2000 was estimated to have 

been over 20,000 birds. In addition, the residence times and the timing of migration of 

different age groups and sub-populations of Herring Gulls were studied. The results 

have been used in the management of gulls at the dumps. The population dynamics 

of Herring Gulls at the Joensuu dump will be analysed more thoroughly by using neural 

computing to analyse over 50,000 cr-sightings from the years 1997 - 2003. 

First use of a capture-recapture model for savannah tree demographics 

G. Lahoreau, J. Gignoux & R. Julliard 

Fire and grazing are the major determinants of the dynamic equilibrium between trees 

and grasses in African savannas. But the impact of fire on tree mortality is poorly understood. 

The Lamto research station in central Ivory Coast provides a model system 

to study the influence of fire given the relative absence of large herbivores. Since 

1962, burnings have also been carried out every January, at the height of the dry 

season. From 1991 to 1994, demographic censuses of trees were conducted every 

June and December. All individuals (including seedlings) were tagged and mapped. 

The status of some individuals, particularly small ones, was sometimes EURING 

2003 – Poster Session 9 unknown because: (i) 10cm-high seedlings can be difficult to 

find in grass layers that often exceed 2m at the end of the wet season; (ii) in December, 

some individuals have already shed their leaves, making it impossible to judge 

their viability. As estimated survival rates derived only from the number of known survivors 

could cause bias (Lebreton, 1992), we attempted to reanalyse the data using 

capture-recapture models. We used the 1993 Burnham model and MARK software to 

estimate age/time-dependent survival of seedlings in nine tree species. Preliminary 

results showed up to a 4-fold increase in mortality for one-year seedlings during the 

dry season, presumably due to fire, with significant differences between species. The 

use of capture-recapture models, common used in animal populations, appears very 

promising to analyse this plant ecosystem. 

Decomposing population growth rate: importance of adult survival, local recruitment 

and immigration in Willow Tits Parus montanus 

Satu Lampila, Markku Orell, Eduardo Belda & Kari Koivula 

We studied population growth rate and its components (adult survival, local juvenile 

survival and immigration) in Willow Tit in Northern Finland. We used capturerecapture 

modelling to achieve unbiased estimates of the population parameters and 

a components of variation analysis to partition sampling variation from process variation. 

During the 12-year study from 1991 to 2002 the population growth rate (λ) was 

close to one with considerable variation ( x = 0.99, coefficient of variation, CV = 0.20). 

50


Population projections, considering time spans of varying lengths, imply that if conditions 

remain as present the population is most likely declining; i.e. the impact of large 

variation. We used multi-strata models to uncover the relative contributions of the vital 

rates to λ. Adult survival had the highest contribution to the population growth rate 

and it was also the least variable trait studied ( x = 0.59, CV = 0.067). Partitioning the 

total recruitment into local juveniles and immigrants revealed that the highest contribution 

to the variation in λ was due to local juveniles. Indeed, the local juvenile survival 

was the most variable of the studied traits ( x = 0.06, CV = 0.61). Survival of willow 

tits after establishing in the breeding population show low temporal variation and 

sets the relative lower limit for population change. Thus we argue that variation in local 

recruitment determines the variation in λ above its relative magnitude set by adult 

survival. We implemented a set of a priori hypotheses to test if environmental factors 

explain the observed temporal process variation in the vital rates. Models suggested 

that adult survival was positively affected by pine cone abundance. However, we 

cannot rule out the possibility that pine cone abundance is correlated with other 

source of food such as invertebrates that is intensely hoarded for the upcoming winter. 

For juvenile survival and immigration pine cone abundance did not explain the 

temporal variation. 

Long-term study of a king penguin population using an automatic identification system 

C. Le Bohec, M. Gauthier-Clerc, J.-P. Gendner, N. Chatelain & Y. Le Maho 

Birds microtagged with transponders under the skin can be identified for life and detection 

antennas allow the automatic identification of the birds with minimal human 

disturbance and presence in the field. The king penguin cycle is unusual because it 

lasts more than one year and birds are not synchronised in their cycle. In addition, 

birds that failed in their reproduction still frequent the colony throughout the year. We 

have employed an automatic identification setup EURING 2003 – Poster Session 10 

for the detection of the passage of free-living microtagged king penguins at Possession 

Island (46°25'S, 51°45'E), Crozet Archipelago. A breeding area with about 

10,000 pairs is connected to the sea by three natural pathways, where antennas for 

bird identification were installed. From February 1998, 900 breeding adults and 1600 

one-year old chicks were fitted with transponders without any external marks. Cohorts 

of chicks will be implanted with transponders each year. This system allows estimation 

and comparison among years of variables such as arrival date for courtship, foraging 

trip duration at sea during incubation, brooding and winter periods, feeding frequency 

of chicks, duration of chick rearing, interannual return rates of adults and immatures 

and reproductive success. Additional studies allow determination of variation 

in body condition of adults, of tick parasitism and predation in relation to year and bird 

location in the colony. The aim of this long-term population study is to determine the 

reproductive strategy of these long-lived birds and the effects of age and climatic 

changes on their breeding performance. 

