NOAA has decided to put Punxsutawney Phil through a reality check. Each Feb. 2, since 1887, the famous groundhog from Pennsylvania has been making predictions for the coming of spring, based on whether or not he sees his shadow.

"It's time to put Phil to the test," said Tom Ross, a meteorologist with NOAA's National Climatic Data Center in Asheville, N.C. Ross compared Phil's predictions with the U.S. national temperature since 1988. "Phil showed no predictive skill during these years," Ross said.

Ross added that since 1960, the U.S. national average temperature in February was above average about 53 percent of the time compared to 47 percent of the time below average. The statistics for March are even a bit warmer. March was above average 68 percent of the time and cooler only 32 percent during the same period.

Climate watchers who wish to do their own analysis comparing Phil's predictions with the historical temperature data housed at NCDC can do so online. Visit NCDC's Groundhog Day Web site at: http://www.ncdc.noaa.gov/ol/climate/extremes/2001/groundhog/groundhog.html

The site also presents background and folklore of Groundhog Day, links to other groundhog sites, and links to climate data from around the country.

Phil's prediction for this year will be announced on Feb. 2. Meanwhile, NOAA's Climate Prediction Center's 30-day outlook calls for below normal temperatures over the northwestern United States and the coastal areas of California and Alaska; warmer and drier than normal conditions in portions of Florida; warmer than normal temperatures from the Southwest to the Southeast and northward to the Great Lakes, Ohio Valley, and Middle Atlantic Coast.

"The theme of the Winter Outlook continues to be variability in temperature and
precipitation regimes," said Jim Laver, deputy director of NOAA's Climate Prediction Center. "This means periods of above, near and below normal conditions are expected for both parameters across the U.S. during the next several months. However, we do not expect a repeat of anything like the magnitude and duration of the November-December cold spell."