Model Output Discussion - January 12th.

[snow drift]

Those meteociel charts colours are a bit scewed and do tend to make it look a touch colder than it is, but even so -6c @ 850hpa is generally pretty reasonable for snow.

Looking at the differences between the models this morning, even at 96 hours they're diverging, which goes a long way to explain the very different outcomes on offer by the time we reach 168 hours plus!

Taken from the model comparison page:

http://www.netweathe...modelcomp;sess=

The model comparison page is superb and clearly highlights where things differ. Today will give us some answers I feel.

[blitzen]

The model comparison page is superb and clearly highlights where things differ. Today will give us some answers I feel.

Is it possible that the Michael Fish latest already has?

Edited January 13, 2012 by Blitzen

[Tom Montalbano]

Correct me if I am wrong here but the 00Z ECM/GFS runs come out at midnight and are only relieased at the times we see them. If that is the case and haveing read the 3-5 day outlook on the metoffice site for my region the ECM model would seem to be more inline with their thoughts up to Tuesday anyway. They mention a cloudier day with the chance of rain later. The GFS chart for Tuesday looks very settled with a huge high on top of the country so I would think rain would be hard to find, however the ECM chart on Tuesday shows a more unsettled chart with the high getting squashed by an approaching system for the Atlantic. So hopefully we could get a better mid range forecast today if my thoughts are correct.

Hi GSL,

Yes the METO suggest rain spreading east on Tues and that was updated around 6.20 or so this morning, as you state certainly doesnt correlate with the GFS view of things but I was surprised that UKMO at t120 looks quite a bit different to last nights FAX for midday Tues. It looks like a halfway house between GFS and ECM.

Yes a nice run from the 00z ECM, although imo not quite as good as their 12z from yesterday.

Slightly encouraging that the GFS 00z ensembles dip to around -5 850s mid-term, although rise later on but as models cant seem to agree beyond t72/t96 nothing to get hung up on, yet.

Yet more drama to come today, no doubt!

Regards,

Tom

[sebastiaan1973]

EC32 shows cold weather (1-3 too cold) till the end of the month for large parts of the continent. In eastern Europe even colder. From 16-23 it's around average in the northern parts of the UK and Ireland. After that in the whole of the UK 1-3 too cold. Entering february there's not a clear signal.

[Frosty.]

The Ecm 00z has not been thrown off course overnight, it is still promising a very cold spell starting in 6 days and intensifying, the ecm would bring heavy snow to the uk by the end of next week and into the following week with bitterly cold NEly winds and some very cold 850's close to minus 10 T850's, the cold would briefly slacken as a reload would then be in progress but the jet profile on the T+240 ecm chart looks full of potential for further reloads of wintry weather with bitterly cold air rushing southeast out of canada and high pressure squashed into the sw atlantic. The Gfs 00z again is nowhere near as promising as high pressure lingers and lingers but is followed by more of a nwly or nly blast eventually but of much shorter duration for the south, a 24 hr blast for southern england but 48-72 hours for scotland before it turns less cold and then it all goes back to average zonal after that so it's the ecm which would deliver what all cold fans are craving.

[shedhead]

Morning All

Well the models remain entrenched in their positions this morning, whilst I haven't seen the ECM beyond T192 it matters little in the circumstances. The ECM and GFS disagree strongly in how they see the AO going - the GFS has a deep low pressure coming out of Svalbard at T168, whereas the ECM is building pressure to form an over arch link with pressure build over Eastern Greenland.

Such pressure build cannot happen if the GFS right, only temporary ridging, The GEM supports the GFS and it is hard to say which way the UKMO would go without seeing more of the 'picture'.

The METO have access to data that we don't see and that has proved invaluable to them in producing such accurate mid-range forecasts over the last couple of years. I would suggest that if it fails to support the ECM in its update at Lunchtime today, then we must see the GFS as being the much more likely outcome.

Regrettably I have to agree with this. ECM continues to look increasingly wintry run by run, but at the same time it continues to look increasingly isolated with the evolution it's touting. Yes the MO update will be of interest later, but so will the NOAA discussion, which basically ditched yesterdays 00GFS after day 4. It may only go out to T+144hr, but UKMO looks a big concern to me if looking for longer term cold, then again how many times has that been said over past Winters in similar situations to this?

Edited January 13, 2012 by shedhead

[Cecil Hogmanay Melchett]

Well ECM looks very nice to me eyes this morning and would signal SNOW for large parts of the UK towards the end of the run as the trough sinks and heights build to the NW.

