Modeling soil erosion and sediment yield under climate change: a comparison of RUSLE and MUSLE integrated with SDR using variable soil data resolutions.

Abstract

Climate change exacerbates soil erosion by intensifying extreme events, which weaken soil structure and increase surface runoff. In developing countries, resource limitations constrain mitigation efforts, accelerating land degradation, reducing agricultural productivity, and increasing reservoir sedimentation. The accuracy of erosion modeling under these conditions is also influenced by the resolution of input data, particularly soil data, which affects spatial representation and predictive reliability. The semi-arid Bouregreg Watershed in northwestern Morocco faces severe erosion-driven degradation, underscoring the need for advanced modeling for effective assessment and mitigation. The study first compared the Revised Universal Soil Loss Equation (RUSLE) and the Modified Universal Soil Loss Equation (MUSLE) using the high-resolution INRA soil dataset (1:50,000). RUSLE with INRA data (RUSLEINRA) was then compared to RUSLE with the global FAO soil map (1:1,000,000) (RUSLEFAO). RUSLEINRA yielded more accurate estimates, with average soil loss for the period 09/01/2016–08/31/2021 of 15.56 t ha⁻¹ yr⁻¹, versus 10.24 t ha⁻¹ yr⁻¹ for RUSLEFAO, while MUSLEINRA estimated 11.40 t ha⁻¹ yr⁻¹. Sediment yield (SY) was calculated using a modified Sediment Delivery Ratio (SDR) of 20.62% and validated against observations at the Sidi Mohammed Ben Abdallah (SMBA) dam, showing strong alignment with RUSLEINRA. The study also projects erosion and SY for the period 09/01/2021–12/31/2040 under SSP126 and SSP585 scenarios, with future soil loss estimated between 31.54 and 37.04 t ha⁻¹ yr⁻¹, indicating a climate-driven increase. These findings highlight the urgent need for effective soil conservation and watershed management, offering key insights for decision-makers addressing reservoir sustainability, sediment control, and erosion in semi-arid regions.