Model Thread - 14th Jan - Mk2

[Thundery wintry showers]

We are now looking at a 36-hour north-westerly, which is similar to what my assessment of the likely outcome was a couple of days ago, but yesterday evening's runs (with the exception of GFS) suggested something rather more sustained, about 3 or 4 days or so.

I can understand the various posts saying "victory to ECM/GFS" as the reality to my mind is that they have "met in the middle"- GFS was probably right to indicate a flatter pattern but somewhat overdid the flatness of it. I won't be surprised though if ECMWF has over-reacted to the GFS signal and may tone up the polar outbreak a little bit as we approach the time.

[If Wishes made Weather]

Surely though on the models people must see that there are two reasons that anything remotely severe (be that wind, cold, heat, whatever) that is being indicated in F1 won't happen:

1 - its F1, meaning that the general pattern being suggested will be revised many times between +240 and +0, and,

2 - everything will always reduce to, or near to, the average. Thus, the predicted 10c above average, is most likely to end up being +2 above average etc...

Thus to see that both GFS and ECM have reduced their extreme suggestions to a middle ground is not overly suprising.

[shedhead]

ECM now very much come into line with the GFS with a far flatter pattern, fair play to the GFS. That is not to preclude any colder shots still from variations of cool zonality, for example both the GFS/ECM are still fairly cool for the north.

Still not a hint of anything more than a little below average on the models this morning sadly...UKMO is best but even that is not being as agressive now.IOndeed 06z GFS is back to very mild by 168hrs!

Pattern isn't too poor for the north, but pretty shocking for the south now.

The latest overnight runs do look to be something of a blend tho kold, remember only 72hrs ago GFS had high pressure virtually dominant out to T+384hrs, but without question ECM has moved more towards GFS than visa versa. I'd like to think that given the way the last 4 or 5 days have panned out we'll see a rather more measured, cautious overall approach in future to the model that shows the coldest evolution, especially when said model is ploughing a rather lone furrow, but as ever time will tell.

[kold weather]

I can understand the various posts saying "victory to ECM/GFS" as the reality to my mind is that they have "met in the middle"- GFS was probably right to indicate a flatter pattern but somewhat overdid the flatness of it. I won't be surprised though if ECMWF has over-reacted to the GFS signal and may tone up the polar outbreak a little bit as we approach the time.

Yes, somewhere in the middle is the likely correct assesment, though I'd personally say the ~ECM has moved more than the GFS in the last few days, if you compare the 00z ECM from 2-3 dasys ago to todays, its massivly different, whilst the GFS from a few days is probably only a little more amplified and showing more of a PV split.

[BLAST FROM THE PAST]

Re Kudos to GFS etc, that is very much incorrect as yet, as indeed to any model. The main set ups on Fri was for 168-192 and thus we are at 144 so far overnights. The UKMO is most amplified with ECM lesser so but closer together and still have decent set ups possible to follow. Tonight we will be 120-144 and tomorrow morning 120 timeframe and less etc. Beyond this point as we stand is far from sorted or settled. Yes NW flow looks agreed but amplification and follow up are not.

The depth of the trough digging into Europe by the UKMO is very interesting indeed.

BFTP

[Buzz]

Surely though on the models people must see that there are two reasons that anything remotely severe (be that wind, cold, heat, whatever) that is being indicated in F1 won't happen:

1 - its F1, meaning that the general pattern being suggested will be revised many times between +240 and +0, and,

2 - everything will always reduce to, or near to, the average. Thus, the predicted 10c above average, is most likely to end up being +2 above average etc...

Thus to see that both GFS and ECM have reduced their extreme suggestions to a middle ground is not overly suprising.

It's FI, not F1

Edited January 15, 2012 by Buzz

[fromey]

BREAKING NEWS FROM OUR CORRESPONDENT IN THE USA!This just in from NOAA.

