TERRY'S TAKE ON WINTER 2021-22...

Cedar Rapids Iowa April 10th, 1973

Hey everybody, TSwails.com needs your help. For the site to continue, I am asking those of you who use, rely on, and appreciate the work and value to consider a voluntary subscription fee of $12 dollars a year, $1 dollar a month to pay for its expenses. If I can't meet my fund raising goal the site will be discontinued in November. Please click on the secure green box below for more information on how you can make a contribution. Thanks to all, T. Swails

THE 2021-22 WINTER OUTLOOK

With the arrival of October, my thoughts are increasingly drawn towards the winter ahead. This is an exciting time of year for me as there is nothing more challenging than forecasting the complex interactions that drive winter storms. Individual snow events are only on my radar for a few days (perhaps a week), and yet their implications often remain unknown right up to the very end. That said, if it's that difficult to forecast a snowstorm over a period of days, you can only imagine how hard it is to to make a seasonal forecast for winter encompassing three months. I'm the first to point out that I'm a meteorologist, not a climatologist and this type of outlook is on the fringe area of my expertise. Making a winter forecast in early October is fraught with peril so keep in mind what you are about to read is nothing more than an educated guess. If it doesn't pan out, neither you or I should lose any sleep over it.

Having laid down my qualifier, I do want you to know that I have given the topic of winter countless hours of thought and will supply you with solid arguments that support the theories I'll identify below.

THE SOUTHERN OSCILLATION

The first place I always begin a long range winter forecast with is by determining the state of the upcoming ENSO. ENSO stands for (El Niño-Southern Oscillation) The ENSO is a recurring climate pattern involving changes in the temperature of waters in the central and eastern tropical Pacific Ocean. During periods ranging from about three to seven years, the surface waters across a large swath of the tropical Pacific warm or cool by anywhere from 1° to 3°C, compared to normal.

This oscillating warming and cooling pattern, referred to as the ENSO cycle, directly affects rainfall distribution in the tropics and can have a strong influence on weather across the United States and other parts of the world. El Niño and La Niña are the extreme phases of the ENSO cycle; between these two phases is a third one called ENSO-neutral.

So we've established the fact that El Nino and La Nina events are essentially sea surface temperature anomalies that exist near the Equator. Strong events, whether they be associated with El Nino or La Nina, flood the atmosphere with unusual amounts of energy that in turn control weather patterns in a somewhat predictable way, especially in the winter and spring. Strong ENSO events can have global implications with catastrophic results that range from excessive precipitation to drought. Temperatures are significantly altered as well. Weaker versions of ENSO are far less predictable and can be influenced by other factors such as atmospheric blocks and sea surface temperatures in other parts of the globe. With 70% of earth covered in water, sea surface temperatures play a large role in our overall climate.

The first order of business then is to figure out the state of this years ENSO and its strength. With unanimous consensus, all of our climate models indicate a developing La Nina is in the works (a build-up of cold sea surface water in the Equatorial Pacific), so that mystery is relatively clear. What isn't is the duration and intensity of the La Nada. The tropical Pacific constitutes a large geographical region making it tough for models to accurately pinpoint how cool the water will be, the spatial extent of it, and how long it remains in place. This graphic from NOAA shows the cold water associated with a La Nina (in blue) already in place over all of the central and eastern Pacific.

That certainly resembles the example of what NOAA uses to define La Nina below .

Forecasters at NOAA expect further cooling with the La Nina steadily strengthening through the fall and lasting through winter of 2021-22.

Predicted model plumes show a weak to moderate La Nina at its peak during the months of November through January before weakening come spring.

Another point worth mentioning is that ENSO can be measured and the location of the coolest water has important implications. The large expanse of water that encompasses ENSO is divided into boxes that have designations. Having a weak or moderate La Nina in the central Pacific is generally considered more favorable for a normal or colder than normal winter with normal to above normal snowfall. The central Pacific is considered region 3.4. Currently the western and central regions are forecast to have the coolest sea surface water departures during winter.

