Winter Model Discussion 12Z 22/1/13 onwards.

[TEITS]

Rather than posting endless depressing charts I shall just post one this morning.

http://cdn.nwstatic.co.uk/ensimages/ens.20130124/00/prmslReyjavic.png

Pretty poor if you're looking for blocking to develop within this region. Only two members reach 1030mb.

So at it stands this morning a continuation of W,lys looks likely within the short, medium range.

[Gibby]

Good morning folks. Here is my take on the 00z outputs from GFS, UKMO, GEM and ECM for this morning Thursday January 24th 2013.

All models show a cold and static airflow over the UK today and at fiest tomorrow as a weak area of high pressure slips South down the North Sea. After a rather cloudy start for many today the cloud should break latr with some sunshine. It will stay very cold with a sharp frost tonight for many. Tomorrow an active front moves into the UK from the West bringing increasing Southerly winds and rain to Western fringes in the day. Later this extends steadily East turning to snow as it does so, with some heavy and disruptive snow for a time in Northern and Eastern areas. It will clear from the West later and become less cold as winds veer West and decrease temporarily. Through Saturday as this trough staggers away East and decays a new vigorous depression moves into the NW with strong winds and heav rain for all this time over Saturday night bringing a rapid thaw to all before clearer, windy and showery weather arrives from the West on Sunday with some heavy and wintry showers in the North and West in a cold and blustery West to SW wind.

GFS then takes us through next week with a lot of windy weather. It will be come much milder, relative to conditions now with temperatures close to the seasonal normal or a little above in the SE for a time, though always offset by the wind. Rain at times would prevail for all, heaviest in the North and West. Through FI today the trend is to maintain the unsettled and windy theme with temperatures slowly dropping to just below the seasonal normal with further rain at times and no doubt some snow on Northern hills.

The GFS Ensembles confirm what I have stated above with the operational being a good representative of he pack. Rain at times is the order of the period with the Atlantic well in control throughout.

The Jet Stream flow shows an angling of the flow NE in the coming days towards the British Isles and later towards Northern Europe before settling to a typical sine wave style pattern at around 50 to 55 deg North by the middle of next week and beyond.

UKMO for midnight on Wednesday shows deep Low pressure to the North of Scotland with a spell of very windy and unsettled weather well entrenched over the UK. It will be often cloudy with spells of wind and rain driven quickly East over the UK with brighter and more showery spells in between. Temperatures would be near to normal but cold enough for some wintry showers at times on Northern hills.

GEM shows a windy spell of weather too with winds always between South and West carrying troughs NE over the UK with spells of rain followed by fresher showery weather with temperatures well up to the seasonal normal or a bit above in the SE for a time.

ECM this morning shows an unsettled pattern too with a West or SW flow bringing rain at times in normal temperatures. Later in the run the flow tilts more West or just North of West with a trend to rather colder air with some wintry showers on hills later. This trend continues at Day 10 with pressure having fallen in the Med and High pressure in mid Atlantic allowing the next depression to probably slip more SE directing a cold Northerly flow across the UK in the days that follow the expiry of the run.

In Summary the thaw is just around the corner and committed to by all models for late Friday and Saturday. The models then settle us into a milder and windy spell with areas of rain and strong winds interspersed by periods of sunshine and showers.in temperatures very close to normal though it will still feel chilly in the strength of the wind. Longer term GFS show little change to the progressive pattern with rain at times while ECM shows signs of change again late in its run as the jet tilts SE and colder winds from high in the North Atlantic are directed over the UK. Although not as pronounced as last night's 12z output a Northerly could develop behind the incoming depression at Day 10. The only negative though there is no Northern blocking to tap into at this juncture so the UK may have to wait at least two weeks before our next serious shot at real wintry weather again.

