In depth Model Discussion

[Glacier Point]

Please keep this thread for technical model discussion. Post your general model comments in the relevant thread. Thank you.

One of the key signals we are looking for during early winter is a movement of the polar vortex eastwards away from Greenland towards northern Scandinavia and Russia.

November and the first few days of December have been characterised by strong thermal anomalies at all levels in this area (60 E- 120E) supporting +ve height anomalies and blocking. This has effectively held in place the polar vortex over Greenland maintaining a strong thermal gradient aloft and flat jetstream across the Atlantic with notable -ve heigh anomalies centred Iceland:

This large warm air anomaly may be attributed to seasonal conditions enhanced by the lag effect of elevated hemispheric temperatures and the impact of a stratospheric cooling event during October and November. All of this is consistent with the westerly QBO phase which remains at values around 10.

http://www.cdc.noaa.gov/Correlation/qbo.data

If we maintain this pattern, there will be no cold incursions during December and temperatures will remain slightly above average.

However, there is evidence of some weakening in the height anomalies across this crucial area to our far NE associated with cooling air. Take the last 10 days in November...

Polar thermal profile (T. anomalies 65-90N):

Polar height anomalies

This is consistent with analogues for December under matching QBO and El Nino conditions suggesting a transference of the vortex and dropping off of heights in the key area, possibly delayed by 10 days as a result of elevated hemispheric temperatures:

We are already beginning to see this in GFS runs, the first possible attempted breakdown now:

Note the development of colder air aloft into Scandinavia and deeper cold at the pole. This is a positive sign of change.

This change is likely to take some time to effect, possibly over several weeks with several attempts before the final displacement of the vortex. At this stage we are looking at t288 (15/16th December) as being the possible significant transition so this is beyond publically available ECM outputs and the variability of GFS runs gives some cause for concern although it has been showing in a number of recent runs.

The sudden death of the MJO has complicated the timing of any possible pattern shifts although any revival of the wave (if) when it enters the Pacific could enhance the transition to a colder winter pattern.

This could of course be complete rubbish and the warm air / +ve height anomalies re-establish to the NE. We are talking a long way out here, but any continued suggestion of transference of the vortex towards Scandinavia would be very encouraging for cold lovers.

To emphasise the uncertainty going on here, compare 04/12/06 500 hPa height anomalies for GFS and ECM 8-10 day:

No joy there on the ECM just yet - strong blocking Eurasian ridge.

GP

Edited December 4, 2006 by Glacier Point

[Stratos Ferric]

Interesting analysis there GP. I have commented elsewhere, so far without any response, on the assymetric lok of the polar ice cap at present, this exacerbated by a hug anomalous warmth in the waters of N Russia. I'm wondering whether this sin't skewing the vortex along the meridian, down 180 E/W. The net effect of this, of course, is to set up reinforcement on the ground (NE Canada = cold, E Asia = cold), thus attenuating the set up further so that a degree of inertia establishes itself.

At this early stage in the winter my view is that the surface cold feedback is all in the wrong place at present for us to get a half-way decent winter worth the name. It's early enough for this to change, but each passing day makes the change less likely, or at least less likely until much later in winter - which would, of course, align well with the pattern of three of the past four years.

[West is Best]

Interesting analysis there GP. I have commented elsewhere, so far without any response, on the assymetric lok of the polar ice cap at present,

It's way beyond my competence, but there are one or two who post techie stuff on NW who ought to be take on your extremely interesting thesis I'd have thought.

I've a fairly senior climate scientist from the Met Office coming for dinner on Wednesday so with your permission I might ask them for their thoughts on it? Have to be careful though - I wouldn't want them stealing your original idea and publishing a landmark paper in a year's time on it!

Edited December 4, 2006 by Paul

[Stratos Ferric]

It's way beyond my competence, but there are one or two who post techie stuff on NW who ought to be take on your extremely interesting thesis I'd have thought.

I've a fairly senior climate scientist from the Met Office coming for dinner on Wednesday so with your permission I might ask them for their thoughts on it? Have to be careful though - I wouldn't want them stealing your original idea and publishing a landmark paper in a year's time on it!

I very much doubt that there's anything original or novel in my thinking, but in any case, feel free to discuss. Whilst we know the polar cap has retreated, it would be interesting to know whether it has retreated symetrically. I know that Philip Eden postulated that one reason why our northerlies are now less potent than they were in the late 60s is because the ice has retreated much further N and NE since then. It not just the opening up of a more active surface, it's the inference that the existing surface further south must also have got warmer.

