[Model output discussion - Closing in on Christmas]

I'm always encouraged by a cluster when both the Op and Control are singing from the same hymnsheet - often these two runs can resolve the mesoscale features which go on to have a huge effect on the latter part of the runs better than the rest of the suite due to their higher resolution.

Taking a look at the wider view of the 0z EPS clusters:

To be honest, by +240 clusters 2 and 4 (remembering cluster 4 contains both the control and op) look very similar to me, and I would personally group those together. If anything I am even more encouraged by cluster 2 than I am cluster 4.

[Stratosphere and Polar Vortex Watch 2020]

Lots going both from the bottom up and the top down.

I would usually caveat this by saying single operational run at long lead time, etc. but checking back through recent runs this really is pretty consistent

Absolutely enormous warming up at 1mb, that's +35c you can see in there (remember, this not an anomaly chart, that is actual temperature) from about day 11 onwards. Warming is still ongoing by the end of the run:

Seeping down into the levels below too:

Back at the very top, this leads to what looks to be a Wave 2 attack by the end of the run, but I could be seeing things there and it may instead be an all-encompassing Wave 1 - difficult to tell whether that is two distinct waves:

We then seem to have the more widely advertised bottom-up pressure which leads to at least a partial split from 100mb:

All the way up to 20mb by the end of the run:

The Wave 1 pressure pretty apparent there too at 20mb.

So we have a situation where we have top-down warming absolutely battering the vortex from the top, and tropospheric forcing causing some fairly significant bruising to the vortex from the bottom. Assuming that current forecasts are correct, the question really must become when, not if, the vortex falls entirely to its knees. I'll leave that to someone far braver than me.

 

 

 

[Stratosphere and Polar Vortex Watch 2020]

Very impressed with this as an ensemble mean:

A slight note of caution re: Strat forecasts being more accurate - to a certain extent, yes, mostly because in the stratosphere you don't have to put up with the unpredictable pitfalls of great big mountains sticking out from a flat surface, or vast oceans interacting with the air aloft. However of course the stratosphere forecasts from NWP modelling aren't just a standalone forecast at a single layer of the atmosphere, they're entirely coupled to the forecast through the troposphere too - and given this latest warming is likely triggered by a +EAMT in the main, caution should still be taken even at these timeframes (although as @chionomaniac alluded to over in the mod thread yesterday, it's always encouraging to see these warmings counting down with the days, rather than staying stuck out at T360+). So whilst the overall dynamics are easier to predict in the stratosphere, you still have to account for the entropy below.

And whilst the temperature profile looks impressive overall, remember temperature alone will not show you the state of the vortex:

By the end of the 0z op, the vortex is disturbed, but not entirely defeated. Further disruption will be needed, but fortunately the warmings look set to continue for the foreseeable

[Model output discussion - Closing in on Christmas]

1 minute ago, TSNWK said:

And by the same token can we assume that when mjo is a COD phase which I understand is not good for coldies that we are still able to get cold..

Absolutely - I think we can all be guilty about religiously following the MJO forecast charts, or looking at the 10mb temperature charts looking for an SSW, but as we have seen with some of the recent output (whether it proves to be correct or not), sometimes you can just get something cold spring up out of seemingly nowhere.

I guess it's what keeps us all nerdily interested in the weather - if it was entirely predictable what would be the point

[Model output discussion - Closing in on Christmas]

re: the MJO question - it's best to think of all the factors which get mentioned on here (e.g. QBO, SSW's, AAM state etc.) as singular parts of a great big machine. The more of these you have in your favour for cold, the higher the chance of cold there is. However, no one  single factor is definitely going to drive us towards another 2010. By having each one in your favour it's essentially like buying lottery tickets - theoretically the more you buy, the better the chances are that you will win, but there is no guarantee.

So having the MJO in an amplified, particular phase in a particular season certainly gives you a better shot of particular longwave pressure patterns (in the case of winter, as a very rough guide, something somewhere between phases 6-1 gives you a shot of something more amplified), but there is no one size fits all answer to the equation which is the weather. What I'm cautioning against here is assuming that what can be a driver of global longwave patterns will definitely deliver cold to the UK is a dangerous game to play (e.g. La Nina or MJO phase 6 = early winter atlantic amplification)

[Model output discussion - Closing in on Christmas]

3 minutes ago, LRD said:

A pretty cold Xmas Eve

What does that Atlantic High do next? It's like an EastEnders cliffhanger. Are Den and Angie still in it?

