Janaki Sandamali, Lana Narine

Forest aboveground biomass (AGB) is a crucial indicator for monitoring carbon and requires accurate quantification. This study aimed to advance AGB estimation using open access Earth observation (EO) data and cloud computing, focusing on Alabama, USA. The specific objectives were to: (1) develop a workflow for creating a 30 m forest AGBD map with GEDI, using GEE, (2) evaluate and compare GEDI-derived maps from ecoregion-specific models with estimates derived from a generalized modeling approach, and (3) compare GEDI-derived AGBD map with existing field inventory data and global AGB product. Utilizing GEE, GEDI footprint-level (~25 m diameter) AGBD was extrapolated with EO and ancillary data by employing random forest machine learning. Two estimation approaches were assessed: statewide and ecoregion-specific models for Alabama's six ecoregions. Ecoregion models showed superior accuracy (R²: 0.34–0.73; RMSE: 49.09–53.78 Mg/ha) compared to the statewide model (R²: 0.32; RMSE: 70.48 Mg/ha). Validation with Forest Inventory and Analysis data and the European Space Agency Climate Change Initiative AGBD yielded R² of 0.50 and 0.81, and RMSE of 33.95 Mg/ha and 83.12 Mg/ha, respectively. The study underscores the importance of ecoregion-specific modeling and demonstrates the potential of open-access EO data and platforms in advancing AGB estimation.