Spatiotemporal assessment on above-ground biomass and carbon stock dynamics in Bhawal National Park of Bangladesh using Gedi Lidar and Sentinel-2 data

Abstract: Forests are essential for regulating the global climate and carbon balance. Monitoring Above-ground Biomass (AGB) and its Density (AGBD) at high resolution is needed for sustainable forest management, biodiversity conservation, and carbon accounting to support climate change mitigation. The present study aimed to assess the spatiotemporal changes in above-ground biomass (AGB) and carbon stock in Bhawal National Park (BNP) between 2019 and 2024. This involved integrating data from the Global Ecosystem Dynamics Investigation (GEDI) and Sentinel-2. A Random Forest (RF) regression model was applied to estimate AGB by combining the GEDI Level 4A footprint data with spectral and topographic variables derived from Sentinel-2 and digital elevation model (DEM). The model showed good predictive performance, with R² values above 0.80 for both years. The results revealed a significant decline in total AGB from 355,242.50 Mg in 2019 to 307,035.78 Mg in 2024, corresponding to a carbon stock loss of 22,657.16 Mg. Spatial analysis showed that high-biomass zones have shrunk, while low and moderate-biomass areas have expanded, especially along forest edges and in degraded areas. Field validation confirmed a moderate match between estimated and observed AGB values. The study identified deforestation, land-use change, and fragmentation as the main drivers of biomass decline. These findings highlight ongoing forest degradation in one of Bangladesh’s most important Sal forest ecosystems. This emphasizes the urgent need for improved forest monitoring and conservation strategies. Integrating GEDI and optical satellite data offers a cost-effective approach for large-scale forest assessment, supporting national carbon accounting, REDD+ implementation, and sustainable forest management. 

Keywords: Above-ground biomass, carbon, GEDI LiDAR, Sentinel-2, forest