Model discussion: Hunt for cold - into mid-January

[Bradley in Kent]

Some very interesting and insightful comments regarding the current setup which I'd like to add on to.

If we are ever to get in the game for any decent cold, we need high pressure in our vicinity and not down in the Med or Azores. Most of the time, these heights end up slipping back south again, but sometimes they don't and end up going to where coldies want it. The position we're in at the moment is high pressure right over us with low heights to the north and Russia. 

This weekend is important coming up, as a low spins up behind our current high pressure with another high behind it.  This low is currently modelled to be a weaker feature but we're yet to see the exact strength. 

Current runs are suggesting this low will effectively get sandwiched and squeezed between the two high pressures. These themselves are boxed in by low heights in North America to the west and Russia to the east. The main question comes on Sunday, as to what happens when this low gets squeezed. Will it get squeezed and pinched out southwards allowing heights to slip of the top of it and link with heights over Central Europe where they can then retrograde Northwestwards? Or will the low be squeezed out of the sandwich to the north, which will bolster heights further south whilst at the same time allowing low pressure systems to barrel over the top and hence end up at square one again. 

It all comes down to timing, we might be lucky and have heights retrograde just as there is a lull in cyclogenesis off North America. Or we might not, and low pressure just comes barging across the Atlantic pushing aside any heights back down to the south because they didn't have the chance to build a wall south of Greenland. 

[Snowman.]

2 hours ago, feb1991blizzard said:

Must admit these NH patterns keep presenting their selves (very decent ones for the date), you would normally say that at some point before mid March a potent cold spell is almost inevitable, but given recent years i will stop short of that, i will instead suggest that the actual model watching season will be interesting for the duration.

Agree with this, high pressure seems to consistently want to move north just the PV is running over the top. If we catch a break or some help from the other side of the pole we may get somewhere. 

[jules216]

For those who are too much hung on the stratospheric developments. Look,zonal winds way above average speed,yet 10 day EPS have USA cold and most of Európe. Even now winter storms in US and my daily mean over last 5 days has Been -5°C. So much for strong vortex in the heart of winter. Yet post SSW last winter we had record mild 2nd half of February. IT was actually colder before onset of SSW then post SSW. As much as EP La Nina in -QBO is failing to produce SSW almost first tíme since this combo is present IT is not necessery relevant to weather here as cold spells are here without them. Maybe not record cold but at least seasonal. It proves that these fancy strat talks in autumns dont need to mean anything re.surface outcomes.

[PiscesStar]

1 hour ago, Bradley in Kent said:

Some very interesting and insightful comments regarding the current setup which I'd like to add on to.

If we are ever to get in the game for any decent cold, we need high pressure in our vicinity and not down in the Med or Azores. Most of the time, these heights end up slipping back south again, but sometimes they don't and end up going to where coldies want it. The position we're in at the moment is high pressure right over us with low heights to the north and Russia. 

This weekend is important coming up, as a low spins up behind our current high pressure with another high behind it.  This low is currently modelled to be a weaker feature but we're yet to see the exact strength. 

Current runs are suggesting this low will effectively get sandwiched and squeezed between the two high pressures. These themselves are boxed in by low heights in North America to the west and Russia to the east. The main question comes on Sunday, as to what happens when this low gets squeezed. Will it get squeezed and pinched out southwards allowing heights to slip of the top of it and link with heights over Central Europe where they can then retrograde Northwestwards? Or will the low be squeezed out of the sandwich to the north, which will bolster heights further south whilst at the same time allowing low pressure systems to barrel over the top and hence end up at square one again. 

It all comes down to timing, we might be lucky and have heights retrograde just as there is a lull in cyclogenesis off North America. Or we might not, and low pressure just comes barging across the Atlantic pushing aside any heights back down to the south because they didn't have the chance to build a wall south of Greenland. 

This has been nicely summarised. Thanks for the post and I agree with  lull in the jet stream this might also assist with retrogression? 

