Model output discussion - is the beast awakening?

[Updated_Weather]

Just now, MattStoke said:

Why do people pay more attention to the older version of the GFS than the newer version? Surely the newer version verifies better. A complete waste of money otherwise!

Yep agreed.. must be a reason why it’s being upgraded maybe we are seeing why!

[Chertseystreamer79]

4 minutes ago, Djdazzle said:

Although the 6z isn't great, at least it hasn't continued the trend seen on the previous two runs.

The last 4 pages in a nutshell. Thankyou. That's all we needed to see for now.

[TSNWK]

6 minutes ago, MattStoke said:

Why do people pay more attention to the older version of the GFS than the newer version? Surely the newer version verifies better. A complete waste of money otherwise!

We had a debate along these lines yesterday. Personally I struggle with accepting the para for two reasons.

We don't get to see where it sits within the para ensemble member.. we could be seeing tweeted runs to support testing 

Secondly.  It's parr and test.  If it was ready then why keep the legacy GFS going. 

Edited February 2, 2021 by TSNWK

[Radiating Dendrite]

5 minutes ago, BLIZZARDMAN said:

I present perturbation 1... JFF of course

Well that is a small improvement to start - I do not think any of the ensemble runs got that cold on the 0z.

[Hodgheg]

5 minutes ago, MattStoke said:

Why do people pay more attention to the older version of the GFS than the newer version? Surely the newer version verifies better. A complete waste of money otherwise!

Not necessarily, it’s a parallel for a reason. 
 

Looks like the drama continues for the rest of today! 

[IDO]

2 minutes ago, Vadoseflame said:

Yep agreed.. must be a reason why it’s being upgraded maybe we are seeing why!

Even the gfsP has relatively short-lived cold spell and ties in with the ecm mean:

With a touch of spring! So hopefully we do get this 5-7 day cold spell. 

[SqueakheartLW]

Looks like a mixed update for me again after yesterday's 12z updates.

GFS 18z

The less I say about this one the better really and I was hoping for a nice colder average to end the day on. What I ended up with was another downgrade to the cold again to -7.0 which is still -5.3 below the long term mean but was the least cold run of the day

GFS 00z

After the disappointing end to yesterday's GFS runs time for a new day and a fresh start, NOT !!!. The peak of the colder spell on this run is at 126 hours away so getting closer at least.

GFS is continuing to push this easterly ever further northwards so now starting to look like even I am going to end up missing out too if this trend continues. The 850hpa temperature average this time was only -6.5 which is still -4.0 below the long term mean but is the least cold GFS update I have featured so far and is not very promising for us anymore. Maybe the wet snow I saw this morning could end up the only covering of snow I see this entire winter, that's how unlucky I have been in Scunthorpe this winter despite some favourable conditions at times in other areas even a few miles away. The anomaly chart clearly shows that milder area over Europe getting ever closer to SE UK.

Mildest 3 charts from GFS 00z at 126 hours away

With the trend less cold the milder charts are really starting to look milder now

3rd     Member 9     850hpa temp -1.7

2nd     Member 13     850hpa temp -0.3

1st     Member 27     850hpa temp 0.5

Coldest 3 charts from 00z GFS at 126 hours away

These charts are getting less cold now with the general trend of the recent GFS runs

3rd     Member 6     850hpa temp -10.8

2nd     Member 24     850hpa temp -11.2

1st     Member 30     850hpa temp -11.6

Mildest doom and gloom chart from 00z GFS

Member 25     +282 hours     850hpa temp 8.1

Coldest fantasy charts from the 00z GFS

Two runs get the honours here for coldest charts on the 00z GFS

Member 24     +132 hours     850hpa temp -12.5

Member 30     +132 hours     850hpa temp -12.5

GEM 00z

After the disappointment of the GFS 18z and 00z lets see if GEM has fared any better. Today the peak of the cold spell on GEM is at 150 hours away but is shared with two other times of 162 and 168 hours away too but I have used 150 hours away since it is the first one. The peak of the cold averages out at -8.6 which is -6.0 below the long term mean. A slight downgrade on the 12z GEM of 0.5 but at least no where near as bad as the GFS has got