During a weak to moderate La Nina, the polar jet is frequently aimed at the upper Midwest which can at times bring significant outbreaks of cold air. However, there are intervals where it retreats allowing brief but welcome breaks from the cold. Overall though, the signal is for near to slightly colder than average temperatures assuming all things are equal.

This graphic measuring moderate La Nina's in ENSO 3.4 (which is indicated this year) shows the cold centered over the upper Midwest through the Great Lakes and into the Northeast.

Winter precipitation is typically about average, perhaps a bit above normal from my area east. Individual storm tracks not known at this time and can cause notable variations.

For all La Nina's this is what average snowfall looks like. Again, where individual events set-up can skew the averages. Moderate La Nina's tend to be a bit snowier than average.

Below you can see average La Nina winter temperature anomalies over the period 1949-50 through 2017-18. In my area 9 winters were colder than average, 7 near average, and 6 warmer than average. 2008-09 and 2000-01 were the coldest and 2011-12 was hands down the warmest.

THE QUASI-BIENNIAL OSCILLATION (QBO)

Another factor that can give us a clue about the winter ahead is the QBO (Quasi-Biennial-Oscillation). Considered one of the most remarkable phenomena in the Earth's atmosphere, the QBO originates in the stratosphere high above the equator where strong zonal winds blow in a continuous circuit around the Earth. The QBO oscillates back and forth blowing in an east to west direction for a few months, becomes neutral and weak for a few months, and then reverses blowing in a west to east direction. The whole cycle takes about 26 months to complete. The direction and intensity of the QBO plays an important role in the evolution of the winter pattern. It's been shown that winters with strong negative values during winter significantly increase the chances of cold and snow here in the central U.S. The majority of the climate models are indicating at least a weak to moderate QBO during the winter months.

The 700mb upper air pattern during winters with a weak to moderate La Nina and an east QBO looks like the bottom right panel. High pressure resides off the west coast and the mean trough is centered over the upper Midwest. Good if you like winter. The west QBO on the left is a much warmer solution and is not forecast.

Winters where the QBO averaged negatives of -10 to -15 C. December through February, had temperature anomalies that looked like this. The QBO ensembles shown above are leaning that way.

When the QBO averages -15 to -20 the anomalies are even colder.

TELECONNECTIONS

That leads us into another area of climate drivers known as teleconnections. There are several that I discuss regularly, especially during winter when their skill levels are greatest. The problem with teleconnections are that they are difficult for models to forecast beyond 7 to 14 days. At this distance from winter, we can only make broad assumptions as to what phase they will be in at any point in the winter based on the trends we've discussed above, in particular sea surface temperatures. Consistently warmer or colder than normal water temperatures have proven tendencies for a specific type of teleconnection to form and perhaps persist much of the winter.

One that I will rely heavily on when the time comes is the MJO (Madden Julien Oscillation). Essentially the MJO is a disturbance that moves eastward through the global tropics roughly every 30-60 days. The convection it generates feeds back energy which passes through one of eight geographical regions known as phases. Each phase is known to produce specific regions of above and below normal temperatures and precipitation across North America at a specified time of year.

During the heart of winter, phases 8, 1, and 2 are the holy grail of cold. Winters where the MJO is consistently in those phases tend to be harsh and snowy. Phases 4, 5, and 6 are the opposite. Below is the MJO phase diagram from the winter of 2017-18. If you follow the dotted lines you will see that from December 23rd to January 5th the MJO traveled through phases 8, 1, and 2. The resulting period was one of the coldest on record in many parts of the nation.

In fact, that period in the Midwest was the coldest ever measured in many areas. If it wasn't #1 in a specific region it was in the top 3.

Look at the anomalies while the MJO was in phase 1/2. The MJO is a great tool when it comes to determining trends and pattern changes.

