Resolving Vegetation Condition and Biogeochemical Processes Using Hyperspectral-brdf Inverse Modeling

Complex variation in terrestrial biogeochemical cycles results from climatic, edaphic, biotic and human processes operating at multiple spatial and temporal scales. This variation cannot easily be resolved at the landscape, regional or global level without an integrated effort involving field research, modeling and remote sensing. To date, the role of remote sensing in terrestrial biogeochemical research has been limited to land-cover change analysis and estimation of two functional properties of vegetation: leaf area index (LAI) and fractional photosynthetically active radiation absorption (fAPAR). Traditional land-cover remote sensing provides the minimum information on the extent of major vegetation types needed to constrain ecosystem models to actual conditions (Wessman and Asner, 1997). Some ecosystem models also utilize remotely sensed estimates of LAI and fAPAR to derive plant carbon uptake or net primary productivity (NPP), which then feeds into soil carbon, nutrient and water algorithms that calculate a wide range of biogeochemical fluxes (Running et al, 1994; Field et al, 1995).