Winter Model Discussion - The Hunt For Cold - 12th Dec

[Frosty.]

a few days ago thought you was saying no chance for cold for the next two weeks

peter gibbs, the mild lover just said mildish until xmas or even the new year but the models don't look that bad, certainly a chance of cold weather returning as shown on the ecm 12z last night, ecm 00z and gfs 06z, even the latest meto hinting at frosts returning with hill snow in the north. Lots of uncertainty still since the crucial T+96 to 120 range could swing things either way and hopefully towards cold.

[Snowman.]

it will be early to mid january before we see the effects of a strat warming.

True still very nice to see. As it needs to work through the layers of the strat for a full effect.

Anyway NO ONE get their hopes up for any kind of cold spell I have learnt that the hard way with all these flops recently.

Edited December 14, 2012 by SN0WM4N

[AugustSky]

I completely disagree that dec 10' was a rare event. It was simply just that we had the right Synoptics in the right place. The same could and can quite easily happen again!

I agree. I think our weather patterns are changing and that can mean anything is possible.

[The Eagle]

This would be a fantastic position to be in on Christmas eve. All up for grabs in my opinion.

[nick sussex]

peter gibbs, the mild lover just said mildish until xmas or even the new year but the models don't look that bad, certainly a chance of cold weather returning as shown on the ecm 12z last night, ecm 00z and gfs 06z, even the latest meto hinting at frosts returning with hill snow in the north. Lots of uncertainty still since the crucial T+96 to 120 range could swing things either way and hopefully towards cold.

That looks a rather premature call especially as theres lots of uncertainty upstream in the USA with low pressure development off the eastern seaboard, this impacts the down stream pattern at a critical time.

From the latest update for Maine, maybe Mr Gibbs should have muzzled it until the upstream pattern is sorted out.

FORECAST CONFIDENCE QUICKLY BECOMES MUCH LOWER INTO LATE MON. THE

SENSIBLE WX FOR EARLY NEXT WEEK REVOLVES AROUND THREE PIECES OF

ENERGY MOVING THRU THE DEEPENING EAST COAST TROF. HOW EACH MODEL

HANDLES THE MERGING OF S/WV ENERGY IS PRODUCING A WIDE VARIETY OF

SOLNS. THE 14/00Z ECMWF DEEPENS THE SECOND WAVE IN RAPID SUCCESSION

BEHIND THE MON WAVE...WITH THE TRAILING S/WV NOT PHASING UNTIL

WELL OUT TO SEA. THE RESULT IS COASTAL CYCLOGENSIS TUE...NOT WED

AS PREVIOUSLY FORECAST BY NWP. CONVERSELY...THE 14/00Z GFS

CONTINUES TO DELAY THE AMPLIFICATION OF THE SECOND WAVE UNTIL THE

THIRD CATCHES UP. IN THIS INSTANCE A LARGE COASTAL LOW PRES

CONTINUES TO HOLD OFF UNTIL WED. WITH THE LOW FORECAST

CONFIDENCE...HAVE BLENDED HPC POP WITH PREVIOUS FORECAST TO

MAINTAIN CONTINUITY. THE RESULT IS CHC POP FOR MOST OF THE MIDWEEK

PERIOD...WITH SN INTERIOR AND MIX ALONG THE COAST. ATTM MUCH CAN

CHANGE AND MUCH WILL DEPEND ON HOW THE FIRST WAVE OF LOW PRES PLAYS

OUT SUN INTO MON.

[Kentish Snowman in Yorks]

Peter Gibbs on BBC 1 has just said the weather this weekend (westerly dominated) is likely to last into next week and maybe even lasting to Christmas a big call that from Peter

Well to my eyes, nothing has changed since yesterday in the sense that a 'green' Christmas would not be a surprise in the slightest given the current output, and today's 6z GFS ensembles do not scream cold either;

However, all the time the blocking remains to our North and East, any slow down or amplification of the Jet means we could be in business in terms of a cold set up at a relatively close time frame, so I would agree - a big call from Peter Gibbs today as it was from Paul Hudson last night.

The modelling of the interaction between the block to the East and the Atlantic in what is clearly a titanic struggle has caused much confusion for the models over the last few weeks, and will continue to do so until the outcome is decided.

I would put the odds on a 'white' Christmas (be that frost or snow) at around 40% at the moment, with an average to mild Christmas looking slightly more likely.

Obviously there is a third outcome that we are left in a no mans land for the foreseeable and through the Christmas period, without any real Westerly or Easterly influence, although that would probably lend itself to the mild / average camp.

All very intriguing. Lets see what the 12z's bring...