When you see the MJO forecast to enter the cold phases with any degree of amplitude during the heart of winter look out. Now of course, MJO forecasts don't go out that far yet so it's a mute point unless you consider the correlation that a cold May leads to a cold December. The thought is that the high amplitude MJO that creates the chill in May, months later results in an MJO cycle that is in a cold phase when December arrives. If the theory holds, combined with the evidence above, this December has the potential to be cold allowing the winter to get off to a fast start. Here's the extent of the below normal temperatures this past May 2021. The MJO was amplified in cold phases.

I'm going to take this a step further too. Cool May's generally lead to a warm October's and I see that's happening this year. When October's are warm in the central and eastern U.S., analogs support a cold December in the Midwest and Great Lakes. Here's the 30 day temperature departures from the EURO weeklies through October.

Other key teleconnections exist and one to watch this year will be the PNA (Pacific North American Oscillation). When there's a large pool of warm water off the west coast of North America as there is this year, it often supports a positive PNA. This produces a persistent ridge over the western U.S. and a downstream trough in the central or eastern part of the country. That supports cold air intrusions and opens the door for respectable winter chill. See the warm SST off N. America below.

The Pacific North America pattern typically remains in a given phase anywhere from a few days, to a few weeks, and has a reputation for being unpredictable which makes it an unknown factor this early in the game. If you want cold, you want to see the PNA in a positive state. Many times the +PNA occurs in conjunction with the colder phases of the MJO. Below you can see the orientation of the jet stream with both the positive and negative phases and why the phase matters.

The last two teleconnections that are major pattern drivers are the AO (Arctic Oscillation) and the NAO (North Atlantic Oscillation). The Arctic Oscillation plays a vital role in driving the climate of the mid-latitudes and is particularly useful in long range forecasting. The AO is one of Earth’s most important atmospheric climate cycles. It occurs over the Arctic and influences mid-latitude weather patterns over the entire globe.

On average, there is consistent low pressure centered over the North Pole. The core of this low pressure is located near the stratosphere and is known as the polar/vortex. It tends to be much stronger in winter than it is in summer, and thus the influence of the Arctic Oscillation is greater in winter than it is in summer.

The Arctic low pressure generates a west to east circulation around the periphery of the vortex and causes winds to generally blow from west to east. The strongest winds in this circulation make up the polar jet stream. The changes in the strength and position of the polar jet have a major influence on the weather patterns here in the Midwest.

When pressure north of the Arctic Circle is lower than normal, the Arctic Oscillation is considered positive. A positive AO leads to a stronger and unbuckled polar jet ( what we call zonal flow). The more positive the AO index, the stronger and straighter the jet. The repercussions here in the Midwest are that the polar jet maintains in a stable, northerly position that tends to keep the coldest air confined to the Arctic. Simply put, the +AO is generally associated with above normal winter temperatures.

Conversely, when pressure north of the Arctic Circle is higher than normal, the Arctic Oscillation is deemed to be negative. A negative AO leads to a weaker, buckled polar jet. The more negative the AO index, the stronger the associated cold air outbreaks.

With the polar jet weakened by higher pressure, it allows cold air to drain out of the Arctic. That means that much of the mid-latitudes experience colder than normal weather during the negative phase of the AO. It can on occasion result in stratospheric warmings which can displace the entire polar vortex (the core of the coldest air in the Arctic) and transport it to the upper Midwest. Our worst Arctic outbreaks are associated with stratwarms.

The Arctic Oscillation typically remains in a given phase from a few days to a few months, and is usually unpredictable beyond 14 days. In winter, however, a strong positive or negative phase tends to persist for many weeks and can sometimes be predicted with lead times of 2 to 3 weeks. Below you can see the recent history of the two phases of the AO going back to 2015. Notice there are periods where one phase is more dominate than the other.

Unfortunately it's a bit too early to get reliable forecasts regarding the AO but when the time comes this teleconnection is a very important driver and one you will hear me talk about as winter comes on. The EURO weeklies do show a trend for the AO to become strongly negative in early November, (as far out as I can get). The million dollar question is will that happen and will it remain there on a frequent basis December through February. At some point this needs to happen for winter to gain a foothold.

