Flood hazard vulnerability assessment in the Sile-Sago watershed, Rift Valley Basin, Ethiopia

Abstract

Flood hazards (FHs) are exacerbated by heavy rainfall and pose serious threats to life, social amenities, and the environment. This study was targeted at identifying and mapping factors triggering, analyzing, and mapping vulnerabilities of FHs in the Sile-Sago watershed. Data were acquired from the field, interviews, and satellite images. The 13 parameters contributing to FH's vulnerability included rainfall, soil types, elevation, slope, proximity to rivers, drainage density, population density, land use or cover, topographic wetness index, vegetation density, etc. Data were resampled into 30 m resolution before conducting the weighted overlay process. The research utilized MCDM and AHP models, based on literature, survey, and expert evaluation, to establish weight scores for ranking factors, enhancing its uniqueness and establishing the relative importance of each factor. Then the weighted overlay operation was used to map the vulnerability levels of the FHs in ArcGIS environment. The study results revealed that nearly one-third (32.5 %) of the area falls in the high (21.5 %) and very high (11 %) vulnerability zones. These areas are characterized by intense rainfall, dense human settlements, and the highest drainage density. The model was validated with a consistency ratio (CR) of 0.034 (3.4 %) and a receiver operating curve with an area under curve value (ROC-AUC) of model prediction accuracy of 0.81 (81 %). Both the CR and ROC-AUC values indicate that the model performance is highly acceptable. Successful management of floods hinges on anticipatory prediction, early warning systems, preparedness, and disaster risk management strategies. Building gabions, canals, bridges, and other physical structures must be constructed by individuals, local and regional governments, and non-governmental organizations as recommended protection and prevention strategies for FHs in the watershed.