Winter 2011-12: Forecasts versus Reality

[John Farley]

A month or two ago, there was some discussion on this board of how badly long-range forecasts were off for Winter 2011-12. Now that the data are in for climatological winter (December, January, and February) and I've had a little time to analyze the actual patterns compared to what was forecast, I've put together a Web page with some maps comparing the actual winter temperatures and precipitation to long-range forecasts by the NWS and Accuweather. Both predicted patterns that would be fairly typical in a La Nina winter, which this was, and both were off by quite a bit. The maps show the specifics of this, and I speculate about why the forecast may have been wrong, most likely due to a fairly persistent positive phase of the Arctic Oscillation. You can view the Web page at:

http://www.johnefarley.com/winter-fcst-actual-11-12.htm

[David Wolfson]

Interesting discussion, John. Continuing on the previous discussion you mention, I think there's a big question of causality implied in the common conclusions that weather did so-and-so because of the AO. As suggested by this winter's season, the predictive power of the (N)AO hasn't been well demonstrated by properly-applied statistical research. IMO. I suspect that a stronger case can be made that fluctuating mid-latitude weather patterns cause changes in the (N)AO. FWIW.