AD Model Builder: A tool for fitting custom built highly parameterized nonlinear 

models 

Mark Maunder 

AD Model Builder (ADMB, Otter Research, http://otter-rsch.com/admodel.htm) is becoming 

the predominant programming environment for producing complex, highlyparameterized 

fisheries stock assessment models. It has been used to fit complex 

nonlinear models with thousands of parameters simultaneously to multiple types of 

data and to fit nonlinear models with fewer parameters to hundreds of thousands of 

51


data points. ADMB is the combination of a code template and a set of libraries for 

C++. The code template and specific ADMB keywords reduce the amount of coding 

needed to implement a model. The C++ libraries supply a set of routines that are 

used to fit models to data. The features of ADMB include an efficient function minimizer, 

a MCMC algorithm for Bayesian integration, matrix algebra, automated likelihood 

profiles, parallel processing, and random effects parameters. ADMB uses automatic 

differentiation for exact derivatives, which makes the minimization procedure 

more efficient and stable than other packages that use finite difference approximation. 

ADMB provides a flexible stepwise process to sequentially estimate the parameters, 

and allows the placing of bounds on all estimated parameters that restrict the range of 

possible parameter values. The MCMC algorithm implemented in ADMB has jumping 

rules that are based on the variancecovariance estimates at the mode of the posterior 

distribution and starts at the mode of that distribution, which makes the algorithm 

more efficient (i.e. reduces the burn-in time). Randomeffect parameters are implemented 

with Laplace’s approximation with automatic second derivatives. All the underlying 

code used to define the model and objective function is coded in C++ (mainly 

C). Therefore, ADMB is very flexible, and is useful for custom-built models. ADMB includes 

parallel processing to make computationally intense models more practical. 

Use of log-linear models to analyse habitat selection of individually marked song 

thrushes Turdus philomelos 

Will J. Peach 

Understanding habitat utilisation and selection patterns at different scales continues 

to be a fundamental objective of many applied biologists. Compositional analysis 

(Aebischer et al., 1993, Ecology 74, 131-1325) is now widely used but can be problematic 

to implement when habitats are not available within some home ranges. Loglinear 

models offer great flexibility in EURING 2003 – Poster Session 11 analysing 

habitat selection data and testing covariate effects (Manley et al. 1993, Resource 

Selection by Animals. Chapman & Hall; Green et al. 2000, J. Zool. 250, 161-184) but 

have not been widely adopted by biologists. Here I use log-linear models to analyse 

habitat selection patterns of radio-tagged song thrushes Turdus philomelos, a species 

that is red-listed in the UK following a population decline of approximately 70%. 

Thrushes were located during two successive breeding seasons in two farmland 

populations – one stable and one rapidly declining. Four thousand fixes were available 

from 86 tagged individuals. We used log-linear models to analyse variation in fix 

density across habitats. The model accounted for differences in ranging behaviour 

between nesting and non-nesting periods and in the numbers of available fixes per 

individual. It also tested the influence of a range of potential covariates on the pattern 

of fix density including tag attachment method, year, season, landscape and study 

area. Randomisation tests in which the individual bird was the unit of replication were 

used to assess statistical significance. Fix density varied significantly across habitats 

but not between study areas or any of the other covariates. Large differences in 

habitat utilisation between study areas mainly reflect differences in habitat availability. 

Numerical problems arose when there are no fixes in one covariate level. Log-linear 

models provide an efficient and flexible means of analysing habitat selection data. 

There is a pressing need to compare the performance of compositional analysis and 

log-linear models, and to develop user-friendly software that includes randomisation 

significance testing and the testing of covariate effects. 

