Winter Model Discussion 18Z 12/2/13 onwards.

[Chris W]

GFS 06z at 126 Hours:

That easterly lasts for days, becomes a slightly milder south-easterly then a cool easterly again into FI after which a bit of a northerly beast.

Suffice to say it is a cold run influenced by a strong block.

[snowking]

Right then, 6-10 day means:

We are familiar with the shorter term story now (minus convective potential - which does seem to be disappearing it has to be said), but the interest for snow fans is now on the lowering heights to the south - as alluded to by GP, Phil, and Ian F this morning already

Both suites keen on developing a west based negative NAO, with likely blocking structure based between Greenland and Canada, but crucially, the low heights to our south stay in a fairly solid position - i.e. they keep us on the colder side of things, with the likely result from todays output being LP's sliding somewhere close to S England, pulling in colder air from the near continent

ECM at day 10 suggestive of this too:

So, whilst 850's in this situation may not look that impressive:

Even at 0-2 below average (which places London, for example, at around -3 to -5c), with a very shallow continental feed, the likely LP systems running into the UK would bring largely snow inland.

The 6z GFS, to this extent, up to +240, is a very plausible run.

So, a shift in emphasis today really away from a convective easterly, and more towards frontal incursions from the south-west (again!). We could end up in similar territory to the January spell. Despite what looks like a downgrade on the face of the operationals this morning, for snow fans if anything its a slight upgrade. Risky stuff relying on frontal incursions, but potentially high payoff

SK

Oops, I forgot to add this:

Just about as solid an agreement as you will see for an negative NAO - no surprise to see the GFS op out on its own towards the end once again!

We see our initial east based neg NAO:

Before it transfers west:

Edited February 15, 2013 by snowking

[R.P. McMurphy]

Wow what a run for N England.

http://www.wetterzen...s/Rtavn2281.png

Most would be digging themselves out if this was to occur especially those in the E.

Gutted that probably won't come off, but would be a big snowfall for the North West / Northern England.

[LincolnSnowstorm]

Sorry: meant to include link. We always get them direct from NOAA at usual place: http://ready.arl.noa...v/READYcmet.php

nice link thanks

[Methuselah]

Considering a week or so ago there were "winters over" claims....I would take solace that there is very little in the models this morning to back that claim.Indeed plenty late winter weather to enjoy.

Are you sure about that, Dr Sno?

Anywho, the weather certainly has the potential to deliver some snow; but, as TWS rightly points-out above, it also has the potential to disappoint...Meanwhile, the models abound with a million-and-one indications...

[Froze were the Days]

How much credibility does this 06z run have? I should think not much, so we're hoping the euro's edge towards this - the last time this happened I believe was back in early December when the ecm back tracked big time. Met office seem pretty convinced there won't be much of an easterly (again).

Edited February 15, 2013 by Froze were the Days

[Methuselah]

Yes the gfs is now the form horse, it was lagging behind but it's a sheepskin nose-band ahead of the others, the easterly is nailed, the devil will be in the detail but that's not going to be known for a few more days.

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

[UV-RAY]

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

No, no RP, it is only to be ignored when showing mild.

[Froze were the Days]

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

[Froze were the Days]

Exactly...not sure if frosty was saying this tongue in cheek, I hope so!

[Severe Siberian icy blast]

Right then, 6-10 day means:

We are familiar with the shorter term story now (minus convective potential - which does seem to be disappearing it has to be said), but the interest for snow fans is now on the lowering heights to the south - as alluded to by GP, Phil, and Ian F this morning already

Both suites keen on developing a west based negative NAO, with likely blocking structure based between Greenland and Canada, but crucially, the low heights to our south stay in a fairly solid position - i.e. they keep us on the colder side of things, with the likely result from todays output being LP's sliding somewhere close to S England, pulling in colder air from the near continent

ECM at day 10 suggestive of this too:

So, whilst 850's in this situation may not look that impressive:

Even at 0-2 below average (which places London, for example, at around -3 to -5c), with a very shallow continental feed, the likely LP systems running into the UK would bring largely snow inland.

The 6z GFS, to this extent, up to +240, is a very plausible run.

