Performance evaluation of HEC-HMS model for continuous runoff simulation of Gilgel Gibe watershed, Southwest Ethiopia

Abstract

Hydrological models are widely used for runoff simulation throughout the world. The objective of this study is to check the performance of the HEC-HMS model for continuous runoff simulation of Gilgel Gibe watershed. It includes sensitivity analysis, calibration, and validation. The model calibration was conducted with data from the year 1991 to 2002 and validated for the year 2003 to 2013 period using daily observed stream flow near the outlet of the watershed. To check the consistency of the model, both the calibration and validation periods were divided into two phases. The sensitivity analysis of parameters showed that curve number (CN) and wave travel time (K) were the most sensitive, whereas channel storage coefficient (x) and lag time (tlag) were moderately sensitive. The model performance measured using Nash-Sutcliff Efficiency (NSE), Percentage of Bias (PBIAS), correlation coefficient (R2), root mean square error (RMSE), and Percentage Error in Peak (PEP). The respective values were 0.795, 8.225%, 0.916, 27.105 m3∙s-1 and 7.789% during calibration, and 0.795, 23.015%, 0.916, 29.548 m3∙s-1 and -19.698% during validation. The result indicates that the HEC-HMS model well estimated the daily runoff and peak discharge of Gilgel Gibe watershed. Hence, the model is recommended for continuous runoff simulation of Gilgel Gibe watershed. The study will be helpful for efficient water resources and watershed management for Gilgel Gibe watershed. It can also be used as a reference or an input for any future hydrological investigations in the nearby un-gauged or poorly gauged watershed.