[Glacier Point]

Some negatives and positives this morning.

Both ECM and GFS show a broadly zonal pattern associated with a +ve AO and NAO pattern with vigorous hemishpeheric jetstream across the Atlantic, with some deepening of the mid-Atlantic trough resulting in height rises over continental Europe out to 10 days.

In the Pacific the strong jet is approaching split flow as it hits the body of cold water off the western USA and there is a notable increase in the sub-tropical jet flow across the southern US over the t144 period associated with El Nino:

http://wxmaps.org/pix/hemi.jet.html

The polar vortex continues to show positive signs of shifting its core to NW Europe and Russia although the height rises over Greenland indicated by yestedays GFS are not evident today. Notably the ECM solution out to 10 days reduces the intensity of the Eurasian ridge although the jet flow remains strong.

The 00Z GFS was dominated by a split jet flow from t228 onwards with displaced high pressure located over NW France and southern UK. This split flow may be a good sign for later cold and could allow retrogression of highs NW to form transient Atlantic blocks.

With many opportuinities for model instability now starting to appear in the Pacific associated with a split flow scenario and increasing strength of the sub-tropical jet, the more consistent ECM modelling is slightly favoured with low confidence on outputs beyond t144.

GP

[Glacier Point]

As part of my winter forecast, I made use of a constructed analogue based on three years providing solid matches for El Nino, QBO and SSTA.

Two of these years, 1963/4 and 1977/8 have verified extremely well in terms of November's pressure anomalies:

Observed

Constructed analogue

and they look uncannily like the current situation in relation to the polar vortex with 1963 in particular providing an interesting parallel. This is the comparison of the polar thermal profile from mid November onwards:

1963

present

The profile for 1977 was also similar and the analogues suggest a trend for the polar vortex to shift from Greenland to Scandinavia and then east to be replaced by a dramatic change to positive temperature anomalies at mid and upper levels by the end of December.

During 1963 this evolution was quite rapid with the cold stratospheric anomalies transferring downwards to form a new vortex over Siberia within the space of 7 days.

From this:-

To this:-

What relevance does this have to the current situation?

Well, here we have two years which have a strong similarity to the weather we are currently experiencing right now, two years with a very similar profile in terms of the polar airmass and major teleconnections. They suggest a big pattern change is likely during December.

The Arctic Oscillation is projected to go through the roof over the next 5 - 10 days:

http://www.cpc.ncep.noaa.gov/products/prec...ex/ao.sprd2.gif

This could well be the sign of a massive period of volatility hinted at by the analogues.

The timescales do not however fit the historical matches and this is where I think we need to factor elevated hemispheric temperatures and a lag effect. For interest the period 5th to the 14th December of 1963/77 looked like this which should translate to the trend for the second half of December 2006:

...showing big height anomalies to our north and low pressure to the SW.

Then up pops the 12Z GFS.....

perhaps not the perfect match for the composite analogue - but indicative of a dramatic shift nonetheless.

None of this means anything though until it starts to show up in ECM plots as well as the 'edge of the rainbow' GFS.

Keep watching. History at least suggests the second half of the month has at least some potential for a pattern shift. Whether or not we will have to wait and see. Also of interest will be how quickly any (if) any change takes place and at what stage the modelling picks it up - remember the 1963 event took place over a very short period of time.

GP

[kold weather]

Very good piece of info there GP. The set-up in regards to 77-78 is the one that has caught my eye there, esp being El nino was a pretty similar strength and when you take your own obs into account it does look intresting possibly.

The key for me is what the LP and the jet does over the USA between 168-228hrs, we must get a NE movement of the LP exiting towards the east states, that would lead to at least a set-up like the 06z showed, a northerly toppler. If we get lucky and the jet then dives back southwards then pressure will have to rise over NE Canada and with the jet having already gone more meridonal (Hence the Bartlett at 228hrs...) then the most logical set-up to finish with would be a northerly. How long it lasts would be up for debate, I suspect a toppler would be the answer but a 12z set-up would be possible.

The other option is that the jet contniues to just keep with the same pattern and that LP busts through any mid-atlantic ridge that trie sto form and we end up with a flat flow across the Atlantic and a colder set-up would be yet another 5-7 days further on from that.

