The Story Behind The Polar Vortex

Ever since last winter, the polar vortex has taken over as the go-to thought as soon as the prospect of cold weather is brought up. Let's take this time today to go over what the polar vortex actually is, and disprove some inaccuracies I've seen swirling around the world.

1. The Polar Vortex Exists Year-Round
The phrase 'polar vortex' is used to describe a strong low pressure system (vortex) that is semipermanently placed over the Arctic region (polar). In the summer, the vortex significantly weakens, and is replaced by high pressure. During the fall months, however, the vortex gains strength yet again and low pressure dominates the Arctic as the polar vortex gears up for winter.

2. The Polar Vortex Extends into the Stratosphere, Not Just the Troposphere

NCEP

The image above shows the 10-day forecast from the GFS model in the stratosphere, at the 30 millibar level. That's pretty high up the sky when you consider us humans are at the 1000 millibar level. In this forecast, we see relatively weak low pressure establishing itself over the Arctic Circle, with weak high pressure surrounding it. That low pressure body is the "birth" of the polar vortex for this cold season, as it begins to come back to life and strengthen for the winter.

Some people assume the polar vortex is no different than your average storm system that brings snow to your neighborhood. The vortex is actually quite different. In order for it to be the polar vortex, it must exist at both the tropospheric and stratospheric levels. That's how big and controlling it is in the atmosphere.

3. There is only ONE Polar Vortex

This is something I've been hearing ever since January 2014, when this polar vortex business really took off. The belief is that there is more than one polar vortex in the atmosphere. This might stem from the aforementioned idea that some believe the polar vortex is no different than your average storm system, but regardless of its origins, the belief is false.

ESRL

Take a look at the image above. In this image, we see a view of 500mb height anomalies across the Northern Hemisphere on January 7th, 2014. Warm colors correspond to high pressure, which brings warmth and generally quiet conditions. Blues and purples indicate the presence of low pressure, which permits stormy and chilly weather. Just taking a glance over this map, we see one strong upper level low that seems to be the strongest one in the entire hemisphere. That upper level low, seen scraping the northern US and inundating Canada, is the tropospheric version of the polar vortex. As you can see, there is only one low as strong as the one in North America, hence there is only one polar vortex. Those other low pressure systems are just disturbances compared to the polar vortex.

ESRL

To give you a better idea of how there's only one polar vortex, let's take a look at the stratosphere on January 7th, 2014. In this image, we see what appears to be two strong areas of negative height anomalies, being pinched by two bodies of ridging/high pressure on either side of the globe. If you thought that this image shows two polar vortexes, you are incorrect!
What's actually happening is the two bodies of high pressure are trying to tear apart the polar vortex, something that happens rather often each year. The difference between other years and this past year is that the ridging forced the main part of the polar vortex into North America; that other area of negative height anomalies in Asia is just a weaker part of the single polar vortex that was torn off of the main body. Some of you more experienced weather enthusiasts may know this type of set-up as Wave-2 stratospheric activity, as two bodies of high pressure push into the Arctic.

To close out this post, here's a stellar graphic made by the folks at the National Weather Service office in New York in early January, when the hype over the polar vortex was in full swing.

NWS

To summarize:

• The Polar Vortex exists year-round, varying in strength. It does not (and will not) "come back" for the winter, as it's already present and cannot "come back".

• The Polar Vortex exists across both the troposphere (where we live) and the stratosphere (where Felix Baumgartner jumped from his capsule in October 2012); it does not exist at the surface.

• There is only one polar vortex; our world would be much different if more than one polar vortex existed.

Additional points of summary are included in the NWS graphic above.

Andrew

2014-2015 Winter Forecast Update; Colder Outlook Now Favored

This post will examine the factors being analyzed for the upcoming winter, their impacts on the cold season, and anything that may have changed from the issuance of the Preliminary 2014-2015 Winter Forecast to today.

After going over the factors presented in our Preliminary winter forecast, and examining new data presented to me, I’m electing to favor a colder outlook to this winter, over the original warmer projection earlier this summer.

ESRL

The above image shows observed sea surface temperature anomalies across the globe from July 1, 2014, to July 8, 2014. In this image, we notice three particularly interesting features. One of these features is the pool of well above normal water temperatures in the Northeast Pacific into the Gulf of Alaska. This particular mechanism is one of substantial importance, given it also played a significant role in our last winter.

In the winter of 2013-2014, we had this warm pool present throughout the entire season, and it was quite possibly the most influential factor in making that winter as cold as it was. The warm water enabled strong high pressure to form over the Northeast Pacific, something that is continuing even to this day. See the examples below.

ESRL

The image above shows mid-level geopotential height anomalies in early January 2014. Cold colors indicate the presence of strong low pressure, hence below normal temperatures. The yellows and reds indicate high pressure, and warmer than normal weather. This image was recorded in the middle of the extreme cold spell that kicked off the new year, also spawning the ‘polar vortex’ hype. In this graphic, we can see the polar vortex located somewhere in south-central Canada, bringing that extreme cold to much of the United States. We also saw strong high pressure centered over Nova Scotia, and quite extensive high pressure from the Southwest United States to Alaska. It was this second body of high pressure that formed as a result of those aforementioned warm waters in the northeast Pacific, and we saw this ridge stay in place consistently throughout the winter. Now, with the warm waters in the Gulf of Alaska still in place a good year later, I am significantly concerned with the idea that we could see this extensive ridge forming yet again this coming winter.

