Spatiotemporal evolution of soil water erosion in Ningxia grassland based on the RUSLE-TLSD model

Accurate assessment of grassland soil erosion before and after grazing exclusion and revealing its driving mechanism are the basis of grassland risk management. In this study, the long-term soil erosion in Ningxia grassland was simulated by integrating and calibrating the transport limited sediment delivery (TLSD) function with the revised universal soil loss equation (RUSLE) model. The differential mechanisms of soil loss were explored using the GeoDetector method, and the relative effects of precipitation changes (PC) and human activities (HA) on grassland soil erosion were investigated using double mass curves. The measured sediment discharges from six hydrological stations verified that the RUSLE-TLSD model could reliably simulate water erosion in Ningxia. From 1988 to 2018, the water erosion rate of grassland in Ningxia ranged from 74.98 to 14.98 t⋅ha^-1⋅a^-1, showing an overall downward trend. July to September is the period with the highest of water erosion. The slope is the dominant factor influencing the spatial distribution of water erosion. After grazing exclusion, the net water erosion rate in Ningxia grassland and sub-regions decreased significantly. The double mass curves results show that human activities were the main driver of net erosion reduction. The focus of water erosion control in Ningxia is to control soil erosion in different terrains and protect grassland with slopes greater than 10°.