Quite a difference at t+144 on the 18z to the 12z, creating at even bigger proportional change for the UK;

12z at 18.00 Friday

18z at 18.00 Friday

18z turns milder halfway through. So another run re-enforcing how this cold spell really won't be what some originally accepted as a given severe spell.

i don't think anyone ever thought it would be "severe". Even on the coldest runs there was never any deep cold on offer. Just a nice change from a record mild December which is still very much on the cards. 

18z stays cold until the latter part. So another run re-enforcing how this spell really won't be what some accepted as a given mild spell.

That remains to be seen, IMO. If ECM jumps ship tomorrow and follows GFS, then I think what GFS shows may be more likely to verify. But I've long been reluctant to trust GFS on it's own, given it's past track record.

120h fax for wednesday.528 line south of UK

That was badly worded - she was saying the week after, not this coming week. In other words in line with the ECM charts.

only if it verifies... there remains more than enough potential in the models for this to turn into a severe spell. That's not a prediction, just stating there are various plausible weather outcomes over the next week or two, and especially with the cold so close to our east, a severe spell is one of them.

quite an array of fronts, all behind 528 dam air, could be snow anywhere?

18z GFS is better for snow potential compared to the 12z .. Plenty of precipitation pushing into sub zero dewpoints, especially across northern parts.

A very hard frost too as high pressure pushes in.

All in all a pretty wintry run with cold air now making inroads by about t+78. :-) 

ECM 12z ensembles for London:

http://www.weathercast.co.uk/services/ensemble-forecast.html

The most obvious point is the confirmation of how much out of kilter the GFS 12z was in comparison.

Not much change from yesterday's 12z with the dominant cluster staying cold through to the middle of w/c 18th, i.e. 12 days away.  The first signs of uncertainty creep in late on Saturday 16th and then progressively greater uncertainty builds over the following 48 hours.

http://www.wetterzentrale.de/pics/Rtavn6617.gif - Doesn't look that cold to me next Tuesday!?

 How sneaky is that extended cold  snow event at the end from the japs

But that's before the cold air has filtered down.

The progression of maximum temperatures goes - 6 to 5 to 4 to 4 to 1 to 0 to 3 to 4 (Tuesday 12th to Tuesday 19th) for the Home Counties.

And that's GFS, which may well be overstating things a little, certainly if you prefer the ECM evolution.

Sorry, but for me this a short lived cold snap with the most likely snow action up north. The output is inconsistent across all models with Meto refusing to buy in to longevity. The chances of an extremely mild January spell have been plotted by Gfs with something below average plotted by ECM.

in such circs, it may be best to listen to the televised forecasts which so far have hinted at chilly weather with a few snow showers.

JMA pulls off some curious tricks to get to that position including having the trough off the U.S. extend SE toward the low near the Azores. Why it does this when the other models are uninterested, I don't know.

As it happens, that chart you've posted is a great example of a relatively weak ridge that manages to have a significant impact on our weather. These features are often overlooked by the models until relatively short notice but we don't seem to be in a position for one to feature next weekend for example; instead a strong high looks to be on the scene, with the level of upstream amplification key to how that high moves about.

I for one am keeping a beady eye on the behaviour of the trough off the U.S. that looks to develop during next weekend. If the MJO manages to remain at high amplitude just a bit longer than ECM is seeing, there's a chance of another round of decent WAA but this time probably a bit east of the attempt that's starting in 3 days time. This is the sort of thing that no amount of ensemble guidance is likely to foresee with any confidence as an underestimation of MJO amplitude is a common trait of the models in general.

I can't make a confident call either of course - just have to wait and see.

Plenty of interest for many areas before that period of course, particularly Thu and/or Fri depending on which model you look at.

Thanks James

t*ts haven't gone up yet

J F F  A bit of wintry ppn around on Thursday