Mildest 3 charts of the 00z GEM at 150 hours away

3rd     Member 15     850hpa temp -5.0

2nd     Member 11     850hpa temp -4.0

1st     Member 13     850hpa temp -3.1

Coldest 3 charts of the 00z GEM at 150 hours away

3rd     Member 3     850hpa temp -11.7

2nd     Member 8     850hpa temp -11.8

1st     Member 1     850hpa temp -12.2

Mildest doom and gloom charts of the 00z GEM

Two charts unfortunately share the shame of having the mildest charts in the 00z GEM

Member 10     +330 hours     850hpa temp 10.2

Member 20     +342 hours     850hpa temp 10.2

Coldest fantasy chart of the 00z GEM

Member 6     +270 hours     850hpa temp -13.5

GFS Extended

Finally time to get onto the only real good news in this update and this is the GFS Extended. The peak of the cold spell on this run is at 156 hours away.

Today the GFS Extended shows a decent easterly average and combined with the 850hpa temperature today of -10.4 which is -7.8 below the long term mean this is a new low which is -1.1 colder than yesterday's GFS Extended. The anomaly chart shows some nice deep blues over the UK. At least the extended run has given me some hope anyway.

Mildest 3 charts of the GFS Extended at 156 hours away

As the operational run doesn't form part of the extended runs then it's mildest run spot doesn't count so the extended top 3 are as follows

3rd     Member 7     850hpa temp -7.7

2nd     Member 3     850hpa temp -4.5

1st     Member 9     850hpa temp -4.4

Coldest 3 charts of the GFS Extended at 156 hours away

3rd     Member 24     850hpa temp -13.4

2nd     Member 20     850hpa temp -13.7

1st     Member 14     850hpa temp -14.7

Mildest doom and gloom charts of the GFS Extended

As with GEM 00z the GFS Extended has 2 charts sharing the shame of mildest runs in the GFS Extended

Member 4     +432 hours     850hpa temp 8.5

Member 8     +486 hours     850hpa temp 8.5

Coldest fantasy chart from the GFS Extended

Member 1     +300 hours     850hpa temp -15.3

Edited February 2, 2021 by SqueakheartLW

[IDO]

The gefs mean still moving away from an easterly for the south and def no mean easterly on the 06z IMBY; up to t180 below, and obviously none after:

 

[Djdazzle]

With so much going on in the 72-120 period, I think anything after that is guesswork!

[Updated_Weather]

Interesting black line Outlier?

[kold weather]

so like 18z GFS, I think its time to take a deeper look at the 06z GFS, particularly at 96hrs:

Introduction:

Each low will effectively perform so kind of fuijiwara effect around each other and will probably eventually become one gyre somewhere close to the south of England. The key location of where that sets up is totally dependent on the constituent lows and where they end up.

GFS 06z:

The GFS has three lows that all interact with each other to beocme one larger gyre which rotates around each other.

1: Limpet low is further north than all other models and more stretched out with a elongated upper trough. This feature starts to get absorbed into the gyre further west pretty quickly and blocks off any cold air from getting further south and keeps the cold air locked away to the north whilst a fair part of the country remains under the low.

2: Iberian low is modelled more realistically this time round but because on No.1, it doesn't really matter.

3: The Atlantic low is actually a little weaker than the likes of ICON 06z, but because of the location of no.1 it manages to absorb and no.3 become the dominant airflow.

The whole lot once connected like they are at 120hrs will pivot around a central point which will increasingly be located close to the low pressure in the Atlantic. This will bring some cold air further south but probably not enough.

00z ECM:

1: Limpet low further south which keeps it away from the atlantic low for much longer.

2: Iberian low rotates up as per GFS but absorbs the Limpet low this time which pushes cold air in from the east.

3: Atlantic low remains disconnected from the UK low this time so ends up further south-east than the GFS has there isn't that tug from the north as No.1 doesn't end up getting dragged westwards.

ECM synoptically is very different with how it handles no.1, which ends up far enough south to start with that it ends up getting absorbed by No.2 instead of no.3 low which allows the cold air to come in from the east. Eventually No.3 DOES absorb it all and become the central gyre but its far enough south to be no issue.

Conclusion:

The limpet low remains utterly key to the evolution. The GFS has actually shifted overnight from the 18z in that the last set of runs the GFS had the low pressure being way further SW which helped to blow up the Iberian low. This time round it gets stuck over Scotland instead instead and rotates so far NW that it gets absorbed into the flow.

