Data for: Soil organic carbon contents of collected soil samples from China's black soil region

The long-term use of cropland and cropland reclamation from natural ecosystems led to soil degradation. This study investigated the effect of the long-term use of cropland and cropland reclamation from natural ecosystems on soil organic carbon (SOC) content and density over the past 35 years. Altogether, 2140 topsoil samples (0–20 cm) were collected across Northeast China. Landsat images were acquired from 1985 to 2020 through Google Earth Engine, and the reflectance of each soil sample was extracted from the Landsat image that its time was consistent with sampling. The hybrid model that included two individual SOC prediction models for two clustering regions was built for accurate estimation after k-means clustering. The probability hybrid model, a combination between the hybrid model and classification probabilities of pixels, was introduced to enhance the accuracy of SOC mapping. Cropland reclamation results were extracted from the land cover time series dataset at a 5-year interval...., ,

长期农田利用与自然生态系统农田开垦共同引发了土壤退化。本研究针对过去35年间，长期农田利用及自然生态系统农田开垦对土壤有机碳（Soil Organic Carbon, SOC）含量与密度的影响展开探究。研究团队在中国东北地区共采集2140份0~20 cm表层土壤样品。研究通过谷歌地球引擎（Google Earth Engine）获取了1985年至2020年的Landsat影像数据，并从与采样时间匹配的Landsat影像中提取各土壤样品的反射率信息。经k-means聚类分析后，针对两个聚类区域分别构建单一SOC预测模型，进而搭建混合模型以实现精准估算。为提升SOC制图精度，本研究引入概率混合模型：该模型由前述混合模型与像元分类概率结合构建而成。农田开垦的相关结果提取自时间间隔为5年的土地覆盖时间序列数据集。