52


Demographic mechanisms of the population decline of the song thrush Turdus 

philomelos in Britain 

R A Robinson, R E Green, S R Baillie & W J Peach 

In Britain, the song thrush Turdus philomelos is categorized as a species of high national 

conservation concern because of a large population decline during the last 

three decades. We used survey data to calculate a series of annual national population 

estimates for woodland and farmland habitats combined for the period 1964- 

2000. We then used turning points analysis to identify seven blocks of years within 

the period of decline (1968-2000) with uniform rates of population change in the 

smoothed trend. Six of the seven blocks showed declines. We used recoveries of 

song thrushes ringed by participants in the national ringing scheme as nestlings, juveniles 

and adults in April-September to estimate survival rates and modeled survival 

rates separately for the post-fledging period, the remainder of the first year and for 

adults. Daily survival probability was much lower during the post-fledging period than 

in the remainder of the first year or for older birds. There was evidence of variation in 

survival rates among blocks of years with different rates of population change and in 

particular for first year survival. There were significant positive correlations across 

blocks between mean population multiplication rate and both post fledging and first 

year survival. The demographic mechanism underlying the song thrush population 

decline appears to be changes in survival in the first year and perhaps also the postfledging 

period. The environmental causes of these changes in survival are not 

known. Adverse weather conditions contributed, but were not sufficient on their own. 

Changes in farming practice, land drainage, pesticides and changes in predator numbers 

are all candidates. 

Estimating demographic contributions to population growth and decline in a 

salamander: a multistate model with a unobservable state 

Benedikt R. Schmidt, Rainer Feldmann & Michael Schaub 

Amphibian populations are declining worldwide. The reasons for the declines and the 

factors that govern population dynamics are poorly understood. Efficient conservation 

action requires that we understand which stage(s) in the complex life cycle contribute 

most to population growth (or decline) and which stage(s) are most sensitive to environmental 

change. Here, we study the demography of a stationary and a declining 

population (based on unadjusted counts) of the salamander Salamandra salamandra, 

and assess whether an observed decline in one population is due to a change in recruitment 

or adult survival. The two populations were studied for c. 20 years and all 

salamanders were known individually. We use a multistate capture-mark-recapture 

model with an unobservable state to estimate both recruitment and adult survival 

probabilities. Our analysis shows constant adult survival in the stationary population, 

and declining adult survival and constant recruitment in the declining population. The 

analysis also shows that the population growth rate of the declining population was 

not only negative but also steadily decreasing. The results suggest that the population 

decline is due to a change in adult survival, which was probably caused by a change 

in forest management. Our results provide a counterexample to the usual paradigm in 

amphibian population ecology, which states that population growth is determined at 

the larval stage. Our analysis also highlights the value of capture-mark-recapture 

analysis for understanding population dynamics and decline. 

53


Toll-free bands and erroneous reporting behavior: effects on survival estimation 

Joel A.Schmutz 

Previous simulation study by Anderson and Burnham (1980) examined the bias in 

survival estimation when there is a "delay" in the reporting of bands - that is, reporting 

bands with an erroneous and later recovery date than when they were truly recovered. 

Their conclusion was that the bias was small and negligible compared to the 

precision of estimates. Recent analyses of band recovery data from Greater Whitefronted 

Geese suggest that bias may exceed 5%, and this large and biologically relevant 

level of bias may be exacerbated by the advent of band reporting using toll-free 

telephone numbers. Specifically, when analyzing recoveries of birds banded during 

1990-1995, estimated survival rates were approximately 5% greater when using recoveries 

through 2000 (a "non-triangular" data matrix, where years of recovery exceed 

years of banding) than when using recoveries through 1995 (a "triangular" data 

matrix). Broad scale toll-free band reporting started in 1996, and I hypothesized that 

many previously recovered but unreported bands would now get reported by this new 

toll-free system, and importantly, that many would be reported with an inaccurate year 

of recovery. Data simulations using hypothesized amounts of single and multi-year 

"delay" in reporting were able to replicate the results seen with the Greater Whitefronted 

Geese. Band recovery data from other waterfowl also demonstrated this pattern. 

The bias induced by this reporting process is expected to be greater in nontriangular 

than in triangular data sets. Comparing results from triangular vs. nontriangular 

data sets over a long time period will elucidate whether bias from this band reporting 

behavior is relegated just to the advent of toll-free band reporting or, alternatively, 

it has been a long term source of bias in band recovery data sets. 

