General Model Output Discussion

[MR EXTREMES]

Very amusing - the JMA has the Lochness monster about to eat Iceland (serves them right for losing all our councils money).

The actual output on the JMA look pretty good though with no early breakdonw on the Easterly next week.

i dunno the jma not my ideal model i think this cold spell might pop back but its very unlikely to be a major event,

im starting to feel that winter maybe over.

after all philip eden says that early start to winter normally finishes early.

i think the models are leaning to much on the side of cold because its been this way for so long.:unsure:

[VillagePlank]

i think the models are leaning to much on the side of cold because its been this way for so long.:unsure:

Yes, I've long wondered about whether these models are defacto numerical models, or, they have a bit of statistics built in at a higher level. Probably a good idea, since a potential basis of all weather forecasting, with a >50% success rate would be simply to say the weather today will be the same as it was yesterday.

[Eugene]

Very amusing - the JMA has the Lochness monster about to eat Iceland (serves them right for losing all our councils money).

The actual output on the JMA look pretty good though with no early breakdonw on the Easterly next week.

Not sure whether you know this or not but that is last nights run, tonights JMA 12Z isn't out yet.

[mushymanrob]

after all philip eden says that early start to winter normally finishes early.

what, like 1978-9? :unsure:

[Methuselah]

With respect to everyone who makes a forecast on this forum (well, almost everyone) I wonder if that's actually true? I've been around here a long time, and I quickly lost count of the times something's been promised and has failed to be delivered in any way, shape, or form. Yes, we all remember the time x said it would be cold in a fortnight's time, and it was cold in a fortnight's time. But equally, there are hundreds upon hundreds of occasions when the opposite is true. I also think that an unfair emphasis is placed on their forecasting for one particular weather-type; snow. It's not the Met Office's remit to ensure their forecasting doesn't upset those of us for whom snow is the most exicting things in their lives, and I imagine they're no better or worse at forecasting snow than any other weather type.

I'm not against anyone on here, and I'm no great defender of the Met Office, but I really don't get the criticism they get, because I bet you 99/100, the mid-range forecasting is a lot more accurate than anyone's on here.

Hear! Hear! OON... :winky:

[492]

i dunno the jma not my ideal model i think this cold spell might pop back but its very unlikely to be a major event,

im starting to feel that winter maybe over.

after all philip eden says that early start to winter normally finishes early.

i think the models are leaning to much on the side of cold because its been this way for so long.

It has been noticeable that in recent winters as a cold spell comes to an end the models tends to develop renewed cold spells that do not usually materialize. However this year unlike many recent winters there is a good snow cover over virtually all of northern europe and there has been sustained high pressure over northern Russia and northern Siberia with extensive intense cold. So the tendency of the models to bring back the cold from the east next week may well have more substance this year.

[full_frontal_occlusion]

Yes, I've long wondered about whether these models are defacto numerical models, or, they have a bit of statistics built in at a higher level. Probably a good idea, since a potential basis of all weather forecasting, with a >50% success rate would be simply to say the weather today will be the same as it was yesterday.

Latest generation models use ensembles as a way of representing the probability density function of errors for each run. In simple terms 'spread'. This depicts spatial (3D) distributions by introducing a random element - the so called Monte-Carlo type 'seeding' which itself introduces error.

Next generation NWP use 4D-Kalman filtering (a variant on the filters used extensively in missile guidance algorithms).

Ensemble Kalman filters EnKF uses predictive algorithms to plot spatial trajectory as a function of time. i.e. Spatial error as a funtion of time is computed from actual synoptics vs the previous predicted spatial distribution. This generates a spatial compensation function. Feedback loops try to reduce the error to zero.

As you may imagine, this type of algorithm imposes very significant demands on processing power!

ffO.

[MR EXTREMES]

It has been noticeable that in recent winters as a cold spell comes to an end the models tends to develop renewed cold spells that do not usually materialize. However this year unlike many recent winters there is a good snow cover over virtually all of northern europe and there has been sustained high pressure over northern Russia and northern Siberia with extensive intense cold. So the tendency of the models to bring back the cold from the east next week may well have more substance this year.

we must also remember this is prog to be a short lived event maybe even 24 hrs.

the teleconnections from what i can see dont support a sustained easterly.

