ukpaul

JMA looks fun, completely different but is it picking up something important or not?

Are you sure you have that the right way round? The ECM is the model that has picked a pattern and stuck with it for the last couple of days, the GFS has given differing results for the op and a much more variable output from its ensembles as opposed to the ECM. I'm looking fr consistency as things move into a more reliable timeframe and, so far, it's provided that (now, having said that, it's bound to flip totally for the next run!)

By 84h though the GFS appears to have a difference from both ECM and UKMO over Scandinavia and that's one of the areas where I think they are sowing the seeds of their differing evolution further down the line.

Jet Stream's gone walkies by 300h but, to be frank, I think it blew it earlier in the run by pushing it through ever harder.

Looking at the GFS/ECM comparison now at around 138 I tend to believe the ECM as it doesn't push the lows through with as much power and as far East, I've been burned too many times with GFS whipping up storms that get downgraded to believe that it isn't been similarly too hasty here also. By 174 the difference is ridiculous as GFS has piled one on top of the other.

This is misleading, just look where the PV goes, a 'proper' cold spell, rather than anything that tried to get there on a wing and a prayer too quickly has been at around 23rd January for a while and that's continued to be the point of interest as it moves closer. Arctic high not as widespread later on which changes the later part of the ECM run. EDIT: although not the eventual outcome which gives a variation on the 00z 240h.

That's exactly how things were at 168 on the 00z. Look at the Northern Hemisphere view for an overall perspective.

A quick model related question, are the different models reacting to the same data (and the lack of it, where that exists)? If so are the chops and changes at the moment because they are reacting to inconsistent signals which they are all getting at the same time? We see many times where models diverge and I've always presumed that this is because of how they use the data differently, if they are using different data however that would make cross model consistency in quick turnarounds more credible. Just wondering what the ups and downs are a result of.

Just by chance I'd left the wetterzentrale graph of the GFS 00z London ensembles on my computer screen this morning, I updated it to the 12z ones and there was barely any change as I could see. Getting colder with some nice cold uppers from 192h on on a number of runs, some runs that put us into something very much at the opposite mild end such as the GFS 12z op. Just thought I'd mention that there is some consistency, just that the consistency is to have a wide spread from a week out.

Quite a few ensemble members looking to eject the low south eastwards at 90h, quite a few succeed, something to look for in the reliable timeframe.

There is one simple reason and that is because there is no definite outcome beyond anything but a relatively short timeframe at the moment. You are going to get people commenting on changes after 120h which vary wildly and, whilst all are 'correct' in the sense that they are plausible, they will be mutually exclusive. Add to that how you get the occasional misreading or attempt to guess something that we will find out ten minutes later anyway, and it's best to focus on the next five days and what signals emerge during that time that are likely to have an effect later. Anything else is akin to reading tea-leaves. Anyway, I'm quite happy with this evening's output (if we're going to take the mood temperature) but it'd be nice to get something cold to kill off these bugs that I'm picking up.

ECM looking much better in the reliable timeframe (about 96h at the moment it appears). After that then it could go anywhere (see GEFS ensembles!). Equally likely to bring in an easterly or a Bartlett high and each as likely to verify! EDIT: Can't get post 120h to load, bugger. EDIT: EDIT: Got it, hah, well that's yet one more wildly different solution to add to the pot.

The Ensembles are, in that case, being surprisingly bullish about an easterly, only a couple with the same oomph as yesterday's 18z but seven or eight members, and that's just the ones that hit us for a few days, a number also go further south into France instead.

GEFS Ensembles, in the absence of anything definite, seem to be trying every option on for size, just to be on the safe side. It's quite fascinating just how quickly they develop in completely different ways.

Ends up by trying to give us colder westerlies, not sure I have much faith in outputs beyond this weekend at the moment, everything after that is going in different directions.

Shouldn't stop people looking for features that are there or not though. That said, there have been a few humorous misinterpretations that look pretty silly ten minutes later! EDIT: The low that had disappeared on the last runs turning up again ejecting out of the deep south by 138h

Just seems to get to where the o6z was but much earlier, not much confidence beyond 96h at the moment,

Have I missed a new thread or is nobody talking about the 12z? Given that FI is so early in the run we're pretty much at the point of divergence now already. Anyway, there seems to be no low ejecting south east on this run with it coming up against a stubborn high over the UK.

Looks like considerably more than that looking at them coming out, I'd say four or five other runs at least.

About four of the ensemble runs on the 12z showed pretty much the same, so I don't think it will be without support this time.

Been watching that little baby all the way from the deep south. Exactly what the best of the ensemble runs developed.

There is no such thing as 'form', weather isn't a horse you know! Signals change, patterns change, we are at a possible change and any summing up needs to show that.

Credible up to that point but there is no agreement beyond that and definitely nothing to suggest probably milder next week, to be accurate you need to say that there is a chance both of a reversion to milder weather, an increase in the severity of the cold or something inbetween.

Not sure that the UKMO is anything more than non-commital, it doesn't go out far enough. The ECM ensembles are the next pointer.

Fair degree of similarity so far between ECM and JMA. EDIT: Can't believe someone posted the same just before me! Anyway, it appears to be ECM and JMA with GFS and GEM on different tracks.