Prediction of flood frequency under a changing climate, the case of Hare watershed, Rift Valley Basin of Ethiopia

Abstract

Climate change is causing unpredictable fluctuation in the rainfall patterns and, frequent heavy rainfall that led to catastrophic flooding and a significant risk on Hare watershed biodiversity in the Rift Valley of Ethiopia. Prediction of flood frequency under changing climate in the Hare watershed has not been well-studied in spite of the risk it poses to the human environment. Thus, this study aimed to predict flood frequency under changing climate in the Hare watershed relevant for effective flood management and early warning systems. The Soil Water Assessment Tool (SWAT) was used to simulate monthly stream flow. The model was calibrated and validated for the period of 1994 to 1998 and 1999 to 2001, respectively. The stationarity of annual peak flow was analyzed using Pettitt’s test. Flood Frequency Distribution (FFD) software package was deployed to determine the flood magnitude and frequency curve for the baseline period (1980–2009) and a predicted period (2021–2050). The result of the SWAT model performance evaluation statistics showed a high potential in the prediction of future streamflow. In addition, the goodness-of-fit test result showed that the Gumbel and Generalized Extreme Value (GEV) were the best-fit probability distribution for baseline and future flood events, respectively. The predicted annual peak flow results show an increasing trend by 60 and 106% under RCP 4.5 and RCP 8.5 emission scenarios, respectively. The future design flood may increase between 6.5 and 119.0% when compared to the baseline period. The study provides valuable information for policy and decision makers during the implementation of different flood events adaptation and mitigation measures for the Hare watershed.