Model output discussion - 1st December onwards

[Frosty.]

I bring good news for coldies, the Gfs 00z shows colder weather on the way, what a pleasure it is to post charts which are not very mild muck for a change!

Edited December 17, 2015 by Frosty.

[carinthian]

9 hours ago, carinthian said:

Ok, lets look at the far reaches of the latest ECM model at 240 hours. How can we get a Scandi high from this postion. Presently I would calculate 30% post 240 hours. Ideally, I would like to see that Low circulation in mid-Atlantic a bit further west and south with a more pronounced trough to align between The Azores and Iberia. This would help to ridge high pressure northwards and allow colder Arctic feed path on its Eastern flank to sink into Europe. That's the position we have to aspire to turn cold and drier. Obviously more runs will determine the outcome whether a NE block develops but signs are encouraging albeit slow at the moment but could form rapidly with the right circumstances in place.

 C

Morning, a bit nearer to the picture we aspire too  regarding to what I posted last evening. GFS still looking at a rise of pressure to the NE. Still more runs needed but significantly has not dropped the process.

C

 

[Summer Sun]

This mornings GFS op is a clear outlier for something colder on the ens between the 27th / 28th to 31st

 

 

Edited December 17, 2015 by Summer Sun

[Nick F]

Well, the 00z ECMWF det. continues to disappoint wrt to building the Euro ridge north, though the EPS mean at t+240 keen on it, the EPS control in the extended range even punches this ridge to the pole via Svalbard around day 11 - which would surely disturb the PV at bit.

In the more immediate future, rainfall piling up across western areas this weekend is of some concern again. A slow moving frontal boundary with ripples/wave developing along it combined with a very mild/humid southwesterly flow could bring some large totals, though unlikely anything on the scale that weekend earlier this month. Rainfall accumulations from GFS maybe on the conservative side, as they often are, Met Office warnings suggest twice this much over exposed areas

[iamgazza]

41 minutes ago, Summer Sun said:

This mornings GFS op is a clear outlier for something colder on the ens between the 27th / 28th to 31st

 

 

Outlier or Trend setter... lol

[Weather Boy]

3 hours ago, IDO said:

Ignoring the crowd pleasing pub run, which is always advisable....

 

Why?

[knocker]

This morning's ecm anomalies keeps the vortex around NW Greenland and the upstream pattern very much the same apart from losing the upper trough in the SW US. At T240 it has the Atlantic trough running a fair way south from Iceland and the Euro HP ridging NE from the western Mediterranean to the Baltic states. This leaves the UK in a SW flow with slightly above average temps.

By T360 the trough and HP have retrogressed slightly with the southward extension of the trough much reduced. The HP has also realigned and now ridges NE from France to southern Sweden and it's reach north also reduced. Not any huge change to the streamlines and temps still a little above average.

Taking a bit of a punt I suspect the percentage play  from here is the European HP playing a familiar role with systems running in from the west probably effecting the north while the south tends towards more settled weather with temps becoming nearer average.

Edited December 17, 2015 by knocker

[Liam J]

2 hours ago, Weather Boy said:

Why?

It's not called the pub run for nothing! The 18z GFS throws up some wild FI charts at times which don't always fit with the other daily runs.... Also there are one or two members I've seen who don't give the 18z & 6z much credence, not sure why something to do with data iirc, maybe they are poorer for verification?

 

Edited December 17, 2015 by *Sub*Zero*

[Frosty.]

The Gfs 6z op run shows considerable moderation in the persistent very mild mushy pattern from next week onwards with some colder incursions of Pm air making inroads at times, especially across the north / northwest with occasional sleet and snow across northern hills and night frosts, there is some mild weather, I would be very misleading if I said otherwise but at least it's not relentless mild blowtorch conditions during christmas week and on towards and into the new year. It is a very unsettled run through high res so more concerns for renewed flooding across western / northwestern areas but looking into the new year, the 6z shows pressure rising strongly to the east and the T+384 chart left me feeling positive about a potential cold spell evolving during January 2016.

Edited December 17, 2015 by Frosty.

[Weather Boy]

24 minutes ago, *Sub*Zero* said:

It's not called the pub run for nothing! The 18z GFS throws up some wild FI charts at times which don't always fit with the other daily runs.... Also there are one or two members I've seen who don't give the 18z & 6z much credence, not sure why something to do with data iirc, maybe they are poorer for verification?

 

Thanks for responding (although I'd be interesting to see IDO's response).

As I understand it, what you say is anecdotal and therefore not very scientific.  It has been the subject of discussion before and I recall someone (was it Ian F - it might not have been someone quite so authoritative) saying that the 'pub run' label is utter nonsense and has no basis at all.

