中国高精度森林土壤有机碳数据集

本数据集为中国森林土壤有机碳（FSOC）的90米高分辨率三维空间分布图，基于“高精度数字森林土壤数据库构建”项目，整合8,709个国家典型森林土壤剖面数据，采用机器学习驱动的预测性土壤制图框架生成。数据覆盖0-20、20-40、40-60、60-100 cm四个土壤深度，包含每个深度的FSOC含量预测值（单位：g kg⁻¹，GeoTIFF格式，比例因子100）及其90%预测区间的不确定性上下限量化图层，共12个地理空间文件（如0-20 cm层对应FSOC_020_pred.tif、FSOC_020_q05.tif和FSOC_020_q95.tif），采用WGS 1984 Albers坐标系。This dataset is a 90-meter high-resolution 3D spatial distribution map of Forest Soil Organic Carbon (FSOC) in China. It was developed based on the "Construction of High-precision Digital Forest Soil Database" project, integrating 8,709 national typical forest soil profile datasets, and generated via a machine learning-driven predictive soil mapping framework. The dataset covers four soil depth layers: 0–20, 20–40, 40–60, and 60–100 cm. For each layer, it contains predicted FSOC concentrations (unit: g kg⁻¹, stored in GeoTIFF format with a scale factor of 100), alongside quantifiable uncertainty upper and lower limit layers for its 90% prediction interval, resulting in a total of 12 geospatial files. For instance, the 0–20 cm layer corresponds to three files: FSOC_020_pred.tif, FSOC_020_q05.tif and FSOC_020_q95.tif. All files use the WGS 1984 Albers coordinate system.

搜集汇总

背景与挑战

背景概述

中国高精度森林土壤有机碳数据集是一个90米高分辨率的三维空间分布图，基于8,709个典型森林土壤剖面数据，采用机器学习框架生成。它覆盖0-20、20-40、40-60和60-100 cm四个土壤深度，每个深度提供有机碳含量预测值及其90%预测区间的不确定性量化图层，共12个GeoTIFF格式文件，采用WGS 1984 Albers坐标系，适用于精准的土壤碳储量和环境研究。