Unfortunately I suspect the positive tilt on the low coming out of Canada will mean this one is a toppler. But at the day 10 range it's all still up for grabs for now. And in the meantime it's still encouraging to see such multi-model support converging around the christmas period for a colder spell for all - those northerly airstreams can create all sorts of fun

[Model output discussion - Winter has arrived]

14 minutes ago, bluearmy said:

Ec = icon 

further east than before and light 

not the ec op’s finest hour on the short term as it ends up pretty much where the other ops have been for days 

Indeed, a big climbdown from it, and looking at it's 18z vs the current radar returns, it looks pretty accurate (ironically) with it's current track.

 

[Model output discussion - Winter has arrived]

2 minutes ago, bluearmy said:

Ec 6z edged east ....if the 12 z edges a bit further still then it will agree with the 12z icon ! 

Admittedly I don't have access to the 6z ECM, but surely that would need to be a fairly monumental shift from it's 0z position below:

Compared to the 12z Icon:

We're talking about the Western periphery being nudging Coventry on the ECM vs it barely reaching Cambridge on the Icon, so a shift of ~80-100 miles or so?

I appreciate that in the grand scheme of the wider troposphere that's pretty tiny, but at 24 hours out that seems fairly poor performance wise to me one way or another

[Model output discussion - Winter has arrived]

25 minutes ago, bluearmy said:

Latest icon takes that bit of snow through Suffolk and Norfolk before it heads west through lincs and the s peaks 

for the ridiculous ec runs of yesterday to be right, it surely needed to at least have corrected west a bit, not east!

 

ECM is certainly standing on its own right now against pretty much all other modelling. Given the UKV update above it sounds like that is pretty similar to the majority of modelling too.

As ever, radar watching needed, but you do get the impression the ECM has got this consistently wrong.

Ultimately you can probably forgive that in what is designed to be a medium range forecast model (if it indeed does turn out to be wrong), but certainly one to remember if events do transpire as the majority of high resolution modelling currently projects

[Model output discussion - Winter has arrived]

6z Euro4 takes a slight step Westwards, but not by a huge amount (0z v 6z below):

Still has a rather keen snow signal in there too, but from experience the Euro4 tends to overdo the snow extent in these scenarios:

Which when you combine with the forecast Dew Points at that timeframe looks even more suspicious:

[Model output discussion - Winter has arrived]

In the shorter term, taking a look at our little wave feature which brings the possibility of something wintry across a swathe of England tomorrow from the point of view of a selection of high resolution modelling available to us:

The general trend has been to push this feature significantly further East - the holdouts really continue to be Harmonie and the ECMWF.

Of course we all know in reality that with such a complex feature it will be all about radar watching in the morning. Without some substantial intensity to the precipitation I'm still struggling to see much in the way of lowland snow - so all eyes to the radar.

[Model output discussion - Winter has arrived]

Which you can find here:

Model charts for United Kingdom (Significant Weather) | EURO-4 (2 days)

METEOLOGIX.COM

---

EURO-4 (2 days) - Current forecast valid for 12/04/2020, 09:00am of parameter "Significant Weather", model chart for map...

Primarily rain across Southern areas, that is really down to the precipitation intensity  on the western flank being far weaker than on the ECMWF 0s, as mentioned earlier in this thread, and you can see below (Euro 4 left, ECM right):

Let's hope the ECM ends up nearer the mark

[Model output discussion - Winter has arrived]

I think both the models and quite a lot of people in here are overplaying snowfall potential on lower ground - although admittedly we don't get to see the key parameters for some of them that determine the rainfall/snowfall boundary.

But if, for example, the Euro4 is anywhere close to the mark:

Then away from higher ground, there is little prospect of any snow.

Looking at the GFS-derived NMM, we see a similar picture:

It's not until precipitation reaches the uplands of Northern England/Scotland that we start to see any appreciable snowfall:

 

As previously stated we don't get to see the DP's (or even better, the Wet Bulb Freezing Level) for the ECM, but unless it is drastically different to much of the rest of the output, I struggle to see much in the way of snow below at least 200m.

