Saturday, June 15, 2013

Preliminary 2013-2014 Winter Forecast

Hello everyone, and welcome to The Weather Centre's 2013-2014 Preliminary Winter Forecast. This forecast will deal mainly with analogs and long range models, with no official forecast being issued. I will give my educated guesses on the upcoming winter at the end of this post, however. We have a lot to discuss, so let's get on with it.

We will begin with the El Nino-Southern Oscillation (ENSO). The ENSO phenomenon involves anomalous sea surface temperatures in the waters off the coast of South America on the Equator. In the image above, you can see the colder than normal waters between 140W and 80W. These below normal waters indicate the presence of a La Nina. In the winter months, a La Nina brings about cooler weather for the northern US and a snowier winter for the Midwest and Ohio Valley. When the same areas shown in blue in the image above are orange (above normal sea surface temperatures), an El Nino is observed. This flips the effects of the La Nina, giving warmth and below-normal precipitation to the Midwest and Ohio Valley, while distributing big storms to the Northeast and Mid-Atlantic. There is also the option of not having an ENSO phase at all, where the water temperatures are neither below normal or above normal. This phase is called the 'Neutral phase', because it has neutral water temperatures. Last winter, we were in the neutral ENSO phase. Because the ENSO was too weak to have an effect on the global weather, the winter's pattern was more determined by smaller-scale indices, like the North Atlantic Oscillation (NAO), the Pacific North American index (PNA), and the Arctic Oscillation (AO), all of which will be explained later in this post. This winter, long range forecasts are going for either a neutral ENSO winter again, or a weak El Nino. While the long range model forecasts you will see below are more La Nina-ish, there is still a lot of time for long range model forecasts to change.

We will now discuss analog years. These are years that were chosen by me after seeing how close a correlation there was between three different synoptic atmospheric indices in previous springs and those same indices for this spring. Based on a rigorous examination of multiple years, the winters of 1951-1952 and 1962-1963 were seen as the best candidates for detecting what our upcoming winter could be like.

We begin with the mid-level geopotential height anomalies for these two years. For all future images unless noted, cold colors indicate low pressure/below normal, and warm colors depict high pressure/above normal anomalies. In this image, we see quite a few things. The first is multiple areas of high pressure inside or in the immediate vicinity of the Arctic Circle. These high pressure areas appear to be close enough to the North Pole that they are breaking up the polar vortex, which I had discussed in great detail last year. The synoptic effects of this high pressure presence include more accessibility for cold air to shift into lower latitudes from the Arctic Circle, as well as a weaker polar vortex in the stratosphere, which we will discuss in greater detail later on in this post. We see high pressure encompassing the area of Greenland and in the waters to the south of that landmass. The presence of high pressure in that area indicates the presence of a negative North Atlantic Oscillation (NAO) index. When the NAO is negative, the jet stream will buckle south from Canada and give many in the Central and East US a shot at cold weather. The buckled jet stream also opens the door for Nor'easters. Low pressure over North America is amplified by the presence of high pressure over the Bering Sea that appears to extend down into the West Coast. These two high pressure systems cordon off the low pressure into North America. We also see a stormy East Asia, and if you know about the 6-10 day gap between storms in East Asia and storms in the East US, you should be jumping for joy with that development. I will discuss that further later on.

Temperature anomalies for these two years show a chilly North US and a slightly mild South Plains. It looks like the negative NAO pattern we discussed above, combined with the other high pressure system in the Bering Sea and low pressure over North America will allow cold air from Canada to flood south. The coldest weather should stay in the northern tier of the United States, with the general cutoff line starting from the Mid-Atlantic into the Central Plains. The West Coast would also see warmer than normal temperatures if these analogs came to fruition.

Back to the stormy East Asian correlation, it is known that when a storm system appears in East Asia, a storm system with relatively similar strength will appear in Central and/or East Asia. My analog years in the first image showed a very stormy East Asia, and if the negative NAO is able to appear quite a bit throughout the winter (which is plausible, if above normal water temperatures in the Arctic Circle continue), Nor'easters would certainly be a very real possibility. The East Asian correlation also opens the door for potential winter storm events from the Plains to the Northeast- this correlation would benefit many in this nation who love snow.

I want to zoom in on 1962-1963, because after analyzing my three analog years (and then discounting one) up against each other for the three atmospheric teleconnections, 1962-1963 easily won in all three categories. This means that the winter of 1962-1963 could align with our upcoming winter.