The change from slight to very + AO will mean the pattern will almost certainly shift from where it is now, but the exact pattern will depend solely on where that Lp over the states between 168-228hrs actually goes, if the worst option comes off then a zonal and Barteltt/Azores high combo here we come, on the right side and a northerly set-up could well ensure.

By the way, the 1977-78 pattern was one that got more and more snow as the witner went along, and while that amazingly cold, it did feature something for nearly everyone and i think everyone would accept a winter like that one!

Edited December 5, 2006 by kold weather

[BLAST FROM THE PAST]

GP

Thought provoking indeed. My method predicts a rapid change in early Jan. I anticipate a 7 day change too after the Jan 4th ' storm' shall we call it. The lag you mention could bekey here and may i suggest will fit in with my timings.

BFTP

[John Cox]

Thanks GP,

I love reading your educational posts.

A question. I have only been model watching here on nw for 2 years.

These synoptics are totally different to the last two years at this time of the winter. We had meridonal type weather for the last 2 winters with cold but fairly dry weather.

Does the current set up with a potential for change in a couple of weeks open the posibility for a really decent cold winter with snow? Could this be the winter we have all waited for or a possible mild winter with a strong EW jet. To put it another way could it be an all or nothing winter with nothing in between like the last 2 winters.

Perhaps this is not a fair question but it is what is on my mind at the moment?

[kold weather]

I'm not GP but when you have El nino it does open the way for the possiblty of much snowier set-ups as well as cold IF you can get northern blcoking to establish because a El nino tends to favor a strong southern jet stream, which means that when northern blocking gets going you get more vigarous depressions to our south which equals more instablity.

Sadly it doesn't work this way always, esp when you've got a strong El nino, but keep it in the weak-mod range and it does increase the chance.

Also worth noting what the met-office themsleves said about El nino:

"Moderate strength El Niño conditions are expected to continue through the winter and have in the past often been accompanied by colder-than-normal winter spells in northern Europe, particularly in the Scandinavian region. However, this effect does not materialise in all cases, and thus the uncertainty surrounding the influence of El Niño remains key to the forecast and will continue to be monitored."

Edited December 5, 2006 by kold weather

[WhiteFox]

I'm not GP but when you have El nino it does open the way for the possiblty of much snowier set-ups as well as cold IF you can get northern blcoking to establish because a El nino tends to favor a strong southern jet stream, which means that when northern blocking gets going you get more vigarous depressions to our south which equals more instablity.

Sadly it doesn't work this way always, esp when you've got a strong El nino, but keep it in the weak-mod range and it does increase the chance.

Also worth noting what the met-office themsleves said about El nino:

"Moderate strength El Niño conditions are expected to continue through the winter and have in the past often been accompanied by colder-than-normal winter spells in northern Europe, particularly in the Scandinavian region. However, this effect does not materialise in all cases, and thus the uncertainty surrounding the influence of El Niño remains key to the forecast and will continue to be monitored."

The winter of 1977-78 was the coldest on record for a large portion of the Midwest. I also believe that there were some heavy snowfalls on the East Coast. It would be interesting to see the charts for that period if I could find anything for the US! This would help to give an indication of what to look out for over here, and the position of the jetstream.

[John Cox]

Hi Kold, (nice to hear from the US too, White fox)

You are a very good stand in for GP and a great source of information

Good explanation and obviously a big piece of the Jig Saw puzzle.

[BrickFielder]

Excellent analysis by GP but I have a feeling that large scales changes are more likely to occur in January and into February and I will explain why later, although this does not exclude short cold spells happening before then. I think looking at correlations between winter and various indices is a slight mistake in that you probably should have treated early and late winter separately. My reasoning is linked to how stratospheric warming events work. Let us have a quick look at the state of play currently in the stratosphere.

No warming looks to be happening at the moment although the stratosphere typically cools at this time of year it is particularly cold now. This equates to a strong stratospheric vortex and very limited high pressure high latitude blocking with little chance of easterly winds bringing cold air from the continent our way. Put simply this is not a good sign for snow lovers and looks set to continue for the time being. I do expect this to change but not till later on.

