Note: Short term predictions are relatively easy, it remains to be seen if this holds up over the long term. I have my doubts. – Anthony
Guest post by Frank Lemke
The Global Warming Prediction Project is an impartial, transparent, and independent project where no public, private or corporate funding is involved. It is about original concepts and results of inductive self-organizing modeling and prediction of global warming and related problems.
In September 2011, we presented a medium-term (79 months) quantitative prediction of monthly global mean temperatures based on an interdependent system model of the atmosphere developed by KnowledgeMiner, which was also discussed at Climate Etc. in October 2011. This model describes a non-linear dynamic system of the atmosphere consisting of 5 major climate drivers: Ozone concentration, aerosols, radiative cloud fraction, and global mean temperature as endogenous variables and sun activity (sunspot numbers) as exogenous variable of the system. This system model was obtained from monthly observation data of the past 33 years (6 variables in total: the 5 variables the system is actually composed of (see above) plus CO2, which, however, has not been identified as relevant system variable), exclusively, by unique self-organizing knowledge extraction technologies.
Now, more than a year has passed, and we can verify what has been predicted relative to the temperatures, which have really been measured (fig. 1).
Verifying the prediction skill of the system model from April 2011 to December 2012, the accuracy of the most likely forecast (solid red line) remains at a high level of 75%, and the accuracy relative to prediction uncertainty (pink area) is an exceptional 98%. Given the noise in the data (presumably incomplete set of system variables considered, noise added during measurement and preprocessing of raw observation data, or random events, for example), this clearly confirms the validity of the system model and its forecast.
In comparison, the IPCC AR4 A1B projection currently shows a prediction accuracy of 23% (September 2007 – December 2012, 64 months) and just 7% accuracy for the same forecast horizon as applied for the system model (April 2011 – December 2012, 21 months).
The two models, IPCC model and atmospheric system model, use two very different modeling approaches: theory-driven vs data-driven modeling. The IPCC model is based essentially on AGW theory by emission of greenhouse gases, namely CO2, the presented atmospheric system model on the other hand is a CO2-free prediction model. It is described by 5 other variables. The IPCC model shows a prediction accuracy of 7% and the atmospheric system model an accuracy of 75% for the same most recent 21 months of time…
The climate system is a complex system that consists of a number of variables, which are connected interdependently, nonlinearly and dynamically and where it is not clear, which are the causes and which are the effects. The simplistic linear cause-effect relationship “more atmospheric CO2 = higher temperatures” the IPCC model is based on is not an adequate tool to describe the complexity of the atmosphere sufficiently.
Read the complete post here:
http://climateprediction.eu/cc/Main/Entries/2013/1/21_What_Drives_Global_Warming_-_Update.html
MiCro:
Thankyou for your clarification at January 27, 2013 at 3:32 am which says
You make a good and very important point.
Can you write up your work for publication?
Richard
richardscourtney commented
“Can you write up your work for publication?”
I can, and have put some of it together, follow the link in my name.
But I will most likely need some help with some of it, if you follow the link, you can contact me from there, we can take this off line, and see what can be done.
Moe:
I have read your flaming post at January 27, 2013 at 4:01 am.
For the information of others, I am not and I never have been a “PR guy”. And my words to you have been very mild responses to your egregious behaviour.
Your claim that I am a “paid fibber” is two lies in two words. However, I suspect you are a paid troll. And I never – not ever – attempt to “mislead people”. Please refrain from assuming that others do as you do.
Your assertion that you are ” a scientist” is patently ridiculous. Indeed, your persistently stated ignorance of statistics is so profound that it beggars belief.
I have repeatedly explained to you that the statistical significance of a trend depends on the variance of the data set but you persist in demanding – and have yet again demanded – that I define a time frame for a statistically significant trend. I did: I stated that for 16+ years the trend in global temperature has been indiscernible from zero (at 95% confidence).
I will spell it out for you in hope that you will be able to understand (although I doubt it when I strongly suspect you are schoolboy posting from the laptop in your bedroom).
A linear trend is a straight line with formula x = a*y + c
If a set of three or more points lie on that line then they have the trend (i.e. the slope) specified as a.
