That said, we can't be sure right now that GFS isn't falling into the same trap it has fallen into so many times in the past of being too quick to bring about the demise of a block and have the atlantic coming back in quickly. I have seen this happen more times than I care to remember.

We can only hope.  At least the meltdown here is not as bad as netweather.  But the positive is that we can save on heating bills, which is a real issue these days compared to previous cold spells like 2010

The flipside of the stronger Azores High seems to be that it pushes the cold down a bit faster and further in advance, and this morning's GEFS is actually the coldest set of the winter in terms of -10 in London: 22 of 32 members show it, one up from the previous high of 21 just before the data-related flipping at Christmas. 

This time around the changes have been more gradual, but suffice to say my advice, at least for those in the south, is enjoy the cold at the weekend and early next week.

Still plenty of time for tweaks, of course, and once the models have picked up on a return to Atlantic flow they do tend to be a bit too hasty, so I wouldn't be surprised to see the cold hanging on until midweek next week down here.

The ECM ensembles yesterday were interesting, with the ECM-46 showing a very cold week two, but whites and pinks on week 3. This has become a bit of a theme and suggests the Atlantic returning... but a bit later than the op runs are showing at the moment. The 15-day ensembles showed cold or very cold conditions right out to the 12th here (with a median high of 4C and median low of just above zero by then), with most of the scatter on the milder half of the members.

https://charts.ecmwf.int/products/opencharts_meteogram?base_time=202512291200&epsgram=classical_15d&lat=51.3973&lon=0.921564&station_name=Leysdown-on-Sea 

(Other locations are available, use the link above and enter your own town).

It'll be interesting to see whether the 0z set continue the cold picture - I've a nagging feeling the median will be starting to rise a bit and the scatter will transfer to the cold side, something which often happens when a breakdown is being sniffed out. We'll see soon enough!

https://meteologix.com/uk/model-charts/euro/europe/temperature-850hpa/20260110-0000z.html 

The "significant weather" charts, which only go out to 240, show the fronts turning to snow as they move eastwards - a more exciting breakdown than that which the GFS shows.

https://meteologix.com/uk/model-charts/euro/europe/significant-weather/20260109-0000z.html 

Will the ECM shift towards the GFS outcome or is the latter having a wobble? I guess we will find out in the next day or two.

A dying cold front moves south on Thursday, introducing our flow of deep cold air. There's an occlusion close to eastern Scotland too - worth keeping an eye on for those up there!

https://www.meteociel.fr/ukmo/fax/fax60.gif?29-0 

Here comes potential snow on Friday - a small low caught in the flow crosses Ireland...

https://www.meteociel.fr/ukmo/fax/fax72.gif?29-0 

...and it exits to the south 12 hours later. Areas to the immediate NE of that are likely to see snow, but the warm front would suggest perhaps rain is more likely in the warm sector. 

https://www.meteociel.fr/ukmo/fax/fax84.gif?29-0 

There are small troughs in the flow too, but they stay out at sea (for now).

One thing for sure this coming cold spell is not a dry one or talking nonsense of wish bone effect? Those charts plainly show LP right next to us so there should be snowfalls at times especially on Friday with deep cold nights temps over snow fields.  Hope other models to follow ECM today on later runs.

I see a isobar line 1008mb cross over my area in a form of NW flow so hope a Cheshire gap fire up ahead of the front in Ireland.

I think the occlusion will pass by just before the colder air moves in, so away from elevation it would be rain rather than snow I suspect.

On a different note, by day 10 the 00z ECM and GEFS means are poles apart.the difference is remarkable (with the GEFS having changed and the ECM being more consistent with previous runs and the GEFS trending away from cold in the longer term).

GEFS 00z day 10 mean 

ECM 00z day 10 mean 

Is the GEFS sniffing something out, or is it just off on one? Well, the ECM-15 has warmed up too - gone are the median 3s and 4s down here from the 6th-12th, the last few days are now a median 6. The scatter is more evenly distributed now, or even in the colder end - typical of the model when it's playing with an Atlantic breakdown. 

It continues to be a fascinating period of model-watching, though, and even if the Atlantic does come charging back in as per GEFS at least we've several cold days first.

That is indeed one possibility, another could be that it will turn out to be incorrect. It will be impressive if it outperforms the ECM and its ensembles* in this scenario.

