Page 25 of 104 < 1 2 ... 23 24 25 26 27 ... 103 104 >
Topic Options
#1423364 - 15/05/2017 17:28 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: RC]
Nature's Fury Offline
Weatherzone Addict

Registered: 29/11/2009
Posts: 1877
Loc: Brisbane Western Suburbs
Originally Posted By: RC
It is interesting looking at the CFS V2 Nino 3.4 prediction.

Where it is now compared to two months ago.



Some of the latest runs are moderate La Nina. Pretty crazy turnaround.

Top
#1423458 - 16/05/2017 20:37 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Nature's Fury]
adon Offline
Meteorological Motor Mouth

Registered: 19/08/2004
Posts: 5295
Loc: Not tellin!
Pretty much confirms the opinion that a lot of people have of the models. They are reactive. I also think that have some assumptions coded into them to favour Elnino developing. Either that or the models are using theories that are in fact not correct and drivers of climate maybe?

It's not the first time that models have done flips from one to another, some more extreme too. It just says to me that more climatic research must be done on the fundamentals to identify the drivers with more influence.

I doubt whether a totally accurate long range forecast will ever be achievable. The margin for error is just way too small to be able to work out too far.

Top
#1423473 - 17/05/2017 00:25 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: adon]
marakai Offline
Weatherzone Addict

Registered: 05/01/2006
Posts: 1746
Loc: Maryfarms NQ
Originally Posted By: adon
Pretty much confirms the opinion that a lot of people have of the models. They are reactive. I also think that have some assumptions coded into them to favour Elnino developing. Either that or the models are using theories that are in fact not correct and drivers of climate maybe?

It's not the first time that models have done flips from one to another, some more extreme too. It just says to me that more climatic research must be done on the fundamentals to identify the drivers with more influence.

I doubt whether a totally accurate long range forecast will ever be achievable. The margin for error is just way too small to be able to work out too far.


Something that I run into all the time and have done for a very long time on these forums. Models are just Models.
Computer generated outcome based upon pre-ordained outcomes is not science.
You just can't model a chaotic system without including all of the variables, and to date we don't have even half of the variables as a known ingredient for the current models to work with.

Science is not infallible, it never is. Even the best of current cutting edge science is prone to either mistakes, lack of input, or the fact that the future will falsify current knowledge.

Even the most state of the art current Climate models are prone to the information with which they are programmed with. Who could program a Climate Model on the behaviour of Cloud patterns or the behaviour of Humidity over the Equator with greater than 95% confidence based upon current knowledge ?

Reading just the current past 12 months or so of this thread/associated ones is proof positive that not even the best funded and apparent best qualified people on the planet actually don't have a clue as to what will happen until it actually has.

Don't get me wrong here, I'm not claiming any more knowledge than a fence post, just pointing out that the prognostications of those that Taxpayers spend Millions upon Millions on are no closer apparently to actual predictive science wise than we were 40 years ago.

Top
#1423476 - 17/05/2017 07:07 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
ColdFront Offline
Meteorological Motor Mouth

Registered: 29/06/2008
Posts: 17908
Loc: Wide Bay..Near the beach
I suggested the ENSO models were reactive a few years back during the 2010/11 Lanina in here and it was dismissed by some as nonsense. I recall Adon agreeing back then that they are reactive as well as one of the Darwin members.

In the recent strong El Nino the predicted peak was moved upward every week in response to what had already happen. The same happened prior to the aforementioned La Nina.

There is still a chance of an El Nino by Spring , however writing in depth essays about what "will happen" prior to the predictability barrier is probably not a good idea .Especially when you have been attacking another member for a prediction they made sometime way back when.

_________________________
"Don't steal. The government hates competition."

Top
#1423479 - 17/05/2017 08:03 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
retired weather man Offline
Weatherzone Addict

Registered: 01/07/2007
Posts: 4323
Loc: Wynnum
I too have suggested reactive models particularly in the recent situations of El Ninos and have also suggested maybe that certain worldwide changes over recent decades might mean an alteration to some parameters in determining El Nino situations.


