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

[jvenge]

5 minutes ago, Paul_1978 said:

So which version of the GFS should we put credence in? The normal op or the Para? Does anyone know?

The ensembles ;-)

 

[feb1991blizzard]

1 minute ago, Paul_1978 said:

So which version of the GFS should we put credence in? The normal op or the Para? Does anyone know?

Too early to know yet, best way to look at parallel's is to just use them as another model (a new one), another assistance in forecasting, better to have them showing the outcome you want than not but caution until more knowledge is gleaned (stats) plus also bare in mind that they can be making alterations each day for all we know.

[karlos1983]

9 minutes ago, Paul_1978 said:

So which version of the GFS should we put credence in? The normal op or the Para? Does anyone know?

Para 100% 

to be honest though, I'm just happy that we do appear to be getting some agreement that heights are due to rise somewhere close to the uk, with hints starting to appear that pressure could well fall over Europe. A much more positive last 12 hours or so. 

Edited December 6, 2016 by karlos1983

[Singularity]

22 minutes ago, Catacol said:

Hehe - there we go - every cloud has a silver lining! Let's put the gloom of the last few days into the dustbin of history.

Talking of silver clouds - this forecast is good news for all snow hunters. MJO progression back on track... watch for more amplified runs hitting the ops over the next few days.

 

Remember all the MJO forecast sending us spinning down low into 2? Now back around into a higher AAM state. Oh ye of little faith. If the models don't get an initial handle on the most important drivers then op runs have NO chance of being accurate. Human input on this has been superior. Just need the spike to end up a little higher than this below.... probably explains MetO undercertainty in terms of 2 possible outcomes later in the month.

 

You can probably imagine how much I wanted to jump onto this yesterday - but it's the most progressive set of ensembles so I'm waiting to see if ECMF move toward such a quick pace of events today.

You know, it could be inferred that yesterday's 12z GFS was in the slow end of the spread shown there, the 00z the fast end and the 06z somewhere in between.

I think, though, the consensus expectation among pros has been for more in the way of 6/7 activity prior to 8/1 than is shown above. Though either way we have the Indo-Pacific forcing that GP has been keen on in recent weeks due to the link with cross-polar ridge from the Pacific to Atlantic sectors. Good signs of that in recent GFS runs too.

[Tamara]

Last Thursday:

there is also quite a significant risk that the models are both overplaying climatology in terms of VI phasing vs troposphere>stratospheric feedbacks and also too much traction being applied to those (short term) masked tropical signals. That is, too much emphasis on a backtracking Phase 1 MJO signal (as a result of Indian cyclone activity) which will overcook low angular momentum and over-flatten the pattern. (A point that I know I have stressed on my own posts as something potentially coming from at least a couple of weeks back)

The models will slam on the brakes, fast, and back up the road pdq if that is the case, Especially bearing in mind MJO forecasts become highly erratic beyond 5 to 7 days

Sunday:

That low frequency amplification signal has not evaporated courtesy of transitory overriding signal in the I/O. Its still the case that models will be vulnerable in the coming days to overdo the present signals and possibly backtrack to some extent as that low frequency feedback signal in the tropics comes more into returned focus. What that means is a re-amplification of the pattern as the MJO signal regains traction eastwards.  Whilst the current extended ensembles are blowing coolish on only a modest descent back of temperatures from a very mild spell, these will be sensitive/responsive to changes either way in these signals.

Today:

1) Inconsistent and volatile models - both now, and as signposted this morning by others also for the longer range. (But both ECM and GloSea5 both showing though that there is no reason for anyone to prematurely extrapolate any westerly pattern out in terms of several weeks and months).

