General Model Output Discussion

[Optimus Prime]

But without the mild biased baggage of that god awful time :lol:

I don't have the time to decipher what that means. The 'mild' biased comes from a mild biased climate which we are in the British Isles, specifically in recent years.

The models really aren't showing a January 1987 or even down the line, the synoptics leading up to that were entirely different and the climate was entirely colder. What we have is an easterly pattern bringing with it the threat of snow. It's looking much like what we got in February 2005 but with some lower temperatures - so yes a great pattern.

[Geordiesnow]

It's one of those rare occasions is where if you watch a BBC forecast tomorrow morning and see where the front is, you will then find out if the UKMO is now starting to trend with the GFS or not.

If the GFS is right, then it deserves some credit and should start being taken seriously by some on here and you can give slight credit to the UKMO who also predicted for the low to be further north barring the last few runs. However if the GFS is incorrect then the credibility of this model goes even lower however i still think the GFS is a decent model picking out any medium term trends however as it gets closer to the reliable, it does quite often try to over complicate things and we often get inconsistent GFS runs.

Nervous wait for tomorrow runs, how i would love the GFS to be wrong but i got that feeling where it might be right.

[pyrotech]

It's one of those rare occasions is where if you watch a BBC forecast tomorrow morning and see where the front is, you will then find out if the UKMO is now starting to trend with the GFS or not.

If the GFS is right, then it deserves some credit and should start being taken seriously by some on here and you can give slight credit to the UKMO who also predicted for the low to be further north barring the last few runs. However if the GFS is incorrect then the credibility of this model goes even lower however i still think the GFS is a decent model picking out any medium term trends however as it gets closer to the reliable, it does quite often try to over complicate things and we often get inconsistent GFS runs.

Nervous wait for tomorrow runs, how i would love the GFS to be wrong but i got that feeling where it might be right.

Since November i have watched GFS predict somthing in fi then drop it, for the original output to be close to the actual outcome. Strange

[mackerel sky]

Scrolling through the GEFS Ensembles from South to North, it isn't until you get up to the Scottish Borders there is any consistency inside +72hrs. For the Midlands generally I would say there are only about 4-5 perturbations that would support a marginal snowfall event (so I reckon about a 30% chance at face value). It would be a new one to say the least that not a single perturbation correctly predicted the shorter term - for the 18z GEFS ensembles for Northamptonshire, only the coldest member keeps the 850s at ~ -5C for Tues/Weds. But then the precipitation charts on the the European models as clearly pointed out do not support the GFS evolution, nor do the positioning on the fronts on the FAX charts. Quite alot to get resolved in the next 12 hours by the looks of things.

Hope this one works out to everyones satisfication!

Edited December 28, 2009 by mackerel sky

[grab my graupel]

Since November i have watched GFS predict somthing in fi then drop it, for the original output to be close to the actual outcome. Strange

Yes, I agree pyrotech

The same applies to any outliers that may appear as they can often become the trend and shouldn't be dismissed IMO

[remymartin]

Having watched the ups and downs of this model watching malarkey I will just stick to now forcasting. Have hopes raised and dashed in hours, I will start just looking out of the window to see what the weather will be , by the way i want snow :lol:

[cyclonic happiness]

I don't have the time to decipher what that means. The 'mild' biased comes from a mild biased climate which we are in the British Isles, specifically in recent years.

The models really aren't showing a January 1987 or even down the line, the synoptics leading up to that were entirely different and the climate was entirely colder. What we have is an easterly pattern bringing with it the threat of snow. It's looking much like what we got in February 2005 but with some lower temperatures - so yes a great pattern.

I meant that, that time (the mid noughties), was a horrible, mild winter dominated decade, but now we seem to have passed that era and entered the "post even larger teapot era" as some would have it, which I also beleive.

Most people I know have noticed a shift in synoptics over the past 3 years that may not be entirely in keeping with modern thinking about the weather, so maybe the models and their human programmers, are missing a signpost here and there?