That said, we're a few days away - time for some improvements still

[Model output discussion - into the start of Winter]

A lot of sudden interest on what's happening in the stratosphere. There's two things to call out here:

1. We currently don't really have any coupling between the troposphere and the stratosphere (thankfully). If we take a look at the zonal wind cross section:

Whilst the vortex rages in the mid and northern latitudes at 1-30mb at above average levels, closer to the surface we see something slightly more serene. For as long as the vortex above rages, long may that continue. However of course such events as a sudden stratospheric warming can lead to dynamic coupling between the strat and the trop....

2. Should that happen, it may not be all that favourable. The truth about point 2 is that right now we simply don't know whether or not it will be favourable for a number of reasons:

- This is still a relatively minor warming being shown beyond day 10, which may not even come to fruition

- The potential event is really being triggered by a +EAMT event - the strength and duration of which we do not yet know, as it still remains a forecast

- We have such a strong vortex aloft that any warming events may not sink into the core of the vortex, and may simply ride the surf at the outer reaches of the upper vortex before being deflected away

But even ignoring all of these stumbling blocks before we can get to being anywhere near close to knowing what might happen - it takes a lot of work to get all of the right pieces of the puzzle in place to benefit our tiny little island. If what you are hoping for is an SSW that imprints onto the trop directly so that we have the stratosphere, rather than the troposphere, setting the tone for the global longwave pattern, then something like this from one of the previously posted tweets could be a horror show for the UK, with split vortex segments really not where we want them to be:

But to re-iterate, this is all theoretical, and should there be any stratospheric warming and subsequent impact on conditions on the ground come to fruition (but current gut instinct looking at the forecast wave activity would suggest the currently modelled attempt would not be enough to achieve such a scenario), we're looking at least 5-6 weeks away before this likely begins to impact us in any meaningful way.

In the meantime, if it's cold you're looking for, look for something tropospheric led.

[Stratosphere and Polar Vortex Watch 2020]

Unfortunately the bias corrected CFS outlook calls for far more average fare:

 

 

Of even more concern for me is the 10 day ECM forecast of the zonal wind cross section in the NH:

Whilst nothing too drastic just yet, clear signs there of some of the stronger zonal winds starting to imprint down closer towards the surface, with only really some minor Wave 2 closer to the top of the stratosphere attempting to make any sort of inroads on the beast aloft:

Of course all of that is still a forecast so we shall have to wait and see.

In the meantime I would be pinning my hopes far more on a bottom-up process to aid chances of some proper winter weather in the UK as opposed to anything aloft helping us for the foreseeable. 

[Model output discussion - Winter approaches]

Looks like some little nibbles of more seasonal weather on offer in the day 6-14 range it would appear.

The main thing for me still at the moment to turn this from a couple of brief wintry spells into something a little more sustained is still the amount of vortex left around Greenland  an annoyingly small amount of cyclogenesis still remains in that area. This was first apparent to me on last nights EPS Day 10 mean.

On the face of things, a pretty decent mean overall with the main bulk of the vortex shifting towards the Asia/Pacific side of the globe. However, spot the little 1010mb LP feature just SW of Greenland. For that to appear on an ensemble mean shows the pretty strong support for such a scenario, and this sadly prevents us from benefiting further from height rises in NE Canada, for now at least. Had we seen a clean link between heights in the mid Atlantic and Canada, combined with the trough clearing SE-wards, we could have seen ourselves being set up for something a little more substantial.

Are there any more positive signs today? Well a little yes. The day 10 ECM mean, whilst still holding a trough in that position just SW of Greenland, is a little less pronounced than it was last night, perhaps suggested a slight switch towards a more promising outlook overall. And of course the PV is far from the beast it could potentially have been at this stage of the season. 

Right now though the signs are that anything seasonal in the shorter term which relies on things falling into place upstream (i.e. to our NW) will be rather short lived.

[Stratosphere and Polar Vortex Watch 2020]

 

45 minutes ago, s4lancia said:

The connection to the eventual weather on our side of the hemisphere maybe tenuous but far from completely unconnected, so I found this very interesting. I was thinking about last SH spring’s highly anomalous temperatures in their high latitudes and look how this year we have opposite extremes in the upper troposphere to mid stratosphere.

Both way outside the climatic norms.
 

 

 

 

In terms of a connection, this would certainly suggest to me that the there is less ozone transport to the SH this year, although whether this is consequential of greater NH poleward ozone transfer, or a generally weaker BDC overall (which would be expected in a La Niña, wQBO year) I couldn’t say.

