General Model Output Discussion

[Frosty.]

None of the 12z ensembles bring any cold upper air for next week: http://www.netweather.tv/index.cgi?action=ensviewer;sess=

And these are for London, which is the most likely to benefit from any east/southeasterly flow!

What a turnaround from yesterday!

Karyo

Yes Karyo it looks like the writing is on the wall for a wintry spell from the east and we need to look between south and west for our weather in the next few weeks with normal service being resumed for the first time since early/mid november.

[Climate Man]

Yes Karyo it looks like the writing is on the wall for a wintry spell from the east and we need to look between south and west for our weather in the next few weeks with normal service being resumed for the first time since early/mid november.

However, their are signs of both an Arctic High and split in the PV in deepish FI, if this is the start of a trend then it is very positive for the end of January/early February - often a cold time.

[Thundery wintry showers]

Regarding the UKMO's performance at T+144, when I did my model comparisons analysis from March 2006 through to Summer 2007, the UKMO frequently did drop markedly in accuracy between T+120 and T+144 and this drop was more noticeable than with either GFS or ECMWF, so I don't think the criticisms of the UKMO at T+144 were entirely unjustified.

However, those T+144 anomaly correlation archive charts are very revealing. I've checked out most of the months of 2006 through 2009, and the stats consistently have the UKMO about level with GFS with the ECM quite some way out in front. This implies that many people were too hard on the UKMO since they often dismissed it as being less accurate than the GFS at T+144, when it wasn't- only ECM was significantly ahead.

Does anyone have a link to the T+120 (5 day) anomaly correlation archive charts? They would be interesting to see- taking the results of my aforementioned analysis, I would expect to see the UKMO up there with ECM, in line with the fall towards GFS's level for T+144.

But it's also very clear that UKMO at T+144 has caught right up with the ECM in the last 2-3 months. Whether this is a blip, or a sign that the Met Office have implemented significant improvements to their medium-range forecasting model, I don't know, but certainly for now, UKMO at T+144 needs to be taken notice of as much as the ECM at T+144.

[casparjack]

Looking at the gefs ensemble spread upto 180hrs for SLP the main area of spread is near Iceland with very little for the uk and further east, given the set up its safe to say the blocking high will remain over Russia with low pressure unable to cross the meridien. In terms of 850 temps again little spread over the uk and europe. So i think it looks pretty much like a stalemate with the chance that eastern and se areas seeing the coldest of any temps with surface flows occasionally from the Continent, still even with this set up theres a chance of some snow depending on the trajectory of low pressure and the tilt of the trough as fronts try and push eastwards.

..not to mention, that in time with the correct alignment of the various other factors coming together we may well see an intensification of the cold air with uppers and surface temperatures moving to favour cold.

Edited January 14, 2010 by casparjack

[johnholmes]

john, i made a post some while back re UKMO T144 output which didnt get a response. maybe it was a bit manic at the time. my anecdotal impression is that UKMO T144 often looks like an ens mean chart as it isnt keen to over deepen shortwaves nor 'blow up' highs at this range. does this give it an advantage when the stats are done to see how far it is out on pressure readings compared to GFS et all. after all, we know how GFS overdoes central pressure on shortwaves at this mid range.

I honestly don't know ba

[Thundery wintry showers]

I've found the T+120 analysis:

http://www.emc.ncep.noaa.gov/gmb/STATS/html/monarch.html

Interestingly, it shows much the same as the T+144 stats- the UKMO tending to come out similar to, or marginally better than, the GFS with the ECMWF quite a distance in front. Again there is evidence of the UKMO gaining ground over the last 2-3 months but less emphatically than with T+144.

So any theory that the UKMO used to drop markedly in accuracy between T+120 and T+144 seems to have been dealt a huge knock!

Maybe there's something behind what Casperjack says. I don't think the UKMO T+144 outputs are generalised enough to be ensemble mean charts but they often lack detail relative to the UKMO's own T+96 and T+120 outputs, and also the T+144 outputs of ECM and GFS. So maybe there is some kind of ensemble-based "smoothing out"? Like John I honestly don't know. Two possibilities emerge from this, one being that it may help boost the UKMO T+144 accuracy ratings and another being that the lack of detail may make them appear less accurate than they really are.

[Guest North Sea Snow Convection]

I think there have been a few overreactions to the 12z output so far.

In the more reliable time period the high still organises itself to ridge into eastern areas at the very least and the atlantic only comes through after t144. Closer to t168 in fact, which is a time period where all sorts of changes can still occur.

There is still a margin there for just a slight adjutment westwards which will make a huge difference in terms of the UK. Frankly the GFS ensembles suggest little to me - they can flip overnight. We have been in situations where there has been 'apparent' suite agreement for a cold spell within 5 or 6 days with 'apparent' ensemble support and the situation has still imploded within t72 to T96.

I'm really not sure why the converse isn't also true?! Especially this winter - which is hardly a bog standard one by any stretch of the imagination

The same applies with the UKMO. There is still a subtle margin for error after t120 - again which would make quite a difference for a nation the size of the UK

Of course, should the ECM this evening also suggest atlantic inroads from after t144 then the odds are certainly reducing - but even then there could easily be more u turns tomorrow. It has happened many times before, and it can happen again

This isn't over yet - the models are going to easily be prone to overcooking the jet position here and underestimate that huge High. I see no reason why the METO should change their outlook. Not yet anyway.

