Winter 2014-2015 Medium-Term Discussion

[donsutherland1]

GFS Ensembles Now Forecasting a Diving Arctic Oscillation:

 

For several days, run-to-run continuity with the Arctic Oscillation (AO) forecast was quite poor. There were alternating hints of a negaitve or positive AO in the extended range. Yesterday, there was a stronger signal for the development of a negative AO. This morning, there is even stronger agreement for such a development. Such an evolution would be consistent with the idea that a major hemispheric pattern change is imminent.

 

Below are charts depicting the forecast AO, teleconnection cases based on similar EPO, AO, PNA, and NAO forecasts (12/16-31/1981-2013), and the GFS ensemble forecasts.

 

 

 

In short, the much anticipated pattern change appears on track if the latest guidance proves accurate.

[N. OF PIKE]

Don is the Polar Vortex or anything else influencial forecast to be in a favorable position for the -NAO/-AO to drive the Colder air to our side of the pond, or more favorable for Europe.

[uncle W]

the ao forecast is encouraging today...one member goes as low as -4...The nao is also forecast to go negative in the long run...

[donsutherland1]

Don is the Polar Vortex or anything else influencial forecast to be in a favorable position for the -NAO/-AO to drive the Colder air to our side of the pond, or more favorable for Europe.

It's still a little too soon to tell, but at least some of the ensembles suggest that cross-polar flow could begin to develop for North America. Moreover, if the kind of blocking shown on the GFS ensembles develops, odds might lean against storms that cut to the Great Lakes as New Year's Day approaches.

 

Other than the likelihood that the subtropical jet will remain active and North America will be growing colder down the road, it's really too soon to be very confident about storm tracks in the extended range. Interestingly enough, one of the GFS ensemble Day 11 analogs was 12/21/2009. Another was 12/26/1966. So at least there are a few wintry ones in the proverbial mix. I'm sure we'll have a much better idea toward the end of this week.

[salseneca]

GFS Ensembles Now Forecasting a Diving Arctic Oscillation:

 

For several days, run-to-run continuity with the Arctic Oscillation (AO) forecast was quite poor. There were alternating hints of a negaitve or positive AO in the extended range. Yesterday, there was a stronger signal for the development of a negative AO. This morning, there is even stronger agreement for such a development. Such an evolution would be consistent with the idea that a major hemispheric pattern change is imminent.

 

Below are charts depicting the forecast AO, teleconnection cases based on similar EPO, AO, PNA, and NAO forecasts (12/16-31/1981-2013), and the GFS ensemble forecasts.

 

 

 

In short, the much anticipated pattern change appears on track if the latest guidance proves accurate.

Thanks Don for your great insight. 

[Srain]

Some impressive Winter dates of yesteryear showing up in the Day 11+ Analogs again. Some of these years were in the Day 8+ /Day 11+ analogs just ahead of the November Arctic event as well.

 

[donsutherland1]

Some quick evening thoughts.

 

1. I have no changes from my thinking of December 14 that the transition to a colder pattern will very likely be underway by the 17th with a duration around a week +/- a few days (#174).

 

2. The recent frontal passage kicked off the transition.

 

3. The GFS ensembles are suggesting the development of an EPO-, AO-, NAO-, PNA+ set up in the extended range. Should such a pattern evolve, one could witness a possible shot of Arctic air toward the end of December or early January.

[Minnesota Meso]

Some quick evening thoughts.

 

1. I have no changes from my thinking of December 14 that the transition to a colder pattern will very likely be underway by the 17th with a duration around a week +/- a few days (#174).

 

2. The recent frontal passage kicked off the transition.

 

3. The GFS ensembles are suggesting the development of an EPO-, AO-, NAO-, PNA+ set up in the extended range. Should such a pattern evolve, one could witness a possible shot of Arctic air toward the end of December or early January.

 

The ECMWF is suggesting the same thing with the exception that is shows a -PNA, which if I'm right would mean extreme cold through the central part of the country.

 

Edit: Which I think should pop a SE ridge?