On the use of Capture-Recapture data in parameterising a Population Viability 

Analysis model for the Bog Fritillary butterfly 

Nicolas Schtickzelle & Michel Baguette 

In the context of a species-based conservation strategy, Population Viability Analysis 

(PVA) is used to determine the habitat network configuration ensuring the highest 

persistence of a species in a given landscape. Structured population models taking 

local population dynamics into account are effective tools for the conservation of 

threatened species at the very end of the fragmentation process. Nevertheless, they 

require estimates of various parameters at the population level, which may be obtained 

from capture-recapture data. This poster details how Capture-Recapture data 

have been used to parameterise a PVA model designed for a viability analysis of the 

Bog Fritillary butterfly (Proclossiana eunomia). The study system is a metapopulation 

located in a highly fragmented landscape in southern Belgium, where this butterfly is 

used as a surrogate species for wet meadow communities. This metapopulation consists 

of 20 suitable habitat patches spread on 6 km along the Lienne river. Daily 

Capture-Recapture data during 10 generations provided information to estimate the 

following parameters: 

• Survival, catchability, recruitment, daily and total (per generation) population size 

were obtained by live-recapture CJS and JS models, fitted using MARK (White & 

Burnham 1999) and POPAN (Arnason & Schwarz 1999) software respectively. The 

sequence of population sizes during the 10 generations was used to derive estimates 

of density dependence parameters (maximum population growth rate and carrying 

capacity) as well as the magnitude of environmental stochasticity (yearly variation of 

population growth rate). 

54


• Dispersal parameters (emigration according to patch area and dispersal probability 

according to patch connectivity) were estimated using the Virtual Migration model 

(Hanski et al. 2000). 

• Spatial correlation of local population dynamics as a function of betweenpopulations 

distance was determined using yearly frequentation of individual habitat 

patches by butterflies as reflected by number of captures. 

Demographic mechanisms of density dependence in Black Brant 

James S. Sedinger 

Density dependence in long-lived vertebrates is typically manifested in reduced recruitment 

of young, while adult survival is preserved. The Black Brant population 

breeding in western Alaska declined substantially during the late 1970s and early 

1980s. Coworkers and I have studied Black Brant breeding at the Tutakoke River colony 

on the Bering Sea coast of Alaska since 1984. During the study more than 

40,000 individuals have been marked with uniquely coded plastic leg bands. Using 

observations and captures of individuals on the breeding colony, at other breeding locations, 

and on wintering estuaries in Mexico, we have monitored a number of demographic 

parameters as the population increased to near historic levels. Consistent 

with general patterns for other long-lived vertebrates, adult survival did not decline 

during the period of population increase. We detected no density-related effects on 

breeding probability, based on robust design. Using a variant of Burnham models we 

did not detect any density-related variation in dispersal rates. Growth rates of young 

declined during population increase, which was associated with declines in first-year 

survival. There were additional density-related effects on survival, as evidenced by 

the fact that survival in more recent cohorts declined more than would be expected 

based on gosling size alone. Overall, we found no change in adult survival or reproductive 

investment by adults but we did find reduced recruitment at higher population 

densities. 

Relationships between landscape characteristics and duck nesting success in the 

Missouri Coteau region of North Dakota 

Scott E. Stephens, Jay J. Rotella, Mark S. Lindberg, & Mark L. Taper 

For upland-nesting dabbling ducks, nesting success is the primary determinant of 

population growth rates. As a result, a great deal of activity is focused on understanding 

factors related to nesting success rates and most conservation programs 

seek to affect nesting success rates. We developed models to understand and explain 

variation in nesting success rates across a gradient of landscape types in the 

Missouri Coteau region of North Dakota. Ultimately, our goal was to develop statistical 

models that could be utilized in spatially explicit habitat models to guide strategic 

targeting of conservation programs across broad spatial scales. During 2000-2002, 

we collected data on the fates of over 5,000 duck nests from 18 study sites. We developed 

a priori models that represented various hypotheses about how landscape-, 

patch-, and nest-level habitat covariates influence nest survival. Because of the scaledependent 

nature of landscapescale metrics, we measured landscape covariates at 5 

spatial extents (i.e., 10.4, 23.3, 41.4, 64.7 and 93.2 square km). We also took advantage 

of variance components methods to estimate the amount of spatial process 

variation in nesting success rates. Based on the results of our a priori models, we examined 

a limited number of exploratory models. Our best models included only landscape-level 

habitat covariates such as the amount of grassland, grassland edge and 

wetland density. Additionally, models that measured landscape metrics at multiple 

55


spatial scales were dramatically better (∆AIC 32.3) than models that measured all 

landscape metrics at the same landscape scale. We estimate that the landscape covariates 

in our best model captured 90% of spatial process variation in nesting success 

in our dataset. Thus, our statistical model looks to hold great utility for use in 

spatially explicit landscape models to guide conservation programs. We suggest that 

avian researchers interested in relationships between demographic rates and habitat 

characteristics measure landscape variables at multiple landscape scales to determine 

which scales are most relevant. 