Edited January 14, 2010 by North Sea Snow Convection

[Steve Edgefield]

I'm really not sure why the converse isn't true?! Especially this winter - which is hardly a bog standard one by any stretch of the imagination

The same applies with the UKMO. There is still a subtle margin for error after t120 - again which would make quite a difference for a nation the size of the UK

Of course, should the ECM this evening also suggest atlantic inroads from after t144 then the odds are certainly reducing - but even then there could easily be more u turns tomorrow. It has happened many times before, and it can happen again

This isn't over yet - the models are going to easily be prone to overcooking the jet position here and underestimate that huge High. I see no reason why the METO should change their outlook. Not yet anyway.

Very sensible post - subtle changes could make a big difference. It's a difficult call to make at this stage despite what the models show and I'm keen just to see how the High holds up, or doesn't after the weekend. These aren't typical times..

[Guest North Sea Snow Convection]

Very sensible post - subtle changes could make a big difference. It's a difficult call to make at this stage despite what the models show and I'm keen just to see how the High holds up, or doesn't after the weekend. These aren't typical times..

Big differences already with the ECM at t120

[Shiny_Bottom_1]

In the more reliable time period the high still organises itself to ridge into eastern areas at the very least and the atlantic only comes through after t144. Closer to t168 in fact, which is a time period where all sorts of changes can still occur.

Are we pulling the 'FI' card Tamara :ph34r:

The reason is, the GEFS ensemble mean was quite good at the current change...somewhere in the region of 5 days i'd say. It predicted the move above the -5oC 850's for around Thursday (today), and then the rise to more normal 2m temps by the weekend. Synoptically, it may not have been perfect, but overall, I do think its been quite good.

People are often too quick to criticise it, feeling they have to side with either the GEFS or ECM. Not sure why.

12z GFS...sum up is not the best for snow and cold lovers, but again, neither for mild lovers. Looks potentially quite like an average temperature prediction. Could be better...could be worse.

[West is Best]

There's one word to sum up the medium and long term outlook from GFS and UKMO based on their 12z outputs. And the word is ...

Bartlett

Fear and trembling to all cold lovers ...

[Optimus Prime]

There's one word to sum up the medium and long term outlook from GFS and UKMO based on their 12z outputs. And the word is ...

Bartlett

Fear and trembling to all cold lovers ...

That's one of many times you've mentioned that word this week and yet I still fail to see where it's going to come from?

[Cecil Hogmanay Melchett]

There's one word to sum up the medium and long term outlook from GFS and UKMO based on their 12z outputs. And the word is ...

Bartlett

Fear and trembling to all cold lovers ...

ECM is better though WIB.

[Shiny_Bottom_1]

There's one word to sum up the medium and long term outlook from GFS and UKMO based on their 12z outputs. And the word is ...

Bartlett

Fear and trembling to all cold lovers ...

Whilst I would say perhaps a longer shot than an Easterly (i'm not going to say 'the' easterly), its not by a huge amount.

A Bartlett is a possibility (OP - Its behind below you)

Edited January 14, 2010 by Shiny_Bottom_1

[Optimus Prime]

I know where a Bartlet can be situated but look at the low heights towards the east and the angular shape of the high over Scandinavia. I really can't see that forming a conveyor over us but I suppose you never know.

[MP-R]

That's one of many times you've mentioned that word this week and yet I still fail to see where it's going to come from?

Well if we're not careful it might become a Bartlett right over the UK lol.

At present Thursday next week looks the mildest day with perhaps double figures but other than that nothing above 10C for everywhere except the far southwest coastline.

http://www.wzkarten2.de/pics/Rtavn17417.png

[Shiny_Bottom_1]

I know where a Bartlet can be situated but look at the low heights towards the east and the angular shape of the high over Scandinavia. I really can't see that forming a conveyor over us but I suppose you never know.

Ahhhh...but if things 'could only shift a few hundred miles west', then could it not go the other way as well?!

[Cumulonimbus Tower]

ECM is better though WIB.

Yes, along comes the ECM with other ideas... Thankfully.... Maybe quieten the doom and gloom mongers a little

[ChartViewer]

12Z ECM up to T168 much better as it shows Heights to our North East and Lows to our South

Certainly NO Bartlett shown there

[Frosty.]

Well if we're not careful it might become a Bartlett right over the UK lol.

At present Thursday next week looks the mildest day with perhaps double figures but other than that nothing above 10C for everywhere except the far southwest coastline.

http://www.wzkarten2.de/pics/Rtavn17417.png

Looks longingly east at all those blues we won't be getting. I think the writing was on the wall as far as I was concerned after that ECM 12z run yesterday.

[snowingtequila]

Never mind all this talk of Bartletts, look at this nice little low, should create some fun on friday, wouldn't be surpriseed if it deosn't then lead onto dragging some colder air in afer it

[yamkin]

Never mind all this talk of Bartletts, look at this nice little low, should create some fun on friday, wouldn't be surpriseed if it deosn't then lead onto dragging some colder air in afer it

OMG This low would produce snow/cold goodies. If anyone has yet to see Michael Fish's latest weekly forecast, it's a must.

[igloo]

well another mild run by the ECM any possible cold will happen in F1 which chops and changes like the weather its self anyway looks like a simple forecast to make for next week mild wet and windy the normal

[nick sussex]

Can we put this Bartlett talk to bed! there is no chance of this occuring with the upstream pattern, i do wonder whether people just pull this term out of a hat whenever the output looks a bit ropey! which is definitely not the case. I would ask certain people to do some research on what exactly the Bartlett high is before terrifying some net weather members into reaching for the prozac!

[snowingtequila]

well another mild run by the ECM any possible cold will happen in F1 which chops and changes like the weather its self anyway looks like a simple forecast to make for next week mild wet and windy the normal

IT's FI NOT F1

have you actually looked a the run?

I am liking the look of ECM myself, not just on a looking for cold basis but I like the look of those vicious lows wherever they go.

Edited January 14, 2010 by snowingtequila