Model Output Discussion 18th Nov.2013-12z onwards.

[karlos1983]

Personally I'm pretty happy with these charts

 

Baring in mind it is the first week of Winter Proper, things could be a lot worse. I'll take a cold snap thanks. Yes I would prefer sustained cold, but that's not likely yet based on recent models as many have already pointed out.

 

Crikey I dread to think how tense it will be in here if we get a sniff of a Beasterly in the coming winter weeks/months,I fear complete meltdown lol

[Gavin Hannah]

358 users reading this topic!! Crikey, it must be almost winter lol.  

 

You can see on the ensemble where GFS has a cold outlier follow the black line on the 850's from the 7th to 11th the average from all the ensembles is the red line

 

 

 

The vast majority of ensembles are pointing towards a cold shot. The only reservation I have is that at this range, as has been the case so often, we see downgrades far more often than upgrades to projected patterns.

[Timmytour]

Seen a few posts now comparing what the models are showing now for early December with what we ended up with in early December 1962. But such comparisons are not really valid......it would be like saying you can tell what the following winter will be like every time there's a cold late spring like there was this year. And like Spring in 1962 was. <thinly disguised hopeful ramp>  

[bobbydog]

I may be looking at the wrong thing, but the 1962 has HP centered over Poland and today's chart does not?

 

i was talking in general over the whole month. obviously its not an exact match!

[stewfox]

but this is the second consecutive ECM Ensemble suite that has not shown any great support for a prolonged cold period. This morning's was a little better but in my view even that was pretty poor if it was prolonged cold and snow you were after.

 

I get confused about this repeated comment, that the models are poor..... 'if its  prolonged cold and snow you after' .

 

Are we not told in model watching get the cold in place first (JH etc) and then worry about FI and the longevity of the snowman ?

[IDO]

It has to be said that the Dutch ensembles show both GFS and ECM 12z op runs were at the cold end of 2m temps (for that toppler):

The mean is 2-3 c higher, with many members going for milder temps, suggesting there must be many members not even seeing this transient northerly. With the clear split at D5 the mean really is not much help to us and we have to wait for a closer consensus before we can judge this upcoming cold spell.

The spread from D10-15 highlights this even more and clearly ECM is struggling re the medium term pattern. GFS at the moment more coherent (ens wise) for this cold shot in a weeks time; pressure chart (London):

The mist should clear soon.

[phil nw.]

One or two recent posts have been getting a little edgy against others who see things differently.

 

Please respect differing views and if you can support yours with a chart or link to some data then it adds some credit to your post.

 

What we don't need is personal attacks-it wont change the weather.

 

Thanks all.

[on the coast]

A word from the wise. I have followed this forum and TWO from the beginning and after nearly 10 years winter model watching and every chart from all the suites the weather will do what it will do. Sure they give us an idea what may happen but more often than not become alot different nearer the time. Believe me 96 hours is the max to look at from any suite. It will save a lot of heart ache. As regards to the coming cold spell I don't see anything note worthy coming. But hey ho I don't look any further than 4 days. 

[Nick F]

358 users reading this topic!! Crikey, it must be almost winter lol.  

 

 

The vast majority of ensembles are pointing towards a cold shot. The only reservation I have is that at this range, as has been the case so often, we see downgrades far more often than upgrades to projected patterns.

 

Yes, at least the northerly hasn't vanished from this evening's runs - given it still doesn't kick until post t+192 on the 12z runs. There is perhaps cause then to extend 'FI' in the case of northerlies, as models tend to verify them earlier on compared to other synoptic evolutions that bring deep cold - which they can struggle more with, namely blocking that brings easterlies. Though, of course, the northerly plunge may still be toned down or beefed up in the count down, even if the broadscale idea of an arctic plunge towards northwest Europe remains.

 

The 12z GFS and EC ensembles are uninspiring though past the initial cold plunge later next week, mainly due the low heights being retained over Greenland and NE Canada - attributable to the lack of HLB. But I find it interesting this progged northerly, in that the jet stays flat pretty much most of next week, before a large chunk of cold upper air breaks away SE from Greenland towards Nern Europe, suddenly pushing the jet on a NW to SE track rather than W to E. The low heights may remain to our north throughout this arctic plunge, but it goes to show the polar vortex (vortices) are unstable in the higher latitudes.

Edited November 27, 2013 by Nick F

[Mucka]

All good in Principal Snowballs However not entirely true- the long term performance of the 18z & 06z is below that of the 00z & 12z-

Only problem is the page the old regional data was displayed on is now missing...

 

onto tonights models. - Not very good is an easy summary.- thankfully lots will change- the ECM at least is still good & is NOT a toppler at 240 before anyone says it.

 

Cold air as well

http://www.meteociel.fr/modeles/ecmwf/runs/2013112712/ECF0-240.GIF?27-0

-7 c to -9c widespread.... that's some 10 degrees below average for London.

 

S

 

It was an excellent post from snowballz but another thing she may have neglected to look at his how well each suit verifies in blocked and zonal conditions or just with various synoptics. It would be great to have that data but as long as we don't there will still be an element of subjectivity just as I suggest there is when comparing performance between models.

You know how I feel you put a little too much emphasis on that but when it comes to the 06z GFS being overly progressive in blocked patterns then my experience (and subjective opinion) is that it very much is.

