Model Output Discussion October 2012

[Methuselah]

No need to apologise Frosty, ineed it is probably me that should do so.

see the part of a para I've quoted from the link Steve gave above

Avoid the 06Z and 18Z runs when 00 or 12 is available; in the U.S. these runs don't include the weather balloon network data (balloons are only sent up twice per day), and are therefore radically different and more likely to have bias.

I have not seen anything from NOAA that agrees with this view but would be very interested if anyone has anything?

Until then my apologies re missing data Frosty.

I've never seen anything that I would class as 'official'. So, until I do, I'll do no more than keep it in the back of my mind...Are the 'bin-worthy' runs really that much less accurate?

[UV-RAY]

I've never seen anything that I would class as 'official'. So, until I do, I'll do no more than keep it in the back of my mind...Are the 'bin-worthy' runs really that much less accurate?

Certainly during winter it is when critical data from polar regions is amiss.

[AWD]

Another things to bear in mind is the steep temperature gradients that could form over the UK in the coming week or two.

Fortunetly the 06z run keeps things settled enough that it wouldn't really cause an issues, but if subsequent runs show unsettled conditions closer to the UK, or indeed, that trough to the SW of the UK ends up closer than predicted on this run, then some very high rainfall totals could be possible and with ground already saturated with recent rains, it wouldn't take much for more flooding to occur.

You can see how close the ThetaE values are over the UK, with vast differences from north to south.

Something else to keep an eye on IMO.

[SMU]

I've never seen anything that I would class as 'official'. So, until I do, I'll do no more than keep it in the back of my mind...Are the 'bin-worthy' runs really that much less accurate?

twice a day

http://www.wrh.noaa....ntroduction.php 00Z GMT & 12pm GMT to capture data for the 00z run that initialises at 3:30 AM & 3:30 PM

S

Edited October 18, 2012 by Steve Murr

[johnholmes]

Certainly during winter it is when critical data from polar regions is amiss.

I would not use the word 'certainly' perhaps IF assuming there is something missing in a critical area. It amuses me to read that 'data is missing, its crucial etc' but one never sees this argument used in summer, nor indeed when a milder weather pattern is what all the models are suggesting?

Picking up on this idea of the weather balloons are crucial - of course they are. Until GFS decided (NOAA) to go ahead with 4x daily model runs NO OTHER Met centre ever considered this. UK and ECMWF did 2x on 00 and 12z with UK doing a supplementary at 06 and 18z.e ven then, at least when I was involved wind data was routinely done at 06 and 18z across the UK, less so elsewhere.

Maybe all of the centres should go back to just doing 2 runs then the queries about data missing might disappear although I doubt it? Do remember that vital data such as satellites and aircraft reports are available around both those supposed data sparse times of 06 and 18z.

Edited October 18, 2012 by johnholmes

[BLAST FROM THE PAST]

What has to be highlighted is the 'quite significant' difference bwteen 00z and 06z re the t99 SW forming at southern tip of Greenland. Now I can see that getting picked up on gradually over numerous runs but such a stark difference on the very next run?

BFTP

[johnholmes]

Fred how many times does GFS do this, someone should count them up?

Go to Wx Online folks and see how the main models can and do vary over several days for the same date/time ahead.

After I have had some lunch I will TRY and be as objective as I can about whether there is likely to be a cold outbreak next week or not and show reasons one way or the other.

Folks do remember I do enjoy, in moderation cold and snow

.

The 3 main synoptic models

500mb anomaly charts

NAEFS outputs

AO, NAO and PNA

MJO

All shown with my analysis of them and an overall suggestion of what the weather pattern next week looks most likely to be.

I'll do it as a pdf so as not to clutter up the MOD thread and you can either ignore it or read it.

Edited October 18, 2012 by johnholmes

[Liam J]

The GFS 6z appears to have fallen more in line with the 00z runs from the UKMO & ECM - So this would then lead me to perhaps consider that the GFS is performing a backtrack of sorts towards the euro models, rather than just blaming missing data for the output change? (which could be the cause)

Can't wait to set eyes on the 12z's this evening.

Edited October 18, 2012 by Liam J

[Methuselah]

Guys, if there really are data issues with certain model runs, and I'm not saying that there aren't, why is it only pointed-out at times when projected Arctic Blasts look like never materializing? Surely any real issues would cause problems with every type of weather and throughout every season ...

Let's just see what comes our way. And, hopefully, enjoy it when it does?

[Snow and storms]

I wasn't sure where to post this but im looking at the GFS charts and I select one as Prev run and one as Current run. Now sometimes, the 'Prev' chat is being updated and the current chart is showing??

The chart with 'current' is surely the most up to date or am I being stupid?

Also could someone please clear up for me the whole 06z 12z 18z thing? I don't understand it and would really like to look out the latest updates. I saw people on here posting about the 06z just after lunchtime?

Thanks in advance

[syed2878]

hi guys if i can make a sujestion and sorry i know this is off topic but i think it would be a good idea to post the daily prognostic discussions here too.

[Gael_Force]

There must be thousands of flights providing data on an ongoing basis throughout a global 24 hour period. It may be that the 06Z is short of data from specific regions due to no night flights in certain areas and no new sonde data but I feel this is a red herring used when things do not go to plan. Big differences in the 06 qnd 12 runs yesterday - let's see if that remains the case today.

[Liam J]

I wasn't sure where to post this but im looking at the GFS charts and I select one as Prev run and one as Current run. Now sometimes, the 'Prev' chat is being updated and the current chart is showing??

The chart with 'current' is surely the most up to date or am I being stupid?

