Model Output Discussion 13/11/12 18Z onwards

[AWD]

Took part in interesting discussion last night with UKMO colleagues re potential snow possibilities for some areas into Saturday. We didn't bother discussing the longer range stuff thxis time...!!

Are we talking Western Scotland?

http://cdn.nwstatic.co.uk/gfsimages/gfs.20121114/00/78/prectypeuktopo.png

http://cdn.nwstatic.co.uk/gfsimages/gfs.20121114/00/84/prectypeuktopo.png

Perhaps down into Cumbria Saturday night too?

http://cdn.nwstatic.co.uk/gfsimages/gfs.20121114/00/90/prectypeuktopo.png

http://cdn.nwstatic.co.uk/gfsimages/gfs.20121114/00/96/prectypeuktopo.png

http://cdn.nwstatic.co.uk/gfsimages/gfs.20121114/00/102/prectypeuktopo.png

[Frosty.]

I think that perhaps this is a bit of an over-reaction to the current runs

No i've just tried to give a balanced view of what the Gfs 00z ens mean is showing this morning, i'm as big a coldie as anyone but the mean shows the west to east pattern persisting and the cold block to the east moving further away east eventually, even the met man on radio five said the chances of a very cold spell are diminishing, the outlook shows mild and cool zonal alternating and perhaps a seperate anticyclonce developing in FI with faux cold and overnight frosts but I want a big freeze as much as anyone.

[offerman]

AWD, Sorry if i misunderstood I thought Ian was talking about the West Country as thats where i see him forecasting on T.V for points West.

[fergieweather]

Incidentally, worth adding how the developments into Saturday as progged by UKMO-GM and EC also have support from ARPEGE (as shown here:)

[Tim Bland]

The skeleton in the room is still the deep seated cold stratosphere. 2 wave activity will not be enough IMO to prevent it effects filtering through. Are there any examples in the past of proper HLBs forming with such a cold Strat? All other factors are in favor of a blocked winter so I think it will be a cold one if the Strat plays ball

[Weather Boy]

No i've just tried to give a balanced view of what the Gfs 00z ens mean is showing this morning, i'm as big a coldie as anyone but the mean shows the west to east pattern persisting and the cold block to the east moving further away east eventually, even the met man on radio five said the chances of a very cold spell are diminishing, the outlook shows mild and cool zonal alternating and perhaps a seperate anticyclonce developing in FI with faux cold and overnight frosts but I want a big freeze as much as anyone.

I hope you didn't take my post the wrong way. No offence intended.

Let me re-phrase it: the 'better' runs for cold were at such a range that they could only ever sensibly have been interpreted as a signal towards significant blocking. Whilst as it happened they also indicated cold on our shores, it wasn't (in my view) possible to treat that seriously at this range. But there was (and still is) a strong chance of northern blocking in the long term.

I haven't had time to actually view the current run myself, but it sounds as if it doesn't remove the blocking signal, but does indicate the block further east meaning that cold is -if that run were to verify- not here, but further east.

At that range, just like the more favourable runs, it could easily change back. I agree that it makes deep cold less likely, but in my view, it was never that likely in the first place, given the long range.

I therefore advocate a more measured view of both 'favourable' and less favourable runs rather than stepping aboard the emotional rollercoaster.

[January Snowstorm]

For me the biggest problem is the fact that there is absolutely no cold over Europe at the moment.

An Easterly would not deliver even if we got one.

The positive is synoptics are very unusual for the time of year with potential very much there

[SMU]

Just the 06z run- All the ridges mentioned last night developing & notably the greenland ridge with that split low identical to Nov 2010

Also the ECM Ensembles trending curve is nice day 10-15....

S

[Radiating Dendrite]

GFS 6z seems to be much better for cold in the mid range.

It is important to remember that the models will usually pick up on an idea a for a few runs and then drop it before bringing it back in the nearer time frame (especially the GFS).

Look North I say and not East at the moment.

If we do get HLB establishing, then it will be a head scratcher with the current state of the upper start (very cold).

[Craig84]

GFS and ECM in very good agreement i'd say on this run at 216hrs.

http://www.meteociel.fr/modeles/ecmwf.php?ech=216&mode=1&map=1&archive=0

http://www.meteociel.fr/modeles/gfse_cartes.php?&ech=216&mode=0&carte=1

Ok, back in my shell!

[EML Network]

GFS 6z seems to be much better for cold in the mid range.

It is important to remember that the models will usually pick up on an idea a for a few runs and then drop it before bringing it back in the nearer time frame (especially the GFS).

Look North I say and not East at the moment.

If we do get HLB establishing, then it will be a head scratcher with the current state of the upper start (very cold).

