O-T, I know, but I remember that well - I was 6 and I still remember waking up on New Years Day to heavy snow.

GEFS06z showing a big scatter after the next few days.

There continues to be a lot of interesting runs and interesting charts around. Who knows where it will end up. Zonal also still appears to be the lesser favoured option.

Indeed, and the ECM 00z ensemble suite does the same.  Gone is the strong cluster that maintained the cold through to day 15

Still a decent cluster around a cold outlook but the more dominant outlook is a return to normal temperatures

and wetter...

and windier...

Even more scatter in Aberdeen! Some may like to note that both the op and the control keep it cold up north, at least... otherwise, pick a strand of spaghetti, -10 or +10?

Reads persuasively though. So far the analysis looks spot on, but naturally, it remains to be seen how things will pan out in the coming weeks. Manics can be right sometimes, just as a stuck clock is absolutely correct at least twice a day :-)

It's nice to see a decent amount of energy going into the southern arm of the jet for a change. Through most of the recent output heights have been too high to the south and it's been going over the top and down into Scandinavia.  I wonder if other output will start to pick this up or whether it's just an aberration.

Well said that man. Two weeks ago I said the next 10 days were fairly clear cut. Somebody, can't remember who, said it wasn't and I should try looking at the charts. Of course I had and I was struck by the GFS ensembles having all of the mean, operational and control on top of each other and in one tight cluster for 10 days. Anyway thought I would plot each subsequent day and see what happened. In fact it was spot on for 8 days then deviated badly as the uncertainty kicked in. Currently it is different with only 4 or 5 days of strong agreement after which it is less clear but at least there is some consensus for it to return to below average. The overall patterns are certainly making me more optimistic for a colder start to winter than recent years. Ensembles are your friend ;-)

Well, if he's correct about the positioning on the upper ridges then the major building blocks are in place. Anomalously high pressure in the Greenland area should put an upper trough somewhere in our vicinity. Whether that's over us or just to our east or just to our west is the next critical part of the jigsaw.

As he says, he is right out on a limb given that all the major seasonal forecasts predict Europe to be on the mild side.

Would you have made that same comment this time ten years ago, Beast?

Quite possibly yes, but pre-supposing the weather is a risky thing to do.

Just saw this posted on Netweather - on the left is the actual T0 chart today. On the right, the ECM 240 chart from 10 days ago. Models = nonsense!

I think the key point is that you saw 8 days of consistency and now it's 4/5 days before the pattern becomes unclear.  I think it does depend on the pattern: sometimes the ensembles do give the correct picture a week out and sometimes they don't. The problem is we don't know which.

If the operational run picks up a new pattern then quite often so will the ensemble suite; it just depends on what has changed. But we've all seen the ensembles flip; it has happened over the last 24 hours, if not dramatically.

Yes, the ensembles are useful, together with cross-model support. But the further out you look the more cautious you have to be, particularly in unusual synoptic patterns. 

Winter of 78/79. Yup I remember that one very well.

Sorry to go OT - the next one was in December 1981.

A quick look at the GEFS 06h groupings.At 240 h 70:30 majority in favour of  cold and continuing Greenland block.

By 360h split cold;warm is 50:50 although no clear return to mobile Atlantic.

Interesting comparison. I think we have all learnt from experience that anything beyond t+144 is suspect. We still like to look at eye candy in FI, but what are the chances of t+240 being accurate? What ECM showed for today 240 hours ago is more usual than what we actually ended up with - if the models had an eye for statistical accuracy over all else, they would all back the form horse (prevalent westerlies) in FI.

Further variations on the 12Z GFS, The Greenland high is there but what will that low to the west do?

Gfs looking very good at 200h.

Ukmo looks similar to it at 144

Cracking 12z GFS today