If you like a lot of snow, this winter may not be your favorite season.
That’s one of the main highlights from my 21st annual Long-Range Winter Forecast.
After crunching all the numbers and analyzing everything from air patterns to ice in the Arctic, it looks like we’re going to have a mild winter and less snow than our average. I’m going to outline the forecast first and then go into detail about what’s at play.
I’m not expecting that we’ll see a single huge snowstorm. We will likely get several smaller snow events, however.
As for temperatures, overall they’ll be milder, but January and February won’t be as warm as last year. Still, February will be the warmest month of this winter. December will actually be colder than average, though.
Let’s get into the specifics for Philadelphia:
MONTH SNOW TEMPS
Local
Breaking news and the stories that matter to your neighborhood.
DECEMBER 4 in. -1 degree
JANUARY 6 in. +2 degrees
FEBRUARY 4 in. +4 degrees
MARCH 1 in. +2 degrees
TOTAL: 12-18 inches
(Normal, by the way, is 22.4 inches)
As for other cities in our area, here are my estimated seasonal snow forecasts:
- Allentown: 31 inches — 2 in. below normal
- Downingtown: 24 inches — 4 in. below normal
- Wilmington: 12 inches — 7 in. below normal
- Atlantic City: 9 inches — 8 in. below normal
- Millville: 7 inches — 6 in. below normal
- The Poconos: 63 inches — 6 in. below
So how did we get here? Here are the factors I look at to develop the forecast:
LA NINA
It sure looks like a La Nina, a weather pattern that features cooler-than-average sea temperatures, is developing in the Tropical Pacific.
The blue colors on the bottom figure represent below average ocean temperatures. That’s the La Nina (if they were red, it would be an El Nino). The colors aren’t dark blue, which would be moderate or strong.
A small area near the coast of South America is coldest of all. That’s why some are calling this an “East-based Nina.”
There is evidence that East-based Nina’s are more favorable for cold and snow than “West-based.” But if you remove the strong El Nino’s and La Nina’s, the difference is only about 5 inches of snow for the season (on average).
Another question is: Will the La Nina last through the whole winter? The forecast says: Yes...sort of.
Many different computer models are shown above, just like we see with "ensemble forecasts" with winter storms and hurricanes.
A number lower than negative .05 is considered a La Nina (the lower the number, the stronger the La Nina). Most models keep the La Nina weak through the winter. A few strengthen it, and a few weaken it to a “neutral” category.
I’ll use the average, and assume a weak La Nina through the winter. Even if it weakens some, the effects can linger for months afterward.
THE WARM OCTOBER FACTOR
The great local forecaster, Tony Gigi, of the National Weather Service, has done research on the connection between a warm fall and a relatively warm and un-snowy winter.
He uses the combination of October and November. Since we just had the second warmest October ever recorded here, it wouldn’t take a very warm November to put Tony in the milder, less snowy camp.
Me, too, especially after seeing this map of what the warmest fall combinations have looked like for the upcoming winter.
That’s a very mild winter in our part of the country. The warmest month relative to normal was February.
THE AO AND NAO
If you’ve followed winter forecasts over the years, you know how big a factor the Arctic Oscillation (AO) and North Atlantic Oscillation (NAO) can be.
Negative values for either the AO or the NAO are directly related to blocking patterns in the upper atmosphere that end up favoring cold and snow in our area.
Our snowiest winter of all, 2009-10, with a whopping 78.7 inches, was largely influenced by a strong and consistent negative-AO.
The problem is: It’s really hard to predict the AO or NAO months in advance. They tend to be much more variable than other patterns in the oceans.
There have been some recent advances in those predictions, and they suggest our winter will probably be dominated by a positive AO. That would favor mild with less snow than normal.
SIBERIA AND EURASIA OCTOBER SNOW
This is one I’ve looked at for more than a decade, but it only recently has become widely discussed among forecasters.
Dr. Judah Cohen has pioneered this research, which shows that October snow in Siberia and Eurasia is correlated with our winter weather patterns.
His latest report shows a much slower advance of snow than last year:
The left picture is 2016, and the right, 2017.
THE QBO-INFLUENCE FROM HIGH ABOVE
The Quasi-Biennial Oscillation or QBO is an especially interesting factor for several reasons.
First of all, it happens way up in the stratosphere, above all the clouds and storms. The winds stay in the same direction for perhaps a few years, and then suddenly changes to the opposite direction. And it can have significant impacts on weather patterns below.
The positive-QBO represent winds from the west. It tends to lead to a strong jet stream across the Pacific, and brings mild air our way.
The negative-QBO represents the east winds. That tends to lead to weaker jet streams, which then favor blocking patterns in the atmosphere. That, in turn, favors more cold and snow for us.
The QBO is more easily forecast than other oscillations, since it tends to be much more consistent from month to month. It is currently a negative-QBO, and is expected to continue that way. So, the QBO is a rare factor this year favoring more cold and snow.
THE PDO AND OTHER OCEAN FACTORS
The PDO (Pacific Decadal Oscillation) has a big influence on weather in the U.S., since it covers so much of the Pacific Ocean just west of us. There are particular patterns that are called "negative" and "positive," just like other oscillations.
Research has shown that a positive-PDO is more favorable for cold and snow in our area. Tony Gigi notes that this is true for 7 of 10 winters. And 6 of 8 winters of negative-PDO were milder, with less snow. As you see, there are no perfect correlations-they just change the odds.
The PDO is hard to predict months in advance, but it seems to be trending more toward the negative.
I can't help but notice the amazingly warm Atlantic waters just off the U.S. East Coast. Records have been set this year for warmest water in many places. Here is a current map:
That’s a lot of red! There doesn’t seem to be much research into general Atlantic Ocean temperatures, and their effect on winter weather.
Even though most winter weather moves from west to east (making the Pacific important,) a unusual ocean pattern like this has to have an impact. The warmer water will surely add to moisture available for storms, which can lead to more precipitation.
But any ocean wind brings abnormally warm conditions with it. This would likely cause borderline rain/snow type storms and we'd likely wind up in the rain category more often than not. At least that’s my theory, and I’m sticking to it.
ARCTIC SEA ICE
I’ve mentioned this in previous winter forecasts. The rapid ice melt in the Arctic accelerated starting around 2007. It makes the Arctic much different than it was even a few decades ago.
The prevailing theory is that the huge ice melt has led to even more warming in the Arctic than other parts of the world. That leads to less of a temperature contrast. And that leads to a weaker jet stream. Which leads to more blocking patterns in the atmosphere. And if you’re stuck in the wrong part of the block, you can get Arctic blasts and/or snowstorms. This could have something to do with our two snowiest winters ever recorded occurring in the past 10 years, even with a warming climate.
But this year, the ice melt isn’t as extreme as last year. That may cut the influence down a bit.
COMPUTER FORECASTS
Just like hurricanes and winter storms, we have sophisticated computer models that try to forecast weeks, and even months in advance. Their skill has been improving slowly over the years.
It just so happens that just about all the computer models I’ve seen show a mild winter. Here are a few examples for February:
As with the other maps, the yellow is slightly above normal, and the orange and reddish colors are way above normal.
