Model Output Discussion

[nick sussex]

A disappointing gfs 18hrs run which shows why easterlies are a real pain to verify and also so difficult to accurately forecast. I think we've seen here the problem again near Iceland which always seems to be the trouble maker in terms of shortwaves. The euro high still looks just about on but even the gfs is hardly ridging this far enough north to bring inversion cold so altogether its been an evening to forget in the model output.

As for tomorrows output well who knows, we saw the ensembles suddenly switch yesterday with many more going for an easterly, then today they've reacted to more energy spilling eastwards off the main trough, perhaps they've overreacted on both counts, hopefully we should see the models settle down tomorrow so that we can see what they do with the euro high and whether they are determined to continue the shortwave love-in!

[NorthantsSnow]

Next 24-36 hours of model watching are crucial. Not too late to revert back to a colder scenario but at the moment it's looking a lot less likely than it did yesterday. The chances of a cold easterly mid month are probably close to 1 in 10 now compared to 1 in 5 yesterday. That's not to say we won't have another shot later on but personally I wouldn't expect another shot too soon if this first attempt fails and the cold block floods east.

The ensembles will be interesting later because the 12z set were very anti cold and the spread was much smaller than we have seen recently, suggesting increased confidence in the milder scenario winning out.

What a pile of junk! :lol:

Looks like the parallel run could be saving the day....or night, at least its not been up the pub and can take over in just over a week from this sad old beast from the bog.

It looks good on the face of it and certainly the cold does reach us which is more than can be said for the past few operational runs at least but it's not the full throttle easterly we have seen flagged up before. The uppers are still plenty low enough, but the cold air only really skims past us rather than a direct hit. Better than nowt though particularly for the south east, and the cold block is very juicy to look at, just a shame it's not a bit more favourable for the UK in terms of angle.

http://91.121.94.83/modeles/gfsp/run/gfs-1-288.png?18

[NorthNorfolkWeather]

Comparing the Current and Parallel runs, they are miles apart. Will be interesting how they 'splice them' if the 15th is the cut over date and the outputs are opposite ends of the warmth spectrum. -10 uppers compared to +3 uppers is a significant difference

[cooling climate]

At this range, remember we are talking 10 plus days away I would be very surprised if the models

had the pattern nailed.

I would not be at all surprised to see the models revert to a very cold pattern again tomorrow or

monday. This is not just wishful thinking, Arctic highs are notoriously difficult to model plus any

shortwaves at t120 could very well disappear altogether once it gets inside the more reliable time

frame.

Nothing has changed my mind yet that the pattern is not going to change to a much colder and

wintry one.

If by Tuesday the models have not started to backtrack then yes maybe will have to be a lot more

patient.

[nick sussex]

Comparing the Current and Parallel runs, they are miles apart. Will be interesting how they 'splice them' if the 15th is the cut over date and the outputs are opposite ends of the warmth spectrum. -10 uppers compared to +3 uppers is a significant difference

It will be interesting to see what happens with the parallel run as they're desperately trying to improve gfs reliability up against the ecm. You have to laugh down here the P run brings 850's close to -16!

[David-kig]

wat time are the gfs elsembles?????

[Mark Bayley]

Just had a quick scan of the ensembles and there not great. I think the likely scenario is for high pressure to stay close by keeping us settled. I still think its very possible we could be in a cold spell by Christmas. Lets just see how things pan out. Hopefully there will be some surprises :lol:

[Paul]

wat time are the gfs elsembles?????

They're out now:

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

[Frosty.]

I think expectations are getting too high which just makes the current output seem even worse than it actually is. We should be happy there is an end in sight to all the heavy rain and floods with a quiet and chilly spell setting in later next week with some frosts and fog. Nobody and no computer model yet knows where the high will end up and how it will impact on the uk.

Edit..although I think GP knows whats going on with the background signals and the bigger picture, we might just have to wait a bit longer for the beasterly. :lol:

Edited December 5, 2009 by Frosty039

[Bristawl Si]

IMO the important thing to remember is not each run individually, but the trend of many runs.

If we look back 5/6 days were the very cold temps to the east and north east even showing for mid-Dec on the models?

The trend over the last 2/3 days model runs shows very cold temps to our east and on some occasions encroaching on our shores. It's trends what matters, not this run or that run.

I think the models are definitely hinting at a major change in our weather leading up to Christmas. Exactly when it'll arrive we know not yet. If i had to put a date on it, i'd say around 17th/18th December.

[J10]

This is always going to be a problem with a near strong El nino ,the jet will nearly constantly be about in a fairly strong manner, its just whether there is enough blocking around to hold it off, however even then you tend to find patterns can be fluid, even more so then you'd normally expect.

Anyway for now the models seem to have moved away from the cold easterly and instead keen to set-up a high close to the UK. In that set-up it really isn't going to make a huge difference where it ends up as long as its not to ofar from the UK.

As for the GFS temps, I'm never more amazed then to see how often it completely misses inversions even when they have started, got a feeling you will see the same with this one as well.

I think it was 2006 (Late Jan/ Early Feb), when the GFS was forecasting temps of 7-8c for the next day and the actuals were closer to 0-2c, due to an inversion.

[snowlover2009]

have you seen how different the gfs operational is compared to al the other ensembles in athens. way different!

[Snow Queen one]

Ill take the operational run please at least it will give us in the SE some cold weather. I have waited a long time for our cold to come from europe every year they seem to freeze and we just miss out and there is only a few miles between us it is frustrating. I hope this year that we in the SE can tap into the cold just once.

Still lots of time even though the pub run is out all is not lost i don't take much notice of the pub run it does not have as much info fed it than the other runs.

[Ripeberry]

Listen to Bristle boy. He talks sense! Look for the trends not the prozac inducing runs.

