Introduction & Disclosure
My name is Randall Gates Simpson. There is no PhD after my name and I am not a PhD climate scientist. I don't consider myself a traditional "expert" on the subject of SSW's because of my lack of official credentials. I do however think I probably know quite a bit more on the subject than the average person might. I work as a television producer by day, and as a hobby, study climate
and weather on my nights and weekends and have for over 30 years.
In addition to the subject of anthropogenic climate change, I first became interested in SSW's a few years back when it became increasingly apparent to me the extreme effects that SSW's can have on Northern Hemisphere winter weather. It wasn't just their extreme effects that interested me, but rather, the fact that their specific causes were still a bit of a mystery-- and I've always loved a good mystery. Isn't that what is (or should be) at the heart and soul of any good scientist?
In this post I reveal what I think is an original synthesis that gives the full picture of some of the main causes and effects of Northern Hemisphere Sudden Stratospheric Warming events based on my readings of many other papers along with hours of my own original research. For research material I relied heavily on the use of the amazing amount of satellite derived reanalysis data as well as ground based observations. I am heavily indebted to organizations such as NASA, NCAR and NOAA for the data they have made available at my fingertips through the web.
My research took me "virtually" to the other side of the world from where I live in Colorado, to discover some of the causes of SSW events. My research has also relied on men and women scientists who have physically travelled to extremely remote and desolate locations to gather data that I have relied on to create this synthesis. I hope perhaps one day to travel myself to these remote places to see first hand what I feel are the true birthplaces of Sudden Stratospheric Warming events.
My hope in publishing my findings on the web, is that others who may have some knowledge of or interest in the dynamics I describe here, may find my "theory" interesting, but also might find all the inconsistencies and errors I've made. Self-correction in the search for truth is essential, right? Perhaps, if I'm very lucky, I've actually stumbled on to something that might provide a truly greater understanding of these remarkable natural phenomena.
Lastly, I'd like to thank Neven for allowing me to do this guest post on his amazing Arctic Sea Ice blog. His service in the cause of understanding the dramatic and rapid changes going on in the Arctic is to be highly commended.Sudden Stratospheric Warming: A Basic Overview & Description
Since their discovery by Richard Scherhag in 1951, Sudden Stratospheric Warming events(SSW's) have drawn considerable attention and research interest due to their sometimes extreme effects on Northern Hemisphere winter weather. Scherhag and others used balloons with radiosondes and rocketsondes for their research, and the rather limited amount of data, especially limited in spatial and temporal coverage, that could be gathered this way led to interesting early theories of the causes of these events-- everything from volcanoes to solar activity.
These early theories were logical based on the limited data available and also based on the amazing essential core discovery related to SSW's-- the upper and middle stratosphere over extreme high latitudes, at elevations ranging from about 25 to 45 km would warm rapidly, in the period of just a few days. What was also of course clear in these early years of research was that the warming was not coming from the troposphere directly under the warming stratosphere, but from some other source.
This is how the mystery of SSW's began. Then, with the advent of more extensive satellite coverage of the atmosphere in the late 1970's, ever more data was gathered on SSW's and the early theories were modified and abandoned as SSW's were seen to be large scale reorganizations of the winter Northern Hemisphere atmosphere, with effects spanning from the pole to the equator and from the mesosphere to the troposphere.
Here for example, is a comparison of what was happening in the polar stratosphere simultaneously with what was happening over the equator in the stratosphere during the January 2013 SSW:
So as the chart shows us, during the January 2013 SSW, just as air was falling and warming in the North Pole stratosphere, air was rising and cooling over the equator in the stratosphere-- a teleconnected effect covering some 9,000 km! No other atmospheric event of this scope and scale exists on the planet, yet more people are aware of the much smaller hurricane, cyclone, or tornado. And as you'll see, far more people are affected by a large SSW event than a large hurricane, yet the average citizen has scarcely heard of them.
