AccuWeather.com

Other Pieces of the Long-Range Dilemma

Jan 12, 2012; 11:05 AM ET

Thursday, 11:30 a.m.

One of the many hats I have the honor of wearing here at AccuWeather.com is that of long-range forecaster. We have a subdivision that focuses exclusively on forecasts beyond the typical three-to five-day forecasts you hear on the radio, see on television or see in newsprint. It's a fascinating and relatively new part of the weather forecasting business and one that is growing quickly as we get a better grasp on what makes the weather work.

The problem with long-range forecasting, or one of the problems (because there are many), is that the more you know, the more you still don't know. The more we research things, the better we see how the global atmosphere works together. However, since we have a relatively short history of detailed weather observations, the more we uncover, the harder it seems to match up what we see with weather patterns of the past. The sample size is so small that it almost renders it useless, and you end up no better off than if you didn't both looking at the past!

That's not really true, but just look at this winter. Given the fact we were in a second La Nina, one following the pretty strong one going into the winter of 2010-2011, we had an idea about how the winter pattern might evolve. A number of previous winter season came up as 'analog' years, or winters that seemed to be fairly close matches to the current winter.

Up to this point, though, you'd probably have to say the winter season forecast has not been the very best, for any number of reasons. Those very analog years have had only limited usefulness in carving out a forecast, and most have seemingly led forecasters down a rabbit trail. The one that I harped on back in the fall was 2001-2002, and so far that may turn out to be the best fit for this winter.

Anyway, one of the many tools we use in making those long-range forecasts is the weather way up in the atmosphere - at the 10mb level. You'd think at an altitude so high the weather would be a poor term to use - it's cold and dry and with so little air to begin with. But studies have been performed in recent years to try and find some correlation to the changes at those high altitude and the weather here on the earth.

One of them that has been given a fair amount of consideration is an event called stratospheric warming, a time when temperatures (not really the best term to use with so little air, but it is how the parameter is measured) at that level go up. What has been observed over time is that often that kind of change seems to correspond to some form of arctic outbreak.

Early this month, we observed a warming event over Alaska and western Canada. Following the traditional thinking, that should correlate to an outbreak of arctic air some seven to 10 days down the road. Look at what is happening now - much colder air is drilling into Texas and across the Mississippi Valley.

Since that warming, temperatures have slowly been cooling over this area. Here's a look at the most recent 10mb analysis:

The numbers of importance are hard to read, so I'll direct you to this website:

http://weather.uwyo.edu/cgi-bin/uamap?REGION=np&OUTPUT=gif&TYPE=obs&TYPE=an&LEVEL=10&TIME=2012011200

Let me show you a close-up of Alaska into Siberia:

That should give you a better idea of the warming taking place there. I'm new to this line of forecasting, but my sense is that if this the area of warming, I would think that might be a place where I might want to look for a build-up of arctic air in a week to 10 days. In other words, on the 'other' side of the pole, rather than on the Canada side. Now look at two other charts. The first is the Canadian ensemble forecast for next Wednesday evening:

That's a large vortex over northern Canada, removed to the south of the North Pole, and one supportive of a jet stream farther south over the next week. With it getting extremely cold again in Alaska, some of that cold air will flow downstream through Canada and at times move across the border into the upper Great Lakes and Northeast, if not farther south at times. You might also note that there is an upper-level trough in the Gulf of Alaska, and if you look at the GFS ensembles, there's a boat load of cold air being forecast to head southward toward the Northwest this weekend and early next week.

Now look at the same Canadian ensemble forecast for the following Wednesday:

Hmmm... It seems that vortex has retreated to the west and northwest and grown much smaller in size. If that's right, what could it mean? Well, if we tag on to the 10mb forecast thought I had earlier, it might mean the coldest air is retreating back toward Alaska and into Siberia. If you look at the 6z GFS ensemble forecast for Thursday, Jan. 26, this is what you'd see:

Then compare that to another chart, a NAEFS eight- to 14-day temperature anomaly composite:

All I can say is interesting. So many tools, so little time. Clearly the implications are that the next week to 10 days, just about the climatological coldest time of the year, may be as cold as it gets in much in North America for the rest of the month, and that the last third of the month has a growing trend toward warmth once again.

The 10mb forecast theory, for lack of a better term, will be interest to test in the next two weeks. Much has been learned, but much more is to be learned!

The views expressed are those of the author and not necessarily those of AccuWeather, Inc. or AccuWeather.com