High-Resolution Mapping of Soil Organic Carbon in the Northeast China Black Soil Belt Using a Geographically Weighted Gaussian Mixture Model

Soil organic carbon (SOC) is a key indicator of soil fertility and carbon sequestration, essential in regulating biogeochemical cycles and supporting sustainable agriculture. The black soil region of Northeast China constitutes one of the world's four largest black soil zones and is critical for global food security. However, the lack of timely, high-resolution spatial distribution maps for SOC significantly hinders understanding of SOC spatial variability. To address this, we developed a geographically adjusted Gaussian mixture model (GAGMM). This model effectively captures local SOC variation characteristics and demonstrates strong robustness. The GAGMM-XGBoost hybrid model achieved a prediction accuracy (R²) of 0.8, representing an 11.6% improvement over the conventional global XGBoost model. Furthermore, leveraging the optimal GAGMM-XGBoost model, we generated a high-resolution (30 m) SOC spatial distribution map for the Northeast China black soil region. This map provides a robust data foundation for comprehensive evaluation of regional soil resource status. Please cite this dataset as: Tan, Q., & Jing, G. (2025). High-Resolution Mapping of Soil Organic Carbon in the Northeast China Black Soil Belt Using a Geographically Weighted Gaussian Mixture Model (V1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15770209