So, a shift in emphasis today really away from a convective easterly, and more towards frontal incursions from the south-west (again!). We could end up in similar territory to the January spell. Despite what looks like a downgrade on the face of the operationals this morning, for snow fans if anything its a slight upgrade. Risky stuff relying on frontal incursions, but potentially high payoff

SK

Oops, I forgot to add this:

Just about as solid an agreement as you will see for an negative NAO - no surprise to see the GFS op out on its own towards the end once again!

We see our initial east based neg NAO:

Before it transfers west:

What an excellent analysis there thankyou for the time it took you to do that.

We have some deep negative NAO values there at times , and we now look to be entering what could be a prolonged cold spell there with values even out to 3rd march looking negative.

There's nothing wrong with a westerly based NAO so long as like you say we can keep on the northern end of the southerly jet , with low pressure staying in Europe and to the south of uk. If the high pressure cell can set up shop between Greenland and Canada then we may well be facing a snowy spell. West enough to allow the jet a sniff at coming north , but close enough to have an effect on how far north the lows can reach. Like you say , high risk with good results .

As for next weeks easterly its still too early to comment on snowfall and upper air temps , but the set up screams potential , weather that will be enough to deliver the goods or just remain the word we have constantly used this winter but with little prevail is yet to be seen .

[Barking_Mad]

Yes the gfs is now the form horse, it was lagging behind but it's a sheepskin nose-band ahead of the others, the easterly is nailed, the devil will be in the detail but that's not going to be known for a few more days.

It's not where you're from, it's where you're at!

[SLEETY]

How much credibility does this 06z run have? I should think not much, so we're hoping the euro's edge towards this - the last time this happened I believe was back in early December when the ecm back tracked big time. Metro seem pretty convinced there won't be much of an easterly (again).

Very little credibility at the moment on all of it's output!However if it gets this one correct then it's the credibility of the other two models that will have to be questioned,as both of them are showing much less potent easterlies than they were.

However at the time-scale that all the models are showing the easterly ,I would expect all of them to go through several changes in their output,remember once a model has released it's data it's already out of date.The weather does't stand still ,for 6-12 hours like the models do!.

Obviously the closer they get to T+0 the less chance they have of being incorrect in their model output.Worth bearing in mind that next time someone starts fretting about "downgrades" if a model dares to alter its output in the next run

[Captain Shortwave]

No, no RP, it is only to be ignored when showing mild.

Well the runs this morning show why the runs preceeding it needed to be binned, it had the pattern wrong at t96 until today, even nogaps was ahead of the gfs . Now that it has finally made the right choice in the earlier time frames we take the frames beyond seriously

[SMU]

There is now broad consistency within the full Operational model suite, ECM ensemble suite & I guess the MOGREPS ensembles-

Rather than comment on whats infront of us I though I would highlight the 4 key things to look for when homing in on a potent Easterly.

Here taken from the 06z- I believe the 06z represents the general super ensemble mean of the all the models.

1) The core of the low heights & Unstable air, this will also be the lowest temps of the upper air cold pool. These upper air cold pools rarely reach the UK & are often stalled or swerved away into france-

They are driven by the Upper level high & will only reach the UK if squeezed west by the Upper & high-

**IF** this makes it to the UK it will 100% bring a period of snow on its northern flank- & the convection will fire further North.

This happens on the 06z ( but was stronger on the 00z GFS) - very close similarities with Dec 26 2005 & Feb 1 2009-

here is where it lands on the 06z.

http://modeles.meteociel.fr/modeles/gfs/runs/2013021506/gfs-0-186.png?6

Whats also important is the track of this low- the further East & its influence will be low over the UK, my favoured track again for reference is:

Norway, Southern Finland, North Germany- where it may remain in situ for a day- then westwards again- after being squeezed.

2) The upper high that moves North out of the UK must gain the correct latitude, it will eventually be sheared & elongated East- which is good, but for people that still desire snow & live in places like Yorkshire then the highs lattitude is everything!

Whats more critical as well is the western end of the block - being 'lifted' to the right lattitude- we dont want the western end saggy - like the UKMO

http://www.meteociel.fr/ukmo/runs/2013021500/UW120-21.GIF?15-06

This will modify the initial CAA from europe.