2) The message though has remained the same since last Thursday.   Its very much the case that less reaction to the models, and every model suite, is most helpful, and even more helpful to pay attention instead to the signals that are causing the models to oscillate. . In that sense I am maybe less "model-centric" than the majority and (try) to approach this subject from a diagnostic point of view first, and then see how NWP sits next to that. That is not meant in any derogatory way, and to imply that one way of approach  is right or wrong - simply meant to be well intentioned and helpful as hopefully always the case

3) Yes, its very much true that seemingly trivial events far away in the I/O and other tropical waters can result in downstream uncertainty such as we have. Whether there is eastward propagation of these signals has direct bearing on the rossby wave trains which amplify, or otherwise the pattern, and whether they cause atmospheric angular momentum to rise in the former instance, or fall in the second instance

The recent and forthcoming sequences of weather patterns are proving that the tropical and extra tropical atmospheric budget does have a direct influence  - unfortunately from a net weather enthusiasts hopes and wishes point of view that influence can work both for and against those wishes. But after all, if it only appeared to work one way - then it wouldn't be sound science would it? And good reason to be sceptical of the GSDM methodology

I continue to be encouraged by the state of the atmosphere/ocean disconnect and its on-going seasonal impacts. The imminent changes in the weather pattern are directly correlated to the downward correction in angular momentum which is about to occur because of the shorter term unfavourable tropical convection signal. But the models are already showing fickle switches and signals in terms of the "noise" signal vs the low frequency one. The noise one flattens and increases the polar jet and gives the vortex a lifeline - but the low frequency one is waiting to flip that pattern to a more blocked and amplified pattern.

4) Scandinavian Highs can occur during periods where angular momentum dips, but crucially just ahead of significant rises that can initiate cold spells in winter. With default +AAM background inertia c/o of the low frequency signal, the extra energy that goes into the northern arm of the jet under -ve AAM can split as momentum then rises - with one arm heading up around the pole, and then "horseshoeing" around a developing ridge (as hinted in some latest modelling) while the other split heads into the southern stream and provides some underpinning of that ridge.

This is a good demonstration of type of scenario where a lobe of polar vortex over NE Canada and Western Greenland need be no obstacle to blocking developing to the NE and cold air advection occurring - the vorticity tends to retrogress under the above described scenario anyway.  Disruption of that vortex energy occurs, especially if the upstream jet split is weighted appropriately. However, in these circumstances, and in contrary to what one might think, some initial polar jet energy is required for the process to begin.

A lot of uncertainties obviously how this all might play out so far in advance. But, to risk overdoing the repetition, especially if the models have overdone the latest signal, then the momentum might well grow over the coming days to enlarge the longer range signal and increase the current indistinct temperature signals downwards. But then again, I would approach this from a pro-active pre-signalling perspective and not a reactive one each Monday and Thursday.

5) The low frequency signal coming into increasing range will engage a jump in frictional torque (think of these in very simplistic terms as differences of pressure in the global longitudinal pattern) .

When torques drop they indicate falls in angular momentum as easterly trade wind momentum increases and sub tropical ridges develop on the northern flank of those easterlies . When they rise however, they indicate poleward propagation of rossby waves moving from upstream. The further east these get, the greater the amplification ultimately in our sector, and the more that amplification can take advantage of favourable splits in the polar vortex to become high pressure features at a high enough latitude to encourage greater cold air advection.

This type of whole process does not happen overnight however, and is one to view through into theheart of December and should be one to enjoy with interest however things pan out, as long as its audience can keep reading between the lines and keep rolling with any shorter term punches that the models suggest.

@karlos1983  Thanks for your comment. I cause my own brain to hurt - imagine how that feels!

Edit: The other factors relevant in terms of the unstable arctic profile and Kara sea ice were discussed in huge detail in a previous post. Plenty of page space already occupied here...

Edited December 6, 2016 by Tamara

[Frosty.]

The Gfs 6z op really is a cracker if it's chilly settled seasonal weather you are looking for as we approach Christmas!

Edited December 6, 2016 by Frosty.

[booferking]

GFS 06z Ens 8 brings easterly in as early as Day 7/8.

Edited December 6, 2016 by booferking

[karlos1983]

21 minutes ago, Tamara said:

 

This type of whole process does not happen overnight however, and is one to view through into theheart of December and should be one to enjoy with interest however things pan out, as long as its audience can keep reading between the lines and keep rolling with any shorter term punches that the models suggest.

