The Consistent-With Game: On Climate Models and the Scientific Method
February 13th, 2008Posted by: Roger Pielke, Jr.
I have been intrigued by the frequent postings over at Real Climate in defense of the predictive ability of climate models. The subtext of course is political – specifically that criticisms of climate models are an unwarranted basis for criticizing climate policies that are justified or defended in terms of the results of climate models. But this defensive stance risks turning climate modeling from a scientific endeavor to a pseudo-scientific exercise in the politics of climate change.
In a post now up, Real Climate explains that cooling of Antarctica is consistent with the predictions of climate models:
A cold Antarctica and Southern Ocean do not contradict our models of global warming.
And we have learned from Real Climate that all possible temperature trends of 8 years in length are consistent with climate models, so too are just about any possible observed temperature trends in the tropics, so too is a broad range of behavior of mid-latitude storms, as is the behavior of tropical sea surface temperatures, so too is a wide range of behaviors of the tropical climate, including ENSO events, and the list goes on.
In fact, there are an infinite number of things that are not inconsistent with the predictions of climate models (or if you prefer, conditional projections). This is one reason why a central element of the scientific method focuses on the falsifiability of hypotheses. According to Wikipedia (emphasis added):
Falsifiability (or refutability or testability) is the logical possibility that an assertion can be shown false by an observation or a physical experiment. That something is “falsifiable” does not mean it is false; rather, it means that it is capable of being criticized by observational reports. Falsifiability is an important concept in science and the philosophy of science. Some philosophers and scientists, most notably Karl Popper, have asserted that a hypothesis, proposition or theory is scientific only if it is falsifiable.
Are climate models falsifiable?
I am not sure. Over at Real Climate I asked the following question on its current thread:
There are a vast number of behaviors of the climate system that are consistent with climate model predictions, along the lines of your conclusion:
“A cold Antarctica and Southern Ocean do not contradict our models of global warming.”
I have asked many times and never received an answer here: What behavior of the climate system would contradict models of global warming? Specifically what behavior of what variables over what time scales? This should be a simple question to answer.
Thanks!
As often is the case, Real Climate lets their commenters provide the easy answers to difficult questions. Here are a few choice responses that Real Climate viewed as contributing to the scientific discussion:
If Pielke wants to contribute constructively to this area of science, he should become a climate modeler himself and discuss such questions in the scientific literature. Otherwise, unless he can present some strong reason for doubting the competence or objectivity of people who do such work, he should listen to people who do work in the area.
. . .
Roger, your question is rather broad and vague. What aspect of the science are you seeking to falsify? See, that is precisely the problem when you have a theory that draws support from such a broad range of phenomena and studies as does the current theory of climate. It is rather like saying, “How would we falsify the theory of evolution?” When a theory has made many predictions and explained many diverse phenomena, it is quite difficult to falsify as a whole. You may be able to look at pieces of it and add to the understanding. Climate science is quite a mature field; future revolutions are quite unlikely. Changes will come but will likely be incremental. It is very hard to envision a development that would significantly alter our understanding of greenhouse forcing unless our whole understanding of climate is radically wrong, and that seems unlikely.
The good news is that there are a range of serious scholars working on the predictive skill of climate models. And there are some folks, myself included, who think that climate models are largely of exploratory or heuristic value, rather than predictive (or consolidative). (And perhaps a post on why this distinction is of crucial importantce may be a good idea here.) But you won’t hear about them at Real Climate.
Once you start playing the “consistent with” or “not inconsistent with” game, you have firmly placed yourself into a Popperian view of models as hypotheses to be falsified. And out of fear that legitimate efforts at falsifiability will be used as ammunition by skeptics (and make no mistake, they will) in the politics of climate change, issues of falsification are simply ignored or avoided. A defensive posture is adopted instead. And as Naomi Oreskes and colleagues have observed, this is a good way to mislead with models.
One of the risks of playing the politics game through science is that you risk turning your science – or at least impressions of it – into pseudo-science. If policy makers and the public begin to believe that climate models are truth machines — i.e., nothing that has been, will be, or could be observed could possibly contradict what they say — then a loss of credibility is sure to follow at some point when experience shows them not to be (and they are not). This doesn’t mean that humans don’t affect the climate or that we shouldn’t be taking aggressive action, only that accurate prediction of the future is really difficult. (For the new reader I am an advocate for strong action on both adaptation and mitigation, despite what you might read in the comments at RC.)
So beware the “consistent with” game being played with climate models by activist scientists, it is every bit as misleading as the worst arguments offered by climate skeptics and a distraction from the challenge of effective policy making on climate change.
