Model Output Discussion

[TEITS]

Having said that, they put out the advisories for today on Sunday consistently for the same counties, only to completely change it a few hours ago. If I was an end user of their forecasts, I'd be a bit cheesed off, had I made any preparations that might have financial consequences.

Yes tomorrows possible snowfall is a very good example of how it can suddenly change. I've only just looked at the NAE and the 18Z is completely different to the 12Z in that the precip intensifies around the W Midlands. This is the complete opposite to what I thought would happen as I expected the front to weaken as it backed W.

You do have to feel sorry for the Met O because in these situations it makes forecasting virtually impossible. Having said this the uncertainity in the models makes it even more exciting to follow.

[Cumulonimbus Tower]

Yes tomorrows possible snowfall is a very good example of how it can suddenly change. I've only just looked at the NAE and the 18Z is completely different to the 12Z in that the precip intensifies around the W Midlands. This is the complete opposite to what I thought would happen as I expected the front to weaken as it backed W.

You do have to feel sorry for the Met O because in these situations it makes forecasting virtually impossible. Having said this the uncertainity in the models makes it even more exciting to follow.

The NAE has shifted the whole lot north west by about 60 miles a run for the last 12 hours. The sweet spot seems to be right over me by this time tomorrow but i am not counting on it being the final position...back south by about 40 miles i would say

[weather eater]

The thing is WE, I cannot recall a single post from you with any real analysis of the models (yes before it is said, nor have I, I'm simply not clever enough). What I have seen is plenty of posts about how others should not look too far ahead and anything beyond 96 hours is hogwash. Then I hear people say that the models are struggling, only to be chastised with something along the lines of ' they are struggling no more than computer based weather models do beyond x hours', you really can't have it both ways. Same could be said that you might be worth paying attention to if you provided some real analysis of the models, rather than policing the thread for anyone that dares to discuss FI. If only up to T96 was discussed it wouldn't be worth reading the thread, we all know the caveats of charts becoming reality the further out that they go...

You may well be right, I’m rather like you I come to learn, but learning is not helped by bias posts, I don’t think it’s healthy to allow that to continue to happen without saying something because what you end up with is some sort of cold mafia. As for FI charts, for my mind they always struggle at 144hrs plus, sometimes less, my gripe is not that charts beyond that are talked about, that fair enough, it’s how they are given credence when cold is on the menu and dismissed when it’s not, it doesn’t help anyone trying to learn and it certainly contributes greatly to the arguing that goes on, I logged on this morning and the first two pages were fisticuffs in the school yard. I read more than I post because I don't think I'm good enough but that’s no reason why I should have to remain silent about bias posts or have to skip half the thread because of rows about bias posts, maybe I'm rubbish but I try to remain objective when I do comment, and if I’m becoming a bore by constantly making policing style posts then I’m sorry but it’s not in my nature to remain silent and be bullied into excepting the status quo.

[ScottSnow]

I logged on this morning and the first two pages were fisticuffs in the school yard. I read more than I post because I don't think I'm good enough but that’s no reason why I should have to remain silent about bias posts or have to skip half the thread because of rows about bias posts, maybe I'm rubbish but I try to remain objective when I do comment, and if I’m becoming a bore by constantly making policing style posts then I’m sorry but it’s not in my nature to remain silent and be bullied into excepting the status quo.

Calm down, Calm down !!! I don't think anyones being bullied, we are talking about the weather on a forum on the internet. Let's put things in perspective Weather Eater.

FI is relevant in that trends can often be picked out whether it be a mild or cold trend. Most people look out for cold trends and because they like to see snowy weather they probably notice them more.

The weather we've got currently is exciting, with the potential for snowfall almost anywhere in the UK in the next few days. It just goes to show that once cold air is in place then opporunities for snow will almost always arise and its best not too look to far ahead when forecasting snow. Thats why I tend to ignore people who forecast snowfall amounts days in advance and just try to make sure the cold air is in place first.

On the models interesting pattern at the moment and I see no let up in the colder/cooler weather especially in Scotland. The very cold air over Scandanavia may have an impact on our weather at the end of the month with northern blocking in the form of the Siberian High pushing the cold air westwards.

Edited February 17, 2010 by Snow White

[Thundery wintry showers]

In many ways, the power of the Atlantic varying widely in FI isn't surprising, after all the power factor is a function of cos Ø.

As for the model outputs themselves, ECMWF goes mild at around T+216-240 with the jet trending north, and IMHO that evolution is more likely than an eventual easterly (though I think it is unlikely to arise as early as the ECM suggests). Otherwise it stays fairly cold and showery with some more general rain/sleet/snow at times for the south and particularly southeast.

[full_frontal_occlusion]

In many ways, the power of the Atlantic varying widely in FI isn't surprising, after all the power factor is a function of because Ø.

Apologies TWS, gotta pull you up on this one:

The Atlantic varying in FI is a direct function of Heisenbergs uncertainty principle and nothing to do with power-factor which is simply the product between voltage and current in sinusoidal AC systems.

ffO.

