A modified soil organic carbon density model for a forest watershed in southern China

Abstract

In the context of global climate change, correctly estimating soil organic carbon (SOC) stocks is significant. Because SOC density is the basis for calculating total SOC, exploring the spatial distribution of SOC density is more important. In this study, a typical forest watershed in southern China was analysed. An established exponential model that combined the soil erosion, topography, and average annual rainfall in the region to estimate SOC density with varying soil depth was modified by simulated rainfall experiments and 137Cs (Caesium-137) tracer soil erosion techniques. Thus, a modified exponential model for the SOC density in southern China was established. The results showed that the correlation coefficient (R2) reached 0.870 for the linear regression analysis of the simulated and measured SOC densities. The differences between the measured and simulated SOC densities in different soil layers (0–60cm) all passed the independent sample t-test. Additionally, the Nash-Sutcliffe coefficient for the simulated and measured SOC densities was 0.97 in the forest watershed. Furthermore, the application of the modified exponential model showed that the measured SOC densities were in good agreement with the simulated SOC densities in the different forest areas tested. These results illustrated that the modified exponential model could be effectively used to simulate the vertical distribution of SOC density in southern China. Because the parameters in the modified exponential model were easy to obtain, this modified model could be applied to simulate the vertical distribution of the SOC density in different geomorphological areas. Therefore, the results of this study will help to understand the global carbon cycle and provide valuable information for constructing the ecological environment of various landscapes.