Model Output Discussion

[kold weather]

The 18z ensembles coming out, ensemble average looks pretty good to me though at 96hrs, certainly a bit different to the ECM. Control run looks decent with regards to the easterly in the early part of next week.

The 0z runs are going to be so mentally key, but at the end of the day we don't know whats going to happen down the line.

[weather eater]

What’s interesting at the moment is we have,(at least until the parallel takes over completely) two runs to mull over every time the GFS comes out. I find it difficult to believe that neither at any time have blinked and followed something akin to the ECM evolution or backed off the easterly in any way, that seems to me to be quite extraordinary.

[snowlover2009]

I've just looked at the GFS and ECM and they side to the GFS no question . High further north .

There is no way they follow the ECM

He must be reading it wrong then. I reckon ecm will upgrade tomorrow morning, although the 12z ecm could show some very very similar to gfs or ukmo, i cannot see it flipping so quickly to gfs, but who would be surprised.

[davehsug]

I suspect that the next ECM chart will fall into line with GFS and others, meetings and phone calls will be going on now to save face I think so I think that the next ECM will heavily backtrack and fall into line with GFS

Saving face? How many people know that the ECM even exists apart from the 0.1% of the population who are as obsessive as us! The people who do know of it's existence will (or should) also know that it's liable to error, just as much as any other computer model. The fact that there is so much disbelief that it MAY be wrong, goes to show how reliable it usually is, in a way it's a back-handed compliment to the ECM.

[The Eagle]

Hopefully not too many disappointed people in the morning. I expect the ECM to be nearer the mark but you never know. Going on experience the ECM will be right and the 00z GFS will be harsh (thats putting it mildly!). Hopefully both will be crackers and a pig has just flown by my window

BTW Im basing that on the models- nothing to do with Ian Brown's theories which I respect but don't agree with.

Edited December 10, 2009 by The Eagle

[John Snow]

They are at the mercy of the data that is input to the model - the model may be wrong but you can be assured that the experts at the ECMWf wont base their forecasts on that one model run. Also the GFS has been "spectacularly wrong" (as a forecasting model) on many more occasions than the ECM in the two and a half years I have been model watching.

My area will not do well out of the GFS at 168/180, we might get a dusting whereas Newcastle will get 50cm, Yorkshire about 40cm, Fife about 40cm and Lincolnshire about 30cm if the GFS is correct so I am not excited about the GFS more envious lol

[TEITS]

Well i've looked at all the runs again and to be honest for the period between +72 & +144 I think we are going to see a blend of the GFS/UKMO. I said earlier I wasn't convinced the E,ly early next week would be as potent as the GFS suggests and im still not.

Look at the GEFS mean for +96 and to be honest the GEFS mean is similiar to some of the UKMO runs.

http://91.121.94.83/modeles/gens/run/gens-21-1-120.png?18

However look at the GEFS mean at +144.

http://91.121.94.83/modeles/gens/run/gens-21-1-144.png?18

So at the moment my summary for next week would be this based on the models so far. Early next week turning colder with wintry showers. These wintry showers dying away midweek, however towards the end of the week turning colder from the NE with snow showers becoming widespread, heavy and prolonged in places.

This is subject to change!

[Duane S.]

FAX does NOT follow the ECM but is a little bit disappointing nevertheless, I was expecting better.

All in all though another good day and another day closer to it all kicking off.

[stewfox]

absolutely the ecm was the most consistent from the start so theres no reason to suspect its wrong and it is the top model out of them all,

and most of the more expert people are not giving there thoughts its easy to side with the models that give us what we want.

so its better to be realistic,

from tonights ecm there will be colder air but nothing very wintry apart from fog and frost and nice sunny days,

this was always the risk with such lovely recent outputs.

but ecm is number one in the charts i dont really know how the gh will effect us in the future i have a bad feeling this may well lose its grip aswell.

but it is still early days im sure gp would agree that this event was a close call,

but we all know that theres a chance of another shot perhapes a little later in the winter,

like suggested mid january.

This has to be the quote of the week

[kold weather]

Well if the GFS does fail then the GFS, esp the new paralell would seriously lose some of its reputation thats for sure, esp after about 20 solid runs of cold easterlies, I've never ever seen a model so insistant on a certain set-up occuring so we will see!

