Model Output Discussion - Colder but how cold and for how long?

[Rob 79812010]

I've got to hold my hands up here. Was critical of the ensembles graphs last night and said beeb would get it right. They're actually starting to fall in line now with the models

[Nick2373]

00z shows currently where we are and where the next direction of travel may be, like some have alluded to in post look towards Scandi there looks like some sort of pattern change coming up for the 27th onwards but I’m happy with the output it’s dry we aren’t under a Atlantic influence dry just above temps or below that’s good better than being underwater which many have. 

[danm]

1 hour ago, Man With Beard said:

I appreciate it doesn't happen every time but I'm always mindful of January 2010 - the time the entire country (except Bournemouth!) went white due to a feature only modelled within T48 - and that was on a northerly.

Correct, I remember this event only being forecast a day or two before it happened, it ended up giving many in the south a big dump of snow, particularly central Southern England. 

 

Nothing widespread in terms of snow showing at the moment for next week, but that is usually the case except in situations where we have battleground snow events. Many easterlies often look dry, but then tend to produce more snow than is modelled several days away. 

Two other examples of short term forecasted, big snow events from northerlies for areas not prone to snow showers in northerlies - last December's 6 inch dumping for many in the S/SE was forecast at short notice, the embedded cold came from a northerly. 18th December 2010 - another potent Arctic blast that many thought looked dry, except for the usual spots prone to the wishbone effect - yet a circulation developed that again dumped several inches over the south. Not suggesting any of this will happen next week, but you can't write anything off. 

Edited January 12 by danm

[Battleground Snow]

1 minute ago, Mcconnor8 said:

I think it would end up *something* like P34 on the ECM 0z run, ridging back into Greenland with the cold hanging on into Tuesday/Wednesday next week instead.

I think the output is sniffing the scandi heights maybe a couple of days too early, with mjo only getting into phase 6 around the 20th.

With the lag a more realistic timeframe is probably around the 27th/28th @Scott Ingham

If we can keep the pattern cold with wedges until then, that would be a bonus

[That ECM]

I wouldn’t call from 24th onwards. Too much going on in the output to be confident in anything. With the nhp as is a few tweaks good or bad would look very different later. FWIW I’m encouraged by what ukmo shows and how gfs is moving to it. 

[MattStoke]

7 minutes ago, MJB said:

Frontal snow ..............................yeah

 

If that stalls a bit, might get a covering before it goes wet 

Front clearly disrupting up against the high pressure to the east. Probably won't even make it into the UK in reality.

[Midlands Ice Age]

1 hour ago, shawty1984 said:

You can have all the modern technology you want, you are trying to predict the future, that's simply impossible on a scale you're asking for  

I am afraid that we are now getting to the point that it is now the errors and assumptions in the software are causing the greater errors.

Until you get those sorted out you will just magnify the errors, and end up with even more chaos.

Spoken after 40 years in the software industry.

MIA

 

[Ice Day]

Incremental improvements on the GES mean out to 126 vs the 0z. Nothing major, but it's moving towards the UKMO (and Icon?) solution, bit by bit.

 

  

  That's far enough out for me now, the uncertainty kicks in around this point (a bit before actually).  Adios until the 12z's 

[offerman]

9 minutes ago, Midlands Ice Age said:

I am afraid that we are now getting to the point that it is now the errors and assumptions in the software are causing the greater errors.

Until you get those sorted out you will just magnify the errors, and end up with even more chaos.

Spoken after 40 years in the software industry.

MIA

 

Brilliant post, and thank you for confirming all your experience in the industry as that is gold to me and the fact you have acknowledged about the changes and also what needs to be done to prevent this happening so much i.e. software issue. 

[IDO]

Looking at the T210 chart it is horrible; a harbinger of changeable weather for the foreseeable with the tPV circling to the N/NW:

T210: 

Fast Forward three days and deja vu:

T306: 

Consistent stuff from GFS in FI rather than chaos, and usually that means a strong signal.

[bradymk]

An outside shot at the moment but when the ridge from the ESB moves across the Atlantic the end of next week, a stronger ridge early on and more energy going SE (usually poorly modelled)  would allow it to be further north than the current GFS run. Similar theme to ICON 00z and this would give the chance of a decent wedge forming to our north around Day 8, helping deflect Atlantic energy south eastwards. If so, the wedge could then lead to a development of a Scandi High as others have said.

GFS was never going to make it then, but there is a hint of this on the 6z Day 8 & Day 10 charts.

In the next few days, I’d be looking in the models for a cut off wedge forming around Iceland / just to the east of Iceland approx Day 8.

 

[TSNWK]

3 minutes ago, IDO said:

Looking at the T210 chart it is horrible; a harbinger of changeable weather for the foreseeable with the tPV circling to the N/NW:

T210: 

Fast Forward three days and deja vu:

T306: 

Consistent stuff from GFS in FI rather than chaos, and usually that means a strong signal.

Long way back to cold with that.. embrace the next week team...it could be the only decent shot this year

Edited January 12 by TSNWK

[A.J]

well, the GooFuS model moves the channel low further south by the looks of it, and it seems to be a bit further south with each run ......the way its going, If I lived in the Sahara Desert, I'd be getting rather excited and starting to ramp to my fellow camels 

[That ECM]

4 minutes ago, IDO said:

Looking at the T210 chart it is horrible; a harbinger of changeable weather for the foreseeable with the tPV circling to the N/NW:

T210: 

Fast Forward three days and deja vu:

T306: 

Consistent stuff from GFS in FI rather than chaos, and usually that means a strong signal.

Save this one and see if resembles reality. I honestly don’t think it will. Time will tell.

[Derecho]

A huge increase in the uncertainty of the EC ensembles in the space of 24 hours between T96 and T120.

