Rainfall-Runoff Modeling Using Hydrologic Engineering Center-Hydrologic Modeling System Models: Case Study of Bilate Watershed, Ethiopia

Abstract

When the rainfalls on the earth’s surface, some portion of it is infiltrated and evaporated, and the remaining portion is changed to surface runoff. The excess surface runoff may result in soil erosion. In Ethiopia, particularly in the study area, the land use land cover management system is poor and the excess surface rainfall causes the erosion of the land surface, overflow of the river on its bank and flooding problems. Flooding have occurred frequently and lead to Migration of people from their place, and properties have been damaged. The general objective of this study is to simulate daily Rainfall-Runoff using Hydrologic Engineering CenterHydrologic Modeling System (HEC-HMS) for Bilate watershed, Rift Valley River Basin, Ethiopia. Bilate watershed is a part of the Abaya Chamo sub basin in the southeast of the Ethiopian Rift Valley and it is located between longitude (37°30ˈ0" to 38°30ˈ0"E and latitude 6°30ˈ0" to 8°30ˈ0"N) and its total area is about 5500km2 . Understanding the complex relationships between rainfall and runoff processes is necessary for the proper estimation of the quantity of runoff generated in a watershed. The input data used were the spatial data, meteorological data and hydrological data. The Bilate watershed was initially delineated and devided into 150 sub basin and then merged into 6 sub basins using the combination of Arc hydro tool version 10.1 and HEC-GeoHMS extension of ArcGIS 10.1. The Curve number was generated using the union of Hydrologic soil group polygon and land use polygon and the Curve number lookup table that links Hydrologic soil code and land use value. The model was simulated for an input daily rainfall and stream flow data of 21 years (1996-2016). The model calibrated and validated using daily rainfall and stream flow recorded data for 14 years (1996 to 2009) and 7 years (2010 to 2016) for calibration and validation respectively at Bilate tena gauging station, at the outlet of the watershed. CN is found to be more sensitive parameter in the Bilate watershed. The objective functions NSE, R2 and MBE were used to check performance of the model. During calibration the model performance resulted in R2= 0.8674, NSE = 0.837, MBE=60.35. Also during validation R2=0.8615 NSE =0.852 values and MBE=56.50. Also the performance ratings for statistics Observations Standard Deviation Ratio (RSR)=0.163 and 0.148 and PBIAS=6.03% and 6.14% for calibration and validation respectively. These values showed that the model performed well during calibration and validations. The hydrologic model was used for determining the peak flow discharge for return periods of 2, 10, 25, 50 and 100 years and the result was found to be 241.8m3 /s, 314.8m3 /s, 651.2m 3 /s, 738.01m/s and 1152.1m 3 /s respectively in HEC-HMS. Using the Statistical flood frequency analysis, the peak flow discharge of 213.5m 3 /s using log Pearson type-III flood frequency analysis for 2-year return periods. For 10, 25, 50 and 100 years return period, 296.32m 3 /s, 541.63m 3 /s, 652.38m 3 /s and 983.125m 3 /s using log Pearson type-III flood frequency analysis respectively. The highest peak flow recorded was 1152.1m3/s.