The 0z will be very interesting, will I be the onlyone whos going to be worried when waking up tomorrow!

[J10]

The teleconnection forecasts back up what the GFS and all the other sources come up with that are not called ECM, they may be respected but how can they get something like this so seemingly spectacularly wrong ?

We cannot say though that the ECM is wrong until next week, in the same way we cannot say GFS is wrong for the same reason.

However whichever of the two gets it wrong, will indeed have a major credibility issue in this instance.

For cold and snow weather fans clearly a solid GFS and parallel GFS run in the reasonable and indeed reliable time frame.

Well if the GFS does fail then the GFS, esp the new paralell would seriously lose some of its reputation thats for sure, esp after about 20 solid runs of cold easterlies, I've never ever seen a model so insistant on a certain set-up occuring so we will see!

Absolutely

[damianslaw]

The real cold air still remains just out of the reliable timeframe with the GFS and Meto have it pouncing on us by Wednesday. Most models are agreeing that some showery weak front will spill over the high Sunday/Monday bringing drizzle and sleety stuff in places.

I think because the cold uppers are still out of the reliable timeframe we still have some difference in the big three models, with ECM not keen on transporting the cold north easterly in. However, expect them all to be singing from the same hymn sheet by Saturday and hopefully it is the GFS/UKMO synoptical evolution being shown now rather than ECM. Having said that the ECM evolution would not be a total disaster, even if the high eventually sank, the upstream teleconnections read deep negative AO and NAO so a scandi trough could easily form with strong greenland blocking behind.

With the upstream signals I would expect the ECM to follow GFS/UKMO by Saturday with the real cold air and north easterlies by middle of next week... but this is not nailed on the head just yet, but we will know soon enough.. fingers crossed.

[TEITS]

Just enjoy the cooler settled interlude and avoid the wake on here tomorrow morning.

I know this will be removed but I can't let it lie.

You know Ian there was a time when I had a great deal of respect for you. However your comment above only proves that your motivation for posting is to wind members up.

[snowlover2009]

This has to be the quote of the week

I agree LMAO, its ok to have opinions on the models, but saying we have to be realistic and believe the models which is going for the christmas pudding style set up is most likely to come out instead of ukmo, gfs, fax etc. yes there is good number in more of an agreement. if anything is to be believed, people would beleive that all models but ecm would be likely to frutition because of the consitency.

Edited December 10, 2009 by snowlover2009

[stewfox]

We cannot say though that the ECM is wrong until next week, in the same way we cannot say GFS is wrong for the same reason.

I thought the ECM data input was flawed (message from Holland) ?

Not just what we didnt want to see

[Just Before Dawn]

They are at the mercy of the data that is input to the model - the model may be wrong but you can be assured that the experts at the ECMWf wont base their forecasts on that one model run. Also the GFS has been "spectacularly wrong" (as a forecasting model) on many more occasions than the ECM in the two and a half years I have been model watching.

You are right wysiwyg - I think people are mistaking the outputs as Forecasts (they are, of a sort I know) when in fact they are theoretical outputs from computer models and are only as good as the data inputted and like all theoretical modelling output, they require interpretation from experienced human expertise. As a result (particularly given the complexities of the systems they are attempting to model) 'FI' actually begings at t+3 hours, it's just that FI has degrees of reliability.

For what it's worth, the wild array of different solutions the ECM has produced over the last 72 hours in curious and suggests possible data input issues, but before we crucify it, lets not forget that the GFS has blown a raspberry at timespans shorter than these. The truth is, no model is perfect.

I agree with Kold - the 0z runs are pretty crucial now. The 18z outputs are pretty impressive for us in the Lincolnshire Wolds, which are well placed for both cold shots, however experience suggests snow is a matter of nowcasting anyway - more than 24 hours out, and it's a rough guess really.

[Geordiesnow]

My area will not do well out of the GFS at 168/180, we might get a dusting whereas Newcastle will get 50cm, Yorkshire about 40cm, Fife about 40cm and Lincolnshire about 30cm if the GFS is correct so I am not excited about the GFS more envious lol

I wish!!

I agree with TETIS regarding that the first easterly the GFS is predicting probably won't happen, i think the high will be more over us similer to the ECM/UKMO but to be fair, it has been showing for quite a few runs now so maybe they will come more in line by tomorrow morning hopefully.

