Mattias

I actually do believe that the SSW might significantly influence the tropospheric circulation in the near future, but not in the way we usually expect. The rapid recovery of the vortex, the fast transition to NAO+, downward directed EP-flux, eastward phase tilt with height and the predicted vertical wave structure of the zonal wave number 1 component are all ingridients that together indicates that downward wave reflection from the stratosphere might be involved in the transition to NAO+ in the next couple of days. However, the time scales of downward reflection events are much shorter compared to the usual SSW-effects related to downward propagation of zonal mean zonal winds. Thus, I agree that there should be good possibilities for another regime shift in the end of January or more likely in February, related to MJO and AAM forcings.

As I tweeted a couple of days ago (from our newly created weatheriscool account, https://twitter.com/wxiscool/status/1592979954874331138) I believe that downward wave coupling/reflection from the stratosphere might be involved in the interesting and complex evolution forecasted ahead. There is an area with negative vertical windshear present in the upper stratosphere which is known to be able to act as a reflective surface. In addition, the EP-flux charts show episodes of downward directed EP-flux and in high latitudes there is (in the blocked runs) forecasted an easterly phase tilt of the eddy geopotential heights (which indicate downward wave propagation). Usually reflective events are associated with NAO going positive, however that is generally when the original upward wave activity pulse comes from the climatological wave number one pattern in midlatitudes (which is then later reflected back downward in high latitudes, with decreasing geopotential heights around Greenland and rising heights on the opposite side of the pole), see Shaw and Perlwitz (2013), https://academiccommons.columbia.edu/doi/10.7916/D8N58X9W/download. However, the initial wave pulse at the moment is not from the climatological zonal wave 1 and a possible upcoming downward reflection event rather seems to act to reinforce the Scandi block and might also support a retrogression of the blocking pattern towards the North Atlantic. If downward wave reflection takes place, as I believe it does in several of the recent model runs, it is not surprising to see a strengthening SPV at the same time as the tropospheric high latitude blocking pattern strengthens. That would be consistent with the recovery of the SPV and the positive u-mean anomalies forecasted in the stratosphere and the negative u-wind anomalies forecasted in the troposphere. Also, to relate back to the previous post, the shift in position of the main SPV can be related to the downward directed EP-flux, because when the main SPV moves from North America/Greenland, towards the pole and the Asian side, it temporarily brings cold air northwards which thus leads to a negative northward heat flux, consistent with the downward directed EP-flux vectors. There is still a large uncertainty in the forecasts, but it will be interesting to see how it plays out.

I have plotted the corresponding zonal mean zonal winds from the cfs members from last week and a majority of the members now actually predicts a reversal att 60N 10 hPa in late december or early January. By the way, these kind of charts, where you interactively can add or remove forecast and analysis data are now available at http://weatheriscool.com/prod/interactiveTserie.html

That's a very good question. I do not have that much experience of the cfs, but it is of course a bit suspicious when more or less all the members behaves like they do in the chart above. I also found this link(http://www.nws.noaa.gov/ost/climate/STIP/FY11CTBSeminars/jperlwitz_092210.htm) where they confirms that there is (or at least used to be) a bias towards weaker polar vortex states in the CFS so the data should be used with great caution. In this case I think it would have been nice to use some kind of model climate (instead of "real" climate) for comparison to get a more realistic signal.

The ensemble data is from CFS, downoloaded from here: http://nomads.ncep.noaa.gov I did the chart by myself so at the moment it is only on our server, but hopefully it will soon be possible to interactively add more ensemble data to the chart at http://weatheriscool.com/index.php/stratosfaren-zonalvind-10hpa-60n-gfs-ens/

I tried to add some more members from recent cfs runs (all members from last week) and its quite clear that according to cfs the vortex will get a tough start of the season. Interesting to see such an impressive signal and will be interesting to see how it verifies.

The latest runs from both GFS and ECMWF shows a very nice and quick tropospheric respons to the SSW with rising geopotential heights in Barents and Norwegian sea and negative zonal wind anomalies propagating dowards from the stratosphere. .

Nice to see the great action in the stratosphere! Will be interesting to see the tropospheric response. Could also be fun to compare with previous events and to make the comparison easier we have added a new product to weatheriscool.com that might be of interest. The product can be found from the website or following this link: http://weatheriscool.com/index.php/interactive-comparison-gfs-era-interim/ . The product will make it possible to plot charts with geopot. and temperature data at 10 hPa from the ERA interim record and compare it side by side with a chart from the latest GFS forecast. Both charts will have the same projection, colormaps and contour intervals to make the comparison as easy as possible. At the moment only January and February data from ERA Interim is available, but December and March will be added as soon as possible. The product might also have some bugs but hopefully someone finds it useful anyway. For each GFS forecast timestep we have also tried to calculate the most similar geopotential field in the ERA interim record (based on a simple Euclidean distance calculation) and the corresponding date is written out in the box at the bottom of this images: http://weatheriscool.com/index.php/stratosfaren-temp-geopot-height-10hpa/ . The calculation is very simple, just comparing the snapshot of the predicted geopot. field for each timestep, and does not take the dynamical background and so on into account so the result is of course far from perfect but maybe it could give some hints of analouge events.

