Spring Model Discussion 18Z 9/3/13 onwards....

[Summer Sun]

CFS (which nailed the northern blocking this month), has a chilly temperature anomaly for the remainder of the month..

Have to admit its performed well recently Is it showing anything milder for the start of April Paul?

Edited March 16, 2013 by Gavin.

[Paul]

I've always been very sceptical. of the CFS. However, I must say, it has performed remarkably well, over recent weeks...

The problem the CFS has is the way it's presented and used at times - for instance taking the raw charts and picking out a day in 4 weeks/months time to show as a forecast is something that people tend to do and that's not really the way it's meant to work. The raw charts, are not bias corrected so in effect to use them you need a knowledge of the biases in the model and the patience to take several days output to glean a clue as to the general pattern.

Then there's the anomaly maps, these are updated daily for people to see, but no operational centres use the daily output, it needs averaging over several days - for instance NCEP update their CFS forecasts weekly, based on the average of the output across that period.

[snowking]

Pure conjecture at this stage, but tonights 12z ECMWF keeps dew points sub 30F through until 192 from the M4 Northwards, with some snowfall shown initially even south of this around T+144 before turning to rain

What this does do, however, is allow us to verify once again (and as suggested after yesterdays 12z) that in this particular setup, the rain/snow boundary lies roughly at the -2c 850mb isotherm:

SK

[Panayiotis]

The problem the CFS has is the way it's presented and used at times - for instance taking the raw charts and picking out a day in 4 weeks/months time to show as a forecast is something that people tend to do and that's not really the way it's meant to work. The raw charts, are not bias corrected so in effect to use them you need a knowledge of the biases in the model and the patience to take several days output to glean a clue as to the general pattern.

Then there's the anomaly maps, these are updated daily for people to see, but no operational centres use the daily output, it needs averaging over several days - for instance NCEP update their CFS forecasts weekly, based on the average of the output across that period.

Yes the last couple of days it has been consistently hinting that the cold will return, I think every 12z run has done this since Monday Sunday, and it looks like it maybe correct....

I know its not really used for medium term forecasts but if it has called this right then we should start to give it more credibility.

Edited March 16, 2013 by Panayiotis

[UV-RAY]

Yes the last couple of days it has been consistently hinting (in the 12z runs) that the cold will return I think every 12z run has done this since Monday, and it looks like it maybe correct....

I think what we have to remember is the CFS is a long range tool, it has been consistent for the last twelve months IMO, picking up the correct signals a few months in advance. How long this new found stardom will last though is anyones guess.

[Snowman.]

Have to admit its performed well recently Is it showing anything milder for the start of April Paul?

Yes before getting colder again.

Pure conjecture at this stage, but tonights 12z ECMWF keeps dew points sub 30F through until 192 from the M4 Northwards, with some snowfall shown initially even south of this around T+144 before turning to rain

What this does do, however, is allow us to verify once again (and as suggested after yesterdays 12z) that in this particular setup, the rain/snow boundary lies roughly at the -2c 850mb isotherm:

SK

Cutting it close there SK.

[Snowman.]

Would that be a record?

[Snowman.]

[FrenchScotPilot1]

What's an appropriate word 'better' than Fantasy Island?

Edited March 16, 2013 by FrenchScotPilot1

[Summer Sun]

What's an appropriate word 'better' than Fantasy Isalnd?

Amazing, about time, long over due?

ECM ensemble

Control and Op hover just above the zero line after next weeks cold spell hard to tell from when as the dates are not shown

Edited March 16, 2013 by Gavin.

[No Balls Like Snow Balls]

So the cold air is coming back again. Cold uppers entering northern UK from T72. What's uncertain is how far south does it get and do we see milder air pushing up from the south next weekend.

[BlueSkies_do_I_see]

I often see you guys discussing the relative merits of the models and sometimes criticising the GFS model for its poor performance. This article seems to confirm this:

http://www.climatecentral.org/news/storms-highlight-flaws-in-u.s.-weather-forecasting-model-15744

[No Balls Like Snow Balls]

The GFS 12z 850 ensemble for London shows very good clustering for upper air temps to rise from the 21st

[Snowman.]

