Enhanced Retrieval of Forest Biomass Using L-Band SAR and Spaceborne LiDAR Data

Forests, which cover approximately one-third of the planet, play a crucial role in maintaining biodiversity and combating climate change. Aboveground biomass is a key forest characteristic essential for carbon accounting, biodiversity evaluation, and climate change studies. This study aimed to estimate above-ground biomass in tropical forests using L–band Synthetic Aperture Radar (SAR) full polarimetric data combined with canopy height information from the Global Ecosystem Dynamics Investigation (GEDI). The study employed a random forest algorithm, incorporating scattering powers derived from seven-component scattering power decomposition and the forest height obtained from GEDI as input variables. The estimated biomass for the Shivamogga forest was compared with field measurements, yielding a root mean squared error (RMSE) of $20.79 \mathrm{Mg} / \mathrm{ha}$ and a relative RMSE of 18.44%, both of which were within an improved accuracy. Additionally, this study highlights the significant spatial variability in biomass distribution.