Model Output Discussion 1st January 2014-06z onwards.

[Man With Beard]

Couple of points from 00Zs:

1. Uppers not quite as important when the feed is off the continent (though snow cover over low countries would be handy)

2. Small changes to model output will decide whether it snows or not in any location. While the charts look like they do this morning, no-one (north or south) will have any great confidence in whether they will see snow or not until T72. Quite frankly, at the moment it doesn't really matter whether the charts say snow over your house or 300 miles away - they will probably show something different tomorrow!

[shedhead]

Couple of points from 00Zs:

1. Uppers not quite as important when the feed is off the continent (though snow cover over low countries would be handy)

2. Small changes to model output will decide whether it snows or not in any location. While the charts look like they do this morning, no-one (north or south) will have any great confidence in whether they will see snow or not until T72. Quite frankly, at the moment it doesn't really matter whether the charts say snow over your house or 300 miles away - they will probably show something different tomorrow!

More like in another 90 mins...

[chionomaniac]

Well i've lost track in recent winters of the number of times i've checked the forum first thing in the morning thinking ohh dear there must have been some pretty serious downgrades only to take a look at the models for myself and realise that in fact there has been very little change since yesterday and all the deterministic runs are displaying this morning are the variations within the ensembles that we have been seeing for the last 36 hours.

 

Deterministic modelling is not as good as it was at say 0z yesterday but I think some people need to try and get their head around what NWP is trying to model here. Attempting to model the physics of two streams of jet energy travelling in two different directions running in to a blocking high is a ridiculously complex equation. There were various members yesterday trying to impart wisdom from years past to tell everyone that whilst face value model output was looking exceptionally good it would be some days yet before we would be able to determine the final outcome. We see models struggling enough to calculate fairly straightforward situations a week in advance, never mind something like this. When Ian F said yesterday that in terms of ensembles it was essentially a 50/50 split (with just a tad more on to the cooler side of the fence) he wasn't saying that for the sake of it, that really is what you get with situations like this, particularly against the background of how strong the jet stream has been.

 

All I can say is it's a good job I am not a newer member attempting to decipher what on earth is going on because the conflicting messages this morning have been something else. What I would say to any newer members for now is this - wait until the weekend before drawing any major conclusions. The only certainty for now is that temperatures at the surface are going to drop in comparison to recent weeks.

 

On the subject of the actual outputs yes the GEFS ensembles are certainly not as good as yesterday but two crucial points to make here. First of all once again both of the higher resolution outputs (det. and control) are still firmly in the colder camp - when you're dealing with splits in jet energy those are going to be the more likely to resolve things correctly, though even within these as we have seen with the ECMWF this morning, expect some variation through the next few days on how much energy travels north/south. And finally, whilst we're on the subject of resolution, please correct me if I'm wrong but last time I checked the GEFS were running at a resolution of 55km out to 192? Well when you consider that, for example, the NAVGEM has an operational resolution of 35km, and how very little we take it's output seriously, you start to see the potential issues (in the interest of balance it should be noted that the NAVGEM issues could also be model physics related).

 

But overall the point here is not to take any output you see as gospel until towards the weekend when we should hopefully start to see some consistency between outputs.

 

SK

Would definitely agree here Kriss. The overall pattern hasn't changed - over than normal variations expected. I'm just happy to see the Atlantic stop for a week or too and hopefully whilst it does, we see further attacks on the vortex.

Edited January 9, 2014 by chionomaniac

[weather eater]

Well we are starting to see a familiar story this morning, some great runs at days 5,6,7 then the tweaking and the fine tuning start to kick in and all off a sudden things don’t look as good as they did, it all starts to get a bit jam tomorrowish. We need to get the Stella charts into the 72hr range where the tweaks are small. What we have now is a slower evolution, however, the charts for days 7,8,9 are all also going to get tweaked considerably, so I wouldn’t bet on the idea that it’s just a case of waiting a bit longer and that a full blown easterly is still in the bag but just not as quickly.

 

Having said all that the GEM is still good this morning and we still have time for the GFS and ECM to make a correction back towards the kind of charts we saw yesterday morning and it could be argued that the models have just reverted back to a slower evolution, so we just got a bit spoiled yesterday, let’s hope so.

