Stratospheric Warming event on the way?

[icebreaker5221]

The AO has averaged around +1.7 this month. The monthly average will probably finish between +1.5 and +2.0 if the current guidance is reasonably accurate. Since 1950, 11/12 (91.7%) cases that saw the December AO average +1 or above had a January AO > 0. 1979-80 was the single exception.

In January 2009, the AO averaged +0.800.

Bob Chill broadened the sample size by going to Plus 0.8 and got 15 years. 14 of the 15 averaged having a positive AO in January. The only good thing is that some of those years had one week to 10 day negative ao periods embedded within all the positive periods.

That said, the AO has never (since 1950) averaged +1.0 or greater for both Nov and Dec of the same year. So in all of the +1.0 Dec in our historical record, Dec was the first month in which the highly +AO appeared. The fact that it continued into Jan those years is not surprising, given the typical timescales associated with cold stratospheric events.

So perhaps the saving grace this year will be the fact that the AO went so positive so early.

[JoMo]

Well the AO has been mostly positive since September with a few minor dips here and there so I'm not sure it matters when it goes positive.

[icebreaker5221]

Well the AO has been mostly positive since September with a few minor dips here and there so I'm not sure it matters when it goes positive.

True, but I was only making my argument for highly +AO events, Sep this year was more of a moderately +AO.

Since we don't have any +1 (or greater) Nov followed by +1 Dec, I decided to look at +1 Dec followed by +1 Jan instead. Doing so, 4/11 cases are followed by a -AO (monthly mean) within 1 month, and 6/11 are followed by a -AO within 2 months. Still not great odds, but I just wanted to note that history would suggest that a 2 month +1 AO is more likely to break down more quickly that a 1 month +1 AO.

[bluewave]

This is even a colder look than during the big +AO during December 2006.

[Matt Rogers]

The EP Flux is trying to turn Poleward on the latest Euro:

Correction: "Latest" on the University of Berlin site (12z yesterday)

[usedtobe]

You beat me to it. The 30 mb temps are forecast to rise to aoa the seasonal normal so the vortex certainly looks like it will be weakening some.

[donsutherland1]

You beat me to it. The 30 mb temps are forecast to rise to aoa the seasonal normal so the vortex certainly looks like it will be weakening some.

I agree. While I don't necessarily believe the warming, even if it culminates in an SSW (still uncertain), will bring about the demise of a general AO+ tendency, I do believe it could lead to episodes of some blocking during January. Such an outcome would be consistent with insights from the 1975-76 and 1999-00 La Niña winters where January was the coldest month relative to normal.

With regard to January 2012, I suspect anomalies in the East would still wind up on the warm side of normal. The observed warming during recent decades, including the last decade in the East, would also probably help tilt the balance toward the warm side of the ledger. A general PNA- regime will probably be the biggest player for January.

Overall, my thinking is January will probably wind up closer to normal in much of the East than December 2011, but still warmer than normal. I also don't expect January 2012 to be the warmest January since 2006 (neutral ENSO + strong EPO+). It will probably be the warmest since January 2008 in much of the East and, in some cities, perhaps 2007 e.g., DCA will probably finish with a mean temperature of around 39°-40° for January even if the month gets off to a warm start as per my thoughts (6z GFS disagrees).

[usedtobe]

I agree. While I don't necessarily believe the warming, even if it culminates in an SSW (still uncertain), will bring about the demise of a general AO+ tendency, I do believe it could lead to episodes of some blocking during January. Such an outcome would be consistent with insights from the 1975-76 and 1999-00 La Niña winters where January was the coldest month relative to normal.

With regard to January 2012, I suspect anomalies in the East would still wind up on the warm side of normal. The observed warming during recent decades, including the last decade in the East, would also probably help tilt the balance toward the warm side of the ledger. A general PNA- regime will probably be the biggest player for January.

