Sudden stratospheric warming: could it lead to a very cold January in D.C.?

A phenomenon known as a sudden stratospheric warming (SSW) event just got underway, and this is sparking some conversation about how cold the pattern in the United States could get in January. Sustained Arctic outbreaks have occurred over the eastern half of the U.S. in the wake of several past SSW events; many times, D.C. has experienced extreme cold. The District – and the East at large – will be turning very cold late this week, though this will not be in response to abrupt changes in the stratosphere.

The thing is, this pattern does not need much help from the stratosphere to get cold in January.  As Wes Junker explained in his discussion a week ago, the Arctic Oscillation (AO) is turning negative. Combine this negative trend with a favorable setup in the northeast Pacific Ocean to transport Arctic air into the Midwest and Northeast, and January holds promise to be a colder-than-normal month in those regions, though perhaps not in D.C. (near normal may be a better bet here). Like December, we should have warmer variability that curtails the stronger cold blasts. A lack of blocking over or near Greenland – indicative of the positive phase of the North Atlantic Oscillation (NAO) – does not bode well for sustained much below-normal periods (the kind that last longer than 2-3 days).

If we were to enter a sustained period of below-to-much below-normal cold — the kind that lasts for a week or longer — it could happen in February (but it is far from a sure bet). By that time, any response from the sudden warming of high altitude air in the polar latitudes (i.e. the SSW event) would have been realized.

History has shown that a lengthy period of much below-normal temperatures typically occurs somewhere east of the Rockies following the start of a wintertime SSW, but the timing, duration and distribution of the much colder air is not always the same. The orientation and strength of blocking features over the high latitudes plays a big role in determining how much and how long Arctic air stays in place.

What is sudden stratospheric warming?

Last January, just as a SSW event was winding down, Climate Central’s Andrew Freedman explored the science behind this phenomenon. During the winter, a polar vortex is usually spinning above the North Pole. This vortex is an area of low pressure marked by very cold temperatures. The stronger this feature, the more cold air stays put in the Arctic.

Large atmospheric waves move upward from the troposphere — where most weather occurs — into the stratosphere, which is the layer of air above the troposphere. These waves, which are called Rossby waves, transport energy and momentum from the troposphere to the stratosphere. This energy and momentum transfer generates a circulation in the stratosphere, which features sinking air in the polar latitudes and rising air in the lowest latitudes. As air sinks, it warms. If the stratospheric air warms rapidly in the Arctic, it will throw the circulation off balance. This can cause a major disruption to the polar vortex, stretching it and — sometimes — splitting it apart.

The polar jet stream thrives on a balance between cold air to the north (Arctic region) and warm air to the south (mid-latitudes). However, the SSW creates an imbalance (warmth to the south and to the north); as a result, the polar jet weakens and the circulation slows down dramatically in the Arctic. The atmosphere continually works to remove imbalances, so, with the polar jet weakened, the midlatitude jet stream gathers strength. Wave-like features become deeper, both the downward dips (“troughs”) and upward bulges (“ridges”). In some cases, surface high pressure will develop and remain in place along an axis from Ontario to Greenland (reflecting the negative NAO pattern), blocking low pressure systems to the south and east across the eastern third of the U.S. In other cases, a blocking pattern will form elsewhere across the globe, with the coldest, stormiest conditions affecting Europe and/or Asia.

When will the SSW start and how strong does it look?

Current observations at the 10 mb level (a very high altitude within the stratosphere) show a strong polar vortex, but this feature is forecast to weaken and stretch out over the next week.

Temperature departure from normal at 10 mb (in the stratosphere), analyzed on Sunday night (left) and forecast for the evening of January 5 (right). The orange and red colors reflect warm air, while the blue and purple colors reflect cold air indicative of the polar vortex. (NOAA Climate Prediction Center)

According to these graphics, this SSW event does not look intense, and — importantly — it does not split the vortex apart or completely dislodge it from the North Pole. One limiting factor is this loop of 10 mb temperature anomalies (departures from normal). See the intense area of warmth (dark red colors) over the U.S. trying to work its way north into the Arctic region? The northward movement stalled around December 20, and that warm area has not made much progress since. This activity lends credence to the model idea that the SSW event may not be a big deal.

How has Washington, D.C.’s weather responded to recent SSW events?

Since 2003, there have been nine separate wintertime SSW events. Some occurred during cold, snowy winters (2002-03 and 2009-10), while others played out in the midst of warm, nearly snowless seasons (2007-08 and 2011-12). Most of the events started in January, and — to the extent that they respond to SSW — D.C.’s temperatures began to show a response about three weeks after the start of the SSW event, in most cases. One of the post-SSW periods (Feb. 4-13, 2012) was warmer than normal, and one period (Jan. 20-31, 2008) recorded temperatures very close to normal. The other seven periods turned out colder than average, with departures ranging from 1.6 degrees to 7.1 degrees below the normal value.

(Rick Grow)

After taking a quick look at the graph, it’s tempting to say that temperatures would be considerably colder than normal more often than not following a SSW event. This is a small sample size, though, and many more SSW events have occurred prior to 2003. What’s more, a long-range forecaster must continuously monitor the current state and forecast trends of various atmospheric signals because it’s one thing to have a SSW event, but quite another to have an efficient delivery mechanism for the cold air.

The following table shows the averages of three signals — the NAO, Eastern Pacific Oscillation (EPO) and AO — during the nine post-SSW periods. Both a negative AO and NAO imply blocking and warmth over the high latitudes, but because the negative NAO indicates blocking over Greenland specifically, the Eastern U.S. more frequently experiences below-normal temperatures in the winter and spring under the -NAO regime. The EPO helps describe the wave pattern over the northeast Pacific; a negative EPO is more favorable for delivering below-normal cold shots to the East, as it supports a ridge over Alaska that can direct Arctic air masses southward into Canada and the U.S.