So the GFS is NOT being consistent run to run despite the way it might appear on the surface, as its getting to the same conclusion with a variety of means depending on exactly what it does with the limpet low. The ECM and indeed all other models keep the UK low far enough SE that we don't make the connection between No.1 and No.3 which keeps the polar boundary moving further south rather than getting stuck over N.England.

So what do I think is going to happen?

I think its really going to depend alot on exactly where the core of the low pressure No.1 ends up. Basically wherever it ends up sitting  then add another 100-150 miles south for a sagging of the whole complex and that will be pretty much your snow line for 120-144hrs.

The good news is the GFS is really on its own, and ironically is now handling the low no.1 *exactly* the same way the ECM did a couple of days ago, so the GFS has backtracked whilst the ECM has shifted towards the OG GFS idea, work that one out! 

ICON 06z has made a move towards the GFS solution with regards to no.3, and that is somewhat worrying because a stronger No.3 means we have much less leeway with how no.1/2 evolve. The stronger no.3 means its reach is larger and the odds of absorption higher, so we need no.1/2 to effectively make up the difference with a good evolution. Luckily the ICON 06z does appear to behave with those lows, but its a sign that the GFS solution is a possible one, even if I think its perhaps not the more likely one. 

EDIT - The GFS ensembles are not going to be worth a dime in this set-up, and probably same for the ECM ensembles. The set-up is so subtle that they will both likely follow their operational as this is likely down to how they are handling the LP within their separate algorithms/data inputs. 

They will all flip at the same time at some point, either towards the GFS or the GFS towards the ECM, but it will probably happen pretty much in one hit through the ensembles and operationals.

Edited February 2, 2021 by kold weather

[Badgers01]

41 minutes ago, mountain shadow said:

 

There has been no comments in the SE England regional thread for 40 minutes, perhaps this comment is better suited for there?

This is not the SE England model output thread.

It’s perfectly legitimate to discuss what the models are showing and the implication for the weather in a certain region !!

[Waterspout]

25 minutes ago, mountain shadow said:

But we know this and have done for ten years, especially the 6z and 18z.

That's why this GFS is again going to be upgraded shortly.

I say it every year, but if the ECM and UKM gave us free access to the amount of data GFS does, we would rarely even refer to the GFS at all.

This was meant to be tomorrow the 3rd, but the upgrade of the IT infrastructure, which is also why some connectivity issues have happened in recent weeks, is behind schedule. It needs to handle twice the Data for v16. This has now moved to “March 2021”. It’s unfortunate as from tomorrow we wouldn’t have had to look at the current version anymore! It would have been consigned to the dustbin, I guess like the 06z Op! As it stand now we have at least another month of v15!

[North East Blizzard]

11 minutes ago, TSNWK said:

We had a debate along these lines yesterday. Personally I struggle with accepting the para for two reasons.

We don't get to see where it sits within the para ensemble member.. we could be seeing tweeted runs to support testing 

Secondly.  It's parr and test.  If it was ready then why keep the legacy GFS going. 

I think someone said that they wanted the para running for 30 full days and something went wrong half way through, so its had to be started again for the full 30 days. May of dreamt that though

There's really nothing in it between the two tbh

[Badgers01]

1 minute ago, Waterspout said:

This was meant to be tomorrow the 3rd, but the upgrade of the IT infrastructure, which is also why some connectivity issues have happened in recent weeks, is behind schedule. It needs to handle twice the Data for v16. This has now moved to “March 2021”. It’s unfortunate as from tomorrow we wouldn’t have had to look at the current version anymore! It would have been consigned to the dustbin, I guess like the 06z Op! As it stand now we have at least another month of v15!

It’s a good reminder that if the upgrade had gone to plan we wouldn’t even be seeing the the current GFS op is showing ! 

[MATTWOLVES 3]

31 minutes ago, Ali1977 said:

Day 6 and the GEFS mean is better than the Op, but not great and definitely not great for anywhere south of the midlands!! 

Big big 12zs, later

Yes I would agree its gonna be interesting to see the UK and Euro 12s..not so sure the GFS though,which will probably come out with yet another evolution...this time a Euro slug or a  Bartlett....oocchhh drop me points for just mentioning that dreaded insulting B word in here..

[sunnijim]

30 minutes ago, MattStoke said:

Why do people pay more attention to the older version of the GFS than the newer version? Surely the newer version verifies better. A complete waste of money otherwise!

I read yesterday that the Para is going to replace the GFS op in the Autumn of 2021.

The blog I was reading talked of the 'input' being superior.

