Model Output Discussion - Into 2024

[Mark wheeler]

I think I am having my own little bit of a wobble this morning. GFS not as good for the south with more encroachment from higher uppers . There is a rain to snow event in the south on the GFS control. ECM slightly better than last night's and Gem seems good . 
 

Mogreps not looking quite so pronounced but still looking broadly supportive still but I noticed not all supportive of cold clearing the south coast . Still in the game and will remain positive that this is still well up for grabs .

Edited January 7 by Mark wheeler

[nick sussex]

The Pacific Ridge  is causing issues and the ECM seems unable to decide how much impact that will have .

It is at least an improvement on last nights run so reasons to be cheerful .

Earlier it’s also expanded the area of -10 850s moving westwards . It took its time , hardly been a good few days for it ! 

Edited January 7 by nick sussex

[IDO]

They are good runs in getting a cleaner split-flow of energy from the UK High, though there is no complete UK high retrogression. So, the charts we saw 3-4 days ago with a stonking GH are unlikely, as they showed a full retrogression of the high. I would guess that less energy means that without further help, the clock is ticking on GH's longevity.

Another fork in the road is the migration of the Azores Low. GEM keeps it furthest away by D8, then GFS, but EC pushes it into the UK, delaying the colder air by two days. With that GH clock ticking down, that is likely two days lost to the cold:

ECM is the third chart. Note: There are still fewer yellows on the ECM chart on the tip of Greenland, so the pattern collapses earlier.

It's all out in FI, so we must wait and see where this fork takes us whilst models hopefully firm up on a cleaner height flow to Greenland pre-D8. Wedges after the GH dissipates may prolong the cold for 2-3 more days. The GEFS mean suggests a return to a familiar pattern in FI with no re-amplification on the Pacific side, instead a wedge sending the tPV towards our NW:

Edited January 7 by IDO

[northwestsnow]

5 minutes ago, nick sussex said:

The Pacific Ridge  is causing issues and the ECM seems unable to decide how much impact that will have .

It is at least an improvement on last nights run so reasons to be cheerful .

Earlier it’s also expanded the area of -10 850s moving westwards . It took its time , hardly been a good few days for it ! 

Hear hear ..

EC is better than yesterday's frankly disappointing runs..

Still divergences 120-144 so still hard to call the specifics ..

 

[Met4Cast]

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

[Lukesluckybunch]

4 minutes ago, northwestsnow said:

Hear hear ..

EC is better than yesterday's frankly disappointing runs..

Still divergences 120-144 so still hard to call the specifics ..

 

We await the ensembles..have a feeling it will be mild outlier

[WinterOf47]

1 minute ago, Met4Cast said:

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

The models are not high enough resolution to pick these events up. I was saying this the other day here that a little smudge of activity on one of the main models can be impactful locally. I noticed then a streamer over the South East was forecast for tonight but that seems to have moved to daylight hours tomorrow so I doubt will come to much other than a wet mix. I used to live downwind of Georgian Bay in Canada and they had mesoscale predictive radar out to 48 hours that was much better at picking up local activity…which regularly dumped 20cms on us, when the likes of the GFS had nothing.

[northwestsnow]

2 minutes ago, Met4Cast said:

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

Yes, hopefully the South will do well tomorrow ..

We've not had much snow here but down south the lads and lassies defo deserve a change of luck.

Fingers crossed ...

[Chesil View]

Ecm.gfs ukmo and gem are roughly similar at 168 this morning. Beyond that as Carinthian said earlier various option are on the table.

For me this is a step backwards from the clean evolutions of a couple of days back. Ecm in particular pushing the arrival of any real cold from the north back to day ten yet again and that is only for the north.

It could have it right and the evolution to full on uk cold may just be a bit slower.

Overall though I'm less convinced of a UK wide cold evolution this morning from a southerners point of view and you don't get much more southern than where i am overlooking Chesil beach.

I'll reserve judgement for now as the ensembles are yet to be viewed and there are a lot of moving parts in the evolutions showing now compared to the cleaner evolutions across the models a couple of days ago.

Edited January 7 by Chesil View

[That ECM]

4 minutes ago, Met4Cast said:

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

Agree on both points. Ukmo T168 for me illustrates your last point. 
 

For our hobby of looking at output it can’t be any more interesting with so much going on and variables on a cold theme. T144 plus will be fi for a while and T24 will be fi when looking at detail at some points.

[northwestsnow]

EC mean looks a downgrade to me.

[feb1991blizzard]

Just now, northwestsnow said:

EC mean looks a downgrade to me.

