Abstract:
The purpose of hydraulic analysis of floods is to improve our understanding of flood behavior and to develop strategies for managing and mitigating flood risk within a floodplain. This study aimed to conduct a hydraulic analysis of a flash flood event in Narok, Kenya, using topographic data collected by an unmanned aerial vehicle (UAV). The channels in the town were dry, making it an ideal opportunity to use UAV to map the channel and floodplain and generate topographical variables for hydraulic modeling. Detailed flight plans were created to cover the built-up area of the town, and the topography was reconstructed from the UAV survey data to generate a 5 cm Orthomosaic and a 5 cm Digital Surface Model (DSM). These topographic datasets were used to extract all the necessary geometric data, including a channel centerline, bank lines, channel cross-sections, blocked flow areas, details about the bridges and land use and land cover (LULC) of the floodplain. HEC-RAS model was used to simulate the flow through the channels during the flood event on November 27, 2019. The inundation depths reached 4 m at their highest point and extended up to 290 m from the channel at their furthest point. The model results were validated using survey data of observed wrack marks and photographs taken by local non-experts. This study demonstrates the potential of UAVs to improve flood simulations by providing accurate and detailed data collection at a lower cost and with fewer resources. In addition, the usefulness of the UAV-derived data is demonstrated beyond extracting river cross-sections, which previous related studies were limited to. Integrating UAVs into flood management strategies has the potential to lead to more effective and efficient flood management in the face of increased risk of flooding.