Basic SSW Characteristics & Classification
Like many atmospheric phenomenons such as hurricanes and tornadoes, SSW events exist on a continuum of size, intensity and effect, even though from the largest to the smallest, they all share a basic set of features, namely:
1) They primarily take place during the Northern Hemisphere winter. Small and infrequent SSW's do occur over the south pole, but, as you'll see, there are precise reasons why they are mainly a Northern Hemisphere phenomenon, displaying yet one more reason why the planet is biased toward the advection of energy toward the North Pole versus the South Pole (but more on this later).
2) They cause a rapid rise in stratospheric temperatures over extreme northern latitudes.
3) They cause a rapid rise in stratosphere pressure over the extreme northern latitudes.
4) They cause some level of wind pattern disturbance over the pole, with the larger ones displacing, disrupting, or outright destroying the polar vortex.
Here are some examples of 3 general levels of SSW's: Small, Medium or Intermediate, and Large. First, a series of rather small SSW's between January and the end of March in 1996:
Here's a moderate SSW that occurred at the beginning of February 2001:
And finally, here's a very large SSW that took place in January 2006:
Major Winter Weather Disruptions at Lower Latitudes
If all SSW's did, was affect the upper or middle stratosphere over the pole or even the equator (as shown in the chart above), they would remain an interesting atmospheric curiosity or mystery much as Scherhag found them in 1951. What we've learned in the past few decades however, is that they have enormous and profound effects (especially the larger ones) on the entire weather patterns of the Northern Hemisphere winter. These effects begin with the descending air and increasing pressure that comes with the sinking air over the pole. Here is the pressure anomaly chart of the January 2013 SSW:
Along with this sinking air, we see the corresponding disruption of the Arctic polar vortex as shown in the zonal wind anomaly chart from winter 2013:
During a large SSW, the breakdown of the polar vortex reverses the normal westerlies in high latitudes (westerlies partially driven by the counter clockwise flow of wind around the polar vortex). When the vortex breaks down or is disrupted, the winds shift rapidly (as shown in the chart above) and come from the east. In places like northern Europe, when the winds come from the east of course that brings cold air from Siberia, stormier weather and much colder temperatures. This large scale shift of NH winds from westerly to easterly affects hundreds of millions of people from the associated extreme cold, snow, and general misery.
Though certainly declines in Arctic sea ice may be changing the patterns of NH weather, the research on these effects is ongoing. In terms of the winter of 2013, we can trace the change in England's weather this winter almost to the day of the onset of the large SSW event over the pole. If you look at the chart above for the SSW that peaked around the 6th of January, and then take a look at the Central England Temperature record for the winter, you'll see this:
SSW's and the Arctic Oscillation Index (AO)
The misery that SSW's can bring to lower latitudes due to their wind and pressure effects can be measured quite readily in the useful Arctic Oscillation Index (AO). Rather than go into a detailed explanation of the AO Index, for the purposes of this paper it suffices to say that generally high pressure over the pole relative to lower latitudes leads to a negative AO Index and lower pressure over the pole relative to lower latitudes leads to a positive AO Index. Generally speaking, when the AO Index is high, we can expect generally better and warmer weather over lower latitudes during that period in the winter, and when the AO Index goes negative or even very negative, we can expect generally nastier and colder weather over lower latitudes.
When a large SSW event occurs such as we had in 2013, warm and descending air over the pole compresses even more, warms even more, disrupts the vortex (as shown above) and turns the AO Index negative. What needs to be clear however is that the AO Index is of course a tropospheric reading, and the SSW begins in the stratosphere and the pressure effects can work their way down into the troposphere and last over a period of many weeks.
Here's a most revealing chart showing a direct chronologically aligned comparison of the timing of the early January 2013 SSW seen in the normalized zonal GPH anomaly (top) aligned with the AO Index (center), and the Central England Temperature (bottom) from January 1, 2013 through April 1, 2013:
The correlation between the three is remarkable and directly related to the initial event-- the SSW that began in early January when England's weather turned cold and a prevailing wind began to blow from the east as the vortex was shattered and the AO Index turned to the negative.
