Mapping Forest Aboveground Biomass Using Multisource Remotely Sensed Data

The majority of the aboveground biomass on Earth’s land surface is stored in forests. Thus, forest plays a critical role global carbon cycle. Yet accurate estimate (FAGB) remains elusive. This study proposed new conceptual model to map FAGB using remotely sensed data from multiple sensors. model, which provides guidance for selecting data, based principle estimating ground allometry, needs species, diameter at breast height (DBH), and tree as inputs. Based we used multiseasonal Landsat images provide information about species composition forests area, LiDAR canopy height, image texture ratio two spatial resolutions crown size, related DBH. Moreover, added RaDAR volume model. All layers were fed Random Forest (RF) regression was carried out eastern North Carolina. We USFS Inventory Analysis plots train test performance. best achieved an R2 0.625 with root mean squared error (RMSE) 18.8 Mg/ha (47.6%) “out-of-bag” samples 30 × m resolution. top five most important variables include 95th, 85th, 75th, 50th percentile heights points their standard deviations 85th heights. Numerous features Sentinel-1 C-Band SAR, 8 imagery along very high-resolution selected. But importance metrics dwarfed all other variables. More tests places broader range more diverse are needed evaluate input