Edited by retired weather man (17/05/2017 08:08)
_________________________
Wyn Nth 2017-Jan107.6(158),Feb24.4(149),Mar410.2(123),Apr8.8(98),May32.2(101),Jun91.4(75),Jul16.6(57),Aug6.4(45),Sep4.2(33),Oct217.4(93),Nov41.8,YTD961.2(1043.4)

Top
#1423529 - 17/05/2017 14:31 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: marakai]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7155
Loc: Adelaide Hills.
Originally Posted By: marakai
Models are just Models.
Computer generated outcome based upon pre-ordained outcomes is not science.
You just can't model a chaotic system without including all of the variables, and to date we don't have even half of the variables as a known ingredient for the current models to work with.

Science is not infallible, it never is. Even the best of current cutting edge science is prone to either mistakes, lack of input, or the fact that the future will falsify current knowledge.

Even the most state of the art current Climate models are prone to the information with which they are programmed with. Who could program a Climate Model on the behaviour of Cloud patterns or the behaviour of Humidity over the Equator with greater than 95% confidence based upon current knowledge ?

Reading just the current past 12 months or so of this thread/associated ones is proof positive that not even the best funded and apparent best qualified people on the planet actually don't have a clue as to what will happen until it actually has.

On that I agree, but hey, this is my opinion. Facts are needed smile . We need to understand how a model works smile to have any sort of confidence in it beyond random guess work.

Top
#1423573 - 18/05/2017 00:44 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
MAFILI Offline
Cloud Gazer

Registered: 24/11/2011
Posts: 19
Loc: São Paulo/Brazil
Modulation of Bjerknes feedback on the decadal variations in ENSO predictability

Abstract

Clear decadal variations exist in the predictability of the El Niño–Southern Oscillation (ENSO), with the most recent decade having the lowest ENSO predictability in the past six decades. The Bjerknes Feedback (BF) intensity, which dominates the development of ENSO, has been proposed to determine ENSO predictability. Here we demonstrate that decadal variations in BF intensity are largely a result of the sensitivity of the zonal winds to the zonal sea level pressure (SLP) gradient in the equatorial Pacific. Furthermore, the results show that during low-ENSO predictability decades, zonal wind anomalies over the equatorial Pacific are more linked to SLP variations in the off-equatorial Pacific, which can then transfer this information into surface temperature and precipitation fields through the BF, suggesting a weakening in the ocean-atmosphere coupling in the tropical Pacific. This result indicates that more attention should be paid to off-equatorial processes in the prediction of ENSO.

http://onlinelibrary.wiley.com/doi/10.1002/2016GL071636/abstract;jsessionid=5AB589F88B6C18D981C03578DE0D3073.f03t02

Top
#1423690 - 19/05/2017 07:57 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 2751
Loc: Buderim
What does 'the models are reactive' mean beyond they change their forecasts up and down as new data is entered into the model?

Top
#1423691 - 19/05/2017 08:04 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: marakai]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 2751
Loc: Buderim
Originally Posted By: marakai


Something that I run into all the time and have done for a very long time on these forums. Models are just Models.
Computer generated outcome based upon pre-ordained outcomes is not science.
You just can't model a chaotic system without including all of the variables, and to date we don't have even half of the variables as a known ingredient for the current models to work with.

Science is not infallible, it never is. Even the best of current cutting edge science is prone to either mistakes, lack of input, or the fact that the future will falsify current knowledge.

Even the most state of the art current Climate models are prone to the information with which they are programmed with. Who could program a Climate Model on the behaviour of Cloud patterns or the behaviour of Humidity over the Equator with greater than 95% confidence based upon current knowledge ?

Reading just the current past 12 months or so of this thread/associated ones is proof positive that not even the best funded and apparent best qualified people on the planet actually don't have a clue as to what will happen until it actually has.

Don't get me wrong here, I'm not claiming any more knowledge than a fence post, just pointing out that the prognostications of those that Taxpayers spend Millions upon Millions on are no closer apparently to actual predictive science wise than we were 40 years ago.



Models are calculations based on the laws of physics, and are just as much science as any other prediction made based on scientific principles. Very simple weather models could be run on pen and paper, but more calculations provides more accuracy, and so computer power is used to do better than even the biggest army of humans with pen and paper calculations could ever hope to achieve.