[Thundery wintry showers]

The original post was meant to be a joke using the Greek letter as a pun on FI- not meant seriously!

[stewfox]

Yes tomorrows possible snowfall is a very good example of how it can suddenly change. I've only just looked at the NAE and the 18Z is completely different to the 12Z in that the precip intensifies around the W Midlands. This is the complete opposite to what I thought would happen as I expected the front to weaken as it backed W.

Its what most expected to happen including the BBC, a fizzle out event as it back west

Whats caused the change ?

[cooling climate]

Latest fax t84 shows better heights over Scandinavia and more suppressed track of the

lows across the UK, even more so than their own model.

I could suggest the mid and long term trending milder from the south or southwest but

that would be wrong, the outlook is cold with much colder weather likely as we go through

next week and into the first week of March.

[TEITS]

.

I could suggest the mid and long term trending milder from the south or southwest but

that would be wrong, the outlook is cold with much colder weather likely as we go through

next week and into the first week of March.

If we look at the ensembles then there is a definite trend of it turning milder. However I don't think it will be as straight forward as the ensembles suggest and these could be wrong. Lets remember when the mild S,lys were forecast not one ECM member out of 50 suggested anything cold and yet they were all wrong!!

The UKMO +120 is a good example of how the pattern is pushed further S as it enters the reliable timeframe.

http://www.wetterzentrale.de/pics/Rukm1201.gif

If we cast our minds back to the +168 charts a few days ago then that LP was supposed to be further N bringing mild SW,lys. However as the track is further S this isn't the case. What we are seeing in the model output is the trend outside of +144 to bring a return of mild SW,lys. However the reality is once this enters the reliable timeframe the LP systems are further S maintaining the cold pattern.

At the moment the UKMO seems the most consistent and this I feel is the model to watch not the ensembles.

I will add that this winter has been a great learning experience. What I have learn't is when we have a -AO and a S,ly tracking jet the models are too progressive in tracking LPs too far NE. Seems within the programming of these models is a bias to bring a return of our default weather pattern.

Edited February 17, 2010 by THE EYE IN THE SKY

[cuckoo]

The original post was meant to be a joke using the Greek letter as a pun on FI- not meant seriously!

Ah! now i get it. I read the letter as theta not phi.

[cooling climate]

If we look at the ensembles then there is a definite trend of it turning milder. However I don't think it will be as straight forward as the ensembles suggest and these could be wrong. Lets remember when the mild S,lys were forecast not one ECM member out of 50 suggested anything cold and yet they were all wrong!!

The UKMO +120 is a good example of how the pattern is pushed further S as it enters the reliable timeframe.

http://www.wetterzen...cs/Rukm1201.gif

If we cast our minds back to the +168 charts a few days ago then that LP was supposed to be further N bringing mild SW,lys. However as the track is further S this isn't the case. What we are seeing in the model output is the trend outside of +144 to bring a return of mild SW,lys. However the reality is once this enters the reliable timeframe the LP systems are further S maintaining the cold pattern.

At the moment the UKMO seems the most consistent and this I feel is the model to watch not the ensembles.

I will add that this winter has been a great learning experience. What I have learn't is when we have a -AO and a S,ly tracking jet the models are too progressive in tracking LPs too far NE. Seems within the programming of these models is a bias to bring a return of our default weather pattern.

Better heights to the north and east on the GFS 06z run with continued suppression and

eastward movement of low pressure systems. The GFS operational run loses its way past

t144 trying to build heights up from the south which is unlikely as GP said due to the strong

jet stream.

Both UKMO and the GFS now showing better heights to the north, northeast so hopefully the

ECM will come on board soon.

[kev238]

I said yesterday we could be sleep walking into some snow events-- the one for today seems like it has not really got going yet- the one for tomorrow is now being mentioned on the BBC and is on the Met Office site as an amber warning. I like it when these turn amber as they are indicative of increasing confidence in a situation. As it happens, I am travelling from Nottingham to Gloucester tomorrow afternoon so it could be a very interesting journey.

The models have a very Feb 09 look to them with similar areas perhaps being hit by snow events. Infact, this could be the best chance of snow of the winter for the area of Nottingham, Derby, Leicester- amounts for us so far have been quite small compared to many other areas. Equally- it could be too marginal or the precipitation could miss all together! The weather will keep us on our toes in the next week or so!

[kold weather]

Interestingly we don't want the models to trend any further south, because any further south and the set of LPs between 120-168hrs will miss us totally...and there is IMO the correct solution now being picked up by the models for a somewhatr milder flow to come in once that set of LPs coming through moves out though I do think thats a brief affair rather then sustained...

Still the models this morning open up a real chance for a couple of snow events, with even a real threat of a snow event for the south looking at the models BUT they must not trend any more south otherwise we could end up with this cold spell going out with a whimper...

[full_frontal_occlusion]

The original post was meant to be a joke using the Greek letter as a pun on FI- not meant seriously!

Doh!....