USED THE 00Z/15 ECMWF FOR THE PRELIMINARY FRONTS AND PRESSURES FOR

DAYS 3 THROUGH 7. THE ECMWF IS BEST CORRELATED WITH THE GEFS MEAN

AT THE MEDIUM RANGE...WITH THE OTHER DETERMINISTIC

MODELS...INCLUDING THE GFS...DIFFERING ENOUGH OVER SEVERAL KEY

AREAS OF CENTRAL NORTH AMERICA TO CAST SUSPICION ON THEIR

SOLUTIONS. THE GFS IS A FLAT OUTLIER WITH THE WAVE CROSSING THE

EASTERN GREAT LAKES AT THE START OF THE PERIOD...WITH THE GEM

GLOBAL ALONE IN CONSTRAINING THE POLAR AIR NORTH OF THE CANADIAN

BORDER FOR THE DURATION OF THE FORECAST

hi complete noob here

if what NOAA saying is correct then this must have implications down stream and to what its output is showing.

is this why the GFS is showing a differing output?

cheers fromey

[IDO]

GFS 06z... Another run and another possibility after next weekend. Still continuing to model the brief PM a day before ECM (which ties in with the timing {prev posts}) and both will probably continue to tone it down; and anyway a 24 hour N'Westerly, wow. Flat pattern and only glimpses of cold, out into FI; never quite gets to us while the PV is weakening. Then the PV strengthens and that is that; Azores HP re-establishes and heading into Feb we have a strong PV, flat turbo charged Jet stream well to the North of the UK. Despite what some members will continue to say, that run is as bad as it gets if taken at face value. And I also looked at the bigger picture:

http://nwstatic.co.uk/viewimage.pbx?type=gfs;date=20120115;time=06;ext=372;file=npsh500;sess=b92929e80e1d3dc5ae0a68e1496e5a37;

Of course little tweaks can affect the picture later on but we are model watching and as far as the GFS Op run has been going over the last few days it is trending very poor for cold. There are still model differences and we can only hope ECM upgrades their 0z later today (before we write Jan off).

[Paul White.]

Yes, somewhere in the middle is the likely correct assesment, though I'd personally say the ~ECM has moved more than the GFS in the last few days, if you compare the 00z ECM from 2-3 dasys ago to todays, its massivly different, whilst the GFS from a few days is probably only a little more amplified and showing more of a PV split.

I really do think that a line needs to be drawn under this GFS V ECM debate. This thread was brought down yesterday due to what appears to be a very emotive subject. surly now best to wait for the forecast period to lapse, and then we look at the verification stats before proclaiming that certain models have the period nailed.

Regards Paul

[Uncle_Barty]

Surely though on the models people must see that there are two reasons that anything remotely severe (be that wind, cold, heat, whatever) that is being indicated in F1 won't happen:

1 - its F1, meaning that the general pattern being suggested will be revised many times between +240 and +0, and,

2 - everything will always reduce to, or near to, the average. Thus, the predicted 10c above average, is most likely to end up being +2 above average etc...

Thus to see that both GFS and ECM have reduced their extreme suggestions to a middle ground is not overly suprising.

For future reference.... It is Fantasy Island, not Formula One.

I can understand the various posts saying "victory to ECM/GFS" as the reality to my mind is that they have "met in the middle"- GFS was probably right to indicate a flatter pattern but somewhat overdid the flatness of it. I won't be surprised though if ECMWF has over-reacted to the GFS signal and may tone up the polar outbreak a little bit as we approach the time.

I have seen a few posts saying a certain model has "won" - lets get this right - no model has "won" until verification, I certainly wouldn't be calling "victory" for any model when we are still 5 days out!

It is looking like we are not in for an extended cold spell on recent output, but there is still a long way to go and things could change again. Weather, eh?!

Edited January 15, 2012 by Jim_AFCB

[swilliam]

Yes, somewhere in the middle is the likely correct assesment, though I'd personally say the ~ECM has moved more than the GFS in the last few days, if you compare the 00z ECM from 2-3 dasys ago to todays, its massivly different, whilst the GFS from a few days is probably only a little more amplified and showing more of a PV split.

Why does it matter which one has moved which way and by how much - the interpretation is entirely subjective depending on whether you are considering will it snow and how long will be cold where I am to how flat/amplified the flow is or to the bigger picture (including NA and then to the whole NH). It appears to me that the difference between the models was due to small differences in the way and timing of ejecting the low from the Conus. Depending on this gives larges changes in the weather we can expect in our neck of the woods. This 'discussion' only leads to argueing as can be seen by the subsequent posts. If you want to compare the models use the official verification stats which paint quite a clear picture I think.