No scientific prediction can take into account all the possible variables, but science identifies what are considered the most important variables and ignores the rest. For example calculations of the motion of bodies under gravitational interactions - which can be shown to be inherently chaotic for any number of bodies from 3 or more - can be quite accurate ignoring relativistic effects etc.

It is easy to look at the past history of ENSO predictions for some models such as ECMWF, and see that in 95% of times, the actual result has been within the range forecast by the model.

Top
#1423693 - 19/05/2017 08:28 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 2751
Loc: Buderim
Current forecasts suggest that the surge in trade winds is now winding down. Which would make for a relatively minor interruption to the development of an el nino. Comparing the recent trade wind pattern to years such as 2006 or 2009 shows much less westerly activity in those years than we have been experiencing, and significant periods of normal to above normal trades in between. We need to see a lot more strengthening of trades than we have seen so far to make a significant difference. I think the recent update to EC which has reduced the forecast el nino intensity by a small amount is a pretty much reasonable response to current conditions.

Top
#1423694 - 19/05/2017 08:28 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Mike Hauber]
ColdFront Offline
Meteorological Motor Mouth

Registered: 29/06/2008
Posts: 17908
Loc: Wide Bay..Near the beach
Originally Posted By: Mike Hauber
What does 'the models are reactive' mean beyond they change their forecasts up and down as new data is entered into the model?


It means some people place way too much faith in them given that by your own admission "accurate" means within a 50% variable. In other words , if they suggest warm neutral and we get a record el nino that's within their forecast because they forwcast warm?

Yes,all models work on modifying output when input varies /changes, however in the case of ENSO they are sometimes spectacularly wrong (including EC) when we exit the "predictibility" barrier as was seen recently. They are very much reactive.
_________________________
"Don't steal. The government hates competition."

Top
#1423722 - 19/05/2017 10:53 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Mike Hauber]
snowbooby Offline
Weather Freak

Registered: 11/05/2016
Posts: 108
My twenty cents.Seems to me the attitude to modelling is all wrong - the output of models may be "wrong"

Originally Posted By: Mike Hauber
Originally Posted By: marakai


You just can't model a chaotic system without including all of the variables, and to date we don't have even half of the variables as a known ingredient for the current models to work with.



No scientific prediction can take into account all the possible variables, but science identifies what are considered the most important variables and ignores the rest.


but the expectations that are so often disappointed here and elsewhere are what was "wronger" in the first place.

"Reactive" implies at least that data are belatedly “found” to justify a change in output after the event. That would have to be doubtful.

Top
#1423727 - 19/05/2017 11:41 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Mike Hauber]
Mega Offline
Meteorological Motor Mouth

Registered: 03/02/2003
Posts: 6619
Loc: Maryborough, Wide Bay, QLD
Originally Posted By: Mike Hauber
Current forecasts suggest that the surge in trade winds is now winding down. Which would make for a relatively minor interruption to the development of an el nino. Comparing the recent trade wind pattern to years such as 2006 or 2009 shows much less westerly activity in those years than we have been experiencing, and significant periods of normal to above normal trades in between. We need to see a lot more strengthening of trades than we have seen so far to make a significant difference. I think the recent update to EC which has reduced the forecast el nino intensity by a small amount is a pretty much reasonable response to current conditions.


EC is stronger and just keeps the trades going:



However it is interesting you bring up 2006 as the subsurface looked quite similar then than it does now.

Top
#1423729 - 19/05/2017 11:41 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
Mike Hauber Offline
Weatherzone Addict

Registered: 13/07/2007
Posts: 2751
Loc: Buderim
Originally Posted By: ColdFront

It means some people place way too much faith in them given that by your own admission "accurate" means within a 50% variable. In other words , if they suggest warm neutral and we get a record el nino that's within their forecast because they forwcast warm?
.


Not sure what you are referring to. Maybe I said that 50% of the time the actual result will be warmer than forecast, and 50% cooler?

I maintain that in the vast majority of cases the actual result is within the boundary of what is forecast, especially for better models such as EC which I feel is very good at generating a realistic range. Model forecasts should be looked at as a range, and not as a single precise value.