[Methuselah]

Anyhoo, back to the models...

I'm minded to think that there must be a systematic error in the models (not in one of them, but in all of them). How else does one explain the fact that all of them modelled a raging BFTE - that never actually happened?

It also seems to fly in the face of the popular contention that any model that shows any possible breakdown (in any cold spell) is demonstrating some sort of 'default to mild', or is somehow 'not very good at modelling Easterlies'...But what if there is no 'westerly bias' at all and the models are, in fact, too good at modelling Easterlies? Or that if the real bias only lies within the weather itself? What then?

What made them all forecast a long, cold blast-that-never-was? Any ideas?

Edited December 14, 2012 by Rybris Ponce

[chionomaniac]

Please keep to the topic, thanks. Any chat about years gone by will magically disappear.

[carinthian]

peter gibbs, the mild lover just said mildish until xmas or even the new year but the models don't look that bad, certainly a chance of cold weather returning as shown on the ecm 12z last night, ecm 00z and gfs 06z, even the latest meto hinting at frosts returning with hill snow in the north. Lots of uncertainty still since the crucial T+96 to 120 range could swing things either way and hopefully towards cold.

Hi,

Brave prediction if he said that. The Russian High is well established for so early in the winter period. Normally, does not show its hand over the UK longitude until Mid or late winter period. So if it continues to hold its ground it will become a powerful force in the outlook period with wet conditions turning quickly to ice and snowfall.

C

[Frosty.]

Hi,

Brave prediction if he said that. The Russian High is well established for so early in the winter period. Normally, does not show its hand over the UK longitude until Mid or late winter period. So if it continues to hold its ground it will become a powerful force in the outlook period with wet conditions turning quickly to ice and snowfall.

C

funny thing is, what PG said conflicts with the meto update to some degree, anyway, a large element of doubt in the models for next week, let alone xmas week with a tricky period coming up and still a chance of cold air making inroads at times, hill snow in the north and a return of frosts seems a reasonable assumption in the mid range. The last thing I see on the latest models is a straightforward milder zonal spell until the new year, Mr gibbs comment today will hopefully backfire on him soon enough.

PS. Blizzard warning for shetland tonight with severe gale force easterly winds, the last bit of fight in the cold snap will take place up there tonight but a less cold 6c in the far north tomorrow with rain turning more showery and winds backing sly and easing.

Edited December 14, 2012 by Frosty039

[nick sussex]

Anyhoo, back to the models...

I'm minded to think that there must be a systematic error in the models (not in one of them, but in all of them). How else does one explain the fact that all of them modelled a raging BFTE - that never actually happened?

It also seems to fly in the face of the popular contention that any model that shows any possible breakdown (in any cold spell) is demonstrating some sort of 'default to mild', or is somehow 'not very good at modelling Easterlies'...But what if there is no 'westerly bias' at all and the models are, in fact, too good at modelling Easterlies? Or hat if the real bias only lies within the weather itself? What then?

What made them all forecast a long, cold blast-that-never-was? Any ideas?

Interesting question, I don't think theres any default mild setting but just the programming deals better with a more typical zonal pattern.

Perhaps theres an inherent problem with the actual programme. In terms of the non event easterly that was perhaps down to a lack of observational data in a key region that didn't pick up the shortwave until much later, however what was the real body blow was that in conjunction with a change in the upstream pattern.

Originally a developing shortwave was expected to head ne towards inland Canada, this amplified the upstream pattern sharpening troughing to the west, this resulted in less eastwards energy and some stretching of the trough with energy being displaced se.

What transpired was a flatter system running more east into the back of the trough, at that point game over.

I don't think the models handle easterlies well, if they did then the modelling of these at earlier timeframes would survive just like zonal set ups do at that range.

[Summer Sun]

If you use the live rewind feature on the BBC iplayer you can see the forecast here - http://www.bbc.co.uk/iplayer/tv/bbc_one_london/watchlive - rewind to 13:29 its available for 2 hours after the broadcast

[Catacol]

Um, sorry - what proves that the MetO 'put very little energy into serious seasonal forecasting at present'? The fact that you disagree with it?

The fact that they were no more accurate than the "free" models that we are privvy to with regard to this week, and the fact that they have gone from cold to mild in their extended outlook with a sentence at the end for 5 - 12 January that says "no clear signal." We have heard this morning that EC is showing a signal for height rises again at the end of the period, and I'll bet that this ends up factored into their extended outlook in a week or so.

So... if all they are doing is reading 2 models (EC and Mogreps) and taking a rough guesstimate between the 2 that says to me not that much effort. Go and read the Strat thread or GP's thoughts on winter and you will find effort. The Met are so highly accountable that they dare not release a forecast with a significant chance of being incorrect, so they simply dont bother beyond releasing computer % predictions to paying customers.