Building off of this idea, in addition to the water temperature pattern in the Pacific going relatively unchanged since last winter, the upper air flow has remained relatively unchanged as a result.

Tropical Tidbits

This forecast image, valid on Monday, July 14, shows the same 500mb geopotential height concept that we saw earlier in this post, with blues indicating low pressure and reds indicating high pressure. In this forecast image, once again, we see strong high pressure stretching along the western coast of North America, with another body of high pressure being observed over Nova Scotia and far eastern Canada. And in a situation just like the polar vortex plunge last winter, we see anomalous low pressure crashing into the United States. This upper air pattern is almost (keyword almost) exactly like the one we saw last winter, bringing into question if we could see this pattern persist into next winter as well. In my opinion, if the sea surface temperature anomalies remain the same into next winter as they are now, I see no reason to discount the idea that the coming winter could be like last winter; a cold, potentially brutal one.

ESRL

Adding to the pieces of this puzzle is the body of significantly above normal temperature anomalies near Greenland. This is a change in what we saw from last winter. This past cold season, we observed primarily below-normal water temperatures near Greenland, resulting in some very stormy periods in the northern Atlantic. However, now that we’ve entered July, we now find ourselves with a formidable body of positive water temperature anomalies, something likely to change the game for the coming winter.

When warmer than normal water temperatures exist near Greenland, it provides a similar opportunity to that in the northeast Pacific, where high pressure can form. In this instance, high pressure can form over Greenland, and actually force the jet stream to buckle in the East US, resulting in an influx of colder than normal air to that area, as the image below shows.


NWS

 We didn’t have that warm water in place last year, meaning such a sequence of events couldn’t happen often, if at all. However, now that the anomalies have reversed, we’re in a much better position to see the jet stream buckle more frequently over the East US, resulting in another cold winter.

On a side note, we also see persistent above normal water temperatures near Nova Scotia, which could provide the mechanism needed to create high pressure in that area. This could then combine with the warm SSTs in the northeast Pacific to create an environment very similar to the one observed with the polar vortex incident.


DMI

We now continue on with a look at temperatures in the Arctic Circle. The graph above shows observed temperatures in Kelvin over the Arctic Circle (colored red), from the first day of 2014 to present day. We also see a green line, indicating the average Arctic Circle temperature for that given day, allowing us to compare with the observed temperatures to detect any anomalies. The consistent blue line highlights the freezing temperature mark. On this image, we see Arctic temperatures have been steadily below normal this summer, something that does not bode well for us in North America.

In order for winter to even be cold to begin with, one needs cold air. It seems simple to understand, but now we need to understand where this cold air will come from. By using this graph above, we can get an early glimpse at how much cold air the upper latitudes may be storing up for this winter. In this instance, below-normal Arctic temperatures in the summer might tell us to prepare for a cooler than normal winter ahead, while a warm Arctic summer could indicate a warmer winter ahead. Going by this guideline, we might be best off preparing for a chilly winter.

Now, last but definitely not least, we will go over the evolving situation in the Equatorial Pacific.

CPC

The image above shows equatorial upper-ocean heat anomalies, in Celsius. This means that the temperature anomaly for the upper portion of the Equatorial Pacific is calculated, giving a helpful indicator to the state of the El Nino-Southern Oscillation (ENSO) phenomenon. In this case, positive anomalies indicate the presence of warmer than normal waters, and potentially an El Nino. Similarly, below normal upper ocean heat anomalies highlight below normal water temperatures, potentially also the presence of a La Nina.

Looking at this graph over the past year, particularly since February 2014, we see that the upper ocean heat anomalies have skyrocketed, up to abut 2.0º Celsius above normal in late March 2014. It was at this time that we were expecting a strong El Nino to evolve for the coming winter. Things have certainly changed since then. The warm waters began to hit the surface this past spring, but couldn’t hold their ground, per se. As a result, we have seen the positive anomalies drop off to neutral territory, indicating that any possible El Nino no longer theoretically exists. Time will tell if we see another body of warmer than normal water temperatures pushing to the surface, possibly creating an El Nino. If that does not happen, we can expect the winter pattern to be dominated by more mesoscale (smaller-scale) features, such as the bodies of warm water temperatures in the northeast Pacific and around Greenland. This could then lead us into - you guessed it - a cold winter.

All of this is expected to change as we progress through summer and into fall; this is merely an update to the situation, since many things have changed since the initial issuance of our Preliminary 2014-2015 Winter Forecast.

Andrew

Upcoming Winter Shaping Up Similar to Last Winter; 2014-2015 Could Be More Severe

Note: The idea of this winter being more severe than last is a preliminary finding based on the work presented in this post. It is by no means set in stone, and is subject to revision.

After a look back at July 2013, I'm finding that July 2014 sea surface temperature anomalies are not only similar to last July, but could be foreshadowing a more severe winter than last.