Consider what is probably the worst EC forecast I can find, from March 2014. The actual result was still within the range, even if only just barely. More common is the forecast from March 2016 where the actual result is pretty much in the middle of the range.

If the actual result was not at the bottom or top of the range occasionally, and is always near the middle, then I'd criticise the model for producing too wide a range. If the correct range is being forecast then there should be at least some instances where the actual result is at the extreme values of this range.

Top
#1423747 - 19/05/2017 13:07 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
ColdFront Offline
Meteorological Motor Mouth

Registered: 29/06/2008
Posts: 17908
Loc: Wide Bay..Near the beach
Originally Posted By: ColdFront
Originally Posted By: Mike Hauber

In both cases the actual result was within the range forecast by EC, which is what is expected in weather modelling. Currently that means anything from barely warm neutral to very strong el nino.


That's quite a hedge at 50%. Toss a coin?


This.
_________________________
"Don't steal. The government hates competition."

Top
#1423834 - 20/05/2017 14:37 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: ColdFront]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7155
Loc: Adelaide Hills.
Weather systems are not chaotic* -- they are deterministic. The "chaotic" part comes from human beings, through science, not being able to precisely replicate the climate's behaviour at each time step...because it is impossible by default! No amount of technological advancement is going to yield a perfect forecast, only more accuracy and narrower margins of error! We have a robust-enough understanding already to get some accuracy, it's a question of building on that smile .

*Although I will acknowledge it's a thought-provoking idea.


Edited by Seira (20/05/2017 14:46)

Top
#1423845 - 20/05/2017 17:46 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Seira]
Snowy Hibbo Offline
Weather Freak

Registered: 30/12/2016
Posts: 314
Loc: Matlock, Victoria.
Originally Posted By: Seira
Weather systems are not chaotic* -- they are deterministic. The "chaotic" part comes from human beings, through science, not being able to precisely replicate the climate's behaviour at each time step...because it is impossible by default! No amount of technological advancement is going to yield a perfect forecast, only more accuracy and narrower margins of error! We have a robust-enough understanding already to get some accuracy, it's a question of building on that smile .

*Although I will acknowledge it's a thought-provoking idea.

I definitely agree with you. We will never get a perfect forecast, there's always errors with humans(robots might be able to get a perfect forecast, but that's a completely different argument...). We just need to attempt to fix our mistakes, in this case do more research.
_________________________
Long term forecaster
http://longrangesnowcenter.blogspot.com.au
Just ask, I'm more than happy to provide.

Top
#1423863 - 21/05/2017 02:29 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Mike Hauber]
marakai Offline
Weatherzone Addict

Registered: 05/01/2006
Posts: 1746
Loc: Maryfarms NQ
Originally Posted By: Mike Hauber
Originally Posted By: marakai


Something that I run into all the time and have done for a very long time on these forums. Models are just Models.
Computer generated outcome based upon pre-ordained outcomes is not science.
You just can't model a chaotic system without including all of the variables, and to date we don't have even half of the variables as a known ingredient for the current models to work with.

Science is not infallible, it never is. Even the best of current cutting edge science is prone to either mistakes, lack of input, or the fact that the future will falsify current knowledge.

Even the most state of the art current Climate models are prone to the information with which they are programmed with. Who could program a Climate Model on the behaviour of Cloud patterns or the behaviour of Humidity over the Equator with greater than 95% confidence based upon current knowledge ?

Reading just the current past 12 months or so of this thread/associated ones is proof positive that not even the best funded and apparent best qualified people on the planet actually don't have a clue as to what will happen until it actually has.

Don't get me wrong here, I'm not claiming any more knowledge than a fence post, just pointing out that the prognostications of those that Taxpayers spend Millions upon Millions on are no closer apparently to actual predictive science wise than we were 40 years ago.



Models are calculations based on the laws of physics, and are just as much science as any other prediction made based on scientific principles. Very simple weather models could be run on pen and paper, but more calculations provides more accuracy, and so computer power is used to do better than even the biggest army of humans with pen and paper calculations could ever hope to achieve.