Do you see it differently?

[nick sussex]

Re my previous post, a good example of why we must always look upstream:

That was the ECM forecast, note the amplified trough about to phase with the PV.

And this is what happened:

The flatter system, less amplification upstream:

We may still have got a better cold spell if only one thing went wrong, even if the shortwave appeared to the north with less energy heading eastwards we had a much better chance of not seeing the energy heading sw phasing with the Atlantic energy.

I know people might think I drone on a little too much about the upstream pattern but really its absolutely crucial because of the effects it has in Europe.

Edited December 14, 2012 by nick sussex

[Bradowl]

Peter Gibbs on BBC 1 has just said the weather this weekend (westerly dominated) is likely to last into next week and maybe even lasting to Christmas a big call that from Peter

Watched it and the word he used was "Possibly' which means we'll probably be having a another freeze.

Hopefully!

[recklessabandon]

Anyhoo, back to the models...

I'm minded to think that there must be a systematic error in the models (not in one of them, but in all of them). How else does one explain the fact that all of them modelled a raging BFTE - that never actually happened?

It also seems to fly in the face of the popular contention that any model that shows any possible breakdown (in any cold spell) is demonstrating some sort of 'default to mild', or is somehow 'not very good at modelling Easterlies'...But what if there is no 'westerly bias' at all and the models are, in fact, too good at modelling Easterlies? Or hat if the real bias only lies within the weather itself? What then?

What made them all forecast a long, cold blast-that-never-was? Any ideas?

A question (kind of) that I raised last year. There is inevitably a bias toward the locating of Meto MAWS (Marine Automated Weather Station), they are all in the western half of the UK's waters, similarly the North Atlantic is loaded with lightships and MAWS, hardly surprising when what, 80% of our weather comes in from that direction. By comparison as I understand it the Meto have no MAWS in the North Sea. There are of course weather data stations in the North Sea and off the Norwegian coast, generally all associated with the oil installations, so inevitably there will be data source cluster then comparatively large gaps between them. I dare say that data sources towards the Barents and Arctic Ocean will be limited to research vessels and experimental stations.

I would suggest that this is why in times of 'zonality' the ensemble clusters are 'tight' due to the wealth of data feeding into the models, whereas easterly/north easterly flows are interpreted on data sets which are poorly spread in a geographical sense by comparison.

Understandable, if you have a limited budget, where would you spend your money.....probably not to the NE of the UK to keep us snow chasers happy

[Bonjangle]

Just read this on TWO .

Just as the weather prospects for the Christmas period were starting to look fairly certain, the computer models have thrown another curve ball at us, with some indications of cold weather from the east returning at least for a time. At the moment I’d say the odds are still in favour of the cold being held back, or only briefly returning, but things do need watching. The 12z GFS / GEFS 850hPa temperature plot for Leeds doesn’t really show anything too dramatic, although a couple of the GEFS runs do bring in colder air for longer, and the GFS operational run shows a cold incursion, but not lasting long.

[The Northern Ramper]

Peter Gibbs on BBC 1 has just said the weather this weekend (westerly dominated) is likely to last into next week and maybe even lasting to Christmas a big call that from Peter

That's far from a big call, it's a safe call.

[Adam lufc]

The call from p.gibbs just knocks the stuffing out of most really doesn't it. 2weeks of mild/average dross wasted. Hope to cling onto the smallest trends by the models of potential easterly flows from undercuts! Plz ecm.........

[Zenarcher]

Model changes of the last 24 hours,

ECM for the 22nd,

Yesterday

Today

Today shows the Atlantic low pressures to be weaker also better heights over our NE and N.

GFS for the 22nd,

Yesterday

Today

Today the Russian high is 10mb higher but things are being pushed back slightly more East because of the Atlantic low. Looking at the FI run from yesterday then today's on the Northern Hemisphere charts today's does look better less PV and a 1045mb high over the pole.

UKMO for the 19th,

Yesterday

Today

Today the Russian high stronger and more further West, Atlantic low made slightly weaker and also been pushed West. High pressure over the NE of Canada stronger and the low trying to enter the Atlantic is a lot weaker.

[Methuselah]

Can we please stick to the models? We have a place for dissecting BBC weather forecasts...And this isn't it!

All aboard the Skylark!

Edited December 14, 2012 by Rybris Ponce

[phil nw.]

We will close this one in a few minutes and start a new thread for the 12z`s.

[phil nw.]

OK locking this now then-new thread here folks.