Winter Model Discussion 18Z 8/1/13. The hunt for cold (is closer).

[adalton]

Irish Met Update seems to also be ignoring the GFS....

Sunday will start mainly dry and bright with frost and ice clearing slowly but temperatures will struggle to rise significantly. Winds will back northwesterly, light to moderate during the day. During the evening, outbreaks of rain will turn persistent in the northwest and west, then becoming more widespread across the country as the night goes on. As the rain comes further east, it will be likely to turn to sleet and snow, especially on higher ground. Widespread rain and sleet is likely for much of Monday, with snow possible down to low levels and perhaps accumulating in places. This will clear overnight into Tuesday; severe frost will result and ice or lying snow will persist. Tuesday and Wednesday will both be very cold with ongoing severe night frosts persisting into daylight hours. Most of the country will be dry but easterly breezes are likely to bring wintry showers to eastern counties later with further snowfall possible there.

[rory o gorman]

The irish met use the ECM only. Dont even look at the gfs in creating a forecast.

[Met4Cast]

Not to any great degree though? And it wasn't without support.

The milder run isn't without support either, though. I don't think we can ignore the GFS because it's an "outlier" when we've practically been warshipping the GFS in it's last few runs because it's shown us colder weather.

The ECM and UKMO remain good for cold weather so we can take some comfort in that, we also have the GFS Ensembles trending colder, ironically just when the Op decides to throw a couple of milder runs out. Hopefully the GFS is having a bit of a tantrum and will be back on board later today.

[BLAST FROM THE PAST]

By a clear rank mild outlier? come on fred, cheer up, deep cold is coming, the words of netweather and the majority of the 00z models with the biggest plus being the even more wintry met office upgrade hot off the press.

Yep, I have my overall outlook which is block to show/form/get attacked from WSW and GHP to reassert and strengthen from say 22nd ridging SE, I see the models BUT I'm a tad concerned as RJS has been even closer this winter. The UKMO model creeks a bit. Maybe just the jitters. However its within my view that there will be a LP squeeze through to the NW of Scotland early next week and this has always been a worry as if we don't see it play ball and we don't se the HP link up to Greenland to from a block, then it could go the way of the GFS. No panic, just musing here.

BFTP

Edited January 9, 2013 by BLAST FROM THE PAST

[mulzy]

Cold ECM London Ensembles (though not as cold as last night's set)

[Glacier Point]

If I could allay a few fears here..

The 10 hPa and 100hPa operational 06z still looks consistent with overall strat developments, and general signal for ridge in the NE Atlantic locale.

t300 GEFS mean height anomaly very well defined, ridge signal eastern side Greenland and deep trough over Europe including UK.

The operational GFS and some ensemble members maybe picking up a reasonable idea for shortwave development between Iceland and Greenland, which makes next week's forecast that bit more problematic. Clearly the quicker the cold uppers become embedded the better for snow prospects next week. Interestingly NOGAPS which had championed this idea has now binned it 06z. Just demonstrates the massive spreads that currently exist in the North Atlantic in the medium range and large scale inter-run disagreement in NWP.

That doesn't alter the general trend here. Cold to begin this weekend, continued signal for a slight moderation in this before cold becomes more aggresive once more, timed for second and most significant downwell from the SSW. Note how the GEFS and also ECM ensembles begin to re-intensify cold beyond day 10.

Bottom line - don't mistake a tendency to moderate cold next week (and undoubted model variability to expode that out of all proportions) with a failed cold spell.

[MR EXTREMES]

Exactly right. What I have found in all the years of model watching is +72 and below is when the models show much greater consistency. When you go from +72 to +144 this begins to fall away especially between +120 & +144. Obviously it also depends on the projected pattern because sometimes the consistency can vary between this period. So based on this we still have plenty of time for the models to trend either way.

Just to illustrate my point about +72 and im sure some of you like me have told friends and family about impending cold spells which have gone pearshaped? Now in recent years I have learn't to keep quiet and I never tell anyone until a cold spell is being projected at +72. This makes sense really because apart from the BBC weekly forecast on Sunday they rarely forecast beyond 3 days.

Back to the models and another scenario which nobody has mentioned is just maybe we shall see a blend of the ECM/GFS. Now you may ask how can we have a blend when they are so different. Well my answer is possibly we could see a brief return of milder weather into next week which is swiftly replaced by a return to colder weather.

which i might add is pretty normal for uk winter well a cold winter lol.

[tight isobar]

Hope this helps, with a bit of context for you too:

Outlier

Technically, it's a term which mathematicians - like moi - or statisticians attach to a reading which doesn't tie-in with the rest; a bit of a black sheep which has got lost somewhere. In meteorology, the term 'outlier' is be used in the context of assessing support for the various model runs. These model runs (sometimes up to 50 different ones) are compiled into ensemble charts, so that they can be visually analysed against one another. It is at this point that you'll see those 'black sheep', where a run goes off on one, relative to the rest. In the main, those are then discarded as outliers, ie: runs without support of the collective suite.

