Wednesday, December 20, 2017

Potentially Significant Arctic Cold Outbreak To End December

Model guidance is beginning to hint at the potential for a significant blast of colder than normal air temperatures in the final days of December, and atmospheric oscillations appear to support such an outcome.

Source: PSU
We first take a look at the geopotential height anomalies at the 500-millibar level, valid in the 8-10 day forecast period. On the left-most panel, we see the ECMWF's forecast. The middle panel and right-most panel shows the GFS and CMC model forecasts, respectively. All models agree on the evolution of a ridge of high pressure blossoming over the Arctic Circle, which will force the tropospheric polar vortex to be displaced to lower latitudes.

Model guidance diverges in where the "strongest" piece of the polar vortex will end up, although with these longer-term forecasts such discrepancies are to be expected. The ECMWF model favors the strongest negative anomalies in western Europe, with lesser negative anomalies contained in southern Canada as that Arctic ridge becomes cut off from the mid-level flow. This results in a zonal jet stream (almost directly west-to-east) orientation, supported by some moderate ridging in the Pacific waters just west of the West Coast. This is the "warmest" of the three model forecasts.
The GFS and CMC agree that the Arctic ridge will not be cut off from the mid-level flow, and will instead force the jet stream far north. The CMC even goes as far as to propel a Rossby Wave (a very nice-looking, "textbook" one, at that) north through Siberia, helping to push that jet stream to the north and thus make the flow much more meridional ("wavy") in the mid-latitudes.
While the GFS keeps the deepest negative anomalies in western Europe, the absence of a cut-off flow means the jet stream buckles south and allows much colder than normal air to circulate around the northern third of the U.S. A slight ridge in the Atlantic also helps to push that lobe of the polar vortex a little further south in North America.
The CMC is undoubtedly the most severe of these three scenarios, combining a strong ridge over the Arctic Circle, a positive PNA-esque pattern and a stronger ridge in the Atlantic to shunt the deepest core of the tropospheric polar vortex into the United States, now affecting a good half of the country, at least. For the moment, I would side with the GFS solution, just as it's between the two extremes of a rather-nonchalant ECMWF forecast and a severe CMC forecast.

Source: PSU
A look at the individual GFS ensembles, as well as the mean top & center, shows a general consensus on cooler than normal weather affecting all of the country east of the Front Range by December 29th, with the coldest anomalies spotted in the Plains and Midwest, as well as southeast Canada. You see some more extreme solutions in there, as well as some more transient cold shots, and for this reason it's a good idea to go with the mean, which indicates cooler than normal weather is possible in this timeframe. Intensity will, of course, be forecasted more accurately in the coming days.

Source: PSU
By January 2nd, to kick off the new year, GFS ensembles again hint at a shot of colder than normal air for the eastern 2/3rds of the country. As you can see, however, there is considerably higher uncertainty with this one due to magnified discrepancies between ensemble members, so again we will opt to monitor this period simply for the potential of cooler than normal weather.

The models can be notorious for showing one thing while the opposite comes to fruition. Atmospheric oscillations & teleconnections appear to argue in favor of a GFS-like solution for the end of December, however.

Source: CPC
The GFS ensembles are shown above projecting the phase of the Madden-Julian Oscillation (MJO) through the first few days of the new year. Recall that the MJO is based on the strength and location of thunderstorms near the Indian Ocean and north of Australia, and each different phase signifies a different location of these thunderstorms.

Each different phase also has different effects for weather here in the United States. Note that Phases 8, 1 and 2 are most favored for cooler than normal weather in the central and eastern U.S., while Phases 4, 5 and 6 are favored for warmer weather. It just so happens that the GFS ensembles favor a Phase 8 or Phase 1 orientation of the MJO for this late December period, when models foresee colder than normal weather in the United States. It must be noted that other forecast models, such as the ECMWF and its ensembles, foresee weaker activity, to the extent that thunderstorm activity in late December is too weak to "officially" resemble a Phase 8 or Phase 1 orientation. In sum, while there are model discrepancies to be worked out, the MJO could very well be in a very supportive phase for this outlook of colder than normal weather at the end of the year.

Source: JMA
You may recall I talked about a warming event in the stratosphere nearly a month ago, which came to fruition while I was studying for finals. The results of that warming event are shown above in a graph of the 30-millibar temperature reading since September 2017, with the average value shown on that gray line.

Source: CPC
30-millibar temperature anomalies
As shown above, it seems that the main portion of that warming event did not come to pass until the early slice of the month, whereas I had expected it to occur at the end of November or perhaps the first day or two of December. Alas, if meteorology were a perfect science I wouldn't have any posts to write, because we would all know what would happen!

With the occurrence of that warming event around December 7th - December 12th, maximized in the December 10th - December 14th period, we could then extrapolate the tropospheric effects to become evident two to two-and-a-half weeks later. This puts us in roughly a December 21st through December 30th period for the first effects, which looks to verify per that three-panel forecast graphic at the top of this post. The effects could really become visible here in the troposphere around December 24th - January 1, 2018 period, going along that two to two-and-a-half week frame from the maximum warming. This general timeframe fits squarely in the eyes of model forecasts for some much colder than normal weather in the U.S., like we saw earlier in this post, and provides additional evidence that this blast of Arctic air may come to fruition, at least to some degree.

To Summarize:

- Model guidance is advocating a potentially significant outbreak of colder than normal weather in the eastern 2/3rds of the country for the last several days of December.
- While details (exact location, intensity of cold air, etc) remain to be figured out, atmospheric oscillations, including the stratosphere, appear to agree with this colder than normal weather event.
- This remains some time off. As such, while models like the GFS show high temperatures of -15 just west of Chicago, IL in this period (for example), we will see significant and dramatic change in such forecasts before we reach a more accurate projection.

Andrew

No comments: