theweatheroutlook

Remove ads from site

sizzle
21 October 2015 09:10:20


It's not sizzle but the one magic rule to follow is that the GFS tends to flatten a pattern whilst the ECM over amplifies one.


Originally Posted by: Solar Cycles 

NOTED ! learn something new everyday,. tho all I was saying was that it was the GFS that picked up on the freeze in 2009. in which MR BRIAN was confident that the GFS was right which it was, he did mention this on the other thread winter prospects thread,

Solar Cycles
21 October 2015 09:12:26


NOTED ! learn something new everyday,. tho all I was saying was that it was the GFS that picked up on the freeze in 2009. in which MR BRIAN was confident that the GFS was right which it was, he did mention this on the other thread winter prospects thread,


Originally Posted by: sizzle 

Lol, I'm no weather guru it's just something I've noticed over the years.

Polar Low
21 October 2015 09:28:41

cant conclude just yet Brian a lot of uncertainties that will affect the down stream pattern.


 


WPC's Extended Forecast Discussion

// document.write(getCalendarStyles());
// ]]>
// var cal1x = new CalendarPopup("calarea");

// ]]>

// if (top != self) {top.location.href = self.location.href;}
// ]]>











UNCERTAINTIES OVER THE PACIFIC REMAIN SIMILAR TO PREVIOUS DAYS
WITH THE PRECISE DOWNSTREAM IMPACTS OF INCORPORATING TYPHOON
CHAMPI INTO THE WESTERLIES STILL A QUESTION MARK.  CLOSER TO THE
WEST COAST THERE IS STILL WIDE SPREAD FOR A SYSTEM THAT MAY
APPROACH THE PAC NW DURING THE WEEKEND.  ONE IMPROVEMENT VS PRIOR
DAYS IS THAT THERE SEEMS TO BE MORE AGREEMENT ON A TRACK BTWN
40-50N LATITUDE VS A TRACK INTO THE NERN PAC.  STILL THERE IS POOR
AGREEMENT ON STRENGTH AND EWD EXTENT BEFORE WEAKENING.  CONTINUE
TO LEAN AWAY FROM INDIVIDUAL SOLNS THAT ARE STRONGER/CLOSER TO THE
COAST SINCE THE UPR ENERGY IS HEADING INTO A MEAN RIDGE.  THE 12Z
ECMWF SCENARIO PROVIDED THE MOST REASONABLE OPERATIONAL
ENHANCEMENT TO RECENT ENSEMBLE MEANS.  THE NEW 00Z GFS SFC LOW
STRAYS TO THE SRN FRINGE OF THE MODEL/ENSEMBLE SPREAD. 

BEHIND THIS SYSTEM CONSENSUS CONTINUES TO SHOW AN AMPLIFYING
TROUGH THAT COULD INCORPORATE HURCN OLAF IN SOME FASHION.  LATEST
GFS RUNS HAVE BEEN PARTICULARLY ERRATIC... THE 12Z VERSION CLOSE
TO THE MEANS BUT THE 18Z RUN QUITE SLOW AND THE NEW 00Z RUN
FLAT/FAST.




 


 


 


 



Today's ECM0z ramping up the mildness. As I said yesterday the colder outcome it was showing came with a serious health warning due to lack of support. That's a pretty decent slice of pizza.



Originally Posted by: Brian Gaze 

idj20
21 October 2015 09:45:22


 


Indeed.


I rate the ECM by far the best model available followed by the Met/o and then the GFS.


Originally Posted by: Charmhills 



In terms of medium range forecasting (this is me talking about the 10 days mark) in the Autumn and Winter season, I find that the GFS seems to have a better understanding of the Atlantic, as after all the west-to-east airflow is the "default" setting anyway so it ends up being correct more often than not. 
ECM is probably better than GFS at picking up unusual changes in the medium range time frame (northern blocking pattern, that kind of thing) and if it does get carried forward into the reliable time frame then we can pat it on the back for spotting that the soonest, but since it is the exception rather than the norm, it doesn't always verify in the end. I'm not suggesting it's being a rubbish model, far from it as I include that in my own thoughts, but generally I tend to "believe" the GFS over ECM.
GEM seem to love those little dartboard lows which rarely verifies.
As for NAVGEM, that's about as reliable as a chocolate fireguard but I view it for comedy purposes anyway.
Our own UKMO does an admirable job but it only goes up to six days at a push (or at least what I can see) but I do rely on it for short range forecasting anyway.