My understanding of how stratospheric warming events work starts with the conditions of high latitude blocking. High latitude blocking can lead to cold for the UK when we get high pressure to our north ,over Scandinavia or mid Atlantic linking to a Greenland high. This high pressure blocking is significantly enhanced by a weak stratospheric vortex and warmer than normal air just above the tropopause (low stratosphere). We know that these warming start off in the top level of the stratosphere and move downwards over a period of just less than a month. Looking at the current state with little warming in the upper stratosphere there is not much hope of a warming reaching lower levels in the next few weeks. Put simply a stratospheric warming moves down from the upper stratosphere to turn the AO (Artic Oscillation Index) or NAM (northern annular mode) negative and stratospheric cooling truns them positive.

Many readers will perhaps be familiar with all that I have mentioned so far, hopefully this bit will be a little new as I try to explain some thoughts about how the warming events start. The source of stratospheric warming events seems likely to be planetary waves which start in the upper troposphere. Here is what a planetary wave looks like in the upper stratosphere identified concentrations of trace elements.

Not only do these waves have a latitude and longitude element but as they break they transfer heat upwards. I think we are porbably around day 30 on the downward transfer of a cold event (another 30+? days to go)

From this we can see that these waves start very low down in the atmosphere and travel all the way up and back down again. These waves do not always behave like this and will quite often be reflected back down from the tropopause . Current explanations for whether the waves are reflected or not center of the strength and direction of prevailing stratospheric winds. Statistics show that during strong westerly QBO (Quasi Biennal Oscillation) stratospheric winds in early winter all planetary waves are reflected back from the tropopause and no stratospheric warming events occur. This changes later in the winter with statistics showing the QBO has no affect in late winter. This is why I think the polar vortex is unlikely to weaken much during December. Comparing planetary wave ones from this year and last we can see that there is a distinct lack of wave activity this year( Note Wave 2 also looks meagre this year).

Here is a quick quote from some reference material on the subject.

Major winter stratospheric warmings preferentially occur during the easterly phase of the QBO, Holton and Tan (1980).

In a usual northern-hemisphere winter, several minor warming events occur, with a major event occurring roughly every two years. One reason for major stratospheric warmings to occur in the Northern hemisphere is because orography and land-sea temperature contrasts are responsible for the generation of long (wavenumber 1 or 2) Rossby waves in the troposphere. These waves travel upward to the stratosphere and are dissipated there, producing the warming by decelerating the mean flow.

There exists a link between sudden stratospheric warmings and the quasi-biennial oscillation: If the QBO is in its easterly phase, the atmospheric waveguide is modified in such a way that upward-propagating Rossby waves are focused on the polar vortex, intesifying their interaction with the mean flow. Thus, there exits a statistically significant imbalance between the frequency of sudden stratospheric warmings if these events are grouped according to the QBO phase (easterly or westerly).Stratospheric Sudden Warmings are twice as likely to occur in El Nino winters than in La Nina winters, thus providing a mechanism for the possible influence of ENSO on the polar stratosphere.

One thing you should pick out from that quote is the warming by decelerating the mean flow and it is by looking at the heat flux at the lower stratosphere that we get a real clue as to the source of planetary waves. We can link the heat flux directly to the NAM and here are a few charts and slides from Polvani and Waugh 2004 which help to explain the link.

To sum up we should be looking at the current AO for clues to the AO in about sixty days except when there is a strong westerly QBO in early winter as now. A positive AO now and strong vortex in theory might equate to the opposite in February. Put simply conditions look favourable for a cold February.Interestingly this would seem to show that we should have two month phases of weather and that seasonal weather should repeat its self every year until we have a strong westerly QBO.

Please note that this post is for peoples enjoyment and is exploring avenues of understanding that I am not an expert on, so I am more than likely a little wrong in my conclusions.

[phil nw.]

Thanks Brick,interesting stuff.There`s been much discussion about the Arctic Oscolation in various threads lately and the connection to the upper atmospheric temps.Its good to learn a little more about why and how these temps. change periodically.

[summer blizzard]

Current modeling supports a possible high pressure inversion next week, what do the big guns have to say on this???

[Stratos Ferric]

Current modeling supports a possible high pressure inversion next week, what do the big guns have to say on this???

Not sure I'm a big gun, but my view is that a HP coming off the Atlantic islikely, at the very least, to have patchy cloud around it. If we get clearances than cooler nights should follow, but nothing particularly cold - the air mass is a warm source.

[BrickFielder]

First sign that the stratospheric cooling may be coming to an end.

Indications are that any warming will take at least a month and a half to reach the bottom of the stratosphere.

This means any large scale blocking or high pressure pattern is unlikely until towards the end of January.