[wolvesfan]

Morning all,

Its been a good couple of weeks for us cold fans but it's all set to go milder over the weekend,by next week GFS goes for daytime maximums into double figures which will feel pretty mild compared to of late,

Dosn't last though as we get a succession of lows tracking close to the UK keeping things mobile and cool.NAO going positive indicative of lower heights to our NW being dominant.however it does head back neutral towards Feb.

Ensembles show the 850 temps to drop after the brief rise so nothing mild on offer for too long,

All this for me shows there's still plenty of cold thats spewed into the mid lattitudes for us to tap into.

[Mark Neal.]

NAE 6z shows The Rain hitting the colder air on the leading edge to turn to snow. Firstly at 12pm on Friday over Scotland and Northern England.

Then by 6pm on Friday evening across much of Northern England & some central areas..

By Midnight on Saturday morning the snow has reached much of Eastern England. With rain still in the SW & much of southern england.

Edited January 24, 2013 by Mark Neal.

[Summer Sun]

Shorter term ECM ensemble

Its not hard to spot the outlier there

Longer term ensemble

Looking at the NAE this cold spell is going out with a bang before rain arrives during the weekend (late Saturday early Sunday is when the heaviest is likely to arrive) and washes all the snow away increasing the risk of flooding

Edited January 24, 2013 by Gavin.

[Supercell 89]

The models are a depressing viewing this again this morning but i wouldn't be surprised if come next week temperatures and conditions are closer to average. Still don't think this winter is done with us yet and think maybe after week or so we may see some changes. Went to a climate discussion at our Uni the other day where they have been doing research on the stratosphere and what they have said about next month was very much singing from the same Hymn sheet as CH and GP and that a north-easterly pattern could well develop next month.

Still a shame that this current cold spell is coming to an end, it has been fantastic in my view. Most of the country has seen snow at some point over the last week and again most of the country has seen lying snow cover for several days at least (i know there are some locations who have been so lucky with snow) and daytime temperatures particularly away from the South-West have struggled to reach much above freezing for at least a week. With one last burst of snow for some likely tomorrow i'd say been pretty good if you ask me

[TEITS]

I pity anyone travelling on Friday evening/night based on the latest NAE. Looking at the precip totals I feel higher ground in N England/Scotland could see a foot of snow and even high totals at sea level.

http://expert-images.weatheronline.co.uk/daten/proficharts/en/nae/2013/01/24/basis06/ukuk/rsum/13012606_2406.gif

Another positive is im not expecting a rapid thaw behind the front and it will take until Sunday morning before we pull in much milder air.

[Frosty.]

I don't agree with steve, for some this has been the best cold spell for many years and far snowier than 2010 for some, for others it's been a disappointment but you will always get these regional variations in the uk. As for the rest of winter, too early to write off wintry prospects but the next week or so looks average to mild from saturday pm onwards but there are signs beyond T+168 hours of the jet tilting more nw/se and more southerly tracking with a good chance of mid atlantic ridging opening up the chance of some polar maritime/arctic incursions during feb/march, and march can produce very wintry weather if the synoptics are right, I have a feeling the last quarter of winter and early spring will make up for the relatively poor end to autumn first month of winter.

Edited January 24, 2013 by Frosty039

[TEITS]

I don't agree with steve, for some this has been the best cold spell for many years and far snowier than 2010 for some,

I feel its important to differentiate between the actual weather and the synoptics. As you rightly point out for some its been even better than 2010 and that has certainly been the case for my location with regards to snowfalls. However synoptically it hasn't been anywhere near as good from a blocking perspective. Many have complained about the lack of very cold temps and this is largely due to the synoptics because upper temps have not dropped below -9C and cloud cover has been persistant. Our snowfalls have also occured mainly due to frontal systems which always brings higher temps, dewpoints compared to a convective N/NE/E,ly.

Personally I prefer the classic convective snowfalls from the E/NE with much colder upper temps and more importantly dewpoints even if this means lower snow depths.

Back to the models and I strongly urge members to follow the other NAE runs today especially those who plan on travelling tomorrow evening or know friends or family who are. At the moment only the SW and those S of the M4 look to avoid the snow.