ESRL

Shown above is a reanalysis image of sea surface temperature anomalies from July 1, 2013 to July 8, 2013. During that timeframe, we saw a large body of warmer than normal waters controlling the northeast Pacific into the Gulf of Alaska, even stretching down towards the coastal waters near southern California. We also saw a pocket of well above normal water temperatures west of Japan, with opposing cooler than normal waters to the east. We also saw the La Nina-like water pattern along the Equatorial Pacific, which helped keep us in a Neutral-ENSO winter for this past cold season. Transitioning to the Atlantic, we saw a snaking line of below normal SST anomalies, occasionally interrupted by small bodies of positive water temperature anomalies, which were mainly confined to the northwestern Atlantic.

Now, let's compare all of that to what we've seen so far this year.

ESRL

The image above now shows sea surface temperature anomalies, still on the same intervals, but now valid from July 1, 2014 to July 8, 2014. Looking to the northeast Pacific, we see that the large body of above normal-temperature waters has not only persisted, but has now intensified and expanded its influence. This was a major factor in last winter's atmospheric flow, and I don't expect things to be any different this winter. We now see the Bering Sea flooded in warmer than normal waters, though just what impact this particular feature may have is unclear right now. We see a similar opposing water temperature pattern to the west and east of Japan, though a substantial change in the Equatorial Pacific is the presence of warmer than normal water temperatures as opposed to the below normal temperatures last year at this time. This would typically indicate the presence of an El Nino, but with uncertainties as to if it will sustain itself, and if the atmosphere will actually "recognize" the presence of an El Nino, some things still need to be sorted out.

Now that we've gone over the similarities, why am I thinking that this winter could be more severe than last?

1. Above Normal Water Temperatures in the Northeast Pacific
Last winter, we experienced extreme bouts of cold, many of which were due to the proximity of the United States to the polar vortex, a massive, semi-permanent low pressure system stationed over the Arctic Circle. The polar vortex was persuaded to dip south into North America, thanks in large part to a massive ridge that constantly formed and re-formed along the west coast of the continent. Because the jet stream went up with the ridge pushing northward, it had to also dip down somewhere, and that somewhere became the United States. The big ridge originated from the body of warm waters in the northeast Pacific last winter, which is why I'm so concerned seeing that mechanism still in place today. If those warm waters stay in place in the northeast Pacific into the winter, they could make that big ridge re-appear, possibly stronger than last winter, leading to an even colder winter this year.

2. Above Normal Water Temperatures near Greenland
Something we didn't have last winter was a swath of above normal water temperatures near Greenland. This prevented high pressure from forming over that area, something that would have raised the chances of an even colder winter than what we ended up seeing. This year, as the July 2014 SST image shows, we do now have a large body of above normal SST anomalies near Greenland. This significantly raises the potential of high pressure forming in that area, buckling the jet stream to the west and resulting in an even colder winter possible. This is just another mechanism I'm watching this winter that could cool down the temperature forecasts even further.

Whether all of this ends up happening, or doesn't happen at all, is yet to be seen. The point is, based on sea surface temperature anomalies, this year is looking a lot like last year at this time. And according to preliminary looks at some key SST anomaly features, things could be more rough than last winter.

Andrew

Unseasonable Cold Blast Looming; Pattern Similar to Last Winter

An unseasonably cold airmass looks to wash over the northern United States, in a pattern eerily similar to the one seen this past winter.

CPC

The Climate Prediction Center's 6-10 day temperature outlook shows significantly above normal temperatures across the West US, most severe over Washington, Oregon, Nevada, Idaho and California. In response to this warmth, we see a deep airmass of unseasonably cool temperatures pushing south across the Midwest, with states like Minnesota, Wisconsin, Illinois, Iowa, South Dakota, Nebraska, Kansas and Missouri all affected on the highest level. Warmer than normal weather looks to retreat to the Gulf Coast and coastal regions along the Eastern Seaboard, but the main story here is indeed the colder than normal weather.

Drag me to storm models

Shown above is a long range forecast of temperatures on the morning of July 15th, in the middle of this unusually cold spell. We see temperatures on this morning plummeting to as low as the mid-40s in the Midwest, where the heavy blankets might need to make a surprise appearance. Temperatures in the far northern Plains into the upper Midwest might even flirt with the low-40s, possibly even into the upper-30s if there will be clear skies. Those finer details will need to be ironed out in days to come, but the general idea is that things are looking pretty cold for a wide swath of the country in the next week or two.

PSU

Oddly enough, the atmospheric pattern behind this expected cold blast is quite similar to the pattern we observed this past winter. On the top-left image, we see the mid-level atmospheric flow valid on July 14th. Here, we can see a strong vortex dropping anomalously south from Canada, nearly pushing into the United States. If you recall, we had the polar vortex take a very similar path down south more than once last winter, which is how the weather got so cold so often. So what's provoking this to happen again, only this time in mid-July? The same thing that made it happen six months ago. We see a very strong ridge pushing north across the northeast Pacific and into the Gulf of Alaska, which is how the West US should end up with those much warmer than normal temperatures. And, bringing things back full-circle, that ridge is likely being caused/enhanced by the body of above-normal water temperatures in the Gulf of Alaska that we targeted as the mechanism responsible for the brutality of last year's winter. The latest water temperature anomaly image is shown below, which identifies the body of much warmer than normal water in the northeast Pacific.