[RONALD B]

The ECMWF is suggesting the same thing with the exception that is shows a -PNA, which if I'm right would mean extreme cold through the central part of the country.

 

Edit: Which I think should pop a SE ridge?

hmm no it want . the real question,  is  the euro right?  the other models an ensembels are saying neutral too postive . the euro has a  weel known bias it tends too favor the sw during active patterns an loses the pna some times !

[donsutherland1]

The ECMWF is suggesting the same thing with the exception that is shows a -PNA, which if I'm right would mean extreme cold through the central part of the country.

 

Edit: Which I think should pop a SE ridge?

Some ensemble members are pointing to a PNA-. I suspect that the worst of the cold outbreak will likely start in the central CONUS and then expand eastward. If some of the more extreme AO forecasts verify, a long-lasting cold period could evolve for the eastern third-to-half of the CONUS, with a few short periods of milder weather.

[Srain]

Don, I agree that the guidance remain on track for a significant pattern change during the post Christmas time frame. The Updated CPC Day 11+ Analogs are trending toward some very memorable dates that include the Major late December Arctic Outbreak of 1983. I also suspect the very chilly temperatures will settle E of the Continental Divide and progress toward the East Coast as we head into early January. If the pattern unfolds as expected, we could see a prolonged cold spell as the blocking regime matures and reloads with additional shots of very cold air settling as far south as the Lower Rio Grande Valley. If the active Southern storm track continues, bouts of wintry weather may be an issue along the Gulf Coast States.

[burgertime]

Don, I agree that the guidance remain on track for a significant pattern change during the post Christmas time frame. The Updated CPC Day 11+ Analogs are trending toward some very memorable dates that include the Major late December Arctic Outbreak of 1983. I also suspect the very chilly temperatures will settle E of the Continental Divide and progress toward the East Coast as we head into early January. If the pattern unfolds as expected, we could see a prolonged cold spell as the blocking regime matures and reloads with additional shots of very cold air settling as far south as the Lower Rio Grande Valley. If the active Southern storm track continues, bouts of wintry weather may be an issue along the Gulf Coast States.

814analog.off-1.gif

 

 

That 12/30/09 analog is interesting. Right after a big wrapped up storm and it was cold enough in the south to keep snow on the ground for two weeks. While the models don't look as cold as that Jan. I was thinking the other day of how what the models are spitting out reminded me of that period just with the overall look. The bad thing was that there was so much cold any energy that rolled through went into the meat grinder and was squashed to Cuba. Having that type of flow with a less suppressed look would spell fireworks for sure. 

[donsutherland1]

For Dr. Cohen's blog:

 

The AO will remain positive for the next 5-7 days but then turn sharply negative near or just after Christmas. This negative state is expected to remain in place through the end of December and into early January...

 

The state of the AO in the long-term (i.e., into mid-January) is uncertain given the lack of forcing from the stratosphere and current model run-to-run volatility.

 

http://www.aer.com/science-research/climate-weather/arctic-oscillation

 

While this stratospheric assessment, which continues to be supported by the ECMWF (12/18 run), remains unfavorable, Dr. Cohen does note that he thinks a sudden stratospheric warming (SSW) event is probably more likely late in January than in early January. Of course, it should also be noted that SSW events can trigger deep blocking, but they can also be a response to deep blocking. Therefore, the absence of such an event does not necessarily mean that there won't be significant blocking. It does mean that there will be greater uncertainty.

 

In short, a pattern change still looks likely to occur with possible Arctic air getting involved toward the end of December. Its duration remains somewhat uncertain and it will depend on the extent and duration of blocking that develops.

 

In any case, SSW events are very difficult to predict with any degree of reliability in the medium- or longer-term.

[GaWx]

So is the CFSv2, but at least its been validated. You're just looking for something to validate your own desire, which is foolishness.

