Stratosphere Temperature Watch 2014/2015

[Yarmy]

Posted this on the solar activity thread, but I think it has relevance here too:

 

http://www.photonicsonline.com/doc/nature-communications-solar-activity-impacts-polar-ozone-0001

[bluearmy]

I stand to be corrected but I believe most of the seasonal models have upper strat modelling included? Recretos seems to indicate in his post that I am wrong here so I await more info

[neige57]

I stand to be corrected but I believe most of the seasonal models have upper strat modelling included? Recretos seems to indicate in his post that I am wrong here so I await more info

 

oh yeah? but then how a model can you know whether there will be a warming (high or low) in the stratosphere 2/3 months before maturity if it included this factor in? It's just not possible so a model we know what it's worth, nothing

[blizzardof96]

Heres the latest sunspot number trends from the BOM Australia site. Notice the recent spike in the september number after a first max back in Feb/March of this year.

2014 82.0 102.8 92.2 84.7 75.2 71.0 72.5 74.7 87.6

 

Although sunspots have been slowly decreasing since early October, if we do see another spike I found some interesting trends. In the 4 years with an ONI between 0.3 and 1.1 in the DJF trimonthly period, a -30mb QBO and a Jan monthly SSN at or above 100 we get this result at 500mb.

 

Notice the -NAO/AO

 

March was particularly cold in both the S UK and North America

 

30mb Stratospheric Temperatures

 

Also found it interesting how 3/4 of these years had a CW in November:

 

1958-59 had a CW in Nov/Dec with a FW in March. 

1968-69 had a CW in November.

1969-70 had a SSW in January.

1979-80 had a CW in November and a FW in March.

 

This increase in sunspots is a possibility but a low one at this point. I'll add some of the other analogs I'm watching and see what we get at 500mb. The main focus of my years was -QBO,monthly SSN between 40-90, weak el nino or warm neutral conditions while considering other factors like the PDO, Walker Cell, AMO, KOE,ONI trends etc(with some obvious exceptions within some of the analogs). There were some really interesting results for the UK especially in Dec-Jan.

 

Dec: 

 

Jan is strongly -WPO,-NAO/AO,-PNA,-Ve over the UK:

 

Feb:

 

March:

 

Strat was very warm overall with strong wave one.

GPH

30mb Air Temperature

 

Cheers from Canada and hopefully we can share some of the snow this winter,

Blizzardof96

[gottolovethisweather]

Heres the latest sunspot number trends from the BOM Australia site. Notice the recent spike in the september number after a first max back in Feb/March of this year.

2014 82.0 102.8 92.2 84.7 75.2 71.0 72.5 74.7 87.6

 

Although sunspots have been slowly decreasing since early October, if we do see another spike I found some interesting trends. In the 4 years with an ONI between 0.3 and 1.1 in the DJF trimonthly period, a -30mb QBO and a Jan monthly SSN at or above 100 we get this result at 500mb.

 

Notice the -NAO/AO

ybOs5WJv2f.png

 

March was particularly cold in both the S UK and North America

B0E1PnTCUAA9xXy.png

 

30mb Stratospheric Temperatures

omr3G3K9AP.png

 

Also found it interesting how 3/4 of these years had a CW in November:

 

1958-59 had a CW in Nov/Dec with a FW in March. 

1968-69 had a CW in November.

1969-70 had a SSW in January.

1979-80 had a CW in November and a FW in March.

 

This increase in sunspots is a possibility but a low one at this point. I'll add some of the other analogs I'm watching and see what we get at 500mb. The main focus of my years was -QBO,monthly SSN between 40-90, weak el nino or warm neutral conditions while considering other factors like the PDO, Walker Cell, AMO, KOE,ONI trends etc(with some obvious exceptions within some of the analogs). There were some really interesting results for the UK especially in Dec-Jan.

 

Dec: 

410bx5Ew7H.png

 

Jan is strongly -WPO,-NAO/AO,-PNA,-Ve over the UK:

El2uYLAiGp.png

 

Feb:

ZddM5zlVt6.png

 

March:

wC1pjGbyi0.png

 

Strat was very warm overall with strong wave one.

