This project will combine research on the management of large, real-time, data streams with research on algorithms for deriving rainfall estimates from raw radar data, to develop an information technology framework for use in the analysis of rainfall data from the NEXRAD radar network. It includes research on and the development of metadata for the NEXRAD raw data feed, archived NEXRAD data and derived data products, client libraries for retrieving metadata and data, and client applications for quality control, data preparation, and the generation of rainfall maps. If successful, the results should greatly reduce the time and effort required to locate relevant NEXRAD data and to convert it to rainfall fields that can be used to drive hydrologic models. The project will also help standardize the process of rainfall field generation, increasing the comparability of results from different hydrological modeling groups. The framework will include a browser for radar data, a map display of rainfall products, and an analyzer that permits the overlaying of geo-referenced rainfall fields, digital elevation models, and surface channel networks. Standard and customizable methods will be available for quality control and rainfall field generation by users of the system. Included in the tools planned for development is software to create merged rainfall products when more than one radar covers a region. If successful, the project will have a potential impact on hydrological science by providing the data that is the primary driver of dynamic hydrology in a fashion similar to the way weather data is available to weather researchers. This, in turn, will facilitate better prediction of soil erosion, flooding and landslides as well as improve environmental resource management. Because metadata is being included in THREDDS catalogs, NEXRAD data will become accessible through DLESE, the Digital Library for Earth Science Education.
