The damage caused by soil erosion to global ecosystems is undeniable. However, traditional research methods often do not consider the unique soil characteristics specific to China and rainfall intensity variability in different periods on vegetation, and relatively few research efforts have addressed the attribution analysis of soil erosion changes in tropical islands. Therefore, this study applied a modification of the Chinese Soil Loss Equation (CSLE) to evaluate the monthly mean soil erosion modulus in Hainan Island over the past two decades, aiming to assess the potential soil erosion risk. The model demonstrated a relatively high R2, with validation results for the three basins yielding R2 values of 0.77, 0.64, and 0.78, respectively. The results indicated that the annual average soil erosion modulus was 92.76 thm-2year-1, and the monthly average soil erosion modulus was 7.73 thm-2month-1. The key months for soil erosion were May to October, which coincided with the rainy season, having an average erosion modulus of 8.11, 9.41, 14.49, 17.05, 18.33, and 15.36 thm-2month-1, respectively. September marked the most critical period for soil erosion. High-erosion-risk zones are predominantly distributed in the central and eastern sections of the study area, gradually extending into the southwest. The monthly average soil erosion modulus increased with rising elevation and slope. The monthly variation trend in rainfall erosivity factor had a greater impact on soil water erosion than vegetation cover and biological practice factor. The identification of dynamic factors is crucial in areas prone to soil erosion, as it provides a scientific underpinning for monitoring soil erosion and implementing comprehensive water erosion management in these regions.

Rain on snow (ROS) is a complex phenomenon leading to repeated flooding in many regions with a seasonal snow cover. The potential to generate floods during ROS depends not only on the magnitude of rainfall but also on the areal extent of the antecedent snow cover and the spatio-temporal interaction between meteorologic and snowpack properties. The complex interaction of these factors makes it difficult to accurately predict the effect of snow cover on runoff formation for an upcoming ROS event. In this study, the detailed physics-based snow cover model SNOWPACK was used to assess the influence of snow cover properties on converting rain input to available snowpack runoff during 191 ROS events for 58 catchments in the Swiss Alps. Conditions identified by the simulations that led to excessive snowpack runoff were a large snow-covered fraction, spatially homogeneous snowpack properties, prolonged rainfall events, and a strong rise in air temperature over the course of the event. These factors entail a higher probability of snowpack runoff occurring synchronously within the catchment, which in turn favours higher overall runoff rates. The findings suggest that during autumn and late spring, flooding due to ROS is more likely to happen, whereas during winter a coincidence of the above conditions in the study area is quite rare. For example, an autumn event which occurred in October 2011 resulted from a combination of spatially homogeneous snowpack conditions following a recent snowfall and high, but not exceptional rainfall, and led to major flooding. The results of this study provide key factors to assess in advance of an incoming ROS event and emphasize the importance of detailed snow monitoring for flood forecasting in snow-affected watersheds.