Estimating Spatiotemporally Continuous GEDI Aboveground Biomass Density During 2015–2020 From Multisource Data Using Machine Learning and Deep Learning

Wang, Xia and Zhang, Yihang and Atkinson, Peter and Zhang, Kerong (2026) Estimating Spatiotemporally Continuous GEDI Aboveground Biomass Density During 2015–2020 From Multisource Data Using Machine Learning and Deep Learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 19. pp. 3839-3857. ISSN 1939-1404

Abstract

Accurate estimation of forest aboveground biomass density (AGBD) is essential for quantifying terrestrial carbon exchange. NASA's global ecosystem dynamics investigation (GEDI) LiDAR mission has ushered in a new era of high-quality AGBD estimation from space. Owing to GEDI's LiDAR data collection, one of the GEDI AGBD products is presented as 1 km spatial discontinuity grid, limiting practical application. This study proposes a method based on machine learning and deep learning to estimate spatiotemporally continuous GEDI AGBD and its associated uncertainty (standard error) during 2015–2020 using multisource optical, synthetic aperture radar, LiDAR, topographic, and climatic data. Moderate-resolution imaging spectroradiometer (MODIS) vegetation continuous field (VCF) was used as an indicator to select unchanged GEDI AGBD samples in 2020 for each year from 2015 to 2020, resulting in 1800–2400 samples per year. The selected annual training samples were then used to estimate annual spatially continuous GEDI AGBD and uncertainty. Experiments in China’s Han River Basin demonstrated that integrating all available datasets resulted in a more accurate spatial continuous GEDI AGBD map in 2020 compared to those using any single dataset. Convolutional neural network (CNN) considering spatial neighboring information (CNN_spatial) outperformed 1-D CNN and four benchmark machine learning methods of extreme learning machines, generalized regression neural network, support vector regression, and random forest with an R2 of 0.8232 and root-mean-square error of 27.7557. Using the unchanged GEDI AGBD training samples resulted in more accurate GEDI AGBD maps (R2>0.82) during 2015–2020 than using the training samples from 2020 alone. From 2015 to 2020, 2.16% of AGBD pixels in Han River Basin experienced a relatively significant increase, while 1.24% showed a relatively significant decrease. Compared with the ESA_CCI and China’s AGBD products, our estimates achieve better accuracy relative to the field plot data. The proposed method offers a solution to generate high-quality spatiotemporally continuous GEDI AGBD in large-scale complex forest landscapes.