Model Output Discussion - Deeper into Autumn

[Ali1977]

2 minutes ago, Mike Poole said:

That doesn’t happen anymore.  Since the FV3 upgrade, the GFS maintains high resolution throughout the 16 days of its run.

Both GFS and the // end exceptionally well.  There is possibly a bit of an all roads lead to Rome about this, because, as I mentioned yesterday, we do have a reasonable window of time with favourable drivers.  Hope we can make it count.  

Strat warming looking stronger now on //:

This followed by a SSW would work

[Frosty.]

The GEFS 12z mean in early December pretty much mirrors the ECMWF 0z mean in the same timeframe, ergo, the transition to more settled / blocked & colder is pretty much complete….hopefully setting us up for something decidedly wintery further down the line! ️ 🌫 

[Methuselah]

Yes, everything seems to be heading toward the same final destination -- blocking and eventually some cold easterly winds setting in. But timing, as always, is going to play havoc with our moods?

    

    

If you like the idea of -10C uppers traversing the North Sea, then this (given both the earliness of the season and the aftereffects of last summer) could be the Golden Ticket?

[Downburst]

Quick look at verification of the models. 5 day and 10 day, 10 day having a very poor result. One poster mentioned GFS post 168 hours, after 168 hours or 7 days the model has a far lower resolution in the physics, from 28 kilometers to 70 kilometers. So it is fair to say you can look at trends etc, but after 168 hours and with the charts below showing the 10 days skill you have to say it is still true that 5 to 7 days is really important. All sorts of things are missed after 7 days, that is simply a fact.

 

5 DAY

10 DAY

[TSNWK]

4 minutes ago, Downburst said:

Quick look at verification of the models. 5 day and 10 day, 10 day having a very poor result. One poster mentioned GFS post 168 hours, after 168 hours or 7 days the model has a far lower resolution in the physics, from 28 kilometers to 70 kilometers. So it is fair to say you can look at trends etc, but after 168 hours and with the charts below showing the 10 days skill you have to say it is still true that 5 to 7 days is really important. All sorts of things are missed after 7 days, that is simply a fact.

 

5 DAY

10 DAY

Thanks it was me on resolution post day 10. But @Mike Poole informed no drop in resolution since fv3 upgrade a couple years back..

Confused.com

[Tim Bland]

Ooosh… Now we’re talking !! 

[iowpompeylee]

There must be a lot of new folk in here. This type of rollercoaster/flip flop happens every single winter! I’ve been on here for many years and it’s all about trends this early out. Very encouraging signs of something cold afoot and I’ve seen a lot of forecasters mention it! Crossing the t’s and dotting the I’s will come in the coming days! Good to be back chasing this early on!!! Usually have to wait to jan/feb time! Enjoy 

[Downburst]

8 minutes ago, TSNWK said:

Thanks it was me on resolution post day 10. But @Mike Poole informed no drop in resolution since fv3 upgrade a couple years back..

Confused.com

Sorry Mike is wrong on that front. you can check it out on their site., “The model is constantly evolving, and regularly adjusted to improve performance and forecast accuracy. GFS is a global model with a base horizontal resolution of 18 miles (28 kilometers) between grid points. Temporal resolution covers analysis and forecasts out to 16 days. Horizontal resolution drops to 44 miles (70 kilometers) between grid points for forecasts between one week and two weeks. “

[TSNWK]

14 minutes ago, Tim Bland said:

Ooosh… Now we’re talking !! 

Probably correct south

[Frigid]

GFS 00z 

GFS 12z

Crazy differences in fl. A complete 180 

Unlikely as of now, but great to see them crop up 

[minus10]

Gfes 12z  z500 anoms still looking reasonable for beginning of December...upto t300

 

[ANYWEATHER]

I really don't know why people get hung up on each and every model run. The trend, dare I say it is very good for cold lovers. All the ingredients are coming togeather......Writing off the current synoptics is rather like writing off a cake you are going to bake with the finest ingredients, before you put it in the oven to Bake

[RJBingham]

39 minutes ago, TSNWK said:

Thanks it was me on resolution post day 10. But @Mike Poole informed no drop in resolution since fv3 upgrade a couple years back..

Confused.com

Would be good if the MOD thread could be spit into two 1-7 days and 7 days onwards.

[bluearmy]

52 minutes ago, Downburst said:

Quick look at verification of the models. 5 day and 10 day, 10 day having a very poor result. One poster mentioned GFS post 168 hours, after 168 hours or 7 days the model has a far lower resolution in the physics, from 28 kilometers to 70 kilometers. So it is fair to say you can look at trends etc, but after 168 hours and with the charts below showing the 10 days skill you have to say it is still true that 5 to 7 days is really important. All sorts of things are missed after 7 days, that is simply a fact.

 

5 DAY

10 DAY

 

47 minutes ago, TSNWK said:

Thanks it was me on resolution post day 10. But @Mike Poole informed no drop in resolution since fv3 upgrade a couple years back..

