Submitted by Dr. Clive Best
The first IPCC report in 1990 chaired by Prof. Houghton made a prediction for a rise in global temperatures of 1.1 degrees C from 1990 until 2030. This prediction can now be compared with the actual data as measured up to now (May 2011).
Figure2: Comparison of yearly HadCru & UAH data with IPCC 1990 predictions
These results have been derived as described below. You can see the results here
http://clivebest.com/blog/?p=2208
regards
Dr Clive Best
Mr. Mosher:
I have suggested to Dr. Best that he update his charts with the appropriate scenario. Perhaps we will get a clearer picture of the situation.
RR Kampen says:
June 10, 2011 at 2:22 am
“As stated last year, the Scenario B in that paper is running a little high compared with the actual forcings growth (by about 10%) (and high compared to A1B), and the old GISS model had a climate sensitivity that was a little higher (4.2ºC for a doubling of CO2) than the best estimate (~3ºC).”
You are committing the same egregious error as Mosher, using some of the conditions in your conditional as excuses for incorrect predictions. If you want to play that game, admit that you cannot make any predictions whatsoever and, therefore, you are not practicing science or anything resembling. You are locked into the closet of the Gaia Modelers, contemporary alchemists.
steven mosher says:
June 10, 2011 at 10:28 am
“So, You have one scenario that projects 400ppm by 2010 and another that projects 390. Since we cannot control this experimental variable, we have to make projections. what happens at 360? 390? 400? 500?”
Well, that helps. Since you do not use scientific method or its terminology, let me rephrase the matter another way? What rational basis have you explicated for selecting the projections? What are the critical principles embodied in this rational basis? How are your results use to criticize you choice of projections and your rational basis? Can you relate any of this to predictions about the real world?
You have pretty much said that you do not practice science. Now I am wondering whether you have any rational procedures at all.
Steve, you’re way over stating your case.
What is the emissions rate in the lower projection curve?
The 100 year stuff is nonsense. There is no case for large and long lags. If you said “it’s plausible”, I wouldn’t have embarrassed myself by laughing out loud here at the bar.
Plausible in the sense that you can’t prove a negative, anything is possible, and nothing is certain.
You do great work and are very smart. You’re making yourself look like a moron here though.
steven mosher says:
June 10, 2011 at 11:31 am
You do not have a clue what this means. You have just expanded your equivocation on the logical word ‘conditional’ by using the word ‘scenario’ as meaning the same as ‘conditional’.
Sir, the simple truth is that if you use conditional statements about the future, whether to make predictions, specify scenarios, or whatever, and you include one condition that describes something that is unpredictable, as you say volcanic eruptions are, then you have made your conditional statement worthless. Now it predicts nothing. It fails to specify any scenario whatsoever. Why are you making your conditional statements worthless? Because you want that clause in there about volcanoes to use as an excuse when your conditional statement proves to be false.
And I see that the grand excuse of “uncertainty” has cropped up again. Great. Do you want to talk about Pascal’s Wager? You must because he invented the decision matrices used by the folks who obsessively study decision under uncertainty. However, none of them claim to be doing empirical science.
Peter says:
June 9, 2011 at 3:41 pm With respect, doesn’t the failure of these predictions also point to a failure of the notion that climate is somehow deterministic? Shouldn’t we revisit the view that climate is essentially chaotic and, therefor, unpredictable?
Absolutely. This is the elephant in the room. You can not predict a chaotic system and even the modellers agree that the system is chaotic. So…………there is not only no point to the exercise even if you DID understand every variable, there is even less point when you actually have no idea about the major variables in the system.
Also, did you know that many of the models show cooling trends? These are dismissed as outliers, (of course, without any statistical testing to identify them….) and the data is cherry picked. A prime example of poor science
Uncertainty in predictions of the climate response to rising levels of greenhouse gases’, by Stainforth et al. (2005)
Link to abstract http://www.nature.com/nature/journal/v433/n7024/full/nature03301.html
You will be able to find free versions on the net, I can’t link to them from here without invoking the Gods of (C).
A classic example of manipulation. My favourite quote, page 2 right hand column has to be….
“This lack of an observational constraint, combined with the sensitivity of the results to the way in which parameters are perturbed, means that we cannot provide an objective probability density function for simulated climate sensitivity.
Nevertheless, our results demonstrate the wide range of behaviour possible within a GCM and show that high sensitivities cannot yet be neglected as they were in the headline uncertainty ranges of the IPCC Third Assessment Report” (Stainforth et al. 2005).