Now let's talk about the NAO, (North Atlantic Oscillation). The NAO is a blocking pattern that allows the polar jet to merge or phase with the sub-tropical jet. The polar jet supplies the cold and the other the moisture. If the two can team up it's a recipe for winter storms. Additionally, when the NAO is neutral or negative it increases the chances of large scale Arctic air masses entering the eastern 2/3rds of the nation. Overall, I believe the NAO becomes a greater factor mid to late winter than at its onset. The NAO phase is determined by measuring the difference in sea level pressure between Gibralter and SW Iceland.

Below you can see the look of a negative NAO. The block is centered near Greenland which buckles the jet stream in from the northwest. That opens the door to the Arctic and allows cold air to drive into the heart of the nation, often times leading to snow near and north of the storm track.

The NAO is a bit of a wild card and certainly is a challenge to forecast far in advance. Going back to the winters of 2018-19, 2019-20, and 2020-21 you can see the NAO did not spend much time in the negative phase and overall those were not real tough winters, especially December and parts of January. I expect that to change this December, especially if we can get the negative NAO to assist. I do look for more blocking in NE Canada so the NAO could be a bigger factor this winter than in previous ones.

SEASONAL CLIMATE FORECASTS

Now that we've established the drivers, early trends in modeling, and the importance of teleconnections in a winter outlook, lets take a look at some of the forecasts that a currently in place from our climate models. Since it's early, most of these outlooks are for the first half of winter, generally November-January.

The EURO is usually a good place to start so lets begin with it. Here's the mean 500mb jet stream projections. There's a ridge in the west and a tough in the central and east. Lots of blocking in Canada. That points to cold air intrusions east of the Rockies

The temperature anomalies show no cold air anywhere over North America and we know that physically that's simply not going to be the case. For whatever reason the EURO is just awful at depicting cold air in its seasonal outlooks. What I look for are the areas showing warmth knowing that there is going to be an opposite reaction where cold would be found somewhere else. To me that spot is in the center of all that white...which is essentially the Midwest. By just following the edge of the warmth (the yellow color) from the northern Rockies to Texas you can see the alignment of the jet and if indeed the model is right, I feel confident there will be cold driving into the central U.S.

The EURO shows a slight trend for below normal precipitation but if the cold comes as the model depicts, a fair amount of what occurs could realistically fall as snow.

The CFSv2 (the U.S. climate based model November-January) is a little less amplified but still shows ridging over Alaska which implies a downstream trough in the heart of the country and that's where one would look for cold.

Looking at its temperature anomalies it clearly shows the door open to cold but is probably underdone with the extent. I think the blue colors just to our north have a reasonable chance to be even further south. The inherent density of cold air masses alone should drive the cold south in a pattern such as this. Something else than can play a major role is snow cover in Canada. It's too early to know how that will develop in coming weeks but suffice it to say, the more there is and the further south it sets up, the better the chances for polar air masses to invade the Midwest.

The CFSv2 differs from the EURO in the sense it has a wetter look. I can't say that will be the case but I'm always in favor of an active storm track, especially if it is to the south which is where you want it for snow.

The Canadian long range model November-January looks good too. The mean trough is over the central U.S. with the ridge off the west. That's the recipe for cold air

It shows normal temperatures as well which I would shade more towards the colder side of the ledger knowing the challenges models have with seeing below normal temperatures at lengthy distances.

It also has the mean storm track close to the region but definitely centered south. That's a good snow maker track.

The JMA (Japanese Meteorological Agency) has an outlook that goes December-February. The 500mb jet (top left) again shows the western ridge and the trough situated over the Great Lakes. Top right its pressure tendencies at 850mb show high pressure off the west and east coasts. The lower pressure between the two would be the trough. You can also see in this panel the mean 850 wind flow out of the NW. Nothing warm comes from the NW in winter.

The precipitation panel on the bottom left shows above normal amounts from the central Midwest into the Northeast.

