LAUSR

Rainfall erosivity is an essential factor affecting soil erosion, which is expected to change under global climate change. Despite the existence of numerous rainfall erosivity models, there remains a scarcity of research focusing on the accuracy of multi‐time scale models. In this study, the subtropical regions of central China (Hubei Province) were selected, where the simulation performance of six widely employed rainfall erosivity models was investigated using daily precipitation data from 70 meteorological stations spanning from 2000 to 2020. Using the optimal model, Kriging interpolation and the Mann–Kendall test revealed significant temporal and spatial variations in rainfall erosivity and density. The results show that: (1) the daily rainfall erosivity model was more suitable for simulating rainfall erosivity in Hubei Province. (2) The mean annual rainfall erosivity in Hubei Province was 5894.25 MJ·mm·ha−1·h−1·a−1, with large variations across regions. (3) Rainfall erosivity and erosivity density showed significant differences between different seasons, and soil erosion was most likely to occur in summer (June, July and August). (4) The spatial distribution pattern of rainfall erosivity and erosivity density was highly consistent: the long‐term high levels of rainfall erosivity and erosivity density were in Xianning City, southeastern Hubei Province, and the soil erosion risk was high. The findings of this study offer valuable insights into the selection of rainfall erosivity models in subtropical mountainous and hilly areas and provide a reference for assessing soil erosion risk and formulating control measures.