Model output discussion - into 2018

[Singularity]

20 minutes ago, Catacol said:

Any idea what's going on with this? The cache issue is bizarre - @Singularity - sorry - you are correct - my WDT graphic is NOT the one that then appears after the save. Only work around seems to be to save and then upload the saved image. I'll try and adjust my initial post....

 

EDIT: Grrr - cant edit my original cos time has gone. GP has posted the proper one I had meant to show above. I'm off to have another moan at WDT....

Ah the fun of science 

I have been under the impression that a bit of a drop in GLAAM was always expected as the MJO moved through the western IO, with the hope being that trends would reverse again as the MJO propagated east. Last I could see before the H-W plots went AWOL (they do pick their moments don't they?), GEFS was still not going with that, hence the renewed downward tendency it produces, but ECM was taking it nicely east with scope for the GWO 3-4 to follow relatively quickly.

All that seems particularly clear right now is that the situation is unusually complex. Good luck everyone .

[Ali1977]

2 minutes ago, Daniel* said:

“The next 12Z will be crucial” is a bit irksome the same has been said day on day, surely if it was so crucial we would have firmed up by now, if you see GFS it’s been correcting a lot more than ECM.

I’ve not had a look at GEFS but that’s good to hear^ 

After a few runs easing the cold spell I’m fully expecting a step back colder again , hoping yes but there’s still a good chance looking at some ENS.

[Man With Beard]

3 hours ago, Glacier Point said:

Yesterday I was sceptical. Even more so now.

GFS drop in angular momentum is now inching forward day by day sufficient to be taken notice of.

That ensures us in a GWO Phase 2 type scenario, cyclonic but still with the ridge tantilisingly to the north-east. The lack of any real depth of cold the crucial thing here.

After day 10, the lowering AAM trend should abate and steadily reverse. With the eastward progression of the tropical wave, westerly inertia should pick up signalling a movement in the GWO towards phases 3 and 4. Phase 4 being the phase in which a proper easterly could take root - but stress, that is some way off yet.

Hi GP - Hopefully a silly question!! You mention the GFS tendencies for angular momentum - does the ECM (/other models) show the same? 

[johnholmes]

15 minutes ago, Daniel* said:

“The next 12Z will be crucial” is a bit irksome the same has been said day on day, surely if it was so crucial we would have firmed up by now, if you see GFS it’s been correcting a lot more than ECM.

I’ve not had a look at GEFS but that’s good to hear^ 

That phrse,'... crucial' is getting tiresome. Why is it crucial, crucial to what? How many times have the models altered their predictions in the last 100 pages in this thread.

Why not wait and think about a whole run, think again, then post, instead of posting after every frame on every run.

That applies in my view to a fair number on here.

No apologies for making this post. It would be a good idea to compare like with like, show the differences, as some do, so we can all see why a particular model is so crucial on any particular run.

Just so everyone can do this the link below allows you to compare models over differing time scales.

https://www.netweather.tv/charts-and-data

Just have a look at all the data there, think about it and then post-please!

[Chesil View]

Just now, johnholmes said:

It's also the wrong thread!

Ho John appreciate your last two posts but how about giving your ex pro's view on possible developments next week

[karlos1983]

Your post could come across as a bit condescending if you don't mind me saying so john, some people might not have the time do look at different runs, compare and contrast, and I don't see the issue with commentary on each model run, that's how I learnt what little I know compared to the powers that be in here, and it's what this thread is about! Some are talented in here, others are just here to enjoy the output and add if they feel comfortable in doing so, your post would put a lot of people off from posting if you ask me.

Anyway, the 12z will be important imo because we are getting ever closer to this developing area of heights towards our North, the closer we get, the less room for improvement, one thing looks certain at this stage, it will feel bitter this weekend.

 

Edited January 4, 2018 by karlos1983

[Nick L]

X run is crucial if you want to drive yourself insane obsessing over tiny details at 5-10 days out. But if you take an objective overall view then it really isn't. 

[phil nw.]