Last one I promise.

Edited March 16, 2013 by SN0WM4N

[Captain Shortwave]

Last one I promise.

CFS abuse is not good for the mind

Evening all, pretty interesting weather happening next week as the mild blitzkrieg seems to be hitting problems with the cold digging trenches over the UK. Looks like a good ding dong will happen.North of the line will see snow, south will be mild and no mans land will only be suitable for those with webbed feet.

[Country Boy]

cant belive what i have just seen, are these up to date charts ??

if so i think its shovel into the car boot again

http://weather.unisys.com/gfsx/9panel/gfsx_850_9panel_eur.gif

[mushymanrob]

chris.

its not about dismissing the colder runs but casting doubt over the validity of the most extreme outputs regardless of what they show.

imho its a complete waste of time trying to big up the outcome of one favoured run when what is important is to try to interpret what is most likely to occur. the fact is that as things stand atm, the ecm is the most extreme of options predicted. but no doubt those with an agenda will jump ship and talk about the most favoured data scource when one becomes available. personally i cant see the point, we want the truth dont we?

[frosty ground]

chris.

its not about dismissing the colder runs but casting doubt over the validity of the most extreme outputs regardless of what they show.

imho its a complete waste of time trying to big up the outcome of one favoured run when what is important is to try to interpret what is most likely to occur. the fact is that as things stand atm, the ecm is the most extreme of options predicted. but no doubt those with an agenda will jump ship and talk about the most favoured data scource when one becomes available. personally i cant see the point, we want the truth dont we?

History shows us all that extrems happen. Saying over and over agin that it won't happen without any charts or links is boring and trolling. Simples

[Mark Neal.]

1055mb over Greenland. I can remember a few months ago most on here were searching for the glory of winter synoptics for us in the UK. Some amazing charts for the time of the year!

Iam someone who usually likes to start seeing spring sunshine at this time of year, but because March cold on this scale and the cold to the NE and amazing Northern blocking i am intrigued, those looking for warm weather it will come, and hopefully we will have a decent summer! Not a few days of settled weather in spring and again later in September!

Enjoy

[JP1972]

History shows us all that extrems happen. Saying over and over agin that it won't happen without any charts or links is boring and trolling. Simples

Whoaah, Mushy isn't a troll!

I've been lurking on here for years and yes he does prefer milder weather (no crime in that) but he is being fairly realistic and objective, hardly merits the "troll" badge does it??!!

[frosty ground]

Whoaah, Mushy isn't a troll!

I've been lurking on here for years and yes he does prefer milder weather (no crime in that) but he is being fairly realistic and objective, hardly merits the "troll" badge does it??!!

repeating the same thing over and over again when run after run is showing the opposite is trolling. Objectivity is a good thing. Saying some wont happen based on it being extream is t objective nor is it scientific.

[I remember Atlantic 252]

those uppers look pants on Friday 22nd on 18Z, for the south, not following ECMWF this at all

[Snowman.]

Whoaah, Mushy isn't a troll!

I've been lurking on here for years and yes he does prefer milder weather (no crime in that) but he is being fairly realistic and objective, hardly merits the "troll" badge does it??!!

Agreed mushy isn't a troll just prefers mild weather.

[JP1972]

repeating the same thing over and over again when run after run is showing the opposite is trolling. Objectivity is a good thing. Saying some wont happen based on it being extream is t objective nor is it scientific.

And sometimes human instinct is a reasonably good tool in forecasting.

After all, the Met Office use it along with the models

[chris78]

repeating the same thing over and over again when run after run is showing the opposite is trolling. Objectivity is a good thing. Saying some wont happen based on it being extream is t objective nor is it scientific.

Not sure that holds water, it depends what you think the models are for and how much faith you have in them......there have been countless times models consistently shown extremes only for them not to verify.....so its not scientific to accept output, they are a computer model not a force of nature