For my location it shows

Light snow 2c Sunday in a SEly

Light snow 1c Monday SEly

same temps for me too, then up to 4c on Tuesday.  BBC shows similar rising to 8c by Sat with little or no precipitation... Hence my thoughts that it's a slow sinker

Same here... SW winds by Thursday with a high of 6C, then 8C by Friday.

Indeed Karl.  There are SIX clusters of ECM 120hrs postage stamps which only serves to underline the HUGE levels of uncertainty from that model:

From IF on NW

"Increasing signs the cold will be shunted away early next week, so we will *eventually* lose these 'awkward' forecasts!"

So expect to see changes sooooooooooooooooooooooon

Sorry for O/T

But then we have this

http://www.bbc.co.uk/weather/features/35319682

Very latest from Alex Deakin on BBC says mild air "probably" replacing cold by end of next week.  Lots of model talk in this forecast.  Impressive.

www.bbc.co.uk/weather/features/35319682

Coming back to the ECM ensemble data - the only clear thing is the lack of clarity. If people think that the op run is representative of the majority of the ensemble members in the longer term I am afraid they are unable to read that output accurately and their other analysis would have to be seen as very suspect.

Anyway the pub run will trickling out soon. Will it hold my surprises? Will it be another random evolution or more typically mobile conditions followed by some random blocking in FI in traditional GFS style. Stay tuned.

Indeed it was and showed why all the uncertainty, couldn't even show a chart for Monday

That is an absolutely fantastic forecast ! Great explanation. 

And there's the source of some of this confusion, just left of centre: http://www.wetterzentrale.de/pics/Rtavn361.gif

By T48 Alex is in exactly the same place as on the 12Z.

That does explain much. I did wonder if this hurricane was the fly in the ointment.

The difference is in the ensembles. GEFS has been bouncing uncontrollably, whether you look every 6 hours or every 24 hours. The ECM ensemble, particularly the median, has been consistently showing the widespread 3 or 4C which is currently prevailing at the surface.

The problem is that we don't get to see the detailed ECM ensemble until hours after the GEFS, so it tends to feel like "yesterday's news", but that's the price you have to pay for taking the time to get it right.

Indeed Steve. No willy waving there just the facts about the current situation. Anyone able to get AD a login for here? 

Alex gives Alex a headache 

Yes just watched that. It's great when media posts aren't actually off-topic as he gave his detailed thoughts on the reasons behind the model uncertainty.

Long may these types of forecasts continue 

I knew Hurricane Alex was unusual, just not 4 times in 150 years unusual!  No wonder the models are struggling 

Interesting.  The lastest update to the MetO not-a-forecast for Evesham has max temps for next Weds at 4c instead of 5c

But also now has snow for Sunday morning .....


Great things are afoot Earthman!

18z out to T+96 now.

At T+90 it doesn't look much different to the UKMO 12z, just the remnants of a filling LP to the NW of Scotland.

Ho hum, when will the default kick in....

Very different to temps for MBY on the Beeb site

Quite different at T114. Where the 12Z had a deep "round" low south of Iceland, the 18Z has a squashed and disrupted trough. 

Not that it really helps us though!

Indeed - it's entertaining watching GFS back-track very slowly.  It's still pushed the LP much further eastwards than UKMO but by T+126 it's almost reduced to a trough extending from the Atlantic. Also heights are better to our north than on the 12z.