GPH

E4OTaeF0C1.png

30mb Air Temperature

baURW7S4o_.png

 

Cheers from Canada and hopefully we can share some of the snow this winter,

Blizzardof96

 

Greetings and many thanks for your analysis. You certainly received far too much snow than your quota should allow you last Winter as the UK and for that matter, vast parts of Europe received next to nothing. This time, it will be ours, well maybe. 

Edited October 16, 2014 by gottolovethisweather

[neige57]

Heres the latest sunspot number trends from the BOM Australia site. Notice the recent spike in the september number after a first max back in Feb/March of this year.

2014 82.0 102.8 92.2 84.7 75.2 71.0 72.5 74.7 87.6

 

Although sunspots have been slowly decreasing since early October, if we do see another spike I found some interesting trends. In the 4 years with an ONI between 0.3 and 1.1 in the DJF trimonthly period, a -30mb QBO and a Jan monthly SSN at or above 100 we get this result at 500mb.

 

Notice the -NAO/AO

ybOs5WJv2f.png

 

March was particularly cold in both the S UK and North America

B0E1PnTCUAA9xXy.png

 

30mb Stratospheric Temperatures

omr3G3K9AP.png

 

Also found it interesting how 3/4 of these years had a CW in November:

 

1958-59 had a CW in Nov/Dec with a FW in March. 

1968-69 had a CW in November.

1969-70 had a SSW in January.

1979-80 had a CW in November and a FW in March.

 

This increase in sunspots is a possibility but a low one at this point. I'll add some of the other analogs I'm watching and see what we get at 500mb. The main focus of my years was -QBO,monthly SSN between 40-90, weak el nino or warm neutral conditions while considering other factors like the PDO, Walker Cell, AMO, KOE,ONI trends etc(with some obvious exceptions within some of the analogs). There were some really interesting results for the UK especially in Dec-Jan.

 

Dec: 

410bx5Ew7H.png

 

Jan is strongly -WPO,-NAO/AO,-PNA,-Ve over the UK:

El2uYLAiGp.png

 

Feb:

ZddM5zlVt6.png

 

March:

wC1pjGbyi0.png

 

Strat was very warm overall with strong wave one.

GPH

E4OTaeF0C1.png

30mb Air Temperature

baURW7S4o_.png

 

Cheers from Canada and hopefully we can share some of the snow this winter,

Blizzardof96

 

 

superb analysis and I totally agree with you. Now it only remains to wait for the value of the OPI index and will be fixed

[Skin]

Looking forward to reading all of your posts, everyone!  This has been a go-to thread for much of my understanding of SSWs. 3rd Winter with you guys and cant be more excited! 

 

PS: Can this thread also cater to United States folks as well? the internet knows no boundaries haha. Maybe include some updates as to what both sides of the pond can expect. 

[Mark Bayley]

Interesting nature review on the role of snow cover and Arctic sea ice in forcing recent trends to more extreme weather in the mid-latitudes. The links to stratospheric warmings is covered in brief, although being a review paper it summarises the pathways and links of Arctic change to recent mid-latitude weather extremes. This paper is very accessible (i.e. not over technical)

 

http://web.mit.edu/jlcohen/www/papers/Cohenetal_NGeo14.pdf

Edited October 16, 2014 by Mark Bayley

[gottolovethisweather]

Looking forward to reading all of your posts, everyone!  This has been a go-to thread for much of my understanding of SSWs. 3rd Winter with you guys and cant be more excited! 

 

PS: Can this thread also cater to United States folks as well? the internet knows no boundaries haha. Maybe include some updates as to what both sides of the pond can expect. 

 

Of course and I had better get back to lurking as I am far behind in the knowledge stakes than a great many of the contributors in here and I could seen as derailing the thread a wee bit. I will however continue reading all the reports from everyone all across the world. 

 

Welcome aboard and please send us UK and European residents something of note this time around, we can be a jealous lot you know. 

