Thursday, October 24, 2013

Final 2013-2014 Winter Forecast

"Slow start to winter should deliver harsh January, February for Central US..."

Hello everyone, this is the Final 2013-2014 Winter Forecast from The Weather Centre. This post will finalize my projections for this winter, with 3 month-averaged temperature, precipitation and snowfall graphics. Month by month descriptions will be written below each graphic. If you do not wish to see the discussion, you may scroll down to the graphics to see the forecast itself.

There are a lot of factors to look at, and we're going to start out with the El Nino-Southern Oscillation.

The El Nino Southern Oscillation (or ENSO) phenomenon involves anomalous sea surface temperatures across these four regions on the chart above. When sea surface temperatures are below normal in this graph, it is said that a La Nina is in place. In similar fashion, above normal sea surface temperatures (SST's) in the Nino 3.4 region correspond to the term 'El Nino'. Both the La Nina and El Nino have different effects on the United States- an El Nino brings about a snowy and cold East Coast, while allowing warm and dry weather to flourish in the Plains and Midwest. A La Nina permits cool and snowy weather to hit the Ohio Valley and Midwest, while warming up portions of the Southern US. Looking at the chart above, from TropicalTidbits, we see a neutral ENSO, meaning the water temperatures are not cold enough to be declared a La Nina, and are not warm enough to be called an El Nino. Because the ENSO is in a neutral state, we are once again going to see other atmospheric patterns affect our winter this year, like last year, because the ENSO is not strong enough to have a pertinent influence on our weather.

An animation of subsurface water temperatures under the ENSO regions shows disagreement over what phase the oscillation is currently in. This confusion confirms the idea that we are in a neutral ENSO state, and a lack of any agreement over temperature anomalies underwater suggests the rest of the fall (and most likely throughout the winter) will also see this neutral ENSO condition.

We will now move on to something called the Quasi-Biennial Oscillation, or QBO for short.

This phase space, drawn up by Mike Ventrice, illustrates the strength and phase of the Quasi-Biennial Oscillation throughout the past several months. As the chart shows, the QBO has been slowly traversing the middle phases of this diagram, to where we are now entering Phase 6 at modest strength. Extrapolation and a glimpse at the current QBO status through other diagrams tells me the QBO will be able to move through Phase 6 in the early part of winter, before transitioning to Phase 7 in time for the January-February part of the winter.

Each QBO phase results in different atmospheric effects during the winter months. The diagram above (also made by Mike Ventrice) shows 500 millibar height anomalies for the November-March period for each QBO phase. If we take a look at Phase 6 on the right-hand side of this image, we see that ridging across the Gulf of Alaska and troughing in the waters just west of the Bering Sea is favored. This sort of pattern can lead to lower pressures in the Rockies, which can then lead to ridging in the East US. However, the ridging in the Gulf of Alaska appears to be close enough to the West Coast so that the East US ridging is suppressed. It is not completely suppressed, however, as deep negative height anomalies are shown across Greenland. This Pacific pattern and strong +NAO indicator (the troughing over Greenland) tells me we will see a pattern that supports a warm-ish pattern in the East US, with some stormy weather across the eastern Rockies/Plains. Bear in mind that the strongest warmth ought to be kept down in the Southeast and maybe portions of the Mid-Atlantic, but I'm not seeing a full-on torch for December.

When the QBO hits Phase 7, you can see that the situation dramatically changes. Height anomalies in Phase 7 do a complete reversal and give way to a quite favorable winter pattern. Extreme ridging is spotted across the Arctic Circle, and this gives an indication that the chance for those stratospheric warming events are enhanced for January and February, when the QBO does enter Phase 7. At the same time, ridging is also observed across the western coast of North America, which is a key component to a colder pattern for the Central and East US. While I don't have temperature charts for each phase of the QBO, I would expect solid below normal temperature anomalies for the Central and East US during Phase 7. If we go solely by the QBO, I would be calling for a warm early half of the US and -- dare I say it -- harsh second half of winter.