ESRL

Andrew

April 16-17 Potentially Significant Snowstorm

I'm monitoring the potential for a significant snowstorm on April 16-17.

We begin by looking at the GFS model. The GFS model shows a strong storm system dropping down from Canada, taking advantage of the cold air already displaced south in the United States and producing an intense snowstorm in the Upper Midwest. The GFS indicates we may see amounts over 12 inches across northern Wisconsin and upper peninsula of Michigan, with spotty amounts nearing or exceeding 18 inches possible. Given model guidance tendencies to be too snowy/too cold in their forecasts, as well as the fact that models may be overdoing snow because they're stuck in a wintry pattern when we're really transitioning to spring (though the snow falling out my window begs to differ), I'm skeptical of these 18 inch-plus amounts. However, if the storm is as strong as projected, a big snowstorm would be expected. I'm just skeptical of temperature profiles right now.

The NAM model, a short-range, higher-resolution sibling of the GFS model, is actually showing similar amounts. I'm expressing surprise here because the NAM is notorious for over-doing snowfall guidance, in that it can (and frequently does) forecast more snow to fall than what actually does fall. However, comparing the GFS and NAM models, they are pretty similar. While this raises confidence in this outlook, it's nothing to get too excited about. I'm still very skeptical of this event, and I want to emphasize how this has a high 'bust potential' (meaning the forecast has the potential to 'bust', or fail). However, if the forecast ends up to be on target, then I am on board for a big snowstorm. With spring, model guidance starts to struggle due to the pattern changing, and that messes up model guidance. This could be a flash in the pan, or something more.

Shown below is a hand-drawn image of amounts as depicted by the GFS and NAM, for easier interpretation.

To summarize:
• There is the potential for a significant snowstorm in the Upper Midwest on April 16-17.
• Amounts may exceed 12 inches in many places.
• I am skeptical of this forecast, as model guidance may be too cold and may be forecasting too much snow to fall.

Andrew

April 14-15 Potentially Significant Snowstorm

Model guidance is suddenly converging on a potentially significant late-season snowstorm.

The image above shows the most recent forecast from the GFS model, highlighting something I discussed the other day: a potential late-season snowstorm that may drop significant amounts of snow. We had seen the ECMWF model dropping similarly high amounts of snow, as the hand-drawn graphic below shows. Remember, this was yesterday's ECMWF MODEL forecast, NOT my forecast.

Yesterday's ECMWF model snowfall map

While these late-season snowstorms are typically weaker, due to warmer soil temperatures that reduce sticking snow, as well as a genrally unfavorable environment, the Gulf of Mexico can provide a basis for intense amounts of moisture and fuel. If there's enough cold air in place, a strong snowstorm can result.

Tropical Tidbits

There is some merit to the idea that we may see a strong storm system, as the ECMWF is suggesting. Shown above was the observed 500mb height anomaly chart over the Western Pacific on April 6th, where cool colors depict negative height anomalies/stormy weather, and warm colors represent quiet, warm weather. On the morning of April 6th, we saw a trough moving through Japan, possessing a pretty decent strength. This is significant, as it directly relates to what we may see in this April 13-15 timeframe .There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. Thus, we can expect a storm in the US on an April 12-16 period, and the April 13-15 ECMWF snowstorm falls right within this timeframe. Additionally, this trough in Japan brought along some pretty cold weather, which COULD contribute to additional chances for snow in this potential storm system. 

The image above shows the projected mean surface pressure by the ECMWF ensembles, as well as the spread among all ECMWF ensemble members in the shown colors for April 14th. We first observe how the ensembles are much less enthusiastic about the idea of a storm system for this timeframe, only bringing a minimum 1004-1002 millibar swath of low pressure over an area from northeast Indiana into southeast Michigan. This sort of weaker projection is to be expected, as these ensemble means take the average of 52 ensemble members, unlike the ECMWF model, which only shows one solution. The next item we observe is the swath of oranges and yellows over the Plains and Midwest. These warmer colors indicate a higher spread in the ensembles; in other words, the ensemble members are in higher disagreement with one another over this higher spread envelope, thus indicating increased uncertainty. However, looking back at the post on this storm from the other day, we note that the uncertainty has decreased, with the ECMWF ensembles showing mainly yellows (decent uncertainty), rather than a swath of oranges, which depicted even higher uncertainty. This bodes well for the idea of a snow event.

If we look at the projected ECMWF ensemble mean 850mb temperatures over North America for April 14th, we find a startling scene. The ensemble mean actually seems to support an upper Midwest snow event, with the freezing line located just west of the dashed black line (marked as the number 0 ). I don't have access to precipitation products that display each ECMWF ensemble member, but just analyzing this map alone, it does seem like the ECMWF ensembles would support at least a slight snowfall event.

The image above shows each individual ensemble member from the GFS Ensemble set, valid for this timeframe. In many of the ensemble members, we see substantial precipitation located to the west of that yellow-ish/brown-ish line, which defines the line between freezing and non-freezing temperatures about 5,000 feet above the ground. It's important to note that the ensembles are still pretty spread out, despite decent agreement on the precipitation location, and it is because of this spread, among other things, that I'm very hesitant to support a snowfall event for this time period.

