General Model Output Discussion

[izi]

watching the bbc forecast, they self admint the low could be 150-200 miles north or south, so mebs the GFS could be right!?

Edited December 27, 2009 by izi

[johnholmes]

yes Nick-the differences in the major items within a 72 hour time frame are quite large

precip - not the same area

850mb temp not the same

pressure not the same

something in one of the models has to give it would seem

UK Fax charts should be interesting as they update

[john mac]

Well i have to say that i am surprised,i thought the 18z would come morwe into line with the euro's, perhaps it will take a couple more runs or may be it wont happen at all

[Supercell]

Write the GFS of at your peril?

Well... this will either be the making of or the breaking of the GFS. In terms of credibility at least. I have always tended to follow the GFS, maybe not neccessarily always for initial building blocks and patterns but moreso for the detail in sticking to its guns and setting a trend for the others to eventually latch on to.

Edited December 27, 2009 by Supercell

[TEITS]

Whats baffling is this isn't the GFS vs ECM or UKMO. This is the GFS disagreeing with every single model output at +72hrs. Very bizzare indeed and not something I have seen before.

[snowmadchrisuk]

Oh Dear the 18z has Tuesday, Wednesday and Thursday warmer in the South than it was today. 6 Degrees at 6am Wed/Thur in the Midlands. I don't know if were on for a Mild week or a Cold one.

[cyclonic happiness]

Here's a scarey though....maybe the gfs is right??!!

After all everyone thought it was a fortnight agao when it was showing the currant cold spell and the other models were having a wobble.

Edited December 27, 2009 by cyclonic happiness

[Geordiesnow]

No way i dont think the low is going to be that deep. The GFS is over doing the low definately

What makes you think that?

GFS often over powers the lows it has to be said but we can't say without much confidence that the GFS is wrong.

Mind you even at 24 hours, the differences between this GFS run and the UKMO run is shocking in terms of the strength of the low.

Dismiss the GFS output at your peril!

[kold weather]

Its an amazing difference John, though I remember feb 2007 was a similar event and the models were generally too far north with that set-up...infact I remember the GFS had snowfall reaching south Scotland, in truth the band struggled to get past the Midlands.

I suspect the GFS will be doing the same this time round.

Low pressure still looks further south then the poor set from the GFS ensembles at 12z so the FI should turn out ok...but we shall see. I'm looking for winds with an easterly kink, be it either from the SE/E/NE for early Jan.

[yamkin]

watching the bbc forecast, they self admint the low could be 150-200 miles north or south, so mebs the GFS could be right!?

GFS 18Z is being too progressive with the low. BBC24 weather forecast by Rob @ 2258hrs mentioned lots of snow on Tuesday night in Wales.

[Guest]

Here's a scarey though....maybe the gfs is right??!!

Let's hope not. Otherwise I'll be needing a JCB to dig myself out and there will be a lot of unhappy Welsh, English and possibly N.Ireland people on here. It's only one run which is at complete odds with every other model output.

One run doth not a winter break.

[kold weather]

Well... this will either be the making of or the breaking of the GFS. In terms of credibility at least. I have always tended to follow the GFS, maybe not neccessarily always for initial building blocks and patterns but moreso for the detail in sticking to its guns and setting a trend for the others to eventually latch on to.

to be fair to the GFS though even the ECM ensembles show a rather interesting scatter for the midweek period, there are milder options even on the ECM as well, this is going to be a nowcast situation I suspect.

http://www.knmi.nl/exp/pluim/vijftiendaagse/index.html

[phil nw.]

yes Nick-the differences in the major items within a 72 hour time frame are quite large

precip - not the same area

850mb temp not the same

pressure not the same

something in one of the models has to give it would seem

UK Fax charts should be interesting as they update

Latest fax John T60hrs.

http://www.meteociel.fr/modeles/ukmo2.php?ech=60&carte=2000

As you say a big difference.

[on the coast]

Well... this will either be the making of or the breaking of the GFS. In terms of credibility at least. I have always tended to follow the GFS, maybe not neccessarily always for initial building blocks and patterns but moreso for the detail in sticking to its guns and setting a trend for the others to eventually latch on to.

GFS called the last cold spell well if i remember when others wobbled and this is why i would never

rubbish the GFS at the moment. It might be way of the mark but the question has to be there? All eyes

on the 0z runs.

