Close Window

Univ. of South Carolina: "Development of a GIS database for high-risk dams in South Carolina"

Final Report

Forecasters and hydrologists at the Columbia NWSFO are routinely challenged by heavy precipitation events and the ensuing threat of dam failure. In the southeastern U.S., abundant moisture, intense surface heating, and tropical storms during summer and fall months combine to produce heavy rainfall across a region characterized by a disproportionately large number of small dams, many of which are susceptible to failure. The WSR-88D system has provided forecasters with high resolution precipitation data-a first step in mitigating flood hazards. However present manual methods used to identify the river basins and dams affected by heavy precipitation are too slow to provide adequate warning to the public. Incorporating precipitation products from the WSR-88D system into geographic information systems (GIS) could help NWS forecasters identify more precisely where heavy precipitation has fallen in relation to specific dams and, therefore, to issue more accurate flood forecasts and more timely warnings.

The effort was an extension of a previously successful project that incorporated WSR-88D digital precipitation array data with other spatial databases using ARC/INFO and GRASS. The goal of this project was to build a comprehensive and accurate spatial database of weather spotters, emergency management personnel, population, roads, hydrography, and dams and flood-prone in Columbia County and Richmond County in Georgia. To that end, the researchers developed an efficient method for ingesting hourly digital precipitation estimates into an object-oriented GIS (ArcView) and built high resolution spatial databases for the two counties within the WSFO CAE county warning area. These databases, built with Arc/Info GIS software, can now be displayed as coverage using ArcView. Coverage can be chosen individually for display with hourly precipitation estimates from the 88D. An Avenue script is currently being written to automate the transfer of real-time precipitation data to a separate displayable coverage to help forecasters during extreme meteorological events. This interface will allow forecasters to construct real-time precipitation maps and examine spatial patterns in the context of high-risk dams, vulnerable populations, and local spotters. Most importantly, the study serves as a prototype for other counties, and for other forecast offices that may choose to employ this technology.

These efforts have produced a system to help forecasters identify more precisely where heavy precipitation has fallen in relation to many important features. This should lead to more accurate flood forecasts and more timely warnings.