In spite of all the variations within the operational runs leading up to this weekend and the colder period,the ens have been pretty solid on it being around 2/3 days duration before we lose the cold uppers from the east.

UK T48hrs    UK T120hrs 

 

Apart from the day 6 onwards of the ECM op runs we never really had any hope of a classic easterly this time.The Scandi block orientates poorly for us so we rely on a cold surface feed off the continent as we go further into mid-week with a mainly dry picture.

To be fair both ens  sets have been consistent with the approach of the Atlantic around 10/11th as the block just gives a little and we pick up that se flow..

These are the last 4 days ECM mean charts from the 00z runs all showing the same idea.

Currently it looks like we may see the approach of Atlantic fronts from the south west which could well become slow moving as energy is still shown to disrupt se into Europe after a quiet rather cold period.

There is no real sign of the block to the ne disappearing to the end of week 2 in the ens but whether that's a good thing from a snow lovers pov is doubtful if it remains as it is.We could be looking at a protracted spell of non-descript and quiet period in a no mans land between the block and a weaker Atlantic.

ECM/GFS Mean charts T240hrs.

I don't like this pattern- either we need the block to fight back and evolve into a full blown easterly- which is a big ask looking a the Vortex to our nw-or let's have stronger trough disruption se and create some interest that way.At least with the heights persisting to the ne there's a better chance of the second option currently!

 

[Timmytour]

I think the crucial runs depend on the timeframe of model divergence and what happens to make them diverge. As that time gets nearer into the reliable, then obviously what the models start to show get more crucial.

Of course it depends on what you regard the reliable to be.

And even then, with cross model agreement, no guarantees are involved beyond T-6 I'd say!  Even the development of today's system bringing strong winds to the southern half of the country was subject to a revision by Meto within 24 hours of arrival

[Frosty.]

Just flicking through the GEFS 6z perturbations, P5 grabbed my attention..

[Glacier Point]

44 minutes ago, Man With Beard said:

Hi GP - Hopefully a silly question!! You mention the GFS tendencies for angular momentum - does the ECM (/other models) show the same? 

I know WSI provide a GEFS based AAM / GWO product, and also provide EC based Hovmollers for zonal wind analysis and forecast, although I've not seen a GWO ECM product. It would be doable if you had the right data to work off. EC will show the same negative anomaly across 20N-30N as GEFS.

[mulzy]

43 minutes ago, ShortWaveHell said:

This is what Iv been doing myself @johnholmes without getting embroyeled in the thread , been comparing GFS 12z with 12z and likewise for ECM , the GFS updating every six hours seems more of a hindrance if you follow it every update rather than benefit ! Especially for a complete novice trying to learn like me ! 

I have never unsterstood that?  Why can’t we compare each and every run?  Seems truly bizarre.

[jvenge]

4 minutes ago, mulzy said:

I have never unsterstood that?  Why can’t we compare each and every run?  Seems truly bizarre.

I have a job and family. Dunno about anyone else...

 

[karyo]

1 minute ago, jvenge said:

I have a job and family. Dunno about anyone else...

 

Exactly! I may do what John says when I retire...

[Paul]

9 minutes ago, mulzy said:

I have never unsterstood that?  Why can’t we compare each and every run?  Seems truly bizarre.

Following every run is fine if they're viewed as part of the wider picture. But if you follow each run and read each update as gospel, you're going to be jumping around like a cat on a hot tin roof, particularly at longer range!

[SMU]

& were off ( IKON )

1-2 runs of sharpening should see some of the pessimists withdraw their negativity...

 

[Met4Cast]

20 minutes ago, mulzy said:

I have never unsterstood that?  Why can’t we compare each and every run?  Seems truly bizarre.

I agree - Comparing a run to something that's 24 hours out of date has never made sense to me either. Perhaps in the olden days when models were in their infancy it made sense but certainly not anymore. 

There's been a definitely shift towards the GFS but not completely. We've still got the block but it's definitely leaning more towards battleground snow (yet again, great if you live up North, rubbish for us snow starved Southerners) rather than a cold, convective Easterly which many of us were hoping for.