One only has to look at the GFS ensembles to see the op was pretty much a mild outlier during the transition period.

Its game on this morning IMO and if the meto dont start suggesting a colder contiental flow by tomorrow id be suprised!!

[mulzy]

ECM ensemble maps look decent to me

day 8: http://www.wetterzentrale.de/pics/Reem1921.html

day 10: http://www.wetterzentrale.de/pics/Reem2401.html

Maybe a worry about the lack of heights to the north....

[sebastiaan1973]

http://www.ecmwf.int/samples/d/getchart/catalog/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!168!pop!od!enfo!enplot!2012011300!!chart.pdf

4 clusters, oper and control are more or less the same (same cluster) together with 20 ENS members.

[nick sussex]

Optimism dented this morning as the ECM loses its main support the UKMO which has wobbled, it hasn't gone fully to the GFS but its certainly backtracked to a mid point solution.

The GFS continues from its outputs of recent days, there is likely to be a reforming of that PV at least temporarily and so the window of opportunity isn't going to last that long so its important that before that happens you need the pattern as far west as possible.

The mystery deepens! Here you go the ECM ensemble maps, every single member places troughing to the east and displaces the high to the west and out of 50 members and the control run the sum backing for the GFS at 168hrs is 0!

Very bizarre!

http://www.ecmwf.int...t!2009112700!!/

Given those ensembles I'd be surprised if the UKMO update doesn't mention some colder weather.Regardless of what the ECM does after 168hrs in terms of the PV its hard to imagine how not a single ensemble member would back the GFS at that timeframe.

Edited January 13, 2012 by nick sussex

[Radiating Dendrite]

I am no expert so apologies if I am wrong but looking at the current GFS run from this morning at T+0 it shows uppers of -4 to -8 across much of Northern Europe and precipitation of snow across the same region.

So why would it not be snow from the ECM 192 chart? What's different?

I just said it would be a bit marginal.

The difference with Europe is that the feed is Continental (i.e. it is not warmed up by water) and that can support snow at higher uppers of -3/-4.

I think people may be getting a bit carried away because we have been snow starved.

The ECM is a good run, but I am not blown away by it and it is not going to be a repeat of last December is it was to come off. We never get truly cold uppers (below -10c) and with the temp difference between the North Sea and Upper Air at its lowest around now (with the North Sea being cool) then I am unsure as to how much convection would be generated, unlike last Nov/Dec when the temp difference between sea and air was upto 20c.

Additionally, later on in the run we end up with slack Northerlies once the initial "blast" has passed through. This would leave us cold but dry inland. As IF has also mentioned, we need to see the PV stay subdued and without further stratospheric warmings it could re-invigorate and crash the party again.

Lots of things still need to fall in place, plus the ECM could be completely wrong anyway!

[TEITS]

After looking at this mornings output I feel im going to sit on the fence rather than predict which model is right. Due to the disagreement I wouldn't be surprised if all the models have this wrong.

One positive though is although the GFS/ECM disagree both show a cold outlook and for the first time this morning the GEFS mean for my location drops below -5C. I will add the GEM doesn't back the GFS and I have no idea why Ian says it does.

Just seen Nick S post.

Thats incredible not one ensemble supports the GFS. Don't think i've ever seen this happen before!

Edited January 13, 2012 by THE EYE IN THE SKY

[cooling climate]

The ECM t240 chart is interesting in that the weather could go either way after this. One option is that another segment of the vortex perhaps splitting and dropping southeast with pressure rising to the northeast prolonging and intensifing the cold spell.The other option is that milder air is introduced from the west with the colder air being mixed out and an quick end to the cold snap.

At this point in time and with the way this winter has gone so far I would take a two or perhaps three day cold snap with the possibiity of snow showers and lying snow than nothing at all although I would not be surprised to see the colder option developing.

Edited January 13, 2012 by cooling climate

[shedhead]

After looking at this mornings output I feel im going to sit on the fence rather than predict which model is right. Due to the disagreement I wouldn't be surprised if all the models have this wrong.

One positive though is although the GFS/ECM disagree both show a cold outlook and for the first time this morning the GEFS mean for my location drops below -5C. I will add the GEM doesn't back the GFS and I have no idea why Ian says it does.

Indeed, GEM is far more ECM'like at T+144hrs, but at the same time UKMO is far more GFS'like. Something has to give here very shortly, but because both positions are so diverse we may well ultimately end up with some sort of blend as you suggest.