Overall, my thinking is January will probably wind up closer to normal in much of the East than December 2011, but still warmer than normal. I also don't expect January 2012 to be the warmest January since 2006 (neutral ENSO + strong EPO+). It will probably be the warmest since January 2008 in much of the East and, in some cities, perhaps 2007 e.g., DCA will probably finish with a mean temperature of around 39°-40° for January even if the month gets off to a warm start as per my thoughts (6z GFS disagrees).

Don, I pretty much agree with your thinking about Jan.

[VAwxman]

EP flux vectors are still mostly equatorward, although a not as strong as the first 10 days of Dec. One thing to note is that the upward flux is very strong now, which means that planetary waves are reaching the stratosphere and doing it's dirty work there...which current observations confirm. If EP flux vectors slowly change to a more poleward direction in the next 15-20 days, then probabilities of a strat->tropo propagation will be good

Hey wxmx, what is the link for these images? Thanks!

[Typhoon Tip]

It's been there for weeks, actually. It's also coupled up against an equally or even more prominent negative anomaly associated with the core PV situated over/near the NE Greenland region.

What I find ...perhaps somewhat more intriguing is that as of very recently, like 2 days ago ... the temperature progs from the GEFs mean have said cold feature warming by some 10 to 15C, while maintaining the warm region, from D4 - 7. That's a new behavior, and "might" signal a relaxation in the U-mean wind about to take place.

Need a negative U anomaly to really signal a warming event - high correlation there..

[dtk]

Hey wxmx, what is the link for these images? Thanks!

The plots are from JMA, available here:

http://ds.data.jma.g...s/clisys/STRAT/

[Amped]

Finally I can tell what a friggen Wave1 pattern looks like. Thanks everyone I've asked over the years for not explaining it.

Might as well guess on the Wave 2 pattern rather than wait 3 years for it to occur with strong a amplitude.

[WxUSAF]

That's certainly a wave 1 dominated structure with maybe a hint of wave 2. But I don't know enough to understand if that's a good thing or a bad thing for breaking up the PV from hell.

[H2Otown_WX]

For people learning about SSWs...we want to see blues around 60N/10mb in the above graph. Blues in the upper stratosphere from 50N-poleward are depicting the first stages of a stratospheric warming.

Just trying to wrap my head around this..why is it that we want an east wind for warming?

[okie333]

Just trying to wrap my head around this..why is it that we want an east wind for warming?

Because it usually means that the PV has either split or been displaced south of 60N.

[BethesdaWX]

I agree. While I don't necessarily believe the warming, even if it culminates in an SSW (still uncertain), will bring about the demise of a general AO+ tendency, I do believe it could lead to episodes of some blocking during January. Such an outcome would be consistent with insights from the 1975-76 and 1999-00 La Niña winters where January was the coldest month relative to normal.

With regard to January 2012, I suspect anomalies in the East would still wind up on the warm side of normal. The observed warming during recent decades, including the last decade in the East, would also probably help tilt the balance toward the warm side of the ledger. A general PNA- regime will probably be the biggest player for January.

Overall, my thinking is January will probably wind up closer to normal in much of the East than December 2011, but still warmer than normal. I also don't expect January 2012 to be the warmest January since 2006 (neutral ENSO + strong EPO+). It will probably be the warmest since January 2008 in much of the East and, in some cities, perhaps 2007 e.g., DCA will probably finish with a mean temperature of around 39°-40° for January even if the month gets off to a warm start as per my thoughts (6z GFS disagrees).

Not sure if I agree with this, a couple things especially regarding a potentially very cold February.

1) The +AO explosion this season occured in early to mid November, powerful +AO vortex has been in place from November to today (December 22nd). The strong Vortex was evidentin mid November and very unusual to have a strong +AO last more than 2 months which is about the limits to how long they've gone historically. No strong vortex period has ever lasted close to 3 months solid, or even 70 days for that matter. If it lasts into mid January that'll approach record levels for the existance of a strong vortex. The existance of the heavy vortex itself usually is a 2 month thing.