[mackerel sky]

Mind you here are the charts for the 5th and 6th of February this year

On the 5th we had 7 inches of snow, on the 6th we had 8 inches of snow. In both instances the automated GFS based forecasts went for Rain on the 5th and Sleet on the 6th

http://www.wetterzentrale.de/archive/ra/2009/Rrea00220090205.gif

http://www.wetterzentrale.de/archive/ra/2009/Rrea00220090206.gif

http://www.wetterzentrale.de/archive/ra/2009/Rrea00120090205.gif

http://www.wetterzentrale.de/archive/ra/2009/Rrea00120090206.gif

Difference was we did have the colder air established at the surface... and existing snow cover... we lost ours here over Christmas!

Here's a vaguely similar chart from 9 years ago to the day that gave a cold Christmas week and snow, although that wasn't marginal

http://www.wetterzentrale.de/archive/ra/2000/Rrea00120001227.gif

The MetOffice is better at forecasting marginal frontal snowfall events from the south or southwest than automated GFS output will ever be. However, in my eyes in February they pretty much decided the event was over once the snow had stopped falling in London. They only put warnings out for here once we were all stuck.

Edited December 28, 2009 by mackerel sky

[kumquat]

Excellent posts Mackerel - always enjoy your injection of realism. If anyone wants a realistic version of how the BBC use the models for forecasting including the NAE, then they should check out Ian Ferguson's posts earlier in the South West thread.

Oh - and Steve Murr's hand crafted maps look very much like those Ian Ferguson was showing in there. The initial setup seems pretty nailed with GFS having to jump backwards over hoops but only for the orientation of the initial low. But it's only a computer prediction and will expand upon starting data. This always gets the "missing data" stories flowing. I do wonder if there is some truth in this. Which is why I worry 33% that GFS has it right. I notice a little low in Italy over the Alps on GFS 18z T=00 which could start the butterfly effect. Not present in the other models at all.

Edited December 28, 2009 by kumquat

[mackerel sky]

Excellent posts Mackerel - always enjoy your injection of realism. If anyone wants a realistic version of how the BBC use the models for forecasting including the NAE, then they should check out Ian Ferguson's posts earlier in the South West thread.

Thank's for that. I do have a good memory, so the chart archive is an excellent resource for trying to compare synoptics! More4 have been repeating "Snowstorm" today which Channel 4 broadcast originally after February's fortnight of cold and snow in what was other a fairly unremarkable winter, saying it was a one in twenty years event. It was the most lying snow i've witnessed in the 9 years I've lived in Northamptonshire, fair enough, but I think those remarks only really relate to London. In hindsight, given winter so far and the commentary trying to compare a few days of unusually heavy snowfall in London before the emphasis shifted elsewhere for longer to the winters of 1947 and 1963 looks a bit daft.

Notice the the chart archive is from the 00z, lets see if there is any sniffs of it being somewhat more realistic in the short term. I guess about an hour to go... mum and baby sleeping on the sofa. Visiting parents sleeping upstairs... me awake.

Edited December 28, 2009 by mackerel sky

[kumquat]

Thank's for that. I do have a good memory, so the chart archive is an excellent resource for trying to compare synoptics! More4 have been repeating "Snowstorm" today which Channel 4 broadcast originally after February's fortnight of cold and snow in what was other a fairly unremarkable winter, saying it was a one in twenty years event. It was the most lying snow i've witnessed in the 9 years I've lived in Northamptonshire, fair enough, but I think those remarks only really relate to London. In hindsight, given winter so far and the commentary trying to compare a few days of unusually heavy snowfall in London before the emphasis shifted elsewhere for longer to the winters of 1947 and 1963 looks a bit daft.

Notice the the chart archive is from the 00z, lets see if there is any sniffs of it being somewhat more realistic in the short term. I guess about an hour to go... mum and baby sleeping on the sofa. Visiting parents sleeping upstairs... me awake.

