Exploration of factors limiting biomass estimation by polarimetric radar in tropical forests

Direct inversion of radar return signals for forest biomass estimation is limited by signal saturation at medium biomass levels (roughly 150 ton/ha for P-band). Disturbing factors such as forest structural differences-and, notably, at low biomass levels, terrain roughness, and soil moisture variation-cause further complications. A new and indirect inversion approach is proposed that may circumvent such problems. Using multifrequency polarimetric radar the forest structure can be assessed accurately. Ecological relationships link these structures with biomass levels, even for high biomass levels. The LIFEFORM model is introduced as a new approach to transform field observations of the complex tropical forest into input files for the theoretical UTARTCAN polarimetric backscatter model. The validity of UTARTCAN for a wide range of forest structures is shown. Backscatter simulations for a wide range of forest structures, terrain roughness, and soil moisture clearly show the limitations of the direct approach and the validity of the proposed indirect approach up to very high levels of biomass