Now this mention of the association between SSW's, the disruption of the Arctic vortex, and a negative AO index is not to suggest that SSW's are the only thing that affects the AO index, just like it would be wrong to suggest that all low pressure and rain over the Gulf of Mexico is related to hurricanes. What it does show, is that when a large SSW event occurs (which is every few years on average) it will tend to dominate the Arctic weather and lower latitudes during that time period.
So now that we've taken a look at some of the effects of SSW's on the NH winter weather, as interesting and far-reaching as they can be, let's get to what I consider the even more interesting part-- the causes of SSW's.
Following the Warm Air Back to Asia
There are a few clues that one can use to trace the origin of the descending warm air over the pole, but before looking at that, let's talk about that descending air in general. Of course the rapid descent of air in the stratosphere compresses and warms, but a key clue comes from knowing that it was already anomalously warm before it began to be compressed. What warmed it and where?
Thanks to satellite reanalysis data, we can in fact see exactly where the warm air came from before reaching the polar latitudes. As this animation so excellently displays, a wave of warm air moved up from South Central Asia in the weeks prior to the SSW event that occurred in early January 2013. This wave in fact seems to almost "explode" up into the stratosphere, almost like the warming you'd get from a volcanic eruption (again, hence the reason for early theories on SSW's looking for a volcanic connection).
Again, this bubble of warm air rises up to become a thermal wave over South Central Asia on about December 21, 2012 and seems to nearly explode to the north and to the east. The rapidity with which this wave forms, indicates that some very strong dynamical event was forcing warm air up into the stratosphere from the troposphere. It then moves north and east at 10 hPa stratospheric levels and higher. Here's different view of the same thermal wave on December 28, 2013:
I consider the chart above to be what I call the "standard fingerprint" of the thermal waves that I investigated as far back as reliable reanalysis data is available. Up until now I've been using the latest 2012-2013 winter as an example of the thermal wave that originates over South Central Asia, but, before continuing to explore in much greater detail the formation of this wave, I want to show you a view of a similar wave, or "standard fingerprint", that occurred during a SSW event in the winter of 2002-2003:
Again, a similar development in the thermal wave as it develops at lower latitudes and moves north. This 10 hPa anomaly chart has been constrained at over Asia from 0E to 80E. Again, I found similar waves are seen in the weeks before every SSW event, minor or major, over the NH that I was able to find reliable data on. Using reanalysis data back to 1948, I found no example of the thermal wave forming anywhere else other than over Asia prior to NH SSW events. Now, once the wave forms and moves north it can actually rotate to the east around the pole (and even migrate back west), before penetrating the vortex at some point and descending, but in my research, the initial formation of the thermal wave was found to always be over Asia. Here's one developing before the big SSW event of January 2009:
In considering the development of these thermal waves at mid-stratospheric levels over Asia, there are several factors that must be considered. First, such a rapid build-up of rising warm air can obviously not come from evaporation as normal adiabatic cooling would occur and we would see the exact opposite condition such as we find associated with the normal convective processes in thunderstorms or on a larger scale in Hadley Cells.
The second consideration is the rapidity necessary to bring such a large mass of warm air upward into the stratosphere. Some process must essentially "launch" this warm air, as we see it seems to "explode" into the stratosphere. Finally, the air must be relatively warm to begin with (anomalously warm relatively to normal stratospheric temperatures at 10 hPa), for no matter how rapidly it is launched or how warm it was, it will expand and cool by some amount, no matter what. So in our search for the source of the thermal wave, we are looking for a source of warm air and a mechanism by which that air is launched rapidly upward into the stratosphere.
Again, It was this basic realization that led a few early researchers to consider volcanoes as a possible cause of SSW's. Other factors however did not support the volcanic theory, not the least of which was the lack of other normal signs associated with volcanoes such as elevated levels of volcanic aerosols.
Looking for High Positive Omega
The vertical rising or falling of air in the atmosphere is known as omega and is measured in Pascals per second (Pascal/s). A high positive omega over a geographic region would indicate that a large mass of air was rapidly moving upward in that region, and likewise, a negative omega would indicate that a mass of air that was falling or moving downward.
