Arctic Oscillation and Polar Vortex Analysis and Forecasts

December 5, 2016

Special blog on winter retrospective can be found here - http://www.aer.com/winter2016

Dr. Judah Cohen from Atmospheric and Environmental Research (AER) recently embarked on an experimental process of regular research, review, and analysis of the Arctic Oscillation (AO). This analysis is intended to provide researchers and practitioners real-time insights on one of North America’s and Europe’s leading drivers for extreme and persistent temperature patterns.

I plan  on updating the weather discussion every Monday.  Next week I will be attending Fall AGU and disruption to the blog is likely.  Subscribe to our email list or follow me on Twitter (@judah47) for notification of updates.

Summary

  • The Arctic Oscillation (AO) is currently neutral and is predicted to trend positive week one and then trend negative week two.  I continue to expect large model volatility in the near term due to challenges predicting downward propagation of circulation anomalies related to the ongoing weak polar vortex (PV) event.  Model forecast are highly uncertain for next week.
  • The neutral AO is reflective of mixed pressure/geopotential height anomalies in the Arctic and across the mid-latitudes. However with positive heights over Greenland and Iceland and negative heights in the eastern North Atlantic, the North Atlantic Oscillation (NAO) is currently negative.  However the NAO is predicted to trend positive this week similar to the AO.
  • I continue to believe that variability associated with the AO does not currently best explain anticipated weather across the Northern Hemisphere (NH).  Instead I believe that the circulation associated with the polar vortex (PV) best explains the large-scale weather patterns across the NH continents.
  • The main PV center is predicted to reside over northwest Siberia for the upcoming week with a second more minor center over the United States sliding west into the eastern North Pacific with a Canadian high-pressure center separating the two.
  • In the near term this favors cold temperatures across Northern Asia and for Arctic high pressure to build across Alaska and northwestern Canada, with pieces of the Arctic high predicted to slide into the Western US over the next two weeks.
  • We heavily rely on October Eurasian snow cover in producing our winter forecast.  Typically the peak hemispheric response to snow cover variability is mid-January to mid-February.  This winter I believe the peak response will be from early to late December, or six weeks early.  This introduces greater than normal uncertainty in our winter forecast based solely on snow cover.

Impacts

It is my opinion that for the past two months the hemispheric atmospheric circulation experienced a nearly textbook response to extensive Siberian snow cover as outlined in Cohen et al. (2007).  I hope to provide support for that claim shortly either in the blog our on twitter (@judah47).  But as I discussed previously, there can be too much of a good thing, at least pertaining to the forecast, and that the response to snow cover variability was so efficient that the cycle of atmospheric responses is greatly accelerated.  Typically the hemispheric lower tropospheric response to snow cover occurs in mid-January and peaks in the weeks from mid-January through mid-February.  However this winter, I would argue that the lower tropospheric atmospheric response initiated the first few days of December and will likely unfold through the end of the month.  I would argue that there is unlikely to be a direct dynamic atmospheric response to snow cover variability in the months of January and February, creating additional uncertainty to the winter forecast.

The best analog that I can think of is the winter of 2009/10, when also the atmospheric response to a rapid advance in Eurasian snow cover was accelerated and peaked in December  (for a diagnosis of that winter see Cohen et al. 2010).  In that winter, a second sudden stratospheric warming (SSW)/PV weakening event occurred in February that helped the forecast to verify.  There are two important differences this winter compared to that winter.  The first is that winter was an El Niño winter and this winter is La Niña.  That favors a northwestward shift of the coldest temperature anomalies from the Southeastern US to the Northern Plains and Central Canada, which can be seen in the different forecasts from winter 2010 and this winter.  A second difference between the two winters, in 2009/10 the quasi-biennial oscillation (QBO) was easterly and this winter it is westerly (though it’s behavior is highly anomalous).  An easterly QBO is thought to favor SSWs and a westerly QBO to inhibit SSWs.  Therefore it is possible that a second SSW is less likely this winter compared to 2010, though SSWs occur in both phases of the QBO.