[IDO]

The control for the ECM appears to have flipped on this morning's run with regard to an early easterly. Last night:  

 

Changes for this morning compared to last night (change in temp): T162:  

 

It and the op IMO seemed to be too progressive with the pattern change and this morning's op to me looks more like many of the GEFS from yesterday.

 

The change in the mean from last night compared to the 0z is a good trend: T192:  

 

So cooler-colder weather is on the way, but still much to be decided with regard to surface conditions UK wide.

Edited January 9, 2014 by i'm dreaming of...

[SMU]

Well how would I describe this mornings runs -

If yesterdays suites were the verve cliquet then todays runs must be something like lansen!

Lets hope we dont end up with co ops own champagne-

Both the GFS & ECM thrown a bit more energy NE and keep that coldest air to the NE although still fantastic for ne england and E Scotland.

However at least to offset thst somewhat all 3 models offer cold out to and past day 10 - all bit by varying routes with the risk of mild sir actually higher from the SE as opposed to the traditional west.

UKMO to come, then onto another set of 12s although saturday will be the real landing zone for accuracy for todays 120 & 144 will be 72 & 96 with minimal wiggle room.

The prospect there for lots of snow still with fortune swinging to the NE of england this morning - however that can quite easily swing back tonight .......

S

Edited January 9, 2014 by Steve Murr

[garbagebags]

Can I ask, I see quite a lot of people showing the GEM model over the past few days. Whereas only a month ago it was ECM, GFS and MET that people looked at. Has the GEM become a lot better recently or is it just showing what people want to see? Serious question.

[Cecil Hogmanay Melchett]

Would definitely agree here Kriss. The overall pattern hasn't changed - over than normal variations expected. I'm just happy to see the Atlantic stop for a week or too and hopefully whilst it does, we see further attacks on the vortex.

Agreed C- the trouble is the changes (subtle or not)are moving in the wrong direction for cold and are currently making the all the difference between a wet landscape and a white landscape.At this distance it will no doubt change again but members commenting on the current output is fine,surely.

Hopefully more swings in the cold direction will be forthcoming..

[Troubleatmill]

has anyone else noticed the issue with the 00z ensembles (GFS), certainly the operational doesn't tie in?

[weather eater]

Can I ask, I see quite a lot of people showing the GEM model over the past few days. Whereas only a month ago it was ECM, GFS and MET that people looked at. Has the GEM become a lot better recently or is it just showing what people want to see? Serious question.

The GEM is a decent model and I’m sure Steve can point you at some excellent verification stats, however, yes of course people are going to post up GEM charts when it is showing the best evolution, who can blame them.

[shedhead]

Can I ask, I see quite a lot of people showing the GEM model over the past few days. Whereas only a month ago it was ECM, GFS and MET that people looked at. Has the GEM become a lot better recently or is it just showing what people want to see? Serious question.

I think it's a bit of both to be honest, but to my mind the GEM is not consistently good enough despite appearing extremely good from time to time - the difficulty is picking out those 'extremerly good phases, which is one of the main reasons why I only give it a cursory glance each run.  

Edited January 9, 2014 by Paul
Removed pointless sarcasm

[Cecil Hogmanay Melchett]

has anyone else noticed the issue with the 00z ensembles (GFS), certainly the operational doesn't tie in?

Looks about right to me? -6/-7 850's on the 17th?

[draztik]

Can I ask, I see quite a lot of people showing the GEM model over the past few days. Whereas only a month ago it was ECM, GFS and MET that people looked at. Has the GEM become a lot better recently or is it just showing what people want to see? Serious question.

can of worms!!you have a point, but lets wait until later before we deconstruct what the models are telling us. We will know more after the midday runs! But alarm bells should be a' ringin!

[shedhead]

has anyone else noticed the issue with the 00z ensembles (GFS), certainly the operational doesn't tie in?

If you look closely at the Op line it splits into 2 at 00hrs on 15th.

 

Edited January 9, 2014 by shedhead

[The Eagle]

00z ECM ensemble mean is an improvement on the 12z. A silver lining.

