Towards Better Utilization of NEXRAD Data in Hydrology: An Overview of Hydro-NEXRAD

With a very modest investment in computer hardware and the open source local data manger (LDM) software from UCAR's Unidata Program Center, an individual researcher can receive a variety of NEXRAD Level III gridded rainfall products, and the unprocessed Level II data in real-time from most NEXRAD radars. Additionally, the National Climatic Data Center has vast archives of these products and Level II data. Still, significant obstacles remain in order to unlock the full potential of the data. One set of obstacles is related to effective management of multi-terrabyte data sets: storing, compressing, and backing up. A second set of obstacles, for hydrologists and hydrometeorologists in particular, is that the NEXRAD Level III products are not well suited for application in hydrology. There is a strong need for the generation of high-quality products directly from the Level II data with well-documented steps that include quality control, removal of false echoes, rainfall estimation algorithms with variety of corrections, coordinate conversion and georeferencing, conversion to a convenient data format(s), and integration with GIS. For hydrologists it is imperative that these procedures are basin-centered as opposed to radar-centered. Thirdly, the amount of data present in a multi-year, multi-radar dataset is such that simple cataloging and indexing of the data is not sufficient. Rather, sophisticated metadata extraction and management techniques are required. The authors describe and discuss the Hydro-NEXRAD software system that addresses the above three challenges. With support from the National Science Foundation through its ITR program, the authors are developing a basin-centered framework for addressing all these issues in a comprehensive manner, tailored specifically for use of NEXRAD data in hydrology and hydrometeorology. Through a flexible web interface users can search a large metadata database base, managed by a relational database, for subsets of interest. Well-chosen and documented defaults are provided for the flow from unprocessed NEXRAD data to basin-centered rainfall estimates at a desired space-time resolution. In addition to the web interface, there are web services that provide access to scripts and compiled programs.