Land_Ecosystems.pdf - S?TE

More documents

Recommendations

Info

224 EQS SCIENCE P..eN - HAPrER 5 distribution functions (BRDFs) from MISR are integrated with the MODIS land-cover product. This advanced landcover product should first be available around 2000 (Lambin and Strahler 1994). Townshend (1996) will pro duce a MODIS-derived land-cover-change product specifically with the 250-rn channels of MODIS for highresolution tracking of regional land cover and land-use change. Potential, or climatically-defined land cover is be ing computed by a number of biogeography models, the most well developed being the Biogeochemical Informa tion Ordering Management Environment (BIOME)2 model (Prentice et al. 1992) and the Mapped Atmosphere- Plant Soil System (MAPSS) model by Nielson (1995). Good definition of potential land cover is a prerequisite for land-cover change analysis and will be used by the Moore-IDS and Schimel-IDS teams for biospheric mod eling. FGLJIE 5.1 2 5.2.3.2.1 Landsat and high-spatial-resolution land science Although EOS is predominantly planned as a global-scale science program, some of the satellites will produce im agery at high spatial resolution. For example, Landsat-7 will produce data at 10- to 30-rn spatial resolution with a 16-day repeat cycle. The long history of Landsat science and applications has illustrated that this spatialltemporal combination is best used for regional land-cover map ping. The data volume is too high for global use, and temporal constraints preclude seasonal time-series analy sis. However, the high spatial detail makes Landsat-7 the preferred platform for land-cover change detection, par ticularly where human-induced changes often occur at sub-kilometer scales (Skole and Tucker 1993). Opportunities to improve mapping of land cover and vegetation structure, such as stem biomass and forest Global land cover classification from satellite data The existing global distribution of terrestrial biomes derived from AVHRR data for 1989 (Nemaniand Running 1995).These land-cover data sets are used to quantify land-surface characteristics in global climate, hydrology, and carbon-cycling models. Land-cover monitoring is also impor tant for quantifying deforestation/reforestation rates and land-use change and urbanization. EQS will produce a global land-cover product similar to this at 1-km resolution beginning with the EQS AM-i launch.
density, lie primarily with the high-spatial-resolution sen sors like Landsat, and the multi-look-angle MISR sensor. MISR will provide seven looks at a target in one over pass with its multiple-sensor design. As with Landsat, the long repeat time and high data volume will preclude it from routine full global monitoring in the way that MO DIS is scheduled, but will allow enhanced capability for detection of vegetation variables and change analysis not possible with MODIS. ASTER will also provide spatial resolution of 15- to 90-rn for selected target scenes. MODLAND-Townshend will supplement 250-rn MODIS data with Landsat and ASTER data to produce a highresolution land-cover-change detection that will include a regional change “alarm” to detect areas undergoing very rapid change. This alarm should alert scientists to large land-cover perturbations in uninhabited areas. Dave Skole of the Moore-IDS team will continue to monitor tropical deforestation rates with Landsat and ASTER sensors (Skole and Tucker 1993). 5.2.3.3 Vegetation structure 5.23.3.1 Leaf-Area Index (LAI) Plant canopies are the critical interface between the at mosphere and the terrestrial biosphere. Exchanges of energy, mass, and momentum between the atmosphere and vegetation are controlled by plant canopies. When considering the array of global vegetation, there is an in finite variety of plant canopy shapes, sizes, and attributes. Over the last few decades, ecologists have found that a useful way to quantify plant canopies in a simple yet pow erful way is by defining the LAJ (the projected leaf area per unit ground area). This parameter represents the struc tural characteristic of primary importance, the basic size of the canopy, while conveniently ignoring the complexi ties of canopy geometry that make global comparisons impossible otherwise. Characterization of vegetation in terms of LAI, rather than species composition, is consid ered a critical simplification for comparison of different terrestrial ecosystems worldwide. Because LAI most directly quantifies the plant canopy structure, it is highly related to a variety of canopy processes, such as interception, ET, photosynthesis, res piration, and leaf litterfall. LAT is an abstraction of a canopy structural property, a dimensionless variable that ignores canopy detail