An Update on Verification Efforts for the NWS Probabilistic Snow Forecasts (286)
Philip Schumacher, NOAA/National Weather Service, Sioux Falls, South Dakota
David Radell, NOAA/National Weather Service
In its third official winter season, 60 NWS Weather Forecast Offices (WFOs) produced probabilistic snowfall information as a means to communicate uncertainty in the forecast. The forecast technique utilizes NCEP's Weather Prediction Center's (WPC) probabilistic snowfall percentile accumulation values to create a cumulative probability distribution function at every grid point, utilizing the official NWS WFO snowfall forecast as the modal value. Tenth and 90th percentile graphics, exceedance probabilities, and categorical range probabilities for a variety of thresholds are output and made available to end users for decision making. The 10th and 90th percentile values are communicated as the low end and high end snowfall amounts, respectively, while the official NWS WFO snowfall is communicated as the expected snowfall amount.
Given the large expansion of participants this season, along with the development of a national gridded snowfall analysis, a combination of point and grid based verification was performed over Central and Eastern Region WFOs for several events during the experiment. Summary metrics including bias and mean error, percent correct within the 10th-90th range and cumulative reliability diagrams were generated and used to access accuracy, reliability and if the underlying ensemble distribution is over/underdispersive. Verification results from this past season's winter experiment, and some comparisons with previous season's verification results, will be presented. The performance in areas impacted by lake effect snows will also be examined.