In recent times, soil erosion has become one of the major environmental issues considering large scale anthropogenic activities and frequent extreme weather events globally. These activities and events cause the wearing away of top soil, initiating loss in agricultural land, and triggering disasters like landslides in the hilly regions. So, such regions demand soil conservation measures such as plantation, and construction of check dams. Due to limited resources in developing countries like India, prioritization of land-based on susceptibility to soil erosion makes the implementation of these measures efficient. In the present study, the openly accessible TanDEM-X 90m DEM and TOPSIS-AHP ensemble model are used for prioritization of Upper Kosi Watershed based on soil erodibility. This watershed having fifth (5^th) order perennial Kosi river is a showcase for different landscapes, biodiversity, micro-climate, and hydrogeology. TanDEM-X DEM is preprocessed for stream network generation and catchment delineation which is further used for extraction of 10 morphometric attributes (MAs) of each sub-watershed. AHP model is used for calculating the weights of MAs. AHP model concludes that Watershed slope, Stream frequency, and Drainage density play a greater role in soil erosion as compared to other MAs. Morphometric parameters and its weightage is utilized in the execution of the TOPSIS model to calculate the closest coefficient for each sub-watershed. Further, these closest coefficient is classified into low (0.147-0.167), medium (0.167-0.326), and high (0.326-0.794) susceptible zones for prioritization of sub-watershed based on the geometric mean method. This study highlights that Upper Kosi watershed contains 4.73%, 38.79%, 56.48% of its area in high, medium, and low susceptible zone respectively. The result can assist the decision-makers and planners in choosing suitable regions depending on the available resources for conservation measures in Upper Kosi watershed.