[bluearmy]

not sure why the meto 6/15 dayer got so much 'bad press' earlier. given that a large clump of the ens seemed to take temps above 5c on the london meteogram, i would have thought that the ens mean would have shown us in a stand off between the scandi block and atlantic trough. i would have thought the meto would work from this mean as their starting point for 10 to 14 days. we get a bit carried away with whatever the op (and control) are showing beyond a week. i dont think the pro's take a lot of notice of op runs at that range and work extensively from the ens means. thats the impression i get reading the NOAA output.

if the ens begin to trend towards the op, then expect the 15 dayer to reflect that output, but not before.

[snowingtequila]

we must also remember this is prog to be a short lived event maybe even 24 hrs.

the teleconnections from what i can see dont support a sustained easterly.

the models only really progged it from last night so it is hard to say how long it will last at the moment. I am not so sure about those teleconnections maybe you could elaborate

[PB22]

ECM 12Z isn't off to the best of starts, as it makes more of the Atlantic trough coming through this weekend. Will be interesting to see where it heads beyond T+120.

T+96:

T+120:

ECM T+144:

Shows that the outcome next week is far from certain.

Wonder why its gone quiet in here?

ECM says NO to the easterly:

[cold snap]

ECM 12Z isn't off to the best of starts, as it makes more of the Atlantic trough coming through this weekend. Will be interesting to see where it heads beyond T+120.

T+96: http://www.meteociel...CM1-96.GIF?13-0

T+120: http://www.meteociel...M1-120.GIF?13-0

ECM T+144: http://www.meteociel...M1-144.GIF?13-0

Shows that the outcome next week is far from certain.

Wonder why its gone quiet in here?

ECM says NO to the easterly:

http://www.meteociel...M1-168.GIF?13-0

http://www.meteociel...1-216.GIF?13-12

Paul is it not a case of the evolution just being delayed a little bit now as above?

Edited January 13, 2010 by cold snap

[weather eater]

ECM 12Z isn't off to the best of starts, as it makes more of the Atlantic trough coming through this weekend. Will be interesting to see where it heads beyond T+120.

T+96:

T+120:

ECM T+144:

Shows that the outcome next week is far from certain.

Wonder why its gone quiet in here?

ECM says NO to the easterly:

Yep looking on Meteociel the ECM looks fairly clear in not developing the easterly, but it could be a blip we have seen this a few times from the ECM lately.

[bluearmy]

ECM 12Z isn't off to the best of starts, as it makes more of the Atlantic trough coming through this weekend. Will be interesting to see where it heads beyond T+120.

T+96:

T+120:

ECM T+144:

Shows that the outcome next week is far from certain.

Wonder why its gone quiet in here?

ECM says NO to the easterly:

the same as the 12z GEM. however, its only one run and its a different enough development for it to head east/norhteast rather than se as on a few GFS runs to mean that we shouldnt take too much from it until it repeats. interesting to see the way that it looks very simple on ecm - if the block is held back by the shortwave, the atlantic simply heads on in, if it doesnt as per the 00z run the opposite is true. i am rather dubious that this devloping feature can hold back what appeared to be such a strong irresistable force. we'll see .................

[ChartViewer]

ECM 12Z isn't off to the best of starts, as it makes more of the Atlantic trough coming through this weekend. Will be interesting to see where it heads beyond T+120.

T+96:

T+120:

ECM T+144:

Shows that the outcome next week is far from certain.

Wonder why its gone quiet in here?

ECM says NO to the easterly:

Paul,

I am not as knowledgable as you but I would suggest that with such a HUGE Russian/Siberian/Mongolian

High synoptics showing it 100 miles West or East shift the synoptic pattern hugely....FI starts

at T72 at present because of the above High Pressure cell.

Regards

CV

[PB22]

http://www.meteociel...1-216.GIF?13-12

Paul is it not a case of the evolution just being delayed a little bit now as above?

Perhaps, but compared to the 00Z run, the evolution on this 12z run is seriously delayed if it does end up in a easterly later. ECM basically brings back the Atlantic influence this weekend, and keeps us under the Atlantic influence right out to T+168 and is a poor run for coldies. To compare, UKMO and GFS have us in a easterly by T+144.

