The assessment of the rainwater harvesting (RWH) potential in Hosaena City from medium to large public buildings is the main objective of the study. ArcGIS with Google Earth Pro was used to calculate the rooftop area of the selected building, and RWH from each category of public buildings was evaluated by considering rooftop areas, monthly rainfall, and runoff coefficient. The result shows that the highest amount of rainwater was harvested from the City Administration building in August (197.1 m3), whereas the lowest was collected from the Health Center building in December (1.8 m3). Additionally, the contribution of RWH from individual public buildings for each month was evaluated, and its values ranged from 0.073 to 1.8%. The comparison shows that Wachemo University, Nigist Eleni Comprehensive Hospital, a technical and vocational institute, and teacher training center buildings, could have a significant potential for rainwater storage, ranging from 1.56 to 28.41%. This implies that, besides saving a noteworthy volume of potable water, the excess rainwater can be stored and utilized for the later dry months. As a result, it is recommended that rooftop RWH in Hosaena be the best source of the city's water supply system to reduce the current water supply shortage.

Soil erosion threatens land productivity, water resources, and hydropower sustainability in the Upper Omo Gibe Basin (UOGB), Ethiopia. This study quantified soil erosion dynamics, assess land use/land cover (LULC) changes and climate variability, and prioritized erosion-prone sub-watersheds to support targeted soil and water conservation planning. A GIS-based Revised Universal Soil Loss Equation (RUSLE) model was applied to estimate soil loss for 1991, 2006, and 2020 LULU series. LULC changes were analyzed using multi-temporal satellite imagery obtained from the USGS Earth Explorer and processed in ERDAS Imagine. Sub-watershed prioritization was conducted using the Sub-Watershed Prioritization Tool (SWPT) based on morphometric and topo-hydrological parameters. Future climate conditions were assessed using Regional Climate Models under RCP 4.5 and RCP 8.5 emission scenarios. The results show significant expansion of agricultural and settlement areas at the expense of forest, shrubland and bare land. Classification accuracies ranging from 72 to 86.4%, with kappa coefficients between 0.67 and 0.84. Mean annual soil loss increased from 25.47 t ha ⁻¹ yr ⁻¹ in 1991 to 36.72 t ha ⁻¹ yr ⁻¹ in 2020, exceeding tolerable soil loss limit. Projected temperature increases ranged from 1.5 to 2.2 °C relative to the baseline period, while rainfall showed reductions of up to 31.8%. Erosion hotspot analysis identified SW1, SW3, SW4, and SW5 as highly vulnerable sub-watershed, with SW1 and SW3 ranked as top conservation priorities. Overall, the integrated GIS-based RUSLE-SWPT framework effectively identifies erosion hotspots and guide’s sustainable watershed management to protect the downstream Gibe III hydropower reservoir.