Yes I couldn't agree more with that

On my own little score count I'd say were at 1.2 on the ''Def Con'' scale (20% probability of a cold & snowy spell)

Would be nice to be at 1.3 tomorrow

- "1.3:) A few days later it shows up again ...were now looking at a possible trend...and has more ensembles support"

[Frosty.]

I hope you didn't take my post the wrong way. No offence intended.

Let me re-phrase it: the 'better' runs for cold were at such a range that they could only ever sensibly have been interpreted as a signal towards significant blocking. Whilst as it happened they also indicated cold on our shores, it wasn't (in my view) possible to treat that seriously at this range. But there was (and still is) a strong chance of northern blocking in the long term.

I haven't had time to actually view the current run myself, but it sounds as if it doesn't remove the blocking signal, but does indicate the block further east meaning that cold is -if that run were to verify- not here, but further east.

At that range, just like the more favourable runs, it could easily change back. I agree that it makes deep cold less likely, but in my view, it was never that likely in the first place, given the long range.

I therefore advocate a more measured view of both 'favourable' and less favourable runs rather than stepping aboard the emotional rollercoaster.

Hi weather boy,

No it's fine, I was just hoping to see something more encouraging this morning but my feeling is we are now further away from a very cold end to the month, it looks like a mixed zonal outlook to me but with a good chance of a seperate anticyclone in FI with our own homegrown pool of cold air with some frosty nights and crisp sunny days but the overall pattern looks west to east with a big low to the southwest of iceland being in control and high pressure to the southwest of the uk, i'm hoping for a miracle though.

[Cloud 10]

To bin or not to bin? that is the question!

Maybe not if the NAEFS is barking up the right tree...

[bluearmy]

i keep reading that an easterly wont be cold as there is no cold over europe. in the short term that is true and i wouldnt expect any short easterly flow to bring anything wintry. (would still be cold). if an easterly is sutained for more than a couple of days, you will see a pool of lower uppers appear, either via troughing to the south or north dropping into the block. i'm sure if you look through any CFS run you'll see this solution play out at some point between now and april.

[Timmytour]

I am ever the eternal optimist.....i am able to place more faith in one "deep freeze in FI" run than 10 runs prior to it or subsequent to it :-)

And so I remain convinced that what showed up for +300h 18z run from monday will start showing up in the +200 runs from tomorrow

[pyrotech]

With hype in the media this week, and of course our first hint of a cold spell played with by the models i thought id put this in for new members and a reminder for us more seasoned members, the reminder is valid for myself also.

It also helps explain what im discussing about todays models/

Some good tips when model watching in fi

Look for trends not just in one model and be more sceptical of forecasts that are only on one model. If more models are on board with a forecast solution, then you can be more confident in the forecast. Look at all the models, looking for similar outlook or trend.

Look at again at model trends. If the high pressure moved towards Greenland on this run, what waydid it run before that? GFS, look at a couple days of Same time run ie 00Z for consistency. Some say the 06Z and 18Z runs are less accurate due to less data, i personally find fi in 18z GFS is best for finding a trend, however thats just my preference and why would that be true with less data inputted?

Each model has its own good and bad points, GFS is thought( and proven) to be less accurate than ECM, however i now think GFS does less overcooking of low pressure systems and so their is an improvement ongoing so that can change at anytime, also just because one model may be more accurate than an other, does not mean the less accurate one has got it right this time. With all the models the longer the timeframe the more likely the error margin. 1 day ahead is likely to be rather accurate, where 15 days ahead could be hopelessly so. GFS runs 4 times a day ECM just twice a day. Resolution also changes later in a run so accuracy is further displaced. A number of factors effect model accuracy including accuracy of initial data received, the computer programming errors of which there will be many in such huge complex programs and complexity of atmosphere compared to computing power.

The ensembles help confirm the likelihood of a models output. Ensembles from the same model are run several times with slightly different starting points. This allows for any mistakes in the initial data input. For instance a low centred over the Atlantic may be measured or estimated to be 995mb, but due to restraints on accurate data collection this low may be a little deeper or less deep, so starting point can be changed to show these scenarios, Initially the changes will be rather small over the next 2 days, but these differences can be huge after a week. This also can help determine where fi starts, by seeing where consitency to scatter starts.

The 18z GFS is often called the pub run, it is true it has less data inputted than the 12z but its normally only actually referrred as the pub run when we do not like what it shows, for instance us coldies, like it when it shows the cold rushing in and then refer it as pub run when next evening it shows mild. We all have bias to weather types and i am not putting coldies including myself down just being truthful to help new model watchers.

Does the GFS pick up trends deep in Fi? I think it does, i have watched it and monitored this and often find it picks a trend at days 12-15 that it then drops only to bring it back after ECM picks it up in the more reliable timeframe. I notice this mostly on 18z run, however this makes no sense as it has less data, unless of course the less data allows more accuracy in fi with the data it has due to less processing. Or could it just be that it throws so many scenarios that one has to be something ressembling the outcome.