2) While the strong +AO has been in place for 1.5 months now, changes are evident in the stratosphere now that are signaling the rubber band snap that occured later in 1999/00, leading to a blocky March in that year, and during 1989/90 the same happened. Decembr 1989 was colder than average until the middele of the month, all starting after hurricane Hugo (Irene 2011), then flipped warm from mid December through most of Febuary. We had the strong +AO in place during early/mid November this year.

3) Note when these analog years are showing up..the 11 year pattern for the most part...1967/68, 1978/79, 1989/90, 1999/00...or 1988/89 to 1999/00. See a pattern there? An 11 year pattern? I usually look up in the sky at the bright ball of fire for answers to these things so I could be wrong in the end, but the 22 year cycle flux is, in my view, a bigger factor than many consider. All of these years feature periods of strong +AO, then a flip later on the cold season depending on then the +AO began, which is a mean of 2 months.

4) The eastern US actually cooled from the 1930s/40s into the 1970s before warming into the 2000s, then flat lining with the rest of the globe over the past decade, before the sharp cooling that should occur during the middle of this decade. Any value of significance in the global temperature IMO is insignificant to weight in a forecast at this point unless you prefer for some reason to use a 1950/1980 baseline which I'm not sure why anyone would, the winter temperature is much more variable in the manner of anomaly than the summer temperature. Shouldn't even be brought into discussion from my perspective.

So based on statistics mostly, I think we'll see a flip later in January with a solidly cold February in the East. If I'm wrong I'll bite my tongue.

[icebreaker5221]

Finally I can tell what a friggen Wave1 pattern looks like. Thanks everyone I've asked over the years for not explaining it.

Might as well guess on the Wave 2 pattern rather than wait 3 years for it to occur with strong a amplitude.

Yes, your schematic for wavenumbers 2 and 3 is correct.

Did you ask your question in the met 101 section before?

[DMADKAT]

Not sure if I agree with this, a couple things especially regarding a potentially very cold February.

1) The +AO explosion this season occured in early to mid November, powerful +AO vortex has been in place from November to today (December 22nd). The strong Vortex was evidentin mid November and very unusual to have a strong +AO last more than 2 months which is about the limits to how long they've gone historically. No strong vortex period has ever lasted close to 3 months solid, or even 70 days for that matter. If it lasts into mid January that'll approach record levels for the existance of a strong vortex. The existance of the heavy vortex itself usually is a 2 month thing.

2) While the strong +AO has been in place for 1.5 months now, changes are evident in the stratosphere now that are signaling the rubber band snap that occured later in 1999/00, leading to a blocky March in that year, and during 1989/90 the same happened. Decembr 1989 was colder than average until the middele of the month, all starting after hurricane Hugo (Irene 2011), then flipped warm from mid December through most of Febuary. We had the strong +AO in place during early/mid November this year.

3) Note when these analog years are showing up..the 11 year pattern for the most part...1967/68, 1978/79, 1989/90, 1999/00...or 1988/89 to 1999/00. See a pattern there? An 11 year pattern? I usually look up in the sky at the bright ball of fire for answers to these things so I could be wrong in the end, but the 22 year cycle flux is, in my view, a bigger factor than many consider. All of these years feature periods of strong +AO, then a flip later on the cold season depending on then the +AO began, which is a mean of 2 months.

4) The eastern US actually cooled from the 1930s/40s into the 1970s before warming into the 2000s, then flat lining with the rest of the globe over the past decade, before the sharp cooling that should occur during the middle of this decade. Any value of significance in the global temperature IMO is insignificant to weight in a forecast at this point unless you prefer for some reason to use a 1950/1980 baseline which I'm not sure why anyone would, the winter temperature is much more variable in the manner of anomaly than the summer temperature. Shouldn't even be brought into discussion from my perspective.

So based on statistics mostly, I think we'll see a flip later in January with a solidly cold February in the East. If I'm wrong I'll bite my tongue.

