FLOOD HAZARD AND INUNDATION MAPPING: A CASE STUDY OF UPPER BARO-AKOBO BASIN, ETHIOPIA

Abstract

Flooding which intermittently creates havoc in Upper Baro-Akobo basin of Ethiopia has been studied intricately through GIS integrated with MCDA and HEC-RAS model. The basic aim of this study is to generate flood hazard and inundation area mapping of Upper Baro-Akobo basin. Different data such as spatial, meteorological, and hydrological data were collected and analyzed using GIS integrated with MCDA, and HEC-RAS model. Flood hazard mapping was generated on the basis of integrating of flood influencing factors such as DEM, aspect, curvature, rainfall, slope, TRI, flow accumulation, flow direction, SPI, NDVI, population density, distance from river, TWI, drainage density, LULC, soil type and STI. Among the described flood influencing factors, only ten factors such as rainfall (12.89%), land use (12.43%), slope (12.14), DEM (9.64%), soil type (8.97%), distance from river (7.94%), NDVI (6.05%), Pd (4.78%), TWI (3.96%), and TRI (3.92%) are having major impact on the flood potential in the study area. The flood hazard map shown that, 202.9 (1%) of the areas are classified as very high, while 4442.4 (21.6%), 10515.3 (51.1%), 369.5 (26.1%), and 32.2 (0.2%) areas are classified as high, medium, low, and very low, respectively. This shows that the watershed's downstream was in a high to very high flood threat zone in comparison to the upstream of the watershed. The flood frequency analysis with stream flow data from 1990-2009 at Baro Gambella gauging station was carried out to estimate the expected peak floods of the Upper Baro-Akobo watershed. The peak floods with return periods of 25, 50, and 100 year were calculated by the normal distribution method and the results were 1739.586 , 1820.872 , and 1893.974 , respectively. The HEC-GeoRAS data, such as land use, DEM, and the Triangular Irregular Network (TIN), were used to create the river geometry. The river geometry and stream flow data were input into HEC RAS model for generated of flood inundation area mapping. The HEC-RAS model results indicated that the flood inundation areas under different land uses with the return period of 25, 50 and 100 year, were 390 , 422.76 , and 453.97 , respectively in which annual crop was found to be the most affected by flood. The inundation depth of 25, 50 and 100-year return periods ranged from 0-2.6, 0-2.9, and 0-3.2m depth at the upstream and downstream of the river, respectively. It is suggested that flood protection measures at places with high to very high risks of flooding should be implemented.