Yes horrible trend today with that low blowing up so far west. Ensembles still look ok with ECM and GFS . But have the Ops latched onto something . Hope not

Me too - the OP has trended so much milder - double finger temps in parts of the south a week today if this comes about! - I wonder if the milder ever gets downgraded? Because cold and easterly's or northerlies tend to but when it comes to milder weather which the models show they always come about don't they? Whenever has a mild forecast in the GFS or ECM failed to deliver?

Op chart for a week today: Looking zonal and mild. 

http://www.wetterzentrale.de/maps/GFSOPEU18_171_1.png

They probably have.

Always seems to be a case of 'whichever model doesn't predict a long cold spell' tends to be right.

I'd, unfortunately, expect models to trend towards ECM/GFS18Z by tommorow.

Looking at the GEFS output (for Edinburgh as that's my location) over the last 24 hours the trend has moved from a fairly solid, flat line cold projection in FI (~-7 average 850s), to a much more volatile set with swinging milder and colder perturbations. The mean 850 line has only slightly increased though. Looks like confidence is reducing in what we can expect as we head into February and tonight's ECM 12z and GFS 18z op runs certainly indicate a much flatter pattern with the trajectory of Atlantic systems more W to E rather than NW to SE.

A trend or just more volatility in the MO I wonder?

Honestly think its about 90% of the time. There was an incident last winter when it happened in reverse (probably during the march cold spell) but almost always the establishment of cold northerlies or easterlies are downgraded. On the plus side, the breakdown of long cold spells tends to be exaggerated by the models too (although to nowhere near as great an extent).

For me the rule that has applied in recent days about FI starting at around 6-7 ahead (if even that much) still applies. It could be that the ECM 12z and GFS 18z are picking up on the beginning of a new trend but there is an awful lot more to be taken into account that merely two individual operational runs.

Based on all that has happened in the model output since late last week, I would not be surprised if it is not another case of model volatility. However we shall see what the runs over the coming few days bring us.

I'd add something else. There is a reason most Model viewing websites have 500hpa heights as their shading with SLP as contours. Historically 500hpa heights were always used over SLP because its much easier to predict them, what goes on in the upper atmosphere isn't as complicated (because the density of fluid is lower) so forecasts for 500hpa tend to be alot better than SLP. And the issue I see in many of the outputs at the moment is a good SLP but decidely unfavourable 500hpa height pattern with the 528 dm line hundreds of miles away from greenland. 

I mean not to be a downer, there is still alot of cold weather predicted just no sign of anything long and severe at the same time. Also cold zonality is annoying for NIMBY reasons.

Its still a really snowy breakdown though.

Would be a very significant snow event although its all going to melt pretty quickly!

Yes basically from 18z onwards it looks mild generally and unsettled with no sign of anything cold as things stand right until end of FI on the 18z Op run. I do wonder how the Met Office and longer range models are going to get out of this one tomorrow when they update? Because it was meant to be cold from middle of Jan and that failed then they said end of Jan and now this is going to fail...!? Where do we go from here? this is turned out to be the worse winter ever for the south west! at least if we dont have snow we would get frost but can't even manage a frost either! Utter cr4p this weather not to mention overcast every day!! Look forward to hearing Richard's rant in the moaning trend! 

Spring. 

Already 10c by 0900am in the south and south west:

They'll just change the forecast. To be honest, this is really quite a common occurance and there have been times when the U turn has been far more monumental. I Remember once in the recent past (probably last 2-3 years) every other model except the UKMO had backed away from a cold easterly. The metoffice based their forecast mostly on the UKMO (and the FAX reflected that). Turned out the UKMO was wrong, and all the other models were right. The BBC even went on for a couple of days after still claiming on countryfile it was going to get much colder!

I think the longer you watch the models the more you learn to curb your expectations. If there isn't unanimous inter-model aggrement on HLB giving snow/cold it almost certainly won't happen. Of course what can sometimes happen is we suddenly get something at 144hr that appears out of nowhere that all the models suddenly latch onto. Its great when that happens!

It would be interesting to do a study on this, but I’m pretty sure it’s perception bias... Take this weekend, it trended milder, then the mild was squeezed into a blip... no gnashing of teeth when that happened!

I'm not really sure its trend towards mild so much as trend away from cold. And not just cold but a particular kind of cold that originates from HLB (particularly north easterlies). Last Friday every model was on board with the easterly and then the 12Z ECM came out that threw a major question mark to this. I was not suprised to see that by Saturday the Easterly had been cancelled because I have seen this happen so many times before on here, how many 'failed easterlies' do we see every year? Seems like half a dozen! And how many convert? Maybe 1-2 if we are lucky.

FAO Tally: With the greatest of respect, you are wrong to say that the MetO's prediction of colder weather from mid-January has failed. Haven't you noticed that has got somewhat colder generally in the past week compared to what it was previously and that some areas have just had their first snow of the winter, even if your area hasn't seen any. Yes, we haven't gone into a deep freeze but it has still turned colder than it was for much of the time in December and the first half of January. To claim that their prediction hasn't materialised is plain wrong.

The Para is once again much better doesn't blow the low up out west and is very cold and snowy up to 192h. Come on Para have this one please

Para, Para,  Para, Para, Para

One other point I quickly want to make about 'squeezing of mild'

There are, seemingly, biases that do work in our favour. For example when a HLB is already well established and you get a negitively tilted low moving into it, it almost always tends south. So if you have a snow warning for S scotland 4 days in advance you would expect it to trend towards a snow warning for the midlands 1 day in advance. So these biases against cold do also seem to exist, but I think (and I know its just annectodal empirical evidence) that when the establishment of HLB is concerned the models are far too keen to do it.

One other thing that may be skewing perception here is the counter-intutive nature of conditional probability. Since HLB NErly events are rare even a relatively low false positive rate could yield far more false positives than true positives.

Para is stunning once again would be a big coup if its gone one over its older brother and the ECM Op.

I don't think the Parallel OP has verification stats that are any better than the regular for the UK. So I'd say the para is probably equally likely to be right as the para. My main concern is the ECM is dead against it.

Interesting that last night’s 18z, the para wasn’t so great, and not a patch on the beast that the Op produced!

ECM ensembles look more Para at the moment.  We could do with a decent ECM Op soon though 

That beast was a really freak run though , Para's stats still better than the old Op for what its worth. Crucial day tomorrow methinks as they say.

For Exeter at least it hasn't been cold it's felt cooler yes but 7.5c max yesterday and max of 8c today? Hardly cold for me anyway! Nighttime min of 1c last night? We had -3.5c back on early hours of 27th Oct.

The Met office said something like by mid month there is an "increasing chance of colder than average weather with frost snow and fog etc - and this lasting through the rest of the period!"

The they said increasingly cold later in the period.

BTW just a reminder this was their forecast back in December:-  The fact is we had no frost in December and it didn't turn colder later (post 22nd Dec!) As you may remember and the run up to Xmas and throughout the NY it was mild nationwide. 

UK Outlook for Saturday 8 Dec 2018 to Saturday 22 Dec 2018:

Any more unsettled weather at the start of this period is likely to be quite short-lived. The weather is then expected to turn calmer and more settled, but this will bring increased likelihood of fog, especially later in the month. Temperatures will probably be around average at first, with a few milder days but also some overnight frosts under clearer skies. Things may then start to turn generally colder later in the period.