OK, we will have to disagree then as I think many of the charts in the last few days have shown something different to what you describe.  

If you excuse the analogy but it seems to me that the GFS op is behaving like a petulant child who is being told they must go somewhere because it is the right way and it is screaming no and having a tantrum and digging its heels as it is gradually dragged in the "right" direct by mummy UKMO and daddy ECM. 

Ah, IMBYism would explain your last few posts, but in any case the GFS has been poor for some time regarding snow for your location, rather than it all being about 'this' run 

If it's any comfort it's well in FI when the milder stuff comes in for southern areas. It's still a chilly 4/5C on Saturday for your location going by the 12z.

http://modeles.meteociel.fr/modeles/gem/runs/2016010912/gem-0-144.png?12

Cant complain

http://modeles.meteociel.fr/modeles/gem/runs/2016010912/gem-0-162.png?12

Always liked Canadians

The key is getting some sort of communication between the mid atlantic ridge and the greenland high. The situation over the artic is less favourable now, but it can still happen:

I think the GEM is very unreliable, but my guess is that many GEFS members will handle the shortwave differently.

Yes, I'd agree with all of that and expect the GEFS members to be all over the place again from that stage onwards.

Wondered what you were all going on about for a second there, it's only out to 72hrs on TWO  Looks excellent 

What's causing GEM to behave differently to the GFS 12z this afternoon. Is it just the handling of the shortwave?

I remember on Thursday thinking that by Saturday we should have a really good idea about the upcoming nature of this cold spell, yet I'm still more confused! and unsure what to make of it all...

But I will try my best and sum up how it has gone so far, so...:

• Earlier on in the week some exceptional charts were being produced and there was a genuine chance they could have come off but the reality is this cold spell will be a lot messier (This is what I was pessimistic about before)


• This is clearly a very volatile pattern which is proving a nightmare to model for the GFS, I think the 18z last night was one of the strangest runs I'd seen


• In northern areas the cold spell is still looking good.


• Southern areas it is very borderline at the moment.


• ECM was a good run this morning and I think in the UKMO run is better then this mornings with some subtle differences in the short term.


• Because of this I reckon the cold air mass will hang on into next weekend.


• Although the GFS seems to have done well to pick up on the SLP rise over Spain.

So in the end an absolute mess but interesting to watch how it will develop.

Unfortunately I don't have the time to look at all the ensembles in detail however.

The important thing from our point of view, is that the HP cells are still linked, without the southern cell calving off and heading into mainland Europe as it does on the GFS.

What would be nice would for the WAA to continue into Greenland, sucking the heights northwards with it. But a shortwave could break through and spoil the party as GFS indicates - more runs needed.

The communication between canada and greenland does look stronger than on analysis charts:

GEM2m temperature forecasts are interesting to say the least. At the 500mB level its verification stats are similar to UKM and GFS, but perhaps it isn't as good lower down as a forecasting tool.

Regarding unreliability issues, I will point out that the correlation indices are averaged over the entire northern hemisphere. That doesn't mean in our corner of the world the ECM is necessarily superior to the GFS or the GEM but more importantly the biases are also dependent on the certain situation. The models tend to over or underdo LPs and HPs depending on where these may live. 

Regarding the GEM output, this is a very plausible scenario; I do 'bin' charts that look contrived and the GEM does produce ridiculous charts sometimes but that has not happened in this case.

Honestly both the GFS and GEM scenario look plausible and I see no obvious reason to discount either.

Indeed! GEM has temperatures well into the minus mid-teens for several nights in a row in central Scotland. Some amazing temperatures 

If only 

Crap GFS but the UKmo looks good out to day six for a sustained cold spell. And at day 5 looks snowy for many including the south.

http://www.wetterzentrale.de/pics/Rukm1201.html

For those interested in Thursday, here is a front overlay; its a really messy picture.

GFS take on the potential for Snow on Thursday

Not saying it's right or wrong ......................it really is J F F

Maybe, but that's educated guesswork. ECM & UKMO both show sleet/snow for the south so I think it would be unwise to rule it out so far ahead.

To be fair though northerly sourced cold spells are very rarely going to bring much for Southern coastal areas.

But is the "met office automatic 5 day thing" (sic) remotely reliable or up to date?

Given how the detail is changing from run to run I'd not make any definitive calls 4-7 days ahead.

Very true, even for me we don't do great in a Northerly