Hydro-NEXRAD: Metadata computation and use

To enable the efficient use of the Hydro-NEXRAD system in hydrologic research, the authors are developing and computing a set of metadata. Vast archives of radar-centric NEXRAD data, the highly variable spatial and temporal characteristics of precipitation events, and the existence of anomalous data complicate the process of selecting subsets of the data for specific investigation. Well-defined, pre-computed, radar-centric, basin-centric and point (for a selected set of rain gauge locations) metadata offer the possibility of efficiently selecting data subsets that meet user-specified criteria. Metadata have to be computed quickly so that they do not slow down the data ingest process, but they do not have to be highly accurate. Separating rainfall from clear air conditions and heavy rainfall from light precipitation is often sufficient. In this paper, the authors discuss the technical challenges of computing metadata for the vast archives of NEXRAD data and basic data quality control incorporated into that process. Users of Hydro-NEXRAD will be able to query the database using different criteria based on the metadata or visually inspect the time series of the metadata.