.So what do we see right now.

GFS keeps playing with cold air around the 27th but it tries Northeasterly right thru to a North - Northnorthwesterly.

It really can see possible blocking but can not get a grip of exact orientation so without that the outcome is very different.

Looking at the essembles and 850mb temperatures, latest run has mean and operational running at around 0c but the control runs to -5. The control looks like an outlier but the model run seems to give some credability even thou it is an outlier. Looking at the essembles and fi would in my opinion start around 19th as scatter becomes evident around there. Ecm only runs to 24th but to me again shows some blocking and it will be interesting to see how it evolves.

The jet stream is a mess around 24th on this run, with a split jet above and below UK, this makes cold rather dificult to effect us, Later in the run the jet becomes less fragmented and shows signs of possibly having just one arm and moving South. Alot of uncertainty in the models, i would suggest the trend has been for cooler - colder at the end of the run in fi, which is a positive, but by know means does it that cold is nailed.

Thats my take on the models and explanation on how i use the models to help new members, i am not saying its perfect and welcome any coments from other members.

PS For new members fi = Fantasy island do not put F1 thats racing cars and a common mistake that new members make and some are very intolerant of that.

[Frosty.]

Well it's only the Gfs 06z op run but beyond T+260 it brings an early taste of winter, nothing like the ens mean is predicting but it's an upgrade from yesterdays 6z and 12z Northerly, it also topples as usual but has eventual reload potential, so whilst the major factors are against it (cold strat), it keeps some eyecandy interest in deepest FI, before that it looks like a mostly unsettled pattern with temps generally around average but some milder and cooler days thrown in.

[Buriedundersnow]

http://www.meteo.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

http://www.meteo.psu.edu/~fxg1/ECMWF_12z/htgcomp.html

still alot of blocking being forecast

[samadamsuk]

Hey all

This looks like the evolution we'd expect if a cold spell is on the way from the latest GFS, in keeping with the last few winters. The gyre over the atlantic moving eastwards, strong WAA breaks into the cold pool provided by deep cyclogenis over newfoundland, blocking to the north delevloping, and a northeasterly feed over the uk, flooding cold air over us on the downward side of an elongated trough, squeezed in either side by a blocking high.

Looks like a classic Warm, then wet, then N/Nely, then possibly easterly outbreak.

Shall I order a sledge, or wait for the next run ???

Sam

[weather_it_will]

Ensembles for Midlands showing most staying below the 850 zero line from the 25th November.

http://modeles.meteociel.fr/modeles/gens/graphe_ens3.php?x=269&ext=1&y=104&run=6&runpara=0

[Cloud 10]

Decent looking FI ensemble mean from the 06z with the trough migrating to our east with

heights building in the NW atlantic.

[itsnowjoke]

Hey all

This looks like the evolution we'd expect if a cold spell is on the way from the latest GFS, in keeping with the last few winters. The gyre over the atlantic moving eastwards, strong WAA breaks into the cold pool provided by deep cyclogenis over newfoundland, blocking to the north delevloping, and a northeasterly feed over the uk, flooding cold air over us on the downward side of an elongated trough, squeezed in either side by a blocking high.

Looks like a classic Warm, then wet, then N/Nely, then possibly easterly outbreak.

Shall I order a sledge, or wait for the next run ???

Sam

wait dont jinx it

[Frosty.]

It's worth pointing out that this time last year the models were so entrenched in a mild pattern, even deepest fantasy island was mild, there was nothing of interest for coldies for most of october and november last year, so in that respect, we are in a superior position now, there is something to grasp in lala land at least.

[Robbie Garrett]

Latest MetOffice 30 day points to settled conditions towards end of November and into December.

Calm before the storm, or just a normality in Exeter.

[phil nw.]

It looks like business as usual with typical November conditions for the next 10 days at least.The Euro block continues to show but never becoming an influence other than to maintain our trough insitu just to the west.

The mean operational heights for days 8-10 look very similar with this setup.

So looking at the next few days then based on the 00z outputs the modelling pushes that upper trough in again after a couple of quieter days with rain spreading south east by the weekend.

The block to our east sinks away and we then enter next week in a broad west/south west flow with somewhat cooler air with rain or showers and a chance of some snow over northern mountains at times.

Later in the week continental pressure build again with the return of some milder uppers from the south as we find ourselves briefly at least on the warmer side of the Atlantic trough.

Signs still of blocks building north both to our west and east at around day 10 onwards but i am less convinced today that we will get anything sustainable by month end.The ever increasing cooling over the pole is apparent on the 500hPa modelling. and can`t be ignored.

Edited November 14, 2012 by phil n.warks.