Another factor we're going to watch for this winter is how the solar cycle plays out. We are entering a time period in the next decade or so that will see the sun go into a hibernation, with sunspot levels plummeting to levels we have not seen since the Maunder Minimum. As we enter this upcoming winter, the sun looks like it is starting that downhill trend. The chart above shows the forecasted sunspot cycle, but as the observed sunspot numbers tell, that forecast is not going to verify. The latest sunspot values for September really dropped off, per the most recent dot on the chart above. Upon some research I did a few months back, there is a positive correlation between spiking or plummeting sunspot values, and temperatures a couple months later. This was best shown by the intense sunspot spike we saw prior to the winter of 2011-2012. It's no secret that the winter ended up warmer than normal. In the same context, we saw a spike in sunspot numbers just three or four months into this year, which led to a large heat wave in June into July. Now, September sunspot numbers have dropped off a cliff. Using the research I did on this, we should see a cooler trend in time for December and January. I did express my concern over the warmth potential in December with the QBO, and those concerns are still valid, but we could see this warmth tempered if the sunspot correlation ends up verifying this winter.

An additional benefactor of the low solar activity is the stratosphere. When the solar cycle is at low levels, ozone is able to build up in the upper layers of the atmosphere, as ozone-destroying energy (i.e. certain solar rays and radiation in general) is tempered by the anomalously quiet sun. This helps the warming process in the stratosphere, which then not only weakens the almighty polar vortex, but theoretically raises the chances of stratospheric warming events as well. With the sun quickly dropping off, I have a feeling this ozone build-up will come into play, but it may take the QBO switching phases in Phase 7 before the stratosphere starts experiencing some turbulence in terms of the state of the polar vortex. This leads into the next (and my favorite) topic, the stratospheric implications for this winter.

This is a 240 hour forecast of what is referred to as the lowest part of the stratosphere, the 100 millibar level. 240 hours (or 10 days) from now, we see that the forecast calls for three very distinct things.

I. Intense warming over the Bering Sea.
This has massive implications for this winter, folks. I cannot stress enough what this would mean for the winter season if this part of the stratosphere becomes entrenched in the weather pattern. It is known that warming in the stratosphere leads to lower pressure and cold weather, which is why winter weather people love sudden stratospheric warming events. At the same time, colder areas of the stratosphere indicate a preference for high pressure and warmth, and we will touch on this particular aspect in just a paragraph or two. But for now, focus your attention on the Bering Sea. That intense warming, should it hold its ground in the weather pattern from now through winter, would help stormy prospects in the Bering Sea. With the use of Joe Renken's Bering Sea Rule, which illustrates a correlation between strong storm systems in the Bering Sea and a cooler/stormier weather pattern in the Central/East US roughly 2.5-3 weeks later, the stratosphere may be able to give a helping hand to the chances for some stormy weather this winter, as long as the warmth in the Bering Sea can translate to stormy weather in that body of water. A caveat for this scenario involves that the stratospheric temperature pattern changes and we see the warm Bering Sea temperatures dissipate, which is a real possibility. However, with this warmth in the Bering Sea staying put in the next 10 days (and possibly beyond), it looks to me like this pattern isn't really going to change in the relatively-long range, and that's good news for winter weather folk.

II. Warmth in the Central/East US
The warmth over the Central and East US will also be here to stay for the next 10 (or more) days, and this should contnue the cooler weather pattern we've been having. I'll discuss how this cold weather will affect this winter in the LRC portion of the winter forecast, but for now we'll stick to the stratosphere. If the temperature pattern in the lower stratosphere remains the same throughout the rest of the fall and into winter (which I'm cautiously optimistic it will), expect the brunt of the cold to really hit the Plains, Midwest and parts of the Rockies and Northeast. On a side note, the lack of warmth in the Southeast US may be helping to notify us of what could be ridging in the Southeast this winter, which would favor a storm track through the Midwest and Ohio Valley, but that's speculation at this point.