[nick sussex]

After looking at this mornings output I feel im going to sit on the fence rather than predict which model is right. Due to the disagreement I wouldn't be surprised if all the models have this wrong.

One positive though is although the GFS/ECM disagree both show a cold outlook and for the first time this morning the GEFS mean for my location drops below -5C. I will add the GEM doesn't back the GFS and I have no idea why Ian says it does.

Just seen Nick S post.

Thats incredible not one ensemble supports the GFS. Don't think i've ever seen this happen before!

It's very strange, even if some of those members turn out to not bring much in the way of cold after that its hard to fathom how every single one does one thing the GFS operational doesn't do at 168hrs.

if the UKMO had remained solidly behind the ECM I would be much more optimistic that the GFS has got this one all wrong, as it is I will join you on the fence.

[wolvesfan]

Morning all,

The worrying thing for me this morn from the ECM in particular, is the way it rebuilds the polar vortex,also the blocks in Russia and Asia pull back and we see the cold push back into the pole from the northern lattitudes especially over the United States.I suspect a -ve AO could only be short lived.

We deffinatly need the strat warming to propogate lower down or IMO any cold snap wont have longevity,

[Radiating Dendrite]

Why do we see such differing outputs sometimes though?

Are the models fed slightly different data because they are from differing forecast agencies? Are the programmes that they run on different?

Is the GFS fed Ready Break and the ECM Cornflakes? :lol:

[TEITS]

Why do we see such differing outputs sometimes though?

Are the models fed slightly different data because they are from differing forecast agencies? Are the programmes that they run on different?

Is the GFS fed Ready Break and the ECM Cornflakes? :lol:

Im pretty sure all models use roughly the same data but its how they are programmed that makes the difference. I sometimes wonder though if the GFS, NOGAPS are run on a Sinclair Spectrum. :lol:

[shedhead]

Morning all,

The worrying thing for me this morn from the ECM in particular, is the way it rebuilds the polar vortex,also the blocks in Russia and Asia pull back and we see the cold push back into the pole from the northern lattitudes especially over the United States.I suspect a -ve AO could only be short lived.

We deffinatly need the strat warming to propogate lower down or IMO any cold snap wont have longevity,

Agree with this too, it's definately of concern this morning. ECM at 240hrs does not look good hemispherically for sustained cold, but at this stage I suspect most would want it to verify post T+144hrs and then worry about where we go from there later on.

[cooling climate]

It's very strange, even if some of those members turn out to not bring much in the way of cold after that its hard to fathom how every single one does one thing the GFS operational doesn't do at 168hrs.

if the UKMO had remained solidly behind the ECM I would be much more optimistic that the GFS has got this one all wrong, as it is I will join you on the fence.

The UKMO looks fine to me, OK its not quite as progressive as the ecm t144 chart but its going

the same way and by t168 I am sure we would see pressure building up to the NW and Greenland.

Edited January 13, 2012 by cooling climate

[johnholmes]

Im pretty sure all models use roughly the same data but its how they are programmed that makes the difference. I sometimes wonder though if the GFS, NOGAPS are run on a Sinclair Spectrum. :lol:

agreed, all take the same data but the programming is/will be different. To find out what that is its best to search each centre for what/how they programme their model(s).

Some of it is actually understandable!

[snowking]

Just RE: UKMO halfway house.

If you look at the verification stats that shouldn't actually be too much of a surprise, as the UKMO is about halfway between the ECMWF and GFS in terms of verification, so to a certain extent at days 5/6 this shouldn't be a surprise - its closer to day 3 that the UKMO starts to catch up more with the ECMWF

SK

[shedhead]

The UKMO looks fine to me, OK its not quite as progressive as the ecm t144 chart but its going

the same way and by t168 I am sure we would see pressure building up to the NW and Greenland.

I'm not convinced of that at all CC, looks to me as if the next system is already showing it's hand southeast of Cape Farewell on the T+144hrs, so mild'ish W or SW'erlies predominating thereafter imo.

[nick sussex]

The longer range ECM ensembles solidly behind a cold outlook at least for De Bilt:

http://www.weerplaza.nl/15daagseverwachting/default.asp?type=eps_pluim

Of course given the location of troughing the UK could be on the less cold side of that but those ensembles certainly show strong support for cold conditons just across the sea in Holland.

[sebastiaan1973]

15 days ENS shows a good agreement till the end of next week. Temporarily warmer around midweek, followed by lower temperatures on friday and the weekend.