In the short term, it does appear that the drawn out PV weakening peaked in late November.  With a relatively quiet period in upward Wave Activity Flux (WAFz)/poleward heat flux predicted in the near term, I expect that the circulation anomalies associated with the PV weakening to episodically descend to the lower troposphere throughout the month of December.  This should favor through the month a mean negative AO (though a classic -AO pattern is looking less likely and not suggested by the recent SSW) and cold temperatures widespread across the NH continents.  Based on the recent and predicted near term PV behavior, that includes Siberia and Western Canada and the Northwestern US.

In the near term, with the primary stratospheric PV center over Northwest Asia followed by northerly flow as the PV recovers closer to the Pole, favors cold temperatures across Siberia and Northwestern Asia.  High pressure over Northern Canada and a secondary stratospheric PV center over the US favor building Arctic high pressure across northwest Canada and colder temperatures across the Northern US.  As the stratospheric PV recovers closer to the Pole, predicted cross-polar flow into Western Canada supports persistent cold in Western Canada.

For East Asia northerly flow in the mid-troposphere, favors episodic outbreaks of cold air from Siberia.  For Europe, the recent wind direction around the main stratospheric PV center over northwest Asia has been northerly resulting in the recent period of cold temperatures.  However as the PV center recovers towards the Pole, the wind direction is predicted to turn more westerly, which is a milder wind direction for Europe.  For next week positive geopotential height anomalies centered directly over Europe is predicted, which effectively blocks mild maritime air from reaching Western Asia but not Western and Central Europe and therefore milder temperatures are predicted for Europe.  However the center of positive geopotential height anomalies is predicted to migrate north of Europe after mid-month, which is a more favorable position for the eventual return of colder temperatures.

Recent and Very Near Term Conditions

The AO is currently neutral (Figure 1), reflective of mixed geopotential height anomalies both across the Arctic and across the mid-latitudes (Figure 2).  Geopotential height anomalies are currently positive across Greenland and Iceland with negative geopotential height anomalies near Portugal/Azores (Figure 2), and therefore the NAO is negative.  This is in contrast to recent winters when the NAO has consistently remained more positive than the AO.

Figure 1. (a) The predicted daily-mean AO at 10 hPa from the 00Z 5 December 2016 GFS ensemble. (b) The predicted daily-mean near-surface AO from the 00Z 5 December 2016 GFS ensemble. Gray lines indicate the AO index from each individual ensemble member, with the ensemble-mean AO index given by the red line with squares.

Circulation in the troposphere reflective of the evolving stratospheric PV will dominate the weather pattern across Eurasia.  The main stratospheric PV center is currently spinning over northwest Asia and the deepest negative departures of mid-tropospheric geopotential height anomalies lie directly underneath over Northwest Asia (Figure 2).  Upstream strong ridging/positive geopotential height anomalies centered between Northern Europe and Iceland (Figure 2) has blocked moist maritime flow from Northern and Central Europe.  These are ideal conditions for radiational cooling and below normal temperatures.   However during the week high pressure will settle south and the winds will turn more westerly resulting in a rebound in temperatures across the northern half of Europe (Figure 3).   Undercutting the Northern European ridging, troughing/negative geopotential height anomalies to the south centered over Spain is contributing to more seasonable temperatures (Figure 2) and heavy rainfall.  This will result in seasonable to below normal temperatures for much of northern Eurasia (Figure 3).  Troughing/negative geopotential height anomalies stretched across most of Northern Asia (Figure 2) and northerly flow will help to persist in seasonable to below normal temperatures for much of Northwestern Asia and Siberia (Figure 3). One exception will be Eastern Siberia.  Due to its proximity to positive geopotential height anomalies centered in the Bering Sea (Figure 2), will result in above normal temperatures (Figure 3). Much of southern Eurasia will be dominated by ridging/positive geopotential height anomalies (Figure 2). Therefore temperatures will average seasonable to above normal for most of Southern Asia including East Asia (Figure 3).

Figure 2. 500 mb geopotential heights (dam; contours) and geopotential height anomalies (m; shading) on 5 December 2016 at 00Z. Note the high heights in eastern North America, between Greenland and Scandinavia, Southern Asia and much the North Pacific with low heights over Spain, Western Asia, Siberia and western North America.

