Global and regional soil organic carbon estimates: Magnitude and uncertainties

Globally, soil is the largest terrestrial carbon reservoir. Robust quantification of soil organic carbon (SOC) stocks in existing global observation-based estimates avails accurate predictions in carbon climate feedbacks and future climate trends. In this study, we investigated magnitude and distribution of global and regional SOC estimates (i.e., density and stocks), based on five widely used global gridded SOC datasets (HWSD, WISE30sec, GSDE, SoilGrids250m, and GSOCmap), a regional permafrost dataset from Mishra et al. (UM2021), and a global-scale soil profile database (the World Soil Information Service soil profile database, WoSIS) reporting measurements of a series physical and chemical edaphic attributes. Our analyses show that the magnitude and distribution of SOC vary widely among datasets, with certain datasets showing region-specific robustness. At the global scale, SOC stocks at the top 30 and 100 cm are estimated to be 828 (range: 577–1171) and 1873 (range: 1086–2678) Pg C, respectively. The estimates from GSDE, GSOCmap, and WISE30sec are comparable, and those of SoilGrids250m and HWSD are at the upper and lower ends. The spatial SOC distribution varies greatly among datasets, especially in the northern circumpolar and Tibetan Plateau permafrost regions. Regionally, UM2021 and WISE30sec perform well in the northern circumpolar permafrost regions, and GSDE performs well in China. SOC estimated by different datasets also show large variabilities across different soil layers and biomes. The discrepancies are generally smaller in 0–30 cm than in 0–100 cm soils. The datasets demonstrate relatively higher agreement in grasslands, croplands, and shrublands/savannas than other biomes (e.g., wetlands). The users should be mindful of the gaps by region and by biome while choosing the most appropriate SOC dataset for specific uses. Large uncertainties in existing global gridded SOC estimates are generally derived from soil sampling density, diverse sources and mapping methods for soil datasets. We call for future efforts for standardizing soil sampling efforts, cross-dataset comparison, proper validation, and overall global collaboration to improve SOC estimates.