Is 12Z GFS Arctic Air 11-15-day Period U.S. Forecast Real?

A map of the weather AI-generated content may be incorrect.
Climate Signals Indicate More Interesting Cold Weather for U.S.
11/11/2025, 2:43 pm EST
A graph of different colored lines AI-generated content may be incorrect.
Chicago and Boston Winter 2025-26 Snowfall Forecast
11/17/2025, 4:34 am EST
A map of the weather AI-generated content may be incorrect.
Climate Signals Indicate More Interesting Cold Weather for U.S.
11/11/2025, 2:43 pm EST
A graph of different colored lines AI-generated content may be incorrect.
Chicago and Boston Winter 2025-26 Snowfall Forecast
11/17/2025, 4:34 am EST
Show all
A graph with a line and arrow AI-generated content may be incorrect.

A close-up of a logo AI-generated content may be incorrect.

Climate Impact Company Midday Update

Issued: Friday November 14, 2025

Highlight: Is the 12Z GFS arctic air in the 11-15-day period real?

A graph with a line and arrow AI-generated content may be incorrect.

Fig. 1: The 12Z GFS 15-day East Pacific oscillation forecast.

Discussion: At midday, the 12Z GFS forecasts a whopping -4 value for the East Pacific oscillation (EPO) during the 11-15-day period (Fig. 1). Other models are averaging -2 at that timeframe, about half the strength (although still impressive). The strong (-4) index supports arctic air involvement and on day-15, 12Z GFS indicates zero for morning low temperatures from Colorado to South Dakota (Fig. 2). Not surprising is the 12Z GFS forecast of moderate stratospheric warming across Canada (Fig. 3) which fuels the deep longwave trough at 500 MB causing the cold outbreak (Fig. 4). Additionally, the 12Z GFS upper air forecast enables a substantial gain in snow cover in the 11-15-day period for the North-central U.S. (Fig. 5). Is the forecast overstated? Yes! The difference (from previous forecasts) is the “ridge bridge” across Alaska which forces low-level air trajectories to tap Siberian arctic air pulling that air mass into the U.S. polar vortex. The “ridge bridge” is unlikely due to the cold SSTA across the Northeast Pacific which is well-correlated to an upper trough much of the time. The upper trough is likely, and the attendant cold develops (more) slowly inspired by the snow cover part of the forecast which is correct. Sprawling arctic air is not likely!

A map of the united states AI-generated content may be incorrect.A map of the earth with different colors AI-generated content may be incorrect.A map of weather forecast AI-generated content may be incorrect.A map of the north america AI-generated content may be incorrect.

Fig. 2-5: The 12Z GFS minimum temperature forecast on Nov. 29, 11-15-day stratospheric warming over Canada, the longwave cold trough over the Northwest U.S., and day-15 snow depth.

EIA EndForecast12-Hour Change24-Hour Change30-Year Normal10-Year Normal
11/20115.5-2.5-4.5134.7129.8
11/27112.0+2.8-25.6150.4145.8
12/4156.8+0.7-4.9165.8158.2

Table 1: The midday 12Z GFS U.S. gas population weight HDD projections into early December.