Remote Sensing and Process-Based Modelling

CURRAN, Paul (P.Curran@soton.ac.uk), Department of Geography, University of Southampton,Highfield, Southampton SO17 1BJ, U.K.; DAWSON, Terry P. (terry.dawson@ecu.ox.ac.uk), Environmental Change Unit, University of Oxford, Mansfield Road, Oxford OX1 3TB, U.K.; WICKS, Toby E. (tew@soton.ac.uk), Department of Geography, University of Southampton, Highfield, Southampton SO17 1BJ, U.K.

Key Words: remote sensing, process-based models, LIBERTY, FLIGHT, FOREST-BGC, BOREAS

The use of remote sensing in ecology is predicated both on a need to understand the interaction of radiation with vegetation, and to couple this with aspects of environmental understanding. A key to the ecological use of remote sensing has become the application of process models that enable us to understand the interaction of radiation with a leaf and a canopy and to use remotely sensed data to drive an a-spatial ecosystem simulation model over large areas of terrain. By way of illustration, this paper will focus on the application of three process models to contemporary ecology. The first is the Leaf Incorporating Biochemistry Exhibiting Reflectance and Transmittance Yields (LIBERTY) model that can be used to determine the effect of key variables (e.g., water or chlorophyll content) on leaf spectra. The second is the Forest LIGHT interaction (FLIGHT) model that can be coupled with LIBERTY to determine the effect of key variables (e.g., water or chlorophyll content) on canopy spectra. The model is the FOREST-Bio-Geo-Chemical (FOREST-BGC) model that enables, via the coupling of remotely sensed data with ecosystem understanding, the spatial estimation of vegetation productivity.

The utility of these three models will be illustrated, by reference to research undertaken, as part of an international environmental research programme called BOReal Ecosystem Atmosphere Study (BOREAS). Specifically the LIBERTY model will be used to investigate the spectral sensitivity of conifer needles to changes in foliar biochemical content. The combined FLIGHT and LIBERTY models will be used to determine the degree to which foliar biochemical information is retained in the spectrum of a conifer canopy. The FOREST-BGC model will be driven, using leaf area index estimates from the National Oceanic and Atmospheric Administration's (NOAA) Advanced Very High Resolution Radiometer (AVHRR) data and used to estimate the net primary productivity of boreal forest.

The paper ends with discussions of the modelling approaches mentioned above, to the understanding and use of remotely sensed data from European Space Agency's (ESA) Envisat satellite.