For Further Reading:
Pielke, Jr., R.A., 2003: The role of models in prediction for decision, Chapter 7, pp. 113-137 in C. Canham and W. Lauenroth (eds.), Understanding Ecosystems: The Role of Quantitative Models in Observations, Synthesis, and Prediction, Princeton University Press, Princeton, N.J. (PDF)
Sarewitz, D., R.A. Pielke, Jr., and R. Byerly, Jr., (eds.) 2000: Prediction: Science, decision making and the future of nature, Island Press, Washington, DC.
February 13th, 2008 at 11:42 am
I profoundlly disagree. For several reasons.
(1) Modern science is closer to formal Bayesian reasoning than Sir Karl’s falisfyabilty criterion, and I am under the impression than he moved closer to this position in has latter days. In Bayesian reasoning, various hypotheses are found to be more or less confirmed by the evidence. For two competeing hypotheses this degree of confiramtion can even be measured, from the evidence, via the naive Bayes factor method.
I’ve been following Real Climate for over two years and, being a retired computer scientist, feel I know have a decent grasp of what climate modeling is about. By informal Bayesian reasoning the climate models have demonstrated quite an impressive record (but far from perfect) in predictions made and agreement with the actual climate response.
(2) Various knowledgeable amateurs answered, in quite a bit of detail, the specific question you posed on the Real Climate ‘Antarctric’ thread. You neither thanked them for their useful responses, individually or as a group, on Real Climate, nor quoted anything but those most self-serving to you here.
(3) Policy in many areas is informed by a combination of experiment or observation and some computer modeling. My experience in this region leads me to conclude that the climate models are vastly better (at what they do) than models used regulary for regional resource management decisions. Methinks thou doeth ask for too much!
(4) ‘Consistent with’ is a weak form of saying “the weight of the evidence favors hypothesis H over hypothesis K, disconfirming K by X decibans.” If there actually were a hypothesis K than such a study could be performed, but I should hope not at taxpayer expense: the existing climate models are indeed quite good. On Real Climate there is even an earlier thread regarding attempts to compare many of the models.
But worse, your opinion piece asserts than ‘nothing could be inconsistent’ with climate models. This is clearly nonsense, even if you carefully hedged it in a condition form.
(5) I am forced to conclude you need not read all of the answers to your inquiry before composing this piece; one which certainly appears, to put the matter kindly, to have been done in an off-moment. I am disappointed.
February 13th, 2008 at 11:45 am
I profoundlly disagree. For several reasons.
(1) Modern science is closer to formal Bayesian reasoning than Sir Karl’s falisfyabilty criterion, and I am under the impression than he moved closer to this position in has latter days. In Bayesian reasoning, various hypotheses are found to be more or less confirmed by the evidence. For two competeing hypotheses this degree of confiramtion can even be measured, from the evidence, via the naive Bayes factor method.
I’ve been following Real Climate for over two years and, being a retired computer scientist, feel I know have a decent grasp of what climate modeling is about. By informal Bayesian reasoning the climate models have demonstrated quite an impressive record (but far from perfect) in predictions made and agreement with the actual climate response.
(2) Various knowledgeable amateurs answered, in quite a bit of detail, the specific question you posed on the Real Climate ‘Antarctric’ thread. You neither thanked them for their useful responses, individually or as a group, on Real Climate, nor quoted anything but those most self-serving to you here.
(3) Policy in many areas is informed by a combination of experiment or observation and some computer modeling. My experience in this region leads me to conclude that the climate models are vastly better (at what they do) than models used regulary for regional resource management decisions. Methinks thou doeth ask for too much!
(4) ‘Consistent with’ is a weak form of saying “the weight of the evidence favors hypothesis H over hypothesis K, disconfirming K by X decibans.” If there actually were a hypothesis K than such a study could be performed, but I should hope not at taxpayer expense: the existing climate models are indeed quite good. On Real Climate there is even an earlier thread regarding attempts to compare many of the models.
But worse, your opinion piece asserts than ‘nothing could be inconsistent’ with climate models. This is clearly nonsense, even if you carefully hedged it in a condition form.
(5) I am forced to conclude you need not read all of the answers to your inquiry before composing this piece; one which certainly appears, to put the matter kindly, to have been done in an off-moment. I am disappointed.
February 13th, 2008 at 3:53 pm
Received by email:
“Dear Dr. Pielke:
I tried to post this response to “The Consistent-With Game: On Climate Models and the Scientific Method” at your website, but for some reason my typekey account does not sign me in there.
Dr. Pielke-
I am surprised to find such a disingenuous post at this website. Should you not have at least mentioned those specific (and relevant) replies to your question posted by commentators at Realclimate.org? I am also surprised that, after examining several decades of climate research, you are “not sure” if climate models are falsifiable. Do you really believe that “issues of falsification are simply ignored or avoided” by working stiff climate researchers? Any attendee at a scientific conference on the subject (or, for that matter, any reader of the journals) would be aware of the many dimensions of the subject that are being tested all the time.