 

Truly fascinating post Tamara. Your will to share your thoughts and knowledge is admirable. My brain hurts reading your post, I have to read it a few times. I'm under no illusion as to how much time you must put into studying the signals to come to these explanations. Thank you so very much. I think my head would actually implode if I was to even attempt going that in depth!  

Sounds like our fun will certainly continue, if model watching is your thing!

[No Balls Like Snow Balls]

 

[Trom]

1 hour ago, IM NOT REALLY HERE said:

Can somebody please explain bimodal to me? 

The mode is a way of describing a distribution of data in this case ensemble runs.

mode is the most frequently occurring observation.  Unlike means and medians data can have more than one mode.  Bimodal simply means two modes in other words their are two observations occurring with the same frequency.

In the context of Ian's post it would mean there are two distinct clusters in the ensembles.

[karlos1983]

All of a sudden it feels like the pendulum is beginning to swing back in our favour somewhat (if it's cold you like). Feet firmly on the ground, but from the 00z, 06z and other updates, my optimism has risen! Let's hope the 12z keeps the theme going.

Edited December 6, 2016 by karlos1983

[Snowman.]

Early days but the signal is still there.

Heights north, lower heights in Europe. Now get it to T0 then ill be happy.

Edited December 6, 2016 by SN0WM4N

[knocker]

8 minutes ago, No Balls Like Snow Balls said:

 

But one could equally say  hints at heading into phase 1

Edited December 6, 2016 by knocker

[weirpig]

Will The ensembles paint a lovely picture going forward most have blocking   and a few have a bitter easterly setting in.  Hopefully something to build on.

[Snowman.]

2 minutes ago, knocker said:

But one could equally say  hints at heading into phase 1

Potential 8 aswell 

[karlos1983]

ECM ensembles appear to show a steep drop off at the end of the run! With few heading the opposite direction.

 

GFS ensembles aren't without interest either. Some decidedly cold runs in there! 

 

Edited December 6, 2016 by karlos1983

[Summer Sun]

A continued trend for things to turn colder as Christmas approaches with some of the ensembles heading towards the freezer

[Frosty.]

3 minutes ago, Summer Sun said:

A continued trend for things to turn colder as Christmas approaches with some of the ensembles heading towards the freezer

Things are looking up..or looking down as far as temperatures are concerned beyond the mild spell!

Edited December 6, 2016 by Frosty.

[Methuselah]

My favourite 'silver lining' is the obvious one: irrespective of what the models, the ensembles, the experts or the MJO forecasts suggest, the weather will do whatever it'll do...If I choose not to view a single model run between now and Christmas (hardly likely, I know!) the weather, on the day, will be exactly the same should I torture myself twice each day?

Aren't we gluttons for punishment?

[Snowman.]

Clusters http://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=euro&MODELL=gefs&MODELLTYP=2&BASE=-&VAR=cslp&HH=372&ZOOM=0&ARCHIV=0&RES=0&WMO=&PERIOD= 

 

[knocker]

Forecast temperature plunge continues apace

 

[karlos1983]

2 minutes ago, Ed Stone said:

My favourite 'silver lining' is the obvious one: irrespective of what the models, the ensembles, the experts or the MJO forecasts suggest, the weather will do whatever it'll do...If I choose not to view a single model run between now and Christmas (hardly likely, I know!) the weather, on the day, will be exactly the same should I torture myself twice each day?

Aren't we gluttons for punishment?

Yuh don't say 

inspiring stuff Ed 

Edited December 6, 2016 by karlos1983

[Frosty.]

4 minutes ago, Ed Stone said:

Aren't we gluttons for punishment?

Yes we are Ed but its all in the thrill of the chase, it didn't take long for the encouraging signs to return, hopefully we are on course for a cold christmas and new year, would make a nice change!

Edited December 6, 2016 by Frosty.

[fat chad]

Some good and clear posts this morning. To all those who have contributed to this forum with good explanation's of what they mean, Thank you.

[Methuselah]

2 minutes ago, Frosty. said:

Yes we are Ed but its all in the thrill of the chase, it didn't take long for the encouraging signs to return, hopefully we are on course for a cold christmas and new year, would make a nice change!

Of course. And I'll be torturing myself just the same as everybody else!