For the non climate scientist among us, they admit that there is NO probabillity density function, i.e., that there is no statistical merit to the data sets produced by the model, (based on the Hadley CRU….). BUT on that basis, they then go on to reject the cooling data that the model produced on the basis that they are statistical outliers! That’s right, they state there is no determinable data populations then reject outliers. Quite fantastic. And, Emperors New Clothes they get away with it.
If the model was correct, it may have predicted the cooling as an emergent property of the system from the given starting point which would have been a very interesting result, because if there is one thing that really worries me, it would be global cooling. The possibility is not even mentioned. Observational bias, systematic bias and a very poor understanding of statistical testing, (undergrad engineering math!).
So we know that the system is chaotic. We know it is non linear. Therefore, it can not be modelled. We know nothing about how the feedbacks which are supposed to generate a positive forcing for CO2. We do not understand the role of water vapour, 80% of GHG forcing, we don’t understand ocean heat transport or even if clouds give a positive forcing (insolation) or negative forcing (albedo) But we model this to produce temp predictions to a supposed 95% CI over 100 years with models unable to approximate what we do know, (simple single layer ocean is typical) and make decisions on the data using statistics on data which has no objective identified probability density function?
This is the junk science on which all this is based. It really is quite shocking.
Mosher: “What does that do to your observations?. It cools them with a forcing that was not assumed in the scenario.”
Okay, so we are asking governments to spend trillions of dollars based upon idiotic scenarios that don’t have enough information to predict what they pretend to be predicting. Also, as you said, the effect of a volcano is short, so your idea that it is the reason for the failure of the trend is dubious. Furthermore, a cooling event at the beginning of a time period will serve to give you a stronger upward trend overall.
Mosher: “1. Its hard but not impossible.”
If it were possible and if they had done what was possible, then you wouldn’t be making all of these excuses for why they got it wrong.
Mosher: “Hard things done wrong are.”
Same point. They did it wrong, so it was a waste of time. But maybe you want to say that the earth got it wrong by producing a volcano. Or maybe you want to say that people got it wrong by not producing enough CO2 – even though nothing has been done to mitigate CO2 and “buissness as usual” is what we have in fact been doing.
Mosher: “3. We make predictions all the time under uncertain information.”
So do people with chicken entrails.
Mosher: “5. It depends where the volcano effect occurs, how large it is, etc. having one at the beginning or end of the period is especially problematic.”
Having a big one at the beginning would be like having a La Nina at the beginning. It would increase the slope of your trend.
Mosher: “This takes more care than Clive used.”
You are getting confused Mosher, Clive wasn’t making the predictions, he was evaluating them. It’s not his job to find excuses for why they got it wrong. The IPCC gave themselves space. They had a best, a high and a low. The contingencies should have been included in that range. But all three were wrong. With regard to the primary driver that the IPCC uses to predict further warming, CO2, there has been no change in global behavior. In fact, the rapid expansion of power plants in China and cars in China was likely unforseen. So if you are trying to say that we had less CO2 in 2010 than the IPCC had predicted, you are only pointing out again that their predictions are worthless.
Mosher: “Further, the real effect of C02 takes much longer than 20 years to develop.”
And that lag in development time should have been in the IPCC chart – so again, you are making excuses for them. Also, remember that we are not talking about a CO2 pulse that happened in 1990; we are talking about CO2 that has been added to the atmosphere since long before 1990.
It’s pretty simple Mosher, the IPCC wants to influence governments with their predictions, or projections, (call them what you like) about the future. Up to this point they have shown no skill in anything other than making excuses for why they got it wrong.
Now, if you just want to make the point that this is hard stuff and we should be working on it so that we can get better, then I will agree. But, for christ sake, don’t be trying to stampede the world with uncertainties and over hyped predictions that always turn out to be bigger than the reality.
Doesn’t look like to me the low prediction matches with the data either even if adjusted to the 1990 value. Scenario C shows about a 0.4C rise from 1990-2011. UAH shows about a 0.15C rise from 1990-2011 and Hadcrut about 0.2. The data matches the Scenario C slope up to about 2004 but after that.. the slope turns negative in the data. Sorry, but seems the data is indicating a climate sensitivity factor much less than expected. In additon, since other factors can also have a large effect on the temperature outcome, that would indicate the factor for CO2 cannot be significantly more important than the others.
