Abstract
Three artificial neural network (ANN) processors available as plug-in modules for the Basic ERS & ENVISAT (A)ATSR and MERIS Toolbox (BEAM) were validated in Lake Taihu, China. Mean deviations of reflectance derived from Lakes\Boreal and Lakes\Eutrophic were 10-90%, while reflectance from the FUB-WeW processor showed larger errors. All processors showed underestimates of chlorophyll a (Chl-a), total suspended matter (TSM), and phytoplankton pigment absorption, while particulate scattering values were severely overestimated. None of the readily available MERIS processors is currently able to separate atmospheric and water-leaving radiance over Lake Taihu, while the retrieval of phytoplankton biomass through ANN processors shows promise.