At this point CFSv2 is forecasting a warm December generally, with a Christmas/New Year cool down, and a low confidence, warm January after that. Bet the under if you want, but the house always wins over any probabilistic interval, guaranteed.

scottmartin,

Well, so much for the warm CFSv2 Jan fcast. It has solid cold in the E US now and the coldest anomalies on the planet with 3-4 C below normal for some areas (5.4-7.2 F below normal).

http://www.cpc.ncep.noaa.gov/products/people/mchen/CFSv2FCST/monthly/images/CFSv2.NaT2m.20141220.201501.gif

Day by day evolution of Jan fcast as it gets colder over last week:

http://www.cpc.ncep.noaa.gov/products/people/mchen/CFSv2FCST/monthly/images/summaryCFSv2.NaT2m.201501.gif

[wxdudemike]

Don, what are your thoughts on impacts from the MJO in the upcoming pattern change?  Euro and GEFS seem to be going for warm phases (3-6) but fairly low amplitude.  Do you think this could have any kind "muting" effect on the cold or an effect on the duration of the upcoming pattern?

[donsutherland1]

Don, what are your thoughts on impacts from the MJO in the upcoming pattern change?  Euro and GEFS seem to be going for warm phases (3-6) but fairly low amplitude.  Do you think this could have any kind "muting" effect on the cold or an effect on the duration of the upcoming pattern?

If those MJO forecasts are right, I suspect one might see somewhat of a southeast ridge popping up. It's also plausible that the cold period eases toward mid-January for at least a time. It will certainly be interesting to see how this evolves over time.

[Ottawa Blizzard]

scottmartin,

Well, so much for the warm CFSv2 Jan fcast. It has solid cold in the E US now and the coldest anomalies on the planet with 3-4 C below normal for some areas (5.4-7.2 F below normal).

http://www.cpc.ncep.noaa.gov/products/people/mchen/CFSv2FCST/monthly/images/CFSv2.NaT2m.20141220.201501.gif

Day by day evolution of Jan fcast as it gets colder over last week:

http://www.cpc.ncep.noaa.gov/products/people/mchen/CFSv2FCST/monthly/images/summaryCFSv2.NaT2m.201501.gif

I'm keeping my fingers crossed that we have a pattern flip similar to years like 1911-1912 which saw the warm December followed by a frigid January and February.

[Isotherm]

Some hope for the hopeless as we approach Christmas...

 

 

November 2014's PDO value of +1.72 was one of only 7 years since 1950 in which the November PDO value was > +1.0, indicative of a pattern than tends to feature troughiness in the Aleutians and ridging in the US West Coast.

 

Those years were as follows:

 

1959

1976

1983

1986

1987

1993

2002

 

-100% of the ensuing winters were colder than normal nationwide; 1959-60 was the only year that featured slightly warmer than normal temps in the Northeast, and this was actually due to a strong -NAO pattern that particular winter.

 

 

-6/7 (86%) of the winters featured a -AO overall for DJF. Only 1983-84 had a +NAO/AO couplet, but the winter was still quite chilly in the Eastern US.

 

-The mean 500mb pattern was such that high latitude blocking dominated, with ridging in the west and large scale troughing in the East; a clear -EPO/-AO/-NAO regime on average

 

500mb composite of the PDO analog years:

 

 

 

The one milder exception in the Northeast US. Strong -NAO still dominated.

 

 

 

 

USA temperature departures for the years:

 

 

 

Point being, based upon the strongly positive PDO at the onset of this meteorological winter, the statistics since 1950 suggest a 0% chance for a torch / warm winter nationwide, and a 17% chance for a warmer than normal winter in the Northeast.

 

While the sample size is obviously small, the type of pattern we had/have for late autumn 2014 was one of only several years w/ strongly positive PDO regimes. If this winter turns out warm nationwide, which it shouldn't, it will be the only event to do so since 1950 within a PDO of > +1.0.

 

This encouraging statistic coupled with the fact that -AO regimes have historically coincided with strongly positive +PDOs, should provide some comfort going forward.

[thenewbigmack]

Some hope for the hopeless as we approach Christmas...

November 2014's PDO value of +1.72 was one of only 7 years since 1950 in which the November PDO value was > +1.0, indicative of a pattern than tends to feature troughiness in the Aleutians and ridging in the US West Coast.

