Model Output Discussion

[Bottesford]

Sorry for OT. Where do the "off-topic" weather thoughts some of us inadvertently post in the model thread go to? Are they simply deleted without reason or shifted somewhere? My own recent one has disappeared - it mentioned the GFS (model) a few times.

Just curious.

Mostly they will be moved to the appropriate thread. Or, if we're busy, simply deleted as there's only so much time we all have to spend tidying up. Posting in the correct thread yourselves will allow you to find your posts more easily later on

[johnholmes]

well I said I would stay out of this mad house but some of the comments in here are quite ludicrous-its the weather for heavens' sake-how many of you have a living that depends on what MAY happen beyond 120 hours?

Fine detail be it now or in mid summer is NEVER that good much beyond T+72 and at times much less.

One or two corrections

SC

The models, none of them, are programmed to do anything, even Met with its obvious bias that you see for GW, to do anything other than use advanced mathematics to solve the extremely complex thermodynamic equations from t+00 seconds out to, whatever time scale that centre decides on.

ECMWF is still the leader in the northern hemisphere 500mb charts if you check with the NOAA site that gives the 5 and 6 day averages for all the main models, both in the past 2-3 days and most certainly over the longer time frame.

and this although I do not like making personal comments, but from a once respected NW forecaster really is ridiculous

Ian Brown, on 10 December 2009 - 19:38 , said:I think after this we need to look at the reasons why we cannot get stable High latitude blocking in the winter months, yes the ECM has chopped and changed but the evolution now looks very plausible and the other models are now trending to move the High back SE.

A very sorry episode for the GFS in particular. Best to make the most of at least a more seasonal feel before the return to Atlantic weather.

and to those posters who persist in saying the next xx run is crucial, all will be decided by this or that model on the next run-bunkum.

Edited December 10, 2009 by johnholmes

[Altostratus]

I smile to myself when people on here comment that the ECM is struggling or argue that because it has been the least consistent model from run to run, it can be discounted. At the end of the day, they are just computer models built around the laws of physics. They show it how it is; although they are obviously only as good as the data fed in. Perhaps the ECM has been inconsistent because it is more sensitive and picks up smaller details that the other models miss – I’m guessing it has a better resolution. If so, the ECM is likely to be the more accurate model and may be on to something – it can certainly not be ignored. The downgrade trend has been evident for the last few runs now. There is still time for small changes, but as we get the nearer, the error margin decreases giving less room for manoeuvre. On the flip side, it is feasible that the ECM has been oversensitive and overdeveloped a small feature which has been magnified as the run progressed. If the input data was good and the run isn’t an outlier, this is the only straw I can see us clutching at present.

[cold snap]

All negative at the minute

Can someone with good detail please tell me why ECM is wrong?

Lets lift the spirit here Please

[Mondy]

Mostly they will be moved to the appropriate thread. Or, if we're busy, simply deleted as there's only so much time we all have to spend tidying up. Posting in the correct thread yourselves will allow you to find your posts more easily later on

Thnaks for clarifying.

[yamkin]

Let's cheer up with this from JMA 12Z 192h

[k19]

when is the 18z gfs run released, what time?

[David-kig]

9.30 the gfs 18z comes out

[Frosty.]

All three will be on board but which ship.....

Probably the titanic

We are rapidly approaching the iceberg....I mean, the crucial 00z runs which will probably decide the fate of the run up to christmas, the ecm 12z gives us a chilly week but generally fine with sunny spells and night frosts with fog patches although the southeast flirts with some colder air off the near continent with a chance of rain showers turning sleety by mon/tues. The rest of the week looks set fair everywhere. The other models give us a real blast of winter but perhaps they won't by tomorrow morning.

[londonsnow]

The most anticipated 18Z GFS i can ever remember.

Will it side with ECM and leave UKMO on its own?

I Will be heartbroken If GFS pulls the plug on this.

lol. so anticipated im opening the wine and ordering a pizza!!! trust the gfs, i dont think it will back down. if it does the wine will take the edge off the dissapointment at least

[NorthantsSnow]

when is the 18z gfs run released, what time?

9:30pm it starts. Complete by 10:45pm.

[I remember Atlantic 252]

when is the 18z gfs run released, what time?

9.30-10.45pm www.meteociel.fr

Edited December 10, 2009 by mark forster 630

[weather vain]

Been watching this forum for days now. Can I ask a quick question ....

If we were back 3 / 4 days ago and we wished we could forward time a few days to see if the model came up to its potential. So here we are - would people be happy with the way things have turned out thus far not ?

[k19]

9:30pm it starts. Complete by 10:45pm.

do the frames update one at a time?

[Snowy Bob]

Evening all,

Having been on here every winter for the past few years, one of the main things that I do tend to think is that whatever happens, it will always been interesting and nothing ever goes as it is predicted.

[Harsh Climate]

whens the ecmwf esembles out?

[Solar Cycles]

well I said I would stay out of this mad house but some of the comments in here are quite ludicrous-its the weather for heavens' sake-how many of you have a living that depends on what MAY happen beyond 120 hours?

Fine detail be it now or in mid summer is NEVER that good much beyond T+72 and at times much less.