Period-averaged daily values of the NAO, EPO and AO indices, along with Reagan National's mean temperature departure from normal for the period. Negative index values appear in blue, while positive values appear in red. (Rick Grow)
Period-averaged daily values of the NAO, EPO and AO indices, along with Reagan National’s mean temperature departure from normal for the period. Negative index values appear in blue, while positive values appear in red. (Rick Grow)

When all three signals are negative, as was the case during the 2005 and 2007 periods, our temperatures tend to drop far below normal. Oppositely, when all three signals are positive — like in 2008 —  temperatures tend to be above normal (they were actually near normal  in 2008, but the EPO was just barely positive). Either the negative NAO or EPO can dominate the signaling at times; when the negative NAO dominates in the winter or early spring, D.C. temperatures tend to be much colder compared to normal.

Sometimes, though, when a strongly negative EPO signal couples with modest polar blocking – yielding intense warmth and ridging over Alaska that hooks with warmth/ridging over the North Pole – the East experiences the core of the below-normal cold. The period from Jan. 11-Feb. 18, 2003 offers one of the best examples of this. Arctic cold collapsed southeast under the force of the massive ridge and settled into DC for five weeks. Reagan National recorded high temperatures of 36 or colder on 21 of the 39 days encompassing the period and, on 15 of those days, the mercury dropped below 22.

The top graphic shows the composite 500 mb height anomaly pattern for the period January 11-February 18, 2003. Orange and red colors on this graphic appear over Alaska, indicating a very strong ridge. Some blocking (again, the orange and red colors) extends from there to the North Pole. Very cold Arctic air flowed underneath this “block” and circulated southward into the eastern half of the U.S. (bottom). This is a Celsius scale, but, if converted to Fahrenheit, these temperature anomalies in purple are 6 degrees or more colder than normal. (NOAA/ESRL Physical Sciences Division)

Click on the links below for reanalysis graphics depicting all nine of the SSW events at 10 mb and the associated post-event temperature departure from normal patterns over the U.S.

1. January 4-24, 2003 Event    Post-Event Temperature Anomaly Pattern

2. February 15-March 5, 2005 Event    Post-Event Temperature Anomaly Pattern

3. January 8-28, 2006 Event    Post-Event Temperature Anomaly Pattern

4. December 28, 2006-January 17, 2007 Event    Post-Event Temperature Anomaly Pattern

5. January 10-30, 2008 Event    Post-Event Temperature Anomaly Pattern

6. January 22-February 8, 2009 Event    Post-Event Temperature Anomaly Pattern

7. January 20-February 6, 2010 Event    Post-Event Temperature Anomaly Pattern

8. December 28, 2011-January 17, 2012 Event    Post-Event Temperature Anomaly Pattern

9. January 5-25, 2013 Event    Post-Event Temperature Anomaly Pattern

How might the pattern respond moving forward?

Despite the upcoming SSW event and several recent examples of temperatures falling well below normal three weeks or so after an SSW event, there are a number of factors that argue against a similarly cold response in late January. First, the SSW event — as it is currently forecast — looks rather feeble. In fact, the 10 mb temperature anomaly pattern over the Arctic may take on a shape similar to that observed in January 2008 (top graphic below). During the last third of that month, much of the country – and especially the northern Rockies/Plains and Upper Midwest – experienced unseasonably cold weather. The Arctic chill just grazed us (see the composite map of the temperate anomalies for January 20-31, 2008; this is the bottom graphic below).

The second map below may closely resemble the temperature anomaly pattern we will see for January 2014, but taking into account how the entire month will average out, it looks much too cold over the Deep South and Southeast. It may be closer to the correct idea over the Ohio Valley and Mid-Atlantic, but, even here, is probably too cold. The Northeast (New York City and north) should be a bit colder, experiencing below-normal temperatures. Areas from the northern Rockies/Plains through the Upper Midwest are similarly poised to experience the coldest temperatures relative to normal.

(Both graphics courtesy of NOAA/ESRL Physical Sciences Division)


The negative EPO (Alaska ridge) looms as a threat to deliver strong cold shots to the East throughout January, and it will play a leading role in delivering the coldest air in years to the District late this week and toward Tuesday of next week. This is the one factor that, more than anything else, stands the best chance of keeping us near normal for the first month of the new year. It clearly has momentum on its side. December is on pace to post the most negative average EPO value in recorded history (since 1948). The EPO value averaged -155.6 through December 28, currently ranking ahead of the first 28 days of December 1980, which logged an average value of -154.3.

Bottom line: While it will get very cold early next month, warmer variability should enter the pattern and chip away at the large negative temperature anomaly D.C. will have built up through the first 10 days or so of January. Additional cold shots are likely through the remainder of the month, but the absence of a Greenland block (-NAO) will tend to keep below-normal periods transient. The evolving SSW event is weak and unlikely to result in a sustained much colder close to January, but, toward February, we will have to watch for any signs of the NAO attempting to turn negative. If that happens, D.C. would likely experience an unseasonably cold month.

Show Comments
Most Read Local

Heat Tracker

90-degree days year-to-date
Yearly Average
Record Most
67 (1980,2010)
Record Fewest
7 (1886,1905)
Last Year

At a Glance


74° /92°


74° /91°
Drop 80%


73° /89°


72° /86°


69° /84°
Drop 30%


70° /89°
Drop 30%
National Airport

Right Now

Next Story
Don Lipman · December 31, 2013