 The mid-level atmospheric pattern for the winter of 1962-1963 is one only dreamed about by winter weather lovers. We have so many things going right in this picture, it's hard to know where to start. We'll begin with Greenland. Once again, we have dominant high pressure over the landmass, providing a base for a  strong negative NAO. As previously mentioned, this allows cold air and storms to hit portions of the Central and (especially) East US. High pressure observed in the first image of this post is amplified in the winter of 1962-1963, and this high pressure even extends into the Northeast Pacific and snakes into the West Coast. The placement of high pressure into the Western US allows the development of a positive Pacific-North American (PNA) pattern. In the positive PNA set-up, high pressure in the West US allows the jet stream to stay north of the ridge before plummeting south in the Central US to allow a favorable winter storm track and cold event in the Plains, Midwest and Ohio Valley. The third positive item in this image is the presence of the same two ridges of high pressure I just discussed, and their placement in the upper latitudes. These two areas of high pressure are interrupting the previously-mentioned polar vortex to allow the negative phase of the Arctic Oscillation (AO) to develop. The negative AO indicates high pressure over the Arctic, leading to a southward movement of the jet stream and subsequent outbreak of cold into much of the lower latitudes. This outbreak coud be favored towards North America if the positive PNA and negative NAO are in full bloom at the time that this happens.

Temperature anomalies across the United States for the winter of 1962-1963 were very cold, with temperature departures of 6 to 8 degrees below normal in the heart of the eastern Midwest, Great Lakes and northern Ohio Valley. A general cold trend continues for those in the Northeast, Mid-Atlantic, Southeast and Gulf Coast. The Southern Plains even get a taste of chilly weather. Considering the triple threat of the negative AO, negative NAO and positive PNA evolving (in addition to the stratosphere, which will be discussed next), these very low temperature anomalies are not surprising in the least.

This is the part i'm really excited about. Last winter, I talked a lot about the impending collapse of the polar vortex and how it could move south. It did weaken significantly and eventually collapse in on itself. Movement into Canada did occur, but the vortex did not fully enter the US (portions of North Dakota and Minnesota were clipped by it). This winter, if the winter of 1962-1963 evolves correctly, the polar vortex will be sitting right over Canada and parts of the US. The above image shows geopotential height anomalies for the 100 millibar area, defined as the lowest layer of the stratosphere and the layer most readily checked for placement of the polar vortex and areas it may affect. This image shows dominant high pressure from Greenland to Eurasia, meaning the polar vortex has to be exiled to lower latitudes; it cannot stay in the Arctic Circle with those high pressure systems. Now, if you were alive in the 1980's and still remember that decade, you may recall the month of January, 1985. If you remember it as being absolutely frigid in the last few days of that month, you remember the cold associated with the polar vortex dropping south into the US. This happened as a result of high pressure in the upper latitudes pushing the polar vortex down into the United States. A similar, weaker event of a collapsed polar vortex happened in early 2009, when many in the North US experienced very cold temperatures. If the atmospheric pattern being pictured above were to evolve, not only would we see the polar vortex angled in a position more able to push towards the United States, but prevalent high pressure over the Arctic Circle could weaken it enough to send it towards this nation in weaker (but still frigid) pieces. It's far too early to tell if this analog year will actually work out, and much, much too early to determine where the polar vortex will end up. However, if these analogs are correct, I see no reason to doubt an incoming chilly (and possibly stormy) winter.

Now that we have looked at the past, let's compare the analogs to what long range models have to say.

Shown above are two long range forecast models (GFDL and NCAR), predicting 200mb heights for the month of December 2013. Areas of blue depict general lower heights and thus low pressure areas, while reds indicate high pressure systems. The contour lines are superimposed on the images to help point out areas of differing pressure. There are four regions I want to look at here, so we'll start out with North America.

Both images show a pretty favorable set-up for cold weather as far as the 200mb level goes in North America. We have a large area of rather strong low pressure over eastern North America stretching into New England, and this is reflected by the compression of contour lines in the same region. The low pressure area over New England tells me this area would indeed be favored for at least somewhat-frequent shots of Arctic air, and the tightening of the jet stream favors this theory. The overall picture is what excites me more. If we look into the far Northern Hemisphere we see no area of low pressure- the closest persistent low pressure center is the one in North America. Now, bear with me here, the idea is pretty outland-ish. But if you were to try to place a location for the polar vortex to average out over the month of December, you would want to find the area with the lowest pressure over the general upper latitudes (including North America, northern Eurasia, and Greenland). That said, it would make sense that the polar vortex may very well have a tendency to be pushed further south towards Canada and the US.

