Model Output Discussion - 26Th Dec 21:25-->

[clark3r]

Has anyone looked at the Fax charts for 72/84 hours? They look pretty great to me for more southern areas, though I am not an expert. Looks nothing like the GFS. Expert opinion?

Here:

http://www.meteociel...arte=2000&ech=0

thats the chart we should all be watching! i am not going with the gfs charts at all today fax chart is more reliable as with the ECM!

[mcweather]

What is of note on ALL the models is the intensely cold air that is being shown now over North East europe and Scandinavia. As much as there is talk of this imploding for the UK, it won't take much either for us to be suddenly looking at a Feb 91 or Jan 87 type real freeze.

The chances of such an occurence are steadily increasing at this time. Should we get that model agreement for an initial easterly, then output from around t144 onwards could start to look quite different and ever more exciting.....

Quite so Tamara.

A voice of reason at last.

[Guest North Sea Snow Convection]

Not to bothered about that to be honest Stu, it's more important IMO to bring a bit of balance to the discussion, especially for new members that read 99% of the posts and think another bitterly cold spell has already been rubber stamped.

For the record tho I still keeping a very open mind on the eventual outcome. I think few can genuinely deny that the current set up means all models are struggling for both consistancy and accuracy, but on the other hand I think a clear trend has emerged across the last 24-48hrs that suggests getting and then locking in another protracted cold spell across the south will be anything but easy/guaranteed.

I think that bringing a bit of balance to the discussion involves discussing the colder possibilities too. This time yesterday, I myself was less convinced about the evolution of this cold spell. But the trends now imply, apart from the continuing model disagreement at an early stage - that really properly cold air is going to come into the equation as that trough over Scandi is starting to show a more favorable aspect in terms of advecting the colder air further south. The models will continue to try and be progressive in terms of bringing the next low in from the t144 onwards period, but this is going to get harder and harder as the air gets denser and colder to the NE and the boundary is squeezed further and further south westwards.

There is a real chance that these lows will get disrupted away properly southwards and a sub zero north easterly is then staring us suddenly in the face. That is as possible and achievable as talk about mild breakdowns from the SW.

[john mac]

Yeah everyone keep their feet on the ground, As i said in an earlier post, we will know alot more later today after the ECM12z. At the moment we cannot just dismiss the GFS, we have done this before and had our fingers burned, patience is needed in this situation and as tamara said the euro's will be the favourites but keep an eye on the GFS

[TEITS]

What is of note on ALL the models is the intensely cold air that is being shown now over North East europe and Scandinavia. As much as there is talk of this imploding for the UK, it won't take much either for us to be suddenly looking at a Feb 91 or Jan 87 type real freeze.

The chances of such an occurence are steadily increasing at this time. Should we get that model agreement for an initial easterly, then output from around t144 onwards could start to look quite different and ever more exciting.....

Indeed Tamara and looking at that pool of cold air the upper temps are around -35C. This is an incredibly cold pool of air which would bring max temps of -20C on the continent with min temps of -30C!

This mornings GEM at +144 is why nobody can really be certain at the moment with regards to the outlook.

http://www.wetterzentrale.de/pics/Rgem1441.gif

Im sure some members must be reading this thread and being utterly confused. However what I will say to them is the outlook is very uncertain indeed. We could see a shortlived cold spell or something more prolonged like the ECM. However looking at the GEM you can easily see how this could upgrade over the next +72hrs.

At the moment my preference is towards the ECM for the period between +72 & +144. If NOAA tend to prefer the ECM then so will I!

[mcweather]

I think that bringing a bit of balance to the discussion involves discussing the colder possibilities too. This time yesterday, I myself was less convinced about the evolution of this cold spell. But the trends now imply, apart from the continuing model disagreement at an early stage - that really properly cold air is going to come into the equation as that trough over Scandi is starting to show a more favorable aspect in terms of advecting the colder air further south. The models will continue to try and be progressive in terms of bringing the next low in from the t144 onwards period, but this is going to get harder and harder as the air gets denser and colder to the NE and the boundary is squeezed further and further south westwards.

