Network News

X My Profile
View More Activity
The new Washington
Post Weather website
Jump to CWG's
Latest Full Forecast
Outside now? Radar, temps
and more: Weather Wall
Follow us on Twitter (@capitalweather) and become a fan on Facebook
Posted at 11:30 AM ET, 02/10/2011

Don't blame the Arctic oscillation

By Greg Postel

We've all experienced prolonged stretches of unusual weather. Multi-week cold snaps and oppressive summer heat waves are part of the natural variability in our climate system. They plague populated regions of the globe, and for now, challenge the skill of our weather prediction systems.

One of the more notable stretches of anomalous winter weather in the modern meteorological record occurred just one year ago. For much of the 2009-2010 winter, the eastern half of the lower 48 endured one of the most brutal combinations of cold and snow I've seen in my lifetime. And to a somewhat lesser degree (less snow in the mid-Atlantic), the combination has repeated itself this winter.

To explain long-lasting weather extremes like the ones we've seen the last two winters, meteorological studies often invoke a phenomenon called the Arctic Oscillation (AO), or its close relative the North Atlantic Oscillation (NAO). While there is undeniably some utility in using the AO index to conceptualize the evolution of highly complex weather regimes, and to package bizarre weather patterns into a single number, we need to use caution in doing so and not blame the AO for the unusual weather.

The AO can be thought of as a fluctuation in the shape of the global circulation. The strength of the oscillation, as defined by the AO index,measures the awkwardness of the circulation's geometry by contrasting the weather in polar regions from that observed in more temperate latitudes like ours. Often times when the AO index is negative, frigid air typically found at far northern latitudes has been displaced southward, while at the same time warm air has been moved northward toward the pole.

As schematically shown in the figure below, the negative phase (on the right) is typically associated with a southward penetration of cold and snowy air over the Central and Northeastern United States as well as in Europe. Perhaps not surprisingly, the AO index measured during much of last winter was "off the charts" low. In fact, it bottomed out at values not reached since record keeping began in the middle of the last century.

Positive Arctic Oscillation (left) and negative Arctic Oscillation (right). Source: J. Wallace, University of Washington

This winter, the AO plunged again. During one stretch of remarkably low AO values that began last November and continued through December and January (see image below), much of eastern half of the United States was unusually cold while much of eastern Canada was strikingly warm.

Arctic Oscillation time series in 2010-2011. Source: Climate Prediction Center.

In the picture below, which is a temperature map averaged over the 3-week period from Dec 5-25 2010, areas in blue denote regions of colder than average temperatures and areas in yellow/red likewise indicate warm conditions. During that time, temperatures reached the 40s across Northern Quebec and the shores of Greenland, while the central Florida orange crop was iced over (many cities in Florida had their coldest December on record.).

Near surface temperature anomaly from December 5-25, 2010. Source: NOAA Earth System Research Laboratory

With the AO now reverting to its positive phase and predicted to remain so by models through late February, temperature forecasts from these same models are simultaneously predicting both the northward penetration of subtropical air into the Lower 48 and the retreat of cold air back to the Arctic.

Of course they are. In this example of circular reasoning, the positive AO and the temperature pattern are manifestations of each other. But the fact that they are identified with one another does not imply causality.

This is because the AO really isn't a mechanism that brings us these weird weather patterns. It's a symptom of them. Common wisdom has latched on to the idea that the AO is a dynamical structure- one that moves around the world and occasionally creates meteorological havoc
by producing lengthy spells of uncommon weather. It is not.

Rather, the AO is a response to, or a byproduct of, the interaction among a group of fundamental atmospheric mechanisms. On more technical terms, it can be understood without appealing to fluid dynamics, and it is not a solution to any governing mathematical equation.

The real question is what causes the atmosphere to slosh back and forth in this way that removes cold air from the poles. No one knows for sure. But when weather systems capable of long-range air displacements eject tropical warmth into polar territory and dislodge cold air from it, the effects tend to last more than a few days.

It often takes the atmosphere a while to "fix" itself because the typically observed westerly jet stream is reversed in this situation and instead blows from the east. As smaller weather systems pass through this deformed jet stream, they often just reinforce the flipped pattern.

Scientists are trying to better understand the complicated processes that create these unusual, long-lasting flow geometries, otherwise known as AO phases.

Some believe that the forcing comes from tropical weather phenomena, like El Nino or the Madden-Julian Oscillation. Matt Rogers' recently wrote about a mechanism proposed by researcher Judah Cohen linked to fall snow cover over Siberia. Andrew Freedman has written about possible connections to Arctic ice extent and global warming. Others believe the cause can be linked to the natural life cycle of really big storm systems, such as the blizzard that parts of the Northeast experienced the day after Christmas. Maybe it's all of them and while it is true that the AO index is a very useful statistical concept that gives a name to uncommon weather regimes, it does not explain how or why we get them.

By Greg Postel  | February 10, 2011; 11:30 AM ET
Categories:  Latest, Science  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Trace to 1" dusts region, some schools close
Next: Snow on ground in 49 states, almost none in air


This is a great post, thanks for sharing. I often get into this "chicken and egg" question myself.

During some of our warm winters prior to 09-10, I would often ask "why is it so warm and snowless?" And the answer would be "the NAO and AO are positive." The follow-up of course, is why are they positive? And the answer to that is a less certain combination of ENSO and other phenomena. If you delve deeper and ask "why are we in a La Nina" the answers are even less clear.

It just goes to show that what we know about weather is really just a scratch on the surface. We can use phenomena such as ENSO, NAO, and AO and models that predict them to try to predict conditions and do so with some accuracy, but when it comes down to causation - real, ultimate causation - we have a ways to go!