To sum the seasonal models up......

All show the western ridge and downstream trough that can open the refrigerator door.

None indicate anything warmer than normal temperatures and I feel those are probably a bit too warm.

Three out of 4 models indicate above normal precipitation with the other just a hair below average.

All 4 models indicate the mean storm track near and south of my area, an important location for snow events.

ADDITIONAL FORECASTS

There are other forecasts that have been issued, some scientific and some not so much. I'll let you figure that out. Here they are in no particular order.

FOLKLORE AND TRENDS

There are other ways of making forecasts too by using years of observation which often become what's known as folklore. One of the very best at that is Steve Gottschalk of Lowden, Iowa who for 60 years has researched folklore and taken observations which he uses to make his own outlooks. Steve has been in numerous magazines and newspapers with his outlooks and dare I say, is better than some of the more well known outlooks that get plenty of play. Here is what he has observed so far.

"The spiders are more numerous and coming in earlier than usual this season which suggests a colder winter. I have just 2 woolly bears so far, one says colder and one says warmer. The hot weather has been keeping them away. I will be keeping my eye out for them and keep you updated".

The cricket numbers are way up again this year and they have been coming indoors for several weeks now. The paper wasp numbers are up again too this season and the horse weed is much taller than usual. All 3 of these signs indicate a snowier winter.

The squirrels have been really busy over the past week burying the walnuts and acorns suggesting a colder winter. Some of the flowers have bloomed for a second time this year which brings a colder winter 70% of the time.

"Some weather trends that I have noticed: A warm September following a warmer summer usually means a colder winter about 70% of the time".

Some forecaster's, such as Terry are predicting a warmer than normal October and so far the temperatures are averaging some 10 degrees above typical levels. I have found that 9 of the top 11 warmest Octobers had colder winters especially, when there was a La Nina in play. When the number of 90 degree days during the entire warm season are around 150% of normal (such as this year), the winter is colder 69% of the time. When you add a La Nina to the mix, the chances for a colder winter are even greater

.I have found that when you have a late freeze , I looked at 12 years with them, the winter will be colder during a La Nina event. I also found 14 dry summers like we just experienced and the following winters tend to have normal to above normal snowfall 67% of the time during a La Nina

Steve also adds, "I am thinking the cold weather will be spread out more evenly this winter, not all coming in one month. We have had 7 mild Decembers in a row and we are due for a colder one. Some of the historical trends that I have looked at suggest we may finally see one". The last time we had a bitterly cold January was back in 1982. Looking into Iowa's records going back to 1873, that is the longest stretch we have gone without having a bitterly cold January. I think we are long overdue".

TERRY'S FINAL TAKE:

If you are still with me (congratulations), there's a lot of information and some of it is pretty deep. However, if you want to see the other side of the mountain you've got to climb to the top. I followed all of the trails that I could find to get a glimpse of the truth. From the peak I stand on, here's what I see.

I'm putting a great deal of emphasis on sea surface temperatures as 70% of our planet is covered with water. Using analogs with similar sea surface temperatures gives us a reasonable idea of where the mean troughs and ridges will become established that not only dictate cold, but are precursors of the large scale storm track. If you can nail those features, you can have a pretty good chance of defining regions that have an increased threat of seeing harsh winter weather. With warm pools of water off both coasts of the nation (the Pacific and Atlantic), I expect the mean trough to set up shop somewhere over the central or east-central U.S. For that reason, I think my area is in line for at least slightly colder than average temperatures based on that alone.

Of all the sea surface temperature anomalies I have looked at, October of 2013 matches up best with where we are now in 2021. You will never find the perfect resemblance but there are a lot of similarities between those two years, especially in critical areas. In the comparison below, you can see I've highlighted regions of warmer and colder water (W and C). Both 2013 and this year will be second year La Nina years (back to back) The 2nd year La Nina (cool water) stands out over the tropical Pacific near Nino 3.4. Cool water is also evident off western Canada into the Gulf of Alaska. Further west, warm water is evident south of the Aleutians. Warm water is also prevalent from the Gulf of Mexico into the Atlantic all the way up the east coast, (warmest near New England. That's a pretty strong SST analog.

