General Model Output Discussion

[Cloud 10]

Here it is: http://en.vedur.is/w...ntic/#type=prec

Karyo

thankyou.

[johnholmes]

I continue to be amazed at people complaining about precip forecasts for 24 hours and in some cases even further out.

Get real-they are no different in predicting showers or precip totals in the summer so why highlight winter?

Weather models are fine for giving 'an idea' of when and where precip is likely but no more than that. At T+24 they will normally give a better indication of spread and intensity. As to forecasting snow then THINK how many different variables there are to get right. In summer there is just one-is the cloud thick enough to give rain.

In winter that is the start,

then the zero isotherm,

your height above sea level,

closeness to the sea,

850 mb temperatures,

1000-850mb thickness,

and to a lesser extent the 1000-500mb thickness,

temperature,

dewpoint,

wet bulb temperature

etc

get the idea?

Its immensely complex on top of the complexity of modeling the overall weather pattern correctly in the first place.

Use sat piccs-radar, especially NW 5 minute version-use the Extra/NMM charts.

on a more positive note-since using NW Extra, for the life of me I cannot recall seeing daytime max and min values over such a long period being predicted to be so low-quite amazing predictions for cold and it will be interesting to see how close to the mark they are.

A truly fascinating winter spell developing for pretty well every part of the UK.

Edited January 2, 2010 by johnholmes

[Guest North Sea Snow Convection]

The upcoming sequence of this cold spell is not going to have the depth of cold that Jan 87 or Feb 91 has (you only have to look at the 850 temps out to t168 to see that on the ECM for eg) but in terms of length it is certainly set to be longer and more sustained than either of those. In terms of what has already been since before xmas and the fact that some parts of northern Uk have never realy got any milder, plus the recent New Year snows, then the case to be said here has even more grounds

[johnholmes]

To be fair the ECM is not a short-range forecasting tool. It was not designed for that purpose.

The ECM is a medium-range forecasting model best utilized for the 72 - 168hrs range if I am not mistaken.

the clue is in the name

European Centre for Medium Range Forecasting

It was set up to run forecasts, chiefly from T+72 to T+240, to try and make forecasts in that period of the same accuracy as the period T+12 to T+72.

Obviously they run forecasts prior to the times I've given but their main emphasis is on the 72-240 period.

[karyo]

the clue is in the name

European Centre for Medium Range Forecasting

It was set up to run forecasts, chiefly from T+72 to T+240, to try and make forecasts in that period of the same accuracy as the period T+12 to T+72.

Obviously they run forecasts prior to the times I've given but their main emphasis is on the 72-240 period.

You mentioned the possibility of snow showers affecting the northwest 2/3 days ago and I think you were the only one here! This I suppose proves once again that human input and experience is very important, regardless of what the model precipitation charts show!

Karyo

[meerkat63]

the clue is in the name

European Centre for Medium Range Forecasting

It was set up to run forecasts, chiefly from T+72 to T+240, to try and make forecasts in that period of the same accuracy as the period T+12 to T+72.

Obviously they run forecasts prior to the times I've given but their main emphasis is on the 72-240 period.

Thank you for the clarification there John.

I knew that it was designed for up to at least T+168 although I was unsure how far beyond that its capabilities extended.

By the way - and off topic, I subscribed to NW Extra the other day - with radar and NMM charts. First class and well worth it. I know that several others have said the same thing but I must concur!

[Thundery wintry showers]

On my update to the monthly forecast (which also gives a summary of my current thoughts- which should please a lot of snow lovers living in the east) it was quite difficult to decide which year to say "severest January wintry spell since", because January 1987's easterly outbreak was much more severe than what we have coming up- but also shorter lived. A closer potential analogue may be January 1985, with less severe but more prolonged cold, delivered by a combination of high pressure and north-easterly winds on its southern and eastern flank.

Bear in mind that there is huge uncertainty over when the mild air will eventually try to push in. I have opted for a milder second half, resulting in a monthly temperature not excessively low- though still sub 3C CET and unusual for recent years. But I am guessing there based on past analogues and some of the longer-term signals from the teleconnections and NOAA, and it's quite possible that the milder weather could be delayed by a week or more, in which case we could end up with a January quite similar to that of 1985.

