Temporal and spatial variations in the mean residence time of soil organic carbon and their relationship with climatic, soil and vegetation drivers.

Although efforts to understand the drivers of the soil organic carbon (SOC) in terrestrial ecosystems have paid off, little is known about the temporal and spatial variations in the mean residence time (MRT) of SOC and how MRT varies with the climatic, soil, and vegetation drivers. In this study, the MRT of SOC from 1376 samples of the 0–20 cm soil layer from the main global biomes was derived based on the simultaneously measured SOC pool and soil carbon fluxes. We also evaluated the factors controlling the variations in the MRT of SOC. The results indicate that, across the different biomes, the MRT of SOC (ranging from 0.47 to 56.75 yr) exhibits large variations at the global scale and shows strong quadratic relationship with latitude. The MRT of SOC exhibits a significant and negative trend from the 1980s to the 2010s. The MRT of SOC shows strong negative and nonlinear relationships (P < 0.001) with mean annual temperature (MAT) and annual precipitation (AP). The MRT of SOC is also highly significantly (P < 0.001) correlated with soil total nitrogen (STN) (R 2 = 0.281), bulk density (BD) (R 2 = 0.097), tree age (TA) (R 2 = 0.099), and litter fall (LF) (R 2 = 0.076), respectively. MAT, AP, STN, BD, TA, and LF are the dominant drivers of the MRT of SOC in global models; however, the model including MAT, AP, STN, and BD (T&P&N&B) performs best among the models with different predictors. Moreover, the T&P&N&B model performs better for simulating the MRT of SOC in grassland than in other biomes. Further analysis indicates that simulating the MRT of SOC using biomes-specific models well explains the variations in the MRT of SOC, with an R 2 value of 0.708. • The MRT of SOC shows strong quadratic relationship with latitude. • The MRT of SOC exhibits a significant and negative trend from 1980s to 2010s. • The variations in the MRT of SOC are controlled by climatic, soil and vegetation drivers. • The biome-specific models well explain the spatial and temporal variations in the MRT of SOC. [ABSTRACT FROM AUTHOR]