Still, it's only January.. 

[Duane S.]

10 minutes ago, Danny* said:

I agree - Comparing a run to something that's 24 hours out of date has never made sense to me either. Perhaps in the olden days when models were in their infancy it made sense but certainly not anymore. 

There's been a definitely shift towards the GFS but not completely. We've still got the block but it's definitely leaning more towards battleground snow (yet again, great if you live up North, rubbish for us snow starved Southerners) rather than a cold, convective Easterly which many of us were hoping for.

Still, it's only January.. 

I also agree. I have always found it strange, without really questioning it properly.

I understand there is different types of data fed in, depending on which run it is (06z 12z 18z 00z), but at the end of the day the most (IMO) important thing (for eventual accuracy) is for the starting data to be as accurate as possible? Therefore, surely, it is more accurate to do a run to run comparison, as opposed to matching a run that, yes has the same set of data fed in, but has starting data 24 hours out of date.

[SnowBallz]

7 minutes ago, Paul said:

Following every run is fine if they're viewed as part of the wider picture. But if you follow each run and read each update as gospel, you're going to be jumping around like a cat on a hot tin roof, particularly at longer range!

This.

For me, it's a case of micro and macro analysis; I personally find trends are easier (ha!) to identify and analyse when considered within a wider gamut, otherwise I find that you can become lost within the noise and data scatter of intra-run variance - something which I pay little close attention to.  There are various techniques which can be adopted and one of those is time benchmarking (comparing 12z to 12z, etc) 

Lastly, GFS gets a fair bashing on here, but what I would reinforce is that its output is absolutely considered within MO, albeit within the context of holistic NWP output; it isn't the case that it is ignored, more that its output is considered against other more locally sensitive products.  WRT such products, without dangling any carrots, there are signs of some confidence around a colder signal - but the timing of this remains less certain.  Again - and to reinforce the analytical approach - this is something of an emerging trend, and is probably better described as a moderation of confusion, following a period where no particular outcome held greater confidence.

I wouldn't go buying sledges just yet, mind.  It is vitally important to see whether such an aforementioned trend consolidates in future runs, or whether it remains rogue and undecided.  As a coldie, I am obviously favouring a colder outcome! 

As you were...

SB

[Tamara]

A fair bit of reference to a crossroads theme in various ways at the moment (relevant to the synoptic pattern) but if I may, yet again, refer back to a post in mid December, I think we see these themes being played out at the moment.

Posted December 15, 2017

.

This, to repeat yet again as in all other posts, is something of a make or break in terms of whether the atmosphere continues to go along the Pacific "destructive" Nina tropical forcing path (hence the evolution to a poleward Pacific ridge, westwards shift in the polar vortex to the NW, and ascendant heights to the NE) or the more traditional La Nina Pacific ridge (which pressurises the downstream vortex leading to increased polar jet flow upstream, and increased temperature gradient, and flatter faster northern arm jet flow over a de-amplified Atlantic ridge)

There is even a hybrid version of the two that is conceivable - implying some Atlantic trough disruption, heights attempting to rise to the NE, but with angular momentum still not sufficiently elevated relative to La Nina to allow the influence of the Canadian vortex bleeding too much energy into the Atlantic to back the cold west. I believe a few of the recent winters since the cold period up to 2012/13 have illustrated the latter two less favourable (for cold weather wishers) themes and which have been a bane to the majority of the seasonal audience of this thread.

In simple language it explains the mixed competing NWP messages that have been suggesting quite different things evolving - and means :

The difference between a La Nina Pacific ridge pattern hybrided by an opposing (almost El Nino like) tropical convection pattern which triggers poleward propagating rossby waves to amplify the tropospheric pattern and further perturb the polar vortex at higher latitude through +MT events

and,

The much more traditional La Nina rossby wave train which bolsters sub tropical ridges in both the Pacific and Atlantic, increases polar jet flow and an associated downside +NAO signature

-----------------------------------------------------------------------------------------------------------------------------------

Please note the bolded section of the post especially.