Interesting nature review on the role of snow cover and Arctic sea in forcing recent trends to more extreme weather in the mid-latitudes. The links to stratospheric warmings is covered in brief, although being a review paper it summarises the pathways and links of Arctic change to recent mid-latitude weather extremes. This paper is very accessible (i.e. not over technical)

 

http://web.mit.edu/jlcohen/www/papers/Cohenetal_NGeo14.pdf

 

Saved and downloaded, Thank you Mark.

Edited October 16, 2014 by gottolovethisweather

[chionomaniac]

Heres the latest sunspot number trends from the BOM Australia site. Notice the recent spike in the september number after a first max back in Feb/March of this year.

2014 82.0 102.8 92.2 84.7 75.2 71.0 72.5 74.7 87.6

 

Although sunspots have been slowly decreasing since early October, if we do see another spike I found some interesting trends. In the 4 years with an ONI between 0.3 and 1.1 in the DJF trimonthly period, a -30mb QBO and a Jan monthly SSN at or above 100 we get this result at 500mb.

 

Notice the -NAO/AO

ybOs5WJv2f.png

 

March was particularly cold in both the S UK and North America

B0E1PnTCUAA9xXy.png

 

30mb Stratospheric Temperatures

omr3G3K9AP.png

 

Also found it interesting how 3/4 of these years had a CW in November:

 

1958-59 had a CW in Nov/Dec with a FW in March. 

1968-69 had a CW in November.

1969-70 had a SSW in January.

1979-80 had a CW in November and a FW in March.

 

This increase in sunspots is a possibility but a low one at this point. I'll add some of the other analogs I'm watching and see what we get at 500mb. The main focus of my years was -QBO,monthly SSN between 40-90, weak el nino or warm neutral conditions while considering other factors like the PDO, Walker Cell, AMO, KOE,ONI trends etc(with some obvious exceptions within some of the analogs). There were some really interesting results for the UK especially in Dec-Jan.

 

Dec: 

410bx5Ew7H.png

 

Jan is strongly -WPO,-NAO/AO,-PNA,-Ve over the UK:

El2uYLAiGp.png

 

Feb:

ZddM5zlVt6.png

 

March:

wC1pjGbyi0.png

 

Strat was very warm overall with strong wave one.

GPH

E4OTaeF0C1.png

30mb Air Temperature

baURW7S4o_.png

 

Cheers from Canada and hopefully we can share some of the snow this winter,

Blizzardof96

Hi Ethan,

 

I would agree with much of what you have posted. The analogues that you choose are pretty much the best ones for conditions likely this winter. I think that I would be surprised to see a super cooling of the strat during December like last year with the associated Vortex Intensification as described by Lorenzo Polvani. More the case of a warmer than average early to mid winter strat as your analogues show. Pointers are really suggesting that this could lead to a GH forming early in winter. Whether or not this occurs I guess we need to keep a close eye on the stratosphere temps come November. The one thing that struck me regarding 2009/2010 winter is the wave breaking into the vortex from Greenland. Now this is not often seen or demonstrated in many papers but is almost the ideal scenario for the UK and will help the east side of the states no doubt. If you remember this winter the stratospheric vortex split from the bottom up (from 100hPa to say 10 hpa in the most extreme splits) and this is completely different from the more traditional type of split seen in a SSW which starts with wave breaking in the troposphere rebounding to the top of the strat around the 'surf zone' before propagating downwards from 1 hPa if the EP flux plays ball.

 

I just wonder whether we will see the Greenland type of split vortex this year.

 

Mind you we are already seeing the top of the strat programmed to take a wave 1 hit even before the end of October - far more encouraging than last year already is forecast at T+240 and not too shabby either.

 

 

Why is it that I think that we should ignore the seasonal model forecasts this year? - they are so far out of sync with the background teleconnective analogues.

 

Ed

Edited October 16, 2014 by chionomaniac

[bluearmy]

the biggest difference from last year is that the 'jam tomorrow ' warming at the top of the strat has come down to day 10 as you illustrated. With further wave breaking likely from trop forcing, we do appear to be in quite a different place to last year.

