Model output discussion - 5th December 2016 - Into Winter we go

[Summer Sun]

2 minutes ago, Seasonality said:

Definitely not a blowtorch Sou-Wester though. 

 

The ens is different to the Op

[MattH]

Just further to a few tweets this morning I don't think we are any the wiser as to how things will pan out as we progress towards and beyond mid-month and towards the Christmas period. The 00Z EC ENS are appalling if you're looking for cold weather. A broad upper trough across the N Atlantic, as we have at present, generally higher heights to the SE most of the time and, overall, a predominantly S or SW'ly air flow over the UK. There is little or no evidence within the EC Clusters to support the height rises that the GFS ENS continues to promote so something has got to give here at some point. The 00Z Det (or HRES) was most certainly at the 'bottom end of the scale' when it comes to being a cyclonic evolution, but when you look at the clusters there really is some 'nasty' looking lows evident too. The clusters, to be honest, for the last week or more have been useless, they have frequently been like Det models in that each day comes and goes and each time the clusters constantly change with little way of picking out any trends. Sometimes the clusters have been broken down into 6 or even 7 'populations' each highlighting a different synoptic outcome, might as well toss a coin and operationally it has and continues to be a highly frustrating and difficult period for forecasting anything beyond 5 or 6 days. Obviously the 00Z EC ENS mean is freely available on the internet and that, overall, tells the story out to 240hrs at least.

I've used the phrase 'make or break' before and whilst each day comes and goes with a lot of uncertainty, there will come a point when the middle and latter half of the month will likely become more certain, whether that is towards a more blocked regime or whether we just end up with a particularly uninspiring evolution which supports average or slightly above average temperatures and generally far more wet and windy weather than was expected.

If someone put me on the spot I would say that the more blocked evolution is ever so slightly more likely than what the EC ENS generally continues to produce longer term, primarily due to some other variables at work (strat, MJO etc), but the very frustrating and uncertain medium and longer evolution isn't going to go away anytime soon I don't think and, from a 'coldies' point of view I wouldn't presume that northern blocking, ridging to the NW etc will come about as, overall, there just isn't enough evidence 'across the broader' to support that, but, as ever, especially given the time of year, fingers crossed for the seasonal Christmas most would want and hopefully no additional repeats of the current weather type as what we have at the moment sure brings back memories of 12 months ago.

Regards, Matt.

[Seasonality]

4 minutes ago, bobbydog said:

Here you go-

http://www.meteociel.fr/modeles/gfse_cartes.php?ech=204&code=0&mode=0&runpara=1&carte=1

It is within the menu of the OP run on there though

Thanks @bobbydog and @Ali1977. Got it.

[blizzard81]

Sounds like the zonal train is coming unfortunately. Like most trains, it was just late.

[Southender]

18 minutes ago, MattHugo said:

Just further to a few tweets this morning I don't think we are any the wiser as to how things will pan out as we progress towards and beyond mid-month and towards the Christmas period. The 00Z EC ENS are appalling if you're looking for cold weather. A broad upper trough across the N Atlantic, as we have at present, generally higher heights to the SE most of the time and, overall, a predominantly S or SW'ly air flow over the UK. There is little or no evidence within the EC Clusters to support the height rises that the GFS ENS continues to promote so something has got to give here at some point. The 00Z Det (or HRES) was most certainly at the 'bottom end of the scale' when it comes to being a cyclonic evolution, but when you look at the clusters there really is some 'nasty' looking lows evident too. The clusters, to be honest, for the last week or more have been useless, they have frequently been like Det models in that each day comes and goes and each time the clusters constantly change with little way of picking out any trends. Sometimes the clusters have been broken down into 6 or even 7 'populations' each highlighting a different synoptic outcome, might as well toss a coin and operationally it has and continues to be a highly frustrating and difficult period for forecasting anything beyond 5 or 6 days. Obviously the 00Z EC ENS mean is freely available on the internet and that, overall, tells the story out to 240hrs at least.

I've used the phrase 'make or break' before and whilst each day comes and goes with a lot of uncertainty, there will come a point when the middle and latter half of the month will likely become more certain, whether that is towards a more blocked regime or whether we just end up with a particularly uninspiring evolution which supports average or slightly above average temperatures and generally far more wet and windy weather than was expected.

If someone put me on the spot I would say that the more blocked evolution is ever so slightly more likely than what the EC ENS generally continues to produce longer term, primarily due to some other variables at work (strat, MJO etc), but the very frustrating and uncertain medium and longer evolution isn't going to go away anytime soon I don't think and, from a 'coldies' point of view I wouldn't presume that northern blocking, ridging to the NW etc will come about as, overall, there just isn't enough evidence 'across the broader' to support that, but, as ever, especially given the time of year, fingers crossed for the seasonal Christmas most would want and hopefully no additional repeats of the current weather type as what we have at the moment sure brings back memories of 12 months ago.

Regards, Matt.

Thanks for the update Matt, although not the sort of update most of us in here want to see unfortunately.

Let's hope the ENS flip in our favour for once, although the form horse suggests otherwise.

Far too early to be calling December a bust cold/snow wise, but we need to start seeing a lot more crumbs than we are now to salvage anything from this month...a model break in order I think and let's see where we are this time next week. At the moment we are just going to get mixed output for mid-long range so pointless trying to interpret it.

Edited December 8, 2016 by Southender

[Singularity]

Honestly, if ever there was a time to take a break from analysing the models and ensembles completely, it's now.

Struggling to understand how the 00z EC ens are as they are. The implosion of the vortex after it tries to ramp up against a disorganised lower-mid-level strat vortex seems logical enough, preventing thermal gradient feedbacks from sustaining a strong jet across the N. Atlantic.

Maybe something weird happened with the observation data for the tropical signals? Though it would have to be a quite strong phase 4 or 5 signal to explain such an insistent westerly outcome in the ens.

You see - in such times trying to interpret the output just leads to muddled minds and in any case the entire suite could well change flavour in the next update. Amazing - but frustrating if you're in the forecasting industry!

[blizzard81]

1 minute ago, Southender said:

Thanks for the update Matt, although not the sort of update most of us in here want to see unfortunately.

Let's hope the ENS flip in our favour for once, although the form horse suggests otherwise.

I agree. Once the majority of the ecm ens flip from blocking to zonal, they hardly ever switch back unfortunately.

[MattH]

Just to put things into context at one particularly time frame, taking the 240hr time frame (18th Dec) and looking through the 00Z GFS ENS there are approx 13 members, out of the 20 population that have a more pronounced blocking pattern over or around the UK, whether that be high pressure over the UK or whether in a more favourable position for colder synoptics. At the same time frame there are, essentially, none  from the 00Z EC ENS. The bulk of the ENS (population 1 out of 3), which is 20 members, have a pattern similar to now with higher heights just to the SE, but with a large scale upper trough to the NW. The other 31 members simply have nothing resembling any type of blocking regime at all.

To say it is definitely a case of GFS ENS v EC ENS longer term is an understatement as clearly one is most certainly interested in a far more anticyclonic evolution, the other, just doesn't want to know. Clearly this is just based on one run, the 00Z, but as I've been mentioning on twitter over the last few days the GFS ENS has shown more consistency over this blocked evolution moving forward, the EC ENS have been all over the place, so take that as you will.

Cheers, Matt.

Edited December 8, 2016 by MattHugo

[weirpig]

2 minutes ago, Southender said:

Thanks for the update Matt, although not the sort of update most of us in here want to see unfortunately.

Let's hope the ENS flip in our favour for once, although the form horse suggests otherwise.

i think i must have read the wrong update,  i read it as clusters are mainly going for a zonal flow  however they are useless at this juncture  and Matt expects ever so slightly to favour a more blocked regime.  Still very much up in the air is how i  construed it.

[karlos1983]

7 minutes ago, blizzard81 said:

I agree. Once the majority of the ecm ens flip from blocking to zonal, they hardly ever switch back unfortunately.

Well I would say it's easy to take the negatives from what Matt has said, but I don't see it that way. I see it still as a very finely balanced setup, with blocking still a possibility as is zonal. Even Matt said that the clusters have been flipping from one run to the next for the past week. 

I hear what your saying that it's rare we flip from mild to cold, but I don't think we can say that from one run. Opportunities are still rife imho. 

Dare i say it..... more runs needed. 

EDIT: Indeed Matt updated as I typed lol. Still all to play for. 

Edited December 8, 2016 by karlos1983

[Seasonality]

19 minutes ago, weirpig said:

i think i must have read the wrong update,  i read it as clusters are mainly going for a zonal flow  however they are useless at this juncture  and Matt expects ever so slightly to favour a more blocked regime.  Still very much up in the air is how i  construed it.

Me too. I think people are automatically seeing the worst.

[johnholmes]

Just to add my penn’orth. As usual with the anomaly charts at 500 mb. Not that there is much to hang a cold hat on so to speak. Just a faint glimmer perhaps on the NOAA output. The chart issued last evening for 6-10 days ahead is, in most aspects, very similar to the one issued 5 days ago, see the last one below.

The ‘glimmer’ is that the +ve height anomaly is now shown as in the southern North Sea not over central western Europe! There is also a slight ridging in the contours east and NE of the UK. Also the contour flow has eased down quite a bit from what was indicated on charts up to a day or so ago. (That is the flow into most of the UK from the Atlantic).

Neither ECMWF or GFS really have this though, the latest EC-GFS is also shown below. Usually NOAA is more like the EC output but it is more like GFS in this comparison.

So no real continuity between the 3 models I use but NOAA is usually closer to what happens so maybe, maybe, just a suggestion that +ve heights at 500mb are trending N, NNE or perhaps NNW with time?

http://www.cpc.noaa.gov/products/predictions/610day/500mb.php

http://mp1.met.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

[Scandinavian High.]

Looks like cold run up to Christmas dry frosty . Also high moving out To the north west

Edited December 8, 2016 by abbie123

[weirpig]

just to compare  GFS and ECM

[Snowy L]

Solar flux and sunspot number on the way down again after that annoying surge that gave us our highest sunspot count since September.

