Winter Model Discussion 06Z 10/01/13>

[UV-RAY]

I've discovered yet more upstream data to inflict on this thread!

Here is the SREF, this is the ensemble system for the NAM.It's basically the USA version of the UKMO NAE .

Enjoy its view of the upstream pattern versus the GFS ! this is from the 03hrs run, still waiting for the 09hrs , that will be interesting as it will be the latest data available from the USA before the evening model runs come out.

These are the respective charts to 03hrs on the 13th:

SREF

GFS

All those lines on top of all the model viewing in the last few days is beginning to take it's toll Nick, can you explain the bottom chart please before my head explodes. Edited January 10, 2013 by Sceptical Inquirer

[WhiteFox]

Having had a lesiurely look at the models this morning many hours after they have completed (really don't understand why anyone would sit impatiently clicking refresh every few seconds - own up if you do!) it is interesting to see what has changed. Too many posts saying "Nothing has changed, GFS and ECM stick to their guns". They do not.

As has been mentioned, the GFS does try to send the key shortwave SE but ultimately fails. It is a small step towards the ECM in this regard and I think Steve M is quite correct: more ensembles will show this solution this afternoon, but the Op may still lag. We will see.

As for the ECM is most definitely not the same. Over the past few runs we have seen it trend towards bringing the jet closer to the south of the UK. THe result of this is that the cold has become more marginal; on the 8/1 we saw deep cold embedded and a strong Greenland High in charge with the jet going well to the south. Yesterday saw the jet creeping closer bringing a series of undercutting systems delivering heavy snow. THis was always going to be more marginal and has been proved so this morning with the snow line being pushed much further north and east than yesterday's runs. In this respect, Ian Brown is correct; it won't take much of a shift for the ECM to bring milder weather in, albeit via a different route than the GFS.

So, the ECM underestimated the strength of the jet earlier in the week, but still sends the energy far enough south to deliver some cold to all areas with the North being more favoured on the overnight outputs.

The GFS seems to have picked up on the strength of the jet earlier, but sends more energy to the north which then cuts off the cold flow earlier than the ECM. The GFS may be inching towards sending the shortwave south east which would mean that it was incorrect with where the main energy goes, but it has been correct with forecasting the increase in the energy.

Just to illustrate the trend in the ECM; T168 on 09/01 12z

T96 on 10/01 0z:

As you can see, the approaching system is further north (less blocking over Greenland, although that has been subsiding for several runs now) and appears to be heading ESE instead of SE on the 12z yesterday.

So, once we finally have a resolution on the shortwave epic for Monday, we will have another epic starting for the Atlantic attack which follows. I'd get your energy back before tackling that one!

[nick sussex]

All those lines on top of all the model viewing in the last few days is beginning to take it's toll Nick, can you explain the bottom chart please before my head explodes.

You're not the only one! lol

If you look at the GFS it takes two shortwaves ne , it weakens the first one which isn't what we want to see.

The SREF takes one deeper low likely to track further towards western portions of Greenland and doesn't drag another shortwave with it.

I don't know how good that SREF is but both it and the NAM both disagree with the GFS on their morning runs at both 03hrs and 06hrs.

PS I've edited the original post, the proper GFS one should now be showing.

Edited January 10, 2013 by nick sussex

[Matt Jones]

That may be so Exeter not using GFS but it was that model that started to rapidly disolve the December cold spell -so it got that spot-on. Sorry to seem a tad contentious but ignoring GFS because UKMO doesn't use it and no doubt driven by the desire of nearly all of us to see something a little bit special in cold weather terms, is colouring our logic possibly. I feel GFS done it again with my only hope being the huge SSW currently inplay throwing a spanner into the computers logarithms meaning they*all* are throwing out wild speculative options ?

I think we have to remember the 90% probability warnings of snow issued friday morning in December by the Met only to be then obliturated by all models that evening, this time around a more cautious 70% has been issued, rightfully so as we all know this is still finely balanced, though the 6-15 day forecast is alot more confident of cold and snow surprisingly.

Massive 12Z's for UKMO ECM later.