III. Cool temperatures over Greenland
Again, cool temperatures in the stratosphere can help prospects for ridging, and this is no different. The forecast for the lower stratosphere suggests we will see a gradual cooling trend over the Greenland area, which, in due time, should lead to ridging over the Arctic land mass. Should this cool-down continue, expect the chances of a negative NAO to rise for this winter.

Finally, bearing in mind the cold weather stratospheric temperature trend that is able to produce ridging, take a look at the 10 millibar (upper stratosphere) forecast below.

That intense cold air mass over the Arctic Circle will be intensifying from now until the 10 day forecast, which is shown above. If this stays into winter, expect the chances of not only a negative NAO, but a negative AO to skyrocket. It remains to be seen if this cold weather will translate to the surface, but for right now, I am encouraged by this development. Further helping my opinion on the stratosphere is the new stratospheric warming event that has just started in Asia, mainly due to persistent positive mountain torque values across Asia. The 10mb animation of stratospheric temperature anomalies is shown below:

While I am skeptical of this warming event having extreme effects on the winter, history points to this coming winter featuring a stratosphere that, at the very least, should not end up anomalously cold. If this stratospheric warming event is to have a big effect on this winter, it should keep the stratosphere in a relatively warm state throughout the winter, possibly in spite of the positive QBO. Take a look below:

These three charts show temperature anomalies from the years of 2005 (top left), 2006 (top right), and the current year of 2013 (bottom). Take a look at the bottom chart. If you look closely, you can see a sharp cut-off of the slight below normal temperature anomalies, and an emergence of slightly above normal temperature anomalies in the stratosphere. I took a look back in history to find years that also featured warming events in the stratosphere in the fall, and the years of 2005 and 2006 were included in this. Glancing over 2005, I circled the warming event in black. Following this warming, the stratosphere did cool for a while, but when December rolled around, the polar vortex never really got going, mainly due to the multiple stratospheric warming events that occurred in the end of the year. Could this be a side effect of that fall stratospheric warming? I'm not in a position to say yes or no, but the possibility to consider is yes, this warming event did set the stage for December to feature a relatively warm stratosphere. It gets better than this- taking a look at 2006 on the top right, the stratosphere was bombarded with at least two big stratospheric warming events throughout January and February. Again, I cannot say if the fall stratospheric warming event influenced these late winter massive warming events out of uncertainty, but I would not rule out the idea that it did. In 2006, this same scenario played out again, with a fall stratospheric warming event leading to a rather warm stratosphere for the end of the year. A look at early 2007 shows that the stratosphere was not nearly as excited with warming events as it was in early 2006, but the January-February 2007 period still featured as many as 4 separate warming events, two of which were decent warming events in terms of how much of the stratosphere was affected. It should be noted that the winter of 2005-2006 had a solid negative QBO, which enabled the polar vortex and stratosphere to naturally be open to stratospheric warming events, but the 2006-2007 QBO is very similar to the QBO we are currently experiencing, and this historical connection bodes well for the idea that while the +QBO should hamper efforts by the stratosphere to enact sudden stratospheric warming (SSW) events (especially in the first half of winter), the stratosphere is likely to be hit with at least one stratospheric warming event, per the factors I have covered thus far.

It's now time to analyze the next piece in the winter forecast puzzle- the SAI.

Next, let's take a gander at a pretty special tool that can be used for forecasting the winter Arctic Oscillation, titled the Snow Advance Index (SAI). The SAI involves monitoring the snow cover anomaly over Siberia during the month of October. In short, if the snow cover anomaly maintains an above normal stance for the month of October, the Arctic Oscillation should be watched for a negative state this winter. On the opposing end, below normal snow cover in Eurasia tends to lead to a positive Arctic Oscillation in the succeeding winter. Per the graph above, which shows northern hemisphere snow cover anomalies from November 2012 to the present day, we see a spike in snow cover anomaly in late September, which has led into strong above normal snow cover anomalies in October... that is, until October 19th. If you look closely, you may be able to see that the snow cover anomalies tanked just around that date, and this is shown by a quick drop on the graphical depiction on top, as well as a drop-off from blue to a tinge of red on the bottom depiction of snow cover anomalies. Medium range snowfall forecasts insist that the snow cover will be built back up in due time, but I am a little concerned that high pressure wavering around Siberia may hurt the snow cover. I expect the snow cover to build back up in the face of this drop-off, so do not be too concerned right now. Overall, the SAI does currently favor a negative Arctic Oscillation for this winter.

