The hunt for cold - Model discussion (late November)

[SMU]

It would help 'yourselves' when referring to back loaded winters & front loaded ones what you actually mean - as some could be referring to cold - Others snow - others blocking... -

If one is just referring to 'blocking' positions then a back loaded winter will always be favoured due to the fact that the vast majority of High lattitude blocking scenarios as well as SSWs occur into the new year ( as well as adding average propergation times ) you already at mid Jan-

However when your talking about potential snow & or cold then you dont need the huge 'showcase charts' just one mid lattitude block would suffice-

That said we are at the tail of November & charts of the 10HPA Zonal wind are being shown to indicate the early season running- well thats totally pointless if the strat is not coupled with the trop-

The AO sitting @-4 will be probably within the top 10 of all years for lowest November values ( poss top 5 ) so as far as the actual atmosphere is concerned the NH taken as a total entity is already in a front loaded winter - its just whether the UK gets to benefit -

The period Dec 7th-17th will probably be the one that decides people perceptions of that however we as said we are already front loaded bearing in mind the standard Dec pattern is around +1 to +2 AO so currently we are probably 2 standard deviations below that....

Just checked in Heathrow so have a good few days -

GFS mean at day 10 & onto day 15 is Easterly @feb1991blizzard

[TSNWK]

17 minutes ago, mountain shadow said:

Unfortunately given the updated Met Office long ranger we now know we can treat these spectacular runs as nothing more than randomness from the logarithms for the next month or so.

I think we're more likely to see a northerly toppler than full blown easterly over the next few weeks.

 

Oh! that's it then... the next four weeks are settled and we can dismiss the next two weeks of GFS / Euros and start over Mid December! - Darn right we will  F5 clicking from 4:00Pm daily at least, looking for those 50 mile swings in the right direction or not! and is that not  half the fun :):)

[frosty ground]

9 minutes ago, Steve Murr said:

GFS mean at day 10 & onto day 15 is Easterly @feb1991blizzard

Easterly? Not sure i see an easterly on the mean

[Ventnor Viking]

2 hours ago, Yarmy said:

Hi, Shaftesbury. I'm not sure that 3 snapshots from a single op run of the 4th-best performing model really indicates anything, but for what it's worth I'm not overly optimistic for December either (the first half more so). This is based on all the output out to 12 days or so. Beyond that there's a ton of scatter on the ens, but there are still a significant number showing high lat blocks and a negative AO:

https://www.meteociel.fr/cartes_obs/gens_panel.php?modele=0&mode=6&ech=300

So not an overwhelming signal for anything, but I've looked at hundreds of these GEFS panels over the years when they are showing nothing but 100% zonal out to 16 days. 

 

 

Hi Yarmy,

I totally agree and good point. It was more to show that these runs chop and change so much and we need to be looking at the broader patterns. Sorry if you have already seen it but I posted a couple of longer range models which show a back loaded winter (blocked pattern, Greenland High and undercutting lows bringing snow) which I think will be the case. The GFS does love the zonal pattern for sure. The high in the arctic is pretty much driving the models such as the GFS and ECM and with every different position they present that high that's when we see some exciting output, both the ECM and GFS are really struggling with it at the moment. Out of those two models I favour the ECMWF just from past experience as the GFS seems to follow this model rather than the other way round.  

Edited November 27, 2018 by Shaftesbury Snow

[bluearmy]

It seems pretty obvious that any relaxation of the jet will allow for an amplification …… whilst some runs were showing that to be early next week, it seems more likely to be later …. if we don't get lucky it may not be at all!, however, i would be very surprised if we have to wait beyond the latter part of next week to see the amplification begin and then it becomes a question of where that will occur - current favourite is to end up over scandi (could be sceuro, could be up through the uk, could even be to our west). so whether we end up in a md dec position where we go proper wintry, surface cold under an inversion, dull and grey or quite pleasant (could be cold and clear, could be mild and clear) remains the big question...…… 

[SMU]

9 minutes ago, frosty ground said:

Easterly? Not sure i see an easterly on the mean

Its there follow it back - crystal clear

 

[snowray]

1 hour ago, frosty ground said:

They don't make spoiler lows like they did in the past.

It seems to me that we may well end up with a Scandi high at some point after mid December, it keeps popping up on the runs regularly now and I noticed the other day when the Italian Meteo guys were doing their winter outlook that they were suggesting what looks like a dominant Scandi block and LP towards the Med, by the look of the anomaly charts that they had up anyway.