A unified framework for modelling wildlife population dynamics 

Len Thomas, Stephen T. Buckland, Ken B. Newman & John Harwood 

We propose a unified framework for defining and fitting stochastic, discrete time, discrete 

stage population dynamics models. The biological system is described by a 

state-space model, where the true but unknown state of the population is modelled in 

a state process, and this is linked to survey data by an observation process. All 

sources of uncertainty in the inputs, including uncertainty about model specification, 

are readily incorporated. We show how the state process can be represented as a 

generalization of the standard Leslie or Lefkovitch matrix. By dividing the state process 

into sub-processes, complex models can be easily constructed from manageable 

building blocks. We illustrate the approach with a model of the British grey seal metapopulation. 

We use Bayesian sequential importance sampling with kernel smoothing 

to fit the model. 

Profile likelihood intervals: a new feature of program MARK to solve a problem of 

“standard theory” 

Gary C. White & Kenneth P. Burnham 

If the MLE of a parameter (such as a survival probability) lies on, or too near to, the 

boundary of a parameter space then what people are taught as large sample (i.e., 

standard) frequentist theory for obtaining an estimated standard error and confidence 

interval can fail abysmally. There is for this situation a good non-Bayesian solution for 

obtaining a confidence interval: the profile likelihood interval. We will give an explanation 

of this interval; how it is computed in MARK will be noted. There is also the Bayesian 

credibility interval as a solution. Under a uniform prior on a probability parameter 

the two intervals are very similar. We use the classical female blackkneed capsids 

data to give examples of different confidence intervals for survival probabilities (S), 

with or without the constraint S


List of participants 

Alisauskas Ray Canadian Wildlife Service 

Prairie and Northern Wildlife 

Research Centre 

Saskatoon, Saskatchewan 

Canada S7N 0X4 

Alpizar-Jara Russell Departamento de Matemática 

Universidade de Évora 

Rua Romao Ramalho 59 

7000-671 Évora, Portugal 

Arnason Neil Department of Computer Science 

University of Manitoba 

Winnipeg, R3T 2N2 Canada 

ray.alisauskas@ec.gc.ca 

alpizar@uevora.pt 

arnason@cs.umanitoba.ca 

Baillie Stephen BTO, The Nunnery 

Thetford, Norfolk IP24 2PU, UK 

Bairlein Franz Institute of Avian Research 

'Vogelwarte Helgoland' 

An der Vogelwarte 21 

26386 Wilhelmshaven 

Barbraud Christophe Centre d'Etudes Biologiques de 

Chizé - CNRS 

79360 Villiers en Bois, France 

stephen.baillie@bto.org 

franz.bairlein@ifv.terramare.de 

barbraud@cebc.cnrs.fr 

Barker Richard University of Otago rbarker@maths.otago.ac.nz 

Belda Eduardo J. Universidad Politecnica de Valencia 

Ctra Nazaret-Oliva s/n 46730 

Ganida (Valencia) SPAIN 

ebelda@dca.upv.es 

Bomberger-Brown Mary University of Tulsa mary-brown@utulsa.edu 

Bonner Simon Department of Statistics and 

Actuarial Science 

Simon Fraser University 

8888 University Dr. 

Burnaby BC, V5A 1S6 

CANADA 

sbonner@stat.sfu.ca 

Brooks Steve Statistical Laboratory 

CMS 

University of Cambridge 

Wilberforce Road 

Cambridge CB3 0WB, UK 

steve@statslab.cam.ac.uk 

Brown Charles R. University of Tulsa charles-brown@utulsa.edu 

Burnham Kenneth P. USGS, Coop Units Program 

Colorado State University 

Fort Collins, CO 80523 USA 

kenb@Lamar.ColoState.edu 

Calvert Anna Universite Laval 

Quebec City, Canada 

anna.calvert@bio.ulaval.ca 

57

Participants 

Cam Emmanuelle Laboratoire Evolution et Diversite 

Biologique 

Universite Toulouse III, France 

58 

emmacam@cict.fr 

Caswell Hal Woods Hole Oceanographic Inst. hcaswell@whoi.edu 

Chernetsov Nikita Biological Station Rybachy 

Rybachy 238535, Kaliningrad 

Region, Russia 

Nchernetsov@bioryb.koenig.su 

Choquet Rémi CEFE / CNRS 

1919 Route de Mende 

34 293 Montpellier cedex 5 

France 

Conroy Michael J. Georgia Cooperative Fish and 

Wildlife Research Unit 

University of georgia 

Athens GA 30607 USA 

Cooch Evan Department of Natural Resources 

Cornell University 

Ithaca, New York 

USA 14853 

Coppack Tim Institute for Avian Research "Vogelwarte 

Helgoland" 