That's my own take on it, anyway.

Folkestone Harbour. 
Brian Gaze
21 October 2015 10:00:11


cant conclude just yet Brian a lot of uncertainties that will affect the down stream pattern.


Originally Posted by: Polar Low 


No conclusions, only probabilities. 

Brian Gaze
Berkhamsted
TWO Buzz - get the latest news and views 
"I'm not socialist, I know that. I don't believe in sharing my money." - Gary Numan
Solar Cycles
21 October 2015 10:14:23


 


No conclusions, only probabilities. 


Originally Posted by: Brian Gaze 

Isn't that what all model predictions are Brian.😜

Brian Gaze
21 October 2015 10:22:18


Isn't that what all model predictions are Brian.😜


Originally Posted by: Solar Cycles 


Of course. That was my point. Yesterday's ECM op runs could still be right but the odds were and are stacked against them. No conclusions. 

Brian Gaze
Berkhamsted
TWO Buzz - get the latest news and views 
"I'm not socialist, I know that. I don't believe in sharing my money." - Gary Numan
kmoorman
21 October 2015 16:57:43
I personally would say that ECM and GFS both have their moments. What GFS allows us to do is to view the full ensemble output, for every run, so every Op run can be viewed in context. Even when the Op shows mild, the control and the rest of the pack might be showing something very different. I'm sure if we could see the details of the ECM ensemble (some of which is publicly available) we'd have a much closer view of their relative accuracy when it comes to northern blocking etc.
Home: Durrington, Worthing, West Sussex. (16 ASL)
Work: Canary Wharf, London
Follow me on Twitter @kmoorman1968
Chunky Pea
21 October 2015 18:07:12

I personally would say that ECM and GFS both have their moments. What GFS allows us to do is to view the full ensemble output, for every run, so every Op run can be viewed in context. Even when the Op shows mild, the control and the rest of the pack might be showing something very different. I'm sure if we could see the details of the ECM ensemble (some of which is publicly available) we'd have a much closer view of their relative accuracy when it comes to northern blocking etc.

Originally Posted by: kmoorman 


 


I would agree. The GFS, I have noticed over the last few winters especially, often out performs the ECM in the longer term, particularly when the ECM becomes bullish about huge northern or eastern blocking. Even this summer just gone and the early autumn so far, I have noted that the GFS had gotten particular details more correct, more often, in the short term for here locally than the ECM, specifically regarding rainfall totals and temps.

Current Conditions
https://t.ly/MEYqg 

"You don't have to know anything to have an opinion"
--Roger P, 12/Oct/2022
Whether Idle
21 October 2015 18:23:01


 


 


I would agree. The GFS, I have noticed over the last few winters especially, often out performs the ECM in the longer term, particularly when the ECM becomes bullish about huge northern or eastern blocking. Even this summer just gone and the early autumn so far, I have noted that the GFS had gotten particular details more correct, more often, in the short term for here locally than the ECM, specifically regarding rainfall totals and temps.


Originally Posted by: Chunky Pea 


Couldn't agree more ECM is far too keen on phantom easterlies that belong only in Fantasy Island.  GFS at day 6 I prefer ECM; though it must be said they are both not without their foibles.  They're only models after all.  Number crunchers with limitations.

Dover, 5m asl. Half a mile from the south coast.
Arcus
21 October 2015 18:24:29


 


 


I would agree. The GFS, I have noticed over the last few winters especially, often out performs the ECM in the longer term, particularly when the ECM becomes bullish about huge northern or eastern blocking. Even this summer just gone and the early autumn so far, I have noted that the GFS had gotten particular details more correct, more often, in the short term for here locally than the ECM, specifically regarding rainfall totals and temps.