 

Edited November 27, 2018 by snowray

[Yarmy]

3 minutes ago, snowray said:

It seems to me that we may well end up with a Scandi high at some point after mid December, it keeps popping up on the runs regularly now and I noticed the other day when the Italian Meteo guys were doing their winter outlook that they were suggesting what looks like a dominant Scandi block, by the look of the anomaly charts that they had up anyway.

 

I'm guessing this is the source:

It's the DJF mean SLP from the Italian Met.

[SMU]

^^^^ Genoa low will do us just fine...

[Methuselah]

12 minutes ago, Steve Murr said:

Its there follow it back - crystal clear

 

I don't bother much with the mean, Steve: the arithmetic mean, between to totally opposed clusters will, I think be almost useless; I'd expect the real weather to tend towards either one cluster or the other, and not to wander around in the vacuum between the two?

That said, stranger things have happened!

[bluearmy]

1 minute ago, Ed Stone said:

I don't bother much with the mean, Steve: the arithmetic mean, between to totally opposed clusters will, I think be almost useless; I'd expect the real weather to tend towards either one cluster or the other, and not to wander around in the vacuum between the two?

That said, stranger things have happened!

there is so much variance at the moment ed that the mean gives an indication of how amplified we might go - years of doing this gives you an idea of what the likely scenario of a mean amplification would be in reality ….

[snowray]

22 minutes ago, Yarmy said:

I'm guessing this is the source:

It's the DJF mean SLP from the Italian Met.

Well done Yarmy, yes that looks like it, blinding.

[SMU]

20 minutes ago, Ed Stone said:

I don't bother much with the mean, Steve: the arithmetic mean, between to totally opposed clusters will, I think be almost useless; I'd expect the real weather to tend towards either one cluster or the other, and not to wander around in the vacuum between the two?

That said, stranger things have happened!

I normally agree however with a Scandi block at 348 which is day 14 you would expect a mean to reflect the long term average as by then most ENS have filtered out the blocking 'noise' - so to have a fairly solid Scandi high & continental cold pool means that there cannot be much 'opposing' data otherwise the signal would be weaker...

Anyway - looking forward to the next couple of weeks, all the info & data highlighted in the Winter forecast is very much live & relative to where we are on todays date stamp- !

Best

S

[mb018538]

The other point for all of the doom mongers is that it's still late November - even if nothing comes up in the next 14 days we will only be 10% through winter. It'd be the equivalent of writing of summer because the first week of June was bad.

[Chesil View]

11 minutes ago, mb018538 said:

The other point for all of the doom mongers is that it's still late November - even if nothing comes up in the next 14 days we will only be 10% through winter. It'd be the equivalent of writing of summer because the first week of June was bad.

A most salient point. Well made

[Methuselah]

18 minutes ago, bluearmy said:

there is so much variance at the moment ed that the mean gives an indication of how amplified we might go - years of doing this gives you an idea of what the likely scenario of a mean amplification would be in reality ….

That's true enough BA, as far as it goes...But using the mean to analyse two quite distinct outcomes would be similar to lumping men's and women's heights together, and deriving the average height of a non-existent hybrid...?

IMO (only MO, I stress) it's very likely (perhaps almost certain?) that, under such circumstances, any amplification will, in reality, not coincide with any model-mean...?So it's not that I dismiss it, not at all; it's just that, in some extreme cases, it might not be representing quite what intuition may suggest...

Anywho...Great discussions today!

[snowray]

10 minutes ago, mb018538 said:

The other point for all of the doom mongers is that it's still late November - even if nothing comes up in the next 14 days we will only be 10% through winter. It'd be the equivalent of writing of summer because the first week of June was bad.

Indeed mb. But then again it does not take long for a month to go by if the Atlantic is driving things, and thats 1/3 of winter gone, it would certainly be nice to see some light at the end of the tunnel by D14. Something thats been missing for way too long is a White Christmas, it would make a most welcome change to see a few flakes by then.

[Ventnor Viking]

56 minutes ago, mb018538 said:

The other point for all of the doom mongers is that it's still late November - even if nothing comes up in the next 14 days we will only be 10% through winter. It'd be the equivalent of writing of summer because the first week of June was bad.

To be fair I don't think anyone has written off winter. I think many of us are hoping for the cold to come through quicker then actually will. In fact, March has actually been one of the snowiest months in recent winters, so there's still loads of time left

Edited November 27, 2018 by Shaftesbury Snow

[frosty ground]

1 hour ago, Steve Murr said:

Its there follow it back - crystal clear

 

 

Don't see it because it's not there, see heights to the south (Blue) , but the mean flow is south westerly (red).

[Mucka]

I'm a bit confused as to where you stand Shaftsburysnow, did you not call zonal through Dec yesterday and write off any pattern change? 