Here's the caveats I'm currently concerned with about this storm.

• This event would be happening in mid-April, rather than mid-January. Thus, the environment is much more hostile to snowfall.
• Model guidance, including ensembles, remain inconsistent with this storm on differing scales.
• Model guidance is notorious for over-projecting cold air coverage and intensity.
• The pattern has been very "wintry" this year, meaning the models may still want to hold on to this "wintry" feel, whereas the actual pattern may be much warmer.

Let's summarize this.
• A potentially major storm system is expected in the April 12-16 timeframe.
• Colder than normal weather is expected to arrive with this storm system.
• Model guidance is hinting at a potentially significant snowfall event during the April 14-15 timeframe, in conjunction with this potentially major storm system.
• There is high uncertainty with this potential event. Caution must be used.

Andrew

April 13-15 Potential Late-Season Snowstorm

WARNING: The chances of a significant snowstorm in April are VERY low. This post is only showing one model's output, and is made to alert people that there could be some snow in this timeframe. I do NOT expect we see these amounts verify exactly.

I'm watching the April 13-15 timeframe for what could be a strong storm system resulting in, yes, snow.

Shown above is an image depicting the most recent ECMWF model snowfall forecast for a potential April 13-15 snowstorm. The exact image showing the ECMWF snowfall output cannot be displayed for copyright purposes. In this most recent forecast, we saw a swath of 2 to 6 inches of snow stretching from northern Missouri into western Iowa and central Illinois, before moving northeast into both land masses of Michigan. Inside that accumulating snow swath, we then see a rather large area of 6 to 12 inches-plus of snow from western Iowa and northern Illinois into southeastern Wisconsin, before continuing into the northern LP of Michigan. As I mentioned above, I do not expect we see these snow amounts hitting for this timeframe; this post is merely to alert people about the potential for snow.

Tropical Tidbits

There is some merit to the idea that we may see a strong storm system, as the ECMWF is suggesting. Shown above was the observed 500mb height anomaly chart over the Western Pacific, where cool colors depict negative height anomalies/stormy weather, and warm colors represent quiet, warm weather. On the morning of April 6th, we saw a trough moving through Japan, possessing a pretty decent strength. This is significant, as it directly relates to what we may see in this April 13-15 timeframe .There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. Thus, we can expect a storm in the US on an April 12-16 period, and the April 13-15 ECMWF snowstorm falls right within this timeframe. Additionally, this trough in Japan brought along some pretty cold weather, which COULD contribute to additional chances for snow in this potential storm system.

Looking at the raw ECMWF surface pressure forecast for the morning of April 14th, we see the storm system at a pretty low minimum central pressure, likely below 1000 millibars, in the middle of Missouri. The ECMWF model may be locked on to this snowstorm idea, but its ensembles are much more uncertain.

The image above shows the projected mean surface pressure by the ECMWF ensembles, as well as the spread among all ECMWF ensemble members in the shown colors. We first observe how the ensembles are much less enthusiastic about the idea of a storm system for this timeframe, only bringing a minimum 1006 millibar swath of low pressure over an area just to the east of the ECMWF model. This sort of weaker projection is to be expected, as these ensemble means take the average of 52 ensemble members, unlike the ECMWF model, which only shows one solution. The next item we observe is the swath of oranges and yellows over the Plains and Midwest. These warmer colors indicate a higher spread in the ensembles; in other words, the ensemble members are in higher disagreement with one another over this higher spread envelope, thus indicating increased uncertainty. While this would initially disprove the idea of a storm system in this timeframe, note how the highest spread is located further west of the ensemble low pressure alignment, more in line with the ECMWF model. All in all, this means that the ensembles are seeing this idea of a storm, but aren't biting yet.

To summarize:
• A potentially major storm system is expected in the April 12-16 timeframe.
• Colder than normal weather may arrive with this storm system.
• Model guidance is hinting at a snowfall event during the April 13-15 timeframe, in conjunction with this potentially major storm system.
• There is high uncertainty with this potential event. Caution must be used.

Andrew

Verification of the 2013-2014 Winter Forecast

Hello everyone, this is a look back at my 2013-2014 winter forecast, and an analysis of how well it performed against what we observed from December 1, 2013 to March 1, 2014.

Temperature Verification

In my Final 2013-2014 Winter Forecast, I projected that we would see the core of the cold centered in the Northern Plains, Upper Midwest, Great Lakes, Midwest, Ohio Valley and into the Central Plains. The forecast also called for warmer than normal temperatures in the Southeast, while average conditions were to prevail over the southern Plains and the Northeast. My projection had the West US in for back and forth temperatures, ending up around normal.

Actual temperature anomalies for December - January - February 2013-2014

Looking back at what actually happened this past winter, we saw the core of the cold in the Upper Midwest into the Midwest itself, with still well-below normal temperatures in the Northern Plains, Ohio Valley, Great Lakes, and Southern Plains. This part of the forecast verified quite well, with the only trouble spot being the Southern Plains, which ended up a bit cooler than forecasted. The Northeast observed overall normal conditions, verifying that part of the forecast. The Southeast did end up a bit warmer than normal, but not as much as I thought it would be. The main trouble spot was in the West, where we saw warmer than normal temperatures instead of the average conditions I predicted. This can be attributed at least partially to the drought, as well as the persistent ridge in the Gulf of Alaska.