Edited November 27, 2013 by Mucka

[damianslaw]

Models continue to suggest a preety chilly theme for early December relative to average- on average it is by far the mildest part of the winter and more often than not can be very mild indeed..when westerlies reach their yearly maxim, it is notable how dry things have been these past couple of weeks and this theme looks like continuing for some time yet. The theme is for heights to remain close enough to the country preventing a surge of milder westerlies/southwesterlies, and upstream signals are pointing to a cold blast from the NW as the trough is pulled down on a NW-SE axis before heights topple in again - but again as we have seen these last 10 days, there doesn't appear to be any signal for a return to strong westerlies thereafter - indeed heights could very easily sit over the country again for some time before another atlantic attack from the NW again and we see a repeat pattern.

 

I wouldn't therefore be surprised to see the models suggesting heights sitting over the country as we approach the middle of December. We are in a very sluggish fluid flow i.e. no major northern blocking, but no real strength in the atlantic thanks to strong ridge in the middle of it.

[Ali1977]

Upgrade on the way....

[TSNWK]

165 on 18:00 Gfs had sopped that low south of greenie, I think that is a good thing and will allow more ridiging north of the high

Edited November 27, 2013 by TSNWK

[Cloud 10]

Colder air spreading down across the country by the middle of next week on the 18z.

 

 

 

Can we get that proper cold air to the North?

[Mucka]

I would much prefer trough to drop South quicker and stop the high trying to ridge back into our SW (bit of race there) but I wont complain too much if we end up here.

 

[Ice Day]

Lining up very nicely at 192.

 

 

Uppers not quite there yet though

 

Edited November 27, 2013 by Ice Day

[SMU]

http://modeles.meteociel.fr/modeles/gfs/runs/2013112718/gfsnh-0-186.png?18

 

GFS & ECM on the same page now with a slight delay in the coldest uppers ( maybe 18 hours)

 

Also both models trying to carve out a GH, but neither getting there- a Scandi high option still on the table.... ( we are talking day 11-12)

 

S

[SMU]

http://modeles.meteociel.fr/modeles/gfs/runs/2013112718/gfs-0-228.png?18

 

NICE.

Proper SNOW cold air-  heavy snow showers merging into longer periods of snow for the usual suspects!

[East_England_Stormchaser91]

Absolute stonker of a Northerly in FI. 

[Cheshire Freeze]

510 DAM down into N England

 

[Matty M]

Looks like the pub run is living up to it's name!

 

[Dancerwithwings]

Blimey!

 

 

You beat me CC...With a chart like that it's worth two

Edited November 27, 2013 by Dancerwithwings

[johnholmes]

Hi,

 

The idea that runs should only be compared relative to the previous output (0z to 0z, 12z to 12z) due to infinitesimally small (magnitudes of 10th/%) idiosyncrasies in data, doesn't stack up. Even if there were data blind spots (often happens) you either run algorithms to blend and normalise it, or you back-fill with prior cleansed 'control' data. The overriding error correction is the updated observational data, which is the precursor to every initialisation.Therein, to discount any run even though it contains perhaps 98% of all operational data, is not advisable. I could sympathise with such a view if such data blind spots brought the scope down to <85-90%, but that simply isn't the case. I often find intra-run variance to be, in the main, anecdotal; for example, verification against the GFS suite (0z, 6z, 12z, 18z) doesn't actually tend to favour any one particular initialisation - they all have, more or less, periods of better performance over each other - which, to be fair, is exactly what you'd expect from a stochastic model.

 

The matter of data density is only a considerable factor when a critical mass is breached insofar as there is a lack of initialisation variables.  For that to happen, there would need to be a serious lack of data (balloons, buoys, aircraft, nautical) and it's very unlikely that would ever happen.  I'm not sure what the operational parameters are to initialise GFS, but ECM is >= 98.25%.  To give context around that, I think I've read in papers that the average over the last three years is about 99.6%.  So, it's a very stable base state model.  I think it dropped to just below 99% when there was disruption to air traffic post eruption of Eyjafjallajokull.

 

So to surmise, yes some runs lack data relative to others, but it's proportionate weight against that of the depth is what matters.  Equally, any such data gaps can be and often are rendered over with algorithms, to help with calculation smoothing.  You tend to get a value spat out at the end of an initialisation which will evaluate confidence in the starting parameter values, and if that's within a tolerance then you can generally have confidence in the run.  I know that's how the developers parse layer code into the stack, and judge its consequential effect.

 

Hope that helps,

 

SB

 

I would agree with most of what you post SB except your first para.

It is well understood in met circles that comparing 00 to 00z and 12 to 12z AT LONG TIME SCALES is preferrable to comparing run to run. I assume this is because each time the same additional data over and above the basic data that every run gets is much more similar. At shorter time scales then your assertion is correct. What is a shorter time scale is hard to define of course. I generally use the former method down to 168h sometimes less then do compare run to run below that, and of course model to model at all times scales. What is being looked for at longer time scales is for similar synoptic patterns or more realistically upper air matches in subsequent runs from the same start time. Closer to 00h then all the models will tend to converge and synoptic detail will become clearer as the time scale decreases.

Edited November 27, 2013 by johnholmes

[Ice Day]

Minus 8 uppers covering the majority of the country at 216

 

[Matty M]

Will probably work out something like that but not quite as intense.

 

2-4 day bitter north/northwesterly.

 

 

Doesnt appear to be anything to sustain the outbreak as the PV re-organises but changes are possible.