Two distinct influences of Arctic warming on cold winters over North America and East Asia

[donsutherland1]

Abstract:

 

Arctic warming has sparked a growing interest because of its possible impacts on mid-latitude climate^{1, 2, 3, 4, 5}. A number of unusually harsh cold winters have occurred in many parts of East Asia and North America in the past few years^{2, 6, 7}, and observational and modelling studies have suggested that atmospheric variability linked to Arctic warming might have played a central role^{1, 3, 4, 8, 9, 10, 11}. Here we identify two distinct influences of Arctic warming which may lead to cold winters over East Asia or North America, based on observational analyses and extensive climate model results. We find that severe winters across East Asia are associated with anomalous warmth in the Barents–Kara Sea region, whereas severe winters over North America are related to anomalous warmth in the East Siberian–Chukchi Sea region. Each regional warming over the Arctic Ocean is accompanied by the local development of an anomalous anticyclone and the downstream development of a mid-latitude trough. The resulting northerly flow of cold air provides favourable conditions for severe winters in East Asia or North America. These links between Arctic and mid-latitude weather are also robustly found in idealized climate model experiments and CMIP5 multi-model simulations. We suggest that our results may help improve seasonal prediction of winter weather and extreme events in these regions.

 

http://www.nature.com/ngeo/journal/vaop/ncurrent/full/ngeo2517.html

[JBG]

What about winters in these areas that are anomalously warm such as 2011-2 or 1990-1, 1991-2, 1994-5, 1996-7, 1997-8, 1998-9, 2001-2 and others? How can AGW cause both cold and warm winters?

[StudentOfClimatology]

This is another chicken-egg dilemma.

It could be that "arctic amplification" is actually product of the same highly meridional patterns responsible for cooler conditions in the mid latitudes, superimposed on the background warming.

[donsutherland1]

What about winters in these areas that are anomalously warm such as 2011-2 or 1990-1, 1991-2, 1994-5, 1996-7, 1997-8, 1998-9, 2001-2 and others? How can AGW cause both cold and warm winters?

 I haven't seen the whole paper, yet.

 

Having said that, it's likely that the scenario the paper is suggesting is probabilistic. There might be some increase in the tendency for the given outcomes, but no guarantee. The kind of situation described e.g.,   anomalous warmth in the East Siberian–Chukchi Sea region may not necessarily be easy to predict, even if Arctic warming increases the probability of such an outcome. Other factors e.g., natural variability, may still result in times when that region is anomalously cool, even if the probability is tilted somewhat more toward warmth in that region than in the past.  

 

An earlier piece dealing with the situation can be found here: http://www.scientificamerican.com/article/weird-winter-weather-plot-thickens-as-arctic-swiftly-warms/. That piece makes an argument why one might deal with persistent ridges (blocking) or troughs (absence of blocking).

 

Finally, the research is continuing on issues related to the Arctic, so more papers may be forthcoming in the next few years.

[Mallow]

This is another chicken-egg dilemma.

It could be that "arctic amplification" is actually product of the same highly meridional patterns responsible for cooler conditions in the mid latitudes, superimposed on the background warming.

 

Not really--there's a decently long lag (weeks to months) involved. Anomalous warmth in certain regions of the Arctic precedes the type of cold outbreaks outlined by this and other studies (including mine, currently in the editing stage).

 

Arctic amplification likely is a result of highly meridional patterns, as you suggest, but in my research I've tended to find that these Arctic warming patterns are in large part caused by tropical convection anomalies. This process takes about one to two weeks. On the other hand, the warm Arctic seas -> cold midlatitudes mechanism described above likely involves the stratospheric polar vortex, and is probably related to Cohen's snow advance index ideas as well.

[Mallow]

What about winters in these areas that are anomalously warm such as 2011-2 or 1990-1, 1991-2, 1994-5, 1996-7, 1997-8, 1998-9, 2001-2 and others? How can AGW cause both cold and warm winters?

 

I haven't looked, but I'd suspect the particular areas of the Arctic Ocean outlined above (Barents/Kara Seas and East Siberian-Chukchi Seas) tended to be cool in those particular winters. In other words, just because the Arctic as a whole is warming doesn't mean everywhere in the Arctic is warm every winter.

[StudentOfClimatology]

Not really--there's a decently long lag (weeks to months) involved. Anomalous warmth in certain regions of the Arctic precedes the type of cold outbreaks outlined by this and other studies (including mine, currently in the editing stage).

Arctic amplification likely is a result of highly meridional patterns, as you suggest, but in my research I've tended to find that these Arctic warming patterns are in large part caused by tropical convection anomalies. This process takes about one to two weeks. On the other hand, the warm Arctic seas -> cold midlatitudes mechanism described above likely involves the stratospheric polar vortex, and is probably related to Cohen's snow advance index ideas as well.

Oh man, this is an obsessive research area for me.

I'm pretty sure the initial arctic warmth you're referring to is forced by planetary wave breaking through the tropopause, into the polar vortex from below. Most of the literature I've read suggests the wave breaking is governed by intraseasonal tropical forcing, operating under a lower-frequency boundary condition set by the QBO and solar forcing. The aggregated effect of these forcing on the Brewer-Dobson Circulation and Eddy Flux(es) may determine whether or not wave breaking activity may initiate a sudden-stratospheric warming.