There is a real chance that these lows will get disrupted away properly southwards and a sub zero north easterly is then staring us suddenly in the face. That is as possible and achievable as talk about mild breakdowns from the SW.

I couldn't agree more Tamara.

I still think that this is the more likely eventual outcome.

[Cecil Hogmanay Melchett]

Im waiting for the meto update.hopefully it will side with ecm/gem which look along the same lines.

GFS much less kenn to bring any sustained cold for the UK.

The 06z ens should be updating and as Mr data said,the control run is much better!!

Quick question,are the control and op at the same resolution?

[Ian Fergusson]

Right, we have now entered the all important 72 hour prediction period. The UKMO have all the best numerical forecasts at hand this morning including snowfall models in place. Its that forecast which is going to be best placed.Active fronts are often easy to position with-in this time span, but may diffuse or become slow moving. For the progess of the front as shown by the latest 06h GFS to be rejuvenated at that pace , a secondary depression has to be in existance. Its simple not there. Back the Exeter boys today folks !!!

C

Agreed.

The latest UKMO MOGREPS probability output for spatial extent (and intensity) of snowfall makes interesting reading alongside the latest GM modelling and simulated ascents for SW Midlands and environs. Understandably 'rough-around-the-edges' but broad agreement for the Tues/Wed event in terms of perhaps 5-10cm snowfall into parts of the SW Midlands, much of central/east Wales and into other parts of the midlands and NW England (northern boundary remains uncertain). The current MO advisory adequately reflects the areas signalled in the GM run (expect this to be further tweaked in due course), albeit ECMWF ENS highlights a fair swathe of not insignificant snow probs across more northern-central stretches of the Westcountry (N Somerset, Bristol, S Glos and northwards, out through N Wilts and up into the Midlands), reflecting the somewhat troublesome pin-pointing of snow risk at a sub-regional scale at the present timeframe. Thus the elasticity remains a forecasting headache, not least given the very prominent temperature gradient being suggested in the UKMO output, which will promise hefty rainfall totals into much of far SW / southern England and a really knife-edge discrimination of where this turns (quickly?) to snow on the leading edge. Interesting stuff. The latest 6-10 day analysis looks equally spicy and will doubtless prove more than a tad newsworthy for various regions.

Best

Ian

Edited December 27, 2009 by Ian Fergusson

[Weather-history]

Here is what GFS 06z last Sunday was predicting for this Sunday

Here is the reality

[Guest North Sea Snow Convection]

I couldn't agree more Tamara.

I still think that this is the more likely eventual outcome.

Hi - in the shorter term, an interesting analysis from Ian F. Irrespective of the broader pattern perhaps being over complicated as I have been suggesting and the probability that imo that the cold air will win out - there is no doubt that pinpointing details and rain/snow boundaries is going to be the hardest job of all over the coming days!!

[londonsnow]

with the model divergence this morning i dont envy whoever is presenting the weekly outlook on the beeb in a short while. look out for the word 'uncertainty', a forecasters favourite in these situations. personally i'm backing the ecm as the forum horse

[Thundery wintry showers]

People need to recognise that when there is a lot of uncertainty, the models' operational runs are going to churn out one of a range of possibilities, and these operationals are bound to change with every run when the range of possibilities is large. I haven't seen much overall evidence of downgrading on the models over the last few days, but what I have seen is a few "fantasy" runs on Christmas Day when the GFS and ECM operationals latched onto the coldest and snowiest case scenario for a run or two, and many people have been considering the latest runs to be "downgrades" because they happen to be less cold and snowy than those "fantasy" runs. It was the same with the NE'ly outbreak on the 17th/18th December.

The first hurdle- do we get a frontal battleground on the 28th-30th followed by a NE'ly outbreak around New Year's Eve/Day- appears to have been cleared, with enormous support from operationals and ensembles. Now attention has shifted to the possible breakdown that GFS shows and UKMO tentatively hints at, as opposed to the sustained easterly spell on the ECMWF. By all means, the uncertainty beyond T+120 is worthy of discussion, but all this talk of "downgrades" of "the ideal cold/snow setup" is somewhat OTT, especially as the last two weeks have perfectly demonstrated that the models are often too progressive at bringing lows in from the SW. Remember that the cold spell that started on the 17th was supposed to break down by the 22nd/23rd, and has ended up failing to break down at all over much of Scotland?