 

 

168 850 average

 

Edited January 9, 2014 by The Eagle

[Frosty.]

Well how would I describe this mornings runs -

Lets hope we dont end up with co ops own champagne-

 

no sparkling blue nun please deep blue coloured uk weather charts will do nicely

 

Edited January 9, 2014 by Frosty.

[Jason T]

which model is that jason?

 

NCEP-ENS-GLB

[Cecil Hogmanay Melchett]

ECM 0z ensembles for Reading are underwhelming hand on heart.(They are actually  coldest sat/sun).

They look coldish but certainly not showing any sort of Beasterly!

http://oi40.tinypic.com/156fp93.jpg

Edited January 9, 2014 by Cecil Hogmanay Melchett

[chionomaniac]

Agreed C- the trouble is the changes (subtle or not)are moving in the wrong direction for cold and are currently making the all the difference between a wet landscape and a white landscape.At this distance it will no doubt change again but members commenting on the current output is fine,surely.

Hopefully more swings in the cold direction will be forthcoming..

That's always the problem with whiteout runs. That then becomes the minimum expectation rather than a blend of all the output. Even some of us old timers on here forget that - I know that I always use the best run as the benchmark but at least I have learnt to temper any disappointment with realism.

[johnholmes]

for what it is worth

 

Thur 9jan

Ec-gfs

No ec this morning (nor last evening), gfs shows upper low beneath ridge over n’ern Scandinavia linking into ridge off eastern seaboard and over Greenland

Upper trough spacings-4-suggests some degree of longevity

All anomaly charts=3 I use show similar so upper pattern shown is highly probable to be as it shows from say about 16/17 jan out towards month end, actually only goes out to 22 at the moment on noaa but seems unlikely any major change in pattern for several days after this date

links belwo

http://www.meteo.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

 

http://www.cpc.ncep.noaa.gov/products/predictions/814day/500mb.php

 

my own view is the majority of you are suffering the roller coaster effects of the synoptic model minute examination of every bit of every run. Step back, post T+144 check like run with like run, below 120h certainly look at each run but take a deep breath, stand back and THINK about what the models, of whatever type, are showing as a PATTERN. Have they ever consistently, note the word, shown deep bitter cold and piles of snow? What is the average pattern they have been suggesting? What are the probabilities of one version or the other. Some of you have mathematical and scientific training so why not use it when looking at the models?

Edited January 9, 2014 by johnholmes

[shaky]

Much better ecm mean this morning compared to the 12z!!colder air further south and West and snowline further south aswell compared to the op!!have a feeling we gona have an upgrade this evening!!

[snowking]

These certainly don't look too disappointing and leave us pretty much where we were yesterday:

 

 

It suggests that at least 50% of the EPS members want to continue sliding enough energy South-Eastwards under the block to give us something of interest, and in all likelihood for that to be a mean chart (accounting for offset from less keen members) possibly a little more than 50%.

 

So, really as we were then from the ECMWF Ensembles - on the colder side of the fence but not convincingly so just yet.

 

SK

 

EDIT:

 

Dutch graphs tell the story well too:

 

 

Encouragingly dew points too (considering we are looking at a shallow Easterly flow) - 80% are sub zero from Thursday to Saturday

 

 

And still strong convergence on an Easterly wind:

 

Edited January 9, 2014 by snowking

[The Eagle]

ECM op run was certainly on the mild side of the ensembles by a good 4c or more in most places. Potentially good news for later on.

 

 

 

 

 

Edited January 9, 2014 by The Eagle

[Man With Beard]

ECM 0z ensembles for Reading are underwhelming hand on heart.(They are actually  coldest sat/sun).They look coldish but certainly not showing any sort of Beasterly!http://oi40.tinypic.com/156fp93.jpg

As others have noted, meteociel shows the mean is far colder than the op. -4 uppers for most from next Wednesday until the Sunday.

[Frosty.]

Holy Cow..this is snow joke guys, the Ecm 00z ensemble mean is FAB-U-LUS. absolute snow fest with ice days galore from this, heck I need a lie down now If this carries on, I might have to ramp this up

Edited January 9, 2014 by Frosty.