I am excerpting from Dr. Cohen’s latest post because of his refreshing candor sharing his thought processes regarding arctic weather patterns. Arctic Oscillation and Polar Vortex Analysis and Forecasts January 29, 2018 Dr. Judah Cohen. (my bolds and images added)
I have really struggled with what to discuss in today’s Impacts section and in the end decided to focus on a feature that gets no respect and to elaborate on last week’s discussion. One big problem has been large model uncertainty and lack of reliable guidance. I think it should be obvious to anyone reading the blog that I am focused on the behavior of the stratospheric polar vortex (SPV) and using variability in that behavior to anticipate large scale climate anomalies across the Northern Hemisphere (NH) on the timescale of days to weeks and even months.
The weather models (I spend most of my time analyzing the global forecast system (GFS) but I do not think that it is limited to the GFS) have been predicting some wild and highly anomalous behavior in the SPV. First the GFS was predicting a SPV displacement into North America (why this is highly anomalous is a good question and not something that I fully understand). Then the GFS predicted a strong warming in the polar stratosphere centered over Scandinavia of the magnitude that is only observed over East Asia and Alaska. The GFS has mostly backed off of these forecasts or at least predicting events of much smaller magnitudes (though it is back in the 12z run).
And looking back at the behavior of the SPV for the winter it can be summed up as unremarkable in many ways. My sense is that the SPV has been stronger than normal for the winter characterized a mostly positive stratospheric AO and cold/below normal PCHS in the stratosphere. Based on that alone one would expect an overwhelmingly mild winter across the NH mid-latitudes. However that would not be an accurate description of the winter.
But in the atmosphere you cannot have low pressure without high pressure. And as we head into the final third or half of winter, I don’t think that one can understand or explain this winter’s temperature variability without focusing on anomalous high pressure in the polar stratosphere. So in the end I have decided to discuss what I like to call the “Rodney Dangerfield” of weather- high pressure because it doesn’t seem to get the respect it deserves certainly compared to low pressure.
My passion for weather began with my love for snow and I couldn’t wait for the next snow opportunity. I grew up in New York City (NYC) where it snows every winter but to get a good snowfall is always challenging and predicted snowfalls more often than not did not materialize because of too much dry air or too much warm air or the storm being too far out to sea…. It became apparent to me having an area of low pressure passing near NYC rarely translated to a snowstorm.
Instead a better predictor of snowfall was the position of high pressure. If Arctic high pressure settled to the north of NYC across Quebec or even Northern New England the likelihood of snowfall greatly increased despite model predicted storm tracks. With high pressure entrenched to the north, good things (as far as snow falling in NYC) happened. So even though meteorologists like to focus on storms and low pressure, in my opinion the key player in whether it would snow or not was the high pressure.
This recognition of the importance of high pressure that began with my passion for weather followed me to my studies. On the regional scale of snowfall in NYC it was the storms that got all the attention and high pressure was neglected (at least that was my impression). Similarly when I began studying winter climate variability on a large scale again my impression was that the focus was on the two semi-permanent large scale low pressures the Icelandic and the Aleutian lows.
There was a third semi-permanent feature that seemed to get scant attention – the Siberian high. When my own research demonstrated a relationship between Eurasian snow cover and winter climate including in the Eastern US to me the obvious link or pathway was the Siberian high. It took many years and many studies to come to the understanding I have today (which remains incomplete) but it is my opinion that the Siberian high is the single most important large scale synoptic feature that influences the variability of the SPV (other climate scientists may disagree with me).
My own empirical observations are that when the Siberian high is shifted to the northwest over the Urals and Scandinavia region, this will inevitably produce increased energy transfer from the troposphere to the stratosphere and more often than not disrupt the SPV. The likelihood of a disruption will increase if the Ural blocking is coupled with downstream troughing across East Asia and the North Pacific or a deeper than normal Aleutian low.
Through the blog I advocate for the importance of SPV variability on sensible weather and whether the SPV is “still” or “disrupted” can have important and large implications for surface weather. As I discussed last week, from the blog it has become obvious to me thinking of the SPV as weak and strong only, or even compositing based on the absence or existence of zonal wind reversals at 60°N and 10 hPa was overly simplistic and probably missed most of the coupling with the troposphere. Instead the position of the SPV and the flow around the SPV were important regardless of the speed of the zonal winds at 60°N and 10 hPa.
But this winter makes me believe that it might even be more nuanced than even the wind flow around the SPV. The precursor to the historic cold in the Eastern US in late December and early January was a Canadian warming/high pressure in the polar stratosphere the third week of December. But as it turns out the most impressive cold anomalies during the month of January are not in North America but Asia. A second warming/high pressure near Eastern Siberia in mid-January accompanied near record cold in Siberia and large parts of Asia.
Figure 12. (a) Forecasted 10 mb geopotential heights (dam; contours) and temperature anomalies (°C; shading) across the Northern Hemisphere for 30 January – 3 February 2018. (b) Same as (a) except averaged from 4 – 8 February 2018. The forecasts are from the 29 January 2018 00z GFS ensemble.
Now a third warming/high pressure predicted back in the western hemisphere across Alaska and Northwest Canada is again a precursor for a return of cold temperatures to the Eastern US and Eastern Canada starting this week (see Figure 12). The location of high pressure/heating in the polar stratosphere is the best explanation that I have for the placement and timing of the dominant cold anomalies across the NH. I have a hard time making the same explanation based on the location of the SPV or the flow around the SPV.
Of course my reasoning is overly simplistic and the resultant weather anomalies are not limited to one factor or influence but rather a combination of many different influences or forcings. As I discussed in last week’s blog an alternative explanation being offered for the return of cold weather to eastern North America is the Madden Julian Oscillation (MJO).
Originally the models and meteorologists relying on MJO forcing predicted a mild first half of February and a colder second half of February. That forecast has changed mostly to a cold February from start to finish. I don’t think that change in the forecast can be ignored or glossed over with the change in timing as an inconsequential detail. The forecast for this week across the US is western ridge and warm with eastern trough and cold, though admittedly the cold is not overly impressive.
Based purely on the MJO the next two weeks should feature a cold Western US and a warm Eastern US opposite of the most recent forecasts. If it is cold in the Eastern US over the next two weeks it is not because of the MJO but in spite of the MJO. Currently the models are not quite sure if the MJO will make it to phase 8 but that phase is related to a warm Western US and cold Eastern US. If the cold persists until the third week of February then the MJO forcing could constructively interfere with the already cold pattern.
If the early arrival of the cold cannot be attributed to MJO forcing then what could be the reason? My explanation is something that I have discussed many times before – the models fail to correctly “propagate down” circulation anomalies from the stratosphere to the troposphere until the changes can’t be ignored. At longer leads the models did not correctly predict the return of Alaska ridging related to SPV variability but corrected at shorter leads.
Thanks Dr. Cohen for illuminating the art and science of studying the weather in its fascinating complexity. More on his forecasting paradigm at Warm is Cold, and Down is Up