Rainfall-Runoff Simulation Using Hydraulic Engineering Center Hydrologic Modelling System For Welmel Watershed, Ethiopia

Abstract

Hydrologic studies on rainfall-runoff have been extensively applied by water resource planners to simulate the hydrological response in many regions around the world to fulfill various desirable needs with the purpose of effective and proper planning and managing of water resources for present and future uses. However, such research does not pay enough attention to the Welmel watershed, Genale Dewa Basin, Ethiopia, which may be affected by water insecurity. Therefore, the main objective of this study was to simulate rainfall-runoff processes using Hydrologic Engineering Center-Hydrologic Modeling System (HEC-HMS) to model and predict stream flow in the Welmel watershed. The input data used were the meteorological data, hydrological data, land use land cover data, soil data and Digital Elevation model obtained from the National Meteorological Service Agency, Ministry of water and Energy, Engineering Corporation of Oromia and Website respectively. The missing value of precipitation data was filled using the XLSTAT, and the consistency of data was checked using a double mass curve. Hydrologic Engineering Center Geospatial Hydrologic Modeling System (HEC-GeoHMS) was used for prepared basin model imported to HEC-HMS. The SCS-CN loss, SCS unit hydrograph, Constant monthly, and Muskingum methods are used to measure precipitation loss modeling, transform modeling, base flow modeling, and flood routing. For model calibration (1990-2010) and validation (2011-2015), hydro-meteorological data were used. The parameters used to evaluate the models' sensitivity were; curve number, initial abstraction, basin lag, Muskingum k, and Muskingum x. The results show that the model was most sensitive to basin lag and Muskingum (K) for this study respectively. During the calibration and validation phase, the performance of the model was assessed by Nash Sutcliffe Efficiency (NSE), Root means square error (RMSE), Coefficient of determination (R2), Percent bias (PBIAS), Percent error in volume (PEV), and Percent error in peak flow (PEPF), indicating NSE (0.702), R2 (0.7156), RMSE (0.5), PBIAS (5.67%), PEV (-5.68), and PEPF (8.76) and NSE (0.735), R2 (0.7559), RMSE (0.5), PBIAS (-2.32%), PEV (2.32), and PEPF (12.21), respectively. The simulated and observed peak discharges differed by 20.5 m3/s in calibration time. This indicates that the peak discharge was well predicted. In the validation period, there was a difference of 22.4 m3/s between the observed and simulated peak discharge. This means that the peak discharge was slightly higher than expected. For this study, calibrated and validated model results showed that the model performed well. The maximum and minimum simulated daily flow potential of watershed was 215.7m3/s and 6.3m3/s respectively. Flood prediction was conducted in the HEC-HMS using 24- hour rainfall depths of 2, 5,10, 25, 50,100,200, and 500 years return period and found to be 139.7 m3/s, 176.1 m3/s, 203.1 m3/s, 243.8 m3/s, 287.6 m3/s, 322.8m3/s, 365.9m3/s, and 418.8 m3/s, respectively. Also using the General extreme value and the General pareto of the Statistical flood frequency analysis, the peak flow discharge for 2,5,10,25,50,100, 200, and 500 year return period were 136.23m3/s, 181.28 m3/s, 210.35 m3/s, 246.24m3/s, 272.27 m3/s, 297.62 m3/s, 322.40m3/s, 354.38 m3/s, and 100.74m3/s, 158.74 m3/s, 191.19 m3/s, 223.20m3/s, 241.20m3/s, 255.13 m3/s, 265.93 m3/s, 276.56m3/s respectively. The minimum and maximum peak flow records in HEC-HMS were 139.7m3/s and 418.8 m3/s, respectively. Therefore, these predicted peak flood will help in water resources and flood management for this study area.