Edited January 24, 2013 by THE EYE IN THE SKY

[TEITS]

Well I've been looking at the charts and coupled with past experience this will be a mainly rain event. Already mentioned this earlier but message was deleted which happens a lot to me, maybe I ruffled a few feathers when I correctly forecast the snow event on the 18th January 2 weeks earlier. mod(s) delete with pleasure as its my last post anyway!

Generally past experience means very little with these types of events. So many factors dictate whether we see rain or snow and more importantly how quickly the snow turns to rain. I have known these transitions to bring only 30mins of snow followed by rain but I have also known them to be all snow followed by it turning milder the next day.

All we can do at the moment is follow the likes of the NAE and see if conditions continue to support snow. During today the NAE might show changes between runs especially where the rain/snow line will be.

As it stands at the moment it isn't until Sunday morning until the much milder air arrives.

[southbank]

we have a very interesting 48 hr period of model watching coming up , think the last time we had an active atlantic front coming up against a block was early 2010 DEC after the heavy snow on the 1st , be interesting if the front moves as quickly as models suggest or will it get slowly as it hits that block

these typpes of set up anything can happen , supposing the front stalls over the SE / eastern parts before the warmer air catches up and mixes out the cold surface?

[SnowBallz]

The irony however is that having said all of that they're still unable to forecast with any great degree of accuracy beyond 4 days at best. In the 1950's the accuracy was at 2 days.

The question is, if you take away the computer model can a forecaster forecast. If they are presented with a chart at 12z today and asked to provide a forecast for 1,2,3 days ahead without the use of a computer can they do it. Half the people in here can interpret the models to provide a forecast and how many of those are expert mathematicians, physicists etc.

All the lay person wants to know is will it be wet, warm, dry or cold, not be baffled with b******t. It's about the end product - an accurate weather forecast; from what I can see, all these advancements in technology, the sciences etc and for what...........to extend the forecasting accuracy by 2 days!

Just seen your reply.

If we look at some verification data from ECMWF - the world renowned pacesetters in mid-range forecasting - you'll see that there is a almost linear growth in forecast verification, in-line with time:

What that chart illustrates is that, back in 1998 the 80% threshold was around 5 days or 120hrs (so, incidentally, I've no idea where you get '4 days today' from) However, by mid-2012, that 80% threshold has been pushed out to 6.5 days, which is 156hrs.

What's also illustrated in that chart, is the tailing-off of growth since around the end of 2009 - with a plateau clearly evident. The reason for this is very simple: ECMWF has exhausted the capability which current supercomputing resources afford them.

There is a clear need to upgrade the hardware, rather than refining current algorithms. However, it is important to finesse algorithms prior to an upgrade, and I note with interest that ECMWF will be introducing further vertical resolution this year; therein - or at least one would imagine - delivering greater depth of data which the refined algorithms can compute.

Even the lowest level of confidence (25%) grows in tandem - now out to around 9 days (216hrs):

Again, that chart illustrates how the bar continually moves forward; 5.5 days (132hrs) in 1998 has now risen to around 8.5 days (204hrs) Again though, we see that tail-off towards the end of 2009. It could be plausibly argued that, as you test more exploratory algorithms, you will naturally increase the error-rate as it's a given that not all algorithm refinement will actually help (indeed, it may hinder - which is what I suspect this plateau effect represents)

For me, I always look for the stories within stories. For example, my interest from these kind of verification charts is more drawn to why it appears model performance markedly drops off in the summer season, relative to its winter performance. A very clear pattern there. Overall, the trajectory is positive, but the hidden pattern still needs addressing (in my opinion anyway)

As you ask (you might've guessed anyway) but I am indeed an expert mathematician - yet that doesn't mean I'm able to, for instance, heavily modify a computer-generated forecast in the way that the professional meteorologists at Exeter frequently do. So, the inference that there is sole reliance by the Met Office on computers to forecast the weather, is flawed and without foundation.

