Sterling, VA verification (courtesy of WFO Sterling, VA)

The entire Sterling VA CWA experienced winter storm criteria snowfall during Dec. 4-5, 2002, except for the area east and south of Fredericksburg VA. At BWI and IAD, 7.4" and 6.2" were received, respectively, representing the heaviest early season snowfall at each location since November 30, 1967. DCA's 6.1" snowfall was the heaviest so early in the season since Veteran's Day 1987.

How did the ensemble QPF verify? The maximum probability for heavy precipitation had shifted somewhat south from the day 2 to the day 1 forecasts, and that precipitation onset was forecast to be later as well. See graphics below for 12-hour precipitation mean and spread from day 1 valid for 24-36 hours, as well as the animations, for a refresher.

It appears that the precipitation, if we assume a 12:1 ratio of snow amount to liquid equivalent, was underpredicted by the SREF in the northern and western portion of the Sterling CWA (and further north as well, with 9" of snow in southeastern PA). Note the 7" at Cumberland where we only put up winter weather advisories. The Martinsburg WV snow amount was pretty much on target. Even at BWI, actual liquid equivalent was 0.96", while the ensemble mean appears to be just over 0.5" over 12 hours (and about 0.6" for 24 hours, not shown). The other two observing stations (IAD, DCA) had about 0.6" from the storm. The rain/sleet/snow transition zone was positioned to the south of the Sterling CWA through most of the event; precipitation changed to sleet and freezing rain at DCA at about 16 UTC on 5 December 2002 for about 2 hours. IAD reported all snow, while BWI mixed with ice pellets at the end of the event. Accumulations of snow of 3-5" to the southeast of DC indicate that the precipitation was diagnosed as freezing rain or rain somewhat too soon relative to verification in these areas. Subsequent glazing was less than 0.25" in thickness.