At the moment the models are consistently agreeing on a generally cold NE flow out to 7 January with the main potential for snowfalls centring on tonight and also the 5th in the north, 6th/7th towards the SE. I am seeing more and more agreement, also, that highest pressure may end up located to the north rather than straight on top of the British Isles in the longer term, allowing cold NE winds and snowfalls for the S and E- though often in this kind of setup we do pick up more modified air from around the anticyclone's periphery which would turn it less cold towards the NE, albeit maintaining cold snowy weather for the SE.

Edited January 2, 2010 by Thundery wintry showers

[kold weather]

Something to note TWS with this set-up is we've had constant -ve NAO which has several times deeper in FI been very under-estimated, until we see some big shift away from the current signal we have in the N.Hemisphere we are going to keep some sort of upper high near the UK...I also would expect an eventual warm-up but the GFS ensembles are once again way too keen on blowing up the low pressure systems and I'd expect the 0z ECM will be close to the eventual evolution which is agreed upon by the 06z GFS upto at least the 12-13th.

If I had to make a punt, then I'd say the upper high will transfer back towards Iceland, maybe a little less cold between say the 12-15th but if we can get some sort of PV movement back towards Scandinavia then I really wouldn't be surprised to see cold air coming down for one final blast before the Hemispheric conditions start to slowly move away from the ideal set-up we have now...I am very happy with my call 5-7 days ago when the models suggested a Euro high developing, I called for a upper high to build near the UK and build towards Iceland, not far off from what the models now suggest.

[Stu_London]

The upcoming sequence of this cold spell is not going to have the depth of cold that Jan 87 or Feb 91 has (you only have to look at the 850 temps out to t168 to see that on the ECM for eg) but in terms of length it is certainly set to be longer and more sustained than either of those. In terms of what has already been since before xmas and the fact that some parts of northern Uk have never realy got any milder, plus the recent New Year snows, then the case to be said here has even more grounds

Indeed, the cold spell will probably go down in history as being at least a month long.

It could be argued that the cold spell started on 14th December and we have now had snow reported as falling somewhere in the UK on 18 consequtive days.

Temperature wise, whilst we haven't seen some of the extremes that people will remember (from the 60s and 80s in particular) it sits ok with anything we have seen since 1992.

Snowfall distrubution will be disappointing for some (not me) but should improve over the next week

[VillagePlank]

Morning, all.

(Sorry about links rather than pictures - uploading images seems to be failing at the minute)

Firstly, I'm very pleased that the Scandinavian High has finally reared it's head on the model output. I felt that GFS wasn't quite cooking the

books properly (as well as other models) by leaving it out.

Right, onto the 6z:

A nice ensemble set with the operational showing that is well in the mix. Perhaps a tiny little bit warm at times (compared to the mean) but, I suspect, this is down to marginal differences with the pressure centres over the Atlantic, and Europe. Main point is that, firstly, this run looks reliable, and, secondly, surely the headline feature of any forecast made, today, that we are going to experience temperature, at 850hPa, of around -10C from the 4th January to the 11th January, with the warmest member reaching a maximum of -5C, and the coldest plummeting to -15C. Is this a first for the internet age? Can anyone recall such a prolonged and cold spell?

Overall Synopsis

Strongest part of the Atlantic jet stream is due to stay very south for the entire forecast period (T0..T180). However, a complex picture emerges which looks to set model watchers at the edge of their seat. The tropopause height (300hPa) is set to fall over Greenland, and increase mid-atlantic - but by marginal amounts. Centres of cyclogenesis, accordingly have all moved a little bit south (one marginal exception) with the Eastern-Seaboard (USA) seeing the strongest cyclogensis by far. As the forecast period rolls on through the jet-stream over Europe stays south and strengthens from time to time. However, it does get weak overall as a trend, which implies a return to high pressure over Europe, especially considereing the height falls over the polar region, in the long term (two-three weeks time)

The one exception is a small bifurcation of the jet-stream, throughout the period, that appears to our immediate NW, and Iceland. This is what will be partially responsible for the dropping pressure over Greenland, and it will also spin up some low pressures, or even just small troughs, that will zoom over the UK (northern parts mainly) One to watch as this could give North and Western parts a nice (or otherwise) surprise that might only be picked up by the models a day or two in advance.