[Methuselah]

That may be so Exeter not using GFS but it was that model that started to rapidly disolve the December cold spell -so it got that spot-on. Sorry to seem a tad contentious but ignoring GFS because UKMO doesn't use it and no doubt driven by the desire of nearly all of us to see something a little bit special in cold weather terms, is colouring our logic possibly. I feel GFS done it again with my only hope being the huge SSW currently inplay throwing a spanner into the computers logarithms meaning they*all* are throwing out wild speculative options ?

I am sure that there many reasons why the professionals (perhaps even those who created the algorithms?) are currently not paying too much attentions to the GFS; and the fact that it was the first to crush our hopes last December, doesn't really figure among those reasons...Like you say, the ongoing SSW (being unprecedented) must be the cause of many a headache...

[Ian Brown]

If people have read some of the things that I have intimated over the years & what ian has confirmed in his posts over the weeks-

* The GFS has poor handling of high pressure systems to the NE- it always has-

* The GFS has a strong atlantic bias- particular on the 06z & 18z suite- especially the ensembles- ie to much energy

* Exeter, whilst viewing the 06z dont take any stock really, all they do however is use it to determine confidence- I have NEVER seen the fax chart modified to the GFS away from UKMO/ECM

* The ECM ensembles & mean are a LOT keener on a the shortwave dropping south than the GFS-

* The fact that the METo have updated next week to ignore the entire GFS shows there is zero confidence-

* THe ARPEGE info MUST be strong-

here is the 12z predictions

UKMO- no change

ECM- westward correction to the UKMO with respect to mondays low-

GFS- a big SW correction & 50% of the ensembles move to the EURO solution.....

S

That is a big call Steve when you look at those GFS ensembles for just T96. I don’t expect retreat from the GFS, and the 12z is often progressive/aggressive with the jet.

It’s hard to know what to expect from the UKMO as it is often slow to pick up on the correct pattern change. But the ECM will I think buckle towards the GFS, there is already no margin for error with how it handles the key timeframes.

[Catacol]

That may be so Exeter not using GFS but it was that model that started to rapidly disolve the December cold spell -so it got that spot-on. Sorry to seem a tad contentious but ignoring GFS because UKMO doesn't use it and no doubt driven by the desire of nearly all of us to see something a little bit special in cold weather terms, is colouring our logic possibly. I feel GFS done it again with my only hope being the huge SSW currently inplay throwing a spanner into the computers logarithms meaning they*all* are throwing out wild speculative options ?

I am not sure how the "only hope" can be a SSW throwing a spanner into the GFS works. Isnt the fact that every other model bar GFS and NOGAPS at the moment is showing continued northern blocking a pretty good reason to suspect that GFS may be wrong? Not saying it WILL be proved wrong... but I dont really understand why everyone this morning is so certain that a GFS solution is the one to follow - especially bearing in mind that the REAL big 2 (ECM and UKMO) have a different outcome.

Also - and this is a personal statement - years of watching GFS has convinced me that what SteveM put out a few mins ago is correct. GFS frequently over eggs the atlantic, and is shocking at getting the track of LP systems that head towards the UK correct. My own observation is that it is particularly poor when systems push up from Biscay, but not a lot better when they are coming from W or NW. UKMO and ECM have been better at getting the details right in this area for years.

[The Northern Ramper]

All those lines on top of all the model viewing in the last few days is beginning to take it's toll Nick, can you explain the bottom chart please before my head explodes.

yeah, and for those viewing on a mobile.

[Timmytour]

I hope the GFS tracks back to what it was showing a couple of days ago, but it's had a few consistent runs since. Consistent doesn't equal right, but as Steve Murr himself was hoping for...

Posted 08 January 2013 - 20:38

As for the greenland high we have cause to be very optimistic, the GFS operationals are now at least 6 runs on the trott for a GH & the ECM has now jumped on board-

The ECM had only just jumped on board. Then the GFS went consistent the other way. So perhaps the ECM will jump off again

The same causes to be very optimistic there are the same reasons to retain a degree of pessimism. I like the reasoning Steve is putting forward as to why the GFS might change, but I do understand the reasoning of those who suggest it might prove itself to be the form horse.

Prior the 12zs anything is possible....and it will be the same situation after them!

[SP1986]

Lovely to see the met office have issued warnings for the whole south of england this mornming from 6am saturday until 6am monday with high chance of snow to these areas and could proberbly go in to tuesday Does that mean they completely dissagree with gfs? or just covering there buts?