Phil's usual well set out post just now rather alludes to the possible no-mans land in the medium term I was suggesting as a possible option then (and as a holding pattern?)   and which I had mused might become in the eventual mix at the time of writing the highlighted post.

I am not the most gifted of computer technicians (*she definitely understates*) but with the appreciated help of @jvenge and @Catacol I also now have sight of the more updated AAM budgets, albeit still rather more lagged than would prefer for best analysis.

They do enough though to point to the 'hybrid option' highlighted back in mid December.

Phase 4 GWO destination still most certainly a plausible road ahead, and all that entails in terms of both tropospheric (and ultimately stratospheric changes) but I am only too conscious that a relatively small section of participants who inhabit the 'fast food culture' of this thread doesn't like or tolerate 'jam tomorrow' propositions.

However this is where a reconciliation of the sober science (exploring the options that various jet stream parameters suggest with the help of the GWO/AAM budgets) presents something of a 'frictional torque' of its own set against the frenzy for it to snow (irrespective of the science) The signals can only meet a specific synoptic response - and this simply cannot always be the most preferred one. At least not the preferred one in this case straightaway - notwithstanding it is hardly the worst immediate outlook anyway!

On the subject of the much more relevant tropical frictional torque- and wrt the analysis the other day in terms of a +ve spike in FT c/o eastward moving tropical convection helping to change the mid latitude profile of amplification into a higher latitude one, then the relative frailties of that not being present at the moment are exposed in terms of polar jet energy vs the need for the extra easterly inertia c/o La Nina forcing to be subdued

This piece of the jigsaw is definitely not completed. Take yet another look at the GEFS portrayal of -ve anomalies around 20N.

The removal of these Nina easterly wind anomalies is brought about precisely c/o the mechanism that rising AAM tendency is associated with rising frictional torque across the tropics. This, to start the process of reversing easterly wind anomalies, and scrub them out with poleward propagating amplifying westerlies.

This marks the start of the process of weakening the La Nina sub tropical ridges which are underpinned by these easterlies and a stubborn tendency for the NAO to be too positive to allow more than a half way process of trough disruption amidst too much cyclonicity tagged to the Canadian vortex being too dominant,  and with too much residual polar jet flow to stabilise Scandinavian High pressure at the correct orientation, latitude and distance to advect deeper cold air.

The elevation of AAM tendency through the advancing tropical forcing to get the GWO to Phase 4 will suffice to get the most stable Scandinavian high, but to take things up another level and allow classic undercutting of low pressure an a fully fledged blocking high, then both the full effects of poleward advancing +AAM anomalies (through a process of +ve frictional torque, leading to +ve MT) is required at the same time as an external stratospheric pathway is created by getting the right higher latitude height anomaly configuration. This in turn leading to SSW potential and fullest submission of the Canadian vortex much more likely.

However, its only possible to work with the cloth that is available. That said, more cloth may well became available during the short term period of interest to keep things occupied - and ought soon enough to allow the short term modelling output thread a lease of life

 

 

 

Edited January 4, 2018 by Tamara

[Johnp]

4 minutes ago, ShortWaveHell said:

This is what Iv been doing but is it beneficial to a beginner ? If there’s a consensus another way im all ears and eyes ? 

Sorry but I think that's a nonsense way of viewing models (just my opinion)!

Any benefit of comparing against a 24 hr old run are nullified by the fact that the newer runs have more recent data. It's surely better to compare to the preceding run when looking at model trends.

Edit - or basically what s4lancia said ?

Edited January 4, 2018 by Johnp

[Mucka]

Lot's of talk about this model and that model being correct, we all get caught up in it and tend to athropomorphise the models.

When I posted the 24h like for like comparisons last night it was clear that the Euros had moved as much toward GFS as GFS moved toward the Euros over the period.

That is the reality, none of the models had it nailed it seems. GFS was too progressive as Steve predicted but ECM and GEM were way too optimistic with undercut and potential retrogression.