Something looks like it will have to give wrt the winter forecasting

[Recretos]

 

Why is it that I think that we should ignore the seasonal model forecasts this year? - they are so far out of sync with the background teleconnective analogues.

 

 

I couldn't have said it better myself. 

And as far as seasonal models go, this was one of the points I was trying to say, about telleconnections, and also their effects/correlations with the stratospheric development. When you have a "normal" or a more stable stratosphere, with less activity or no SSW the models might be right, but when it is unstable, many things can happen, that are beyond the low resolutions of seasonal models (low resolutions relative to the complexity of the events). Seasonal models can simulate SSWs, but on a questionable basis and with questionable feedbacks. 

 

Seasonal models normally have stratosphere included. But, as far as modelling of it goes, they are inferior to the general models like GFS and ECMWF, and even those have a bit of a accuracy problem if we go into deeper FI. 

Stratosphere is a complex thing, tightly connected to the troposphere dynamics. It is not an entity on its own. So when it comes to seasonal forecasts, well, you see, if you get the tropospheric picture wrong, you gonna have the stratosphere wrong, and you might get the winter wrong. So that is one part why seasonal models tend to get the patterns at least a bit wrong. Modelling the troposphere for 2 weeks ahead is challenging, but modelling it months in advance gives you a wide room for error and you can easily undershoot or overshoot important activity for the stratosphere. So seasonal models can easily miss possible MT events that can lead to wave breaking, and since that is a complex thing on its own, it depends on the model physics how it handles any of these possible events that it does create, and even if it handles that, how will it then handle the possible development in the strat. and if it does and makes an SSW, how will it handle the possible feedbacks down into the troposphere, if it has the tropospheric patterns wrong to begin with, if we are looking 2-3 months ahead, where the tropospheric skills are low.

 

As far as seasonal models go, yes they have vertical layers in the stratosphere. The most advanced by far, is the S4 from ECMWF, with 91 layers and a model top at 0.01mb. The rest are much weaker than that. I would dare to say that if we could look at skill scores of seasonal models, we might get a bit lower scores in years with SSWs. Just an assumption.

 

We will see, as always, but I also have a feeling that the models might not be giving as much importance to the stratospheric "imbalance" that is kinda shaping up in the current projections. Of course if it stays as projected, and as far as teleconnections go. 

 

Regards.

 

Edit: 

In case someone wonders about the ECM32, it has 62 vertical levels, with the model top at 5mb. 

The ECMWF ensembles have 91 vertical levels with a model top at 0.01mb (S4 uses the same vertical schematics).

 

Now GEFS has a model top at 2mb, but around December, it is said to get an upgrade of resolution, both horizontal and vertical. So after the upgrade it would have 64 vertical layers with the model top at 0.1mb. Looking forward to that if it happens. 

Edited October 17, 2014 by Recretos

[MattH]

Echoe the sentiments of others. Models are models, whether it's a model looking at 24 hours ahead or 3 months ahead. Clearly even short term models can be incorrect and these seasonal models from the ECMWF have been wrong in times past too, they were nowhere near picking up on the cold and blocking pattern that dominated in December 2010 for example. You can't ignore the initial signs and signals this year that's for sure. As ever it's like trying put a jigsaw puzzle together, if you can find the corner pieces and the edges, then you're well on the way, if you get the analogy just by reading this. However, it would be an even more difficult pill to swallow if this winter turned out anything like last winter after all these initial signs and signals we are looking at. Charts like the below shouldn't be 'sniffed at' whether at this early juncture or whether in late December...

 

http://users.met.fu-berlin.de/~Aktuell/strat-www/wdiag/figs/ecmwf1/ecmwfzm_ha1_f144.gif

http://users.met.fu-berlin.de/~Aktuell/strat-www/wdiag/figs/ecmwf1/ecmwfzm_ta1_f192.gif

http://users.met.fu-berlin.de/~Aktuell/strat-www/wdiag/figs/ecmwf1/ecmwfzm_vt1_f192.gif

http://users.met.fu-berlin.de/~Aktuell/strat-www/wdiag/figs/ecmwf1/ecmwfzm_u_f144.gif

 

As Ed mentioned an a recent post it'll be the progress through November that will be crucial as ever. As we found out last year if the vortex gets 'bedded in' and there's little or nothing to stop it from strengthening then you can almost write-off December as being cold to start with and when I say cold, I obviously mean a synoptic pattern that delivers some 'proper' cold weather to the UK, not a 2 day N'ly behind an eastward moving low. All these initial signs and signals need to continue to develop and come together in the next 4 to 6 weeks to hopefully stop the 'beast above' from becoming anything like it was at the end of November last year.