All you need to do is get down to the quarry area of Irchester Country Park. I remember a few "black runs" there in the 80's lol

[mackerel sky]

All you need to do is get down to the quarry area of Irchester Country Park. I remember a few "black runs" there in the 80's lol

Haha! I take the dogs round there, loose rock and all. Never thought about there as a sleging venue. I used to go on the former rack and pinion mineral railway upto Rosedale, Ingleby Incline - a 1 in 3 gradient for 1000ft down the side of the Cleveland Hills. That gave me concussion.

[cold snap]

Whos waiting on GFS OZ?

Im still up 3.30 am rolling out.hopefully we will get a backtrack by them

[Jayfromcardiff]

Whos waiting on GFS OZ?

Im still up 3.30 am rolling out.hopefully we will get a backtrack by them

I woke up lol and cannot sleep so i thought i would pop on to see what's going on.

[David-kig]

poor run already !!!!! short term very confused

[Mikel Nimbus]

Waiting patiently here for an upgrade.

[David-kig]

poor run short term looks even further north dear me!!! if it couldnt go anything further north it has

really really its even worse then the 18z really doesn't like cold, the gfs has it as wednesday as rain except scotland mainly, im worried now. plus it being even worse then before dear oh dear

Edited December 28, 2009 by David-kig

[NorthernRab]

I'm liking how much further north the low pressure is on this run, really gives Scotland and especially the southern Highlands a battering

[cold snap]

The differences between UKMO and GFS are getting bigger and bigger .one of them is wide of the mark.

Lets hope its GFS .Once again folks in refuses to back down it wants that low further north.

Eyes on 6Z now.

[anvilhead]

00z not looking good for us southerners. still - i believe it will keep the forecast blizzard for nyd (for now!)

[Mikel Nimbus]

Interesting run so far. The secondary low on Tuesday/Wednesday is certainly further north, practically in Scotland! However, 00Z is showing it to rotate the main system further south a lot quicker than 18z meaning it is out of our way a lot sooner. The main low is still well placed over the bay of biscay.

There are also signs of a surface low developing over Poland. This looks like it might increase the flow of colder air towards our shores.

Of slight concern is a weakened Greenland high. Down from 1040mb to 1035mb at one point.

[anvilhead]

as I said earlier, I think forecasters will write off the gfs at thier peril! - very different to other models in the short term, but consistent..

[Mikel Nimbus]

00z is infact turning out to be a slight upgrade. Not brilliant, but better. For a start the channel low by +90 is more elongated, giving regions further south a chance with the colder upper temps. We also keep low pressure to our south east drawing in cold air. 18z yesterday was very keen for it to shift west and join the Atlantic low (highly unfeasible). And to top it off, the northern blocking situation is so that it pushes everything that little bit further south by +138.

00z at least pulls places further south into more marginal set ups instead of nailed on rain.

Edited December 28, 2009 by Mikel Nimbus

[high ground birmingham]

Well short term the GFS is sticking to its guns with the precipitation from Monday night/Tuesday's low making significant inroads into Scotland- really turning into a spectacular cold snowy spell for Scotland.

If the GFS is right short term, it must be said it is the biggest boost to the credibility of the GFS for a long time sticking to its guns against all the other models. It would be a massive victory for the GFS.

Long term, the fact that the low gets so far north does not have a significant impact medium term and long term on the 00Z GFS as the low does not fill over the UK but quickly moves into the North Sea and into Scandi... this has the effect of advecting the colder air quickly west as the low winds up and so a nice easterly shows up on the 00Z.

Regards, hgb

[SMU]

LOL- Well so far this morning the GFS has handed the biggest can of xmas whoop ass ever to the UKMO & every other model.....

Patiently awaiting the end of the week....

Sorry to the People of wales your 40Cm of could be gone without a trace- Hello Lanarkshire....

The consolation here is the picture continues to look better in the mid term-

S

Edited December 28, 2009 by Steve Murr