Much better is to concentrate on what the models are showing, how this is changing, what level of agreement is there (not stating which one is right or wrong - nobody knows) what are the prospects for the weather in the UK based on interpretation of the models.

As far as I can see it is all very unclear beyond the short NW outbreak next weekend and even the intensity of that is still not known. So I am expecting a lot of changes in the medium term output over the next few days.

Edited January 15, 2012 by swilliam

[Frosty.]

The Gfs 06z is a huge downgrade for snow/cold potential from yesterday but northern scotland would have some short lived Pm incursions, generally turning milder through next week and breezier as high pressure loses it's grip and pressure falls from the northwest, high pressure centred to the southwest of the uk is at least partly responsible for stopping any cold air reaching the south so it turns into a fairly mild run for southern england until the euro high builds towards the uk later in FI so there would then be a change to frost and fog but not much chance of snow reaching southern britain on this run and only for the hills/mountains further north.

[MP-R]

Major downgrades from both the models this morning. Well, the GFS never really jumped on the cold band wagon but yesterday the ECM was encouraging for at least 2/3 of the country next weekend. Both show a northwesterly incursion towards the end of the week after becoming milder by Tuesday. Perhaps the main difference is the flatness of the GFS compared to the ECM which seems to show more 'sliding' lows to the northeast of Britain. Either way, high pressure to the southwest as has been mentioned won't aid us in getting any of that cold down from the north.

Best enjoy this seasonal frosty spell because from Tuesday onwards it does look milder and more uncertain.

[TonyH]

My take on the coming week, in spite of based on recent model trends/ outputs:

Weather guide Monday 16th to Sunday 22nd January (west Wales and central Midlands centric!)

Headline: Dry and bright start, then unsettled with variable temperatures

In spite of high pressure last week, nuisance weak troughs plagued west Wales, so that our dry interlude was delayed and spells of drizzle up to Thursday put another 8mm in the Llanwnnen rain gauge. However, this amount is far below a typical weekly deposit for January. The Midlands did though manage a mainly dry week, with less than half a millimetre falling on Coventry in the past 7 days. Dry weather proper has now arrived in Ceredigion for the first time since mid November.

The models have been very interesting to follow over the past week, with the GFS continuing to keep the worst of Winter at bay, while ECM has been consistently threatening a cold northerly blast come the end of the coming week. However, I am inclined to go with the equally consistent but more moderate, milder outlook that the GFS has been delivering, plus that ECM seems just this morning to have moved well away from the very cold, potentially snowy prognosis of a few days ago. Therefore, I foresee no real change to significant cold later this week.

High pressure is still close to England during Monday and Tuesday, so that the pleasant dry, bright spell continues for a few more days. Further frosts by night, down to as low as -3c. A chilly south east breeze on Monday, maxima 4-6c, but less cold on Tuesday.

A deep low pressure then passes close to Iceland bringing with it cloud, wind and rain at times for Wednesday and Thursday. Also it turns mild, maxima on Wednesday above 10c once more. The cold front clears through some time on Thursday, so cooling off, with perhaps some late sunshine and showers . After an early ground frost, Friday sees a chilly blustery north west wind, with sunny spells and some showers, these perhaps wintry over high ground.

A mixed weekend then follows with mild and colder interludes, with a west to north west flow prevailing, and there will be some rain at times but also with drier spells between.

For (hopefully!) supportive charts see the Welsh thread:

http://forum.netweather.tv/topic/71920-wales-regional-weather-discussion/page__st__740

[swilliam]

Looks like the OP is still a bit out of kilter with the ensemble judging the short term ensembles - note the clustering at about -5

[nick sussex]

hi complete noob here

if what NOAA saying is correct then this must have implications down stream and to what its output is showing.

is this why the GFS is showing a differing output?

cheers fromey

We need the pattern in the eastern USA to remain as amplified as possible especially as the low exits the ne around 96hrs, so yes this of course impacts the amplitude of the pattern in the Atlantic which in turn effects the displacement of the Azores high.

These small changes at that point domino into big changes in western Europe, for this reason until we get agreement on the pattern within 96hrs in the eastern USA then the output is liable to be wrong to some degree.

Thats not to say some big ne flow is on the way but that we may see the GFS move a bit more to the ECM/UKMO and hold the pattern further west, the next point of debate will be where low pressure near Iceland goes later, the block to the east could well force some energy more se rather than east.

We'll just have to try and at least get agreement within 144hrs before seeing what could happen there.

[Cloud 10]

Cloud 10, the GFS may have slightly moved towards the ECM, but make no mistake, the ECM has nearly done a 180 on its old solution, there has been only one winner in this battle and thats the GFS.

Anyway I wouldn't get too downhearted about the models, I'd not be all that susprtised to see the models toy with this idea again in another 5-10 days time, still quite a wide range of solutions on the table down the line as well.

If we compare what the 06z gfs and the 00z ecm were predicting for the 19th of Jan on their

runs from 3 days ago compared with their runs from this morning,it seems to me that its the

gfs which has moved the most as it now has the troughing digging south-east.

gfs 3 days ago> now>

ecm 3 days ago> now>

[kold weather]

It appears to me that the difference between the models was due to small differences in the way and timing of ejecting the low from the Conus. Depending on this gives larges changes in the weather we can expect in our neck of the woods. This 'discussion' only leads to argueing as can be seen by the subsequent posts. If you want to compare the models use the official verification stats which paint quite a clear picture I think.

Thats certainly not the only difference the models had a few days ago, maybe the huge upper high that the ECM developed near Scandinavia for example?! Its this feature that has been constantly downgraded and its this feature that was the main driving system for the cold developing. Get rid of that upper high and there is nothing to stop a flatter flow from coming in, and indeed that is what is happening. The low ejecting from Newfoundland is a minor issue compared to the massive over-doing of that particular upper high. Its not by chance that the ECM has been reducing and weakening any cold spell as its constantly weakened that upper high.

Its useful to talk about the differences in the models and why they have changed, we'll never improve our understanding if we don't discuss such topics. However the problem is as you say, some people let it become far too emotive. I think I've been very balanced in what I've said.

Cloud 10, take a look at the 00z ECM on the 13th Jan and the 06z on the 13th Jan and tell me what is closer to being correct for OUR neck of the woods...

Control run is quite nice!

[fromey]

thanks for your reply Nick.

its still confusing as to why some take the GFS as gospel/praise it etc etc when NOAA call it a FLAT OUTLIER, and go with a different blend of other models.

more learning to do for me.

fromey

[shedhead]

The Gfs 06z is a huge downgrade for snow/cold potential from yesterday but northern scotland would have some short lived Pm incursions, generally turning milder through next week and breezier as high pressure loses it's grip and pressure falls from the northwest, high pressure centred to the southwest of the uk is at least partly responsible for stopping any cold air reaching the south so it turns into a fairly mild run for southern england until the euro high builds towards the uk later in FI so there would then be a change to frost and fog but not much chance of snow reaching southern britain on this run and only for the hills/mountains further north.

Hard to argue with your assessment Frosty IF you take the 06 GFS at face value, it's looks about as condusive to widespread cold as a straight N'erly is to a Summer heatwave. However, given the overall hemispheric pattern and the ongoing/expected strat warmings I'm really not sure we can assume any of it past 96hrs is even likely to verify, let alone make the assumption that it will, but of course it cannot be ruled out either. To my mind the next 7-10 days are critical for deciding how the overall Winter pans out, if we get the expected vortex disruption then hopefully Feb will prove to be something of a cold balance, but just like the last week or so it remains very much on a knife edge and we simply cannot rule out ending up with a mild, snowless season for many.

Edited January 15, 2012 by shedhead

[Jason M]

Looking through the output this morning it seems to me that were really no further forward than at this time last week. ECM has steadily downgraded next weekends cold spell, to a cold snap and is now moving towards a situation where were looking at a couple of rather cold days and nothing more.

The one feature that dominates all the output is the Azures high, which despite changes at a Global level isn't showing any signs of going anywhere. Same rule applies now as at any other time during this Autumn and Winter (and last Jan - Feb) which is high pressureto our south = no snow of note. Whilst we seecharts like the one below accross the models, nothing much is going to change.

