model discussion Model output discussion 11th July onwards

[damianslaw]

1 hour ago, Cambrian said:

ECM 12z NH view for day 7 and the Arctic Gorilla getting in some early ice hockey practice I see. Modest Scandinavian / Siberian highs linking up. 

168h

The mid-latitude jet in bits, no sign of any low setting up near Greenland any time soon (quite the opposite in fact).

Out to day 9, Greenland high developing, trying to link up with the “Sciberian”, a block from Greenland through Scandinavia to Siberia. 
216h

Other than the nascent PV wandering around the pole, the meandering mid-Atlantic low in our general locale (will it, won’t it?), the Aleutian low and the one developing off the far east of Russia are the only other pieces of cyclonic action in the whole of the Northern Hemisphere, the latter in the Sea of Okhotsk being the only one undergoing cyclogenesis at that point. 

It’s just so quiet! And seemingly for days on end. 

A northern hemisphere more akin to a PV smashed by a SSW.. rather than one entering Autumn proper. What has happened to the energy in the Atlantic? I wonder if the uniform warm SSTs over the Atlantic as a whole are having an effect?

Edited September 20, 2021 by damianslaw

[Methuselah]

Well, not a bad GFS 00Z to start the day:

    

    

    

Anywho, there's plenty of time for all the northern blocks to dissipate before winter arrives!

And, here's the GEFS 18Z ensemble for Aberdeen:

    

[mb018538]

Let's have a look at what the 00z runs are telling us then...



A very marked temperature contrast showing on the 00z output. In the S/E especially, summery like weather continues for the next 5 days, with temperatures up to 23/24c at times, and staying fairly warm. At the same time, we're seeing afternoon temperatures perhaps stuck at 9c in the far NW of Scotland! Temps gradually increasing the further S and E one heads.

The dry weather looks like perhaps hanging on until Monday now for the majority (far NW Scotland W Ireland excluded), with the odd spot of rain for some before a more active cold front moves through for a good soaking, before the warmer air finally leaves the SE and we see temps drop back into the 14-18c range:

[sheikhy]

Very dry output continues this morning!!!that atlantic has been in sluggish mode for so long!!when is it gona kick into gear!!!

[Evening thunder]

Something odd I've noticed recently, the GFS seems to have incorrect sea temperature data around the UK.

It seems to use the same data as on these charts, up to 6C below average in the channel:

Whereas in reality, SST's are above average around the UK:



This seems to manifest in the GFS showing air temperatures too low over the sea and windward coasts. 3pm today GFS vs ECM:


tomorrow night:



If this is incorporated back into the model, surely this would affect the forward accuracy of the GFS?

Edited September 21, 2021 by Evening thunder

[mb018538]

Another chart showing how the models have really struggled with September so far:

Last Thursday's 500mb update had a trough dominating the UK for this week. The second chart is from last Monday. So we went from settled weather likely on Monday, to unsettled weather being favoured by Thursday.



Last night's update? Back to a ridge again!

[Eagle Eye]

I shall be doing a test on how on average models agree(percentage) and how close those models are to the eventual outcome on those dates. 

Having chosen 2 random days for part 1bto do this test beggining with the first crossover day May 1st, this is only part 1 of my test. 

May 1st

GEM VS GFS VS ECM

day 3

Model agreement: Not much difference at all between GEM and GFS except the strength of the low positioned over northern Europe and Russia, this low had been strengthened sightly towards the centre of it by the GFS, both models and the ECM are relatively similar as well. 

Model agreement: 95%

Eventual outcome : 95%

 

Can only get it at midnight but its close enough model agreement in line with outcome : 95%

GFS: 97%

ECM : 97%

GEM : 91%

Day 5

GFS and GEM weaken the low North of England more than the ECM, differences to the feeding through if the high towards Europe as well. 

Model agreenent: 90%

Outcome :

Model agreement to outcome : 85% ,there is a lot different to the Models in terms of that low's positioning but in model terms it's pretty accurate and so is that high. 