I have always assumed that the pub run is so named not only because of alleged inaccuracy but also because it comes out round about chucking out time!

[nick sussex]

A mixed bag of outputs today. The GFS 00hrs run shows what could happen if that deep low to the west amplifies, the GFS 06hrs run not as good because the pattern is flatter.

The ECM disappointing and it would take a long road to develop anything colder from its T240hrs output.

Much depends on whether that deep low turns up, how far east it is and if it amplifies.

You can see the difference the amplitude of that low makes if you compare the GFS 00hrs run below:

The GFS 06hrs run:

 You can see the difference between those and this comes down to the amplitude, in these situations you want to see at least the more oval shaped and preferably more elongated as this will drive better WAA ahead and develop a stronger ridge further on.

We'll have to see whether the ECM returns to this low as its dropped the idea today. For those looking for colder conditions the quickest route is the GFS 00hrs and so hopes rest on that more amplified low.

[IDO]

4 minutes ago, Weather Boy said:

Thanks for responding (although I'd be interesting to see IDO's response).

As I understand it, what you say is anecdotal and therefore not very scientific.  It has been the subject of discussion before and I recall someone (was it Ian F - it might not have been someone quite so authoritative) saying that the 'pub run' label is utter nonsense and has no basis at all.

I have always assumed that the pub run is so named not only because of alleged inaccuracy but also because it comes out round about chucking out time!

The Pub run is run at lower resolution and with less data (less flights, weather balloons only go up twice a day etc). It is therefore increasingly unreliable with time so should be looked on after D6 with some caution. Of course when the models are flummoxed by possible pattern changes then they become even less reliable, especially the op run, that has a bias to blow up heights in such situations.

There is a review on this by NCEP but only really analysis up to D5 and I have little qualms about that as any bias is compensated by having more up to date data (run 6 hour s later). The conclusions even at D5 is that the 0z and 12z are better than the 06z and 18z. The review is below with a forum discussion also listed:

http://www.emc.ncep.noaa.gov/gmb/wx24fy/doc/GFS4cycle_GCWMB_briefing_13dec2012_fyang.pdf

http://www.americanwx.com/bb/index.php/topic/43144-why-do-people-dismiss-6z-and-18z-model-runs/

[Nouska]

2 minutes ago, IDO said:

The Pub run is run at lower resolution and with less data (less flights, weather balloons only go up twice a day etc). It is therefore increasingly unreliable with time so should be looked on after D6 with some caution. Of course when the models are flummoxed by possible pattern changes then they become even less reliable, especially the op run, that has a bias to blow up heights in such situations.

When did this take place? I know the radiosonde data is only in the 00Z and 12Z but no knowledge of a different resolution between GFS runs.

[Singularity]

ECM's shown us the risk that we wind up at the mercy of the Atlantic trough while the Sceuro high gets its act together. I remember in Nov 2009 how we found ourselves on the eastern flank of storm after storm while the blocking highs developed to the E and then NE. A similar path, with similar outcomes as that blocking drives wave breaks on the vortex, seems an entirely reasonable expectation to me at the moment.

Nov 2009 was the wettest month I've recorded in the past two decades, so it's obviously not at all desirable unless you enjoy wet and wild weather to the extreme (which presumably excludes large parts of the NW for the foreseeable!). The GFS option has the wet and wild conditions out in the Atlantic while we stay largely high and dry - let's hope we can have the pattern set up closer to that than the ECM outcome. Even if we didn't get a pre-SSW (hopefully SSW) easterly, that would have the potential to bring a cooler, more seasonal feel to the weather.

 

I find the GloSea5 update encouraging for a change at some point Jan-Feb, though clearly a few too many ensemble members are holding on to the Euro high further through January than I'd ideally want to see. I know Feb can bring sustained cold, snowy conditions as extreme as Jan with the right setup, but I do tend to get fed up with living in fear of sunny spells during snowy spells at that time of year  Anyway, the strength of the -ve NAO signal for the Feb-Apr period is stunning and implies a considerable hangover from the winter into spring. Perhaps Dec 2015 and March 2016 have traded places? 

[nick sussex]

1 minute ago, Nouska said:

When did this take place? I know the radiosonde data is only in the 00Z and 12Z but no knowledge of a different resolution between GFS runs.

Nor am I, it would be pointless to run the GFS 18hrs operational at lower resolution and if that was the case the accuracy would fall through the floor. And this is not supported by the verification stats.

[Nick L]

All GFS runs are at the same resolution as far as I'm aware. Not sure where this nugget of info has come from.