 

Regards to all,

 

Matt.

[bluearmy]

Is there an archive of zonal wind flow charts in the strat? The Berlin forecasts look quite tepid but i wonder how unusual they are for the time of year.

[chionomaniac]

Is there an archive of zonal wind flow charts in the strat? The Berlin forecasts look quite tepid but i wonder how unusual they are for the time of year.

One can historically get the averaged mean zonal winds from this site Nick to look at previous years.

 

http://acdb-ext.gsfc.nasa.gov/Data_services/met/ann_data.html

 

I have been keeping an eye on this and the mean winds at 10 hPa are dropping off and this is set to filter down the strat.

[Recretos]

Or if you know how, you can get the data from datasets like ECMWF or MERRA and plot it manually like I do usually.

Bluearmy if you need any specific years, tell me and I can make the vertical zonal mean plots similar to FU Berlin.

Edited October 18, 2014 by Recretos

[bluearmy]

Or if you know how, you can get the data from datasets like ECMWF or MERRA and plot it manually like I do usually.

Bluearmy if you need any specific years, tell me and I can make the vertical zonal mean plots similar to FU Berlin.

Not important R, just interested to see how unusual it is in late October to see the zonal winds dropping in intensity in the upper strat coupled with the movement of the strongest flow at the very top away from northern latitudes towards the tropics.

[lorenzo]

The wave 1 activity is pretty unusual in the 35 years recorded in the Merra dataset. Only 6 years even hold a candle to 2014 with similar wave activity during October, even then not as strong as current projections on the Berlin site.

[bluearmy]

The wave 1 activity is pretty unusual in the 35 years recorded in the Merra dataset. Only 6 years even hold a candle to 2014 with similar wave activity during October, even then not as strong as current projections on the Berlin site.

That's the kind of stuff that's interesting to know tony.

I suspect there is some head scratching going on in Exeter regarding what lies ahead with the contrary forecast/telecons. anyone have any 'info' on how the meto strat model sees the upcoming few weeks? snowballz usually has some insight though she may not be able to divulge it.

[Frozen Britain]

Hi Ethan,

 

I would agree with much of what you have posted. The analogues that you choose are pretty much the best ones for conditions likely this winter. I think that I would be surprised to see a super cooling of the strat during December like last year with the associated Vortex Intensification as described by Lorenzo Polvani. More the case of a warmer than average early to mid winter strat as your analogues show. Pointers are really suggesting that this could lead to a GH forming early in winter. Whether or not this occurs I guess we need to keep a close eye on the stratosphere temps come November. The one thing that struck me regarding 2009/2010 winter is the wave breaking into the vortex from Greenland. Now this is not often seen or demonstrated in many papers but is almost the ideal scenario for the UK and will help the east side of the states no doubt. If you remember this winter the stratospheric vortex split from the bottom up (from 100hPa to say 10 hpa in the most extreme splits) and this is completely different from the more traditional type of split seen in a SSW which starts with wave breaking in the troposphere rebounding to the top of the strat around the 'surf zone' before propagating downwards from 1 hPa if the EP flux plays ball.

 

I just wonder whether we will see the Greenland type of split vortex this year.

 

Mind you we are already seeing the top of the strat programmed to take a wave 1 hit even before the end of October - far more encouraging than last year already is forecast at T+240 and not too shabby either.

 

NH_TMP_1mb_240.gif

 

Why is it that I think that we should ignore the seasonal model forecasts this year? - they are so far out of sync with the background teleconnective analogues.