[Methuselah]

Good day, all...

I know it's not always easy (we all live somewhere! :lol: ) but, can you all please try and keep personal preferences out of model analysis??

Please??? :lol: :lol:

[NorthNorfolkWeather]

Agreed.

The latest UKMO MOGREPS probability output for spatial extent (and intensity) of snowfall makes interesting reading alongside the latest GM modelling and simulated ascents for SW Midlands and environs. Understandably 'rough-around-the-edges' but broad agreement for the Tues/Wed event in terms of perhaps 5-10cm snowfall into parts of the SW Midlands, much of central/east Wales and into other parts of the midlands and NW England (northern boundary remains uncertain). The current MO advisory adequately reflects the areas signalled in the GM run (expect this to be further tweaked in due course), albeit MOGREPS highlights a fair swathe of not insignificant snow probs across more northern-central stretches of the Westcountry (N Somerset, Bristol, S Glos and northwards, out through N Wilts and up into the Midlands), reflecting the somewhat troublesome pin-pointing of snow risk at a sub-regional scale at the present timeframe. Thus the elasticity remains a forecasting headache, not least given the very prominent temperature gradient being suggested in the UKMO output, which will promise hefty rainfall totals into much of far SW / southern England and a really knife-edge discrimination of where this turns (quickly?) to snow on the leading edge. Interesting stuff. The latest 6-10 day analysis looks equally spicy and will doubtless prove more than a tad newsworthy for various regions.

Best

Ian

Thanks Ian,

I note with interest that, as always, calling the boundary between rain and snow is very difficult, and it's at this point that the FAX charts and personal interpretation ( and experience) come into play.

I still wonder about the positioning of the 528 DAM, it looks too far north on everything I've seen. Also for those saying 200 miles is a huge difference, we can drive that distance in 4 hours, even on our wonderful roads. Even the highest resolution models only go, I believe, to 8km and the normal GFS is lower resolution than that.

Finally, rain or snow, what will happen will happen, enjoy whatever you get

[mcweather]

Agreed.

The latest UKMO MOGREPS probability output for spatial extent (and intensity) of snowfall makes interesting reading alongside the latest GM modelling and simulated ascents for SW Midlands and environs. Understandably 'rough-around-the-edges' but broad agreement for the Tues/Wed event in terms of perhaps 5-10cm snowfall into parts of the SW Midlands, much of central/east Wales and into other parts of the midlands and NW England (northern boundary remains uncertain). The current MO advisory adequately reflects the areas signalled in the GM run (expect this to be further tweaked in due course), albeit MOGREPS highlights a fair swathe of not insignificant snow probs across more northern-central stretches of the Westcountry (N Somerset, Bristol, S Glos and northwards, out through N Wilts and up into the Midlands), reflecting the somewhat troublesome pin-pointing of snow risk at a sub-regional scale at the present timeframe. Thus the elasticity remains a forecasting headache, not least given the very prominent temperature gradient being suggested in the UKMO output, which will promise hefty rainfall totals into much of far SW / southern England and a really knife-edge discrimination of where this turns (quickly?) to snow on the leading edge. Interesting stuff. The latest 6-10 day analysis looks equally spicy and will doubtless prove more than a tad newsworthy for various regions.

Best

Ian

Many thanks for that Ian.

Nice to have some input from someone with the very latest output. I look forward to the spicy 6-10 stuff as well.

Thanks for taking the time to post

[Glacier Point]

Well the operationals have gone on their traditional festive tangent, but study of the ensemble means has continued to provide good overall guidance on the longwave pattern, which continues to see high latitude blocking and 850 hPa temperature anomalies across NW Europe ranging between -1 and -9C over the 15 day period.