Our second area of interest is the United States. Both models show that low pressure area trying to push into New England, but we already covered that in the above paragraph. The item I want to focus on is the West Coast into Alaska. We see a separation of the contour lines into a looser composition, as well as oranges building into those areas. This indicates the presence of high pressure, and sets up positive phase of the Pacific-North America (PNA) pattern. In the wintertime, the jet stream does exactly what is being depicted above, and this means a warm West US and cold East US. Additionally, storm systems will tend to track in a path favorable for snowfall in the Midwest, Plains and portions of the Great Lakes into the Ohio Valley. The positive PNA helps to usher in lower than normal heights across the East US, and this will deliver us into the next area of interest.

The third area of interest is the Bering Sea. Both models project strong high pressure in the vicinity of the Bering Sea region; the GFDL model (left) has high pressure square in the Bering Sea, while the NCAR model (right) has the high pressure more towards Asia but still influencing the Northern Hemisphere. In both situations, the Arctic Circle appears overcome with above normal height anomalies, indicative of high pressure in the region. High pressure in the Arctic would imply a weaker polar vortex (yes, the same vortex I discussed this previous winter), as well as tendency for cooler weather to make its way down to lower latitudes of the world. If we do have high pressure in the Bering Sea, it would be easier for instances of high pressure to overwhelm the Arctic, and this could (keyword is COULD) allow the polar vortex to be pushed further south and affect lower latitudes in significant ways.

The final area of concern is East Asia. All models I observed for December 2013 depicted a tightening of the contour lines and lowering of height anomalies in East Asia, sometimes even into the northern Pacific. I have described a few times on this blog the idea of the East Asia-East US storm correlation, where a stormy period in East Asia can mean a stormy period in the East US 6-10 days later. This would mean more chances for snow whenever East Asia got active, and if long range models are correct this active period could be quite often. If the tightened jet stream in the North Pacific and stormy East Asia evolves as predicted (combined with the possible +PNA pattern and weakened polar vortex), this winter may help those winter-loving souls damaged by the last couple of winters.

A side-by-side analysis reveals truly stunning similarities- even I did not expect to find such a strong connection. The winter of 1962-1963 and the forecast models have a whole list of similarities. Both indicate low pressure in the eastern portion of North America. The analog year sees this low pressure area as more expansive, but both the analog and forecast model center the low pressure near the Northeast US and Southeast Canada. Both models (and this is the connection that surprised me the most) also have high pressure building from the West Coast into the Bering Sea. But it gets even better than that- the location of high pressure from the West Coast on the forecast model is almost identical to where the high pressure anomalies begin in the analog winter. They both strengthen the high pressure in the Bering Sea, and continue this high pressure through Siberia and northern Eurasia. If that isn't a correlation, I don't know what is. Both items have low pressure in Eastern Asia and into the Pacific. The forecast model is definitely weaker in that aspect, but such small differences like that are to be expected. The general idea is still firm in both articles: lower pressure and a stronger jet stream in East Asia. 
It is also worth noting the differences we have here, as some of them must be addressed in this comparison. The biggest difference is the lack of high pressure over Greenland in the forecast model. I should mention that other forecast models not shown here do have prevalent high pressure over Greenland, but for all intents and purposes we will keep that difference in our back pocket to analyze in a later forecast. The other difference I see at this time is the East Asian pressure difference, and we already went over that in the above paragraph. 

Global models generally favor a cooler winter for the North Plains, Great Lakes, Ohio Valley, Upper Midwest, Northeast and generally places in the top-right quadrant of the United States (if you were to cut the nation into four pieces). Because we're a long time out from the winter, these anomalies are not that significant. As we get closer to December we will start to see these anomalies become more extreme, with the below-normal temperatures getting colder and above-normal temperatures raising their anomaly projection. In the precipitation department, a very La Nina-type set-up appears to be favored, with a stripe of above-normal precipitation from Oklahoma to the U.P. of Michigan. I'm rather skeptical of precipitation forecasts this far out and feel it's best to stick to the analogs and upper-level height anomaly charts we were looking at earlier in this post. Regardless of its accuracy, the premise of a stormy Midwest/Plains and chilly East US can raise excitement for some winter weather lovers.


My analog years are matching up unusually well with long-range forecast models. While these models will change, I find it to be a good sign that there is a rather substantial amount of correlation this far out. My analogs may be revised down the road as we enter summer and early fall, and it is assured that the long range models will change. However, these are my long range estimates for this winter.

-The polar vortex may be at lower latitudes this winter.
-A wetter winter is possible for portions of the Plains and Midwest.
-Cooler weather is possible for portions of the northern US.