Inselstation, D-27494 Helgoland 

Dhont André Cornell University 

159 Sapsucker Woods Road 

Ithaca, NY 14850, USA 

choquet@cefe.cnrs-mop.fr 

conroy@forestry.uga.edu 

evan.cooch@cornell.edu 

coppack@vogelwarte-helgoland.de 

aad4@cornell.edu 

Diefenbach Duane R. Pennsylvania Cooperative Fish drd11@psu.edu 

and Wildlife Research Unit 

Dinsmore Stephen J. Mississippi State University sdinsmore@cfr.msstate.edu 

Doherty Paul Dept. Fishery and Wildlife Biology 


Fort Collins, CO 80523-1475 

doherty@cnr.colostate.edu 

Doligez Blandine Dept. Of Evolutionary Ecology 

Institute of Zoology, 

University of Bern 

Baltzerstrasse 6, CH-3012 Bern 

Switzerland 

Drake Kiel University of Saskatchewan 

Department of Biology 

112 Science Place 

Saskatoo, Saskatchewan 

Canada, S7N 5E2 

Drechsler Martin UFZ-Centre for Environmental 

Research, Leipzig, Germany 

Ebbinge Bart Centre for Ecosystem Studies 

Alterra, P.O. Box 47 

NL-6700 AA Wageningen 

The Netherlands 

blandine.doligez@esh.unibe.ch 

kiel.drake@ec.gc.ca 

martind@oesa.ufz.de 

Bart.Ebbinge@wur.nl


Edwards Holly H. Florida Fish and Wildlife Conservation 

Commission 

111 Eighth Ave. SE 

St. Petersburg, FL 33701 

Efford Murray Landcare Research 

Private Bag 1930 

Dunedin, New Zealand 

Ens Bruno Alterra-Texel 

P.O. Boix 167 

NL-1790 AD Den Burg (Texel) 


Fiedler Wolfgang Max Planck Research Centre for 

Ornithology 

Vogelwarte Radolfzell, Germany 

Fletcher David University of Otago 

Dunedin, 

New Zealand 

Fonnesbeck Christopher Georgia Cooperative Fish & Wildlife 

Research Unit 

Warnell School of Forest 

Resources 

University of Georgia 

Athens, GA 30602 

Francis Charles Canadian Wildlife Service 

National Wildlife Research Centre 

Orrawa, Ontario, K1A 0H3 

Frederiksen Morten Centre for Ecology and Hydrology, 

Hill of Brathens 

Banchory, AB31 5SS, UK 

Gauthier Gilles Department of Biology and 

Centre d' études nordiques 

Université Laval 

Quèbec, Qc, Canada, G1K 7P4 

Gimenez Olivier CEFE / CNRS 



France 

Haas Timothy C. School of Business Administration 

University of Milwaukee 

P.O. Box 742 

Milwaukee, WI 53201 

Heynen Iris Staatliches Museum für Naturkunde 

Schloss Rosenstein, Stuttgart 

Hines Jim USGS-Patuxent Wildlife Research 

Center 

USA 

59 

Holly.Edwards@FWC.state.fl.us 

effordm@landcareresearch.co.nz 

bruno.ens@wur.nl 

fiedler@vowa.ornithol.mpg.de 

dfletcher@maths.otago.ac.nz 

chris@fonnesbeck.org 

charles.francis@ec.gc.ca 

mfr@ceh.ac.uk 

gilles.gauthier@bio.ulaval.ca 

gimenez@cefe.cnrs-mop.fr 

haas@uwm.edu 

heynen.smns@naturkundemuseum-bw.de 

jim_hines@usgs.gov

Participants 

Hiroi Tadakazu Yamashina Institute for 

Ornithology 

Hochachka Wesley M. Laboratory of Ornithology 

Cornell University 

159 Sapsucker Woods Rd. 

Ithaca, NY, 14850 USA 

Hoyle Simon Inter-American Tropical Tuna 

Commission 

8604 La Jolla Shores Drive 

La Jolla CA 92037-1508, USA 

Hunter Christine M. Woods Hole Oceanographic Inst., 

USA 

Jenouvrier Stephanie Centre d'Etudes Biologiques de 

Chizé – CNRS 

79360 Villiers en Bois, France 

Julliard Romain Centre de Recherches sur la 

Biologie des Populations 

d'oiseaux 

55, rue buffon 

F-75005 Paris, France 

Juvaste Risto North Karelia Polytechnic 

Environmental Technology 

Karjalankatu 3, 80200 Joensuu, 

Finland 

Kendall Bill USGS Patuxent Wildlife Research 

Center 

11510 American Holly Drive 

Laurel, MD 20708-4017, USA 

hiroit@jcom.home.ne.jp 

wmh6@cornell.edu 

simon.hoyle@pobox.com 

cmhunter@whoi.edu 

jenouvrier@cebc.cnrs.fr 

julliard@mnhn.fr 

Risto.Juvaste@ncp.fi 

William_Kendall@usgs.gov 

Kery Marc Patuxent Wildlife Research marc_kery@usgs.gov 

Center, Laurel, MD 20708 

King Ruth University of St. Andrews, UK ruth@mcs.st-and.ac.uk 

Lahoreau Gaëlle Laboratoire "Fonctionnement et 

Evolution des Systèmes Ecologiques", 

Paris. 

