Model Output Discussion - 10th Jan Onwards

[Troubleatmill]

Nobody including the MET Know what is going to transpire, they have to give a forecast because that is their Job and a good one they do, but there are no certainties in there. 

 

The 06z ens have a real mixed bag look to them at t168 with an easterly feed or Northesterly or no mans land for a while. until the forces that be decide what rules the roost for our little corner its worth keeping an open mind. Hoping for the best whilst expecting the worst is probably a good idea.

Edited January 14, 2014 by Nicholas B

[Mapantz]

well none of the models have been any good past +96 for a while now so why should we believe what they are predicting for the next TWO weeksDo the met have a crtystal ball?

With their wealth of extra information that they have access to, compared to what joe public get, i'm more than happy to go along with them, regardless of how tricky the situation has been. Their short to mid-range forecasts have been fantastic throughout Autumn and Winter! Edited January 14, 2014 by Mapantz

[Purga]

Odd straggler cold runs but overall very slim chance of anything cold

 

With METO of the same opinion it's best to forget this current tease and wait until February at least.

[bluearmy]

we appear to be going around in circles and then coming back to the same place. the original mean pattern of blocking to out ne, candian vortex around hudsons/baffin and mid atlantic high waxing and waning as the pulses of jet come around the top and then sink into europe. this trough disruption is proving difficult to pin down. last week, the euros decided to drop it towards iberia and now its modelled further east. there appears to be a decent ens agreement that passing just to our east into holland is the current favourite. my take is that the blocking will probably prove rather stronger to our ne and as such, the troughing will be further west. generally in our lontitude. as nw europe cools in comparison to this coming week, we will have opportunities for undercutting if the approach is to our west and we draw in a se flow. if the trough drops in over us, we can still see a pretty cold outcome. if it passes to our east, then we are always receiving a pm flow which is likely to have dp's that bit too high for any widespread snowfall. its certainly looking to continue to be a fascinating period of model watching and you can bet that runs akin to yesterdays ecm op will pop up from time to time. you never know, they may even verify one day !!

[Dancerwithwings]

Talking of the past 96h, Heres how the top three models are dealing with things atm 'Not in order i must add'. Below is what the models were showing last Thursday for this Tuesday.

Make your own mind up, But looking at the UKMO, Well top marks must be said.

 

 

ECM

 

 

UKMO

 

 

 

GFS

 

Edited January 14, 2014 by Dancerwithwings

[johnholmes]

 I think John Holmes was alluding to this a few days ago, that he felt this was the more likely direction of the UK receiving colder temps, rather than from the East or NE. Apologies if i've 'misquoted' JH.

 

yes that was and remains the way I see it, I'll do an update later today-in a rush just now

[nick sussex]

DWW, the ECM I think you've forgotten to put the +hours forward on the first chart, but the ECM still was the most out looking at its prediction, its funny it still always manages to be first in the verification. Unfortunately where as for most of the year most of us aren't looking at the fine detail any errors become very much highlighted in the winter season.

[nick sussex]

The amount of solutions showing at just T96hrs on the ECM postage stamps hardly inspires confidence that the models have got this sorted out yet:

 

http://www.ecmwf.int/samples/d/inspect/teasers/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!96!pop!od!enfo!enplot!2014011400!!/

 

A variation on the tilt and shape of the troughing which impacts on how the following output could develop.

 

Then later out  what happens to the shortwave energy near the UK which again seems to be another point of contention:

 

 

It depends on whether you see a clean split in that and early enough to allow the ridge to the east to extend a little further west and force more trough disruption of the next incoming low off the Atlantic.

 

Today whats happened is that we've seen a slightly flatter pattern over the ne USA  around T144hrs which impacts the energy spilling off the PV chunk to the nw.

 

Will that be the same tonight or will we see more changes?

 

 

 

 

 

[Man With Beard]

Talking of the past 96h, Heres how the top three models are dealing with things atm 'Not in order i must add'. Below is what the models were showing last Thursday for this Tuesday.

