深度学习方法下GEDI数据的天然云杉林地上生物量反演

s the largest carbon reservoir on land, forests play a crucial role in human life and development.Understanding the dynamic changes in forest resources and modernizing their sustainable development iscurrently a significant research focus. This study focuses on natural Picea forests in the Tianshan Mountains anduses ground measurement data, helicopter airborne LiDAR point cloud data, and Global Ecosystem DynamicsInvestigation (GEDI) data to construct a multisource fusion data framework. By utilizing deep learningalgorithms within the AutoKeras framework, the study aims to predict the regression model of multiple relativeheight quantiles of GEDI data and their aboveground biomass in the study area, thereby validating the feasibilityof GEDI data for large-scale aboveground biomass retrieval. The main conclusions are as follows: (1) GEDI dataare highly feasible for estimating forest aboveground biomass. Through automated deep learning algorithms andtraining and verification sets, the overall data achieve a coefficient of determination (R2) of 0.69, 0.63, and 0.67,respectively, along with a mean absolute error of 3.73 mghm2, 4.22 mghm2, and 3.89 mghm2, demonstratinghigh prediction accuracy. (2) Helicopter LiDAR, an intermediate technology for estimating aboveground biomassusing GEDI data, exhibits a single tree recognition accuracy of over 0.75 across the study area. The studysuccessfully utilizes multimodal data fusion to quantitatively describe the structural parameters of the single treefoundation in the study area while verifying the potential of GEDI data for obtaining forest aboveground biomass.Moreover, the study provides a theoretical basis for estimating carbon sources and sinks, biomass, stock, forestmanagement, biodiversity protection, and other projects in similar areas, offering essential guidance, andfundamental data support.