[Seasonality]

Happy ending coming up on the 06z

 

 

[bluearmy]

Ecm op AO/NAO dive negative day 10 and interestingly the GFS also though much less pronounced

this was the first time for a fair while that the ops had done this and one wonders if the fact both did so on the same run should be given more  attention than would usually be the case on ops at that range 

 

[tight isobar]

Hardly perishing cold....

But another possible angle/ evolution' as the models toy with high cell placement. 

Also ignoring the exactions of 850's..to which if evolutionary take that path..thing could become vigorously more interesting in that department. ..

Certainly not writing off notable cold spell later this month. ....'just yet'!

[Methuselah]

Still looking 'interesting' in the later stages, I think: I'll take a slightly displaced AHP over any sort of displaced Bartlett any time!

[bluearmy]

The op looks pretty isolated on the London eps graph re wind speed and precip

the general trend for temps to be higher later days 10/15 than previous suites 

[Singularity]

GEFS has been more keen and quicker to let the Pacific tropical forcing take over from that in the Indian Ocean/Indonesia so that seems a good explanation for them having a blocking signal.

EC ens must be really slow to change things or perhaps even have a standoff situation. The model has a known under-progressive bias in the Pacific sector so that might be important.

Its odd that the ECM set reflects more of a GEFS type tropical evolution, but perhaps the lower-mid stratospheric vortex disruption manages to influence the troposphere enough to produce such changes independent of the tropical forcing... though why the det would do this and not the ens. is beyond me as I thought the ens. had as many strat. levels as the det. - or so I have that wrong?

[Nick F]

I think there is a enough uncertainty and spread beyond day 7, not just from the GEFS members, but also the EPS members too, to calm the worries that mild SWlys are here for the long haul. Though not totally discount the worries either.

The uncertainty seems to stem with how the models modulate deep cold trop vortices dropping down over Canada and Nern U.S. over the next 10 days and consequently how they handle amplification downstream of these vortices over the N Atlantic. One these vortices drops SE over E Canada next 24hrs - introducing cold continental arctic air south across much of northern, central and western U.S. But it's another, much deeper and colder vortex modelled to drop SE then E over Canada next week which could be causing GFS and EC op differences downstream to the upper flow. 00z ECM det. pulled this vortex E out into the  NW Atlantic - flattening the flow, whereas 00z and 06z GFS swing the vortex NE toward Labrador Strait and Greenland, amplifying the flow more downstream over the N Atlantic.

Just look at the 500mb vortices over N America heading toward NW Atlantic on 00z EC, and you might see what I mean

Against the unfavourable background of weak MJO signal, low AAM, wQBO and deep cold over northern N America driving a strong and zonal jet stream, there does appear to be enough volatility to create some amplification downstream over N Atlantic/Europe ... just a case of whether these ridges will set up favourably to advect colder conditions back to us. There are, at least, favourable hints of cross polar ridging appearing on the models later on too - which could help lower the AO index.

Edited December 8, 2016 by Nick F

[snowwman]

I understood that a Westerly QBO is more associated with  extremes of weather- extreme mild or extreme cold, than Easterly QBOs, which, however, tend to give colder weather when averaged. Think 1947, 1979, 1986, 2010.

And if the current, Westerly QBO is record- long, as it appears to be, and there continues to be a strat- trop disconnect, then past about a week, there's ample reason to treat the model output with caution. Were're in uncharted territory.

Edited December 8, 2016 by snowwman

[Man With Beard]

GFS 06Z still on a completely different page to the ECM ... or is it? Heights pulling back to the Atlantic, low heights back over Europe, breakthrough of low pressure into the North Sea and beyond ... ok that's as far as it goes!!

  

The parallel continues to promote a massive storm to our north in deep FI (between Dec 20 and 24) - will be interesting to see if that verifies

What's even more remarkable is the difference between the GEFS and ECM means at T240. I usually see this as a sign of high uncertainty.

  

However, over the years I've rarely seen the GEFS mean be wrong when it produces a 1030mb high, so I'm going to side with the ideas that the Euro High at D8 will move towards a UK High by D10.

And just for fun (some of us need it after the past eight tortuous weeks) - P14 has the eye candy this morning!

 

[Nouska]

4 hours ago, blizzard81 said:

I find the model output absolutely astonishing at the moment. The difference between yesterday's ecm 12z and this morning's 00z is nothing short of amazing. One of the biggest differences I have seen in 13 years of model watching.

I hope they have plenty of 'juice' in the DECIDER function - it is going to be getting a very thorough work out in the coming days - maybe even weeks.

http://www.ecmwf.int/sites/default/files/elibrary/2015/13419-decider-visualising-range-possible-weather-scenarios-medium-range-and-monthly-forecast-periods.pdf

The new monthly update due out tonight will have the same starting data as this last ECM run - does not bode well for a clearer path to beyond the Christmas period.

The late run modelling from ECM does reflect how it saw the week three period, as per last EC46. Deep, cold troughing exiting NE Canada but that was on a steep SE trajectory which allowed heights to build to our north. All a matter of how the big vortex cogs are expected to position from the starting data.

I suspect the models are having real difficulty in not having a coherent tropospheric PV and not being able to create a stable one.