Originally Posted by: Chunky Pea 


A quick scan of the verification stats for Global SLP shows that over the past month the ECM has verified best at T+120, closely followed by UKMET, and PR4DEV (PR4DEV? is that a renamed version of something?), then the Canadian model, then GFS. Extend it out to T+240 and it's still ECM ahead of GFS.


Every time I have checked, ECM has always been outperforming GFS. 


http://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/


 


 

Ben,
Nr. Easingwold, North Yorkshire
30m asl
LeedsLad123
21 October 2015 18:55:39


 


A quick scan of the verification stats for Global SLP shows that over the past month the ECM has verified best at T+120, closely followed by UKMET, and PR4DEV (PR4DEV? is that a renamed version of something?), then the Canadian model, then GFS. Extend it out to T+240 and it's still ECM ahead of GFS.


Every time I have checked, ECM has always been outperforming GFS. 


http://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/


 


 


Originally Posted by: Arcus 


Yes, ECM is king. Look at how well it handled Joaquin.

Whitkirk, Leeds - 85m ASL.
Polar Low
21 October 2015 19:04:42

Again I like the general theme from 12z ecm this evening. if and a big if ,, we could get that going well into Nov that would be a good trend for winter.





Chunky Pea
21 October 2015 19:04:47


 


A quick scan of the verification stats for Global SLP shows that over the past month the ECM has verified best at T+120, closely followed by UKMET, and PR4DEV (PR4DEV? is that a renamed version of something?), then the Canadian model, then GFS. Extend it out to T+240 and it's still ECM ahead of GFS.


Every time I have checked, ECM has always been outperforming GFS. 


http://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/


 


 


Originally Posted by: Arcus 


 


I have no doubt on a global level, but I am going by my own observations of both models for my specific location. Only last week, the ECM forecast maxes here below 10c for a number of days which proved to be totally wrong. The GFS never showed such and proved to be completely right. I have read before that the ECM has a bias towards over amplifying northern or eastern blocking during the winter months and I think this has been proved often enough over the last few winters. The GFS, I have found, is often more realistic regarding this between 168-240 hrs during the season.

Current Conditions
https://t.ly/MEYqg 

"You don't have to know anything to have an opinion"
--Roger P, 12/Oct/2022
LeedsLad123
21 October 2015 19:09:13


 


 


I have no doubt on a global level, but I am going by my own observations of both models for my specific location. Only last week, the ECM forecast maxes here below 10c for a number of days which proved to be totally wrong. The GFS never showed such and proved to be completely right. I have read before that the ECM has a bias towards over amplifying northern or eastern blocking during the winter months and I think this has been proved often enough over the last few winters. The GFS, I have found, is often more realistic regarding this between 168-240 hrs during the season.


Originally Posted by: Chunky Pea 


I'm afraid your own observations are worth nothing, however.. and more to the point, where does ECM forecast maximum  temperatures?

Whitkirk, Leeds - 85m ASL.
Polar Low
21 October 2015 19:13:45

Indeed which means that ecm gets more things right than most "most" of the time.


don't forget gfs does run 4 times a day it has every chance to outwit Ive seen her  leave the back door many times in disgrace.


I cant forgive her all those times it refused to undercut only to change at t76 and all those dart board lows just so silly of her.


I could go but Id bore


Its ecm for me


 


 



 


A quick scan of the verification stats for Global SLP shows that over the past month the ECM has verified best at T+120, closely followed by UKMET, and PR4DEV (PR4DEV? is that a renamed version of something?), then the Canadian model, then GFS. Extend it out to T+240 and it's still ECM ahead of GFS.


Every time I have checked, ECM has always been outperforming GFS. 


http://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/


 


 


Originally Posted by: Arcus 

Chunky Pea
21 October 2015 19:17:17


 


I'm afraid your own observations are worth nothing, however.. and more to the point, where does ECM forecast maximum  temperatures?


Originally Posted by: LeedsLad123 


 


Really? how do you make that out? my observations for my area are quite valid and I stand by them. You saying they are not, however, is worthless.

Current Conditions
https://t.ly/MEYqg 

"You don't have to know anything to have an opinion"
--Roger P, 12/Oct/2022
Chunky Pea
21 October 2015 19:21:43


Again I like the general theme from 12z ecm this evening. if and a big if ,, we could get that going well into Nov that would be a good trend for winter.


