U.S. HDD Forecasts Trend Colder But Not Certain

Headline: Stratospheric warming? Favorable MJO? The U.S. thermal pattern goes cold if snow cover increases.

Fig. 1-2: Mega-cluster ensemble “most likely” North America temperature and precipitation anomaly forecast through the next 5 days.

U.S. day 1-5 discussion: Highlight of the 5-day forecast valid through Thursday night is two storms in the South and East, each with almost the same storm path and attendant weather. The first storm moves through the East today/Monday primarily as a rain event except ending as snow as far south as Arkansas to the Tennessee Valley and mostly all snow across the Northern Appalachians where Winter Storm Warnings are posted. The second storm follows QUICKLY forming in Texas early Tuesday and including both ice and snow across Western Texas. By dawn Wednesday, the storm shifts to the Tennessee Valley generating a lot of rain although snowfall extends from Oklahoma/Arkansas to Chicago at that time. Expect heavy snow in the Ohio Valley Wednesday night with rain moving into the East except freezing rain in New England. Over to heavy rain in eastern New England on Thursday. After the second storm, a widespread colder than normal outbreak Central into the East but arctic air stays in Canada.

Fig. 3-4: Mega-cluster ensemble “most likely” North America temperature and precipitation anomaly 6-10-day forecast.

U.S. day 6-10 discussion: Most important, is the extent of widening snow cover during the period. Temperatures are now consistently cold enough to maintain snow cover. In the 6-10-day period arctic air in Canada will infiltrate the Northern U.S. (late period in New York/New England) as snow cover develops and is maintained. Heaviest snow during the week is across the North-Central U.S. There is reluctance of significant cold to move south of snow cover. The forecast is biased toward the ECM solution.

Fig. 5-6: Mega-cluster ensemble “most likely” North America temperature and precipitation anomaly 11-15-day forecast.

U.S. day 11-15 discussion: There is enough stratospheric warming in the polar region to warrant additional arctic air development in Canada which easily slips into the U.S. due to increased snow cover. Forecast confidence is increasing primarily due to the general agreement on most operational models. However, at 46% confidence, this forecast is not a lock but the trend is considered reliable. Because of the colder forecast, most of the U.S. is cutoff from a moisture source to generate major storms except in the East where a low-latitude snow event could generate.

Fig. 7-8: Mega-cluster ensemble “most likely” upper air pattern across North America in 15 days.

In 15 days: The mega-cluster ensemble indicates two equally valued upper air solutions in 15 days (valid Feb. 6^th). Implied is below average confidence on whether the colder forecasts for late JAN/early FEB can last very long. Option 1 indicates a large polar vortex pattern which would likely generate more snow and cold for the U.S. Option 2 favors a retreating polar vortex to Southwest Canada and re-emergence of a warm upper ridge in the Northeast States. The day-15 forecast is very reliant on the extent of snow cover at that time. If snow cover is sufficiently expansive the colder forecast is correct. If snow cover expansion under-achieves the milder pattern will try to come back. The bottom line here is that the cold pattern ahead into early February is probably NOT a month-long regime and the mild tendency of winter 2022-23 is likely to regenerate later in the month.  

U.S. gas population weight HDD forecast discussion: The consensus of all forecast models U.S. gas population weight HDD projections through Feb. 9 yield a colder pattern ahead. National heating demand is certainly higher than normal for next week. However, note the discrepancy for next week as GFS forecasts near normal demand and GFS ENS/CMC ENS project significant heating demand. The national heating demand trend has increased since Friday for the Feb. 3-9 forecast.

Fig. 9: U.S. gas population weight HDD consensus forecast utilizing all models and comparing with 48 hours ago and the 10-year/30-year normal.