Integration of RUSLE Modelling and Sentinel‐1 and Sentinel‐2 Imagery to Assess Bare Ground Evolution Over a Typhoon Season in the Gaoping Basin, Taiwan

Abstract

ABSTRACT The aim of this project was to establish a set of criteria for predicting where in the Gaoping catchment in Taiwan erosion would be most likely to occur following sustained heavy rain and/or typhoon conditions. The Revised Universal Soil Loss Equation (RUSLE) model was tested with mean rainfall between 2002 and 2018 and for the 2024 rainfall. This model is based on five criteria: rainfall ( R factor), slope ( LS factor), soil characteristics ( K factor), land cover ( C factor) and land use ( P factor). Bare ground and land cover changes to bare ground between 2024 and 2025 were quantified using Sentinel‐2 imagery and GIS tools from ArcGIS. The results showed that following the Krathon typhoon on 2 October, bare ground erosion increased by 55%, 72% and 80% in the Baolai, Duona and Fuxing areas, respectively, compared with the mean bare ground erosion calculated between 2002 and 2018. Within the Gaoping catchment, the factors that most affected erosion were slope ( LS factor with 143% relative deviation) and vegetation cover ( C factor with 135% relative deviation). Further, the RUSLE model was integrated with GIS mapping to observe and analyse, at 10‐m resolution, the spatiotemporal development of soil erosion over a 1‐year period. The RUSLE model performed well in the upper catchment and correctly identified the high erosion risks. However, calculation uncertainties were as high as 36% in areas with little slope. Our analyses were able to predict the erosion changes that occurred in response to the rainfall events of 2024–2025. Therefore, our approach can be used to estimate future soil erosion caused by typhoons and storm rainfall in the Gaoping River catchment.