This is what I'm saying in the Hampton Roads thread in the mid Atlantic section about the 6 to 8 week pattern that will break in the meat of the calender. That would be mid January to early March . That is the time of year you want winter to be strongest. We had a mild November which bodes well for a good winter because if its cold in that month,that usually leads to cold coming too late in spring by time it flips back. I mentioned 1988 / 89 winter which did same thing. Mild all the way through January then BAM! Two back to back weekend snow storms of 15 inches then 9 inches with ice storm in March then 2 inches of snow again in April. This is in Norfolk Virginia area where I live. The almanac says this in its lore. " Ice in November that will bear a duck , nothing there after but sludge and muck." Makes sense when looking at the 6 to 8 week patterns. This positive AO has been rolling since early November. I'm not putting this winter to bed yet. No not at all. We will wait and see what happens. Fun stuff indeed !!

[donsutherland1]

No strong vortex period has ever lasted close to 3 months solid, or even 70 days for that matter.

Check out the 11/27/1988 through 4/14/1989 timeframe. Only 2 days had negative AO readings.

2) While the strong +AO has been in place for 1.5 months now, changes are evident in the stratosphere now that are signaling the rubber band snap that occured later in 1999/00, leading to a blocky March in that year, and during 1989/90 the same happened. Decembr 1989 was colder than average until the middele of the month, all starting after hurricane Hugo (Irene 2011), then flipped warm from mid December through most of Febuary. We had the strong +AO in place during early/mid November this year.

March could well be blocky. Right now, I'm focused on January. I do expect some periods of blocking in January, but remain uncertain whether sustained, strong blocking will set in.

4) The eastern US actually cooled from the 1930s/40s into the 1970s before warming into the 2000s, then flat lining with the rest of the globe over the past decade, before the sharp cooling that should occur during the middle of this decade. Any value of significance in the global temperature IMO is insignificant to weight in a forecast at this point unless you prefer for some reason to use a 1950/1980 baseline which I'm not sure why anyone would, the winter temperature is much more variable in the manner of anomaly than the summer temperature. Shouldn't even be brought into discussion from my perspective.

My reference was to the observed warming that has occurred in recent decades, including the most recent one. Such warming has occurred and it shows up in the reanalysis data.

Hence, in marginal situations, I take that warming into consideration in assigning probabilities. Should a temperature reversal take hold, then I'll take that into consideration, too.

[donsutherland1]

Don, I pretty much agree with your thinking about Jan.

Thanks Wes. It will be interesting to see how things evolve.

[Amped]

Yes, your schematic for wavenumbers 2 and 3 is correct.

Did you ask your question in the met 101 section before?

No it was mainly back in the Easternuswx days.

[JoMo]

Brett Anderson at Accuweather is onboard with SSW causing a pattern change around the second week of Jan.

And he just posted on his twitter about 2 hours ago that the new ECMWF weeklies support very cold air returning to western canada the week of Jan 9th.

[BethesdaWX]

Check out the 11/27/1988 through 4/14/1989 timeframe. Only 2 days had negative AO readings.

Ok, I stand corrected on that generally, though statistically it is rare to see something like that, no? In a -PDO phase, longer stretches of +AO are even rarer, in looking back I cannot find one strong +AO period lasting up to 2 months, while +PDO periods seem to feature them more frequently overall. I have my idea on the mechanism but it's only a theory.

March could well be blocky. Right now, I'm focused on January. I do expect some periods of blocking in January, but remain uncertain whether sustained, strong blocking will set in.

Ok, thanks. I deeply respect your predictions, was just curious.

My reference was to the observed warming that has occurred in recent decades, including the most recent one. Such warming has occurred and it shows up in the reanalysis data.

Hence, in marginal situations, I take that warming into consideration in assigning probabilities.

But isn't that only the case if you keep the old baseline though and don't alter it to fit the "new" normal?

