Estimation of fraction of Absorbed Photosynthetically Active Radiation from multiple satellite data

Fraction of Absorbed Photosynthetically Active Radiation (FPAR) is a critical input parameter in many climate and ecological models. An accuracy of ±0.1 in FPAR is considered acceptable in the applications. However, most of current FPAR products, such as Moderate-Resolution Imaging Spectroradiometer (MODIS) and Multi-angle Imaging SpectroRadiometer (MISR), do not fulfill the accuracy requirement yet. The objective is to develop a new radiative transfer model for FPAR estimation, with broadened surface reflectance database from the time series of twelve years' reflectance data. The model proposed here could successfully identify growing season and the time series curve of estimated FPAR was smooth over years. The R2between estimated FPAR and in situ measurements was improved compared to existing FPAR products.