Estimating biomass consumption and carbon emissions by integrating GEDI, Sentinel-1, and Sentinel-2 data

Accurately assessing biomass consumption and carbon emissions from wildfires is crucial for understanding their impact on global climate change and the carbon cycle. The Global Ecosystem Dynamics Investigation (GEDI) mission provides spaceborne light detection and ranging (LiDAR) measurements of global forests, offering valuable opportunities for monitoring forest dynamics and carbon emissions. This study presents a three-phase model integrating GEDI canopy height products with Sentinel-1 and Sentinel-2 data to estimate biomass consumption and carbon emissions from the Beachie Creek Fire. In Phase 1, canopy height models (CHMs) for pre-fire (R2 = 0.72, RMSE = 8.87 m) and post-fire (R2 = 0.63, RMSE = 9.63 m) were developed using the LightGBM algorithm with GEDI RH95 data and multi-source remote sensing variables. The CHMs were used to predict pre- and post-fire canopy height. In phase 2, a power-low allometric equation was fitted to estimate aboveground biomass (AGB) from canopy height, with RMSEs of 54.85 Mg/ha (pre-fire) and 61.82 Mg/ha (post-fire). In phase 3, carbon emissions were derived from AGB loss using a conversion equation. Compared with other studies, our results better capture the spatial heterogeneity of biomass consumption and carbon emissions, providing more reasonable estimates in areas with high burn severity. The Beachie Creek Fire resulted in 8.81 Tg (±2.9 Tg) of biomass consumption and 14.54 Tg (±4.78 Tg) of carbon emissions. This approach provides an efficient framework for assessing burning emissions across large and remote regions, contributing to climate change mitigation strategies and informing forest conservation and fire management policies.