The assessment of water resources in ungauged catchments in Rwanda

•Climate, physiography and land cover strongly influence the hydrology of catchments in Rwanda.•Using leave-one-out cross-validation, the log-transformed models were found to predict the flow parameters more suitably.•Developed models can be used for estimating the flow parameters in ungauged catchments in Rwanda.•Methodology can be applied in any other region, as long as sufficient and good quality streamflow data is available.

Study regionRwanda is a landlocked country in Africa with precipitation ranging from 800 mm yr−1 in the east to 1500 mm yr−1 in high-altitude regions in the north and west.Study focusStreamflow estimation is an important task that is required in water resource assessments due to its importance in planning, decision-making and economic development. In this study, streamflow characteristics of ungauged catchments in Rwanda were calculated using a regionalization approach based on climate similarity and stepwise multiple-regression analysis. One climatic homogeneous region was identified and datasets of nine gauged stations and general available catchment characteristics were used to develop non-transformed and log-transformed regression models.New hydrological insights for the regionResults of this study show that climate, physiography and land cover strongly influence the hydrology of catchments in Rwanda. Using leave-one-out cross-validation, the log-transformed models were found to predict the flow parameters more suitably. These models can be used for estimating the flow parameters in ungauged catchments in Rwanda and the methodology can be applied in any other region, as long as sufficient and good quality streamflow data is available.