Satellite remote sensing provides the only means of directly observing the entire surface of the Earth at regular spatial and temporal intervals. Key Earth system variables can be obtained from satellite data by integrating appropriate processing, interpretation, and modeling. For example, the amount of photosynthetically active radiation absorbed by plants (APAR) and land surface albedo can be inferred from remotely sensed optical measurements. Radiative transfer model inversion exploits the dependence of reflectance on the relative source-sensor geometry to estimate surface parameters. In contrast, geometrical effects are suppressed in most other approaches. We present an algorithm for the retrieval of fractional APAR (fAPAR), albedo, and other parameters from AVHRR (advanced very high resolution radiometer) reflectance measurements by inverting a modified version of the SAIL (scattering by arbitrarily inclined leaves) canopy radiative transfer model. The model is inverted using an effective bidirectional reflectance factor (BRF) distribution created by aggregating AVHRR data into cells of size comparable to those used in current terrestrial biosphere models (50 × 50 km). Successful inversion results over an area in central Africa are presented and compared with a vegetation index-based analysis and other satellite data. The procedure also provides unique information on phenology derived from timing of changes in leaf optical properties and canopy structure. Our methods are unique in that they explicitly incorporate a priori ecological knowledge in the choice of model parameters and constraints. This approach can eventually be employed at pixel resolution with the EOS sensors, MODIS (moderate-resolution imaging spectrometer) and MISR (multiangle imaging spectro-radiometer).