The 12.1 inches of snow that fell over Madison in the first nine hours of Saturday morning was surely a surprise to many residents while setting the all-time monthly record.
Forecasts made through the late afternoon on Friday, consistent with those made on Wednesday and Thursday, were suggesting that the heaviest band of snow would run southwestward from Racine/Kenosha back to northwestern Illinois and leave Madison with an inch or two.
GOES East Visible satellite imagery from March 25th paired with Day Cloud Phase distinction RGB imagery showing where the heaviest snow fell and a sharp cutoff where no snow occurred.
The forecast was in error not because the heavy snow band was not foreseen, but rather because the band ended up occurring farther northwestward than the computer forecast models had predicted.
This error underscores how hard it is to pinpoint the location of heavy precipitation bands in winter storms, especially rather weak winter storms like the one that hit on Saturday morning.
Even such weak storms have circulations that impact millions of square kilometers, while the precipitation bands impact only a small fraction of that area. In this case, the important structures and dynamics that conspired to produce the snow were well represented in the computer forecasts, but their location was different than the forecasted location for reasons that can be determined only in an after-the-fact investigation of the event.
Though much of the stunning progress in forecasting that has occurred in the past half century has been a result of relentless advances in theory, observational capabilities and the expanding power of computers, an often overlooked ingredient is the grueling detective work that ensues in the aftermath of such forecast errors. Meteorologists undertake “case studies” of such incidents to determine what physical factors were responsible for the weather as it actually occurred and also to better understand how and why the computer models went astray.
This difficult but necessary work informs the future development of these forecast models with the hope of minimizing future forecast errors. Thus, though such incidents as our Saturday snow might inspire the cynical view that weather forecasting is not a scientific endeavor, the exact opposite is actually true.
Steve Ackerman and Jonathan Martin, professors in the UW-Madison department of atmospheric and oceanic sciences, are guests on WHA radio (970 AM) at 11:45 a.m. the last Monday of each month. Send them your questions at firstname.lastname@example.org or email@example.com.