I have taken into account other things, such as the Pacific Decadal Oscillation, and the Atlantic Multidecadal Oscillation. They will not be covered in this forecast, as the PDO, AMO status has not changed since my Official winter forecast.

The last item left to discuss is the Lezak Recurring Cycle. The Lezak Recurring Cycle, or LRC, is a tool developed by meteorologist Gary Lezak that, in essence, can enable forecasters to predict the overall weather pattern months in advance. The gist of the LRC involves a cycling weather pattern that develops in October and November of each year; no pattern is the same from year to year. Around mid November, the LRC begins to repeat, meaning we start to see a similar weather pattern in mid November that we saw in early October. This means that the cycling pattern has begun, and it will continue to cycle on a 40-60 day interval for the next ~10 months before it dissipates over the following summer.

The first item to address is the October 4-6 winter storm, which delivered close to 5 feet of snow in a few isolated spots across western South Dakota. This storm system trekked northeast after strengthening over the Plains and dumping all of its snow. In the wake of this storm, a cold front meandered south through the southern Plains, Midwest and the Eastern US. Just on a very preliminary look at the winter storm, and the environment surrounding it, I would not be surprised to see this storm stay in the Plains and Midwest through the next couple of LRC cycles through this winter. To show why, take a look at the 500mb map for October 5th:
You can see the storm system in the Plains as a closed low, signified by the closed contour lines. There's a reason it went northeast rather than east- there was a ridge of high pressure in the Southeast. This ridge means it really can't just dive south or go due east unless the ridge moved, and in this case, it didn't. That said, considering the LRC should maintain a similar atmospheric pattern through its winter and spring cycles in the upcoming year, the storm system should go northeast again when it re-cycles in mid-November. I don't see this storm system being one that will become a big snow producer for the East, given the position of the ridge in the Southeast, but I would also be surprised to see extreme snowfall amounts like the ones seen in early October from this storm. Depending on how the jet stream changes from now until the next cycle in mid-November, the storm track may change slightly, but the overall pattern should remain similar enough to keep the storm track at least relatively on the same track as we saw in early October. With another storm system about to move into the Plains again, this region is looking at multiple snow and cold chances this winter.

The next piece of the LRC puzzle is the new weather pattern that has developed as of mid-October, and continuing now and into the next week or two. As the 500mb height anomaly chart on the left shows, this new pattern involves strong ridging over western North America into the Gulf of Alaska. This ridging provokes deep troughing in the Central and East US, and this is intensified by the favorable negative North Atlantic Oscillation (NAO) over Greenland. There is little doubt in my mind that we see this pattern cycle again for late November/early December through mid/late December, and again into early February, possibly through a majority of that month. The point is, this cold snap should last a good ~15-20 days when it actually ends, with modified cooler air possibly lasting beyond that ~15-20 day point. This means the LRC, if all goes as is expected, will allow the Central US to experience some frigid weather this winter as a direct result of the LRC. If you live near the Great Lakes, especially Michigan, Huron and Superior, get ready for what could be a big lake effect snow event when this part of the LRC repeats again.

Taking into account all of that information, and then some, here is the Final 2013-2014 Winter Forecast.

Confidence: High
I'm anticipating a warm start to winter, with that idea coming from Phase 6 of the positive QBO as well as the LRC, which should see the warmth observed in early October repeating in mid-November into early December. This warmth should be held up throughout the East US, and into the Great Lakes and Midwest. Expect the northern Plains to escape the worst of the warmth.