Those years were as follows:

1959

1976

1983

1986

1987

1993

2002

-100% of the ensuing winters were colder than normal nationwide; 1959-60 was the only year that featured slightly warmer than normal temps in the Northeast, and this was actually due to a strong -NAO pattern that particular winter.

-6/7 (86%) of the winters featured a -AO overall for DJF. Only 1983-84 had a +NAO/AO couplet, but the winter was still quite chilly in the Eastern US.

-The mean 500mb pattern was such that high latitude blocking dominated, with ridging in the west and large scale troughing in the East; a clear -EPO/-AO/-NAO regime on average

500mb composite of the PDO analog years:

The one milder exception in the Northeast US. Strong -NAO still dominated.

USA temperature departures for the years:

Point being, based upon the strongly positive PDO at the onset of this meteorological winter, the statistics since 1950 suggest a 0% chance for a torch / warm winter nationwide, and a 17% chance for a warmer than normal winter in the Northeast.

While the sample size is obviously small, the type of pattern we had/have for late autumn 2014 was one of only several years w/ strongly positive PDO regimes. If this winter turns out warm nationwide, which it shouldn't, it will be the only event to do so since 1950 within a PDO of > +1.0.

This encouraging statistic coupled with the fact that -AO regimes have historically coincided with strongly positive +PDOs, should provide some comfort going forward.

If you take a look at January only however, it paints a different picture.

[WidreMann]

If you take a look at January only however, it paints a different picture.

Do it without the pre-1950 years in there. You added a bunch of extra years in that were not included in Isotherm's data set.

[thenewbigmack]

Do it without the pre-1950 years in there. You added a bunch of extra years in that were not included in Isotherm's data set.

The point of the post was to highlight the rare occurrence of such a high PDO. Why remove years which fit that criteria?

[WidreMann]

The point of the post was to highlight the rare occurrence of such a high PDO. Why remove years which fit that criteria?

For one, better data. But two, you didn't also include the other data points, which were the actual temperature anomalies and AO/NAO phase (even though they can be surmised, maybe, from the map you posted).

[thenewbigmack]

For one, better data. But two, you didn't also include the other data points, which were the actual temperature anomalies and AO/NAO phase (even though they can be surmised, maybe, from the map you posted).

We can agree to disagree then.

[psuhoffman]

Based on that January pattern feb would have to be Hoth like cold to get to isotherms map.

[Met1985]

Based on that January pattern feb would have to be Hoth like cold to get to isotherms map.

If one were to read Isotherms years you can see were the discrepancy is coming from Macks post is.  

1959

1976

1983

1986

1987

1993

These are the years that isotherm posted from the previous 50 years with a PDO of greater than 1 in November. Mack years are these. 1937, 1922,1987, 1907, 1935,2003, 1988, 1905, 1977, 1930, 1994, 1998, 1906, 1960, 1984, 1927.

[Rainman]

We can agree to disagree then.

Your map implied that singling out January changes the signal, but you altered the dataset without saying anything, lol.

[Fozz]

Yeah, tampering with the dataset without explaining is a very disingenuous way to analyze a long-range pattern.

[franklin NCwx]

If one were to read Isotherms years you can see were the discrepancy is coming from Macks post is.

1959

1976

1983

1986

1987

1993

These are the years that isotherm posted from the previous 50 years with a PDO of greater than 1 in November. Mack years are these. 1937, 1922,1987, 1907, 1935,2003, 1988, 1905, 1977, 1930, 1994, 1998, 1906, 1960, 1984, 1927.

yeah good post. It's not apple's to apple's.

[thenewbigmack]

Your map implied that singling out January changes the signal, but you altered the dataset without saying anything, lol.

Only using post 1950 dates when there are more occurances to use seems more like altering the dataset if anything.

[griteater]

Jan for the pre 1950 ++PDO Nov years added by bigmack

 

 

Jan for the post 1950 ++PDO Nov years from Isotherm