Validation of gridded precipitation datasets for flood-typing in six basins in the contiguous U.S.

The U.S. Geological Survey (USGS) and the U.S. Army Corps of Engineers (USACE) are collaborating with the U.S. Federal Emergency Management Agency (FEMA) on improving flood-frequency analysis methods to account for mixed populations arising from different flood causal mechanisms. Precipitation data at different timescales are widely used in flood-typing studies. Various gridded precipitation datasets were validated by comparison against station observations to support flood-typing over six pilot regions in the contiguous U.S. (CONUS), where flood-typing approaches will be initially tested. The six pilot regions are (1) the Delaware River, (2) the Iowa River, (3) Puget Sound, (4) the Red River of the North, (5) the Trinity River, and (6) the Upper Colorado River. Various precipitation datasets derived from station, radar, reanalysis data, or combinations thereof, were validated in terms of their ability to capture the spatiotemporal characteristics of daily precipitation as well as multi-day (1–14 day) extreme precipitation events. The datasets were validated by comparison against gage data from the NOAA Global Historical Climatology Network daily (GHCNd) for the periods 1981-2013 and 1998-2013. Taylor diagrams and the Kling-Gupta efficiency (KGE) metric were used for validation.

This data release consists of three tables in EXCEL spreadsheet format:

Normalized Taylor diagram statistics for daily precipitation by season in each pilot region, for the two periods (All_precipitation_statistics.xlsx).
Normalized Taylor diagram statistics for multi-day extreme precipitation in each pilot region, for the two periods (Extreme_precipitation_statistics.xlsx).
List of GHCNd stations used for evaluation of daily precipitation by season and for evaluation of multi-day precipitation performance in each pilot region, for the two periods (GHCND_stations.xlsx).