So, when there are a set of points in a time series it can be determined if they approximate to a linear trend by means of a least-squares fit. This determines the line through the points which provides the lowest value for the squared values of the points from the line. This line is the trend.
PLEASE NOTE THAT THE TIME LENGTH OF THE TIME SERIES IS NOT RELEVANT.
Obviously, a least-squares analysis will provide a result for a whether or not there is a valid trend. The validity of the trend is indicated by the square of its correlation coefficient (r^2). An r^2 near 1.0 indicates the data provides a ‘true’ (i.e. useful) trend and a r^2 near 0 indicates the apparent trend is not valid. I can’t post equations here so you will need to look up how to calculate an r^2 for yourself.
The standard deviation (s.d.) of the data can also be determined and is the inverse of the variance of the points from the determined trend. And confidence limits can be determined for the trend. In principle, any confidence limits can be calculated. Strict sciences (e.g. physics) use 99% confidence. This states there is a 1 in 100 chance that the trend is not within those limits, Climatology is less rigorous and normally uses 95% confidence limits. This states that there is 1 in 20 chance that the trend is within those limits.
Any scientist would know this and would not pester about provision of a time frame for a statistically significant trend, but may have requested an r^2 statistic for the stated trend.
Another of your several errors was your claim that every molecule of anthropogenic CO2 accumulates in the atmosphere. That was so ignorant an error that I ridiculed it. But I notice that Lord Stern has recently made that same error and Willis Eschenbach has very recently posted an article on WUWT about his making it.
You are a know-nothing, anonymous pest trying – with some success – to disrupt this thread.
I will not feed your ignorant, irrelevant and stupid trolling with any more replies whether or not they are flaming posts aimed at me.
Richard
Moderator:
My reply to Moe seems to have gone in the ‘bin’. Please retrieve it.
Richard
Oh dear.
In my anger, I made some serious errors in my post at January 27, 2013 at 11:42 am
I wrote
This determines the line through the points which provides the lowest value for the squared values of the points from the line.
but intended to write
This determines the line through the points which provides the lowest value for the squared values of the distances of the points from the line.
I wrote
This states that there is 1 in 20 chance that the trend is within those limits.
but intended to write
This states that there is 1 in 20 chance that the trend is not within those limits.
Sorry. I should not rapidly type the keyboard while angry at outrage.
Richard
========================================================
It would seem that “Mother Nature” doesn’t often cooperate with “Model Nature”.
Gunga din says ‘Get with the program! Do we have to educate every noob who comes along??’
Well it appears you are one of th noobs (your term) that need educating. Your answer was wrong!
Richard, Ganga fas had a go and got it wrong, perhaps you can grab a stats package and some satellite data (there is enough) and tell him how long a time span you need to get a statistically significant answer to determine the Earth’s temperature trend with confidence. BTW, you claim that is not time dependant is incorrect. You are dealing with a time series, and statistical significant answer can be achieved by selecting an appropriate time period.
Moe:
Your post at January 27, 2013 at 5:28 pm says in total
I have never heard of “Gangs fas” or what he/she did or said that was “wrong”.
I went to the bother of explaining how a time series is analysed to determine a linear trend and to assess if it is statistically significant (see my post addressed to you at January 27, 2013 at 11:42 am with corrigendum at January 27, 2013 at 12:19 pm).
You have not stated any error in that explanation because it is correct.
Instead, you have repeated your mistaken belief that the statistical significance of a time series is a function of the analysed time period. It is NOT: it is a function of the variance of the data (as I explained with sufficient detail for it to be understood by an average 12-year-old).
Your reply demonstrates that the explanation is beyond your understanding.
Clearly, you need to waste less time blogging from your bedroom and to spend that time in your mathematics lessons at school which is where you should be.
This thread is not the place for little boys to pretend to be grown ups, and you are wasting space on the thread. Stop skiving off from school and learn what you are trying – and failing – to talk about.
Richard
===========================================================================
Actually, I didn’t say that. (See my comment “Gunga Din says:
January 27, 2013 at 12:22 pm”) However I do not disagree with what D. B. Stealey said.
(Can I interest you in some Mint Toe-Paste? I need to use it myself at times. 😎
MiCro:
re your post at January 28, 2013 at 7:18 am.
Please see the inbox of your blog’s contact facility.