3) Iberian Energy, People often overlook this in favour of the eyeline always on the Scandi high- however the energy moving North out of Iberia is whats supporting the high-

The best & most prolific Easterlies come with the sqeeze of isobars across europe to the UK are a result of the polar cell force moving south V the energy within the ferrell cell moving North-

that way the high is 'stuck' it cant sink.

Heres 3 classics-

http://modeles.meteociel.fr/modeles/reana/1952/archives-1952-3-29-0-0.png

http://modeles.meteociel.fr/modeles/reana/1987/archives-1987-1-12-0-0.png

http://modeles.meteociel.fr/modeles/reana/1991/archives-1991-2-7-0-0.png

This is usually the weakest of all features modelled, & also one that invariably is our stumbling block- notably because the energy isnt always deep enough to compensate the force moving south out the pole-

4)Southern Atlantic energy- Often ABSENT in recent winters & being replaced by the ever present Azorews high ridging North to kill the convection.- This seems to be well placed this time around-

Again, to build & sustain a decent easterly the track across Europe we need 'forcing' moving the cold west, then another area of energy across the Brest peninsula.

What will happen is the energy will drive north, but stop somewhere in this location due to the force of the high-

In recent winters the scandi high has had LOW heights positioned on the western end of Norway-

http://modeles.meteociel.fr/modeles/reana/2012/archives-2012-2-3-0-0.png shown here-

Which impedes the blocking-

this is so crucial that this region is blocked as the ridge will normally extend out towards iceland- as a result the southern energy can slide south under the block, where as no blocking the energy will recurve north over the UK = end of cold.

This is the GFS energy plot at 174

http://modeles.meteociel.fr/modeles/gfs/runs/2013021506/gfs-5-162.png?6

notice the energy approaching the UK from the SW & then being sheared away east again.

So in summary-

The pattern longwave is nailed- Upper high over scandi & low heights into europe-

What we need is consistency of the 4 key features of a snowy easterly to varify.

1) Upper high in the right lattitude

2) Iberian Energy

3) Southern jet energy

4) Upper air cold pool moving west to the UK & a westerly track of the low.

I hope this insight has helped some viewers.-

This is also why the UKMO are tentative around forecasting cold & snow from the east, because without a CLEAR signal of all these coming together at day 6/7 then its pointless.

Even at day 5 the track of the low can vary quite a bit- so UKMO sitting on the fence is the right choice.

regards

S

[bradymk]

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

No, no RP, it is only to be ignored when showing mild.

The reason the GFS was ignored was not because it was not showing cold but because it was in complete disagreement with all other models. ECM, EC ens, UKMO-GM and GEM all agreed yesterday of a cold easterly with GFS completely out of kilter. This was ignored because of science, not because of cold bias. It would be the same for any model however recently it always seems to be the GFS because it has been poor this winter. The reason it is not being ignored today is because it is now in broad agreement with all the models and now we are seeing just variations of the same theme.

[shaky]

Wow what a run for N England.

http://www.wetterzentrale.de/pics/Rtavn2281.png

Most would be digging themselves out if this was to occur especially those in the E.

Not just for north england but the whole of england!!

[Simonl]

Good to see the NAE at +48 supporting the 06Z GFS.

http://expert-images...021706_1506.gif

Im going to keep my eye on this chap over the next 24hrs.

TEITS

Would you be kind enough to briefly (!) how you read/interpret this and what are you looking for? I am not a novice per se at reading charts but I don't understand the NAE ones at all! lol. Also, Isn't GFS still playing catch up with the euros? If this is the case and it appears so isn't it like looking at history already? the NAE charts (If you see what I mean?!)

Thank you

Simon L (formerly known as Snowfish)

[No Balls Like Snow Balls]

The short ensembles are out. It looks like the op could turn out much colder than the mean in the extended.

IMO that was to be expected.

Edited February 15, 2013 by No balls like Snow Balls

[fergieweather]

The 00z UKMO-GM now looking much more in tune with other solutions mid next-week and contra it's predecessors has received no modifications this time around from Exeter.