Warm temperatures and high pressure over northern Canada in the polar stratosphere is reflected in the troposphere with ridging/positive geopotential height anomalies  centered near the Aleutians.  This is forcing a deepening trough/negative geopotential height anomalies in Southwestern Canada and the Northwestern US (Figure 2) and well below normal temperatures (Figure 3). To the south and east of the trough/negative geopotential height anomalies, ridging/positive geopotential height anomalies are located across the Southwestern US and eastern North America (Figure 2).  Positive geopotential heights will result in mostly above normal temperatures for much of the Southwestern US and eastern Canada (Figure 3).  However colder air filtering in from the US will keep temperatures more seasonable in the Eastern US (Figure 3).

Figure 3. Forecasted surface temperature anomalies (°C; shading) from 6 – 10 December 2016. Note the warm temperatures across much of Eastern Canada, Central Europe and China with cold temperatures in Alaska, Western Canada, the Northwestern US and Siberia. The forecast is from the 00Z 5 December 2016 GFS ensemble.

Cold temperatures will support new snowfall across Siberia and Northwestern Asia but a milder pattern will cause snowmelt in Northern and Eastern Europe (Figure 4). A turn to colder temperatures for Southern Canada and the Northern US will support mostly new snowfall across the region (Figure 4).

Figure 4. Forecasted snow depth anomalies (mm/day; shading) from 6 – 10 December 2016. Note the snowfall over parts of Northern and Western Asia, Canada and the Northwestern US with snowmelt in Eastern Europe. The forecast is from the 00Z 5 December 2016 GFS ensemble.

Near-Term

1-2 week

The AO is predicted to remain positive for next week (Figure 1).  The positive AO is a result of mixed geopotential height anomalies across the Arctic basin and mostly positive geopotential height anomalies across the mid-latitude ocean basins (Figure 5a). However with neutral to positive geopotential height anomalies predicted near Iceland and Greenland, the NAO is predicted to remain closer to neutral this period.

Figure 5. (a) Forecasted average 500 mb geopotential heights (dam; contours) and geopotential height anomalies (m; shading) across the Northern Hemisphere from 11 – 15 December 2016. (b) Same as (a) except averaged from 16 – 20 December 2016. The forecasts are from the 5 December 00z GFS ensemble.

The main stratospheric PV center near northwest Asia coupled with persistent strong ridging/positive geopotential height anomalies stretching from Central Europe to Greenland, will anchor deep troughing/negative geopotential height anomalies across Western Asia, centered on the Ural Mountains (Figure 5a). This pattern will favor above normal temperatures for much of Europe and below normal temperatures for Western Asia (Figure 6).   The strong troughing across Western Asia will help force ridging/positive geopotential height anomalies downstream over Eastern Asia (Figure 5a). Under the influence of positive geopotential height anomalies, temperatures will average above normal for Central and Eastern Siberia and much of China and Southeast Asia (Figure 6).  Northerly flow upstream of a closed low in the Sea of Okhotsk (Figure 5a) will bring seasonable to below normal temperatures for Northeast Asia (Figure 6).

Figure 6. Forecasted surface temperature anomalies (°C; shading) from 11 – 15 December 2016. Note the warm temperatures across much of the Southwestern US, eastern Canada, Europe, Siberia and China with cold temperatures in Western Asia, Alaska and Western Canada. The forecast is from the 00Z 5 December 2016 GFS ensemble.

Coupled with strong positive geopotential height anomalies across the central North Pacific, strengthening northerly flow and deepening negative geopotential height anomalies in the stratosphere over Northern Canada and Alaska will contribute to similar troughing/negative geopotential height falls across Northwestern Canada and the Northwestern US in the mid-troposphere (Figure 5a).  With maritime Pacific air blocked from penetrating into Western North America coupled with northerly flow downstream from the central North Pacific ridge, below to possibly well below normal temperatures for Western Canada and the Northwestern US are predicted (Figure 6).   A strong Jet Stream along the US-Canadian border will transport modified Arctic air into the Eastern US where temperatures will average below normal (Figure 6).  Relatively mild temperatures will persist in far Eastern Canada (Figure 6) where southwesterly flow is predicted (Figure 5a).  Positive geopotential height anomalies and a fast Jet Stream will transport milder air into the Southwestern US where temperatures will average above normal (Figure 6).

