Mapping the Forest Height by Fusion of ICESat-2 and Multi-Source Remote Sensing Imagery and Topographic Information: A Case Study in Jiangxi Province, China

Forest canopy height is defined as the distance between highest point of tree and ground, which considered to be a key factor in calculating above-ground biomass, leaf area index, carbon stock. Large-scale forest monitoring can provide scientific information on deforestation degradation policymakers. The Ice, Cloud, Land Elevation Satellite-2 (ICESat-2) was launched 2018, with Advanced Topographic Laser Altimeter System (ATLAS) instrument taking task mapping transmitting data photon-counting LiDAR, offers an opportunity obtain global height. To generate high-resolution map Jiangxi Province, we integrated ICESat-2 multi-source remote sensing imagery, including Sentinel-1, Sentinel-2, Shuttle Radar Topography Mission, age Province. Meanwhile, develop four extrapolation models by random (RF), Support Vector Machine (SVM), K-nearest neighbor (KNN), Gradient Boosting Decision Tree (GBDT) link ICESat-2, spatial feature imagery. results show that: (1) moderately correlated age, making it potential predictor for mapping. (2) Compared GBDT, SVM, KNN, RF showed best predictive performance coefficient determination (R2) 0.61 root mean square error (RMSE) 5.29 m. (3) Elevation, slope, red-edge band (band 5) derived from Sentinel-2 were significantly dependent variables model. Apart that, one that relied on. In contrast, backscatter coefficients texture features Sentinel-1 not sensitive (4) There significant correlation predicted measured field measurements (R2 = 0.69, RMSE 4.02 m). nutshell, indicate method utilized this work reliably distribution at high resolution.