Analyzing common social and physical features of flash-flood vulnerability in urban areas

Flash flooding events, with their intense and sudden nature, present unique challenges for disaster researchers and emergency planners. To quantify the extent to which areas impacted by flash flooding share similar social and physical features, the research uses community-scale open-source and crowdsourced data and k-means clustering. Crowdsourced data helps reveal the social and physical vulnerabilities of a community to flash flood impacts which could better inform decision-makers who must allocate limited resources, have a spatial understanding, and aim to reduce future effects of flash floods. The research evaluates the impacts of Tropical Storm Imelda on Houston Metropolitan and Hurricane Ida on New York City. It develops a combined flash flood impact index based on FEMA claims, 311 calls, and Waze traffic reports which is able to capture a combination of crowdsourced data for the societal impact of flash flooding. K-means clustering evaluates a community's socio-demographic, social capital, and physical features to the combined flood impact index. To ensure accessibility and replicability to different types of communities, our research uses publicly available datasets to understand how socio-demographic data, social capital, and physical connectivity and development affect flash flood resilience. The findings provide a framework to identify potential flood impacts using historic data.