Make your own mind up, But looking at the UKMO, Well top marks must be said.

ECM

UKMO

GFS

Yes indeed. Aside one run last week, the UKMO was almost always closest to the mark as far as it went out. And IMO this is often the case for the UK, whatever the global verification stats say.

Thank you for doing this, I'd love to see more analyses like this, often we leave charts at t120 to go off searching for the next cold shot and we never see which model really was closest

[ARNOTT ad]

It seems to me that were all straw clutching at the slightest changes in model output in the last week or so. The ecm last night was a glimmer of hope but is now more or less back to square 1. The charts frustratingly seem to be stuck in a rut with the uk in no mans land and in n out of milder uppers compared to East and west of us.
I for one think it's game over for this winter because the met office have resumed there status quo forecast well into feb (when days get longer and the sun stronger). They have mongreps and other information at there disposal which is obviously showing no potential. Whilst at the same time we try to find any crumb of comfort from the ecms easterly and gfs NWesterly etc which just simply is very unlikely to come to fruition So a very poor winter in my books.

[Dancerwithwings]

Nick sussex, You mentioned "  I think you've forgotten to put the +hours forward on the first chart,..... I know the others had 13:00hrs showing, But on the ECM It was on 1:00 hrs for the 00z output and i couldn't get 13:00hrs for today ...... If that's what your on about that is

[blitzen]

No but they correctly predicted no cold this week compared to what the models were showing for a few days last week

 

They also have there in house models like MOGREPS

No cold this week for where?   3.3c max here yesterday and currently sitting at 2c.( and has been all day)   Certainly chilly up here and I am in one of the milder areas.

[syed2878]

again the few models showed the cold outcomes has backtracked the next day that's y last nights ecmf wasn't worth getting exciting about. The ecmf been quite poor this winter the cold cynoptics allways ben in the unreliable timeframe of 144h-240h. Until the met come on bord forecasting a cold spell it's not worth getting excited over.

[Man With Beard]

OK one more comparison if that's ok. Below is the ECM forecast for tomorrow morning at T24:

 

And now the ECM postage stamps for tomorrow as they were showing last Friday morning (T120)

 

http://www.ecmwf.int/samples/d/inspect/teasers/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!120!pop!od!enfo!enplot!2014011000!!/

 

Out of 50 members, I would argue just members 2 and 5 got the low next Iceland more or less correct. That's a really shocking return at just 5 days out! The majority of the members - in unison, I would emphasise - had the low hundreds of miles too far west - you might have thought (as indeed I did) that there were enough members going for a held-back pattern to be fairly confident in this verifying - but no - ensemble agreement was clearly not a precursor for a confident forecast. Certainly the UKMO operational outshone all but two of the fifty ECM ensemble members.

 

Looking ahead once again - although I never write off a Scandi High, as they always seem to have secret powers to defy the Atlantic and branch west - I think my lesson from the above is to imagine the ECM pattern 300 miles further east in its latter stages - if it still looks good, game on - if it then looks like game over, be very, very cautious until T96 or even T72. Something to remember at about 6.40 this evening!

[Cheshire Freeze]

OK one more comparison if that's ok. Below is the ECM forecast for tomorrow morning at T24:

 

And now the ECM postage stamps for tomorrow as they were showing last Friday morning (T120)

 

http://www.ecmwf.int/samples/d/inspect/teasers/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!120!pop!od!enfo!enplot!2014011000!!/

 

Out of 50 members, I would argue just members 2 and 5 got the low next Iceland more or less correct. That's a really shocking return at just 5 days out! The majority of the members - in unison, I would emphasise - had the low hundreds of miles too far west - you might have thought (as indeed I did) that there were enough members going for a held-back pattern to be fairly confident in this verifying - but no - ensemble agreement was clearly not a precursor for a confident forecast. Certainly the UKMO operational outshone all but two of the fifty ECM ensemble members.