Paul,

I am not as knowledgable as you but I would suggest that with such a HUGE Russian/Siberian/Mongolian

High synoptics showing it 100 miles West or East shift the synoptic pattern hugely....FI starts

at T72 at present because of the above High Pressure cell.

Regards

CV

Hi CV,

I would tend to agree with your sentiments...after all, we don't often see a 1060mb Siberian High on the charts!

Looks like ECM is going to have another go with the high at T+192:

Nope, ECM not interested tonight:

T+216:

Will be interesting to see the ensembles later.

[Guest]

Perhaps, but compared to the 00Z run, the evolution on this 12z run is seriously delayed if it does end up in a easterly later. ECM basically brings back the Atlantic influence this weekend, and keeps us under the Atlantic influence right out to T+168 and is a poor run for coldies. To compare, UKMO and GFS have us in a easterly by T+144.

I can't quite believe I am going to post this, however, here goes:

What would John Holmes say about comparing the 00z to the 12z?? Haven't we all been here before? Compare like with like, 12z to 12z.

And Paul, I do appreciate your knowledge in these things is far superior to mine, but please stick to the rules which the know-nothings like me should stick to :lol: :lol:

[PB22]

I can't quite believe I am going to post this, however, here goes:

What would John Holmes say about comparing the 00z to the 12z?? Haven't we all been here before? Compare like with like, 12z to 12z.

And Paul, I do appreciate your knowledge in these things is far superior to mine, but please stick to the rules which the know-nothings like me should stick to :lol: :lol:

Well that's me told! Sorry!

OK, to compare to last night's ECM, then tonight's has definitely backed away from the easterly. Last night's ECM had us under an easterly by T+168, tonight's T+144 hasn't.

[Guest North Sea Snow Convection]

The transition on the ECM from t144 to t168 is very dubious. There will be much colder members in the ensembles than the operational.

Although I have suggested that the energy to the north west needed watching yesterday, this output won't verify as it has overcooked the troughing to the north west.

Edited January 13, 2010 by North Sea Snow Convection

[weather eater]

I can't quite believe I am going to post this, however, here goes:

What would John Holmes say about comparing the 00z to the 12z?? Haven't we all been here before? Compare like with like, 12z to 12z.

And Paul, I do appreciate your knowledge in these things is far superior to mine, but please stick to the rules which the know-nothings like me should stick to :) :lol:

Fair enough, the ECM 12z is nothing like last nights, which I believe, had the easterly showing.

[TEITS]

What would John Holmes say about comparing the 00z to the 12z?? Haven't we all been here before? Compare like with like, 12z to 12z.

If you want my opinion i've never agreed with John H about comparing like with like. This has never made any sense to me and never will.

Back to the ECM and whilst the 12Z is disappointing I remain confident about the E,ly. If the 12Z was showing a repeat of Jan 1987 everyone would be urging caution. So why should this be any different!

[johnholmes]

On this image of Sundays UKMO chart

What is the significance of the black line that separates the blue isobars from the orange?

its the Standard Atmospheric Pressure for the whole world which ( ICAO) (International Civil Aviation Organisation) use.

1013.2mb

Edited January 13, 2010 by johnholmes

[Frosty.]

Well that's me told! Sorry!

OK, to compare to last night's ECM, then tonight's has definitely backed away from the easterly. Last night's ECM had us under an easterly by T+168, tonight's T+144 hasn't.

I guess it's all over then BOOHOO :lol:

Even without the E'ly it would be a fairly cool week with some overnight frosts and the ukmo 12z looks promising at T+144 hours, also the gfs 12z still looks good to go for an easterly next week but if it's over, we have had a nice long freezing spell and maybe it's time for the wildlife to enjoy more temperate conditions.

[weather eater]

I guess it's all over then BOOHOO :lol:

Even without the E'ly it would be a fairly cool week with some overnight frosts and the ukmo 12z looks promising at T+144 hours, also the gfs 12z still looks good to go for an easterly next week but if it's over, we have had a nice long freezing spell and maybe it's time for the wildlife to enjoy more temperate conditions.

Or it could be a blip, as I say the ECM has done this a few times this winter. Identify a trend, go off the idea for a run or two, and then come back to the initial trend, one run tells us nothing.

[The PIT]

ECM having none of the GFS return of the easterly.

So it's either going to go back to average and stay that way or back to average then back to cold.