Richard
Richard, you are incorrect in asserting that statistical significance is not a function of time. Repeatly stating this the more your math (and scientific) credibility suffers. You said your self that statistical significance cannot be established with sufficient confidence in 16 years. However extending the time period you will get statistical significance.
I am surprised that you keep this up instead of doing the analysis. You’re a scientist (in what area I wonder) so you should know how to calculate significance. But if you wish to continue to show your true colors to the threaders, I will continue to point out your error.
You could do the right thing and work out how far you must go back in the time series to get acceptable significance and do the analysis. Check what the trend is with sufficient confidence and let everyone know what it is.
I will give you two hints. It is longer than 17 years and the result will show you the Earth is warming.
============================================================
I think Moe may have confused me with D.B. Stealey.
(BTW I’m honored.)
Mint Toe-Paste not only makes the experience of tasting one’s toes less unpleasant but also serves as a lubricant to make it easier to extract one’s foot from one’s mouth. The only thing that defeats it is pride.
Moe,
Since you want a longer trend than 16 years [and I agree], here is a very long term trend chart.
That chart has one very noticeable feature: the long term global temperature trend rises along the same trend line, whether CO2 is low or high. Despite the recent increase in CO2, there is no acceleration in global warming. [In fact, global warming has stalled for the past 10+ years.]
That tells us unequivocally that rising CO2 does not cause any measurable global warming.
You write:
“Check what the trend is with sufficient confidence and let everyone know what it is.”
I have shown the long term [natural] global warming trend. It shows that the ≈40% rise in CO2 has not caused global warming to increase. At all. Either you can free your mind from the pervasive, anti-science AGW propaganda, or you can’t. But facts are facts, and the facts show conclusively that CO2 has had no measurable effect on global temperature.
Moe:
At January 28, 2013 at 1:12 pm you begin your ridiculous post saying
Say what!?
1.
I did NOT merely merely assert that statistical significance is a function of the variance of the data and NOT of the length of analyses time. I explained it but you have been incapable of understanding the elementary statistical procedure which I provided for you.
2.
I did NOT say “statistical significance cannot be established with sufficient confidence in 16 years”. I REPEATEDLY DENIED IT. This is merely another in your series of lies.
3.
Only you have made the ignorant and silly error of claiming that the length of time (not the r^2 statistic) defines the validity of a determined trend. Indeed, I doubt you did as I advised and looked up its equation.
I advised you to stop wasting time on the computer in your bedroom and to spend the time in the maths. lessons at school where you should be. Clearly, you have rejected that advice.
Read the explanation of basic regression analysis which I provided for you.
Long before there were “stats packages” which you say I should obtain (but have), I was using a cylindrical slide rule to conduct regression analyses in my lab. One needs to understand what one is doing and how to do it in order to obtain a regression that way. Indeed, one needs to be able to assess the likely result to within an order of magnitude because a slide rule doesn’t provide the decimal point. And you say I need to press a button to get a result! Yes, I can do that, but I know what the program is doing, and you keep proclaiming you don’t.
Stop wasting your time stating falsehoods here. I am sure the pocket money you get paid for trolling is useful, but the education you would get in school would be much more useful to you for your adult life.
And I again tell you that your superstitious belief that “The Earth is warming” is nonsense. There has been no discernible trend (at 95% confidence) in global temperature for 16+ years so if the Earth is warming or cooling it is not possible to know it.
Richard
DBStealey says:That tells us unequivocally that rising CO2 does not cause any measurable global warming. Unfortunately you must not have heard of the greenhouse effect. You are confused in thinking that CO2 is the only reason the earth should heat. There many influences as you know because each year is not a replica of the previous. The CO2 effect is overlaying these other influences causing the trend that you have correctly identified as hearing.
Oh dear Richard I think I have identified why you are confused. The data is a function of time (hint the temperature series is a TIME series). Therefore going back in TIME will give you statistical significance that you will be confident with. I am surprised that you are continuing to deny this, but if you must show you ignorance of stats that is ok by me. (Where on earth did you study this? I think your cylindrical (I think you mean circular) slide rule must have let you down.)
