MERIS land products: principles and evaluation of performances
Frédéric Baret(1) , Marie Weiss(2)
, Kathy Pavageau(1)
, David Beal(1)
, Beatrice Berthelot(2)
, and Peter Regner(3)
(2) Noveltis, Parc technologique du Canal, Toulouse, France
(3) ESA-ESRIN, Via Galileo, Frascati, Italy
Algorithms have been developed to estimate vegetation biophysical variables (the products) from MERIS top of canopy atmosphere observations. This includes the leaf area index (LAI), the fraction of photosynthetically active radiation absorbed by the canopy (fAPAR), the cover fraction (fCover), and the canopy integrated chlorophyll content (LAI.Cab). The algorithm is based on the training of neural networks over an extensive data set representing a large variability in canopy characteristics made of radiative transfer model simulations (SAIL, PROSPECT). The architecture of the back-propagation neural network was optimized for each biophysical variable and provides good theoretical performances for fAPAR and fCover, LAI and LAI.Cab. This version of the algorithm was implemented within the BEAM toolbox.
MERIS biophysical products including LAI and fAPAR have been compared to other similar products (MODIS LAI-fAPAR, CYCLOPES-VEGETATION LAI-fAPAR and MGVI) over the BELMANIP network of sites for year 2003. Results show good temporal consistency of our product when the clouds are efficiently screened. Some discrepancies are also observed between the several products that are further analysed as a function of surface type. Direct validation was finally achieved by comparison over a set of ground measured values of LAI, fAPAR and fCover thanks to VALERI and additional available sites. Conclusion is drawn on the accuracy of the MERIS biophysical products and the way to improve their performances.