[Cold Winter Night]

2 hours ago, minus10 said:

I was under the impression that all  models used the same base data as it is shared, maybe i'm wrong on that. Still find it incredible that there can be such differences at relatively short timescales as is happening...

Most of the observations going into the models are from the same sources, but the key issue is that observations are not the starting point for the models. That's a common misconception about weather models.

The models use direct observational data (from weather stations, ships, airplanes etc.), indirect observational data (satellites) and the previous model forecast to calculate a model atmosphere at T+0h. The initial model atmosphere is effectively an estimate of the real initial state of the atmosphere.

So at 0h you're not looking at the latest observations, but it's a model calculation already.
That's why all models and ensembles are different. Different grid systems, different resolutions, different number of atmospheric layers, different data assimilation methods, different parametrisations for atmospheric processes etc. generate a different starting point.

Then as time progresses, the many differences between models will result in only more divergence between them, all the more so in complicated setups, as we know so well.

[TSNWK]

16 minutes ago, jules216 said:

For those who are too much hung on the stratospheric developments. Look,zonal winds way above average speed,yet 10 day EPS have USA cold and most of Európe. Even now winter storms in US and my daily mean over last 5 days has Been -5°C. So much for strong vortex in the heart of winter. Yet post SSW last winter we had record mild 2nd half of February. IT was actually colder before onset of SSW then post SSW. As much as EP La Nina in -QBO is failing to produce SSW almost first tíme since this combo is present IT is not necessery relevant to weather here as cold spells are here without them. Maybe not record cold but at least seasonal. It proves that these fancy strat talks in autumns dont need to mean anything re.surface outcomes.

You must be rubbing your hands with glee with 3 drops if vortex into your area forecasted next couple of weeks.. supported in part by us in the UK providing the high pressure association.  You owe us @jules216

[damianslaw]

It's an anticyclonic outlook, not of the cold or particularly mild category though northern parts are experiencing milder air from off the Atlantic.

The key forcing is from the PV which is separating into two lobes but not cleanly enough to enable heights to build far enough north to enable something colder from the north. It is a near miss on this occasion, the cold looks destined for east Europe and we will end up sitting under the high.

All eyes then on whether we maintain the separation which would most likely maintain the current synoptics for a while longer or we see more energy transfer back west to Canada allowing the jet to ride in over the top. 

It is an unusual sluggish pattern and I commented on whether the east QBO combined with la Nina imprint and state of MJO are weakening the energy in the jet.. we've had one named storm so far this season only... 

Edited January 12, 2022 by damianslaw

[feb1991blizzard]

14 minutes ago, Cold Winter Night said:

Most of the observations going into the models are from the same sources, but the key issue is that observations are not the starting point for the models. That's a common misconception about weather models.

The models use direct observational data (from weather stations, ships, airplanes etc.), indirect observational data (satellites) and the previous model forecast to calculate a model atmosphere at T+0h. The initial model atmosphere is effectively an estimate of the real initial state of the atmosphere.

So at 0h you're not looking at the latest observations, but it's a model calculation already.
That's why all models and ensembles are different. Different grid systems, different resolutions, different number of atmospheric layers, different data assimilation methods, different parametrisations for atmospheric processes etc. generate a different starting point.

Then as time progresses, the many differences between models will result in only more divergence between them, all the more so in complicated setups, as we know so well.

Are you sure that is the case?, i didn't think previous runs had any baring on the current run

[Cold Winter Night]

6 minutes ago, feb1991blizzard said:

Are you sure that is the case?, i didn't think previous runs had any baring on the current run

Yes. It's called data assimilation.

But it's only to analyse the initial state of the atmosphere. Not later on.

For the initial state of the 12z, they use the short (12 hour) forecast from the 0z, update it with observations to correct errors, and calculate the new initial state for the new forecast (which in turn will be used for the next 0z).

Data assimilation | ECMWF

WWW.ECMWF.INT

---

To make a forecast we need to know the current state of the atmosphere and the Earth's surface (land and oceans). The weather forecasts produced at ECMWF use data assimilation to estimate initial...