Hope it’s the former!

[Model output discussion - Winter approaches]

Always be wary of any output with the draining low heights away from Greenland but with little margin for error beyond day 4 - there has been a constant tendency for modelling at this timescale in recent years to exaggerate any height rises in the Greenland locale, with slow corrections towards lower heights the closer we get for T+0, leading to a less amplified Atlantic pattern overall.

I’ll remain sceptical until this starts to show up within the T+72 timescale with some consistency

[Model Output Discussion - Into Autumn 2020.]

7 hours ago, bluearmy said:

two days on and the features  A and B advertised by the ec spreads are firming up - the Siberian side lost its shape on the spreads day 9 so it’s no surprise to see that area less certain, run to run 
 

     

I'm a big fan of an Azores high in this position, not only because it offers the possibility of my favourite type of autumnal weather - morning frosts/fog, sunshine by afternoon (though this is always very uncertain with warmer SST's around our island), but because particularly when it is propped up by LP through the med, it leads to all sorts of interesting shorter and longer term possibilities, particularly if we can combine it with an Aleutian trough as the overall longwave pattern rotates eastwards around the northern hemisphere...

[Early run up to Winter 2020/2021 discussion]

The other thing that's worth remembering to highlight how too much focus on one teleconnection can be misleading, in amongst 2010/11 being suggested as an analogue partly on the basis of La Nina, is that from memory, we did stand on the verge of another major freeze during February which the UK was unfortunate to miss out on being just on the western periphery of the initial action. 

We started to see the core of the vortex shift towards the Western hemisphere, allowing for high pressure build to our N and E during week 2:

This wedge started to become more substantial through week 3 and left the UK right on the precipice of another snowy spell wtih Atlantic systems looking poised to duck underneath the expanding block:

Sadly we ended up with just a touch too much of a positive tilt to the jet and the Atlantic systems came roaring through:

 

Whilst that ultimately failed, you can see how close we were to a return to something substantially colder again through February - so there's no magic bullet to analogue matching, particularly when taking one or two teleconnections in isolation.

 

[Model Output Discussion - Into Autumn 2020.]

12 minutes ago, bluearmy said:

as November drifts towards and into December we are in a + NAO …… not exactly raging but with high heights to our south …….

Thanks for clarifying Nick, appreciate the ECM website data only runs out to week 4 in the main, but seemed too be too much of a sudden switch from what was being shown at week 4 to then be a raging +NAO - sounds like back to standard autumnal fare (which is what is probably to be expected at week 5 with a move back towards climatology)

[Model Output Discussion - Into Autumn 2020.]

2 hours ago, jules216 said:

According to EC 46d by the 4th week of November we are back in raging +NAO. 

 

Are we?

Usual caveats apply in terms of whether such forecasts verify at this range, but to me that looks like more of a tendency towards high pressure/-NAO than towards any sort of raging positive NAO at face value

[Stratosphere and Polar Vortex Watch 2020]

 

2 minutes ago, Griff said:

"Thin blue lines represent the individual ensemble members, thick blue line the ensemble mean; red lines show the mean and the 10th and 90th percentile of the model climate." 

In a good way it looks quite average?  

Yeah it is broadly in line with the extended GFS outlook, very slightly below average as an ensemble mean but nothing too much to write home about in either direction:

I'm still absorbing the wealth of new data which has become available, but another one that caught my eye (and is broadly in line with expectations of a Nina winter with a front-loaded potential feel to it)

Certainly a tendency there towards a negative NAO through October and November, slightly less so for December, and then Jan and Feb look to tend towards more of a +NAO outlook

 

[Stratosphere and Polar Vortex Watch 2020]

On the 7th day of October, the ECMWF gave to me...

https://apps.ecmwf.int/webapps/opencharts/products/extended-zonal-mean-zonal-wind?area=nh&base_time=202010050000

The 45 day 10mb ensemble plot for free

[Snow & Ice coverage in the Northern Hemisphere Winter 2020/21]

Very useful free output now available from the ECMWF freebies for snow cover forecast - usual caveats apply about the below charts, being a single operational run etc.

https://apps.ecmwf.int/webapps/opencharts/products/medium-snow-sic?base_time=202010070000&projection=opencharts_north_pole&valid_time=202010070000

Looks like plenty of snow to come south of 60N further east into Asia, we could do with more of a push westwards for the second half of the month