Figure 7. Forecasted snow depth anomalies (mm/day; shading) from 11 – 15 December 2016. Note the new snowfall over Siberia, Northern and Eastern Europe, Southern Canada and the Northern US. The forecast is from the 00Z 5 December 2016 GFS ensemble.

Northerly flow across Scandinavia, Eastern Europe and Western Asia should transport some colder air to better support new snowfall across these regions (Figure 7). Overall lowering geopotential heights in Canada and the Northern US and expanding cold air will favor new snowfall across Southern Canada and the Northern US (Figure 7).  This is important because the deeper the snow cover across Canada and the Northern US, the more likely are cold temperatures to persist across the region.

3-4 week

Mostly positive pressure/geopotential height anomalies are predicted to cover most of the Eastern and Western Arctic but not the Central Arctic. This coupled with mixed pressure/geopotential height anomalies across the mid-latitude ocean basins this period (Figure 5b), contributes to a near neutral AO (Figure 1). Positive pressure/geopotential height anomalies are predicted for Greenland and Iceland that may bias the NAO negative.

Little change is predicted in the circulation pattern over Eurasia with a deep trough/negative geopotential height anomalies over Northwestern Asia (Figure 5b) and positive pressure/geopotential height anomalies on either side of the trough over Europe and Eastern Asia (Figure 5b). Mild temperatures near the highest geopotential height anomalies in Northern Europe and Eastern Siberia will sandwich below normal temperatures extending across from Western Asia to Central Siberia (Figure 8). The positive height center is predicted to migrate further north compared to the previous period which may allow a more northeasterly flow across Southern Europe and cooling temperatures (Figure 8).  A persistent polar low in the Sea of Okhotsk (Figure 5b), will help to drive some colder Siberian air into Northeast Asia (Figure 8).  A zonal Jet Stream pattern will keep temperatures close to seasonable across Southern Asia (Figure 8).

Figure 8. Forecasted surface temperature anomalies (°C; shading) from 16 – 20 December 2016. Note the warm temperatures for far Eastern Canada and Northern Europe with cold temperatures in much of Northern Asia and western Canada and the Eastern US.   The forecast is from the 00Z 5 December 2016 GFS ensemble.

Ridging/positive geopotential height anomalies just west of Alaska and or in the Eastern North Pacific are predicted to help persist northerly flow and negative geopotential height anomalies across Canada (Figure 5b).  Therefore mostly below normal temperatures are predicted for Alaska, Canada and the Northwestern US (Figure 8).  Building positive geopotential height anomalies building in from Greenland will help to persist relatively milder temperatures across far Eastern Canada (Figure 5b).  Predicted rising geopotential height anomalies across the Southern US could result in seasonable to above normal temperatures for the Southern US (Figure 8).

Figure 9.  Forecasted snow depth anomalies (mm/day; shading) from 16 – 20 December 2016. Note the snowfall over Scandinavia, Eastern and Central Europe, Siberia, Western Asia, Southern Canada and the Northern US. The forecasts are from the 00Z 5 December 2016 GFS ensemble.

New snowfall is predicted across northern Eurasia, though no strong signals emerge.  New snowfall could potentially expanded further west across Europe (Figure 9).  Across North America new snowfall is predicted near the core of the Jet Stream in Southern Canada and the Northern US (Figure 9).

Longer Term

30–day

The latest plot of the tropospheric polar cap geopotential height anomalies (PCHs) shows normal to positive/warm PCHs in both the troposphere and the stratosphere (Figure 10).  The warm PCHs in the mid-stratosphere peaked at the end of November consistent with the current negative stratospheric AO trending slowly positive through the third week of December (Figure 1).

Figure 10. Observed and predicted daily polar cap height (i.e, area-averaged geopotential heights poleward of 60°N) standardized anomalies. The forecast is from the 00Z 5 December 2016 GFS ensemble.