On a self-grading scale, I might put this forecast at an A-minus, maybe down to a B-plus.

Precipitation Verification

For the precipitation forecast, I anticipated above normal precipitation from the Central Plains into the Midwest and Great Lakes, believing that was where we would see the primary storm track. I also believed a drier than normal winter was in store for the Southeast, with neutral conditions in the East and West.

This part of the forecast actually busted quite badly. The Central Plains saw below normal precipitation rather than the above normal precipitation that was forecasted, though we did see verification in the Great Lakes. Overall, the East Coast didn't stray too far from what I thought would happen, ending up only slightly above normal overall compared to my projected average conditions. The West US definitely busted hard, with a serious drought taking over instead of my forecasted average conditions. The Southern Plains also ended up drier than normal, while the Southeast did not.

On a self-grading scale, this portion of the forecast likely deserves a D-plus.

Snowfall Verification

The snowfall forecast called for snowier than normal conditions over the Plains, Upper Midwest, Midwest and Great Lakes. The rest of the country was pegged for average conditions, with the East Coast possibly in line for some big storms.

I do not have a verification image, but I can affirm that the Midwest and Great Lakes regions verified accurately in this portion of the forecast. The East Coast ended up a bit snowier than I thought it would, and the West definitely got much less snow than what is considered average.

I would put the accuracy of this forecast at a B even.

Overall Forecast Verification

The overall forecast called for some big snowstorm potentials in the Northeast, with a warm and dry winter in the Southeast. The risk for a warm and dry winter existed in the Southern Plains, with a cold and snowy winter outlined for the North Central US. Slightly below average temperatures were anticipated for the Rockies, while a warmer than normal winter was projected for the Pacific Northwest. I feel verification of this graphic went better than I actually thought it would, with most areas seeing at least partial verification of this winter forecast.

For the overall forecast, I believe a grade of B-plus would be appropriate here.

What did you think of the accuracy of this forecast? Comment below with your thoughts!

Andrew

March 25-27 Potentially Significant Blizzard

I'm watching the potential for a potentially significant snowstorm that model guidance believes will become a blizzard.

The image above shows 500mb geopotential height anomalies from the latest run of the GFS model over the Western Pacific, valid for March 20th. As this image shows, we see strong negative height anomalies present over Japan on March 20th, indicative of a rather strong storm over the area. There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. So, if we take this March 20th storm and move it ahead 6-10 days, we find a potential storm system around the March 26th through 30th period.

Let's take a look over current model guidance. Shown above is the latest forecast from the Canadian GGEM model, valid on March 26th. We see an extremely strong storm system positioned just offshore of the Northeast, dumping heavy precipitation on the order of over 12" on inland areas, with multiple FEET of snow falling on coastal locations. The GGEM model has already been shown to be suffering from convective feedback issues on this run, meaning precipitation amounts are grossly exaggerated and incorrectly altered due to this specific bias. Thus, this forecast run is overdoing precipitation amounts and is not correct in its forecast at this time. What I do agree with, however, is placement of the storm. We look to be in a northwest flow regime, a pattern notorious for providing a very favorable environment for East Coast snow events, and we'll discuss this a bit later.

We now move on to the ECMWF model, which also shows a very strong storm system displaced just east of the Northeast coastline. This is of similar strength to the GGEM model and would likely be putting down some pretty darn heavy snow amounts, but because we lost access to the main provider of our weather maps, we can't give an indication as to just how severe this storm looks to be on the ECMWF. However, once again, we're seeing this strong storm in the midst of a northwest flow pattern, as we see the dashed lines dropping down to the southeast from Canada. This is crucial to the forecasts of a strong East Coast storm.

The American GFS model is far less enthusiastic about this storm, only putting down modest precipitation along the coastline and sparing inland regions from any accumulating precipitation, snow or otherwise. This is considered the outlier forecast when taking into account how the GGEM, ECMWF and ECMWF Ensembles believe we'll see a strong East Coast-hugging storm, but in the grand scheme of things, there is high model agreement on a storm at least threatening the East Coast in this timeframe, and that's what we're looking at here.

Take a look at this graphic above. This image shows a few things important. The main item we see first are the negative 500mb and positive 500mb anomalies, marked by contour lines in blue and red, respectively. Just glancing over the image, we find a sustained northwest flow scenario unfolding, with deep negative height anomalies pressing into the East US, with high pressure surging northward in the West US. The second item we need to discuss are the dates on the bottom right hand corner. They are analog dates, listed from top to bottom, and the date at the top means its pattern is most similar to the pattern forecasted in the graphic itself, which is valid around the March 26-30 period, right on top of the timeframe outlined for this potentially significant storm. Look closely at the dates- are you picking up on one particular date in that list? If not, check out the second date from the top. That pattern was deemed to be the second most similar to the pattern expected to unfold as this storm system impacts the East US. And that date - March 16, 1993 - spawned the 'Storm of the Century'.