The GFS doesn't necessarily refute the chance of snow on Monday/Tuesday, but I suspect it will be fairly borderline, that said I think at the moment we can summise there is a fair chance that many areas in the northern sector of the precipitation have a high chance of snow.. where that sector lies still needs to be ironed out.

[nick sussex]

Wow that was a very bullish forecast from DB on the BBC, I'm quite surprised at the level of certainty shown for next week.

They have obviously completely disregarded the GFS, thats a brave call.The plot thickens!

[tight isobar]

Well either the gfs, is immenent for a MAJOR flip,or most (if not all other output) has gone completely bonkers...think the earlier...be a very interesting 12z..

[Southern Storm]

Wow that was a very bullish forecast from DB on the BBC, I'm quite surprised at the level of certainty shown for next week.

They have obviously completely disregarded the GFS, thats a brave call.The plot thickens!

Yes nick just seen this and he seemed confident about cold lasting into next week,

Also the new model you posted from the us, is it a high resolution model?

[Harsh Climate]

I think it will all be settled once the 12zs come out. Can see the ECM/UKMO trending towards the gfs but if this doesn't happen and the gfs makes a last ditch backtrack, I think we will be out of the woods and it will be plain sailing for a good cold spell next week.

Also a lot of selective posting going on here, picking parts of a model that support cold whilst ignoring the potential negatives. When If it was the ECM looking bad they would be bringing up how bad the ECM was late December..

[surrey slush]

Wow that was a very bullish forecast from DB on the BBC, I'm quite surprised at the level of certainty shown for next week.

They have obviously completely disregarded the GFS, thats a brave call.The plot thickens!

Heh heh, they've probably been looking at the same upstream data you have, if the double shortwave never forms on any of the models but the GFS, It can hardly come steaming through our block... (Hopefully haven't misunderstood what you were posting)

[SupaSnowy]

Well I'm finding this very interesting, either way one model has to be right, looking forward to this evening. Hopefully we will finally have some sort of idea and agreement on all models, or maybe not haha. Of cause I would like the UKMO or ECM to be right as you can tell by my name.

Gotta say though GFS is being very stubborn and could well turn out to be correct again. But with pretty much everything else going against it surely there is something wrong? Fascinating times.

Edited January 10, 2013 by SupaSnowy

[Paul T]

As far as models go, the NAE @ 48hrs is much more accurate than the GFS @48hrs (due to resolution, I beleive??) so lets have a look:

Mains points:

The NAE has the Low in the atlantic much further North - it is just cut off, but this could be potentially 750 miles different.

The GFS has a much more clumsy apporoach to the low over the SW of the UK. The NAE drives this to a colder solution over the UK @ 48hrs.

The low over Iceland on the NAE is 250 miles further south than the GFS.

Now, assuming the highly regarded short term NAE is correct, then these differences at 48hrs will have a massive butterfly effect down the line.

Edited January 10, 2013 by Paul T

[Paul White.]

Had a couple of texts from friends/family - 'why haven't you told me about this' etc....

Told them milder on Wednesday.

I can't believe they can readily disregard the GFS when the ECM could easily have gone that way and I believe will - later.

Ian if you have called this right then I will total respect for you. You have been brave in sticking with your resolve it what looks like a wall of different evidence. The UkMO ECM bbc GP sm and countless other models. Calling for cold IMO with only the gfs to back you up its very brave. Good luck !

[SupaSnowy]

[quote name='

Now, assuming the massively highly regarded short term NAE is correct, then these differences at 48hrs will have a massive butterfly effect down the line.

[bluearmy]

Had a couple of texts from friends/family - 'why haven't you told me about this' etc....

Told them milder on Wednesday.

I can't believe they can readily disregard the GFS when the ECM could easily have gone that way and I believe will - later.

maybe they are looking at the high res usa modelling of the upstream and disregarding the gfs/gefs solution as a result ?? thank goodness we have nick s on here posting so they can see them !