We do have some incredibly knowledgeable members on here (GP, Chio to name but two) who obviously maintain a deeper, more scientific interest in meteorology - indeed, some might say, pushing the boundaries beyond merely looking at the prima facie output. Others don't share that deep interest or understanding; they just want to know if it's gone to snow or not. Which is fair enough, but I think there is enough space for everyone to contribute and feel valued, whatever their persuasion.

I've tried to make this post less 'wordy', without it sounding condescending. The truth is: mathematics are playing a vital part in our understanding of the physical world, and the application of it is only going to increase over the coming years. Technology is ever advancing; for instance, I read yesterday how scientists have now developed a means to encode DNA. The mind-blowing significance, is that they could store the worlds media ever created - in HD - in barely a cupful of it.

SB

[nick sussex]

In the shorter term still uncertainty with tomorrows snow.

The latest NAE downgrades the risk for more southern and far se areas and the GFS 06hrs run also downgrades the wintry potential for the whole UK.

I don't know whether RH on the BBC is still using earlier data because that still looks very wintry for the south and se.

In the medium term still uncertainty with how amplified the pattern will be upstream this effects the low moving east and engaging the PV lobe in Canada.

NOAA lean a bit more towards the ECM ensemble mean as its more amplified.There is likely to be some pulling back of those low heights over Greenland around 168hrs, so if we're to see more interest develop it will be around the 168 to 192hrs timeframe as that deep low engages that PV lobe.

Edited January 24, 2013 by nick sussex

[MR EXTREMES]

well the latest gfs is very flat indeed very westerly flow i agree theres not much of a single to show a change in sometime yet but we all know how things change.

[TEITS]

Looking at the GEFS mean in the medium range and you can clearly see where its all gone wrong because what is being projected is the complete reverse of what was being suggested around 7 days ago.

It was said that the Canadian Vortex would weaken and the Siberian Vortex would become the dominant one and would move SW into Scandi and that we could tap into very cold NE,lys at the end of Jan.

http://modeles.meteo...-21-1-192.png?6

If you look from +0 to +192 the complete opposite has occured. The Siberian Vortex weakens and moves back NE whilst at the same time the Canadian vortex becomes dominant.

This leads on from what I said last night. Whilst the Stratosphere has an definite impact, predicting how this impacts the PV is just as uncertain as any anything else. So if you want to know why we're looking at mild W,lys at the end of Jan instead of bitter NE,lys there is your answer.

Edited January 24, 2013 by THE EYE IN THE SKY

[Hotspur62]

Have watched this thread for last 3 months and am in awe of some of the knowledge on here that has helped me understand alot more what drives the weather in the UK.This is my first post on this thread(Have stuck to posting on Regional Thread for SE).Have just looked at latest GFS model and although it looks pretty Zonal,apart from next Tues/WEd next week which look above average in terms in temps, i don't think it looks disastrous from a cold perspective.As others have said(all with far more expertise than me!!),if the piece of Vortex shifts across Atlantic quickly which it looks like doing,then in the medium term,i think by say 3rd/4th Feb we may be in with a chance of Heights building in Atlantic.

I think next 120-144hrs are very important in future developments for last 1/3 of UK winter.I still have a feeling that we have another 7-10 cold spell possible, but i am not sure if it will be with a Greenland High or another Scandi High.Think developments over next 7 days will give us more answers to which one it will be.Having enjoyed the current cold spell which is due to end for most by tomorrow night,i would like one more cold spell before Spring comes.

Well,i have given my 2 bobs worth now and will wait to see what happens and continue to watch this thread with great interest as i feel this winter really has been an interesting one and one that brings out the passion which we all have for the British weather!!

[chionomaniac]

Snowballz.

I would say that the summer reliability may drop off due to the lack of stratospherically driven weather pattern (which is fairly predictable). The same may occur when we have a SSW - effectively we enter a summer pattern up above for a period - hence the record high shannon entropy recently.