Shortwave

Essentially, these are embedded flow disturbances which disrupt the overriding synoptic pattern. In meteorology, it's wise to bear in mind two patterns: the overriding parental one (zonal, blocked, etc...) but then there is also a child meso pattern, which is often most influenced by shortwave disturbances. Shortwaves are notorious for their synoptically subtle influences which - once amplified - can, and often do, completely change a pattern. To spot them, look for little hook/kinks in the isobars. But shortwaves are not always bad news, and sometimes synoptically you need them, in order to progress or move to a more advantageous pattern. But timing and location is absolutely crucial, and this is what can cause such utter frustration. Nick S loves a good shortwave

Toppler

A term only ever used in reference to Northerly's. Essentially, a 'toppler' is a short-lived - or even non-existent Northerly - which is synoptically spoilt by high pressure ridging from the Atlantic. A toppler can, nonetheless, 'produce the goods' - only that it's short-lived (perhaps 24-36hrs) and often belies a fairly mobile pattern. Not the best winter synoptic, although I'd hazard a guess they're fairly common for the UK; at least, more common than, for instance, an Easterly flow - which, by contrast, is characteristically colder and has much more longevity associated with it.

Hope that helps you - and indeed others - with a few of the myriad of meteorological terms.

SB

nicely summed up.:-).

[WhiteFox]

Trouble is, the GFS had been a cold outlier for several runs against it's ensembles

The Op run was indeed generally amongst the coldest prior to its latest wobble and I had a feeling that we would soon see a warmer run on the operational. What I din't expect was to see the Op run flip to the mild options and the ensembles start trending back to cold!

Anyway, the huge scatter on the ensembles only highlights the undertainty in the models. I don't think we will have a clear idea of next week until this Friday as the timing of the potential jetstream split starts moving into the higher resolution models and we have an idea of the starting point.

[nick sussex]

Okay I've looked at all GEFS members and the better synoptics for the UK are correlated with whether the models pick up energy from the sw USA and transport this eastwards.

I shall explain with the following maps,

GFS 96hrs

GFS 120hrs

Now look at the eastern USA and further to the sw, note the GFS brings one shortwave ene and picks up energy from the sw transporting this also eastwards.

You must if you want to see the ECM solution verify have a single low with no trailing energy:

Here you can see the ECM 96hrs

ECM 120hrs

You can see the ECM brings the low ene separating out the energy, it transports it eastwards as a single low with no trailing energy.

Once the shortwave picks up that sw energy this will alter its track and development.

Edited January 9, 2013 by nick sussex

[Southender]

If I could allay a few fears here..

The 10 hPa and 100hPa operational 06z still looks consistent with overall strat developments, and general signal for ridge in the NE Atlantic locale.

t300 GEFS mean height anomaly very well defined, ridge signal eastern side Greenland and deep trough over Europe including UK.

The operational GFS and some ensemble members maybe picking up a reasonable idea for shortwave development between Iceland and Greenland, which makes next week's forecast that bit more problematic. Clearly the quicker the cold uppers become embedded the better for snow prospects next week. Interestingly NOGAPS which had championed this idea has now binned it 06z. Just demonstrates the massive spreads that currently exist in the North Atlantic in the medium range and large scale inter-run disagreement in NWP.

That doesn't alter the general trend here. Cold to begin this weekend, continued signal for a slight moderation in this before cold becomes more aggresive once more, timed for second and most significant downwell from the SSW. Note how the GEFS and also ECM ensembles begin to re-intensify cold beyond day 10.

Bottom line - don't mistake a tendency to moderate cold next week (and undoubted model variability to expode that out of all proportions) with a failed cold spell.

Thanks GP.

So in short a cold spell this weekend for a few days, turning slightly more average for a few days, then the potential for a proper cold outbreak thereafter.

[smhouston]

Surely the 06z isn't an outlier? Yes it's at the top end of the ensembles, but there are other ensembles that support it:

Obviously a massive split though

[bradymk]

@GlacierPoint Thanks GP, interesting that nogaps 00z picked this up and dropped it on 6z. Just a quick question, are you leaning towards ECM or GFS at this stage regarding next week.

[Southender]

Surely the 06z isn't an outlier? Yes it's at the top end of the ensembles, but there are other ensembles that support it:

Obviously a massive split though

Obviously it has support. But it is an extreme mild solution.

I would say a "warm up" of sorts looks possible mid-week, still below average conditions. But possibly transitionary as we head back towards colder conditions by the following weekend. Which could be deeper and more prolonged.

Edited January 9, 2013 by Shrimper

[karlos1983]

Obviously it has support. But it is an extreme mild solution.