Overall Synoptics

A transient Atlantic block emerges as heights in this area increases comparitvely to surrounding areas. This is not likely to last long, but it does set up the main message from GFS: Low pressure to dominate Europe for the forseeable future (less than two weeks) This sets the UK up nicely for northerly and easterlies, and depending on resulting shape of the pressure centres north-westerlies for the forecast period. Very late in the forecast period, watch for high pressure to form and move over the northern part of Britain, and pressure to drop over Greenland. This is the result of the weak jet-stream slither appearing to our north-west. This is certainly not set in stone (T=FI) but it paves the way for a change in synoptics. ie the Scandi high builds for a time, emerges over UK, before collapses to create a Northern block, and allows Greenie high to re-emerge. This is what GFS says, and, in my opinion, it's not convincing, since the low pressure areas form without the support of the jet-stream - so what causes cyclogenesis?

A better bet is to take a look at the weather over the next three days and assume that that's what it will be like for ten days give or take a snow shower here, or a snow shower, there.

Main risk for end of cold spell is high pressure blocks forming around the UK, and then dominating the UK for a time. Not convinced, to be honest, but it's still there.

More later when I can figure out how to upload images!

Edited January 2, 2010 by VillagePlank

[Thundery wintry showers]

Just seen that extended ECM output - a real "armageddon" setup for south-east England and East Anglia in particular. I am not expecting something as pronounced as that to come to fruition (I think high pressure will tend to be a little further south) but it is notable that the models in general are trending towards a prolonged spell of east-north-easterly winds, again quite similar to January '85.

[pyrotech]

Well all the models show a trend into fi of cold. What a fantastic set up. Like john so sensibly said percipitation from 24 hrs plus is not worth worrying about. Last night we had a nice surprise down in Cardiff. These set ups for such deep cold are rare in the last two decades and what the models show is something to be totally pleased with. The prcipitation will come and due to the cold any lying snow has more chance of sticking around longer than a few hours that we are so used too.

What the models do show is very cold dangerous conditions. Cutting people out of cars is the worst part of my job, this morning many roads and a main duel carriageway into Cardiff were completely closed due to numerous accidents on the ice. what i see on these models today is percipitation falling at short notice on frozen surfaces. I can say with experience that gritting can fail for patches of road due to lumps of salt clogging up momentarily or percipitation washing it away from road then freezing.

It is not the white snow covered roads that catch you out as you can see it, its the normal roads you think are fine, speed up and it catches you out.

This is the weather i love cold and possibly snowy and the models show a dream come true, but there really is another side to this forthcoming period so when you look at the models and see this fantastic week plus ahead, take in what your seeing and remember it when driving. Rant over ( just horrible to attend these incidents)

The fax charts show interesting developments towards the middle of the week for Wales and Southwest (IMBY) I mention that as rest of the UK looks pretty much fantastic for snow opportunities throughout the next few days.

Huge agreement too 120hrs then different ways of keeping the cold in place, but importantly the consensus is that the cold trend continues. Infact the ECM and GFS are extremely similar in that time frame and then heights over scandi are again similar after but even thou models are consistant with that its fi. Huge agreement of a sustained cold spell and snow for many or even most.

NAE charts show good agreement with the two main models in near timeframe and assembles are coldest that i can remember. fantastic spell of wintry weather ahead and all models singing from same songbook. Any breakdown is well into FI and does not deserve mentioning except to say that any breakdown whenever it comes is going to be exciting.

Looking further ahead this cold spell could get yet another reload, in FI 240hrs pressure tries to rise in the North and fall over Southern Europe. Its something that we need to watch. I believe in 63 it was a case of a cold 3 months with reload after reload and although i am not jumping on the 63 bandwagon there is potential for a cold spell with less cold ( not mild) throughout this month. Prediction for the main Jet to move North over the UK is a short lived spell deep in Fi and i think we will not see it at all. What ever fi brings these charts are to be savoured and kept to look back on.