Then there are winter temps over the past decade which have been cooling in the latest update, and the same circulation mechanisms don't operate from summer to winter:

http://www.esrl.noaa...mit=Create+Plot

Also, what do we define as a period of "normals"? If I select 1997 through 2002 the temp during that period is warmer than the past several years, the fluctuations are constant so it makes it more difficult to determine the true mean is over time:

http://www.esrl.noaa...mit=Create+Plot

http://www.esrl.noaa...it=Create+Plot\

Maybe you're right about everything, but I don't see much point in weighting the less than 1/2 degree of warming seen year round, nor the cooling seen during the winter.

[Clinch Leatherwood]

Don,Wes....

I have to think we're looking at mid January for anything meaningful in terms of a change.

sadly the old saying drought brings drought is somewhat applicable to what we have now. We are running so far behind on snow and the SST's are so warm, it's going to be hard to overcome.

[Geos]

Not sure if I agree with this, a couple things especially regarding a potentially very cold February.

1) The +AO explosion this season occured in early to mid November, powerful +AO vortex has been in place from November to today (December 22nd). The strong Vortex was evidentin mid November and very unusual to have a strong +AO last more than 2 months which is about the limits to how long they've gone historically. No strong vortex period has ever lasted close to 3 months solid, or even 70 days for that matter. If it lasts into mid January that'll approach record levels for the existance of a strong vortex. The existance of the heavy vortex itself usually is a 2 month thing.

2) While the strong +AO has been in place for 1.5 months now, changes are evident in the stratosphere now that are signaling the rubber band snap that occured later in 1999/00, leading to a blocky March in that year, and during 1989/90 the same happened. Decembr 1989 was colder than average until the middele of the month, all starting after hurricane Hugo (Irene 2011), then flipped warm from mid December through most of Febuary. We had the strong +AO in place during early/mid November this year.

3) Note when these analog years are showing up..the 11 year pattern for the most part...1967/68, 1978/79, 1989/90, 1999/00...or 1988/89 to 1999/00. See a pattern there? An 11 year pattern? I usually look up in the sky at the bright ball of fire for answers to these things so I could be wrong in the end, but the 22 year cycle flux is, in my view, a bigger factor than many consider. All of these years feature periods of strong +AO, then a flip later on the cold season depending on then the +AO began, which is a mean of 2 months.

4) The eastern US actually cooled from the 1930s/40s into the 1970s before warming into the 2000s, then flat lining with the rest of the globe over the past decade, before the sharp cooling that should occur during the middle of this decade. Any value of significance in the global temperature IMO is insignificant to weight in a forecast at this point unless you prefer for some reason to use a 1950/1980 baseline which I'm not sure why anyone would, the winter temperature is much more variable in the manner of anomaly than the summer temperature. Shouldn't even be brought into discussion from my perspective.

So based on statistics mostly, I think we'll see a flip later in January with a solidly cold February in the East. If I'm wrong I'll bite my tongue.

I find the 11 year pattern very interesting. These strong +AO events match up nicely to when the solar cycle intensity is running up to its peak.

[BethesdaWX]

I find the 11 year pattern very interesting. These strong +AO events match up nicely to when the solar cycle intensity is running up to its peak.

I agree. Even more significantly, try comparing the difference from one solar cycle to the next, to the NAO state and you find a nearly perfect correlation...as in, greater difference between two cycles = -NAO in the second (following) cycle, while two similar cycles = +NAO in the second (following) cycle.

Take a solar cycle, and move onto the next...if that solar cycle is similar in strength to it's prececessor, the NAO is positive, regardless of how strong the cycle is (only if the climate is at equilibrum). If there is a large difference the strength of a solar cycle to it's predecessor, in either direction, the NAO is always negative.

Example, cycle 22 was similar to cycle 21, so the NAO in cycle 22 was positive. Varying cycles from the predecessor = -NAO. +NAO = stable climate, -NAO = unstable climate, look how many "moving parts" the pattern hs with strong arctic blocking..