[kold weather]

EVerything thus far in the run is a downgrade for cold, the low is further north earlier on, it takes longer to get out of the way of the UK and the upper high over Greenland is not quite so favorable in terms of its slant, I think its 50-50 as to whether the GFS shows anything decent in Fi, may show us just on the wrong side of any cold boundary in Europe...

Also add the LP in the Atlantic isn't as good either...I think this is going to be a rather poor run...we shall see though, the upper high over Greenland could save it.

Edited December 27, 2009 by kold weather

[Geordiesnow]

GFS 18Z is being too progressive with the low. BBC24 weather forecast by Rob @ 2258hrs mentioned lots of snow on Tuesday night in Wales.

Thing is though, as Izi menturned even Rob said the PPN band could be to the north and to the South and on the 21:27 forecast, the confidence of where the low will be is only medium!

GFS could be right albeit, i do think it is over-deepening the low. Oh well, we shall see by morning because the differences of having a slightly weaker low and more southerly position is huge regarding snow potential.

[TEITS]

GFS called the last cold spell well if i remember when others wobbled and this is why i would never

rubbish the GFS at the moment. It might be way of the mark but the question has to be there? All eyes

on the 0z runs.

The GFS has its wobbles as well. Remember the way it backed off the E,ly showing a very slack, non descript E,ly. This upgraded as the event came closer.

Having said this I would prefer to see model agreement. I wonder if the new GFS model is being run by a ZX Spectrum.

[kev238]

Well I think the best thing to do it keep a close eye on the bbc forecasts and also on the met office site. If GFS does turn out correct, then the real story will not be one of snow but of pretty serious floods.

[cold snap]

I know this might sound a bit stupid but the people involved in GFS will they be looking at the other models and see that not one other model goes with them?

Or will they not care less?

[Spurry]

Well I think the best thing to do it keep a close eye on the bbc forecasts and also on the met office site. If GFS does turn out correct, then the real story will not be one of snow but of pretty serious floods.

And some serious backtracking from the Beeb

[Guest North Sea Snow Convection]

There is so much variation and ridiculous over elaboration of this low dumbelling around in quicksand between each GFS output within its own set of solutions - let alone its being totally out of kilter with the other models!

[NorthantsSnow]

What I don't understand is how the GFS could be so different to all the other models I have seen at such an early timeframe. This scenario is almost unprecedented since I've been following the charts.

It must surely backtrack at 00hrs. Either that or every single other model must!

[NorthNorfolkWeather]

GFS called the last cold spell well if i remember when others wobbled and this is why i would never

rubbish the GFS at the moment. It might be way of the mark but the question has to be there? All eyes

on the 0z runs.

Yep, weird that everyone is deriding it when 2 weeks ago it was the hero of the hour for calling the spell in the first place.

A thought about data collection for ALL of the models, in an earlier thread on the models, someone posted that many flights, which are a big proportion of the collected data will be further south than normal going across the Atlantic due to the jet. I know that it should affect all models equally, but could this be part of why the GFS is on it's own? It's recently been upgraded, the algorithms have a data set which is different to most of what's been happening since the models started, it's trying to cope in the only way it knows, defaulting Zonal weather asap -- it's still learning

[johnholmes]

GFS called the last cold spell well if i remember when others wobbled and this is why i would never

rubbish the GFS at the moment. It might be way of the mark but the question has to be there? All eyes

on the 0z runs.

looking through e scores of charts I copy and save for the weekly lrf I would say that is not really true.

By T+240 EC and GFS had cold rather than mild in their song sheet. Prior to that and its always extremely difficult with GFS to pick out its main trend beyond T+240. I recall it giving very mixed signals.

As to that T+60 Fax chart-I would back that rather than GFS and don't worry overmuch about where the so called magic 528DM thickness line is.

That is far less important in frontal situations such as this than the 1000-850mb thickness line. Unfortunately the only real clue we have to that is within Extra and that is from the GFS output.

and PLEASE don;t restart the red herring about lack of data.

Neither NOAA nor UK Met nor ECMWF, if you exchange e mails with them, subscribe to that idea.

Edited December 27, 2009 by johnholmes

[rich1]

someone may need to give their computer a kick or something

If it is wrong, there will be some things for the programmers to sort out

does turn colder at the back end of the week however, for how long we shall see

hopefully this wont verify, dont fancy spending all week stuck in the house due to rain