GEM: 87%

GFS: 85%

ECM: 83%

Day 7

Model agreement: 80%

Still a lot the same right now, the centre of that low is different on the Models by quite a big margin and that East Pacific low is different as well. 

Outcome : 

There is a lot different with the strength of the low here not only that but the depth of the East Pacific low across north America

Model outcome : 70%

Day 9

Model agreement : There is a lot different now but the principle is still the same 55%

Compared to outcome :

There is a big different to the Models here, including the strength of the high going up in Europe and the strength of the Arctic low's, thee incursion of the low across Greenland has definitely retreated. 

50%

GFS: 55%

ECM: 48%

GEM: 47%

2nd day run

15th May

Model agreement : 90%

Outcome: 85%

17th May

Model agreement : 85%

Outcome : 83%

19th May

Model agreement : 80%

Outcome : 70%

21st May

Model agreement : 55%

Outcome : 45%

Thats it for now, part 1 done

[minus10]

So for the first day of October the ecm0z brings a not insignificant HP to the NE stretching into Greenland with trough over us and to SE and easterly wind potential. 

I know its day 10 but ecm has been on this signal before. Prior to that a nice late summery spell for many. 

Certainly not your expected Autumn fayre and nothing is certain it seems presently. Enjoy the next few days. 

[Frosty.]

Where did all these posts come from? …anyway, looking at the 0z output, the Atlantic is still floundering and the ECM looks very disrupted to say the least, these are not typical autumn charts!  

 

[SLEETY]

Models all over the place at the moment, supposed to be unsettled by Thursday, but high pressure not pulling away now, looks like the 10 day trend forecast from last Wednesday was actually correct in the end! 

[Cambrian]

0z ECM and UKMO ops - Good agreement at 144h for the building of an extensive belt of high pressure from Scandinavia through northern Russia.

 

Doesn’t look to be moving away any time soon, maintaining a strong influence for several days to come. Intensifying more over Scandinavia by day 8, and still there ridging to Iceland at day 10, as low pressure nestles in on its southern flank over the UK and Europe, bringing us into an easterly. 

 

All this descending air over Eurasia cools the T850 to below normal for an  extensive area covering most of Russia and Europe, including the UK and Ireland as shown by the 850 hPa anomaly chart for 240h. 

0z ECM 240h 850 hPa temps, and the anomaly chart

 

Some relatively cold uppers over Eastern Europe (-5 in the mix) would be heading our way by then. So if the block does remain in place, some tentative signs of the potential for the set-up to lead to some colder than average temperatures over Europe heading into October, particularly as the daytime warming wanes. 

[carinthian]

4 hours ago, jon snow said:

Where did all these posts come from? …anyway, looking at the 0z output, the Atlantic is still floundering and the ECM looks very disrupted to say the least, these are not typical autumn charts!  

 

Aye, we have seen this pattern  for the Atlantic trough to disrupt as you say for many months now. Mid latitude Quasi Resonant Amplification ( QRA ) as some of our European experts report could hold sway for a lot of snowfall in Western Europe/UK this coming winter, especially the Alps as they mention ( if things stay as they are ) ! This is very unusual.

C

[mb018538]

5 hours ago, Evening thunder said:

Something odd I've noticed recently, the GFS seems to have incorrect sea temperature data around the UK.

It seems to use the same data as on these charts, up to 6C below average in the channel:

Whereas in reality, SST's are above average around the UK:



This seems to manifest in the GFS showing air temperatures too low over the sea and windward coasts. 3pm today GFS vs ECM:


tomorrow night:



If this is incorporated back into the model, surely this would affect the forward accuracy of the GFS?

I wouldn't really trust the GFS on a lot of things!



At the moment it reckons nearly the whole country should be covered in 100% cloud cover. That chart would look terrible at first glance. 

In reality there's high cloud floating about with a bit on infill down in the S/E corner, but otherwise it's a fine and nice day. Nothing like what that suggests. 

