Land use land cover (LULC) dynamics is the main environmental challenge that mainly affect soil erosion and sediment yield in Ethiopia. Thus, this study aimed to evaluate LULC dynamics on sediment yield and soil erosion in Barkana watershed. The main data sources for image analysis were Landsat TM (1993), ETM+(2008), and OLI (2023). While rainfall data, soil type and texture, Digital Elevation Model (DEM), and LULC data were used to generate the erodibility (R), erosivity (K), slope-length (LS), land cover (C), and management (P) factors for RUSLE Model. Soil erosion and sediment yield were estimated using the RUSLE, Sediment Delivery Ratio, and Sediment Yield (SD) algorithms. These data sets were analyzed using ERDAS Imagine and Arc GIS software accordingly. Moreover, a total of twelve key informant interviews (KII) were selected, three from the upper, four from the middle and five from the lower part of the watershed purposively depending on the spatial extent of these parts. The KII have been used to triangulate the secondary data analysis. The Maximum Likelihood Algorithm (MLA) of supervised classification was used to classify Landsat images from three separate time periods (1993, 2008, and 2023) into six key land use/cover classifications (agricultural, bare land, built-up, forest, grazing land, and shrubs). Consequently, throughout these years (1993 2023), forest, grazing land, and shrub declined by 3156.8ha (12%), 1584ha (6%), and 868ha (4%), respectively, while agricultural, bare land, and built up rose by 3742.5ha (14.3%), 488.2ha (2%), and 1378.5ha (5.3%). Throughout the study periods, the annual rate of change for different land uses and land covers varies. For instance, the yearly increases in built-up, bare land, and agricultural land were 0.48%, 0.06%, and 0.18%, respectively. In 1993, 2008, and 2023, the mean annual soil loss increased by 27.95, 41.98, and 59.34 t/ha/yr, respectively. The sediment yield increased by 8.6, 12, and 15.7 t/ha/yr during these times on the same trend. The average amount of soil erosion in 1993, 2008 and 2023 was employed to identify hotspot regions for watershed planning and management strategies. As a result, five priority classes (very high, high, moderate, low, and very low) are applied to the entire watershed. The first and second priority classes that needs attention together covered about 4,035ha (15.4%) of land. Moreover, the highest rates of soil erosion (more than 50 t/ha/yr.) were observed in very high, high, and moderate priority levels. As a result, these areas require quick attention.