No scientific prediction can take into account all the possible variables, but science identifies what are considered the most important variables and ignores the rest. For example calculations of the motion of bodies under gravitational interactions - which can be shown to be inherently chaotic for any number of bodies from 3 or more - can be quite accurate ignoring relativistic effects etc.

It is easy to look at the past history of ENSO predictions for some models such as ECMWF, and see that in 95% of times, the actual result has been within the range forecast by the model.


I would beg to disagree here:

Quote:
Models are calculations based on the laws of physics, and are just as much science as any other prediction made based on scientific principles.

Models are just 'Models' They may have some grounding in Science per se, but in Science when something does not work out or is disproved/falsified it is thrown out and a new avenue of causation is entertained until it is either accepted or otherwise disproved.
You just don't tweak the variables of an experiment until it tells you what you want it to. If a Model hindcasts well but is unable to accurately predict, then it is for all means and purposes useless for it's stated purpose other than as a base for further research practices.

Quote:
No scientific prediction can take into account all the possible variables, but science identifies what are considered the most important variables and ignores the rest. For example calculations of the motion of bodies under gravitational interactions - which can be shown to be inherently chaotic for any number of bodies from 3 or more - can be quite accurate ignoring relativistic effects etc.


I agree completely with your first sentence, but then I find myself questioning the ability of said models to interpret things such as localised humidity and the variables of cloud cover in tropical thunderstorms, the cooling effect they have in areas such as Darwin on a semi daily basis throughout the wet season up there and multiple other places along and either side of the Equator as the MJO rolls through such areas.
The onset of monsoon activity in the north of our country alone , and the localised effects such as lightning storms placing particulate matter into the atmosphere through sparking bushfire's and on and on and on.
Models as you said just sort of take the basics, without all the variables into account, and they are based through sheer data limitations alone upon a very small window of time compared to the magnitude of just the last hundred thousand years, the last million years, or even the just the last ten thousand years since the last real ice age.
Thirty to Fifty years of all our quasi accurate climate data compared to even the history of Human agriculture of a mere 10-15 thousand years, is not exactly cause for celebration nor is it cause for blindly accepting the data that a computer spits out after being asked to compute an algorithm based upon inputs that are unable to even compute cloud cover.

Sadly though the whole of the human race is currently asked to accept that a bunch of scientists who sit around on the taxpayers dollar and create all these 'Climate models' which are consistently wrong and have zero skill at all of predicting our future weather.

And yet every week day, the likes of Koshie or Trioli are harping away about something that has no actual scientific proof at all.

Poor bugger me my country!!!

Top
#1423873 - 21/05/2017 12:37 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: marakai]
Seira Offline
Meteorological Motor Mouth

Registered: 27/08/2003
Posts: 7155
Loc: Adelaide Hills.
Originally Posted By: marakai
And yet every week day, the likes of Koshie or Trioli are harping away about something that has no actual scientific proof at all.

You’ll have to enlighten me on that one smile .

Top
#1423876 - 21/05/2017 13:42 Re: Climate Driver Discussion 2017 (Enso, IOD, PDO ,SAM etc) [Re: Seira]
Snowy Hibbo Offline
Weather Freak

Registered: 30/12/2016
Posts: 314
Loc: Matlock, Victoria.
Originally Posted By: Seira
Originally Posted By: marakai
And yet every week day, the likes of Koshie or Trioli are harping away about something that has no actual scientific proof at all.

You’ll have to enlighten me on that one smile .

Kochie and Trioli are both on breakfast TV(Channel 7 and ABC respectively). I certainly don't think breakfast TV shows science correctly.
_________________________
Long term forecaster
http://longrangesnowcenter.blogspot.com.au
Just ask, I'm more than happy to provide.

Top
Page 25 of 104 < 1 2 ... 23 24 25 26 27 ... 103 104 >


Who's Online
1 registered (1 invisible), 225 Guests and 3 Spiders online.
Key: Admin, Global Mod, Mod
Today's Birthdays
beefy, falay002, Paul79, Ruckle
Forum Stats
29401 Members
32 Forums
23726 Topics
1468992 Posts

Max Online: 2925 @ 02/02/2011 22:23
Satellite Image