Some more pictures about the downward propagation/reflection of the wave 1 pattern that's been discussed above. By filtering out only the wave 1 component of the geopotential height field and plotting for 30 and 100 hpa in the same picture (see below) the predicted phase shift from westerly to easterly by height can be clearly seen, indicating a shift from upward to downward propagation. Contours showing 30 hPa and filled contours 300 hPa (note: scale not the same for 300 and 30 hPa, but amplitude not very important here since the phase tilt is what is meant to be shown). The shift from upward to downward propagation can also be seen by looking at a cross section for the wave activity flux. In the first figure below, showing the forecast for 22st of December (some day before the wave 1 pattern in the stratosphere culminates) there is upward flux from the troposphere to the stratosphere. A couple of days later there is instead downward directed EP-flux in the lower stratosphere (from around 60N and northwards). When the direction of the wave activity flux changes the stratospheric vortex can recover while the zonal winds in the troposphere is decelerating which is shown in the last image below showing the zonal mean zonal winds from GEFS at 10 and 300 hPa.

Just want to add that I dont think that this possible downward propagation directly will result in a favorable pattern for cold weather in Europe since the rising geopotential heights (according to the easterly phase tilt by height during downward propagation) probably will be in the "wrong" location, at least initially. But if the rising geopotential heights can spread across the arctic area as in GFS 12z it would of course be a great thing.

There is a lag between the stratosphere and the troposphere so I think the downward propagation now suggested in the models will be of the negative zonal wind anomalies seen in the stratosphere during the next couple of days, se image below. My interpretation of this situation is that the current disturbance of the stratospheric vortex is driven by upward wave activity flux and the wave 1 pattern with negative zonal wind anomalies in the stratosphere culminates in a couple of days from now. Then the models suggest downward propagation of the pattern back to the troposphere, with a deceleration of the zonal winds in the troposphere in the end of December and begining of January . This downward propagation is also related to downward wave activity flux in the lower part of the stratosphere which means that the stratospheric vortex can gain in strength from above when the upward wave activity flux fades. Its not fun to see the uptick in the zonal winds in the stratosphere but in this case it might be a price we have to pay to get something tropospherically out of the current disturbance of the stratospheric vortex, but that's just my thoughts.

GFS 12z seems to be a bit extreme, some more images shown below, but GFS ensemble also has a very strong signal with a majority of the members now going for a reversal at 10 hPa 60N. .

The gfs ensemble now goes for a reversal at 10 hPa 60N. Still very far out in the forecast and latest GFS not as extreme, but it seems to be good agreement regarding the increased wave 1 activity, latest gfs shown below. Interesting times ahead!

In the short term forecast, looking att zonal mean zonal winds and EP-fluxes, there seems to be some downward wave reflection from the lower strat with EP-flux convergence and deceleration of the zonal winds north of 60N in the troposphere. Maybe that could be contributing to the building of the greenland blocking during the next couple of days? Thats my interpretation, but since I am no expert it might be wrong. Further out in the forecast the downward reflection in the lower strat disappears and some gfs runs have shown some signals of weakening of the upper strat, but yet not much to feel very excited about.

I also think we should have some patience, the signal for the end of November and beginning of December is indeed very promising. Of course the reliability in the extended range forecasts is limited, but it should also be remembered that the predictability is highly flow dependent. Even though the models have shown some large spread lately I think there are some good reasons to believe in the quite strong signals that are shown in the outer range of the 10 days forecasts now. One reason is that we are seeing an increased amplitude of MJO in phase 6-8 which is correlated with rising heights in the Greenland area with a lag of about 10 days, which is consistent with the signal in the models. It has also been shown that the skill in the extended-long range forecasts is increased when there is a significant MJO-amplitude in the initial state, which is the case right now. Secondly the weak stratospheric vortex and the displacement towards Scandinavia and northern Russia should favor some high latitude blocking in the second half of November. The zonal winds at 10 hPa 60N today is actually weaker than all years in the Era interim record for this date! To me the evolution in the stratosphere seems to be quite similar to 2009 when we we had a strongly displaced vortex in the mid November.