[m1chaels]

Just seen your reply.

If we look at some verification data from ECMWF - the world renowned pacesetters in mid-range forecasting - you'll see that there is a almost linear growth in forecast verification, in-line with time:

Thanks for this informative post but a couple of comments.

The paths of improvement appear constant at first glance but more recently appear to follow a pattern of sharp increases then levelling off (even falling). Could this be because the models are ignoring some 'external factors' (possible examples ocean currents, polar ice libido, solar activity) which actually impact on the atmospheric physics' in a manner that means that over time a model optimised against a steady state background will decline in accuracy?

When looking at model accuracy is this assessed over the globe or hemisphere as a whole? In which case a model that verifies well at this macro level might still show a lower level of accuracy or even a specific bias for individual regions (such as the UK)?

(Mods please feel free to move to the correct thread)

Edited January 24, 2013 by m1chaels

[Climate Man]

As far back as 1983 I studied BBC forecasts for a month in each season. I devised a marking scheme for comparison with actual - Summer showed the lowest level of accuracy.

[High Altitude]

Just seen your reply.

If we look at some verification data from ECMWF - the world renowned pacesetters in mid-range forecasting - you'll see that there is a almost linear growth in forecast verification, in-line with time:

What that chart illustrates is that, back in 1998 the 80% threshold was around 5 days or 120hrs (so, incidentally, I've no idea where you get '4 days today' from) However, by mid-2012, that 80% threshold has been pushed out to 6.5 days, which is 156hrs.

What's also illustrated in that chart, is the tailing-off of growth since around the end of 2009 - with a plateau clearly evident. The reason for this is very simple: ECMWF has exhausted the capability which current supercomputing resources afford them.

There is a clear need to upgrade the hardware, rather than refining current algorithms. However, it is important to finesse algorithms prior to an upgrade, and I note with interest that ECMWF will be introducing further vertical resolution this year; therein - or at least one would imagine - delivering greater depth of data which the refined algorithms can compute.

Even the lowest level of confidence (25%) grows in tandem - now out to around 9 days (216hrs):

Again, that chart illustrates how the bar continually moves forward; 5.5 days (132hrs) in 1998 has now risen to around 8.5 days (204hrs) Again though, we see that tail-off towards the end of 2009. It could be plausibly argued that, as you test more exploratory algorithms, you will naturally increase the error-rate as it's a given that not all algorithm refinement will actually help (indeed, it may hinder - which is what I suspect this plateau effect represents)

For me, I always look for the stories within stories. For example, my interest from these kind of verification charts is more drawn to why it appears model performance markedly drops off in the summer season, relative to its winter performance. A very clear pattern there. Overall, the trajectory is positive, but the hidden pattern still needs addressing (in my opinion anyway)

As you ask (you might've guessed anyway) but I am indeed an expert mathematician - yet that doesn't mean I'm able to, for instance, heavily modify a computer-generated forecast in the way that the professional meteorologists at Exeter frequently do. So, the inference that there is sole reliance by the Met Office on computers to forecast the weather, is flawed and without foundation.

We do have some incredibly knowledgeable members on here (GP, Chio to name but two) who obviously maintain a deeper, more scientific interest in meteorology - indeed, some might say, pushing the boundaries beyond merely looking at the prima facie output. Others don't share that deep interest or understanding; they just want to know if it's gone to snow or not. Which is fair enough, but I think there is enough space for everyone to contribute and feel valued, whatever their persuasion.

I've tried to make this post less 'wordy', without it sounding condescending. The truth is: mathematics are playing a vital part in our understanding of the physical world, and the application of it is only going to increase over the coming years. Technology is ever advancing; for instance, I read yesterday how scientists have now developed a means to encode DNA. The mind-blowing significance, is that they could store the worlds media ever created - in HD - in barely a cupful of it.

SB

Thank you for taking the time to provide thiis very interesting and informative reply.