The 2nd easterly chance is certainly possible but this can still go wrong, i think we may have to wait a few more days before we know whether we will get a potent easterly with snow or an weaker easterly with cloudy but dry conditions.

[snowhope]

The last 3 nights especially has seen the 18z upgrade the forthcoming week in terms of cold and snow. IF you compare it to the previous days 18z like we are constantly reminded to.

Cant wait to see tomorrows!

Edited December 10, 2009 by snowhope

[snowmadchrisuk]

This has to be the quote of the week

Quote of the Century if your talking about how wrong it is. GFS has been the most consistent it has ever been in all my years of model watching . ECM has been chopping and changing every single run.

[Neilsouth]

NOGAPS 18z now out to 144 and whilst its still playing its own game, it has a high firmly established to the north-east of the UK with -10/12hPa heading for the south-east!!

Nice

[Duane S.]

What’s interesting at the moment is we have,(at least until the parallel takes over completely) two runs to mull over every time the GFS comes out. I find it difficult to believe that neither at any time have blinked and followed something akin to the ECM evolution or backed off the easterly in any way, that seems to me to be quite extraordinary.

This is exactly what I have been constantly saying as well. I have been incredibly impressed with the consistency of the model, like no other I have ever seen in the run up to a cold spell, without qusetion. But of course, at this stage, you can only be impressed with consistency, not yet verification.

Hopefully though we will be lauding it as the king of models in a week's time and not writing it off before it has even begun!

All very intriguing to see how this pans out...

[karyo]

18z ensembles show the operational and control very close together throughout the run! Very close to the mean also!

http://www.netweather.tv/index.cgi?action=ensviewer;sess=

Karyo

Edited December 10, 2009 by karyo

[John Snow]

I wish!!

I agree with TETIS regarding that the first easterly the GFS is predicting probably won't happen, i think the high will be more over us similer to the ECM/UKMO but to be fair, it has been showing for quite a few runs now so maybe they will come more in line by tomorrow morning hopefully.

The 2nd easterly chance is certainly possible but this can still go wrong, i think we may have to wait a few more days before we know whether we will get a potent easterly with snow or an weaker easterly with cloudy but dry conditions.

http://cds047.lo1.hwcdn.net/c9s4a5k3/cds/gfsimages/gfs.20091210/18/159/h850t850eu.png?dopvhost=charts.netweather.tv&doppl=0b34b3014015305c4045300c91803b4740453028&dopsig=2847bbcb8fe3513e99f1f84bbc7a4117

This chart, you are bang in the firing line and with the lake effect off the North Sea you will get the mother of all pastings if my limited understanding of charts is correct

Respected forecasters have talked of you getting "hammered" so my 50cm prediction I believe will not be that wide of the mark

[Snowaddict]

In spite of an excellent GFS 18Z Operational, Control, Parallel Run & Ensemble Mean... I would say be weary, very weary because it's very, very uncommon for ECMWF to backtrack as such a close timeframe to realtime.

The differences are stark this evening, even from T+72..

Again, as what I said earlier, past letdowns have learned us that GFS can backtrack in one run...

As many have said, the 00Z's will be absolutely crucial, along with tomorrow's 12Z's..

In light of every available source of model guidance tonight, including FAX charts, I would lower the potential for a snowy outbreak to 30% & no higher..

It will turn cold, increasingly cold, but whether it will be sustained or snowy is really open to question.

I fear that the ECMWF 12Z run will be nearer the mark, but I hope & pray that GFS is correct.

I think everyone should lower expectations slightly just in case an overnight reversion to ECMWF 12Z does take place.

[kold weather]

It's going to be fascinating as the models go head to head to the final fence, I imagine most will be viewing thru the gaps between thier fingers !

Certainly will be, to be fair the ECM is a very good model and it can be a good trend catcher BUT the GFS and its Paralell along with the UKMO model has been ultra consistant, far more then can be said about the ECM which has been unable to decide.

Will be very interesting to see what comes out on top!

Ensembles still very suggestive of the cold spell occuring as well.

Saying all that, I am still half expecting a downgrade TBH, simply because thats what normally occurs!

Edited December 10, 2009 by kold weather