Evaluation of climate anomalies impacts on the Upper Blue Nile Basin in Ethiopia using a distributed and a lumped hydrologic model

Abstract

Summary Evaluating the climate anomalies impacts on the Upper Blue Nile Basin (UBNB), Ethiopia, a large basin with scarce hydroclimatic data, through hydrologic modeling is a challenge. A fully distributed, physically-based model, a modified version of the Interactions Soil–Biosphere Atmosphere model of Meteo France (MISBA), and a lumped, conceptual rainfall–runoff Sacramento model, SAC-SMA of the US National Weather Service, were used to simulate the streamflow of UBNB. To study the potential hydrologic effect of climate anomalies on the UBNB, rainfall and temperature data observed when climate anomalies were active, were resampled and used to drive MISBA and SAC-SMA. To obtain representative, distributed precipitation data in mountainous basins, it was found that a 3% adjustment factor for every 25 m rise in elevation was needed to orographically correct the rainfall over UBNB. The performance of MISBA applied to UBNB improved after MISBA was modified so that it could simulate evaporation loss from the canopy, providing coefficient of determination (R2) = 0.58, and root mean square error (RMSE) = 0.34 m3/s in comparison with the observed streamflow. In contrast, the performance of SAC-SMA at the calibration run and the validation run is better than that of MISBA, such that R2 is 0.79 for calibration and 0.82 for validation even though it models the hydrology of UBNB in a lumped, conceptual framework as against the physically-based, fully distributed framework of MISBA. El Nino tends to decrease the June–September rainfall but increase the February–May rainfall, while La Nina has opposite effect on the rainfall of UBNB. Based on the simulations of MISBA and SAC-SMA for UBNB, La Nina and Indian Ocean Dipole (IOD) tend to have a wetting effect while El Nino has a drying effect on the streamflow of the UBNB. In addition, El Nino Southern Oscillation (ENSO) and IOD increase the streamflow variability more than changing the magnitude of streamflow. The results provide useful information on the effects of global oceanic anomalies on the hydrology of UBNB.