In other words, meridional mass transport into the polar regions precedes the observed warming, rather than the other way around.

[donsutherland1]

Based on the piece in its entirety, the argument is as follows:

 

1. Arctic amplification has been occurring with the Arctic warming faster than at other latitudes. I looked at the GISS trends 1950-2014 and the data were as follows: Globe: +0.136°C/decade; Northern Hemisphere: +0.150°C/decade; Arctic: +0.282°C/decade.

 

2. During the 1979-1997 period, as global surface temperatures warmed, winters in Europe, East Asia, and the U.S. were also warming; the Arctic had little temperature change with some cooling.

 

3. During the 1998-2013 period, winter surface temperature trends "became conspicuously different" from the 1979-97 trend.

 

4. Arctic warming had become stronger since 1998 with stronger warming trends being observed over teh Barents-Kara and East Siberian-Chukchi sea regions.

 

5. Those regions have witnessed a marked reduction in sea ice concentration.

 

6. Reduced autumn sea ice is "generally followed" by warm Arctic temperatures in the winter, which "often leads to the so-called 'warm Arctic-cold continent' pattern forced via stationary Rossby waves."

 

Observations and model data support a link to reduced sea ice concentrations. The paper also notes that "atmospheric responses to Arctic sea-ice variations are complex."

 

The paper concludes, as follows:

 

This study shows that there are two key Arctic regions where regional warming can induce distinguishable cold winters over northern continents... These results suggest that the regional distribution of Arctic warming may provide additional predictability for intraseasonal to seasonal forecasts in the NH extratropics. These results may also provide guidance for assessing the potential risk of extreme events over regions where current seasonal prediction skill is often poor. In particular, this study suggests that the recent increased frequency of severe winters over East Asia and North America may be partly caused by recent Arctic warming.

[Mallow]

Oh man, this is an obsessive research area for me.

I'm pretty sure the initial arctic warmth you're referring to is forced by planetary wave breaking through the tropopause, into the polar vortex from below. Most of the literature I've read suggests the wave breaking is governed by intraseasonal tropical forcing, operating under a lower-frequency boundary condition set by the QBO and solar forcing. The aggregated effect of these forcing on the Brewer-Dobson Circulation and Eddy Flux(es) may determine whether or not wave breaking activity may initiate a sudden-stratospheric warming.

In other words, meridional mass transport into the polar regions precedes the observed warming, rather than the other way around.

 

As I understand it, the stratospheric connection (the QBO, SSWs, effects on the Brewer-Dobson Circulation, etc) is not necessary for the tropical convection -> Arctic warming mechanism (see the "TEAM" mechanism of Sukyoung Lee et al.). Though the stratosphere may provide a helpful preconditioning state there. Regardless, tropical convection over the proper region, by itself, is enough to amplify the climatological stationary wave, enhancing meridional flow and transporting warm, moist air into the Arctic over specific key regions. Wave breaking may be involved in amplifying the stationary wave.

 

So I completely agree with your last sentence--meridional mass transport into the polar regions does precede the observed warming. What I am (and many other researchers are) finding is that sustained warmth as a boundary condition in certain regions of the Arctic (I don't want to give too much away here, as it's an active area of research and I'd like my paper to be published first ) can excite a response in the stratospheric polar vortex (warmth in key regions leading to a breakdown of the polar vortex, and vice-versa). This effect on the stratospheric NAM then is reflected in the tropospheric NAM (the AO). So in that sense, it's less of a "meridional transport into the Arctic" question, and more of a "breakdown of the stratospheric polar vortex" question, hence the impact on the midlatitudes.

[StudentOfClimatology]

Good luck on your paper, it sounds like it's going to be a good one. I look forward to reading it. I've never looked at antecedent warmth as a significant forcing on PV structure, so maybe I'll learn something.

A lot of my research has revolved around a gradual strengthening of the Brewer-Dobson circulation since the 1990s, and the corresponding change in the structure of polar warming events (fewer Canadian warmings, more frequent wave-2 responses, etc). I wonder if some of the behavioral changes I'm seeing may be related to the factors you've uncovered.

[Mallow]

Good luck on your paper, it sounds like it's going to be a good one. I look forward to reading it. I've never looked at antecedent warmth as a significant forcing on PV structure, so maybe I'll learn something.

A lot of my research has revolved around a gradual strengthening of the Brewer-Dobson circulation since the 1990s, and the corresponding change in the structure of polar warming events (fewer Canadian warmings, more frequent wave-2 responses, etc). I wonder if some of the behavioral changes I'm seeing may be related to the factors you've uncovered.

 

It's all an area of active research, so nobody really knows what the "right" answer is. It's entirely possible I'm wrong, I'm just explaining it as far as I understand it.

[scottmartin49]

An earlier piece dealing with the situation can be found here: http://www.scientificamerican.com/article/weird-winter-weather-plot-thickens-as-arctic-swiftly-warms/. That piece makes an argument why one might deal with persistent ridges (blocking) or troughs (absence of blocking).

 

Finally, the research is continuing on issues related to the Arctic, so more papers may be forthcoming in the next few years.

Here's a link to the latest Francis & Vavrus; http://iopscience.iop.org/article/10.1088/1748-9326/10/1/014005/pdf;jsessionid=74594FFE9D8E08EEBF8A27E3BF4DBAE8.c1

 

...they're really starting to nail it down.