Also could someone please clear up for me the whole 06z 12z 18z thing? I don't understand it and would really like to look out the latest updates. I saw people on here posting about the 06z just after lunchtime?

Thanks in advance

The GFS model is programmed to produce four runs each day - 00z (12am) 06z (6am) 12z (12pm) 18z (6pm) So when you see, for example - 'GFS 6z run' mentioned in posts this refers to the 6am set of charts produced by the GFS model. The charts generally become available around 4.5 hours after the run has initialised (3.5 hours when the clocks go back) I think?

The latest charts can be viewed here on Netweather - http://www.netweathe...ction=nwdc;sess=

Or another popular site ( in French) - http://www.meteociel...6&code=0&mode=0

Sorry this is only a brief explanation, hope this helps.

Edited October 18, 2012 by Liam J

[Frosty.]

A quick look ahead to the weekend from Mr Gibbs shows a fairly nice day on saturday with a ridge over the uk but going downhill from the southeast on sunday with rain spreading north but northern areas still look sunnier on sunday, I noted a change in the track of sunday's low based on the rain distribution, on the gfs 00z it was heading towards southwest england but the latest met data suggests the low will probably be incoming towards the southeast, these rain maps change every few hours so the track of the low will probably alter a few more times before it's nailed. Into early next week it looks like becoming drier, brighter and warmer across the southern half of the uk but northern britain probably won't notice that much difference from current or recent values although a few degrees increase is likely, the main warmth (17-19c further to the south).

Edited October 18, 2012 by Frosty039

[Met4Cast]

The 06z GFS does seem to have trended more towards the ECM this morning, almost ditching the Northerly and sinking high pressure over the UK instead of building heights to the North and West.

Yesterdays 06z GFS dropped the idea of cold flooding down across the UK too, only to be picked up by the 12z, 18z & 00z. It's said compare like with like, so comparing todays 06z with yesterdays would give some idea of consistency, but so would comparing yesterdays 12z with the days before, etc..

ECM has been very consistent with not developing the Northerly and keeping high pressure close to or over the UK, UKMO doesn't seem keen on the Northerly either. With that, I'd say the chances of cold like the GFS 12z have been showing is lower, especially as these have often than not been on the cold side of its ensembles.

[BLAST FROM THE PAST]

Fred how many times does GFS do this, someone should count them up?

Go to Wx Online folks and see how the main models can and do vary over several days for the same date/time ahead.

Indeed John, and I am not calling 06z more dodgy than 00z, just highlighting that one of them is at least quite wrong and IMO that should not be the case within t120 and on a next to each other run. I've gone for blend of ecm and gfs but 06z does look much more like ECM 00z

BFTP

Edited October 18, 2012 by BLAST FROM THE PAST

[Paul]

Here are the GFS data stats for August. As you'll see the 06z and 18z have a lot less land soundings within them than the 00z and 12z runs, but they do have more of some other data sources. According to the stats for this particular month the 6z has the lowest amount of data within it overall.

http://www.nco.ncep....ugust_2012.html

Despite this though, there's no clear correlation with that in the verification stats..

http://www.emc.ncep....tml4x/acz5.html

[BLAST FROM THE PAST]

A quick look ahead to the weekend from Mr Gibbs shows a fairly nice day on saturday with a ridge over the uk but going downhill from the southeast on sunday with rain spreading north but northern areas still look sunnier on sunday, I noted a change in the track of sunday's low based on the rain distribution, on the gfs 00z it was heading towards southwest england but the latest met data suggests the low will probably be incoming towards the southeast, these rain maps change every few hours so the track of the low will probably alter a few more times before it's nailed. Into early next week it looks like becoming drier, brighter and warmer across the southern half of the uk but northern britain probably won't notice that much difference from current or recent values although a few degrees increase is likely, the main warmth (17-19c further to the south).

Indeed Frosty, I mentioned the other day that BBC SE had us as wet on Sunday when you mentioned in your summary re pleasant warm weekend into next week. The model output looked fine but there is a niggly front lurking. Looking at that they have extended the rain area more and looks well wet!!

BFTP

Edited October 18, 2012 by BLAST FROM THE PAST

[johnholmes]

the pdf I promised about GFS 00z V ECMWF and UK Met

GFSvECMWF and UK Met.pdf

[phil nw.]

the pdf I promised about GFS 00z V ECMWF and UK Met

GFSvECMWF and UK Met.pdf

Thanks John.

Apart from those charts you reviewed the ECM and GFS Ens mean over the last few days have never favoured a Northerly across the UK for later next week.

The GFS operational has been very much on the cold side of it`s ens. members these last 2 days-apart from the 006z run today where it seems to have moved closer to other outputs in building heights across the UK and diverting the cold plunge east towards Scandinavia and N.Russia.

Edited October 18, 2012 by phil n.warks.

[IDO]

I think that the GFS 06z still has the pattern of colder air pealing off the Arctic:

http://cdn.nwstatic..../h850t850eu.png

But its the same old story, ditto last year, the cold pooling heads to our east on this run. The Atlantic doing its usual, in pushing from the west, and we miss out, as per the majority of these cold runs.

However we still get a cool easterly from the HP, modelled by GFS and ECMW. Not very exciting, seeing whats on offer, but a good start to the winter season:

http://cdn.nwstatic....erdeenshire.png

However the scatter from the members suggests anything is still possible from T120.

Edited October 18, 2012 by i'm dreaming of...

[Methuselah]

It's only OCTOBER!

[phil nw.]

Whilst it`s quiet it seems a good time to start a new thread ready for the 12z`s.

Closing this in 5 mins.then