This has always been a finely balanced setup and one ALL models struggle to deal with, UKMO has probably been the most consistent but then it only goes out to 144 so the errors can never be as obvious or big as models that go out much further.

Below are the last 4 00z runs from ECM and then GFS for Dec 10th.

 

What can we learn?

1. GFS is over progressive in these situations, Steve is quite right to point that out, it's just a fact.

2. ECM is very erratic, swinging from very progressive to very blocked on whim.

3. Contrast the 2nd and 3rd day ECM with today's, it is a huge back-down today. (Or rogue run)

4. Contrast the 2nd and 3rd day GFS with today's. changes but more incremental and consistent than ECM and never showing a nicely placed and oriented blocking high with Easterly feed.

5. This morning they are closer than any other run, is it really true the GFS has moved so far to ECM and not the other way around or somewhere in the middle?

conclusion?

Stick with UKMO in these situations, don't go beyond 120/144 absolute max, and assume where there is large model divergence neither model is completely correct or completely wrong. 

Of course I won't follow that advice and nor should you, as sensible as it is.

This forum is ultimately about us all enjoying our pastime and sticking our selective necks on the chopping block from time to time by allying ourselves with a particular model or outcome and then spending all our time trying to vindicate that position when it all goes TU.

It is the rampers and the Debby downers that keep this place fun (and sometimes not so fun) and there are so many knowledgeable posters who share their wisdom and perhaps more importantly enthusiasm here.

Finally, this is still not resolved. It is still possible we will get the cold air we desire across the country to permit snow but we may have to take the scenic route rather than hopping on the Siberian Express.

So long as we have blocking so close to the East, it doesn't take much of a redistribution of energy to change things for the worse or better for our little Island.

We could still the high flatten quite quickly if more energy is pushed NE than currently modelled but conversely we can get the block further NW with a much better feed if the models push more energy SE.

It is still finely balanced even if this mornings output has generally gone the wrong way IMO, with a reluctance for the models to force energy SE and undercut becoming an obstinate signal as mentioned last night.
In that regard GFS is currently being vindicated as it never saw the undercut. Caveat 1. Even though if it happens that may be as much by luck and its natural progressive bias than judgement. Caveat 2. Is it still a bias if it proves correct?

Lastly, UKMO.

That may not offer immediate winter joy even though it is chilly and more settled but at least it offers the hope of something better down the line.

The 00z pattern probably would be forced SE but either more amplification upstream day 8+  with the possibility of an Atlantic ridge reinforcing any high or a better jet profile day 6/7 (more energy SE) and things could quickly change in our favour.

 

 

 

 

Edited January 4, 2018 by Mucka

[JBMWeatherForever]

I for one am looking forward to battleground snow followed by a prolonged spell of cold from the east. Unfortunately though I don't think the pessimists will withdraw their negativity, they will just move it on to something else to be pessimistic and negative about !

[jvenge]

4 minutes ago, ShortWaveHell said:

This is what Iv been doing but is it beneficial to a beginner ? If there’s a consensus another way im all ears and eyes ? 

In short, it is unlikely you can pickup real trends in 6 hourly runs, especially when people are looking 6+ days away for those trends, when you take into account mathematical uncertainties vs the benefit of an alleged up to date data source. I think the suggest is more from a sanity point of view than anything else.

Exceptions of course for high res storm tracking and precipitation forecasts.

[comet]

I must admit to being a little more optimistic after seeing the 06z and its ens  so I decided to re-visit the ERSL composite strat charts and they to give more confidence of  a more amplified pattern developing in the mid range. Further downwelling still instigated by strong wave break in early December over Alaska, northern Canada and now a wave 2 type response from Central, northern Europe extending north should make itself more visible in the troposphere in the next 10 to 15 days so do not be surprised to see the emergence of a more amplified stronger block into northern Scandinavia with much more of a possibility of undercutting.

Here's hoping for a much better 12z from the gfs.