Obviously, in the formation of a thermal wave in the stratosphere over South Central Asia we would be looking for a region in the troposphere in the same proximate region with unusually high positive omega. It would have to be an area that displayed what would be much like vertical jet stream or plume of warm air moving rapidly upward in the troposphere, through the tropopause, to form the bubble or thermal wave in the stratosphere.
In looking at the omega reanalysis data for the period around the formation of the thermal wave that formed over Asia in December 2012, we find the following very interesting data that displays the average omega across the entire troposphere from 1000 hPa up to 100 hPa. This chart was further constrained to show the region of interest, namely South Central Asia from 60E to 110E longitude and 0 to 70N latitude:
In looking at this reanalysis chart we see three regions of high positive omega. These regions represent areas where air was moving rapidly upward across the entire troposphere averaged over the month. Each region is worthy of special consideration in their own right, and should be the subject of future research, but for the sake of simplicity, I chose to narrow my analysis down to the area of highest positive omega of the three, which was the one furthest north and west near 85E and 37.5N. Here's a closer view of the omega in that region during December 2012:
Again, for a clearer understanding of omega, in looking at the preceding diagram, one can imagine a stream of air moving right out of the center of the red and orange area, outward from the screen or page. This stream would be flowing up through the entire depth of the troposphere, centered at 85E and 37.5N during the month of December 2012. For the sake of comparison, here's the omega for the exact same region during the month of July 2012:
In comparing the two, we see that the high omega over this region in December 2012 truly is at least a monthly anomaly and not a year-round feature of this region. So we at least now have a potential area for a vertical stream of air moving into the stratosphere to form our thermal wave in December 2012. Let's take a closer look at this region near 85E and 37.5N.
The Taklamakan Desert, Kunlun Shan & Altun Mountains
The area of high positive omega over South Central Asia in December 2012 is centered on the Kunlun Shan mountains and the southwestern part of the Altun Mountain range. These mountain ranges collectively form the southern border of the Tarim Basin and Taklamakan Desert and mark the northern edge of the Tibetan Plateau.
Of course, it is not surprising at all to get a high omega for this area as the Kunlun Sham and Altun mountains rise rapidly from their close proximity to the Taklaman Desert and offer the potentially ideal combination of topographic features to produce strong topographically induced lift to air. Here's a roughly placed overlapping map of the high positive omega in the region during December 2012, in the weeks prior to the SSW of early January:
Note that in the previous image we clearly see the area of high omega to be the southern side of Taklamakan Desert. In the red and orange areas in this region - since this omega is averaged over the month and across the entire troposphere - a large mass of air is streaming at a high rate through topographic forcing, through the tropopause and up into the stratosphere. The prevailing winds on the ground during the winter over the Taklamakan Desert tend to be from the northwest, north, and northeast as you move west to east respectively across the region, as illustrated here (Yoshino, 1991, 1992):
The north to south direction of these prevailing winter winds across the Taklamakan Desert would possibly be a key part of the energy that serves to launch the warmer desert air off the Kunlun Shan and Altun Mountains with the right overall conditions. It is also possible, and much further research would need to occur for verification, that given the general lowering of the winter tropopause and the elevations involved in the region, that jet stream winds could dip lower in the area and be integrally involved in providing some of the energy for the topographic lift.
Topographic Profiles of the Region of High Positive Omega
We've identified the boundary between the Taklamakan Desert and the Kunlun Shan and Altun Mountains as the general region of high positive omega during the month of December 2012. It's now time to look at the region in more detail. Generally, of course, the relatively flat Tarim Basin and Taklaman Desert give way to the Tibetan Plateau as you move south. The Taklamakan Desert is typical for deserts around the planet, with tens of thousands of square kilometers of dunes that look like this:
But as you travel south, those dunes rapidly give way to the Tibetan Plateau. Here's what the transiton zone between the two regions looks like:
And to get an even better feel, here's a Calipso LIDAR image showing the dramatic rise at the boundary between the Taklamakan Desert and the Tibetan Plateau (image courtesy of NASA/LARC):
In the LIDAR image above, we see that the transition between the Taklamakan Desert and the Tibetan Plateau is, relatively speaking, very rapid-- occurring in just few tens of kilometers. The peaks along the boundary soar to well over 6,000 meters high, and represent ideal topographically induced "launch points" to send tropospheric air rapidly into the stratosphere with the proper wind conditions coming off of the Taklamakan Desert.