Now, there are some who are anticipating another Storm of the Century because its date has shown up on these analog images, and they might tell you to prepare for feet of snow. All that we know now is that model guidance is hinting rather consistently at the potential for an all-out blizzard in the East US, specifically the Northeast. As of right now, there is disagreement on who could get hit by the storm, and there's even disagreement on if there'll be a storm at all. As of now, the potential for a significant storm is rather high due to the East Asian correlation, but it remains to be seen if this could be a blizzard for the East Coast, hence all the caveats throughout the post.

Andrew

March 26-30 Potentially Significant Storm System

The potential for a potentially significant storm system over the March 26-30 period remains alive.

Tropical Tidbits

The image above shows 500mb geopotential height anomalies from the latest run of the GFS model over the Western Pacific, valid for March 20th. As this image shows, we see strong negative height anomalies present over Japan on March 20th, indicative of a rather strong storm over the area. There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. So, if we take this March 20th storm and move it ahead 6-10 days, we find a potential storm system around the March 26th through 30th period.

In an earlier post, I had addressed the idea of two strong storms in this timeframe. Model guidance has dropped the idea of two storms on March 20th in Japan, and appears to be going with one storm instead.

Tropical Tidbits

If we look at the long range GFS jet stream forecast for March 26th, the beginning of this four day timeframe where a storm is expected, we see a few things. The primary thing we want to discuss is the active subtropical jet stream. If you recall, there are two jet streams which have the most effect on our weather (there are three jet streams in total): the Pacific jet stream, which comes from the Pacific and is a primary driver of weather pattern, and the subtropical jet stream, usually located in the South US, with less influence but high-impact effects at times. In this case, I'm concerned because an active subtropical jet stream combined with this potentially strong storm system may create a severe weather situation. There are a lot of variables that need to come together for this to happen, but we've at least established the idea of a strong storm system hitting the US in this timeframe. This is indeed something to watch out for.

Andrew

March 26-31 Multiple Potentially Significant Snowstorms

I'm examining the idea of multiple significant winter storms over the March 26-31 period.

The above image shows a recent ECMWF forecast, projecting 500mb height anomalies over the North Pacific on March 20th, where blues and purples show negative height anomalies (cool and stormy weather), and oranges and reds depict positive height anomalies (warm and quiet weather). Looking towards the continent of Asia, specifically zeroing in on Japan, I marked two points of interest using the letter X. These X's denote the location of storm systems in the area. There is a storm system moving east over Japan, tilting negatively as it does so, and a storm to the west of Japan, just beginning to drop south and racing towards the country. There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. The two storm systems both retain impressive intensities, with the bright purples indicating that both systems look to be strong.

About 18 hours later, we see that the situation over Japan has changed. The first storm system that was previously over Japan has shot north, influenced by the even-stronger system shown in green shades south of far northeast Russia. The second storm system that was previously racing towards Japan is now impacting Japan, maintaining rather-strong characteristics, as shown by the tinted purples over the country. This confirms there may be not one, but two storm systems in this timeframe.

Shown above is the ECMWF Ensemble mean forecast of 500mb height anomalies, over North America. The same color rules, where blues indicate stormy weather and oranges show calm weather, still apply. This forecast is valid on March 26th, the beginning of the five day period we're watching for these two potential storms. We see storminess evolving over the West Coast, shown by the darker blues just offshore of California, Oregon and Washington state. We see this provoking slight ridging in the southern Plains, well illustrated by the erosion of blues in that region. We also see an old friend in a piece of the polar vortex still sticking around in Canada. The storminess over the West Coast should enable the high pressure over the Southern Plains to gradually strengthen and push east, as it's possible we see a negative Pacific-North American pattern evolve, which is characterized in the image below.

Typical negative PNA pattern
NCSU

We also see how a lobe of the polar vortex still sits over Canada, as it has been doing for much of the winter, and should continue to do into spring. If this forecast were to verify, it wouldn't be too hard for the storm track to be suppressed, from the Midwest down into the Ohio Valley. This could favor areas that have been hit multiple times already this winter, including the lower Midwest and Ohio Valley regions.

The GFS Ensemble forecast for 500mb anomalies at about the same timeframe have a pretty similar pattern, even though they may seem different at first glance. The PNA remains negative, like the ECMWF ensembles project, as we see strong negative height anomalies over the West Coast. We also see a lobe of the polar vortex positioned in Canada, as we did in the ECMWF Ensemble image. The only 'difference' here is that the ridge we discussed in the ECMWF Ensemble image is more pronounced on this image, and located further east. This solution would likely still drive storm systems into the Midwest, but they would also be suppressed- just not as much as the ECMWF ensembles say they would be. This sort of projection would take the two potentially significant snowstorms into the Midwest and Ohio Valley, as ridging in the Southeast would deter an East Coast solution.

Caveat: The ECMWF projection that says we would get two major storm systems is a long range forecast, and has a chance of not verifying completely. There is a pretty good chance we would be seeing a potentially significant storm system in this timeframe, it's the idea of two storm systems instead of one that isn't really solidified just yet.