Lampila Satu University of Oulu 

Departement of Biology 

PL 3000 

SF-90014 Oulu, Finland 

Le Bohec Celine CEPE-CNRS 

27, Rue Bequerel 

67087 Strasbourg cedex 02, 

France 

Lebreton 

Jean-Dominique 

CEFE / CNRS 



France 

lahoreau@wotan.ens.fr 

satu.lampila@oulu.fi 

celine.lebohec@c-strasbourg.fr 

lebreton@cefe.cnrs-mop.fr 

60


Lee Danny C. USDA Forest Service 

Pacific Southwest Res. Station 

1700 Bayview Drive 

Arcata, CA 95521 USA 

dclee@fs.fed.us 

Lee Derek E. PRBO dlee@prbo.org 

Link William A. 11510 American Holly Drive 

USGS Patuxent Wildlife Research 

Center 

Laurel, Maryland 20708 USA 

william_link@usgs.gov 

Lokki Heikki University of Helsinki 

Dept. of Computer Science 

P.O. Box 26 

FIN-00014 University of Helsinki 


Lukacs Paul Colorado Cooperative Fish 

and Wildlife Research Unit 


Fort Collins, CO 80523 USA 

MacKenzie Darryl Proteus Research & 

Consulting ltd. 

PO Box 5193, Dunedin 

New Zealand 

Maunder Mark Inter-American Tropical Tuna 

Commission 

8604 La Jolla Shores Drive 

La Jolla CA 92037-1508, USA 

Moore Clint USGS Patuxent Wildlife Research 

Center 

Warnell School of Forest 

Resources 

University of Georgia 

Athens, GA 30602, USA 

Morgan Byron Institute of Mathematics and 

Statistics 

University of Kent 

Canterbury, CT2 7NF, England 

Nichols James D. U.S. Geological Survey 

Patuxent Wildlife Research 

Center 

11510 American Holly Dr. 

Laurel, MD 20708-4017 

Heikki.Lokki@cs.Helsinki.fi 

plukacs@cnr.colostate.edu 

darryl@proteus.co.nz 

mmaunder@iattc.org 

cmoore@forestry.uga.edu 

B.J.T.Morgan@kent.ac.uk 

jim_nichols@usgs.gov 

Norris Jim Wake Forest University norris@wfu.edu 

Otis David L. Iowa Cooperative Research Unit 

Iowa State University 

Ames, IA 5011 

dotis@iastate.edu 

Otto Mark US Fish and Wildlife Service Mark_Otto@FWS.gov 

61

Participants 

Ozaki Kiyoaki Bird Migration Research Center 

Yamashina Institute for 

Ornithology 

115 Konoyama, Abiko 270-1145 

Japan 

ozaki@yamashina.or.jp 

Peach Will RSPB will.peach@rspb.org.uk 

Pledger Shirley School of Math. And Comp. Sci. 

