This year, the Old Farmer’s Almanac (OFA) is first out with its winter forecast. Its headline? “Above-normal temperatures (almost) everywhere.” This prediction is actually pretty similar to the National Weather Service’s, which also favors above-normal temperature over the majority of the nation, with the exception of the Southeast.
OFA’s forecast for a mild winter is also in keeping with the long-term trend in which the season has warmed due to climate change.
In other words, it is not an unreasonable forecast.
But it does have some weaknesses.
It mentions the “expected arrival of a weak El Niño” among the rationale for its warm and wet outlook. While it is true El Niño and its associated moisture feed off the Pacific Ocean frequently make large parts of the country wetter-than-normal in winter, the temperature connection is less clear.
“The ‘wet’ aspect of the forecast is reasonable,” says Capital Weather Gang contributor Matt Rogers, who specializes in long-range forecasting. “Weak to moderate El Niño events since 1990 have been mostly wet, but for temperatures they have split 50-50 between cold and warm winters causing the temperature part of the forecast to be more uncertain.”
The OFA outlook also says “a decrease in solar activity” favors milder-than-normal conditions. We have actually noted the decreasing solar activity tends to favor blocking patterns in the Eastern U.S. which can lock cold air in place. So that is another area where OFA’s forecast is vulnerable.
As we’ve mentioned when we’ve critiqued OFA and Farmers’ Almanac forecasts in the past, their specific forecasts for individual days and weeks are most problematic. Weather prediction has not yet advanced to the point in which we can say months ahead of time what specific conditions will be like in a given week or set of days.
While these forecasts are best characterized as pseudoscience, here is what OFA is calling for at particular times in its Atlantic corridor region, which includes the District, Maryland and Virginia: “Winter temperatures will be much above normal, on average, with the coldest periods in early to mid-December, early and late January, and early February. Precipitation will be slightly above normal, with below-normal snowfall. The snowiest periods will occur in early December, late January, and mid-February.”
In our view, these predictions are meaningless. It is worth noting that in most of the recent El Niño winters, the coldest weather with respect to normal has occurred toward the end of the winter, with the mildest weather at the beginning. The OFA forecast does not reflect this tendency.
Even the more generalized outlook for winter, without getting into the specifics, is fraught with uncertainty. If El Niño does not develop as predicted or strange things happen with the polar vortex, for example, the whole outlook could be turned on its head. OFA forecasts never communicate any uncertainty or qualifiers, which is a real weakness.
We must also point out OFA advertises 80 percent accuracy for its outlooks, but has never published data to substantiate this claim.
Last winter, the OFA correctly predicted a mild winter in the eastern U.S. but its temperature forecasts were errant in many other parts of the country. It predicted cold conditions in the Southwest where it was much warmer than normal. And it called for mild conditions in Montana and the Dakotas, where it was the opposite.
OFA’s 2017-2018 winter forecast
Actual temperatures compared to normal
Next week, the Farmers’ Almanac will publish its outlook. Stay tuned for our critique.