The West-based -NAO is depicted by GFS and GEM H5 ensemble mean height anomalies for days 11-15:

http://raleighwx.easternuswx.com/models/cmcensemble/00zCMCENS11-15day500mbHeightAnomalyNH.gif

http://raleighwx.easternuswx.com/models/gfsensemble/00zENS11-15day500mbHeightAnomalyNH.gif

Crucially, that is a flat jet pattern with both model tools extending the trough into the core of Europe which will allow and retain a reservoir of polar air over NW Europe. Prior to this timeframe, both GFS, GEM and ECM ensembles suggest anything between -4 and -9C 850 hPa anomalies to become established across much of northern Europe with southern parts of the UK along the battleground between airmasses.

The teleconnective situation remains unchanged. The AO is severely depressed beyond -4SD. The NAO is also becoming locked negative. Trade winds are increasing ahead of a convective signal returning to the western Indian Ocean which ties in with increased negative mountain and frictional torques sending the Global Wind Oscillation towards phases 7-0 favouring a west-based NAO block. Due to increased angular momentum, fluxes in the coupled ocean-atmosphere system are strongly favouring the northern block to wobble around the North Atlantic, east to west, west to east and so forth for the duration of January reaching a likely peak intensity January 10th - 25th. The sustained outlook I would suggest, is for below or much below average temperatures to prevail.

[huntso]

Was wondering myself when the weather for the week ahead would be on, no Countryfile this evening either

[TEITS]

I know the links below are for the 2m temps on the ECM and the 850s for the GFS. However there is alot of disagreement between the ECM/GEFS ensembles.

http://www.knmi.nl/exp/pluim/vijftiendaagse/Data/15DAAGSE_06260_NWT.png

http://www.wetterzentrale.de/pics/MT8_De_Bilt_ens.png

I personally think some of these problems are caused by how the GFS has been programmed. Due to our default weather pattern being mild SW,lys once we see the reverse of this the GFS struggles. This zonal machine will always attempt to blast any blocking away and remove any cold pooling ASAP. When you have the AO off the scale this will always cause problems.

Im not surprised NOAA nearly always sticks with the ECM. I tell you what if we didn't have access to the GFS then following the models would be alot easier and less stressful!

[BARRY]

country tracks is not on today is it?

forecast after the news at 11.55

[snowmadchrisuk]

I Know some of you think it is a waste of time using some of the smaller models but in this case the point is with regards to the cold air every model (except GFS) is showing pretty much the same thing. From a Midlands point of view these charts give us a great chance of Snow .

The GEM Model :

The JMA Model :

The GME Model :

The UKMO Model :

The ECM Model :

As you can see from these charts , they all go for -5 850's across the Midlands and in some cases further South. I do not think the GFS will be correct on this one , but if it is the new version of the GFS will be a brilliant bit of kit as it goes against every other model at regional level.

[londonsnow]

country tracks is not on today is it?

no, but i believe the weekly forecast will be tagged onto the 11.55 bbc news

[TEITS]

For this to happen they must be considering the milder air also getting reasonably far north, before everything turns around and heads south again later in th week.

I wouldn't imagine too far N otherwise there wouldn't be warnings for Bristol, Bath etc. However I doubt the snow would effect Lancs & Bath so they are covering all bases.

[Paul_1978]

Here is what GFS 06z last Sunday was predicting for this Sunday

Here is the reality

And there is one reason why I try never to look more than 7 days ahead on the GFS! Everyone should take note!

[snowlover2009]

why are people so downhearted? it is only one model that is out on its own going for the mild weather next weekend. ecm and ukmo the biggest models keep it cold and the ensembles. looking at gfs, if there ever was missing data i would have thought the data was still missing looking at those charts.

[Frosty.]

The gfs is coming in for a lot of stick today, I admit to not liking the 00z gfs output this morning and even less liking for the 06z run but it could be right and the other models could all be wrong. We may get very close to dream winter synoptics and yet just miss out by a small margin but that's life..there are reasons to be optimistic with the ecm and gem in particular, the gem 00z is a stellar run but we really have to just wait and see. The potential good news could come from the comparison chart which Mr_Data posted up for the gfs T+168 projection from 7 days ago, you will all notice that it got that a bit wrong.