Signs of a change coming up.Strong Atlantic jet seems set to weaken for a period while the Siberian high seems set to move back west. The jet is also set to put on a burst as it exits china.

Leading to an amplified pattern over the United States and sending the jet north in our location.

I think it is a pattern which will reset back to Atlantic weather for us as canadian cold builds high pressure and straightens the jet out by mixing cold and warm air of the east coast of the United States. The jet burst off the Chinese coast is likely to repeat so our weather is likely to flip flop between Atlantic wet and windy weather and murky foggy high pressure with brief northerlys.

There is an interesting little low pressure forecast to develop over the Azores which by its nature of having some tropical properties may prove unpredictable and a bit of a spoiler or unknown quantity.

It might take of though the Med or the continent rather than fading as currently forecast (Not likely to head our way but worth watching). I expect it to throw the forecasts off a little ,so expect forecast changes over the weekend.

Some chance of snow where we have had very little so far but I doubt that we will have a white christmas.

[The Eagle]

Latest CFS data shows that Feb may be interesting but thats only relative to average so maybe your right re the stratosphere as this would fit nicely into the high latitude blocking. It is becoming increasingly difficult to see where a meaningful cold sanp will come from for the forseeable future:

[Jack Wales]

I notice a lot more mention of ‘Stratospheric Anomalies’ on the forum this year, particularly regarding of the possible effects on our weather at the surface. The analysis and modelling of Stratospheric Anomalies has also formed part of a lot of winter LRF’s this year.

I have done some research and there is a definate correlation between stratospheric warming events and cold anomalies occurring at the surface approximately 3 – 6 weeks later. Conversely when the stratosphere is cold this leads to more zonal weather at the surface with a similar time lag.

The attached chart (Stratospheric Warming) for the JFM period in 2006 shows this clearly. The stratosphere in the NH being cold in Dec 05 entered a warming phase at the beginning of January 06 and remained anomalously warm until early Feb (and beyond in the lower layers). The cold kicked in properly at the surface (in the UK) around early February and remained until April so a clear correlation is visible there.

There is also a correlation with Stratospheric zonal wind conditions which generally turn negative during warming phases with similar time lags affecting temperatures at the surface, see attached (Stratospheric zonal wind) chart.

It is clearly evident that we entered warming phase about a week ago and the signs are that warming is taking place down to the mid layers of the stratosphere (40 hpa), considering the rate of warmth you would expect to see an exponential increase in temperature anomalies in the lower layers over the coming days. Factoring in a four week time lag for this event to take effect at the surface, can we expect to be engulfed in an arctic air mass when the surface synoptics ‘flip’ in early January? I very much doubt it but with the AO likely to enter a more prolonged neutral/negative phase you would assume that the chances of colder weather would be greater.

The above analysis merely reflects anomalies for the Northern Hemisphere alone. When you break this down it actually shows that the stratospheric warming is taking place over Canada/Alaska with UK/Europe remaining in a cold phase.

Link

So I have three questions:

1, Does this mean that any blocking would only occur over N America with W Europe remaining in a zonal pattern?

2, What correlation is there (if any) to the amplitude of the height’s field waves to surface level blocking?

3, What are the caveats which could inhibit stratospheric anomilies from affecting weather at the surface?

I’d be grateful if anyone could shed more light on this.

Stratospheric_Warming.xls

Stratospheric_Zonal_Wind_Anomaly.xls

Edited December 19, 2006 by Jack Wales

[beng]

Just thought I'd bring this thread back up - since Jack's question about whether the stratospheric warming would produce high pressure over Alaska during early Jan - and leave Europe zonal - looks like your question has been answered by the models now. Very interesting though to see the strong relationship between the stratospheric warming and the weather on the ground though.

[beng]

I totally agree with this - and I think it could be the biggest factor that eventually turns things around for us. The latest (strong) warming episode comes in a much more favorable location for this part of the world. I think Feb could be quite different to the rest of the winter.

Tamara

Hi Tamara, yes I'm hopeful that the new warming event is more favourable for us - I think we need to be looking at the 5hPa level to see where the current warming is and if I'm looking correctly - it's shifted a little - although not a lot.

http://www.cpc.ncep.noaa.gov/products/stra..._t05_nh_f48.gif

by 168 hours it looks a bit more promising though - but we could be talking last week in Jan before the effects are felt down here with the PV perhaps shifting towards Scandi

http://www.cpc.ncep.noaa.gov/products/stra...t05_nh_f168.gif

Have to admit though, I'm still getting my head around the stratospheric warming charts so I may be wrong.