Well, there’s no need for multimillion dollar computers and self-opinionated incompetent modellers and their sycophantic defenders. These kind of predictions, projections or whatever weasel words one wants to call them, can be made as good as these models with tea leaves or reading the flight of the swallows, as soothsayers used to do in ancient days. What a steaming pile of bullcrap and even more lousy excuses defending that crap! More then the modellers who did the work in 1990, the staunch defenders of the indefensible seem to have no idea or qualms about scientific method or rational thought.
Tilo Reber:
Thank you for taking the time to respond to my comment. I expand upon the theme of this comment at http://judithcurry.com/2011/02/15/the-principles-of-reasoning-part-iii-logic-and-climatology/ . In brief, IPCC climatology is a pseudo-science that may appear to be a science through the ambiguity of reference by climatology’s language to the associated ideas. An element of this ambiguity is produced by confusion of the disparate ideas that are referenced by the words “prediction” and “projection.” As the predictions of a model make a falsifiable claim, they support the scientific method of inquiry. As the projections of a model make no falsifiable claim, they fail to support this method. Contrary to the understanding of many of the readers of Wattsupwiththat, the major shortcoming of IPCC climatology is not that its theory of anthropogenic global warming is wrong but rather that it cannot be determined to be wrong.
You can not predict a chaotic system
As a statement of fact this is incorrect. You cannot predict the short-term behaviour of a chaotic system with any certainty, but you certainly can over the long term make useful predictions for certain chaotic systems.
We do it all the time. Weather is chaotic. Yet we plant corn at a certain time, in the knowledge that the seasonal weather will very likely be suitable, even though we cannot even predict tomorrow’s weather with certainty. That is because weather is less chaotic than you make out.
While climate may be chaotic (and I note you offer no proof that it is) it clearly operates within certain boundaries. If we move the lines of those boundaries (CO2, sun, cloud cover, volcanoes) then we may be able to predict a change in the averages observed even without clearly understanding the system properly.
Gasses are entirely chaotic on a molecular level. Yet I know that if I double the volume that pressure will fall by almost exactly half.
Tilo Reber says:
June 10, 2011 at 8:40 am
Leif: “I have no particular opinion about his ‘work’.”
Thanks. I understand you better now.
now, do YOU have an opinion about his work? and what PRECISELY do you base that opinion on? Just so we can understand you better….
Looks like a Grand Canyon developing between IPCC projections (prophecies of a pseudo-scientific-religious tangent) and reality. To take a wider view of how AGW Alarmism has evolved, think of endless levels of play on a Sony Playstation as government coins are fed into the Arcade Simulation. Such lavishly funded imaginations.
CO2 was first blamed for Climate Change back in the 1930’s for the Dust Bowl years, then Global Cooling in the 1970’s Ice Age Scare, and now this.
The wide angle reality of Climate Change is a mild warming out of the LIA.
That’s a mild warming corresponding to innumberable up-shots in the Vostok and EPICA cores. The LIA was nothing more than a mild down-tick in the same records, there being hundreds to thousands of them.
The Dust Bowl, Ice Age Scare and the AGW previously imagined wouldn’t make a pixel on the Vostok record.
Geologic Time makes man look puny.
I don’t know who said you cannot predict a chaotic system but Mooloo above gave a good answer.
From Wikipedia
So there are boundaries, and the three body problem and deterministic chaos is discussed well by John Gribbin in his book Deep Simplicity, he is a good journalist and it is worth reading.
It is a fascinating subject that, unlike many other mathematical branches, lends itself to beautiful diagrams like this plot of the Lorenz attractor.
Mooloo says:
June 10, 2011 at 9:52 pm
You can not predict a chaotic system
‘Gasses are entirely chaotic on a molecular level. Yet I know that if I double the volume that pressure will fall by almost exactly half.’
If I grow an elm tree in the lab can I predict what it will look like in 50 years time? Is the climate as complex as an elm tree? If I plant the elm tree outside the lab, how much will that change my prediction?
If, in 50 years time, I am accurate in my prediction, how will I know if I am right?
Leif: “do YOU have an opinion about his work?”
Yes, Leif, I do. And I stated it very explicitly in a recent thread where you took part.
http://wattsupwiththat.com/2011/05/30/kill-it-with-fire/
Leif: “and what PRECISELY do you base that opinion on?”
Given in the thread.
Leif: “Just so we can understand you better….”
No problem. I’m not pretending to straddle any fences.
The thread involves a lot of information exchange, references to other links that discuss the upside down Tiljader proxies, and several posts on my part, so I won’t try to reproduce that for you here. But it’s available to you if you really want to know.
