Several days ago, one of the computer models forecast a major storm for Thanksgiving Day (Winter Storm for Thanksgiving Day?), but I posted it with the disclaimer that the 384-hour GFS (the model I made reference to) was famous for its inconsistency.
Speaking of inconsistency, the latest run (the 288-hour of the GFS) of the same computer model leaves us saying “What storm?”:
Forecast for Thanksgiving Day (12z run of GFS from November 15, 2009)
Inconsistency
Model inconsistency in the long-range is expected from the GFS (and all computer models), so it’s not surprising that the storm it predicted for the middle of the country is no more. These long-range computer models are used much more by us (forecasters) to help forecast the overall pattern, not the specific forecast for a specific location.
Forecast for Fun
Understanding that the specifics are again likely to change, here are a few bulleted highlights of the forecast–just for fun:
- Morning rain in New England to the coastal Mid-Atlantic region
- Chilly with a couple of rain and snow showers in New York state and Pennsylvania
- Winter cold with some mainly light snow from the upper Midwest to the northern Rockies
- Dry and pleasant (with some sunshine) in the South, extending from California to Florida
Next Update on Thursday
We’ll look at the forecast again on Thursday, which will be just a week out. By then, the specifics of the model are more within the range of what can be expected to be reasonable.
–Paul Yeager