 

Ed

Which brought this? [see attached thumbnail] Some mouthwatering pictures to get us in the mood for the season ahead. We are now more than half way through October. Still looking like some warming is going on, but at 2hpa and 5hpa the cooling is ramping up, unfortunately, but looking good at other levels.

Edited October 18, 2014 by Frozen Britain

[sebastiaan1973]

Or if you know how, you can get the data from datasets like ECMWF or MERRA and plot it manually like I do usually.

Bluearmy if you need any specific years, tell me and I can make the vertical zonal mean plots similar to FU Berlin.

 

Is it too possible for you to show the WAF? According to Cohen, WAF will increase in years with a big SCE. Thanks!

[lorenzo]

Is it too possible for you to show the WAF? According to Cohen, WAF will increase in years with a big SCE. Thanks!

 

I had a look into this and hopefully Recretos can apply his dataset genius to that question.

 

From the Linking Siberian Snow Cover to Precursors of Strat. Variability paper, the method is described below.. The calculation begins with the anomalous 100hPa meridional heat flux. Cohen gets this via deviation from the zonal mean of the meridional component of the wind x the deviation from the zonal-mean temperature.

 

The values provided are then averaged per month of averaged between 40N to 80N. The measure describes poleward heatflux and also represents upwelling tropospheric wave energy into lower strat. (EP Flux)

 

Failing that the JMA site has improved it's chart output from previous years. Temp & Vertical component of EP Flux - graphic updated to show waves 1-3 also.

 

A nice Time Latitude hovmoller has also appeared.

 

And the EP Flux looks a lot clearer.

[Frozen Britain]

Nobody has said anything about how good total ozone has been in the stratosphere recently. It has strong ozone for the entire run to t120.

[SMU]

2 points to add on the season forecasts, however 100% pertinent & related to why the ensembles always smooth out anomalies.

 

* Somewhere in the algorithms no matter how forward thinking will believe they are they all have an element of base state data taken from the long term mean, whether that's averaged over 10 years or even 50 -

 

Relating to that unless one of the key forecasting parameters is considerably stacked at one end of the scale either positive or negative then the general forecast will nearly always smooth to the average over time- this is because the model will eventually mute that signal out & blend it away into the 'averageness'

 

The best current 'model' forecast data we only be semi reliable for the next month / possibly 6 weeks where its current starting conditions have more influence in the calculation as opposed to the blended average condition that is built in say 6-8 onwards.

 

* Also using base state data makes a very large assumption ( & indeed carries a huge weighting in the forecast) that what the atmosphere is doing now is highly correlated to what it was doing way back in the base state....

To an extent this is probably very correct for how the whole system behaves technically, with fluid dynamics being a key process within this - However there are variables within the Ground based feedback loops that have changed significantly in the last 10 years- Namely

* polar temperature gradient

* Ice coverage & locations.

 

These provide significant feedbacks into the tropospheric - Stratospheric relationship which based on the short data we have increase the frequency of blocking around 60N.

 

As a result all the seasonal models will be wrong because I don't believe they are dynamic enough to incorporate these recent flux changes. I think the models will need significant updates & changes to take these into consideration resulting in more realistic seasonal outlooks & if the feedbacks are calculated correctly then more specific data relating to height anomalies & subsequent temp anomalies will become apparent.  

 

Its all about changing paradigms........

S

[phil nw.]

Good points Steve.

Of course we can't know if the different agencies already consider and include these in any model upgrades such as the new GFS due soon.

We have seen changes wrt the Stratosphere modelling in recent times so I would imagine forecasters would have an eye on these other factors you mention if they are on the ball.

Edit to add

Further to above this is worth a read from the MO Climate section

http://www.metoffice.gov.uk/services/climate-services/case-studies/improving-our-near-term-climate-predictions

It seems all recent changes are being considered in their LT forecasts.That,s the impression gained on reading that.

They do suggest that lower Arctic ice levels affecting the track of the jet and that it encourages +ve ht anomalies at high latitudes thus increasing chances of Easterly winds in Winter.

Just to say though that any imo Longer term outlooks can only give a general guide on probabilities and as ever should be treated with caution.☺

Edited October 19, 2014 by phil nw.