Assessing hydrological erosion estimation using the Revised Universal Soil Loss Equation (RUSLE) model in Google Earth Engine: a case study of Medjerda River Catchment, Tunisia

This study explores a novel methodology for predicting soil erosion in the Medjerda River watershed, Tunisia, leveraging Google Earth Engine's (GEE) capabilities. By integrating satellite imagery, Digital Elevation Models (DEMs), and advanced geospatial analyses, we demonstrate the effectiveness of GEE as a platform for studying hydraulic erosion dynamics. CHIRPS, Open Land Map, ISDA datasets, and satellite imagery were utilized within GEE to calculate the Revised Universal Soil Loss Equation (RUSLE) for the period between 2000 and 2020 and years marked by historical floods (2000, 2003, 2005, 2009, 2012, 2015). The analysis revealed seasonal variations in RUSLE factors and enabled the generation of soil erosion maps. The RUSLE model implemented within GEE facilitated the rapid production of erosion prediction maps. Predicted erosion rates were highest in 2003 and 2009 (A = 170,26 t/ha/yr and A = 156,41 t/ha/yr, respectively), while the other years presented lower risks. Approximately 98% of the basin displayed low erosion rates (0-5 t/ha/yr), attributed to high vegetation cover, crop cultivation, and gentle slopes. Conversely, 2% of the area experienced moderate to severe erosion (> 5 t/ha/yr) in regions with gullies, barren lands, and minimal vegetation. Areas with the greatest chance of erosion are found upstream of the Oued Medjerda River and in the Tabarka and Siliana basins at higher elevations. This high risk is due to several factors: steep inclines, the type of soil (clayey marl) in the surrounding mountains, and a high R factor (indicating erosive rainfall) even in areas with some vegetation. In contrast, the flat plains around Bousalem have a much lower erosion risk (around 0.03 tons per hectare each year). Interestingly, the risk of erosion is even greater at higher altitudes. This trend is consistent yearly, suggesting that some areas are always more susceptible to erosion. This phenomenon is likely because of a combination of factors, such as the shape of the land (topography), the soil composition (texture), and the type of land cover, all of which contribute to how vulnerable the watershed is to losing soil.