Spatial Soil Erosion Modeling for Improved Land Management in Gumara Watershed, Ethiopia

Abstract

Agriculture output and environmental sustainability are threatened by land degradation, which deteriorates soil resources. In response, soil and water conservation (SWC) initiatives were implemented in Ethiopia. However, soil erosion remains a main challenge to soil productiveness and sedimentation in the country. To determine erosion-prone areas and assess the impact of SWC activities on soil erosion in the country, a revised version of the universal soil loss equation (RUSLE) was implemented. Gumara catchment is an erosion-prone area in Ethiopia. Spatial estimation of soil erosion is essential to conserve, manage, and use soil and water resources. Therefore, research was conducted to evaluate the soil erosion of the catchment using the RUSLE with ArcGIS environment. The results indicated that the soil erosion extent was classified into four categories as per the erosion rate, and 5.45% of the catchment explained very high erosion (> 200 ton ha−1 yr−1) followed by 17.24% classified as high (50–200 ton ha−1 yr−1). And, 24.21% and 53.11% of the catchment had a moderate and slight class, respectively. The spatial distribution of soil erosion in the Gumara watershed could be used to better use soil resources, increase agricultural production, and ensure environmental sustainability. The country’s soil erosion is characterized by various estimates, indicating spatiotemporal dynamics. This is primarily due to the heterogeneity of the different sites, which is primarily linked to varying cover values and management factors. With this information, conservation decisions can be made with greater knowledge by concentrating on important hotspots. Therefore, RUSLE applied with ArcGIS across various land management practices and climate zones is a potential tool for SWC demanding site identification. This continues to be beneficial in the pursuit of sustainable land management techniques for the local people’s long-term well-being.