I am afraid you have said you cannot get statistical significance in 16 years as in ‘there has been NO STATISTICALLY SIGNIFICANT warming in the last 16 years’. Are you backing away from this statement?
As for your statement: And I again tell you that your superstitious belief that “The Earth is warming” is nonsense. ‘ I think you had better have a look at DBStealey’s post at 3:21 where he says:the long term global temperature trend rises along the same trend line,’ Once again your credibility is taking a beating.
Moe,
You are hand-waving. There is no measurable, testable evidence of global warming from CO2. None.
You cannot measure any global ‘greenhouse effect’. That is an assertion on your part, with no measurable basis in fact. It is an assumption, which may have some validity — or not. So if you believe it is valid, post the amount of the ‘greenhouse effect’ that results from X increase in CO2.
You can’t. Neither can anyone else. And if something cannot be measured, it is no more than a conjecture; an opinion.
Your comment is nothing but hand-waving. You write of a putative ‘CO2 effect’ without being able to produce any verifiable, testable scientific evidence or measurements.
If you actually believe that “the CO2 effect is overlaying these other influences causing the trend…”, then what you are doing is repeating scientifically baseless alarmist propaganda instead of thinking for yourself. “Overlaying” is just another baseless assertion. That nonsense may get traction at RealClimate or some other pseudo-science blog, but it fails here at the internet’s “Best Science” site.
There is no verifiable, testable, measurable “CO2 effect”. None. You are like a guy in his bed in a dark room, totally convinced that there is a black cat under your bed. You can almost hear it breathing. But when you get up and turn on the light… there is no cat. And there never was.
Wake me when you can produce any measurable, falsifiable, testable scientific evidence quantifying a “CO2 effect”. Unti then, you are just hand-waving.
Oh dear, DB do you dispute that there is a greenhouse effect?
When Richard comes back with his analysis and declares the Earth is warming you can then come back with an explanation as to why it is warming.
Moe,
As usual, you have the scientific method exactly backward. I don’t “dispute” a putative greenhouse effect, I only ask you to prove that it exists, and that it is measurable. The onus is on the one making the assertion, and your belief in a greenhouse effect is only an assertion. You have no measurable evidence to support your belief.
So either provide empirical, testable, falsifiable scientific evidence verifying a global “greenhouse effect”… or admit that you are operating on a belief system.
The onus is on you, not on scientific skeptics. You are the one making the assertion. Get it now?
DB, you say:’I don’t “dispute” a putative greenhouse effect, I only ask you to prove that it exists, ‘. If we are in agreement, why do you want me to prove it to you. Seems a little redundant.
Where is Richard with his analysis, I want him to eat humble pie and apologise to the readers for wasting their time. It is amazing, how people of low integrity, run off with their tail between their legs when they are found out rather than admit their error and move on.
Yo, Moe,
I don’t know why you are so fixated on another commentator that you feel you must ask me about him. Are you insecure, or what? FYI, I don’t speak for anyone but myself.
Or, maybe it is just a distracting tactic because I challenged you to provide scientific evidence, in the form of verifiable measurements, showing that your greenhouse effect exists?
You made the assertion that a greenhouse effect exists. Now prove it — or admit that it is a conjecture.
DB, you don’t dispute it so there is no need to prove it. I really can see your point. You can just as easily look at the reasons you used not to dispute it as to me giving them to you.
As for asking about Richard, that was more a note to self. I notice when people are shown up on this blog, they slink away to stop drawing attention to their errors.
Just to clarify, Moe is pronounced like Zoe not like toe, but I still appreciate your Yo Moe, it is good. (I guess that should be Yo Mo).
Moe:
In your post at January 28, 2013 at 9:29 pm you say
A note to self is not posted on the world’s most popular science blog.
A note to deceive is posted in such a manner.
I have repeatedly pointed out your several errors but you have not admitted any of them and you have slunk away from admitting your many mistakes.
The most notable of your silly and mistaken notions is that the statistical significance of a trend in a time series is defined by the length of the data set. As I have repeatedly stated – and provided a detailed explanation – the statistical significance is a function of the variance of the data and is determined by the r^2 statistic.
You have refused to admit your silly and ignorant error. Instead, you keep repeating it!
An apology for your behaviour here would be appropriate before Mummy tucks you up for the night.
Richard