such as leaf-angle distribution, canopy height, or shape. Hence the definition of LAI is used by terrestrial models to quantify those ecosystem processes. FPAR is a radiation term, so it is more directly related to remotely-sensed variables such as Simple Ra LAND ECDSYSTEMS AND HYDROLDGy 225 tio, NDVI, etc., than LAI. FPAR is frequently used to translate direct satellite data such as NDVI into simple estimates of primary production. It does not define plant canopies as directly as LAI, but is more specifically re lated to the satellite indices. As remote sensing became an important tool in terrestrial ecology, initial efforts concentrated on measur ing LAI by satellite (Asrar et al. 1984; Peterson et al. 1987). Remote sensing of LAI was first attempted for crops and grasslands, correlating spectral reflectances against direct measurement of vegetation LAI. Various combinations of near-infrared and visible wavelengths have been used to estimate the LAI of wheat (Wiegand et al. 1979; Asrar et al. 1984). Peterson et al. (1987) first estimated the LAI of coniferous forests across an envi ronmental gradient in Oregon using airborne Thematic Mapper Simulator data. NDVI is found to vary monotonically with fraction of vegetation cover for various biome types, a variable related to LAI (Price 1992; Huete et al. 1988; Nemani etal. 1993). In perennial biome types such as forests, LAI re flects climatic optima; warm, wet climates produce forests of high LAI, while colder, drier climates produce lower LAI. Because plants regrow and recycle their canopies on a regular basis, LAI provides a direct measure of the magnitude of biogeochemical cycling of a vegetation type. LAI changes seasonally in crops and annual vegetation types, and the weekly increase in LAI provides a good monitor of crop development. Even in permanent vegeta tion types such as forests, LAI will change interannually with climatic fluctuations, particularly of water balance. Defining the length of growing season of deciduous veg etation can best be done by tracking spring growth and fall senescence of LAI. The advanced biospheric models such as TEM, Century, and BIOME-BGC all use LAI as the vegetation structural variable. Neither LAI nor FPAR are critical vari ables themselves, rather they are both essential intermediate variables used to calculate terrestrial energy, carbon, water cycling processes, and biogeochemistry of vegetation. Although the NPP of grasslands, annual crops, and other seasonal biome types can be estimated by the time integration of observed developing biomass, for biome types such as forests, chaparral, and other ever green broadleafs, permanent live biomass occupies the site continuously, causing annual NPP to not be visible from orbiting satellites. The current consensus is that LAI will be used preferentially by ecological and climate mod elers who desire a representation of canopy structure in their models.
Page 2 and 3: 3 ) 4: 4. UN r - -- - - ?-- - •
Page 5: TABLE CF CDNTENTS Foreword 1 Prefac
Page 8 and 9: Preface PREFACE 3 The EQS Science P
Page 10 and 11: CHAPTER 5 EQS ScIENCE PLAN Land Eco
Page 12 and 13: CHAPTER 5 CDNTENTS (CDNT.) 5.3.2.3
Page 14 and 15: 2C2 EQS SCIENCE PLAN - CHAPTER 5 la
Page 16 and 17: 204 ECS SCIENCE PI.N - CHAPTER 5 ta
Page 18 and 19: 2D& EDS 9CENcE PLAN - CHAPTER 5 dev
Page 20 and 21: 2D5 EQS SCIENCE FN I-IAPTER S FIGuR
Page 22 and 23: 21 D EQS ScIENcE PLAN - CHAprEp S r
Page 24 and 25: 21 2 EDS SCIENCE PIN - CHAPVER S ab
Page 26 and 27: 214 EDS SCIENcE PN CHAPTER 5 eratio
Page 28 and 29: 21 6 EDS SCIENCE PIN - CHAPTER 5 19
Page 30 and 31: 21 6 EDS SCIENCE PLAN - CHAPTER 6 e
Page 32 and 33: 22J EEJS SCIENCE PL..aN - NAPTER 5
Page 34 and 35: 222 EOS SCIENCE PL..N - CHAPTER 5 f
Page 38 and 39: 228 EDs SCIENCE PN - CHAPTER 5 5.2.
Page 40 and 41: 225 EQS SCIENCE PLAN - CHAPTER 5 se
Page 42 and 43: 23 C ED S S C EN CE P L.AN - CHAPTE
Page 44 and 45: 232 EOS SCIENCE PL.AN - CNAPTER S s
Page 47 and 48: 234 EQS SCIENCE PLAN - CI4AP’rER
Page 49 and 50: 236 EDS SIENE PN - CHAPTER S ing a
Page 51 and 52: 23B EQS SCIENCE PLAN - CHAPrER 5 ma
Page 53 and 54: 240 EQS SCIENcE PL..N CHAPTER 5 5.4
Page 55 and 56: 242 EDS SCIENCE PN CHAp-I-ER 5 FIGu
Page 57 and 58: 244 EDS SCIENCE PL.AN CHAP1ER 5 545
Page 59 and 60: 248 CIS SCIENCE PLAN - CHAP-rER 5 S
Page 61 and 62: 242 EDS SCIENCe PtN - CHAPTER 5 TAB
Page 63 and 64: 25D ECE SCIENCE Pi...aN - CI-lAPTER
Page 65 and 66: 252 EDS SCIENcE PN - CHApTER 5 In a
Page 67 and 68: 254 EQS SCIENtE PL.AN - CHAP-rEP S
Page 69 and 70: 258 EDS SCIENCE PiN - CNAPTER 5 Rob
Page 71 and 72: 252 EDS 5cENcE PN - CHAPTER 5 Chapt

Land_Ecosystems.pdf - S?TE

Create successful ePaper yourself

Delete template?

Save as template?