[BrickFielder]

I thought I would take a quick look at tomorrows potential snow, but before I get there let us just take a look at todays charts.Notice that surface temperatures are around 8 degrees apart from a small area around wales. Notice also that the zero degree isotherm is up at around 300m or above. This is why I think today will be margin for snow at best.

Now when we go into tomorrow and we look at comparable charts we can see a lower zero degree isotherm and surface temperatures which are somewhat lower. Notice how just to the north of the precipitation temperatures are too high and the zero degree isotherm above where we woiuld like it, but under the precipitation conditions are condusive to snow.

What I think we might see is a band of rain turning to snow, which will most likely mean snow will take a time to settle.

Drifting snow should not be a major problem due to light winds ,but I notice several convergence areas which if low level lapse rates were steeper I would be thinking thunderstorms. It is worth mentioning the steep mid level lapse rates which I think will make for quite heavy precipitation with a risk of hail. It is the prospect of hail which I think might be the unknown quantity, apart from that I think it will be a bit marginal whether there is snow. I personally am inclined to think some areas will with an area from bristol through to lincolnshire currently the prime areas.

[Tom D]

BrickFielder - it may be marginal, but those charts suggest Leicester to be 7/8c by this afternoon, which it most certainly is not. It's currently snowing.

Interesting temperature chart for tomorrow however; why is there a slice that is much colder and elsewhere?

[mickpips]

Hi Brickfielder, I find it interesting you state surface temps are around 6 degree's. Even your first chart says so yet the current temp here in Leeds is 0.7oC and looking across the Meto site its not above 2 degree's anywhere in Yorkshire.

Not a model discussion post but if the ppn makes it this far north then i would expect snow even to lower levels later today along with tomorrow

[Cheese Rice]

I thought I would take a quick look at tomorrows potential snow, but before I get there let us just take a look at todays charts.Notice that surface temperatures are around 8 degrees apart from a small area around wales. Notice also that the zero degree isotherm is up at around 300m or above. This is why I think today will be margin for snow at best.

While I do agree it will be marginal for all areas I do question the GFS temperature charts, it is currently -0.8c and cloudy, which is far of the predicted 6.c. Many have said before the GFS temperature charts are usually wrong, so are they better than most think?

[Buzzit]

I have stated this a few times in the last month and I will state it again

With the current synoptics and variability in model output the NMM and NAE are not worth the computer processing time they take up.

For instance compare the current NMM for 2pm today with actual temperatures - generally the NMM is a minimum 3c higher than actuals across the UK. Precipitation is also slightly misplaced, which over time will also have a considerable negative impact vs actual evolution

Brickfielder : I think you have given a very good reasoned analysis of the NMM data for tomorrow as it exists - However I think the actual will be entirely different due to the frankly quite poor nature of the data

Edited February 17, 2010 by Buzzit

[MKN]

Hi Brickfielder, I find it interesting you state surface temps are around 6 degree's. Even your first chart says so yet the current temp here in Leeds is 0.7oC and looking across the Meto site its not above 2 degree's anywhere in Yorkshire.

Not a model discussion post but if the ppn makes it this far north then i would expect snow even to lower levels later today along with tomorrow

These are the air temps for ground level today according to the 6z gfs not what was show by brickfielder which was the actual ground temp. Correct me if im wrong but i think thats why they are higher.

[johnholmes]

maybe the chart below will help give an idea on where the precip is likely to track.

Then apply the usual checks for your area if you look like being affected.

check the radar for intensity-slight and it may well stay as rain-mod or higher then if the following are ok then prob snow

zero isotherm

wet bulb

dewpoint

dry bulb

850mb temp

1000-850mb thickness

all found in the Extra charts and the way to use them in the NW Guides

good luck some of you.

be a good idea if I added the chart!

maybe the chart below will help give an idea on where the precip is likely to track.

Then apply the usual checks for your area if you look like being affected.

check the radar for intensity-slight and it may well stay as rain-mod or higher then if the following are ok then prob snow

zero isotherm

wet bulb

dewpoint

dry bulb

850mb temp

1000-850mb thickness

all found in the Extra charts and the way to use them in the NW Guides

good luck some of you.

be a good idea if I added the chart!

so here it is

looking at the NAE and GFS 06Z issue for today then its the NAE which seems to have a better 'handle' on the area where snow is most likely.

For tomorrow then they both suggest a 'centre-line' for the snow area as Bristol Channel to the Wash.

Edited February 17, 2010 by johnholmes

[BrickFielder]

These are the air temps for ground level today according to the 6z gfs not what was show by brickfielder which was the actual ground temp. Correct me if im wrong but i think thats why they are higher.

well spotted. Actaul temps are.

[Craigers]

well spotted. Actaul temps are.

amazing how the temp is 0.4c and its 3pm! just shows you now casting is best bet along with the NMM model!

just to edit ive noticed NMM models are out of place so deffo weird today!

Edited February 17, 2010 by Craigers

[kold weather]

12z GFS this time shunts Saturdays possible event WELL south and makes it a *south coast* event, so clearly some big shifts are happening in the models, for any snow event FI is probably as close as 24hrs right now!