Unraveling soil erosion severity in the endorheic basin of Hodna, Algeria: a spatiotemporal responses framework under LULCC dynamics

Land use and land cover (LULC) dynamics substantially influence environmental change and are key factors in water erosion within hydrographic basins, among other essential processes of global land degradation. In the land degradation concept, water erosion experiences quantitative and qualitative alterations shaped by temporally variable elements such as rainfall erosivity and land uses. Analyzing and predicting the changes in water erosion over time is crucial for efficient water and soil conservation planning and environmental management on a large scale like watersheds. This study aims to investigate the soil erosion severity and responses (SESR) to historical (2000, 2010, and 2020) and predicted (2030, 2040, and 2050) LULC changes in the Hodna basin (HB), Algeria, using the Modules for Land Use Change Evaluation (MOLUSCE), geospatial analysis and the Revised Universal Soil Loss Equation (RUSLE). The Hodna Basin, an arid to semi-arid region spanning 25,856 km², is subdivided into 24 watersheds to assess spatiotemporal erosion severity trends. Results reveal a progressive increase by over 130% in mean annual soil loss from 3248 t.ha⁻¹.year⁻¹ in 2000 to 4346 t.ha⁻¹.year⁻¹ by 2040. Watersheds W11, W12, and W9 emerged as the highest priority due to severe erosion rates with the same trend in studied periods (> 150 t.ha⁻¹.year⁻¹), while W23, W7, and W6 exhibited minimal risk. Bare Land (BL), covering 60–73% of the basin, contributed disproportionately to soil degradation, with erosion rates increasing by 75%, from 28.01 to 49.05 million t.year⁻¹ by 2050. Agricultural Land (AL) showed fluctuating trends, peaking at 4.44 million t.year⁻¹ in 2010 before declining due to conservation efforts. Principal Component Analysis (PCA) underscored strong correlations between LULCC and soil erosion highlighting the influence of some LULC classes among others on soil loss by erosion. The study emphasizes also the critical role of dynamic C and P factors in erosion modelling, with vegetation cover reducing erosion by up to 60% in forested areas. These findings provide actionable insights for policymakers to prioritize interventions, such as terracing, reforestation, and sustainable land management, to mitigate soil loss and enhance regional resilience.