To summarize:
-One or possibly two potentially significant storm systems are expected in the March 26-31 timeframe.
-Both winter and severe weather modes would be quite possible with this event.

Andrew

March 22-26 Potential Weak Storm System

There's the potential for a weak storm system in the March 22-26 timeframe.

WeatherBell

It's looking like we'll be seeing a storm system cut across Japan on March 16th, with rather meager negative height anomalies, indicating a weak storm system. There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. Thus, 6-10 days after the date where we see a storm hit Japan (March 16) brings us to a potential storm system in the US around the March time period.

Looking deeper into that ECMWF forecast, we see strong high pressure over the West Coast. This is strongly indicative of a positive Pacific-North American pattern (positive PNA), which is demonstrated in the above graphic. The positive phase of the PNA sees high pressure form over the West Coast, which then leads to storminess and cold weather in the Plains. This sort of pattern can give snow to the Midwest and Ohio Valley in situations where ridging tries to push north in the Southeast, but can also send wintry weather to the East Coast if we see high pressure over Greenland.

WeatherBell

If we go out in the long range, we can see what the weather pattern may be like on March 22nd, the first day of this window that has been designated for a storm system. We see that the pattern includes warm weather in the West, as persistent ridging continues, but also pushes east into the Plains. More importantly is the presence of a piece of the polar vortex digging well into the Northeast, likely providing the chance for another intense round of cold weather. This sort of pattern tells me we would likely see some severe weather in the South Plains, but that would be about it. This pattern isn't conducive for wintry weather in the Plains, Midwest or Ohio Valley, and the polar vortex will be suppressing the storm south to likely keep it away from the East Coast.

This looks to be a weak storm, and impacts should be minimal on the wintry side. I would monitor severe weather prospects in the near future, however.

Andrew

March 20-24 Potential Storm System

I'm watching out for a potential storm system around the March 20-24 timeframe.

WeatherBell

There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. Using the image above, we can expect a storm in the US 6-10 days after March 14th, which would bring us to a March 20-24 storm. 

The storm looks pretty healthy in this model projection, which could also mean we see some active weather, both in terms of severe and wintry weather. It's still too far out to tell who could get hit, but considering we'll be seeing the Madden-Julian Oscillation in Phase 2 around this timeframe, I would keep an eye out for severe weather.

(The following is copied from the March 4th post)

Paul Roundy

Shown above is a chart of six different types of outgoing long wave radiation (OLR) anomalies. We want to focus on the middle panel of the left-hand side, where it says MJO OLR anomaly on February 16, 2014. On February 16, 2014, we saw negative OLR anomalies just south and east of the subcontinent of India. This location of the -OLR anomalies (which indicate enhanced tropical convection) tells us the Madden-Julian Oscillation was in Phase 2, out of its 8 possible phases. Check out the severe weather event that occurred just four days later, on February 20, 2014.

On February 20, 2014, we saw a large-scale severe weather outbreak, as the chart of storm reports from the Storm Prediction Center shows above. The outbreak was primarily a damaging wind event, as you can see by the large swath of blues across Tennessee and Kentucky, just to name a few of the many affected states. However, we also saw a handful of tornado reports. The fact of the matter is, this severe weather event and the MJO entering Phase 2 are not coincidence. They are correlated with one another. See the AMS (American Meteorological Society) article below:

The Madden–Julian oscillation (MJO) has been linked to weather variability in the midlatitudes via its associated overturning circulations and Rossby wave trains that redistribute the thermal and mass fields at higher latitudes. This work examines the relationship between the MJO and violent tornado outbreaks in the United States. A census of events shows that violent tornado outbreaks during March–April–May (MAM) are more than twice as frequent during phase 2 of the Real-time Multivariate MJO (RMM) index as during other phases or when the MJO was deemed inactive.

If you didn't get what this was saying, don't worry. Basically, it tells us that when the MJO is in Phase 2 during March, April and May, or when we see enhanced convection south and southeast of India, we tend to see the chances of a violent tornado outbreak skyrocket in comparison to the nonactive MJO, or the MJO in other phases. While the February 20th event wasn't exactly a tornado outbreak, it was a severe weather event, and this has me concerned for mid-March.

Let's take a look at the forecasted OLR anomalies for March 12, 2014. Once again, we'll look towards the panel labeled MJO OLR Anomaly. If we look at the forecast by Paul Roundy above, we can see very deep negative OLR anomalies placed just south and southeast of India. These blues are far deeper than what we saw on February 16, meaning that this Phase 2 MJO event could be even stronger, and thus the severe weather event would be even stronger as well. Going by the LRC we discussed above, as well as the AMS document analyzed here, I am concerned that we will see a potentially significant severe weather outbreak for the mid-March period. Because there could be a lag period between the MJO Phase 2 event and the severe weather event of about 4 days, like we saw with the February event, I'll make a tentative timeframe of March 13-18 for a potential severe weather outbreak. This one has me worried, and while you should not be worried yet, it's something to watch out for in the future.

(End copied post)

If we come back to Paul Roundy's forecast and look at the MJO OLR Anomaly panel, it looks like the placement for Phase 2 anomalies has been pushed back to March 16, which would put the possible severe weather threat either just before or during this potential storm system. Lots of things to watch here.

Andrew