The peak in warm PCHs centered on November 26th is followed by a suggested downward propagation of the warmest PCHs.  This downward propagation is referred to as “dripping paint” because it does not descend all at once but episodically in fingers with alternating positive/neutral or negative PCHs.  One downward finger or drip occurred the first week of December followed by a return of normal PCHs predicted for this week.  The duration and amplitude of these pulsing or on again/off again downward propagation of warm PCHs is highly variable and poorly simulated by the weather models.  Therefore I anticipate continued model volatility into the foreseeable future.

Figure 11. Observed and predicted daily vertical component of the wave activity flux (WAFz) standardized anomalies, averaged poleward of 40-80°N. The forecast is from the 00Z 5 December 2016 GFS ensemble.

As I wrote last week the predicted tropospheric circulation for next week and beyond is highly anomalous with simultaneous Aleutian and Icelandic ridging and troughing centered near the Ural Mountains.  This is as close as you can get to a long wave pattern that is 180° out of phase with the climatological wave pattern.  Strong destructive interference between the two should result in below normal WAFz for the next two weeks, consistent with the forecast for below normal upward pulse of Wave Activity Flux (WAFz) or poleward heat transport (Figure 11).

Figure 12. (a) Forecasted 10 mb geopotential heights (dam; contours) and temperature anomalies (°C; shading) across the Northern Hemisphere for 4 – 8 December 2016. (b) Same as (a) except averaged from 9 – 13 December 2016. The forecasts are from the 00Z 5 December 2016 GFS operational model.

Reduced upward WAFz should help the stratospheric PV return to a more normal state with the PV located close to the Pole along with cold temperatures (Figure 12).  However the PV is not predicted to return to a completely quiet and circular configuration but rather a more oblong shape.  This will support cross-polar/northerly flow for Western North America and Western Asia.  And over the next two weeks these are the two regions that the weather models are predicting the largest negative departures in temperatures.  This pattern is reminiscent, to me at least, of January and February 2014 when the PV was elongated to produce northerly flow in central North America and Central Asia.  There is also an interesting kink in the PV predicted near Europe.  It might be interesting to monitor to see if eventually troughing develops over Europe in the troposphere as well.

Going forward there is much uncertainty and much will depend on how much the PV both in the stratosphere and the troposphere strengthen.  With a quieter period predicted in the WAFz, I would expect strengthening in the stratospheric PV in the near term.  If the strengthening PV later extends lower down to the troposphere, then a prolonged mild period could emerge.  But regardless, I consider the next few weeks as the period with the highest probability of a relatively cold period across the NH to occur, followed by a relatively milder period.  If the milder period sets up the next PV weakening event relatively efficiently, then confidence for the winter forecast to verify increases while procrastination of a subsequent PV weakening increases the probability of an overall milder winter.

Surface Boundary Conditions

Arctic Sea Ice

Arctic sea ice growth accelerated this week but still at a record low extent for the date. The greatest negative sea ice anomalies remain on the North Atlantic side in the Barents-Kara Seas (Figure 13). This trend should continue and if the predicted weather pattern verifies, sea ice growth could continue to grow more rapidly on the North Pacific side of the Arctic.  Recent research has shown that regional anomalies are important and the sea ice region most highly correlated with the winter AO is the Barents-Kara seas region where low Arctic sea ice favors a negative winter AO. Given that sea ice is running well below normal if not record low extent, this currently favors a strengthened Siberian high, and a weakened polar vortex/negative AO this upcoming winter.  With such strong negative anomalies in the Barents-Kara Seas, Arctic sea ice anomalies are ideally positioned to force a cold Eurasia/warm Arctic pattern and this winter will be a nice test of the theory.   I am expecting that low sea ice in the Barents-Kara seas will favor another PV weakening this winter.  With anticipation that sea ice anomalies will diminish on the North Pacific side of the Arctic, make Arctic sea ice anomalies less favorable for forcing cold temperatures in eastern North America.   Though regardless of sea ice extent anomalies, in the near term ridging near Alaska favors downstream troughing over Canada and the lower 48.