So varying solar cycle one to the next = -NAO, similar strength = +NAO. Of course a warmer state during higher solar cycles will be more "stable" than a colder state during weak solar cycles, kind of how like winter anomalies vary more than summer anomalies...so strong solar cycles slightly favor stronger positive NAOs than weaker solar cycles, if they're similar in strength.

Two other fascinating correlations to the Sun.

1) The PDO flips every two solar cycles, as does the AMO. -PDO beginning in 1998/99, -AMO beginning soon.

2) You can also go and correlate every ENSO event we've had to the geomag sun on a 6 year lag, the AP index shows this well. Look between 2003/04 and 2009/10 and see the ONI and global temp similarity to the AP index 6 years ago.

I have an older broder' working as an atmospheric scientist, he's a big solar guy and has gone through all this stuff with me, and it's amazing how obvious it becomes when you look into it. I am now convinced every oscillation on this globe is operated by the Sun, as is the global climate.

[Yes_Probably_Maybe_No]

Finally I can tell what a friggen Wave1 pattern looks like. Thanks everyone I've asked over the years for not explaining it.

Might as well guess on the Wave 2 pattern rather than wait 3 years for it to occur with strong a amplitude.

Judging by the chart you posted, I'm gonna guess a wave 1 pattern is simply HP vs LP battling it out near the Pole?

[OKpowdah]

Judging by the chart you posted, I'm gonna guess a wave 1 pattern is simply HP vs LP battling it out near the Pole?

Wave 1 is a displacement of the vortex from the pole. Wave 2 is a splitting of the vortex.

[donsutherland1]

Ok, I stand corrected on that generally, though statistically it is rare to see something like that, no? In a -PDO phase, longer stretches of +AO are even rarer, in looking back I cannot find one strong +AO period lasting up to 2 months, while +PDO periods seem to feature them more frequently overall. I have my idea on the mechanism but it's only a theory.

It is rare. It would not surprise me if there is some impact from the PDO (long-term cycle).

But isn't that only the case if you keep the old baseline though and don't alter it to fit the "new" normal?

Then there are winter temps over the past decade which have been cooling in the latest update, and the same circulation mechanisms don't operate from summer to winter:

In the change over the past decade, I use the 1981-2010 baseline. Also, I compare use decadal changes by month. December (warming), January (warming), February (cooling) in setting another parameter to try to gain insight into marginal/toss-up situations.

Also, when discussing my overall thoughts, my references to above/below normal readings are relative to the 1981-2010 base (most recent 30-year period). Of course, there has always been sizable variability in temperatures (monthly/seasonal). The same applies with snowfall (actually, the standard deviation for monthly/seasonal snowfall is even larger than that for temperatures).

Maybe you're right about everything, but I don't see much point in weighting the less than 1/2 degree of warming seen year round, nor the cooling seen during the winter.

In most cases, the observed decadal temperature trend is not relevant. I only use it if differences are sufficiently small that it becomes a possible factor. ENSO/ENSO evolution, teleconnections, etc., play a much larger role in constructing possible analogs to compare with the modeling/ensembles.

[Bob Chill]

I'll leave the whys to the pros because I really have no idea but these 2 AO dailies are for Nina years when Dec averaged +.8 or higher. There is one more analog year (71-72) but it's not a really good match from what we've already seen since the beginning of Nov.

The AO can be quite stubborn when it's hanging in very + territory early on. Of the 15 years I analyzed, there are more 60-80 day stetches of a genreal +AO regime than I wanted to see.

I created the charts with numerical data.The date axis was a pain in the year so it left it as a 153 day period starting on Nov 1st instead of breaking it out with month and date. Not too hard to convert months in your head. I screen snipped the curent dailies for comparison.

I'm working on charts for the NAO dailies for the same 15 years I picked out to see if anything sticks out.

Many thanks to Don S for linking me up to the daily data!