A 30-m aboveground biomass dataset for China in 2020

准确估算地上生物量（AGB）对于了解陆地碳循环和制定气候政策至关重要。中国地形多样，植被类型丰富，是全球碳储量的重要组成部分。然而，现有的AGB产品在空间分辨率、数据一致性和可获取性方面存在诸多局限，难以全面反映中国不同植被类型的生物量格局。本研究提出并发布了一个高分辨率（30 m）、全国覆盖、整合多种植被类型的2020年中国地上生物量密度（AGBD）数据集。该产品基于激光雷达LIDAR、光学、雷达等多源可开放获取的遥感影像构建而成，实现了对森林、草地、灌木和耕地等多种生态系统的统一建图。我们采用地面样本、国家统计年鉴及空间分布特征对数据集进行了系统验证，并将其与多个现有AGB产品进行比较。结果显示预测模型产生了平均R2 = 0.85和平均RMSE = 31.26 Mg/ha的准确度，中国植被区域的总碳储量为20.20Pg。该数据集为中国的生物量研究提供了一个更新、全面的基线，可能有助于碳核算和生物多样性监测。

Accurate estimation of aboveground biomass (AGB) is critical for understanding terrestrial carbon cycles and formulating climate policies. China, with diverse topography and abundant vegetation types, is a significant component of the global carbon storage. However, existing AGB products have various limitations in terms of spatial resolution, data consistency and accessibility, making it difficult to fully reflect the biomass patterns of different vegetation types across the country. This study proposes and releases a high-resolution (30 m), nationally covered 2020 aboveground biomass density (AGBD) dataset for China that integrates multiple vegetation types. This product is constructed based on multiple openly accessible remote sensing images including LiDAR, optical and radar data, realizing unified mapping of various ecosystems such as forests, grasslands, shrubs and croplands. We systematically validated this dataset using ground samples, national statistical yearbooks and spatial distribution characteristics, and compared it with multiple existing AGB products. The results show that the prediction model achieved an average R² of 0.85 and an average RMSE of 31.26 Mg/ha, with the total carbon storage of China's vegetation regions reaching 20.20 Pg. This dataset provides an updated and comprehensive baseline for biomass research in China, and may facilitate carbon accounting and biodiversity monitoring.