I would say a "warm up" of sorts looks possible mid-week, still below average conditions. But possibly transitionary as we head back towards colder conditions by the following weekend. Which could be deeper and more prolonged.

Well if that's what it takes to get to deep cold then I'll take it thanks very much. We all know getting to deep and prolonged cold is very tricky for our Island in terms of everything falling into place.

[WhiteFox]

Obviously a massive split though

That's the key and has been during this period of turmoil! Looking at the ensemble graph all you can say is that the outlook past Monday is highly uncertain (based on this run of course). The scatter after Monday says ignore all charts after this point until we get a clearer picture.

[Glacier Point]

@GlacierPoint Thanks GP, interesting that nogaps 00z picked this up and dropped it on 6z. Just a quick question, are you leaning towards ECM or GFS at this stage regarding next week.

70% ECM 30% GFS

[Bobby]

After looking through the GFS 06z ensembles while there are some bad runs there are plenty of good runs that go against the OP and with the other models - and there are signs also that even it goes pear shaped early next week there is a strong chance of the cold building again later on. Combined with the other models, ensembles and Pro comments I think things are still looking pretty good. I think the 12z runs today will be pretty good and won't go too much towards the GFS OP.

Was getting a little concerned but those ensembles have plenty of cause for optimism I think combined with everything else.

Very promising upper mean at the end of the ensembles

It could be an awful lot worse.

Edited January 9, 2013 by Bobby

[phil nw.]

There`s usually one model or run that tries to throw us wrt the pattern.Previously it was the ECM and today it`s the 06z GFS operational that looks out of kilter with all other output.

I use the same guidance as always by looking at the trending in the mean outputs day to day alongside Height anomalies.

Here`s some images to show nothing has changed for the medium term,certainly up to 10 days.

The latest means from both runs at 850hPa for a week today.

Just to compare the 006Z gfs Op.

and the latest mean Height anomls for day 10

still a solid -AO and an increasingly-NAO with falling hts over Europe.

To underline how even the 00z GFS Op was unsupported by it`s own Ensls here are the Index graphs

enough blocking to send the jet south with the UK looking cold.

In the short range as well we have good agreement for the cold air spreading west.

We can see from the latest fax and GFS image it`s on it`s way west over the weekend.

as low pressure starts to undercut the building heights to our north.

In summary still some uncertainty with the placement of the surface features ie the exact track of those little lows coming south east but with enough of a mean block modelled to our north the overall pattern still looks cold for next week and beyond.

Edited January 9, 2013 by phil n.warks.

[nick sussex]

@GlacierPoint Thanks GP, interesting that nogaps 00z picked this up and dropped it on 6z. Just a quick question, are you leaning towards ECM or GFS at this stage regarding next week.

The NOGAPS clearly shows the importance of whether one low moves cleanly east or whether you get one with another shortwave transported along at the same time.

[Robbie Garrett]

Hello,

Further to my enquiries. The MetOffice have said the following.

"Looking at the output this morning we see that the ECMWF has very definitely switched to a much colder

solution, whilst the GFS has gone milder, which was a big surprise. As a result we are likely to favour the colder ECMWF solution, although perhaps not as extreme, but it does not make the uncertainty any better."

[MR EXTREMES]

nogaps looks good

[alan776600]

I was wondering if someonme can help me. Ive been watching this forum for a couple of years now and one thing I cant understand is how people will know it will be very snowy from friday night until Sunday? From going by what alot of you have said over time, snow is very hard to predict and normally can only be predicted in a 24h time scale. Many thanks in advance , Alan

[Wivenswold]

I guess this uncertainty is why we love model discussion in the first place. I'd love to be a fly on the wall in mission control, Exeter at the moment.

All of this will, in the long term, help the weather community's understanding of SSW events and also identify weaknesses in the super computers' performances. Every blip like this is for the long-term good of forecasting.

Imagine how dull it would be if we knew, 21 days in advance, that we'd have 10cm of snow between 6pm and 9pm tonight?

Edited January 9, 2013 by Wivenswold

[Zakos]

I think the models have picked up on a signal for an Atlantic attack of sorts. This does not by any means mean that the chance of cold is reduced.

the met have clearly binned the GFS 06z, however they have stated in their output that milder conditions may spread into the far west at times.

The atlantic almost always attempts to attack a block, but I really dont see the atlantic winning this battle. it may have won by and large in December, but the odds are now stacked heavily in favour of northern blocking.

The GFS, as usual has completely overreacted to this signal, and decides to have South Westerlies all the way across Scandinavia.Not a chance.

I wouldn't rule out an Atlantic interlude, although in my view this is unlikely in the foreseeable timeframe.

The 06z however has the cold air completely brushed aside by Monday. I'd bet very heavily against that happening

Edited January 9, 2013 by Zakos