Similarly we haven't had a prolonged hot spell in August for a long time. Maybe not since 2003, although we have had some hot August days.

And back on topic the GFSP has almost the whole country snow covered by 186hrs. Nobody has a clue, least of all me.

A big if and If the Para 12z run is correct then a week tomorrow we could be looking at a stiff northerly and frequent sleet/snow showers with temperatures barely getting above freezing by day:

Gotta love the Para after the northerly an easterly.

First ice day of winter 2018/19 if this Para 12z run comes about:

Let's forget about the 0p run and focus on para? The OP run has been useless this winter!

I've accelerated the TWO GEFS process so charts are appearing about 30 mins earlier than usual this evening. 12z update looks quite mixed.

Moreso in the south, up here in the short term there has been and continues to be a cooling trend, in the longer term not much change.

Given the changes between runs I wouldn’t read too much into the longer outlook. One look at the GFSP will give you an idea of what is possible (whether iti is probable or not I suspect a different matter). I would expect the volatility to continue for a while yet.

I wouldn't ignore any output Tally, but given that the GFS Op is often 4th best on the verification (actually 5th behind GEM sometimes) I would factor that in on where I place my faith in the output.

Snow rows of 98 though not to be sniffed at. It's marginal down here but snow is possible from about the 19th onwards .

Well 12z Op run is far more blocked as you can see compared to the 06z run which shows little if any blocking and just zonality. So which is the best model output would you say? ECM?

I've noticed that too. Easterlies have always tended to occur more often later in the winter, and may now be favoured over northerlies for cold weather - at least in the south - because the Arctic has warmed by so much in recent decades. For a really wintry January in my part of the world, you probably have to go back to 2010.

Average ensemble mean from N. Hemisphere view down at the end of each run! - After this mornings disappointment it's game on!

It's looking more blocked from 00z run with HP becoming more dominant over the Arctic/N.Pole regions as you can see we lose those dark blue colours and the jet taking a more southerly track each run:

00z run

06z

12z

The caveat is still that annoying Azores HP- which is so stubborn to move anywhere!!

In terms of verification it does sometimes vary, but it's usually ECM followed by UKMO (albeit it only runs out to T+144), followed by GFS Para, GFS Op and GEM.

While the 850s may not look mega cold on the GFSP, the low pressure means that the air is colder than you might think, with 510-516dam thicknesses over most of the country, from a northerly.

We seem to have hit rock bottom in terms on Brighton Snow Rows....    58 is the magic number.

Recently at least, at 144hr the GFS op is in 5th behind the GEM(CMC). The GFS(FV3) Which the current parallel GFSP model is a good way clear of the current GFS operational model and is also ahead of the GEM but still behind the UKMO and ECMWF

https://www.emc.ncep.noaa.gov/gmb/STATS/html/new_acz6.html 

Edit: On the subject of snow rows it’s nice to see the one for three days out for here sitting at 20. (Overall total is 175 FWIW)

True, but Tally was referring to the GFSP there, which one would like to think would be somewhat better than the current operational GFS when it comes to verification.

I was looking at the 12z stats Doc, but yes the pattern is similar with GEM outperforming GFS Op on some of the timescales. 

Tally was asking about ignoring the GFS Op in favour of the para, hence my comment relating to ignoring the GFS Op.

Looks like the ECM has a snowy slider at 144h nice upgrade 

I was about to question it, but then I looked at the 850hpa temps and they are satisfyingly low.

Mean gone from -7C briefly to -3C now. I still think Saturday will be about the time we can be confident or not about the rest of January being cold. 

Even the ECM is showing a bombing low diving SEwards over the UK at around the 23rd - just like the GFS's 12z as mentioned by me earlier on this afternoon. Should that verify, I think that will be a bigger issue than short lived transitory wintry events.

Hmmmm. Interesting.