I suspect we could see some dramatic swings in the model output through the weekend with things becoming substantially clearer by Sunday evening for later next week.

GFS is even more uncertain during this time period. Usually by T72 the spread reduces markedly.

Edited January 12 by Derecho

[joggs]

10 minutes ago, IDO said:

Looking at the T210 chart it is horrible; a harbinger of changeable weather for the foreseeable with the tPV circling to the N/NW:

T210: 

Fast Forward three days and deja vu:

T306: 

Consistent stuff from GFS in FI rather than chaos, and usually that means a strong signal.

It's gone reset mode imo. That fly's against a lot of background signals and in the nearer time it's trending towards UKMO.

Don't but into any det runs after around four days imo

[josh-weather]

29 minutes ago, Midlands Ice Age said:

I am afraid that we are now getting to the point that it is now the errors and assumptions in the software are causing the greater errors.

Until you get those sorted out you will just magnify the errors, and end up with even more chaos.

Spoken after 40 years in the software industry.

MIA

 

But model performance hasn't dropped... Has it?

[IDO]

5 minutes ago, That ECM said:

Save this one and see if resembles reality. I honestly don’t think it will. Time will tell.

...and by D16:

Let us hope you are right, but this represents the mean and is consistent with the last 2-3 days' FI output. Looking on Twitter, I do not see anything that would make this an outlier scenario. Since repeated designs are a good starting point for Winter patterns, we got 3+ weeks of zonal after the previous cold spell. But as you imply, just for fun, or not!

[Derecho]

40 minutes ago, Midlands Ice Age said:

I am afraid that we are now getting to the point that it is now the errors and assumptions in the software are causing the greater errors.

Until you get those sorted out you will just magnify the errors, and end up with even more chaos.

Spoken after 40 years in the software industry.

MIA

 

Without working on the ECM, GFS or UKMO models, I don't know how you can prove this without any evidence?

If you were just letting machine learning algorithms take over the models without an understanding of how the output is created, then yes I'd agree... but every piece of software is different.

The errors in forecasting are caused by turbulence and a lack of observations in key areas.

Edited January 12 by Derecho

[That ECM]

6 minutes ago, IDO said:

...and by D16:

Let us hope you are right, but this represents the mean and is consistent with the last 2-3 days' FI output. Looking on Twitter, I do not see anything that would make this an outlier scenario. Since repeated designs are a good starting point for Winter patterns, we got 3+ weeks of zonal after the previous cold spell. But as you imply, just for fun, or not!

Can’t keep up on here let alone going on any other sites so not sure what has been said or not,  genuinely.  I of course could be completely wrong and we end up there. Jeez I hope not.  

[Catacol]

1 hour ago, Cambrian said:

The models are struggling big time. The Scandinavian trough is still very much in attendance and looking at that, a reasonable observation would be that it only needs just a little more amplification in the Atlantic heights to altogether prevent any phasing of the troughs associated with two lobes of the PV and keep us very much under the influence of the cold trough to the northeast. 

This analysis nails it. It will be a very close run thing, and some average temps thrown into the mix in week 2 is an inevitability I think - but cold air will also be present and close by. These are the sorts of scenarios that, in theory, produce the snowiest conditions for the UK. It would be counter intuitive - but entirely possible - that the very cold week next week produces less of a snow impact than the milder one that follows. And then all eyes on the expected impacts of spiking AAM and MJO orbit as we hit February and that lovely Glosea chart.

As for next week - well: there is a huge amount to be worked out. We are not close to knowing what might be embedded in the flow come Mon/Tues/Weds/Thurs nor what may or may not encroach from the SW. This is where the luck comes in - macro scale has done its job and it is time for the micro to sort itself out over a 5-6 day period. Once this warm ridge gets shunted to Greenland then the fun begins...  Let's hope for a fair slice of that required luck because snow hunters on this island, in the best synoptics and large scale pattern in 6 years, deserve a bit!

Edited January 12 by Catacol

[booferking]

Gefs 7 has it right.

[Mcconnor8]

Still some snowier perturbations in the mix, GEFS P13 for example.

 

 

[bradymk]

27 minutes ago, bradymk said:

An outside shot at the moment but when the ridge from the ESB moves across the Atlantic the end of next week, a stronger ridge early on and more energy going SE (usually poorly modelled)  would allow it to be further north than the current GFS run. Similar theme to ICON 00z and this would give the chance of a decent wedge forming to our north around Day 8, helping deflect Atlantic energy south eastwards. If so, the wedge could then lead to a development of a Scandi High as others have said.

GFS was never going to make it then, but there is a hint of this on the 6z Day 8 & Day 10 charts.

In the next few days, I’d be looking in the models for a cut off wedge forming around Iceland / just to the east of Iceland approx Day 8.

 

GEFS Pert 7 has the idea that I am thinking 

Like I said outside chance, but this is what I am going to look for in the next few days to start appearing.

Edited January 12 by bradymk

[Jason M]

One of those situations where lots of short term inconsistency and different routes that can be taken, but all roads seem to be leading to the same place by next weekend.

I suspect next week won't be quite as dry as the charts currently look. Thereafter I was hoping to see height rises towards Scandi but the window for that seems to be closing rapidly and we are not really seeing much if any suggestions of it happening in a meaningful way in any of the ensemble suits (not seen the latest GEFS yet).

So, in summary a decent winter cold spell with some snow about but right now its not looking like anything memorable for most people (deliberate use of 'most' here instead of 'all' before any toys are thrown!). Snow potential could increase next week though and I wouldn't be surprised to see some shortwaves get modelled through the channel even if the main low stays well south.

Longer term modelling is suggesting strongly a much milder, wetter end to the month and it will be interesting to see how that plays out against the background drivers being discussed on here. Place your bets!