You will recall how yesterday’s 12z GM - with faster CAA and considered something of an outlier - wanted a greater degree of upper trough disruption across the Biscay/NW Iberia area through late Sunday into early Monday... consequently, this resulted in backing the continental flow into the UK more by mId-week. However, also recall how (despite fans here!) this more potent-looking solution had scant support from other models, i.e. for a faster transition to a more pronounced E’ly. For that reason, Exeter stayed cautious but the 00z GM is thought to represent a much more realistic fit to the most probable evolution - especially given the support from 00z NCEP and EC Det runs. Thus, Exeter will run with what you see in the raw fields, and little significant modification.... at least for now.

[Frosty.]

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

One of the things I have noticed about the gfs after being on here for over 7 years is that it does tend to be slower to respond to a pattern change and therefore looks clueless as it flounders around but once it latches onto a pattern, it can suddenly pick up the reigns and storm ahead, this is again what seems to have occured but since it's the gfs with 2 extra runs per day compared to the other models, you never know what you are gonna get next.

[Yarmy]

So I hear a lot about how bad the GFS 06z run is (and to some extent the 18z so-called pub run). Being a fairly unexperienced member of this forum and always willing to question to conventional wisdom, I thought I'd have a look for some numbers and found this:

http://www.emc.ncep.noaa.gov/gmb/wx24fy/doc/GFS4cycle_fyang.pdf

It seems that the 06z is indeed the least skilful and the 00z the most; however the differences are very small and have been since 2007. Most interesting to me though was this section:

Question 2

Since the forecast skills of 06Z and 18Z cycles are not as good as the 00Z and 12Z cycles, what is the benefit for running these two extra cycles?

Answer

1. 
   For short-range forecast ( ~ 3 days), the later 06Z cycle does show better forecast skills than the earlier 00Z cycle when both are validated at the same verification time.
   
2. 
   For medium and longer range forecast, the later 06Z cycle is not significantly different from the earlier 00Z cycle.
   
3. 
   Similarly, the 18Z cycle is better than the 12Z cycle for short- range forecast, but not significantly different from the 12Z cycle for medium and longer range forecast.
   

Something to bear in mind perhaps when summarily dismissing a particular GFS run.

[Methuselah]

Well the runs this morning show why the runs preceeding it needed to be binned, it had the pattern wrong at t96 until today, even nogaps was ahead of the gfs . Now that it has finally made the right choice in the earlier time frames we take the frames beyond seriously

But, if the weather treats the current batch 'stellar runs' with the same degree of contempt as it did those that have preceded it, they might (again) all need binned?

[johnholmes]

a very insightful post Yarmy.

Later model runs for trying to forecast within a 72 hour time frame will USUALLY give a better 'feel' for detail than the previous run but it is not always the case. That is the problem knowing when an 06 or 18z run, in that shorter time frame, is giving better guidance than the previous 00 and 12z outputs.

well worth reading the full link SLOWLY as it does dump one or two myths that seem to be popular on here!

this just the summary

I have bolded perhaps the one most often bandied about!

Unless someone has actually done statistics for the European theatre that shows this is not the case it is a point we should all take note of

Summary

33

1. The GFS 06Z and 18Z cycles are not as good as the 00Z and 12Z cycles.

2. The 06Z cycle is the least skillful.

3. Conventional data count of the 06Z is consistently less (~10%) than that

of the other cycles. No large difference in satellite data count is found

among the four cycles.

4. For short-range forecast ( ~ 3 days), the later 06Z (18Z) cycle does show

better forecast skills than the earlier 00Z (12Z) cycle when both are

validated at the same verification time.

5. For medium and longer range forecast, the later 06Z (18Z) cycle is not

significantly different from the earlier 00Z (12Z) cycle.

--- only run the 06Z and 18Z cycles for short-range forecast?

6. In general, both the GFS and ECMWF have the best (better) forecast skills

for the 00Z cycle. However, both model showed worse tropical 850-hPa

winds for the 00Z cycle than that of the other cycles.

Edited February 15, 2013 by johnholmes

[Paul_1978]

Just as I was almost beginning to believe that it should always be ignored...No wonder newbies are getting confused...

Frosty's post was tongue in cheek....