Confused.com

 

34 minutes ago, Downburst said:

Sorry Mike is wrong on that front. you can check it out on their site., “The model is constantly evolving, and regularly adjusted to improve performance and forecast accuracy. GFS is a global model with a base horizontal resolution of 18 miles (28 kilometers) between grid points. Temporal resolution covers analysis and forecasts out to 16 days. Horizontal resolution drops to 44 miles (70 kilometers) between grid points for forecasts between one week and two weeks. “

Lots of out of date information on the net 

gfsv3 runs at 13km out to day 16 at 127 vertical layers 

Edited November 23, 2022 by bluearmy

[Mike Poole]

34 minutes ago, Downburst said:

Sorry Mike is wrong on that front. you can check it out on their site., “The model is constantly evolving, and regularly adjusted to improve performance and forecast accuracy. GFS is a global model with a base horizontal resolution of 18 miles (28 kilometers) between grid points. Temporal resolution covers analysis and forecasts out to 16 days. Horizontal resolution drops to 44 miles (70 kilometers) between grid points for forecasts between one week and two weeks. “

Can you provide a link for that?  18 miles resolution?  That must be from donkey’s years ago.  Think horizontal resolution is now 13 km.

^ Thanks @bluearmy

Edited November 23, 2022 by Mike Poole

[Beanz]

Just now, RJBingham said:

Would be good if the MOD thread could be spit into two 1-7 days and 7 days onwards.

It's been asked every year, but unfortunately it's not a popular choice (but I do agree with you).  I'm sitting here wondering why we're discussing the possibility of cold at 16 days out...I get the 'trends' argument, but every year it's the same formula. 

384hr "BFTE v.2" -> 264hr "MetO and ECM must come on side" -> 240hr "GFS is rubbish" -> 168hr "winter is over" 

 

 

[ScottSnow]

6 minutes ago, ANYWEATHER said:

I really don't know why people get hung up on each and every model run. The trend, dare I say it is very good for cold lovers. All the ingredients are coming togeather......Writing off the current synoptics is rather like writing off a cake you are going to bake with the finest ingredients, before you put it in the oven to Bake

Yes there’s no point in looking at small details in every run. At the moment we should just be looking for the overall theme, which at the moment is for a blocking signal and a gradual cooling down.

Certainly there is potential for colder conditions. Let’s hope that potential is realised. 

Twenty years on here has taught me to err on the side of caution until I actually see the white stuff with my own eyes.

[Trom]

53 minutes ago, Downburst said:

Quick look at verification of the models. 5 day and 10 day, 10 day having a very poor result. One poster mentioned GFS post 168 hours, after 168 hours or 7 days the model has a far lower resolution in the physics, from 28 kilometers to 70 kilometers. So it is fair to say you can look at trends etc, but after 168 hours and with the charts below showing the 10 days skill you have to say it is still true that 5 to 7 days is really important. All sorts of things are missed after 7 days, that is simply a fact.

 

5 DAY

10 DAY

Root Mean Square Error is the standard deviation of observed values around forecast values for out of sample data.  So if the models were perfect they would have a RMSE of zero.  Are you confusing this with correlation which some verification data uses?

[swfc]

2 minutes ago, Beanz said:

It's been asked every year, but unfortunately it's not a popular choice (but I do agree with you).  I'm sitting here wondering why we're discussing the possibility of cold at 16 days out...I get the 'trends' argument, but every year it's the same formula. 

384hr "BFTE v.2" -> 264hr "MetO and ECM must come on side" -> 240hr "GFS is rubbish" -> 168hr "winter is over" 

 

 

Post of the day supposed to be a hobby not a trial. Anyway heres to a blocked December and snow

[Metwatch]

4 minutes ago, Beanz said:

It's been asked every year, but unfortunately it's not a popular choice (but I do agree with you).  I'm sitting here wondering why we're discussing the possibility of cold at 16 days out...I get the 'trends' argument, but every year it's the same formula. 

384hr "BFTE v.2" -> 264hr "MetO and ECM must come on side" -> 240hr "GFS is rubbish" -> 168hr "winter is over" 

 

 

Nothing but facts spoken here

[MATTWOLVES 3]

I always like to view those ens before making any kind of judgement,and tbh nothing as changed this evening. There are strong signals for Blocking to take charge further down the line. The ens show a variation of Heights emerging all over the place..from Greenland to scandy and pushes of Heights into the Arctic area. We could be seeing the vortex coming under all kinds of pressure down the line if some of these ens are close to the mark!

But I still feel this is a walk before you run situation right now,there are blocks being put in place and the possibilities become endless if all falls correctly into place.

But at this very early stage of proceedings I would definitely call this as a very promising situation to be in right now.

But most of all certainly not much in the way of Atlantic oomph and a zonal train taking hold.

To sum up it could become a very interesting run up to Xmas this year with a little bit of luck...but the signs and omans are pretty promising.

The World cup has already shown us that shocks do and will happen...so if we do encounter a severe protracted Wintry spell...dont be that shocked...its sods law that it will eventually happen.

A very good evening to each and everyone of ya.

Edited November 23, 2022 by MATTWOLVES

[swfc]

40 minutes ago, TSNWK said:

Probably correct south

Always it will be slider gate

[sheikhy]

Bye bye gfs and welcome ecm 12z!!!!!!!scandi wall from the east!!!no very cold air yet but i will take that for now!

[Bradley in Kent]

Evening all, hope everyone is well.

Like many in the past couple of days I've started to tentatively look out for the building blocks.

Obviously way, WAY too early to anticipate but put it this way, I'd rather have cold members in the mix rather than complete mild full house. At least there are avenues to go down where there's a chance of something seasonal.

We all know it can and in all likelihood will go wrong, but if it's full house mild there's no route at all where it's a no show for at least another fortnight.

[sheikhy]

Massive absolutely massive high on the ecm!!!and as i mentioned earlier more importantly we get the building block much earlier to my liking!!!surely aint gona sink from here!!lol!watever happens its gona feel a lot colder!