Hot off the press from Larry Cosgrove:
"I will admit that some of the recent weather model output has concerned me. There is the matter of the heat ridge over FL, The Bahamas, and Cuba, which the GFS and GGEM series insist will bring an early end to winter across the eastern half of the nation (say roughly to the right of a Chicago IL to Dallas TX line). But the dissenting ECMWF suite seems to recognize suppression of the subtropical high to the Caribbean Sea side, due partly by intense storms and increased ridging in the vicinity of Alaska and Greenland. Since the European panels have been consistent with this scenario for quite a few runs, I lean toward its colder solution.
With the weakening El Nino trying to assume a Modoki signal before dispersing in late spring (more on that later), there should be more opportunities in February to have cold troughs dominate the lower 48 states. The Madden-Julian Oscillation is not cooperating as of now (weakening and not linked to the polar jet stream), but may do so as a larger trough moves off of the Asian mainland in about a week or so. Last, but not least, the vast snow cover will continue to generate immense cold air that would be pulled south into the U.S. with an active southern/eastern storm track shown by the CFS and ECMWF weeklies. There is a risk for a significant cyclone in this arrangement around February 15 - 17.
But for all of this to happen, two things must be evident. That the European series is correct (because the other main ensemble packages will make snow lovers over the Mid-Atlantic and New England states cry), and the heat ridge complex drops down below the Greater Antilles. The latter feature is the kiss of death for winter, which looks to be the case next month as the subtropical high will almost surely expand or move northward into Georgia and the Florida Peninsula. But before that happens, in theory colder air could still dominate much of the U.S. before about March, when the analog sets shows a strong retreat of cP and cA values."