Originally Posted by: Polar Low 


Is there any evidence to suggest that weather patterns in early November are a good indicator for general trends during the following winter?

Current Conditions
https://t.ly/MEYqg 

"You don't have to know anything to have an opinion"
--Roger P, 12/Oct/2022
Polar Low
21 October 2015 19:47:34

 


It did in 2010 But I don’t have the data Patrick from other cold or rather cold winters perhaps Gav or Kev have that.


https://en.wikipedia.org/wiki/Winter_of_2010%E2%80%9311_in_Europe


 


My general thoughts from the original post are that’s never too early to get a cold snap in the uk and Id much rather see imo what im seeing rather than an onslaught of persistent  s/westerly type stormy weather pattern which is hard to break and takes weeks out of a potentially cold period ie late Nov- end of Dec


 



 


Is there any evidence to suggest that weather patterns in early November are a good indicator for general trends during the following winter?


Originally Posted by: Chunky Pea 

Chunky Pea
21 October 2015 20:06:47


 


It did in 2010 But I don’t have the data Patrick from other cold or rather cold winters perhaps Gav or Kev have that.


https://en.wikipedia.org/wiki/Winter_of_2010%E2%80%9311_in_Europe


 


My general thoughts from the original post are that’s never too early to get a cold snap in the uk and Id much rather see imo what im seeing rather than an onslaught of persistent  s/westerly type stormy weather pattern which is hard to break and takes weeks out of a potentially cold period ie late Nov- end of Dec


Originally Posted by: Polar Low 



I'd gladly take a stormy pattern over the current drudge though, a drudge that does not look like ending anytime soon.

Current Conditions
https://t.ly/MEYqg 

"You don't have to know anything to have an opinion"
--Roger P, 12/Oct/2022
Arcus
21 October 2015 22:16:08
So having established that the top-tastic countdown in verification stats for the well known models is what it more or less always was for the last 5 years (ECM 1st, UKMO 2nd, GEM 3rd, GFS 4th), what is the PR4DEV model?
Ben,
Nr. Easingwold, North Yorkshire
30m asl
nsrobins
21 October 2015 23:10:14

So having established that the top-tastic countdown in verification stats for the well known models is what it more or less always was for the last 5 years (ECM 1st, UKMO 2nd, GEM 3rd, GFS 4th), what is the PR4DEV model?

Originally Posted by: Arcus 


The little I know about PR(4D)EV Ben is that it is an experimental model run in parallel to GFS main using a new 4D spatial reconstruction. 


 

Neil
Fareham, Hampshire 28m ASL (near estuary)
Stormchaser, Member TORRO
Whether Idle
22 October 2015 06:26:58

So having established that the top-tastic countdown in verification stats for the well known models is what it more or less always was for the last 5 years (ECM 1st, UKMO 2nd, GEM 3rd, GFS 4th), what is the PR4DEV model?

Originally Posted by: Arcus 


UKMO is my favourite model as it only goes to 144 hours.  How much simpler winter model watching would be if all the models limited their output to the semi reliable range?

Dover, 5m asl. Half a mile from the south coast.
Gooner
22 October 2015 06:51:15


 


UKMO is my favourite model as it only goes to 144 hours.  How much simpler winter model watching would be if all the models limited their output to the semi reliable range?


Originally Posted by: Whether Idle 


Spot on , stop us posting charts like this that are just outside the reliable timeframe


 


V

Remember anything after T120 is really Just For Fun



Marcus
Banbury
North Oxfordshire
378 feet A S L


Brian Gaze
22 October 2015 08:27:10

So having established that the top-tastic countdown in verification stats for the well known models is what it more or less always was for the last 5 years (ECM 1st, UKMO 2nd, GEM 3rd, GFS 4th), what is the PR4DEV model?

Originally Posted by: Arcus 


It's the experimental GFS model but isn't currently publicly available on NCEP para.

Brian Gaze
Berkhamsted
TWO Buzz - get the latest news and views 
"I'm not socialist, I know that. I don't believe in sharing my money." - Gary Numan

Remove ads from site

Ads