The assessment of synergetic effects of airborne LiDAR, CCD and hyperspectral data for above-ground biomass estimation

The estimation of aboveground biomass (AGB) is critical for carbon accounting and monitoring, especially in the subtropical forests which account for approximately a quarter of the area of China. In this study, the airborne LiDAR, CCD and hyperspectral data acquired simultaneously from the LiCHy (LiDAR, CCD and Hyperspectral) Airborne Observation System were assessed separately or in combination to estimate ABG within a northern subtropical study site. Overall, the results demonstrated that most of the LiDAR, CCD and hyperspectral metrics refined by the PCA analysis have relatively low correlation coefficient with each other (r <; 0.7). The model coupling LiDAR and hyperspectral metrics (R²=0.87, rRMSE=11.67%) had a relatively higher performance than the models fitted by LiDAR only (R²=0.85, rRMSE=12.31%). The combo model which integrated LiDAR, hyperspectral and texture metrics had the highest accuracy for AGB estimation (R²=0.88, rRMSE=11.47%). In addition, the point cloud metrics of 95th percentile height (h₉₅), 25th percentile height (h₂₅), hyperspectral metric of Simple ratio (SR), Normalized difference vegetation index (NDVI) and texture metrics of entropy (EN), second moment (SM) have the relative importance for AGB estimations.