[AGAL]

I am not sure how the "only hope" can be a SSW throwing a spanner into the GFS works. Isnt the fact that every other model bar GFS and NOGAPS at the moment is showing continued northern blocking a pretty good reason to suspect that GFS may be wrong? Not saying it WILL be proved wrong... but I dont really understand why everyone this morning is so certain that a GFS solution is the one to follow - especially bearing in mind that the REAL big 2 (ECM and UKMO) have a different outcome.

Also - and this is a personal statement - years of watching GFS has convinced me that what SteveM put out a few mins ago is correct. GFS frequently over eggs the atlantic, and is shocking at getting the track of LP systems that head towards the UK correct. My own observation is that it is particularly poor when systems push up from Biscay, but not a lot better when they are coming from W or NW. UKMO and ECM have been better at getting the details right in this area for years.

I can only hope you are right about the GFS having its foibals but we went through those in December and it's foibals were not to be denied.

As for SSW: Well its obvious I'm a lay person who maybe was sceptical of the SSW event but if it is a masive a deal as being purported then surely the models can't easily account for this extra dimension, blimey they struggle anyway trying grasp the three dimensional troposphere let alone this stratosphere putting its oar in. Just my humble view and hope

[Weathizard]

Can I just say there has been some fantastic posts on here today, really good discussions going on.

As far as models go, the NAE @ 48hrs is much more accurate than the GFS @48hrs (due to resolution, I beleive??) so lets have a look:

Mains points:

The NAE has the Low in the atlantic much further North - it is just cut off, but this could be potentially 750 miles different.

The GFS has a much more clumsy apporoach to the low over the SW of the UK. The NAE drives this to a colder solution over the UK @ 48hrs.

The low over Iceland on the NAE is 250 miles further south than the GFS.

Now, assuming the highly regarded short term NAE is correct, then these differences at 48hrs will have a massive butterfly effect down the line.

Very good post, good to see the NAE looking a lot better, would be very unlike this model to be so far out.

[WhiteFox]

I can't believe they can readily disregard the GFS when the ECM could easily have gone that way and I believe will - later.

Maybe/maybe not. Looking at the comparison charts for the crucial stage at midnight Saturday, you can see that the angle of the high over Scandinavia has altered slightly tilting more South/North than SW/NE. This is what led some of the gun-jumpers to proclaim that the shortwave was heading SE. If the tilt of the high changes slightly more on the 12z then SE it goes. The ECM still retains the SSE/NNW tilt and remains convinced that this will be the case. I think it would take more for the ECM to go the way of the GFS than vice versa, but the difference isn't great as you point out. I'd say 60/40 ECM at the moment.

GFS 0Z:

GFS 6z:

[Coast]

Model Thread Reminder Post

Please keep your posts on topic in this thread - ie related to and discussing the model output.

If you're feeling like a rant or a ramp about the model output, or want to sound off about the models without actually discussing them then we have a thread setup for that here:

http://forum.netweat...n-ramp-go-nuts/

If you're finding this thread a bit too fast paced and chatty for your liking, we have a slower paced, more in depth thread available here:

http://forum.netweat...-and-summaries/

This thread is very popular and very fast moving, but that doesn't mean it's a free for all so please help keep it interesting and on topic for all those using it by making sure you're post is relevant to the topic before hitting submit.

Thanks!

[BurwellWeatherWatch]

Maybe/maybe not. Looking at the comparison charts for the crucial stage at midnight Saturday, you can see that the angle of the high over Scandinavia has altered slightly tilting more South/North than SW/NE. This is what led some of the gun-jumpers to proclaim that the shortwave was heading SE. If the tilt of the high changes slightly more on the 12z then SE it goes. The ECM still retains the SSE/NNW tilt and remains convinced that this will be the case. I think it would take more for the ECM to go the way of the GFS than vice versa, but the difference isn't great as you point out. I'd say 60/40 ECM at the moment.

fantastic observation. Yes- this will hopefully be rseolved on the 12s but even then it could go to the wire on the 0zs. Hopefully another twitch or more from GFS this afternoon. I'd be inclined more to 70-30 in favour of ECM, particularly in light of latest METO thinking. Wonder if IB's family will be wandering through a snow drift next Wednesday in bermuda shorts and t-shirts

But I don't think there'll be need for any humble pies or castigating of any models whatever the result: the differences are so microscale, that it's not a big deal in the grand scheme of things - just makes a huge deal for our weather in the UK!