Here's a few topographic profiles (with the Taklamakan Desert to the right) of some the typical peaks in the region, all of them in the Kunlun Shan or Altun Mountains:
The Full Picture
SSW's are marked by rapidly descending air over the polar region. This air warms as it descends but we've seen that it was warmer than normal to begin with. Through reanalysis, we have seen that the air traveled to the high latitudes as a stratospheric thermal wave at 10 hPa. In looking at the origins of this thermal wave we find high omega areas just south of the Taklamakan Desert in the Kunlun Shan and Altun Mountains.
When prevailing north to south winter winds blowing across the Taklamakan desert (and including possible energy from jet stream winds) are directed toward the Tibetan Plateau under the right conditions, the perfect topological lifting conditions occur, allowing the warmer air from the desert to ascend rapidly into the stratosphere where it builds into a thermal planetary wave of warmer air that is carried north by prevailing upper level winds.
Upon reaching the higher latitudes of the Arctic, the warmer air begins to descend rapidly, warming even more but also carrying enough momentum during the descent to drag mesospheric air down into the stratosphere as well. This intrusion of mesospheric air during SSW events has been well documented, but is outside the scope of my research. Suffice to say it has been measured by many researchers and can be identified through its unique chemical signature. One excellent recent source for more information on this is Kvissel et al., 2012.
As illustrated earlier, the rapidly descending air in the stratosphere over the Arctic has effects or teleconnections that reach some 9,000 km away to the equator. Here we see that the momentum of the polar descending air pulls the air in the stratosphere up at the equator, such that this air cools and expands and creates a temporary temperature and pressure anomaly.
Again, here's another example of this from the large SSW event that occurred around January 21, 2009 where, in this equatorial stratosphere profile of temperature, you can readily see the effect of air being drawn upward into the stratosphere:
Other Causal Elements of SSW's: The Winter Tropopause
I have only briefly mentioned the tropopause thus far, but it should of course not be left out in a discussion of the full dynamics of SSW's. Certainly, the intrusion of a rapidly ascending stream of warm air into the stratosphere from the troposphere means that this stream must penetrate through the tropopause as it ascends.
We've talked about the unique topological circumstances with location of the Taklamakan desert next to the Tibetan Plateau. The desert provides the ideal source of warm air to be "launched" upward and the rapid and steep mountain Kunlun Shan and Altun Mountains are the perfect launch pads. With the winter winds prevailing from the desert toward the mountains is this region, it only takes the right set of circumstances to initiate the precursor events to an SSW.
We must however add the additional factors of the lower winter tropospheric height and the fact that the Kunlun Shan and Altun Mountains are at an ideal position for penetrating the tropopause with warm air from below as the tropopause drops down rapidly between 30 and 40 degrees, as seen in this simple illustration (B. Geerts and E. Linacre, 1997):
Thus, not only does the tropopause lower in the winter, it also drops down dramatically just at the point where the Kunlun Shan and Altun Mountains ranges are located and ready to provide the necessary launch platform for warm air from the Taklamakan desert. It is quite possible that had these mountain ranges been located five hundred hundred kilometers further south where the tropopause is higher, that the character of NH SSW's and NH winters might be completely different.
This is not to say that the Kunlun Shan and Altun Mountains are the only places in the region that offer a suitable high omega area, for we know there are others, and they need to be investigated and researched as well.
Other Potential Corroborating Evidence
It has been known for quite some time that Greenland ice cores contain a high level of isotopes that can be traced back specifically to the Taklamakan desert and associated Tarim basin. It was only very recently however that I became aware of this fact. Considering that the high omega topographically induced "launch sites" for warm air into the stratosphere exist just south of these regions, and that the air that the Kunlun Shan and Altun Mountains ranges launch upward, would be coming directly from the Taklamakan desert, then it stands to reason that the air would contain traces of the dust of the isotopes from these regions.