 

Looking ahead once again - although I never write off a Scandi High, as they always seem to have secret powers to defy the Atlantic and branch west - I think my lesson from the above is to imagine the ECM pattern 300 miles further east in its latter stages - if it still looks good, game on - if it then looks like game over, be very, very cautious until T96 or even T72. Something to remember at about 6.40 this evening!

 

I've said numerous times that ensembles are virtually useless....I've seen far too many examples of scenarios such as this. The GEFS seem even worse (flip flop flip flop). If we can't have any faith in the operationals at day 4/5 and no faith in the ensembles either, we may aswell just give it up as a bad egg.

 

The ensembles are the result of tweaked starting conditions to attempt to account for modelling errors at initialisation and to try to gauge the sensitivity of a given outcome to such potential changes. However, the atmosphere is a complex entity 20 or even 50 perturbations are a drop in the ocean as to how many potential dynamic vagaries there could be. For instance it's possible that there is one atmospheric nuance which isn't accounted for by the tweaked starting conditions and it is this nuance which is the catalyst for major alterations to the predicted outcome (a keystone variable if you like)....in this case all of the ensembles would be rendered useless.

 

Model accuracy has certainly improved in general...especially within the confines of the usual W-E zonal flow. Any scenario outside of this comfort zone however and I'm not really sure we've advanced very far at all from where we were 20 years ago.

Edited January 14, 2014 by CreweCold

[Summer Sun]

Latest NAO update keeps it at neutral for another 2 weeks at least

 

 

Latest AO update keeps it negative

 

[Man With Beard]

I've said numerous times that ensembles are virtually useless....I've seen far too many examples of scenarios such as this. The GEFS seem even worse (flip flop flip flop). If we can't have any faith in the operationals at day 4/5 and no faith in the ensembles either, we may aswell just give it up as a bad egg.

 

Model accuracy has certainly improved in general...especially within the confines of the usual W-E zonal flow. Any scenario outside of this comfort zone however and I'm not really sure we've advanced very far at all from where we were 20 years ago.

 

Yes I'm coming round to your thoughts on the ensembles, I followed them carefully last summer and they were much, much better, maybe it's just winter battlegrounds they can't do ... As for the op runs, hopefully it's not all doom and gloom! After all, the UKMO managed to nail the pattern for tomorrow run after run (except for that crazy morning last Wednesday!) Although I'm increasingly aware that the ECM op is often too far west in these situations, I do also recall it doing pretty well on the easterlies of last February and March. And of course the famous November 2010 cold spell was counted down like clockwork from T240 by all the ops, including the GFS!! 

 

So ensembles, agreed, rather shaky, but op runs, I still have a little hope!!

Edited January 14, 2014 by rjbw

[nick sussex]

And yet more changes to the trough orientation and position, upstream its a bit more amplified on the GFS 12hrs run at T84hrs.

[The Eagle]

I think this is slightly better at 90 hrs.

[Cheshire Freeze]

And yet more changes to the trough orientation and position, upstream its a bit more amplified on the GFS 12hrs run at T84hrs.

 

Stark changes at just 84 hours out Nick!

[Man With Beard]

I think the GFS 12Z is about to send us an easterly as well - at T96, low pressure swivelling east to west on the northern side, high pressure poised:

 

[winterof79]

Digging further south at 96 than the 06z and better heights to n/e

http://modeles.meteociel.fr/modeles/gfs/runs/2014011412/gfsnh-0-96.png?12

http://modeles.meteociel.fr/modeles/gfs/runs/2014011406/gfsnh-0-102.png?6

[nick sussex]

Stark changes at just 84 hours out Nick!

Yes its much better but we need to see the splitting of that shortwave energy later on, given the way its going I wouldn't be surprised to see the other models still disagreeing regarding the troughing near the UK.

 

No phasing with the upstream low is crucial here.

Edited January 14, 2014 by nick sussex

[Cheshire Freeze]

Yes its much better but we need to see the splitting of that shortwave energy later on, given the way its going I wouldn't be surprised to see the other models still disagreeing regarding the troughing near the UK.

 

Aye a pathetic showing by the NWP at present, I must say.

[The Eagle]

108 - this is close!