[Thundery wintry showers]

For those who don't remember much about the January 1985 setup, here's some charts which show similarities with the upcoming synoptics:

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850107.gif

Cold NE'ly, high sinking from the NW.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850111.gif

High over the UK, interestingly a little further south than the models currently show.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850114.gif

High drifted N, giving very similar charts to what the models are frequently showing at around T+168.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850117.gif

High retrogessed NW to Greenland (interestingly along the lines of what Kold Weather mentioned) with slightly less cold air heading in from SE.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850123.gif

The later progression of Jan '85 had the lows push in from the SW, but these temporarily led to a cold NW'ly with snow showers for northern and western areas.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850202.gif

It turned mild with westerly and south-westerly winds into February 1985.

http://www.wetterzentrale.de/archive/ra/1985/Rrea00119850209.gif

then towards mid Feb 1985 another easterly episode resulted.

That last bit also highlights something that was going through my mind when reading Glacier Point's teleconnections analysis towards February, suggesting the real possibility of a very mild one. If we have the blocking patterns getting a shift on, and some bursts of very mild air and pools of very cold air shifting around the Northern Hemisphere, it doesn't necessarily mean the UK can't get hit by cold pools from time to time, so while much milder for February looks a safe bet, mild and snowless is somewhat less so.

[Thundersquall]

Great 06Z run from GFS which shows this cold spell lasting until at least 16th January.

Basic synpotics are Greenland High and Western Europe depression to out southeast, lasting until about Friday 8th

January, then a Scandi High takes over and linking to another Greenland High by mid-January,

the low to our SE lasting until 10th/11th January, then replaced by low to our SW.

I am taking no notice at the moment beyond 16th January, as this is way, way too far ahead.

Greatest chance of snow for the Southwest, West Country and Wales seems to be Tuesday night 5th into Wednesday 6th Janaury.

Edited January 2, 2010 by Thundersquall

[rbw]

For those who don't remember much about the January 1985 setup, here's some charts which show similarities with the upcoming synoptics:

http://www.wetterzen...00119850107.gif

Cold NE'ly, high sinking from the NW.

http://www.wetterzen...00119850111.gif

High over the UK, interestingly a little further south than the models currently show.

http://www.wetterzen...00119850114.gif

High drifted N, giving very similar charts to what the models are frequently showing at around T+168.

http://www.wetterzen...00119850117.gif

High retrogessed NW to Greenland (interestingly along the lines of what Kold Weather mentioned) with slightly less cold air heading in from SE.

http://www.wetterzen...00119850123.gif

The later progression of Jan '85 had the lows push in from the SW, but these temporarily led to a cold NW'ly with snow showers for northern and western areas.

http://www.wetterzen...00119850202.gif

It turned mild with westerly and south-westerly winds into February 1985.

http://www.wetterzen...00119850209.gif

then towards mid Feb 1985 another easterly episode resulted.

That last bit also highlightss something that was going through my mind when reading Glacier Point's teleconnections analysis towards February, suggesting the real possibility of a very mild one. If we have the blocking patterns getting a shift on, and some bursts of very mild air and pools of very cold air shifting around the Northern Hemisphere, it doesn't necessarily mean the UK can't get hit by cold pools from time to time, so while much milder for February looks a safe bet, mild and snowless is somewhat less so.

Interesting - whilst very much a weather enthusiast at the time I don't recall much that was spectacular in Jan 85 - I guess there wasn't a lot of snow around so I didn't really notice

R

[VillagePlank]

is it me or are models generally showing dry weather with just favoured coastal areas with some snow showers?

i see no eveidence of countrywide snow yet.

pressure is just too high or at best flabby

Too early to tell with any sort of certainty, I think. But, given that it's a North/Easterly, then it stands to reason that the North and East are more likely (but not certainly) to be in the firing line.

[SizzlingHeat]

is it me or are models generally showing dry weather with just favoured coastal areas with some snow showers?

i see no eveidence of countrywide snow yet.

pressure is just too high or at best flabby

I wouldnt suggest pressure is too high, especially mid-week with low pressure and troughs spreading across from the north sea from NE to SW and this could bring snow anywhere across the country. Troughs and streamers may affect a fair few places in the south and east and the band coming down the western side for wales and SW on tuesday night.

Pressure depth is irrelevant if we get an easterly, those 850s are 10-14c will give significant convection and spread far inland over the UK.

The cold is here, snow will come later with unexpected troughs batches of showers

[grab my graupel]

Please find a new thread for the 12z run