Furthermore, it stands to reason that as this air forms the thermal wave that travels north toward the Arctic to descend as a SSW event, then some of that air would naturally make its way into the Greenland glacial ice. There are of course other paths that dust and trace isotopes could take to Greenland, i.e. more westerly across the Pacific and North America, and certainly some of it does take this path, but the existence in high quantities in the Greenland ice raises an interesting potential future avenue of research whereby samples of air could be taken over high latitudes of Asia and Siberia just prior to SSW event, before 10 hPa thermal wave reaches the pole. If these samples contain isotopes from the Taklamakan desert, it would provide some confirmation of the correctness of my overall SSW model.
Summary & Conclusion
My investigation into numerous SSW events using reanalysis data has shown that in every event that I could find reliable data on, the initial locus of the warm air has come from South Central Asia. I investigated one potential source of this warm air among several-- namely at the point where the Taklamakan Desert meets the Tibetan Plateau in the Kunlun Shan and Altun Mountain ranges.
This combination provides an ideal topographically forced "launching pad" for warm air to be lifted up through the lowered winter tropopause into the stratosphere. This air is then advected as a thermal wave toward the northern latitudes where it eventually descends, warms further, and creates an SSW event. This diagram illustrates the overall dynamics of this entire process:
In addition to the extensive reanalysis data, there appears to be tangential corroborating evidence in support of this overall dynamic, such as the high level of dust in Greenland ice cores that has an origin point in the Taklamakan desert and Tarim Basin. This dust could be part of the thermal wave that descends over the Arctic during the SSW and is eventually deposited in Greenland ice.
Future research could focus on gathering mid-stratosphere air samples at 10 hPa during one of the thermal wave episodes, just prior to the SSW event. A high level of dust inside that thermal wave that could be identified through isotopic analysis as coming from the Taklamakan Desert and Tarim Basin would be strong evidence of the origin of the warm air. Other avenues of research should include the analysis of other regions of South Central Asia that also show high positive omega values in the days leading up to the formation of the 10 hPa thermal wave. It is quite possible that several high positive omega areas could be involved simultaneously or each during different SSW events.
Finally, a close analysis of the regions involved in high positive omega events should be conducted to see what meteorological conditions exist to create the wind field energy necessary for topographic lifting. In addition to pressure zones and prevailing winds, energy from the lowering of the tropopause and the jet stream during winter could be an important source of energy for high positive omega events and the precursors to SSW's.
Update #1: I came across the following chart from this site: http://arise-project.eu/atmospheric-dynamics.php
Both the chart and the Arise website are right on target with what we've been talking about related to the causes of some of the various types of waves we can find in the stratosphere and mesosphere (though their mountains are better than in my simple drawing!). In short, there are multiple sources (thunderstorms, cyclones, and yes, topography) that can send waves into the stratosphere.
It should be also be noted that vertically forced Rossby Waves have been discussed quite a bit by some posters related to SSW's, and that indeed, topological factors (i.e. the Tibetan Plateau and other high mountain ranges) can play a big role in these types of waves. Here's a quote from a paper on these types of waves that is worth reading (some rather heavy math edited out):
"Rossby waves are often forced by topography...Rossby waves cannot propagate vertically if the mean zonal winds are easterly, or if they are westerly and exceed a certain speed.This has important implications for the dynamics of the middle atmosphere (defined as the stratosphere and mesosphere). In the summertime the zonal winds in the middle atmosphere are easterly, and so energy from topographically forced Rossby waves cannot reach the middle atmosphere. In the wintertime, however, the zonal winds in the middle atmosphere are westerly, allowing Rossby waves to reach the middle atmosphere and deposit energy. This explains the sudden stratospheric warming episodes (as much as 40-50 K within a few days) observed in the Northern Hemisphere wintertime. This phenomenon is not as pronounced in the Southern Hemisphere because there are not as many topographical features in that hemisphere to generate topographically forced Rossby waves."
The full paper for this quote (with the maths) can be found here: http://snowball.millersville.edu/~adecaria/ESCI343/esci343_lesson11_rossby_waves.pdf