[phil nw.]

4 minutes ago, Singularity said:

ECM's shown us the risk that we wind up at the mercy of the Atlantic trough while the Sceuro high gets its act together. I remember in Nov 2009 how we found ourselves on the eastern flank of storm after storm while the blocking highs developed to the E and then NE. A similar path, with similar outcomes as that blocking drives wave breaks on the vortex, seems an entirely reasonable expectation to me at the moment.

Nov 2009 was the wettest month I've recorded in the past two decades, so it's obviously not at all desirable unless you enjoy wet and wild weather to the extreme (which presumably excludes large parts of the NW for the foreseeable!). The GFS option has the wet and wild conditions out in the Atlantic while we stay largely high and dry - let's hope we can have the pattern set up closer to that than the ECM outcome. Even if we didn't get a pre-SSW (hopefully SSW) easterly, that would have the potential to bring a cooler, more seasonal feel to the weather.

 

I find the GloSea5 update encouraging for a change at some point Jan-Feb, though clearly a few too many ensemble members are holding on to the Euro high further through January than I'd ideally want to see. I know Feb can bring sustained cold, snowy conditions as extreme as Jan with the right setup, but I do tend to get fed up with living in fear of sunny spells during snowy spells at that time of year  Anyway, the strength of the -ve NAO signal for the Feb-Apr period is stunning and implies a considerable hangover from the winter into spring. Perhaps Dec 2015 and March 2016 have traded places? 

Yep pretty much my thoughts too Singularity.

There's little doubt now that some height building is likely to our east post Christmas unless all the model suites suddenly switch.It's then a case of how much amplification can we get.

Already the GFS is toying with different amounts of ridging so the uncertainy will likely continue until we start to get closer and into the higher resolution part of the runs.Like you i would settle for a UK high while we wait hopefully for another attack on the pv from further wavebreaking.

What we don't need is a poorly orientated Sceuro high prolonging this south westerly pattern,eating up part of January.

 

[IDO]

Sorry my mistake re resolution!

Latest verification stats for the four GFS run @ D10:   @D8: 

Edited December 17, 2015 by IDO

[Weather Boy]

1 minute ago, IDO said:

Sorry my mistake re resolution!

Latest verification stats for the four GFS runs: 

Thanks for this.  Looks like all the runs are pretty similar at the moment, but there is slightly greater frequency of the 'pub run' occasionally going badly wrong, which they all do sometimes.  Mind you, that is only the recent stats.

[bobbydog]

7 minutes ago, Nick L said:

All GFS runs are at the same resolution as far as I'm aware. Not sure where this nugget of info has come from.

chinese whispers nick. fergie did clarify it a while back. as far as i remember, they are run at the same resolution and there is no "missing" data as such, the 06z and 18z are only "missing" the balloon data so they are run with the previous balloon data alongside the new info from other sources. they may be slightly less accurate but it is negligible. 

[Nick L]

Just now, bobbydog said:

chinese whispers nick. fergie did clarify it a while back. as far as i remember, they are run at the same resolution and there is no "missing" data as such, the 06z and 18z are only "missing" the balloon data so they are run with the previous balloon data alongside the new info from other sources. they may be slightly less accurate but it is negligible. 

Yep, only a matter of time before the annual "It's Christmas so the models can't be trusted" line is wheeled out!

[Methuselah]

1 minute ago, bobbydog said:

chinese whispers nick. fergie did clarify it a while back. as far as i remember, they are run at the same resolution and there is no "missing" data as such, the 06z and 18z are only "missing" the balloon data so they are run with the previous balloon data alongside the new info from other sources. they may be slightly less accurate but it is negligible. 

Aye, and it's certainly not the cause of all the warm weather we are currently experiencing.

[bobbydog]

3 minutes ago, Nick L said:

Yep, only a matter of time before the annual "It's Christmas so the models can't be trusted" line is wheeled out!

yes nick, yet another myth. actually santa collects data from all over the world in a single night on xmas eve so the model runs are at their most accurate. thats true that is..

[Liam J]

Some positive signs on the ECM monthly for a pattern change into January as we lose the mild air, looks quite stormy for a time with frequent blasts of PM air and the run finishes with a flow from the Arctic and an attempt at a mid Atlantic ridge towards Greenland, Euro high completely obliterated at T762 hours anyway lol... 

[Frosty.]

25 minutes ago, Nick L said:

Yep, only a matter of time before the annual "It's Christmas so the models can't be trusted" line is wheeled out!

Any positive spin for coldies is welcomed since those of us in the southern half of the UK are buried in very mild muck and Saharan sand..Lol. :-)