A comparison of multi-sensor (radar and gage) and gage precipitation estimates at growing season, monthly and daily temporal resolution and county spatial resolution was undertaken for the central United States. Precipitation data were collected from February 2002 through October 2006 from four sources: 1) Stage II radar-only precipitation estimates, 2) multi-sensor precipitation estimates (MPE) based on the Stage III/IV algorithm developed by the National Weather Service (NWS) Office of Hydrology and NWS River Forecast Centers, 3) daily quality-controlled NWS cooperative raingage (QC_Coop) data from the National Climatic Data Center (NCDC), and 4) raingage data from three regional networks operated by the Illinois State Water Survey. The gage data were employed as the reference standard.

Most differences in county-averaged monthly precipitation estimated by MPE and QC_Coop in nine Midwestern states and ~750 counties, were within +/- 15 %, averaging 6 %. Data from 3 regional networks indicated that on a daily basis MPE and gage correspond to within +/- 25 %, and that the MPE were a considerable improvement over radar-only precipitation estimates. Both daily and monthly data showed a tendency for MPE values to be smaller than the gage values for higher precipitation rates.

The correspondence between gage and MPE data was found to vary across the region. Possible reasons for this variation were examined including: distance from the radar, precipitation intensity, season, area coverage per gage, and latitude. Results are presented that partially explain the spatial variation in differences between MPE and gage amounts.