It's gonna be cold at least once or twice this winter. The best cold 'attacks' always come out of the blue!

[kumquat]

have you seen how different the gfs operational is compared to al the other ensembles in athens. way different!

There are real problems here. The start data (for a start!) is way out of kilter.

[NorthNorfolkWeather]

have you seen how different the gfs operational is compared to al the other ensembles in athens. way different!

How can the operational be an outlier from almost beginning to end?

Methinks there's something not quite right about the data input

damn, beaten to it by a Kumquat

Edited December 5, 2009 by NorthNorfolkWeather

[phil nw.]

A disappointing gfs 18hrs run which shows why easterlies are a real pain to verify and also so difficult to accurately forecast. I think we've seen here the problem again near Iceland which always seems to be the trouble maker in terms of shortwaves. The euro high still looks just about on but even the gfs is hardly ridging this far enough north to bring inversion cold so altogether its been an evening to forget in the model output.

As for tomorrows output well who knows, we saw the ensembles suddenly switch yesterday with many more going for an easterly, then today they've reacted to more energy spilling eastwards off the main trough, perhaps they've overreacted on both counts, hopefully we should see the models settle down tomorrow so that we can see what they do with the euro high and whether they are determined to continue the shortwave love-in!

Yeah a step back from an easterly today Nick.

As soon as the shortwaves appeared it was game up for an early shot of cold and those heights over the pole programed for T144hrs on have diminshed.

There`s just too much energy coming off the Canadian vortex to our NW compared to yesterday`s runs.

I think it`s odds on we are now looking at a midlatitude high nearby for a weeks time with some cooler surface temps but any real cold will stay East for now.

Things can change later on though and if the vortex remains fragmented there is always another chance when the jet allows.

[Nick L]

There are real problems here. The start data (for a start!) is way out of kilter.

It's a good job the GFS is being replaced soon!

[SMU]

If your wanting cold out of this then we need a second pulse of WAA out if the atlantic to help push that shortwave back south or wait for it to go west through greenland- that really is our only hope of good cold- I threw that comment out on my last post earlier & there is some represntation of this on the 18z(p)- it the same jet profile as we see at 120 but just retrograded back around 300 miles-

here it is-

http://91.121.94.83/modeles/gfsp/run/gfsnh-0-174.png?18

S

[snowking]

have you seen how different the gfs operational is compared to al the other ensembles in athens. way different!

http://charts.netweather.tv/ensimages/ens.20091205/18/t2mAthens.png

Perhaps one of the best ensemble sets ive seen for a long time. The sooner they ditch the current operational run the better - i'm not sure at present how anybody can place any trust in NWP output, beyond even 3 days its just been pathetic these last few weeks. I think its very much a case of looking at teleconnection and (thanks to the experts in the particular field) the stratospheric analysis, which over the past few weeks has been really rather accurate reading back through some of GP's and Ch's posts.

Theres also been a pretty noticeable decline in the verification correlations in the last few days right across the board

http://www.emc.ncep.noaa.gov/gmb/STATS/html/acz5.html

http://www.emc.ncep.noaa.gov/gmb/STATS/html/acz6.html

A sign of uncertain times

[kumquat]

I think the main problem is lack of land. Lets have a land grab. Can we pave over the GIN??

"There`s just too much energy coming off the Canadian vortex to our NW compared to yesterday`s runs." Good point - well made.

Edited December 5, 2009 by kumquat

[SP1986]

The Athens pressure and atmospheric temperature runs (discounting the operational) suggest that there is no signal for easterlies to reach the UK, if this was the case I would expect low pressure of approximately 1000mB in the Athens ensembles if that was to occur. It to me suggests a strong high pressure never really gets a grip at all.

[pyrotech]

IMO the important thing to remember is not each run individually, but the trend of many runs.

If we look back 5/6 days were the very cold temps to the east and north east even showing for mid-Dec on the models?

The trend over the last 2/3 days model runs shows very cold temps to our east and on some occasions encroaching on our shores. It's trends what matters, not this run or that run.

I think the models are definitely hinting at a major change in our weather leading up to Christmas. Exactly when it'll arrive we know not yet. If i had to put a date on it, i'd say around 17th/18th December.

Quite right BB. We are finally heading in the right direction with trends, even if trend becomes only cold for europe to start with.

So many possibilities fron then on which are better than last month. Cold and dry maybe to start with then anything can happen, very complexed, not given up on easterly that is still possible, even a north blast possible by 23rd. GFS very undecided on what it wants for us, but seems consistant with cold for Northeast Europe. I could not attempt a prediction with this set up, this could easy go anyway amd 850 temps could be anywhere between +5 to -10 by 23rd of December. If i did it would be biased on my want of the easterly, i am one of many on here who often sees what he wants and ignores the models as on a pub run when they say something to the contary.

(Anyway i find the whole weekend GFS more fickle than on week days, one long pub run till monday )The serious point is that its 100% better than we seen so far this winter on all the models.

If all we get is a high with night time frosts at least it would feel like xmas. Seems this year has only had two seasons ( wet and very wet) a little winter frost would be welcome.

Edited December 6, 2009 by pyrotech

[LomondSnowstorm]

If I recall correctly, the 18Z operational did the same last night, with Athens again managing to pick up -10 upper air temps on the operational with every other run showing it above 0C! Let's hope the parallel doesn't do this often. It will also be interesting to see if it has a 'pub run', or whether it can resist a trip down the boozer most evenings at 6!

LS

[smich]

The word "outlier" is used erroneously far too often on here.

Above is an example of a real outlier. Not towards the warm or cold end of the scale, but properly at odds with every other member.

Everyone please note: an outlier

Smich