Yes Darren.  Its a case of hunkering down with a warming Brandy and checking in either once or twice a day after the ensemble data is there to see how representative the `ops are being of their ensemble suites.  This is not a blunt knife-edge either - it looks to be very sharp indeed.

Look at that spread for the 23rd. Shows the uncertainty.  More of a defined cold cluster after that but conditions 23/24th completely up in the air. 

Some very down to earth posts on here today!     Is it Sensible Sunday?  

A run-in wobble is all part of the fun isn’t it? 😂😎

The GEM ensembles are certainly sparkling today - here's London. This is especially interesting as they tend to be less keen than other models (even ECM!) when potential cold weather is in the offing. It's one of the coldest suites I've seen at this range.

And for those wondering what that purple run is doing, it's a cold ridge over snowfields. Here's the noon T2M chart, just for fun! (And yes, that's a -10C noon temperature over the East Midlands!

(Charts from Meteociel)

ECM ensembles look pretty good still.  Op is definitely an extreme run. Huge uncertainty still though. Christmas day onwards remains unknown.  

https://www.meteociel.fr/cartes_obs/ecmwfens_display.php?x=317&y=140&run=0&type=10&runpara=0

I notice public forecasters are very unwilling to stick their neck out and say anything about the Christmas period despite being pressed to do so, another measure of how uncertain the situation is, less than a week out.

FWIW the charts this morning show:

WX summary - Atlantic clearly winning for everywhere S of the baltic with above-norm tems week 1 and even milder week 2. Any v cold air over NE Europe has dissipated. Where the mildness goes, pptn follows.

GFS op - current HP collapses to a col between LPs on Atlantic and over scandi by Fri 24th but interestingly the Atlantic side develops into a very intense but tiny 965mb low over Brittany Sun 26th pulling in NE-lies for northern UK (but milder across Europe to the S of its track). It moves on into Baltic with a legacy of cold over UK before another col Thu 30th. The Atlantic low then moves across Scotland 980mb Sun 2nd, its more N'y track dragging up milder SW-lies

GEFS - agreement on descent to norm temps in S on Christmas Day after which anything could happen, with a very wide spread of ens members (range of 20C by New Year).  Op & control cold for a few days after Christmas then v mild at New Year. In Scotland below norm from 23rd and much better agreement on temps staying there to Thu 30th. Pptn in most runs from Fri 24th onwards

ECM - tracks the LP on 26th further S across France and less deep; then develops an extra LP moving quickly  across S UK 980mb Tue 28th which cuts off cold NE -lies for all but N Scotland then back to a col the following day like GFS

GEM brings in LPs from the Atlantic more slowly and then links them to the Scandi LP as a junior partner, in Baltic wed 29th with cold outflow from the NE covering UK

I’m waiting until 12z runs tomorrow.

I‘m waiting until at least Thursday. 

Yep GEM has been by far the most consistent with this cold spell pretty much rock sold for a week now. ECM and especially GFS have been all over the place.

Great post and sums things up nicely. I also do hope you are right that the GFS is just a case of what seems to be typical for it - a see-sawing compounding effects of subtle and marginal changes of data, leading to forward and reverse gear changes in its outputs like Jeremy Clarkson was back testing a car on top gear. The other models seem to be a little more measured in their Op interpretations...but having said that, the GFS has lead the way in signalling pattern changes before, even if the details were off beam, however, we have had more runs improving the scenario for cold than we have seen bringing in the Atlantic. We have all seen these cold spells evaporate at T60 let alone T96 or more, so much more “fun”to be had ‘til this is resolved. 

Interesting article posted recently about airplane data being short too, which must be a contributing factor in the excessive compounding of projections from more limited data.

I have noticed that about the GEM output too. Has the model been upgraded or just been unusually consistent this time?

In the past it rarely showed the same extent of cold as has been shown in recent days, e.g. the current ensemble data for here (t850hPa)

LOL - This time tomorrow I’ll be saying Tuesday 12z and so on

Fixed your post.  

Seriously though, if nothing else it has been a fascinating few days of model watching. How long to the next run…? 

I think a voucher for plumbing services might be more important, Caz.

😮😉

I am happy with the runs so far , it could well be Tuesday until we have a clue BUT I have seen in the past forecasts that are 50 -100 miles out on the day - in this case that makes a huge difference .

Are we still talking about underwear? 

Once we get into high res territory just three hours…

😉

🤔

😳😳

🤣🤣🤣

No….

ECM also gets messy and turns off the cold tap. But I’m sure it will all look different on the GFS again in a couple of hours.

I think that’s about right, except I think even Tuesday is too early. Unless there’s a very clear trend one way or the other, ie taking the battleground zone well north or, more likely at this stage, south, the finer details won’t be clear until much closer than that; maybe the Xmas Eve output.

There are a number of high res models and I’ve watched them disagree and chop and change really very close to T+0. Indeed it’s not unusual to see differences with the radar images in real time; that unexpected Sunday snow recently being a good example.

It’s worth noting that in a marginal set up, which this might be for some locations, things like elevation and precipitation intensity come into play. Rain to snow through evaporative cooling is not unknown and is more likely with slow-moving or stalling troughs, which might be the situation.