To avoid upsetting anyone i'll show a chart from GEM. This is one of the better ones this morning as well!

http://www.meteociel.fr/modeles/geme_cartes.php?&ech=132&mode=0&carte=0

24 hours later leads to this

http://www.meteociel.fr/modeles/geme_cartes.php?&ech=156&mode=0&carte=0

With that Azures high in position there is only ever going to be one outcome ie a renewed push from the west. Even with things changing globally its perfectly possible that the Azures high will stay in situ regardless.

Still time for me to be proved wrong of course and others no doubt have different views but i can see no evidence whatsoever of any material change to the status quo at this point. I've felt for some time that were headed for winter that will sit firmly in the 'very mild' category and based on our current starting position Feb could really be quite exceptional for all the wrong reasons.

No doubt by this afternoon the next ice age will be showing at 120 hours

Jason

[Frosty.]

It's a shame that all the best posters on here who have made a good case for the uk to undergo a pattern change are possibly going to have the rug pulled from under them, the models this morning are back to where they have been for the last 6 weeks minus that rather colder zonal spell during mid december, relative riches compared to what the models are currently indicating for the next few weeks, still it could just be a model wobble on a large scale as I assume the background signals are still the same.

[IDO]

Surely though on the models people must see that there are two reasons that anything remotely severe (be that wind, cold, heat, whatever) that is being indicated in F1 won't happen:

1 - its F1, meaning that the general pattern being suggested will be revised many times between +240 and +0, and,

2 - everything will always reduce to, or near to, the average. Thus, the predicted 10c above average, is most likely to end up being +2 above average etc...

Thus to see that both GFS and ECM have reduced their extreme suggestions to a middle ground is not overly suprising.

In this case I disagree; a few days ago ECM went with an amplified pattern over N.America that slowed movement west to east allowing blocking patterns to build near the UK. Its subsequent downgrading is not a toning down to an in between IMO; this is an important variable loaded onto the computer run(s) that meant profound changes downstream. If they were wrong it would be nice to know why they amplified it so much initially. What were the background noises? Another poster (Saint Snow at theweatheroutlook) summed it up better than I can:

"

My concern is that, with hindsight, ECM has narrowed that gap between the two lows on each of the last few runs. Rather than just 'flipping' to a new pattern, it's evolved it gradually, until it reached a kind of tipping point where the ridging wasn't there at all and the overall effect on the UK was disproportionately huge."

So although this ECM -v- GFS battleground is more prolonged than any of us expected, it will be of help for the future model watching; if ECM are wrong, then maybe we will only pay attention to sudden amplified patterns, when they are verified. Then there may not be as many disappointed casual viewers.

[swilliam]

Thats certainly not the only difference the models had a few days ago, maybe the huge upper high that the ECM developed near Scandinavia for example?! Its this feature that has been constantly downgraded and its this feature that was the main driving system for the cold developing. Get rid of that upper high and there is nothing to stop a flatter flow from coming in, and indeed that is what is happening. The low ejecting from Newfoundland is a minor issue compared to the massive over-doing of that particular upper high. Its not by chance that the ECM has been reducing and weakening any cold spell as its constantly weakened that upper high.

Its useful to talk about the differences in the models and why they have changed, we'll never improve our understanding if we don't discuss such topics. However the problem is as you say, some people let it become far too emotive. I think I've been very balanced in what I've said.

Cloud 10, take a look at the 00z ECM on the 13th Jan and the 06z on the 13th Jan and tell me what is closer to being correct for OUR neck of the woods...

Control run is quite nice!

Hi Kold

Sorry I I was not really picking on your post I was just a bit fed up of reading a lot of posts which started of similarly beginning which one is right/wrong, talk of climb downs, kick on teeth for the ECM etc. and would just like to concentrate on what is being shown. This of course includes discussing the differences between the outputs and why they might occur and what is significant.

Looking at the longer term ensembles there does seem to be a general trend for slightly milder weather after next weekend. this is probably to do with where that low near Iceland goes. Some nice runs in far FI as the PV generally becomes more fragmented.

[chionomaniac]

I think it is time to draw a line under any differences in the GFS and ECM that have occurred over the past few days and concentrate on what is showing from now onwards.

Thank you.