I do have a question though that I'd like to put out there. Are the models programmed with all the currently known variables that could affect the weather ie. SSW, MJO, GWO, Solar influence (Sun activity), NAO, AO etc etc so that whenever any of these variables change they are reflected in the next model runs output?

Also, is a lack of data variables in some models/runs the reason for such wild differences in output and if so which model has the most variable input?

[Catacol]

I've tried to make this post less 'wordy', without it sounding condescending. The truth is: mathematics are playing a vital part in our understanding of the physical world, and the application of it is only going to increase over the coming years. Technology is ever advancing; for instance, I read yesterday how scientists have now developed a means to encode DNA. The mind-blowing significance, is that they could store the worlds media ever created - in HD - in barely a cupful of it.

SB

Excellent post overall and not too wordy at all - you write a hell of a lot better than most mathematicians I have met! However I wont be following your advice as to my friend's wife and her position at the Met... :-)

Do you think there is a theoretical ceiling to the ability of these ever-tuned algorithms to accurately forecast weather, or are we talking constant evolution to the point that chaos weather theory is a thing of the past?

Edited January 24, 2013 by Catacol_Highlander

[Frosty.]

Looking at the GEFS mean in the medium range and you can clearly see where its all gone wrong because what is being projected is the complete reverse of what was being suggested around 7 days ago.

I choose to believe what's being projected now can easily be reversed in the same way as the above, the patterns are constantly evolving and changing and better times won't be far away. I have seen enough evidence that from around T+168 hours, there will be more energy being pumped into the southern arm of the jet and the chances for colder incursions will increase, at some point we will get the height rises around greenland/iceland.

[SMU]

All-

Just a comment on the short term NAE- surprised it hasnt been picked up--

00z V 06z NAE is all about the positioning of the shortwave trough ahead of the parent low & how far south it digs bfore moving through-

The eastern flank of the sliding shortwave will be the area supportive of snow & the western side will be filled with atlantic air-

so across that low there will be a sharp gradient & so the faster it pushed through the faster it warms up & the snow turns to rain-

I would put MORE stock in teh 00z NAE than the 06z simply because of the reasons I always present on the 06z performance, however because the 00z doesnt outperform the 06z 100% of the time occasionally it will varifiy.

so the 00z at 48 has the trough sitting over wales at 48

http://expert-images.weatheronline.co.uk/daten/proficharts/en/nae/2013/01/24/basis00/ukuk/prty/13012600_2400.gif

the 06z has the trough sitting flatter over the midlands & running through faster-

http://expert-images.weatheronline.co.uk/daten/proficharts/en/nae/2013/01/24/basis06/ukuk/prty/13012600_2406.gif

Simply if people want a longer snow event then take the 00z NAE,00z GFS

& even better the 00z ECM

S

[Hotspur62]

I choose to believe what's being projected now can easily be reversed in the same way as the above, the patterns are constantly evolving and changing and better times won't be far away. I have seen enough evidence that from around T+168 hours, there will be more energy being pumped into the southern arm of the jet and the chances for colder incursions will increase, at some point we will get the height rises around greenland/iceland.

That's what i like,glass half full as opposed to Half empty!!The next weeks model watching will be very interesting and hope it gives the coldies amongst us positive news!!

[Captain Shortwave]

I not overly worried about the output yet. Correct me if I'm wrong but during the first warming event which started at the end of December, didn't the tropospheric vortex kick into life and temporarily strengthen before it eventually weakened and split? I'm just saying as models don't learn unless new algorithms are inputted into their database so maybe we will see the same thing. The warming of the Canadian sector causes a temporary increase in activity of that Vortex segment. We are at the point where this happens in the reliable/semi reliable timeframe. So looking at fi beyond 200 hours might be pointless as the results will be skewed towards a dominant canadian vortex which when that time actually comes might not be the case. The vortex weakened eventually last time, it will happen again hopefully.