As the QBO transitions to Phase 7, and the LRC goes into its next cycle, the temperature trend ought to be on a decisive colder path. The Plains region into the Midwest and Great Lakes should see the brunt of the cold in January, with slight ridging in the Southeast offering the opportunity for warmer than normal temperatures in that area. Depending on if the negative Pacific Decadal Oscillation is still present by the first month of 2014, this ridging in the Southeast could be enhanced, and the overall storm track would be affected. Prospects of a negative Arctic Oscillation for this winter, as well as the heightened chance of stratospheric warming events in the latter half of the season should see a colder trend in much of the nation.

More of the same from January is a good summary of what the nation can expect in terms of temperatures. The stratosphere should be in a decent amount of turmoil for February, as long as the current stratospheric temperature trends and changing QBO phase cooperate. Once again, eyes will be on the Plains, Midwest and parts of the Northeast and Great Lakes for the brunt of the cold for the last month of traditional winter.

Confidence: Medium
*Due to low confidence, this discussion will be grouped into the three month December-January-February period.

Signs are pointing to ridging in the Southeast this winter, and that should enable a rather dry set-up to be provoked in that region of the nation. The ridging in the Southeast, aided by what should be a stormy Bering Sea, ought to outline a swath of above normal precipitation that could stretch from the Oklahoma panhandle to western New York. This would all depend on the eventual storm track for this winter, which has yet to be determined due to a few select atmospheric factors that I will need more time to monitor before making decisions that pertain to the storm track. Confidence goes down for both coastal regions, as it is (decently) possible we see a stormy East Coast and possible we see a dry West Coast. Again, I will need more time to decide what the forecast will be for those two regions, but if I had to choose, I would expect a rather wet trend for the East Coast and maybe a drier trend along the West Coast (but that is in question due to the potential storminess of the Bering Sea and how it could affect high pressure prospects in the Pacific Northwest).

Confidence: Medium to Medium-High
*Due to low confidence in conjunction with the precipitation forecast, and because the snowfall pattern will follow the storm track (which is somewhat TBD), the snowfall description will be grouped into a three month December-January-February description.

The lake effect snow machines will be in overdrive this winter, as the Lezak Recurring Cycle and end-half of winter pattern allows intense lake effect snow events to develop. Lakes Michigan, Superior and Huron will be affected the greatest, so long as the current outlooks for the core of the cold to stay in that region locks in place. Lakes Ontario and Erie should not see as intense of lake effect snow outbreaks, but the coastlines of those lakes will still experience some hefty lake effect snow events. As far as synoptic snowfall projections, current trends highlight the Plains, Midwest and western Ohio Valley for the best chances of above normal snowfall. The Southeast Ridge, stratospheric temperature trends and LRC ought to see this happen, but I am not confident enough to say it will happen, purely because of the 'little things', in terms of mesoscale teleconnection patterns that have yet to be determined for this winter. The East Coast will have the opportunity to see some hefty snow events this winter, and if you really want to, you could envision an above-normal snowfall outline across the coastal regions of the East Coast in addition to the Plains/Midwest. I may have to adjust the southern Ohio Valley's snowfall projection in the event my idea on the storm track changes, and the same may be said for the northern Rockies. Any adjustments will most likely be made in mid-November, when I will re-examine the forecast. Do not expect a whole other winter forecast to be issued at that point; if any changes are made, it will just be a single, moderately-long post.

Here is the overall graphic for the winter of 2013-2014, subject to slight changes.

Well, there you have it, folks. This concludes The Weather Centre's Final 2013-2014 Winter Forecast, and thus concludes the release of winter forecasts for this year. As I see fit, adjustments will be made and the changes will be published, but the Final winter forecast usually heralds the end of the publishing of major winter forecasts.

If you have any questions, don't hesitate to comment below, comment on our Facebook page, or tweet us.


This winter forecast took into account data produced by: Joe Renken, Eric Webb, Steven DiMartino, Michael Ventrice, and Josh Herman.