[Shiny_Bottom_1]

Interestingly, since Darkman posted the above chart (now updated?), things dont look to continue with unpresidented warmth. The UK mainly stays in the above average category, although this doesnt look anything like 2006.

Back end of winter and into Autumn also shows possible signs of going below average, which in a way will kind of link into February cooling.

Also interesting that quite a fair bit of Eastern Europe goes below average for the last 2/3 of summer and first 1/3 of autumn.

The warmth over Russia disappears too throughout the summer.

I think after 2006, 2007 will be an interesting year to see how it actually pans out.

[Glacier Point]

A lot is going on now suggesting we are on the verge of a second, more progressive pattern change favouring cold.

The QBO has declined to weak values, especialy over the last ten days or so. This reanalysis shows the change on anomalies of 30mb zonal winds with an already weak QBO in decline:

Weak westerly QBOs have a major bearing on the state of the polar pressure anomaly or the AO with the breakdown of the polar vortex related to the state of the QBO - a weaker QBO gives a better chance of the emergence of +ve height anomalies over the North Pole.

Reanalysis of the 7 Februarys with a weak westerly QBO suggests the following 500 upper pressure anomaly pattern:

Points to note - the ridge in the NE Pacific, strong trough into Canada, deep -ve anomalies into Scandinavia and ofcourse a strongly -ve NAO signal in the Atlantic. Interestingly this suggests that the weakening QBO tends to favour a splitting of the polar vortex into Scandinavia and North America.

The solar flux is an important consideration for the state of the QBO and for the two years in our reanalysis when there was a solar minima (1962 and 1964 (1964 was also a weak El-Nino), the reanalysis looks like this:

Similar but with the NAO less -ve and the Atlantic block closer to the UK although this is a low sample.

Looking at the polar thermal anomalies, we have seen a dramatic change during mid January. Compare the 13th January profile with the 21st January:

Showing an extension downward and narrowing of the columns of cold air with two distinct columns towards Canada and Scandinavia. Coincidence or as a direct result of the weakening QBO ? I suggest the latter.

With the ECM and GFS Ensembles continuing to support the development of the Alaskan ridge forcing height rises over the Pole and deepening trough into the eastern USA and Canada, we have potentially the embryo of a sustained -ve NAO phase. The next evolution to look for is height rises over southern Greenland as the jetsream is forced southwards as the trough across the USA deepens sufficiently to bring the sub-tropical jet into play.

The worst case scenario I see is a blocking high to the WNW of the UK as the AO is not strong enough to recover from the earlier stratospheric cooling. The best case scenario for cold is for a sustained -ve NAO with a strong Greenland block in place presenting numerous opportunities for depressions to try and undercut from the SW - the huge snow potential I mentioned a few days ago.

Now before we get too carried away, this is just around the t240 margin so plenty to be cautious about and I would again urge some consideration that this type of synoptic need not deliver brutal cold to the UK as we can be located just out of its reach whilst the near continent freezes. For the record, the weak QBO composite predicts deep cold to penetrate Scandinavia and sub-normal temperatures for the southern UK:

GP

[VillagePlank]

Feb, to be slightly drier, and warmer than normal

Mar, slightly wetter than normal with S/E England,Scotland, and Ireland (including Eire) posting negative anomolies for the month.

The time for prolonged snowfall, then, it seems, is best left, once again, to March, I'd suggest that this season is again shaping up rather than same as recent years (the even larger teapot?) where significant wintry weather is left until right until the end of winter.

Coupled with GP's (as always) excellent post, I'd suggest a very cold end to Feb and beginning of March as a realistic possibility how we can match up GP's QBO, and CFS charts; which is, once again, the pattern of recent winters.

Perhaps a little too early to call ; so don't shoot the messenger However, given the warm(ish) outlook for the first week of Feb, it seems, at this stage, to be the only sensible solution.

Edited January 25, 2007 by VillagePlank

[kold weather]

Hi GP, yes as i was saying on the other thread nearly all the teelconnections over the past 7-14 days have flipped somewhat away from where they were, thats a massive global change!

