Flash-Flood Monitoring and Modeling in Kentucky

This research project focused on the evaluation of hydrologic issue of flash-flooding in the state of Kentucky. The primary objectives of this project were the following:(1) to initiate the establishment of a hydrologic database archive necessary for characterizing rainfall and runoff associated with flash-flooding; (2) identification of appropriate modeling approaches for evaluating site-specific flash-flood runoff behavior. Specific tasks accomplished to meet the objectives include the following:(1) development of a rainfall and streamflow data archive using existing measurement gages and identification of the rain gage data from two sources for preliminary quality control; (2) identification of the spatial and temporal characteristics of rainfall at daily, seasonal, and annual time scales; (3) definition of the characteristics of runoff associated with flash-flooding; and (4) initiation of a review of flash-flood runoff modeling approaches for small watersheds. Flash-flooding is one of the most costly natural hazards nationally each year. In 1990, it resulted in I 09 deaths and damages of $625 million (Kentucky Engineer, 1992) in the United States. In 1992, a single flash-flood event in Bar-Creek, Kentucky resulted in four deaths and displaced approximately 54 families (National Weather Service, 1992). Due to the short response time associated with the watersheds prone to flash flooding, rainfall data must be collected rapidly in real-time and flood estimates computed accurately in order that adequate and timely warnings may be issued. Accurate estimates of flash-flood water levels require site-specific information describing the hydrometeorologic conditions and physiographic characteristics of the watershed for use of high resolution runoff modeling approaches. Monitoring rainfall events that lead to or cause flash-flooding is necessary to identify the rainfall characteristics associated with flooding and flash-flooding. The ultimate objective in a flash-flood warning system is to provide increased warning time to residents to allow them to escape the rapidly rising water. National-level agencies are only beginning to address the issue of real time, high resolution flood forecasts meeting the needs of state agencies and local-area residents. Other issues, beyond the scope of this work, must be addressed and resolved before such a system can be significantly mitigate flash-flood losses. This project addresses the initial step toward establishing such a system by compiling a flash-flood precipitation and runoff database for Kentucky from existing gage networks, and quantitatively defining the behavior of precipitation and runoff