Today you seem to be saying that you favour a back-loaded Winter (Climatological norm) but we can't write off any cold in December?

Are you still writing off a pattern change based upon this mornings output or not? 

This mornings output was disappointing because it did not advance last nights improvements and trended more toward mid lat blocking than high lat blocking but that could easily go back the other way this afternoon and evening.

We are currently looking for a blocking pattern to establish as we head into 2nd week Dec, I think that still looks favourite but whether that results in cold and snow is another matter.

Personally I believe we will see a pattern change to blocking toward end 1st week Dec which will of course increase the chance of cold reaching us but by no means guarantee it, beyond that the current signals are for another zonal spell before the teleconnections work their magic later in Winter which would go against my idea we will get other blocking opportunities later in the month (Dec) but I certainly wouldn't ever be married to long range generalities either way.

I''m much more interested in looking for pattern changes and cold chances within the 2 week time-frame and even those can and do go pear-shaped.

 It will take more than a few dodgy runs before I wave the white flag (though we do need keep a white flag handy in this game)

 

Edited November 27, 2018 by Mucka

[Kirkcaldy Weather]

some nice charts from the 00z run of the GEFS Legacy 

 

a few with some cold pooling over Europe 

 

[ArHu3]

1 hour ago, Broadmayne blizzard said:

A most salient point. Well made

But cold takes far more time to arrive and from February onwards the sun gains ever more strength. Otherwise it's just a few days but for a prolonged winter period we need winter to come in by January 

[Ventnor Viking]

35 minutes ago, Mucka said:

I'm a bit confused as to where you stand Shaftsburysnow, did you not call zonal through Dec yesterday and write off any pattern change? 

Today you seem to be saying that you favour a back-loaded Winter (Climatological norm) but we can't write off any cold in December?

Are you still writing off a pattern change based upon this mornings output or not? 

This mornings output was disappointing because it did not advance last nights improvements and trended more toward mid lat blocking than high lat blocking but that could easily go back the other way this afternoon and evening.

We are currently looking for a blocking pattern to establish as we head into 2nd week Dec, I think that still looks favourite but whether that results in cold and snow is another matter.

Personally I believe we will see a pattern change to blocking toward end 1st week Dec which will of course increase the chance of cold reaching us but by no means guarantee it, beyond that the current signals are for another zonal spell before the teleconnections work their magic later in Winter which would go against my idea we will get other blocking opportunities later in the month (Dec) but I certainly wouldn't ever be married to long range generalities either way.

I''m much more interested in looking for pattern changes and cold chances within the 2 week time-frame and even those can and do go pear-shaped.

 It will take more than a few dodgy runs before I wave the white flag (though we do need keep a white flag handy in this game)

 

Just to clarify again.

A back loaded winter means cold - blocked - snow in the second half of winter. I referred to a major pattern change going in to mid to late Jan onwards. I posted this morning charts to back this up...please take a look. I also called zonal up to 144hrs on yesterdays 12z GFS.

I definitely did not rule out any cold in December , however I did say any cold wont last long in December and would probably be transient and weak. There wont be the major patter change just yet though that will happen in January based on the attached. There will be attempts at it though as the NH profile is out of sync.

To anyone else who is referring to my posts please quote them as you will see they are quite different to what is being said.

Does this now make it clearer? I think you have also contradicted yourself there. Please quote where you think I said things and I will happily clear up any confusion  

The links - Canadian model from dec to feb and then JMA dec to feb

 

 

Edited November 27, 2018 by Shaftesbury Snow

[Ross Andrew Hemphill]

Potential for some snow in Scotland this weekend on the GFS 12z, down to about 300m so definitely one to watch in the next few days!  

[Interitus]

3 hours ago, Yarmy said:

Presumably this is driven largely by MOGREPS. It's a shame that Ian F got chased from here, as he used to have good insider knowledge on confidence and the rest.

There's actually some archive data for MOGREPS now publicly available at the following link, albeit in netCDF format so you'd have to write some code to read it and render it.

https://aws.amazon.com/public-datasets/mogreps/

I don't know how up to date it is, but it does say this 'Data is currently updated infrequently with regular updates coming soon'.

 

Edit: Just googling around and found the following tool called Panopoly. Anyone used it?

https://www.giss.nasa.gov/tools/panoply/

 

Panoply doesn't seem to like the files, but here is a quick render in IDV for the forecast temperatures at 4pm on July 1st 2015 from earlier same day (July record max) -

MOGREPS doesn't run in to long-range forecasts, and these archived forecasts comprising a massive amount of data represent something of a curiosity for other than serious research.