The idea of an analog is that by looking at the past, one can get a glimpse of the future. It's best if you can get multiple analogs that point to the same result. 2013 being the closest match to current ocean temperatures I wanted to show you how the following winter turned out. As you can see it started fast in December and kept right on going with below normal temperatures December through March. February was bitterly cold over the central U.S. with the mean trough locked in place.

2013-14 was a very snowy winter as well over much of the nation, especially the northern half. In the Quad Cities the period December-February was the snowiest on record with 53.8". It was also the 6th coldest with the average temperature of 16.6 degrees. Below you can see the seasonal snowfall measured in 2013-14

Here was NOAA's outlook for that winter. Might have under done the severity a bit which underscores the degree if difficulty with seasonal forecasts.

To be clear, I am not predicting or expecting anything like 2013-14. It represents the worst case scenario and many things would have to align just right for that type of winter. But, I think it tells us that at least in terms of SST's we have some things going for a healthy winter.

Additionally, if the teleconnections I discussed such as the MJO and AO can consistently stay in cold phases, we could actually have a severe winter but there's too much doubt this early in the fall to say that with confidence. However, there are some analogs that show that trend as well.

With the storm track aligned the way I think it will be, I believe that winter precipitation has a good chance of averaging close to normal. However, a small deviation in the storm track could tip the scales one way or the other. The depth and amount of cold air we have access to will determine how much moisture is available and the ultimate position of the storm track. For snow lovers a track that curves up from Missouri into the Ohio Valley is optimal and that option is on the table. If though, the cold air is deeper or the trough is further east, that is a drier solution. No matter how it turns out, it's my belief that if colder temperatures prevail, even with somewhat below normal precipitation snowfall would still be above normal as a larger share of what does occur comes as snow.

I also think that with the cold February and May we experienced followed by what's going to be a warm October, prospects are good for a fast start to winter. There is considerable support for that in years with those trends. I think the bigger question is whether or not we can consistently sustain wintry conditions into January and February? There are usually periods where we get some breaks and thaws and I do think there will be some ebb an flow but not as much as in recent years. In fact, I expect at least one, perhaps two periods where we have significant Arctic outbreaks, one of those could be in the latter half of December! I would not be surprised to see a stratwarm and a visit from the Polar Vortex.

What can go wrong? Lots of things. First of all I could simply be reading the tea cards wrong. Or I could be caught up in wishful thinking (wanting something too much). That's called "wish-casting" and I'm keenly aware that it can bias opinions so I try very hard to look at both sides of the issue using antilogs.

My worst nightmare though centers around the La Nina ending up stronger instead of weak to moderate. That would be a major problem and could turn the winter significantly warmer but the consensus among numerous models is strongly against that happening. There's always the danger with seasonal outlooks too that the data is flawed leading to inaccurate modeling. That is why you look at ensembles and multiple factors instead of just sea surface temperatures, one analog, or a single teleconnection. It's my belief that this year there's far more evidence among all the factors for a tough winter as opposed to one that's weak.

To sum it up, I'm on the record for a faster start to true winter in December and one that averages somewhat colder and snowier than normal. In other words it won't be a cupcake. As always, there will be surprises and as momma used to say, "seasonal forecasting is like a box of chocolates, you never know what your getting till you get it". That's my story and I'm sticking to it until I'm proven wrong. Roll weather and white gold...TS

Hey everybody, TSwails.com needs your help. For the site to continue, I am asking those of you who use, rely on, and appreciate the work and value to consider a voluntary subscription fee of $12 dollars a year, $1 dollar a month to pay for its expenses. If I can't meet my fund raising goal the site will be discontinued in November. Please click on the secure green box below for more information on how you can make a contribution. Thanks to all, T. Swails