 

[Methuselah]

FYI: 

 

[Lukesluckybunch]

afternoon all..so the 6z gfs was a nice run for cold for me,temps averaging 4-6c throughout the whole run,also with some cracking ens in the mix..25,27,29, being very good! ICON 12z is not good in any way high comes in then quickly flattens.im looking at the timeframe now of between 144-168 when the high over the uk and pressure building in the atlantic at the same time link up,ECM was showing somthing like this morning.lets hope they can link up then head northwest!

[TSNWK]

36 minutes ago, Cold Winter Night said:

Yes. It's called data assimilation.

But it's only to analyse the initial state of the atmosphere. Not later on.

For the initial state of the 12z, they use the short (12 hour) forecast from the 0z, update it with observations to correct errors, and calculate the new initial state for the new forecast (which in turn will be used for the next 0z).

Data assimilation | ECMWF

WWW.ECMWF.INT

---

To make a forecast we need to know the current state of the atmosphere and the Earth's surface (land and oceans). The weather forecasts produced at ECMWF use data assimilation to estimate initial...

 

Thank you.. really useful post.. I can see  the logic in starting with a previous forecast then updating with actuals to then provide an initial state for that run.. good way to see and apply any early trends..

[jules216]

1 hour ago, TSNWK said:

You must be rubbing your hands with glee with 3 drops if vortex into your area forecasted next couple of weeks.. supported in part by us in the UK providing the high pressure association.  You owe us @jules216

We may need 2 or even 3 bites at Cherry IT looks that even on 17th the cold will miss even us, the cold high is not exactly exciting as much as if we were to get hit by low heigts, its only a glancing blow, so focus may shift to day 9-10 at earliest. Its like UK, a proper northerly is so rare, everything drifting too far east.

[northwestsnow]

 

Icon 12z 180.

It's more amplified than the 6z run at said timeframe.

Hoping the trend continues this evening ..

Edited January 12, 2022 by northwestsnow

[Lukesluckybunch]

2 minutes ago, northwestsnow said:

 

Icon 12z 180.

It's more amplified than the 6z run at said timeframe.

Hoping the trend continues this evening ..

yes exactly nw ,heights also building towards east seaboard at same time.hopefully will give the high towards iceland more of a boost

[feb1991blizzard]

57 minutes ago, Cold Winter Night said:

Yes. It's called data assimilation.

But it's only to analyse the initial state of the atmosphere. Not later on.

For the initial state of the 12z, they use the short (12 hour) forecast from the 0z, update it with observations to correct errors, and calculate the new initial state for the new forecast (which in turn will be used for the next 0z).

Data assimilation | ECMWF

WWW.ECMWF.INT

---

To make a forecast we need to know the current state of the atmosphere and the Earth's surface (land and oceans). The weather forecasts produced at ECMWF use data assimilation to estimate initial...

 

Here's the analysis chart for the 18z run.  

 

[Staffmoorlands]

UKMO much milder next week

[Lukesluckybunch]

gfs setting up nice here at 168,heights going to greenland

[northwestsnow]

Well GFS is a trier !

day 7 ...

[TSNWK]

Let’s see if attempt 2 does any better.. looks a bit higher into Greenland.. and up to this point there had been a tiny back west of the pattern.

Edited January 12, 2022 by TSNWK

[Battleground Snow]

3 minutes ago, Lukesluckybunch said:

gfs setting up nice here at 168,heights going to greenland

Probably topple again, but a bit better than the 06z

I am encouraged by the negative AO. Latest update has it even more negative than yesterday too.

Edited January 12, 2022 by Battleground Snow

[Ali1977]

Incoming cold by the looks of this 

[TSNWK]

Difference here is that vortex is spilitting might come let the heights in

[TSNWK]

And we get this at day 8! Team we have another chase………….

 

[CoxR]

this has near miss after near miss written all over it...