One or two corrections

SC

The models, none of them, are programmed to do anything, even Met with its obvious bias that you see for GW, to do anything other than use advanced mathematics to solve the extremely complex thermodynamic equations from t+00 seconds out to, whatever time scale that centre decides on.

ECMWF is still the leader in the northern hemisphere 500mb charts if you check with the NOAA site that gives the 5 and 6 day averages for all the main models, both in the past 2-3 days and most certainly over the longer time frame.

and this although I do not like making personal comments, but from a once respected NW forecaster really is ridiculous

Ian Brown, on 10 December 2009 - 19:38 , said:I think after this we need to look at the reasons why we cannot get stable High latitude blocking in the winter months, yes the ECM has chopped and changed but the evolution now looks very plausible and the other models are now trending to move the High back SE.

A very sorry episode for the GFS in particular. Best to make the most of at least a more seasonal feel before the return to Atlantic weather.

I worded it badly John, what I meant is, these sort of evolutions are not the models forte. They do very well at modelling most things, this sort of set up isn't one one of them!

[NorthantsSnow]

do the frames update one at a time?

That is correct. With so many frames to rollout however they do tend to appear quite quickly once the run gets going. But to be honest, not sure I have the stamina tonight.

[Alex95]

whens the ecmwf esembles out?

I don't know when they come out, maybe someone else knows? But I definitely think that people should wait until they come out before even suggesting which run to disregard, the ECM, GFS or UKMO the prime contenders. No run should be disregarded for a couple of days yet, but once the ensembles are out I could forgive anyone for backing one model more strongly than another.

Also, does anyone know where you can access the ECM ensembles?

Edited December 10, 2009 by alza

[beng]

http://www.cpc.noaa.gov/products/predictions/814day/500mb.php

This chart does not really fit with the ECM

[Harsh Climate]

I don't know when they come out, maybe someone else knows? But I definitely think that people should wait until they come out before even suggesting which run to disregard, the ECM, GFS or UKMO the prime contenders. No run should be disregarded for a couple of days yet, but once the ensembles are out I could forgive anyone for backing one model more strongly than another.

I totally agree. Wouldnt it be funny if this latest ecmwf run was a massive mild outliar and the rest cold

[cooling climate]

I have just copied this from NOAA this evening.

TODAY'S NUMERICAL MODELS ARE IN FAIRLY GOOD AGREEMENT ON THE EXPECTED 500-HPA

HEIGHT PATTERN THROUGH THE 6-10 DAY PERIOD. THIS PATTERN FEATURES UPPER LEVEL

TROUGHING OVER THE EASTERN PACIFIC AND MUCH OF ALASKA, UPPER LEVEL RIDGING OVER

THE ROCKIES AND CENTRAL U.S., AND UPPER LEVEL TROUGHING OVER THE EASTERN THIRD

OF THE CONUS. TODAY'S 06Z GFS ENSEMBLE AND 00Z CANADIAN ENSEMBLE PREDICT MORE

UPPER-LEVEL TROUGHING OVER THE EAST THAN THE 00Z ECMWF ENSEMBLE, WHICH DEPICTS

A LESS AMPLIFIED PATTERN WITH A BROAD RIDGE COVERING MOST OF THE U.S. THE 06Z GFS ENSEMBLE HAS SHOWN A VERY CONSISTENT TREND TOWARD AMPLIFYING THE PATTERN,

WITH THE UPPER-LEVEL RIDGE BUILDING WESTWARD INTO THE PACIFIC NORTHWEST, AND

THE UPPER-LEVEL TROUGH BUILDING SOUTHWEST INTO THE SOUTHEAST AND SOUTH CENTRAL

U.S. IN ADDITION, THE 06Z GFS ENSEMBLE MEAN SOLUTION IS VERY SIMILAR TO THE 06Z

DETERMINISTIC GFS SOLUTION, FURTHER INCREASING CONFIDENCE IN THE GFS SOLUTION.

THEREFORE, TODAY'S OFFICIAL 500-HPA BLEND FAVORS THE 06Z GFS ENSEMBLE.

[weather eater]

I think some are forgetting this is a week away, so until Tuesday (at the very earliest) we will have a better idea of what may (or may not) happen. Some members are jumping the gun at a few dodgy runs, as members keep saying there will be highs and lows from each run, all the models are chopping and changing so there is little if any certaincy as to what will happen at then end of next week. All we know is it will get colder.

That’s a sound post; we are in danger of forgetting just how close to an event we often need to be before we get model agreement. Normally I trust the ECM over the GFS, if it had shown a consistent alternative evolution I would jump to back it, but it has failed to do so. Even at 96hrs the differences are quite stark, if the ECM proves right over the GFS, UKMO, not to mention most of the minor models, I will be very surprised.

[Harsh Climate]

http://www.cpc.noaa.gov/products/predictions/814day/500mb.php

This chart does not really fit with the ECM

lol couldnt be more different. interesting times ahead.

[Iceberg]

Here's the ECMWF pressure plotted onto the GFS ensembles for a random town south of the UK, (this shows how far south the HP is likely to fall.

It's quite clear that ECM really doesn't have very much if any support.

I would be reasonably certain that the ECM run has little support from the EPS. It would be very unusual for the two sets of Ensembles to be so far out at this time range.