Victoria University of Wellington 

P.O. Box 600 

Wellington, New Zealand 

shirley.pledger@vuw.ac.nz 

Pradel Roger Montpellier, France pradel@cefe.cnrs-mop.fr 

Pulido Francisco Max Planck Research Centre for 

Ornithology 


pulido@vowa.ornithol.mpg.de 

Rahbek Carsten Copenhagen Bird Ringing Centre 

Zoological Museum, University of 

Copenhagen 

Universitetsparken 15 

DK-2100 Copenhagen O, Denmark 

Reed Eric Canadian Wildlife Service 

351 St Joseph Blvd. 

Hull, Quebec K1A 0H3, Canada 

Rivalan Philippe Laboratoire Ecologie, Systematique 

et Evolution 

Universite Paris-Sud, bat. 362 

91405 Orsay, France 

Robinson Rob BTO, The Nunnery 

Thetford, Norfolk, IP24 2PU, UK 

Rodriguez Marco A. Department de chemie-biologie 

Universite du Quebec a Trois- 

Rivieres 

Quebec, Canada 

crahbek@zmuc.ku.dk 

eric.reed@ec.gc.ca 

philippe.rivalan@ese.u-psud.fr 

rob.robinson@bto.org 

marco_rodriguez@uqtr.ca 

Rotella Jay Montana State University rotella@montana.edu 

Royle J. Andrew U.S. Fish and Wildlife Service andy_royle@fws.gov 

Samtmann Sébastien CEPE-CNRS 

27, Rue Bequerel 

67087 Strasbourg cedex 02, 

France 

Saurola Pertti Finnish Museum of Natural 

History, P.O. Box 17 

FIN-00014 Univ. of Helsinki 


Schaefer Thomas Max Planck Research Centre for 

Ornithology 


sebastien.samtmann@c-strasbourg.fr 

pertti.saurola@helsinki.fi 

schaefer@vowa.orbithol.mpg.de 

62


Schaub Michael Swiss Ornithological Institute 

CH-6204 Sempach 

Switzerland 

Schmidt Benedikt Zoological Institute 

University of Zurich 

Switzerland 

Schmutz Joel United States geological Survey 

Alaska Science center 

1011 East Tudor Road 

Anchorage, Alaska 99503, USA 

Schtickzelle Nicolas Catholic University Louvain 

Biodiversity Research Centre 

Croix du Sud 4-5 

B-1348 Louvain-la-Neuve 

Schwarz Carl J. Statistics and Actuarial Science 

Simon Fraser University 

Burnaby, BC, Canada V5A 1S6 

michael.schaub@vogelwarte.ch 

bschmidt@zool.unizh.ch 

joel_schmutz@usgs.gov 

schtickzelle@ecol.ucl.ac.be 

cschwarz@stat.sfu.ca 

Sedinger Jim University of Nevada Reno jsedinger@cabnr.unr.edu 

Senar Juan Carlos Natural History Museum 

P. Picasso s/n 

08003 Barcelona, Spain 

jcsenar@mail.bcn.es 

Siriwardena Gavin BTO, The Nunnery 

Thetford, Norfolk, IP24 2PU, UK 

Smith Barry D. Canadian Wildlife Service 

Pacific Wildlife Research Centre 

5421 Robertson Road 

Delta, B.C., V4K 3N2, Canada 

Sokolov Leonid Biological Station Rybachy 

Rybachy 238535, Kaliningrad 

Region, Russia 

Spina Fernando Istituto Nazionale per la Fauna 

Selvatica 

Via Ca' Fornacetta, 9 

I-40064 Ozzano Emilia (BO), Italy 

Stauffer Howard B. Humboldt State University 

Arcata, California 95521, USA 

Stephens Scott Ducks Unlimited Inc. 

Great Plains Regional Office 

Bismarck, ND 58503 USA 

Tavecchia Giacomo IMEDEA 

07190 Esporales, Spain 

Thomas Len Centre for Research into Environmental 

and Ecological Modelling 

(CREEM) 

University of St. Andrews 

gavin.siriwardena@bto.org 

barry.smith@ec.gc.ca 

lsok@bioryb.koenig.su 

infsmigr@iperbole.bologna.it 

hbs2@humboldt.edu 

sstephens@ducks.org 

g.tavecchia@uib.es 

len@mcs.st-and.ac.uk 

Thomson David L. Netherlands Institute of Ecology d.thomson@nioo.knaw.nl 

63

Participants 

Thorup Kasper Zoological Museum 

University of Copenhagen 

Denmark 

Traylor Joshua, J. Department of Biology 

University of Saskatchewan 

c/o Canadian Wildlife Service 

115 Perimeter Road, Saskatoon, 

Saskatchewan 

S7N 0X4 Canada 

van der Greft Janien Centre for Landscape Ecology 

Alterra, P.O. Box 47 

NL-6700 AA Wageningen 


van Noordwijk Arje Netherland Institute of Ecology 

Boterhoekse straat 48 

NL 6666GA Heteren 

Viallefont Anne IUT Lumière 

Université Lyon 2 

69676 Bron cedex, France 

White Gary C. Dept. of Fishery & Wildlife Biology 


Fort Collins, CO, USA 

Yoccoz Nigel Norwegian Institute for Nature 

Research 

Polar Environmental Centre 

N 9296 Tromsø, Norway 

kthorup@zmuc.ku.dk 

joshua.traylor@ec.gc.ca 

Janien.vandergreft@wur.nl 

a.vannoordwijk@nioo.knaw.nl 

aviallef@univ-lyon2.fr 

gwhite@cnr.colostate.edu 

nigel.yoccoz@nina.no 

64

The quantitative study of marked individuals in ecology, evolution ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?