Well heat being conducted up from the ground underneath lying snow can complicate things, I suppose. Snow that's in contact with a warm surface is I suppose going to start to melt from the bottom up.

2 pages on the thread on the 12z's is it Dec 2010 again?

GFS12z is another level if you love a very cold spell. I mean come on.....not sure I've seen a colder ensemble for years? And approaching relative reliability but in these synoptics, until its 3 days away its risky to call 

Only GEM is against the trend but its been on its own a lot, that'll come in line surely soon, or is it a trend setter?

Nice LP over France, keeping the flow nice and easterly 

I'm pretty sure there will be further twists and turns in the MO though, before then. All you can say this far out, is that it will become generally colder as we head into the last week of January. How cold, and for how long, to be determined. 

You also have sublimation.

KMA too, and I think KMA is a version of the metoffice model they use in South Korea.

True, and the snow gradually compacting over time as the air is squeezed out of it.

The ECM ensemble has trended cooler, though still not as cold as the GEFS.

Is something odd going on with those or is it me?

When I look at the main run it has -6/-7 over London at T162, the ens set doesn't:-

On the ECM op front re Jhall's comment - I thought the op looked really interesting but actually it's much more bland than the charts suggest IMO. The first 2 LPs bring nothing really bar some cold rain although deeper in FI there is snow for the far south and it is generally cold at the surface from 26th onwards to the end of the run. 

I think they are the ECM AI ens not your normal set.

My location 

https://www.wetterzentrale.de/en/show_diagrams.php?model=ecm&lid=ENS&var=2&bw=1&geoid=50625 

Yeah I know - Brian put the AIFS ens up, the chart I linked is the AIFS control I believe but the ens plot isn't anything like the run. 

That'll teach me to assume and not check the chart you posted. Definitely something amiss as you say. 

Oh what I wouldn't give, to be able to worry about either of those things! 😀

The ENS isn't plotting the AIFS operational.

Yes it pulls a rabbit out of the hat and prolongs the cold past the end of the run. 

Looking at the top down (NH) view around T192 on GFS it is quite remarkable synoptics. 1055mb high centred right over the north pole and reverse zonality around basically the whole northern hemisphere. Can this really be right?

That’s a very interesting observation Doc, since you mention it - I am not surprised you think the  0z shows a more Atlantic influence and the 12 Z quite the opposite. That does seem to be the case in recent days. Do you have a theory why that might be? Which is likely to be correct? Would it be an average of both or is the 0z the one with the full data set? Therefore meaning the less favourable runs are likely to be correct. I hope it’s an average or indeed the 12z is the one with the fullest data- but fear (from bad experiences of old) that the former is correct! ?