[damianslaw]

No comment on UKMO and GFS 12z.. probably because they are tempering the high pressure influence through remainder of the week. Atlantic trough slightly further east which reduces heights instead quite a moist SW airstream likely into western parts Saturday with the trough pushing in during Sunday. SE parts should hold onto fine warmer weather..  but the clock is ticking, and there seems no escape from the Atlantic by Monday for all. Can still change.. but those heights to the east are not near enough to kick the trough aside. 

Edited September 21, 2021 by damianslaw

[carinthian]

4 hours ago, carinthian said:

Aye, we have seen this pattern  for the Atlantic trough to disrupt as you say for many months now. Mid latitude Quasi Resonant Amplification ( QRA ) as some of our European experts report could hold sway for a lot of snowfall in Western Europe/UK this coming winter, especially the Alps as they mention ( if things stay as they are ) ! This is very unusual.

C

Amazing there we go again with latest from GFS.

C

[sheikhy]

9 minutes ago, damianslaw said:

No comment on UKMO and GFS 12z.. probably because they are tempering the high pressure influence through remainder of the week. Atlantic trough slightly further east which reduces heights instead quite a moist SW airstream likely into western parts Saturday with the trough pushing in during Sunday. SE parts should hold onto fine warmer weather..  but the clock is ticking, and there seems no escape from the Atlantic by Monday for all. Can still change.. but those heights to the east are not near enough to kick the trough aside. 

Really!!maybe the gfs but the ukmo i have just viewed has heights further west and a stronger scandi high!!infact its waaaaay stronger than gfs for the same time?!!!!gfs looks pretty darn dry as well!!couple of bands of rain that breeze through but nothing substantial!

[damianslaw]

13 minutes ago, sheikhy said:

Really!!maybe the gfs but the ukmo i have just viewed has heights further west and a stronger scandi high!!infact its waaaaay stronger than gfs for the same time?!!!!gfs looks pretty darn dry as well!!couple of bands of rain that breeze through but nothing substantial!

The trough is elongated NW-SE destined to settle over UK as shown on UKMO regardless of heights to the east being further west at the same time heights build to our west. Trough has nowhere to go but over UK. 

Edited September 21, 2021 by damianslaw

[sheikhy]

3 minutes ago, damianslaw said:

The trough is elongated NW-SE destined to settle over UK as shown on UKMO regardless of heights to the east being further west at the same time heights build to our west. Trough has nowhere to go but over UK. 

My worry is that we get a middle ground and trough settles over the uk!!Lets hope we get either a stronger push of heights from the east or the atlantic going just about enough east to stop the fronts from stalling over us!!

[minus10]

Interesting this feature from gfs 12z on 2nd October. Looks like a sudden appearance of a tropical storm just off coast of East Anglia.

It quickly dissappears and will be gone next run. Dont often see that...

After a few days of more unsettled weather (particularly for nw) gfs brings back HP again over us...

 

...and warmer temps....JFF....

[Methuselah]

There'll be no moans from me, should anything like this verify!

    

[damianslaw]

12 minutes ago, Ed Stone said:

There'll be no moans from me, should anything like this verify!

    

Mmm 7th October.. all conjecture..  the ensembles suggest trough settling over UK and may take time to be dislodged. Not sure the developing west Russia high will quickly be replaced by a trough to allow heights to settle in behind. 

[heath]

Is it just me or is this a bit unusual for late September? Any comments from the knowledgeable? 

[Tim Bland]

19 minutes ago, heath said:

Is it just me or is this a bit unusual for late September? Any comments from the knowledgeable? 

https://www.meteociel.fr/modeles/archives/archives.php

you tell us? Above is a link to all charts back to the 1800s. I don’ think it’s that unusual for September to be honest. The PV usually takes until end of Oct or early November to get going. I remember many year us coldies getting excited in Autumn by all the blocking and lack of PV but by November it appears and it’s business as usual lol 

[heath]

11 minutes ago, Tim Bland said:

https://www.meteociel.fr/modeles/archives/archives.php

you tell us? Above is a link to all charts back to the 1800s. I don’ think it’s that unusual for September to be honest. The PV usually takes until end of Oct or early November to get going. I remember many year us coldies getting excited in Autumn by all the blocking and lack of PV but by November it appears and it’s business as usual lol 

Thanks. I was thinking of the lack of cyclonic activity in the NH rather than the PV. Sampling history randomly it looks a bit unusual. 