Antony,
Interestingly did the BBC website produce data and a story claiming the opposite.
see: http://www.bbc.co.uk/news/science-environment-13719510
They use HadCRUT3 data which show still increasing temperature trend. Any views on this?
Rein
Phil Jones has rendered this post moot with a quick flash of his hands. http://www.bbc.co.uk/news/science-environment-13719510
And with this, he confirms the prior IPCC predictions as being correct—this year, but not last year.
And the best line of the story is aptly the last, “And a new initiative to construct a global temperature record, based at Stanford University in California whose funders include “climate sceptical” organisations, has reached early conclusions that match established records closely.”
So you see, move along, nothing to see here but alarmism.
FACT
Global mean temperature pattern is cyclic with a slight overall warming of 0.6 deg C per century: http://bit.ly/cO94in
Tilo Reber says:
June 11, 2011 at 12:46 am
I’m not pretending to straddle any fences.
Neither am I. What I’m saying is that for me to form my opinion I would have to look at the data myself. Otherwise I would, like you, just be parroting other people’s opinion.
Leif Svalgaard says:
June 11, 2011 at 7:50 am
Tilo Reber says:
June 11, 2011 at 12:46 am
“I’m not pretending to straddle any fences.
Neither am I. What I’m saying is that for me to form my opinion I would have to look at the data myself. Otherwise I would, like you, just be parroting other people’s opinion.”
Isn’t it frustrating to talk to Leif? That’s because he is a scientist. Science never promises a rose garden.
Keith Minto says:
“June 10, 2011 at 11:25 pm I don’t know who said you cannot predict a chaotic system but Mooloo above gave a good answer.
From Wikipedia
An early proponent of chaos theory was Henri Poincaré. In the 1880s, while studying the three-body problem, he found that there can be orbits which are nonperiodic, and yet not forever increasing nor approaching a fixed point.
So there are boundaries, and the three body problem and deterministic chaos is discussed well by John Gribbin in his book Deep Simplicity, he is a good journalist and it is worth reading.
It is a fascinating subject that, unlike many other mathematical branches, lends itself to beautiful diagrams like this plot of the Lorenz attractor.”
With the greatest of respect, I don’t think you understand the problem.
There is no probablility density function.
Therefore, the model output is random junk.
By definition, you can not predict a set of variables in a radom system. Quoting bits of Wikipedia which you half understand won’t change that.
Read the paper I posted, you tell me how you can obtain statistical data to within a 95% CI assuming the data is normally distributed when it is by the authors admission, randon, then come back to me.
Leif: “What I’m saying is that for me to form my opinion I would have to look at the data myself.
Otherwise I would, like you, just be parroting other people’s opinion.”
Based on the opinion you gave here, you didn’t have to look too hard before you parroted Mosh’s opinion.
Leif: “If as Mosh points out, the two temperature curves were shifted to go through the 1900 point, the low IPCC prediction looks right on.”
That level of “looking at the data” was something that you could have done very easily on the “Kill It With Fire” thread and on any number of McIntyre threads.
Leif wants to look at Mann’s whole data, methods and code to make an opinion. Somebody forgot to tell him that this was the issue all the time with MBH, M&M, Hokey Stick and the team’s shenanighans etc.,, that Mann never showed his data and work and hid it. That was one of they key reasons for birth of Climate Audit. The evidence of what the team did to hide the data and the tricks were employed were all over Climategate mails. I guess Leif needs a refresher course on these issues, not / sarc..
But when it comes to Climate Models, he’ll accept Mosh or any other believer’s words and does not need to see any data, methods etc. Yes, very scientific indeed.
Tilo Reber says:
June 11, 2011 at 8:54 am
Based on the opinion you gave here, you didn’t have to look too hard before you parroted Mosh’s opinion.
Leif: “If as Mosh points out, the two temperature curves were shifted to go through the 1900 point, the low IPCC prediction looks right on.”
I looked at the data [the curves] and saw for myself what the effect of shifting the curves would be, so you can spare us your snide remarks. My opinion happened to agree with Mosh’s, but did not depend on his.
That level of “looking at the data” was something that you could have done very easily on the “Kill It With Fire” thread and on any number of McIntyre threads.
But I didn’t bother. I have better things to do than look at Mann’s stuff [perhaps you have not]. When asked, I would happily defer to all those experts you mention, but it would not be my opinion, because [as I said] I do not have an independent opinion on this.