Global distribution and predictors of the mineral-associated to total soil organic carbon ratio: an indicator of soil carbon stability
Jiangjia Zhao, Budiman Minasny, Raj Setia, Zhenghu Zhou, Chengjie Ren, Asim Biswas, Zhongkui Luo, Lianqing Zhou, Zhou Shi, Songchao Chen
Abstract
Soil organic carbon (SOC) is the largest actively cycling reservoir of terrestrial carbon (C), and its stability plays a crucial role in carbon-climate feedback. However, the global distribution and environmental controlling factors of SOC stability across Earth's biomes remain unclear due to data limitation and complex interactions between climate, soil properties and land cover types. This study synthesized 11,495 globally distributed observations to the proportion of mineral-associated organic carbon (MAOC) in SOC (f MAOC ), an indicator of soil carbon stability, in topsoil (0-30 cm) and subsoil (30-100 cm) across diverse climate zones and land cover types. We identified the key factors using forward recursive feature selection (FRFS) with the Random Forest model, and generated global distribution of f MAOC in topsoil and subsoil at 1 km resolution. Our results showed that a greater proportion of SOC in Earth's mineral soils is stored as MAOC in the subsoil (78.62%) compared to the topsoil (66.62%). Across climate zones and land cover types, temperate forests exhibited notable variation between topsoil (59.65%) and subsoil (73.73%), whereas cropland showed minimal difference, with subsoil only 6.54% higher than topsoil. The tundra in Boreal was the most vulnerable to C loss, with subsoil showing lower f MAOC than its topsoil counterpart. The f MAOC was jointly influenced by climate, soil properties, and land cover types, clay (contributed for 12.12% in topsoil and 14.83% in subsoil) and mean annual temperature (8.50% in topsoil and 14.07% in subsoil) were the predominant controlling factors at two depth intervals. Meanwhile, nitrogen deposition (8.63%) was important in determining f MAOC in topsoil, while elevation (12.63%) was important in subsoil. A high spatial heterogeneity and regional variation of f MAOC were observed at a global scale. These findings underscore the importance of quantifying and modeling the f MAOC to accurately predict the response of SOC storage to global change, ultimately informing strategies for SOC sequestration, climate change mitigation, and sustainable land management.