Which begs the question why don't we all start with the Para now as regards our first port of GFS call?

The GFS op is already regarded as second class,that is my understanding.

[MJB]

Don't shoot me 

I am thinking we will see a change from ECM this evening towards GFS , it followed GFS for the Easterly and caught up , I think it will do the same here . GFS running 4 times a day will always show different scenarios . With us seeing ECM twice a day its a day behind with regards what we see. 

Sadly I think GFS has sniffed something out............just because it's on it's own doesn't mean it's wrong , UK weather law would suggest it's right.

I do hope I am wrong 

[topo]

I believe GFS is way wrong here.. It tries as always to overcomplicate things.. Stay simple as ECM and UKMO

[frosty ground]

24 minutes ago, IDO said:

The gefs mean still moving away from an easterly for the south and def no mean easterly on the 06z IMBY; up to t180 below, and obviously none after:

 

DO you think the mean will be representative of the actual conditions?

[That ECM]

26 minutes ago, IDO said:

The gefs mean still moving away from an easterly for the south and def no mean easterly on the 06z IMBY; up to t180 below, and obviously none after:

 

Meto got you with a max of 1 on Monday with a strong north easterly. 

Edited February 2, 2021 by That ECM

[Buzz]

Just now, MJB said:

Don't shoot me 

I am thinking we will see a change from ECM this evening towards GFS , it followed GFS for the Easterly and caught up , I think it will do the same here . GFS running 4 times a day will always show different scenarios . With us seeing ECM twice a day its a day behind with regards what we see. 

Sadly I think GFS has sniffed something out............just because it's on it's own doesn't mean it's wrong , UK weather law would suggest it's right.

I do hope I am wrong 

Why not temper your abject fears and reduce your stress by looking at the Parallel instead?

[Yarmy]

Why bin the GFS for the Para when the former is verifying better in the 5-6 day range currently?

These stats jump around all the time. A week or so ago, the Para was verifying better (slightly). There's no objective reason to have any more confidence in it than the old one.

[sunnijim]

5 minutes ago, MJB said:

Don't shoot me 

I am thinking we will see a change from ECM this evening towards GFS , it followed GFS for the Easterly and caught up , I think it will do the same here . GFS running 4 times a day will always show different scenarios . With us seeing ECM twice a day its a day behind with regards what we see. 

Sadly I think GFS has sniffed something out............just because it's on it's own doesn't mean it's wrong , UK weather law would suggest it's right.

I do hope I am wrong 

I'm sure all of us have the same nagging fear, a high risk high reward situation like this doesn't leave much 'wriggle room'

I certainly remember the GFS 18z sniffing out the ruination of a freeze a few years back before any other model.

Oh,how we laughed at the silly old 'pub run' at the time.

Come Morning,we were all viewing a 'car crash' across the board!

Edited February 2, 2021 by sunnijim

[kold weather]

7 minutes ago, MJB said:

Don't shoot me 

I am thinking we will see a change from ECM this evening towards GFS , it followed GFS for the Easterly and caught up , I think it will do the same here . GFS running 4 times a day will always show different scenarios . With us seeing ECM twice a day its a day behind with regards what we see. 

Sadly I think GFS has sniffed something out............just because it's on it's own doesn't mean it's wrong , UK weather law would suggest it's right.

I do hope I am wrong 

The thing is MJB as my post above shows the GFS HAS shifted every run.

Its subtle and perhaps not noticeable enough but its far from consistent with how its evolving the pattern believe me.

For example, the 18z GFS shoved the limpet low well to the SW of the UK and gets absorbed into the flow that way.

Todays 00z stalled it out over Wales and drifted it before getting absorbed.

Todays 06z GFS take sit over Scotland before moving it westwards and eventually south-westwards towards Ireland.

Every single run is evolving the low differently from the last.

The ECM/UKMO/GEM mind you are being utterly rock solid and have had the same solution now for the last several suites, there are slight variations in the exact locaiton but all have kept the same pattern. GFS has shifted the low almost 500 miles from the 18z run to now, hardly a sign of confidence within its modelling.

Funnily enough the GFS 06z GFs is almost identical to that bad 12z ECM 2 days ago in the way it handles the low!

All that is not to say the GFS has not sniffed something out. Maybe it has? Watch for the limpet low over the UK, as long as that behaves we will get something down into our shores, its just how good it gets.

Edited February 2, 2021 by kold weather