Quantitative analysis of impact of human activities on soil erosion using the RUSLE model in a typical karst area in Guizhou, China

Human activities have significant influence on soil erosion in karst areas. The spatial and temporal evolution of soil erosion in Guizhou Province was evaluated using the Revised Universal Soil Loss Equation (RUSLE), which revealed an increasing trend in the initial data analysis for the soil erosion modulus. To disclose the impact of human activities on regional soil erosion, the soil erosion in 2000, 2010, and 2020 was analyzed. The results show the following: (1) The average values of the soil erosion modulus in the study area for 2000, 2010, and 2020 were 4.479, 4.945, and 5.806 t·hm^-2·a^-1, respectively; when considering human activities without the influence of rainfall erosivity, these values were 4.679, 4.963, and 4.799 t·hm^-2·a^-1. The influence of human activities on soil erosion is gradually becoming a positive force. (2) According to the Spearman regression analysis, the top four factors related to soil erosion in 2000 and 2010 were soil loss risk (E, 0.721 and 0.737), anti-erosion factors (Pr, - 0.236 and - 0.221), rock exposure rate (0.222 and 0.279), and altitude (0.210 and 0.195). In 2020, the top four factors were Pr (0.725), land surface temperature (LST, 0.268), NDVI (- 0.232), and E (0.186). In the first two stages, soil erosion is closely related to natural factors, while in 2020, soil erosion is more closely related to human activities. (3) The geographically weighted regression (GWR) showed the highest range of regression coefficients for Pr (150), followed by E and NDVI (25), rock exposure rate (10), and land surface temperature (LST) (1.5). The rainfall erosivity is increasing annually as a consequence of global climate change. This rise in rainfall erosivity has resulted in a corresponding increase in soil erosion in the study area, which obscures the positive impact of human activities in the reduction of soil erosion.