Flash Flood Prediction on Shoal Creek: or Why Radar Rainfall?

Central Texas is often called “Flash Flood Alley” because of its frequent, intense storms. While large events seem to happen every decade, lesser events also cause public safety concerns. The City of Austin operates a Flood Early Warning System (FEWS) to reduce public exposure to flash flood hazards. The system is operated and maintained by the Watershed Engineering Division of the Watershed Protection Department. FEWS offers advanced information and warning of urban stormwater flooding based on radar rainfall and distributed physics-based hydrologic modeling. Radar rainfall detection and forecasting leverages rain gauge networks of the City, Lower Colorado River Authority, Upper Brushy Creek and NOAA. An innovative hydrologic information system built upon radar hydrology and Internet information access was implemented in 2004. It has proved to be useful during storms producing heavy precipitation and flooding. This paper describes the off-line configuration and online experience of this approach in Shoal Creek.The Shoal Creek watershed is a thirteen-square mile basin in the heart of Austin. It has a troublesome history of deadly flash floods. The May 1981 flood that took nine lives in this watershed triggered the establishment of the FEWS. The watershed is 43% residential and is home to approximately 65,000 people. A greenbelt path with wooded areas and parks follows much of the eleven-mile channel. The greenbelt path eventually enters into the urbanized mixed-use developments of downtown where businesses and high rise residents co-exist. The greenbelt is then connected to a larger system of greenbelts that run the course of Lady Bird Lake. While citizens enjoy the natural environment, danger can present itself quickly.Late afternoon on May 23, 2009, the Saturday of Memorial Day weekend, FEWS personnel received a text alert that heavy rainfall was moving into the Austin area. One of the team's responses was to check the web-based system for additional information. A steep rise in the hydrograph produced by the real-time continuous Vflo™ model was the basis for notifying the on-duty officer to clear pedestrians and cyclists from the paths and to deploy a photographer to take pictures of the flood's arrival. This photo documentation, radar rainfall data, hydrologic model prediction and verification, display system, and notifications will be presented as a case study in real-time stormwater management.