As you say I don't actually see why some people are so down about the global set-up ,its got huge potential. i wrote a post in the model thread however i think its probably just as relevant in here, this is my ideas on why the models will struggle ove rthe next week or so:

If you actually take a moment to look at the global set-up you can easily see why ther eis such uncertainty and why you can't rule out anything. Super +ve AO has now ready to be replaced by a -ve AO however at the same time the stratosphere is cooling again, so how long to hold the -ve AO?

Needless to say the models have often been WAY to fas tin changing set-ups, the same may hpapen wit hthe -ve flow now.

Secondly you've got El nino which has weakened massivly. its been the main driver of the jet however with some rather severe weakening over the past 7-10 days the jet is therefore likely to weaken down in Feb like tinor mally does anyway, but a -ve AO means more colder air will be able to bne tapped into whjich will start to possibly re-tighhten the gradient inbetween Lp's and strengthen them a little again, but to what extent and how fast does the neutral signal emerge?

Then you've got that huge block that has formed and it is very strong and obviously as we saw in December the models are not great at handling these featurews, esp when they dictate the global flow like this one is doing so, obviously if the models aren't able to handle these blocks well then there will be much disagreement about how the block shape is kept therefore a different....

Polar vortex position which will obviously have a huge effec ton our weather, a different block placement will end up shifting the PV in a different way, esp as the PV is a major feature that is being progged to form currently over Canada. So does the huge block topple and force the vortex towards Greenland like the 12z GFS op run does or does it stay where it is and the block stays in the same place like the colder runs in the ensembles?

The other major problem is as ever the global models biases. It should be noted here that there are three biases that could make the difference:

1: In the longer term the models tend to put to much energy in the northern arm. This may be even more likely then in recent months where the +ve AO has actually played into the models hands somewhat in terms of this bias but now we ar ein a -ve AO set-up the models may well start to find themsevles over-egging that northern jet and pressure being higher then expected over the pole.

2: Leaving too much energy to the Sw of the main polar vortex. what this does is pump up ridges beyond what would normally be the case in a flat W-E flow. This can do one of two things, either give the models too much of a easterly flow over Europe when we have a mid-atlantic high OR over-do the strength of any Euro high.

3: Digging the cold air too far south. This one sort of links in with 2 because the models over-do the cold air dgging down from the arcitc. This can mean that sometimes the models overdo cut-off lows aweay from the arctic.

So what does this mean for the models?

Now when you link those three you see why this could cause problems for the models. Firstly if the models overdo the energy in the northern arm of the jet then that means that any northerly flow is goin to dig far too hard southwards which means that not only will Lp's become stronger then what actually normally occurs but any northerly flow for the Uk is also going to be over-done and more importantly past 168hrs it starts to make even more serious problems because with the model over-doing the energy in the jet it means too much cold can come south. This then means that there is a tighter gradient then is the actual case which powers up LP's too much. This is where No.2 comes into play. because there is more energy present the PV to the SW of any block (though this does happen to the SE as well sometimes...hence why sometimes potent easterlies turn out to be a bit toothless.) it tilts the jet stream flow to a SW-NE direction.

Sometimes this is the case as this is the global flow generally anyway however the problem for the next 2-4 weeks arrises in the models because the the cut-off low to the SW/SE of the huge block will stick around for too long and exert to o much energy away from the block meaning the models will destroy the block too fast and then all the sudden the jet can change and go to a more zonal flow...however in truth because of all of the above it doesn't happen this way.

will these biases make any real difference to the models in the end???

Maybe not. the eCM is probably the best at handling thses biases thoguh it suffers from No.2 quite badly at times, No.1 is the major problem wit hthe GFS and therefore as a result gets all 3 of the other major biases as a result.

Ironically the real problem for the models is that all three of these thnigs are a result of each other. if you get one then you'll probably get all three to some extent which is why the models dont do so well past 180-240hrs on the global pattern and until they can sort out those biases, they never will do that well either beyond that timeframe.

So when you do eventually add all these various parts you can see why the models may well struggle wit hthe global and therefore the local pattern ove rthe next 7-14 days esp. I think the set-up may well be like what you suggest GP. Sevral runs have hinted at thwe classic weak El nino set-up and thats the one that has my favor at the moment like you I'd guess. However because of biases and the uncertainty with global pattern you can see why thereis such scope in the current ensemble members and models, everything from a beast from the east to a bartlett is up for offer however the models may well go with their normal thing of over-doing the jet...

Edited January 25, 2007 by kold weather