Posted by: jahutch | February 10, 2011 12:05 PM | Report abuse

It probably isn't El Nino--we had a very strong El Nino during the mild, windy, rainy winter of 1997/98 and only a tenth inch of snow was recorded here the ENTIRE winter.

The MJO connection is worth looking into. We just had a strong phase 8 Kelvin wave the past month or so. The wave is vanishing, the MJO is weakening--and we're losing our cold air by the weekend with about six inches of snow to go before we can reach "average" snow for Reagan. By now it's a race till the end of
"meteorological" winter, though astronomical winter still has over a month to go.

Posted by: Bombo47jea | February 10, 2011 12:13 PM | Report abuse

I really enjoyed this article. It's good to finally read something about the unusually cold weather in the lower 48 as of late without having the article go into why this proves or disproves global warming.

Thanks for sharing.

Posted by: nlcaldwell | February 10, 2011 12:15 PM | Report abuse

Nice explanation. I assume the same logic applies to North Atlantic Oscillation, that it is a measurement of some effects rather than a cause?

Posted by: eric654 | February 10, 2011 12:28 PM | Report abuse

Greg, another possible cause that some think might plays a role in the arctic oscillation is stratospheric warming events (Baldwin and Dunkerton). There are even some that think the sun may play a role in the NAO so there probably are lot of possible contributing factors that nudge the poles towards one phase or another. However, I'd also argue that once you have the arctic oscillation really strongly negative it usually lasts for awhile like last year and early this winter and that once it is in place, the blocking does play a role in where storms might track and on the potential for a big snowstorm especially when coupled with el nino. While last years combo of el nino and a strongly negative ao may not have been the cause for any one storm. the combo sure set the table and when random chance brought shortwaves towards the east coast, it raised the probability of one of waves producing a major storm way above climo. To a guy trying to forecast several weeks in advance it is a pretty nice way of explaining why the pattern might stay cold in the east for awhile. Interesting article.

Posted by: wjunker | February 10, 2011 1:25 PM | Report abuse

Wouldn't warm ocean water due to the reduced ice cap be a cause, punching a hole in the dense cold air that is behind the vortex's existance in the first place?

Posted by: ender3rd | February 10, 2011 1:46 PM | Report abuse

Hi Wjunker,

yes, strat warming events are linked to trop anomalies via the downward development of critical layers (wave activity absorption zones). Baldwin et al have done great work on this. But my point is the AO is not a physical system. It's not like MJO/ENSO. PNA is a better analogy. The persistence of one sign or another of the AO index can be related to a feedback between eddies (traveling storm systems)and the basic state (jet stream). And the AO index itself is like a thermometer that measures temperature anomalies that are, in turn, driven by mechanisms like these. It does not explain why we get what we get .


Posted by: gregpostel | February 10, 2011 2:15 PM | Report abuse

Then who should I blame? George Bush or Sarah Palin?

Posted by: getjiggly1 | February 10, 2011 3:28 PM | Report abuse

What do you think of several published studies blaming cow farts for the extreme weather patterns of recent years?

Also, since the human population as of 2010 is close to 7 billion do you believe human farts also contribute to changing global climate and weather conditions?

Posted by: montana123 | February 10, 2011 3:37 PM | Report abuse

Having just come back from an interview with Mike Halpert at NCEP today at the same time Greg posted this article I found it very interesting. Have to tune in Monday at 11 on 7 for "How did La Nina become La Bomba" but I think several points are worth mentioning, even if I don't enough time to mention all on TV. First - weather happens. Every time we have some extreme (care to define "extreme") pattern or weather event, folks want to know, "What caused that". "Why are we having these extreme snowstorms last winter and this winter"? Weather happens. Why was the AO index persistently negative for months rather than weeks? Is this yet another indicator of what appear to be longer duration patterns and is the atmosphere responding to or adjusting to the relatively rapid changes in latitudinal temperatures? Are the westerlies indeed moving farther north? Are we likely to see more cut off circulations in the mid-latitudes (hello Stu Ostro TWC) as a result and is the persistence this winter and last of the -AO somehow related to this or Siberian snow cover? Lot of theories as Mike talked about today but no firm answers. Important to note what we do know, or are very confidient we know, about weather atmosphere-ocean-land etc. links. But also important to let everyone know what we don't know. Why a persistent -AO, sure not a reason but a signal? We don't know. What if the theory of decreasing Arctic sea ice is shown to hold some water (sorry less ice) . . . increasing probabilities of -AO in the coming decades and mild winters in Greenland while the major population centers of the east get more cold winters? Whoops I've already run over time.

Bob Ryan

Posted by: rtryan1 | February 10, 2011 7:57 PM | Report abuse

Bob Ryan!!!!!!

Posted by: KRUZ | February 10, 2011 8:19 PM | Report abuse

Hi Bob

Yep ... it sure seems like there is ton we have yet to learn about why the blocking frequency changes the way it does. From a forecasting perspective, I've seen more high amplitude (blocking) episodes in the last 2 winters than I can ever remember. Makes me wonder ...

Thanks for reading and submitting a response.


Posted by: gregpostel | February 10, 2011 9:29 PM | Report abuse

so to make sure I understand you.

The AO is like a bubble of cold air at the pole that sometimes wobbles about for reasons that we don't know, BUT there are some good guesses why. Kind of like a drunken uncle at a party crashing about from the point of view of a five year old kid.

Posted by: JC17 | February 10, 2011 10:42 PM | Report abuse

Resonant forced Rossby modes?

Posted by: imback | February 10, 2011 10:43 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company