(*less cold in the longer term but still below average)

That's yesterday's mean today's is much flatter . Another likely win for AIFS.

https://www.wetterzentrale.de/en/topkarten.php?map=1&model=ecm&var=1&run=0&time=240&lid=AVG&h=0&mv=0&tr=24#mapref 

Automated output has nudged the temperatures up progressively too. Still have three days (Sat to Mon) with a max of 1C which I suppose is a relatively decent cold snap by recent standards. 

It says Tuesday 00z on it?

The difference in heights over Greenland is notable as are the significantly colder t850s.

It doesn't for me, maybe an issue my end. But we know ECM has an issue with over doing heights in Greenland that's why it will almost certainly fall in line to AIFS later. It's ensembles are starting to already. 

https://www.wetterzentrale.de/en/show_diagrams.php?geoid=50893&model=ecm&var=2&run=00&lid=ENS&bw=1 

The fog is beginning to lift, and I’d suggest that option 2 is now strongly favoured, if ever it wasn’t. I won’t go OT, but this is a metaphor for so many things.

This is the ECM ENS mean at t+240.

The ensembles are almost identical at this point to the 12z set for your location (in fact a little bit colder in the short to medium term). The latter part of the 00z ensembles are not shown on WZ at time of writing.

As for the ECM having issues with modelling over Greenland, I wasn’t aware of that statistic, have you got a link?

FAX - current HP regressing to S Greenland tomorrow, allowing LP 976mb to bring strong N-lies down across Scotland by Thu 1st, less so in S England as they get diverted SE-wards; an interesting small disturbance in the flow crosses England NW  to SE on Fri 2nd before the really cold air digs in behind a cold front.

GFS Op - like FAX though the really cold air doesn't affect Ireland and far W Britain. On Tue 6th all  the cold is moved on E-wards as Atlantic air circulates around the 'top' of the Azores HP. This HP drifts in across Europe to 1040mb Switzerland Sat 10th, ridge to S England and mild W/NW-ly winds further north, persisting until an Atlantic trough develops into a Channel-running LP 980mb Tue 13th. This brings colder weather to Scotland while further S another S-ly tracking LP is ready to arrive after Thu 15th.

ECM - agrees generally with GFS though the Azores HP is slower to move in and only reaches Spain by the 10th. This ensures a much stronger W-ly flow across Britain, with LPs in that flow bringing some polar maritime to Scotland and potential snow for the hills e.g. from the LP 970mb N Ireland Mon 12th (might even get some rain in S England from its associated trough?)

GEM - distinctly different after Tue 6th as LP swoops in from Greenland with associated cold air and is positioned over the Highlands 970mb Thu 8th - unsettled for all and looking like a big snowdump over the hills. 

AIFS - London, maxima dropping to give ice days over the weekend and early next week, back up to 5-10C with rain at intervals from Tue 6th (the milder days being the rainy ones). Edinburgh, a mild New year's Day, but otherwise ice days to Tue 6th, then 3-5C maxima with quite frequent rain

GEFS - cold (5-8C below norm) with milder exception Thu 1st and good ens agreement to Tue 6th, then less agreement but generally milder with mean returning to norm (op noticeably mild, not much in the way of cold outliers) and some rain appearing on many days in most runs. Snow for N Scotland from Jan 1st else best chance of a little snow is on the east coast as the milder spell starts. 

The prospects of extended cold weather after next weekend have been written out of the script.

It's certainly something I've noticed, I wish it didn't,  I'm not making it up.

AIFS has the breakdown of the cold on the 7th, I'd be very surprised if the ECm doesn't follow it in the next day or 2.

https://www.wetterzentrale.de/en/show_diagrams.php?geoid=50893&model=aifs&var=2&run=0&lid=ENS&bw=1 

It may very do but to be honest I’m surprised that people put so much faith in relatively untested “AI” models over ones which have been tried, tested and verified for decades.

i have never really noticed the ECM having issues modelling the Greenland region compared with other models. I’m not saying you are making it up just rather it may be a perception thing and it’s not possible to say without any stats. I will try to find some.

I have just come across this article about models’ accuracy, strengths and limitations and thought it was a really interesting read.

https://www.getambee.com/blogs/best-weather-forecast-models 

A fair few years ago now, another lad and myself resized, rotated and overlayed at UK fax chart with the actual corrosponding model output from the UKMO and the match was perfect. No signs of modification at all except the addtion of drawn in fronts, which I do appreciate a lot. Having said that, I am sure there are pattern modifications here and there but I don't think that they would be that significant.