MAPPING FLOOD INUNDATION AREA USING REGIONAL FLOOD FREQUENCY ANALYSIS IN UNGAUGED SHAFE RIVER

Abstract

Flood inundation modelling is needed for rivers to simulate and forecast the possible impacts of floodplain development in an ungauged river. The hazard of flooding occurs due to heavy rainfall, which affects downstream agricultural areas and households to displace from the areas. The goal of this study is to estimate flood inundation mapping of ungauged rivers using regional flood frequency analysis in Shafe River. Hydrological and hydraulic modelling was conducted for flow generation and flood mapping in the Shafe River using the HBV IHMS and HEC RAS models, respectively. This research demonstrates a variety of approaches that can be used to perform regionalization of typical ungauged rivers, where stream flow data are available from hydrological related catchments. The various methods are evaluated by comparing the goodness of fit of an array of hydrologic distribution functions. Flood inundation modeling was performed based on steady and 1D unsteady flow for Shafe River. In this research, the HBV-IHMS model was calibrated in four (Bilate at Alba, Bedessa, Hare, and Gidabo) catchments of the Abaya Chamo lakes sub basin. The runoff data from (1991-2006) was divided into three periods. The first warm-up was the model (1991), model calibration period (1992-2000), and model validation (2001-2006). The model's effectiveness was assessed using the RVE relative volume error (RVE) and the coefficient of Nash-Sutcliffe model efficiency (ENS), which both showed high model estimate performance within the range of 0.65 to 0.79 and -10% or +10%, respectively. Using General extreme value based goodness of fit test flood extreme value for the respective recurrence interval 5,10,25,50 and 100 was 40.02, 56.14, 60.30, 71.81 and 88.15m3/s, respectively. The HEC-RAS model outcome indicated that the flood inundation mapping area for 5, 10, 25, 50, and 100-year recurrence intervals were 151.75, 159.10, 179.72, 187.49ha for steady flow and 161.75, 179.72, 193.85, 202.00 and 228.06ha for unsteady flow analysis respectively. That result revealed that, at a return period of 100 years, the highest water depth generated by the unstable flow model is 4.2 m, while the steady flow model produces 3.8 m. And also the percentage of the entire flooded area during 100 return periods was 1.21 and 1.41, respectively. Because of this, 1D unsteady flood modeling for Shafe River was estimates better than steady flow analysis. Therefore, flood control techniques will be required for managing floods in Shafe catchment.