Yes - yet another one.

[loafer]

10 minutes ago, Met4Cast said:

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

I agree - I pointed out yesterday that sea temps are near record highs, so the convective energy is much more than usual.

[Lukesluckybunch]

Ecm mean not great..looks a bit flat to me..and the op is colder than the mean that's worrying

Edited January 7 by Lukesluckybunch
Spelling

[That ECM]

2 minutes ago, northwestsnow said:

EC mean looks a downgrade to me.

Agreed but not much?

[northwestsnow]

00z mean day 10 doesn't even get -5 uppers south of the Scottish border 

Yesterdays day 9

Yesterday's day 10

Edited January 7 by northwestsnow

[Met4Cast]

4 minutes ago, northwestsnow said:

EC mean looks a downgrade to me.

Interested to see what the spread is here, definitely looking for wedges now and ensemble means will be very smoothed out. 

I think the individual EPS members will be a better guide over the means beyond day 5/7.

Edited January 7 by Met4Cast

[Lukesluckybunch]

1 minute ago, That ECM said:

Agreed but not much?

Still a bit of ridging into greenland..but we need to work on that

[Chesil View]

1 minute ago, northwestsnow said:

00z mean day 10 doesn't even get -5 uppers south of the Scottish border 

Yesterday's day 9

Indeed NWS .Scott Ingham day steadily being pushed back on the ECM.

[northwestsnow]

Clusters will perhaps reveal more ..

Let's see ...

[Mike Poole]

1 hour ago, That ECM said:

T168 all 4. never going to be the same but on the same page from a nhp pov.

Just catching up this morning.  So the bottom line out to T168 on the op runs seems to be:

ECM is an upgrade on yesterday, this is a big plus in this timeframe.  It still looks messy North Greenland.  UKMO looks like a downgrade.  GEM is very good, and GFS looks decent and in keeping with recent runs.  What I would say is that there is plenty of uncertainty at a week out, the differences in behaviour North Greenland to Arctic high (or lack of it) is laughable really.

Longer term, I do think ‘whatever’ is driving this pattern, seems to be a slacker driver than was shown a few days ago.  This brings into play high risk high reward snow scenarios, at the expense of potential longevity of the spell.  But on that timescale, the Azores low needs to move east with the southern arm of the jet or it will inflate heights to our south and put southern extent of the cold in peril.  

Edited January 7 by Mike Poole

[nick sussex]

We should bear in mind and something often highlighted in the NCEP extended forecast discussions .

The ne Pacific area has a large drop in observational data .

[That ECM]

6 minutes ago, Mike Poole said:

Just catching up this morning.  So the bottom line out to T168 on the op runs seems to be:

ECM is an upgrade on yesterday, this is a big plus in this timeframe.  It still looks messy North Greenland.  UKMO looks like a downgrade.  GEM is very good, and GFS looks decent and in keeping with recent runs.  What I would say is that there is plenty of uncertainty at a week out, the differences in behaviour North Greenland to Arctic high (or lack of it) is laughable really.

Longer term, I do think ‘whatever’ is driving this pattern, seems to be a slacker driver than was shown a few days ago.  This brings into play high risk high reward snow scenarios, at the expense of potential longevity of the spell.  But on that timescale, the Azores low needs to move east with the southern arm of the jet or it will inflate heights to our south and put southern extent of the cold in peril.  

You said in one post what I tried to say in 5  

[winterof79]

33 minutes ago, Met4Cast said:

Modelling appears to have vastly underestimated this mornings convective shower streamer across the SE (all rain currently) which might bode well for tomorrow, wouldn't surprise me if one or two spots saw a bit of a surprise tomorrow morning!

Longer range modelling is good this morning, MOGREPS/NAEFS remain cold, GFS is okay and ECM is an improvement. I think we've lost a real solid Greenland high overall now so we're going to have to hope we see a decent wedge of heights instead & that still seems likely for the time being.

Probably the most accurate short term precip model is the WRF-HMM

[feb1991blizzard]

1 minute ago, nick sussex said:

We should bear in mind and something often highlighted in the NCEP extended forecast discussions .

The ne Pacific area has a large drop in observational data .

I have always said this is the reason we cannot achieve anything remotely approaching perfection with extended range forecasting, the computing power is there, the understanding of science is there, the programming skills are there, but you will carry on getting diminishing returns without huge increases in weather stations and balloons, potentially in next few years you will have the capability to reduce grid size and increase no. of vertical layers on global models to that of the ukv etc, and still run them to D16, but unless you have obs data, you'll be leaving a lot of the grid blank so there's no point.