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

There's some proper beauties in that. Somnetimes in these sort of set-ups the high can be too far south and it's the SE that gets the coldest temps, whilst up here it's less cold (850's) as lows pass SW-NE above the block and drag milder air into the mix. The trend in these ENS is for the high to be a bit further north, so no lows - and the Manchester ENS are just as cold (actually perhaps a little colder)

Indeed. The potential for a cold easterly outbreak is still there across the MO, but there is so much that can go wrong as can be seen in recent model runs. The age old problem of not enough Atlantic energy going south, aided and abetted by the Azores High poking it's nose in, could very well scupper things again. Will be interesting to see how things develop, but my gut feeling is that this will be another case of close, but no cigar. Time will tell.

It been a while those runs the cold will start from 20-21 Jan with the peak coldest on 26-27th Jan, no push backs so hope this count down successfully so we can tap -15C as it needed with this global warming as the recent -10 to -11C here still delivered less cold temps 1-2C compare to -1 to -2C in the past thanks to global warming effects.  It would be good to have a prolonged cold to very cold period from 20th Jan to early Feb then we all happy with it to focus Spring after that.

If Only...

I thought it was funny how with the ECM 00Z shows the -10c isotherm just reaches the UK but comes to a halt with cold air to our North east is so near yet so far@ 312z.

That'd be a 6-incher down here, an absolute peach of a run and a textbook midwinter easterly - something we've not seen since the 90s.

If it were possible to bank a run, that'd be the one!

At T+144h we see that bubble of deep warm air over svalbard

https://modeles16.meteociel.fr/modeles/gfs/runs/2026011306/gfsnh-13-144.png?6 

The idea is that it drags the deep high northwards, while we end up with a cold cutoff that is pushed westward towards the UK, that would be the system containing the cold air at lower levels.

There are a few different ways it can go, if you want a more intense cold cutoff then the warm air should be further north; however that also opens the risk that the atlantic is able to have more of an influence. The warm bubble further south means a short fetch easterly is more likely but the uppers will probably be milder. 

I suspect we may get some resolution on this perhaps by tommorow's 12Zs. Then the key step will be at T+120 with much less scope for things to be dramatically different. However, the caveat is that Svalbard is pretty much inside the arctic where uncertainty is extremely high so pattern changes even with the more predictable upper trop can happen at lower lead times.

Didn't the last cold spell start with the Op, then a few other members of the GFS joining in, still with much doubt (although the Op resolutely stuck to its suggestions) until finally there was wider consensus?

A few other Ps have now joined the Op's repeated suggestions and the FI mean 850mb in the ENS at 0z dipped to -5 again.

Like to see the mean in an hour or so when the graphs roll out.

Very  nice set of 06z ensembles as it show near -12C  here and -14c over London, both should deliver deep proper cold.  Notice the mild spike then shot down to cold spike but both have some support and not total outlier both mild and deep cold.  Like other post above similar to Feb 1991 had a very mild spike for 1 day at 8C follow by -5C 3 days later.

No back track on the date line so hope now the count down commence.

https://www.theweatheroutlook.com/charts/gefs/gefsens850london0.png 

Chichester, of all places, at -12.9C on the 06Z operational run for Jan 26th! And control run not far behind

https://www.wetterzentrale.de/en/show_diagrams.php?model=gfs&var=2&geoid=50482&lid=ENS&bw=1 

A nice birthday present - and yes, it is my birthday😊

You probably actually remember 1947!

A lot of people on the other channel getting very excited. But I will hold fire just yet,

So many things that can go wrong 

AIFS 6z shows what can go wrong too much energy coming out of Nova Scotia and the low blasts the block and cold away before it even gets to the UK. 

Can anybody remind me the difference between the Op, the Control and all of their remaining runs showing on those graphs? To what extent are the Control and Op better predictors than any of the other lines?

The op is the highest-resolution run. It has unaltered starting conditions.

The control is run at the same resolution as all the other ensemble members, but like the op it also has unaltered starting conditions.

The other 30 runs have their initialisation data tweaked.

Theoretically, the op run is the best, but in practice the difference is meaningless. UKM are ditching the “op run” concept this year, I think.

Please forgive my ignorance, but why do they tweak the starting conditions for the other runs, surely the start is the only thing known to be true? What would be the point in running a simulation based on data known to be false at the start?

It's the only way to generate ensembles, otherwise every member would be exactly the same. 

The idea is that if small changes at the start mean big differences say 4 days later, you know the forecast confidence is low. If they all look more or less the same then conversely you know you can regard the forecast with more confidence. 

Lots of missing data when the run is initialised, due to patchy coverage, so they don't really know that the starting data is particularly accurate.

Tweaking the start point allows models to gauge whether slight alterations have a big impact further down the line. 

EDIT: What Darren Retron said 🔝

If you think about it the start can never be fully true because it is impossible to know the exact state of the atmosphere. 

Must... resist... (It's hard not to get a little excited, I have to say, as these midwinter easterlies are all but extinct.)

Here's the MC "LCC" view, which gives a better overview than the normal view. 

https://images.meteociel.fr/im/89/7306/gfseu_0_306mob2.png 

-17c 850s in the east.

https://www.wetterzentrale.de/en/topkarten.php?map=20&model=gfs&var=2&run=12&time=360&lid=OP&h=0&mv=0&tr=3#mapref