[Eagle Eye]

On 21/09/2021 at 08:30, Eagle Eye said:

I shall be doing a test on how on average models agree(percentage) and how close those models are to the eventual outcome on those dates. 

Having chosen 2 random days for part 1bto do this test beggining with the first crossover day May 1st, this is only part 1 of my test. 

May 1st

GEM VS GFS VS ECM

day 3

Model agreement: Not much difference at all between GEM and GFS except the strength of the low positioned over northern Europe and Russia, this low had been strengthened sightly towards the centre of it by the GFS, both models and the ECM are relatively similar as well. 

Model agreement: 95%

Eventual outcome : 95%

 

Can only get it at midnight but its close enough model agreement in line with outcome : 95%

GFS: 97%

ECM : 97%

GEM : 91%

Day 5

GFS and GEM weaken the low North of England more than the ECM, differences to the feeding through if the high towards Europe as well. 

Model agreenent: 90%

Outcome :

Model agreement to outcome : 85% ,there is a lot different to the Models in terms of that low's positioning but in model terms it's pretty accurate and so is that high. 

GEM: 87%

GFS: 85%

ECM: 83%

Day 7

Model agreement: 80%

Still a lot the same right now, the centre of that low is different on the Models by quite a big margin and that East Pacific low is different as well. 

Outcome : 

There is a lot different with the strength of the low here not only that but the depth of the East Pacific low across north America

Model outcome : 70%

Day 9

Model agreement : There is a lot different now but the principle is still the same 55%

Compared to outcome :

There is a big different to the Models here, including the strength of the high going up in Europe and the strength of the Arctic low's, thee incursion of the low across Greenland has definitely retreated. 

50%

GFS: 55%

ECM: 48%

GEM: 47%

2nd day run

15th May

Model agreement : 90%

Outcome: 85%

17th May

Model agreement : 85%

Outcome : 83%

19th May

Model agreement : 80%

Outcome : 70%

21st May

Model agreement : 55%

Outcome : 45%

Thats it for now, part 1 done

 

 

Part 2 of my self research

Same models in same order as before

8th June start date

10th June

Model agreement : 90%

Outcome :  85%

12th June

Model agreement : 80%

Outcome : 73%

14th June

Model agreement : 65%

Outcome : 52%

16th June

Model agreement : 45%

Outcome : 37%

Run 20th June

22nd June

Model agreement : 95%

Outcome : 88%

24th June

Model agreement : 79%

Outcome : 72%

26th June

Model agreement : 52%

Outcome : 47%

 

28th June :

Model agreement : 42%

 Outcome : 40%

Total percentage average at 

48 hours

Model agreement : 92-94% on average

Outcome : 88-90%

96 hours

Model agreement : 80-87%

Outcome : 72-80%

144 hours

Model agreement : 58-72%

Outcome : 47-57%

192 hours

Model agreement : 44-55%

Outcome : 40-50%

As far as I can see June seems to have more differences within it compared to May, perhaps the positoning of the Arctic low's which seems to be the main difference is hard to predict further into Summer. The high's positioning is also hard to predict especially around 144 hours compared to 96 hours, after 4-7 or so days the diffence in percentage slow's down. Model agreement is definitly much more than the outcome however interestingly towards day 7 the gap closes down.

 

GEM closer to now isn't as trustful as GFS and ECM, GEM is varying a lot more compared to the other models in terms of how close the model is at certain hours. ECM is an underrated model in terms of being up there with as close as GFS in the outcomes. GFS does prevail most of the time as the turtle style model however in June it does seem to be less of a model for high's over the UK.

That's all for now