Lessons Learned from Multifaceted Flash Flood Damage Surveys and Implications for Future Forecasting Systems

Several significant flash floods driven by extreme precipitation events occurred in the south-central United States during the spring of 2024. A series of multifaceted post-event flash flood surveys were conducted on six of these events. These surveys included a combination of on-site ground evaluations, uncrewed aerial systems (UAS) flights, high-resolution satellite tasking, data collection from broadcast and social media, information from local officials, results from online societal surveys, and traffic mobility data. The severity of each documented impact was quantified using an updated version of the Flash Flood Severity Index (FFSI). This information was used to analyze and create a more complete picture of the impacts of precipitation-driven significant flash flood events across Arkansas, Oklahoma, and Texas. These data points with corresponding FFSI values were also gridded for comparison with products from the Multi-Radar Multi-Sensor (MRMS) and Flooded Locations and Simulated Hydrographs (FLASH) systems.

A number of lessons were learned from conducting these flash flood surveys as well as from the application of FFSI for quantifying the spatial severity of each event. This included 1) the identification of significant impacts (i.e., FFSI = 3 or greater) and how/where these features were with respect to vulnerable locations (e.g., low-water crossings, drainages and culverts, low-spots), 2) how significant impacts are viewed through the analyses of topographic features coupled with ground and aerial survey data, 3) the relationship of flash flood impacts to precipitation and hydrologic-based products, and 4) how these overall findings can be incorporated into future large-scale hydrologic modeling (e.g., FLASH) for improved prediction of the occurrence and the potential severity of flash flooding. Lessons are also learned from the societal and traffic data obtained from these cases, and how it can drive efforts to incorporate a more human-centered, impact-based application to future flash flood modeling, prediction, and warnings across the United States.