Figure 13. Observed Arctic sea ice extent on 4 December 2016 (white). Orange line shows climatological extent of sea ice based on the years 1981-2010. Image courtesy of National Snow and Ice Data Center (NSIDC). Snow and Ice Data Center (NSIDC).

SSTs/El Niño/Southern Oscillation

Equatorial Pacific sea surface temperatures (SSTs) continue to be weakly cooler than average (Figure 14) and most winter ENSO forecasts are for weak La Niña conditions. Though SSTs along the equatorial Pacific have warmed recently adding doubt to whether the forecasts of La Niña will verify.  La Niña conditions favor a negative Pacific/North American (PNA) pattern that produces cold anomalies in the northwestern US and warm anomalies in the Southeastern US.   Winter has begun with the PNA firmly negative.  In the extratropics the most dramatic change over the past month has been the rapidly cooling SSTs across the mid-latitudes of the North Pacific.  These cool waters seem to be a result of the very cold temperatures that developed across Siberia this October and November.  The cold air across Siberia both being advected out across the North Pacific and strengthening the westerlies across the North Pacific that have dramatically cooled SSTs in the North Pacific.  The cooler SSTs have erased the southern edge of the persistent warm “blob” in the eastern North Pacific, which could paly a role in ridging in the Eastern North Pacific this winter.  Finally the cool SSTS in the mid-latitudes surrounded by warm SSTS on either side more closely resemble the positive phase of the Pacific Decadal Oscillation (PDO), which also favors the opposite pattern of La Niña.

Figure 14. The latest weekly-mean global SST anomalies (ending 3 December 2016). Data from NOAA OI High-Resolution dataset. The tropical Pacific shows La Niña SST structure with cool waters near the equator in the eastern and central tropical Pacific. Warmer than normal waters also extend into the subtropical North Pacific and near Alaska with cooler than normal waters from Northern Japan to British Columbia.  Well above normal waters extend across the subpolar North Atlantic near Greenland and north of Iceland.

Figure 15. Past and forecast values of the MJO index. Forecast values from the 00Z 5 December 2016 ECMWF model. Yellow lines indicate individual ensemble-member forecasts, with the green line showing the ensemble-mean. A measure of the model “spread” is denoted by the gray shading. Sector numbers indicate the phase of the MJO, with geographical labels indicating where anomalous convection occurs during that phase. Image source: http://www.atmos.albany.edu/facstaff/roundy/waves/phasediags.html.

I also note that northern North Atlantic SSTs are warmer, with cold SST anomalies weaker and shifted south compared to the same time last year. In fact the cold “blob” is nearly gone.  Potentially significantly warmer SSTs this winter could favor opposite conditions this upcoming winter with a weakened longitudinal temperature gradient, a weakened Jet Stream, a negative NAO and a colder Europe.  And so far this fall, the NAO has had a negative tendency.  But it is still very early and there are many other complicating factors still unknown. Also air and ocean temperatures remain near record levels across the NH and without dynamically forced cold (i.e., a weak polar vortex), the streak of warm weather will persist.  The colder North Pacific SSTs and the warmer North Atlantic SSTs show how sensitive extratropical oceans are to atmospheric forcing while arguments of the ocean forcing the atmosphere are more uncertain.

No phase of the Madden Julian Oscillation (MJO) is clearly favored over the next two weeks (Figure 15).  Therefore little impacts from the MJO are expected in the near term. 

Northern Hemisphere Snow Cover

Snow cover across Eurasia remains near decadal highs. In contrast, North American snow cover advance has been slow and recently achieved a record low extent for the date.  However with expanding cold air across Canada and the Northern US, North America snow cover is likely to advance more rapidly in the near term.  Also the predicted teleconnection forecast of a negative PNA and NAO favors a rapid expansion of snow cover across North America.

Snow cover advance across Eurasia continued consistently above normal for the entire month of October.  Also because much of the advance has occurred at latitudes south of 60°N, the snow advance index is also well above normal.  Above normal snow cover extent, especially south of 60°N, favors a strengthened Siberian high, cold temperatures across northern Eurasia and a weakened polar vortex/negative AO this upcoming winter followed by cold temperatures across the continents of the NH.