So I must be mistaking the purpose of your question. Perhaps it is simply that you are disgruntled because the Real Climate people do not support your view that “climate models are largely of exploratory or heuristic value, rather than predictive”? Why not just say that, and then defend the proposition?
Best regards,
Pat Cassen”
February 14th, 2008 at 12:50 am
Comments sent in by email from John Christy, UAH:
“Roger:
Our paper on tropical upper air trends sought to test climate model output in a way you describe today. We asked this question: If a climate
model’s tropical surface temperature trend since 1979 matched observations, would the model’s tropospheric trend match observations?”
This is a test to see if a model has realistic vertical heat transport processes and thus whether it could be trusted to properly accumulate heat
over decades of execution. This is also a way test the region of the atmosphere that has (in models) the largest and clearest signal of
greenhouse warming (tropical troposphere).
We gathered 67 model runs to have an extremely robust test statistic (more than used in the IPCC or CCSP).
As it turns out, the mean surface trend of the model runs indeed matched the surface trend since 1979. Great … we then could proceed with our
test. Well, the models’ mean tropospheric trend did not match (much warmer than the observations.) Thus what the models predicted for post 1979
tropical tropospheric warming was statistically significantly much more than observed. We concluded that the models’ vertical heat transfer
processes were not realistic and that they allow too much heat to accumulate in the troposphere relative to the real world.
RealClimate did not understand the test and twisted themselves around trying to defend model output. They tried to compare the observations
with the entire RANGE of model outputs which included some models with virtually no surface trend since 1979 (and thus little tropospheric
trend). We compared apples with apples, i.e. models and observations which showed very similar surface trends and then checked the tropospheric
trends.
I thought the test was simple, clean and a good example of falsification of a model hypothesis regarding the vertical structure of heating. I
guess it was too simple. As we say (see page with section 4.2, paragraph on right starting with “Additionally …”) … Our experimental design is
more rigorous. We are comparing the best possible estimate of model-produced upper-air trends that are consistent with the magnitude of the observed surface trend.”
February 14th, 2008 at 1:52 am
David B. Benson
You are suggesting that the climate models are sufficient. Your entire argument stems from your first point of using Bayesian reasoning. There is a problem with that. Assuming the modelers are using Bayes’ Theorem, then the modellers would be able to assign qualitative probabilities to each possible outcome. When a broad range of similar but opposing behaviours are said to not be inconsistent, that is not the full story. The probabilities are needed as well. For example, the IPCC reported that 5 models all predicted that warming due to anthropogenic increases in greenhouse gases would have a specific effect on the temperature trends in the tropics’ troposphere. That must have had a fairly high probability for that to be included in the IPCC report.
Unfortunately, the tropics did not follow the model predicted trend. The temperature trend in the tropics did not conform to the models’ prediction. To suggest that this observation is not inconsistent is fine, but then, using Bayes’ Theorem, you must then indicate what the probability level is of the observed scenario.
“Not inconsistent with” does not mean “likely.”
Restating Dr. Roger Pielke, Jr.’s question using your terms, at what probability level would an observation have to be to disprove the models?
February 14th, 2008 at 12:40 pm
Whoa, Mr. Benson! I need some supporting evidence for your claim “my experience in this region leads me to conclude that the climate models are vastly better (at what they do) than models used regulary for regional resource management decisions. ” I happen to be a scientist involved in regional and local resource management decisions. There are vast numbers of models, sedimentation models, air quality models, biodiversity models, vegetation models, etc., used in resource management decisions. Are you saying the average of their individual accuracies of predictions is less than the average of climate change models? What are the models and accuracy estimates that you used to make this claim?
Or is this just a drive-by shot at another set of scientific disciplines?
Sharon Friedman
February 14th, 2008 at 3:01 pm
Hi Roger,
I’ll probably have more comments on this (bet you can hardly wait!), but for tonight, “And out of fear that legitimate efforts at falsifiability will be used as ammunition by skeptics (and make no mistake, they will)…”
Ummmm…yeah! Isn’t that what a real scientist should ***welcome***?
Do you notice any scientists who report on the evolution of humans (for example) worrying that their research “will be used as ammunition” by 6000-year-old earth creationists?
Of course not. Because they’re real scientists, practicing real science. That is, they’re trying to honestly answer a question. They’re not trying to provide ammunition to support a predetermined position.
“Climate science” (as practiced by the IPCC and the “climate change community”) has as a primary purpose to provide support for a predetermined position that greenhouses gas emissions–and especially CO2 emissions–must be dramatically reduced in the next few decades. (Or sooner.)
February 15th, 2008 at 7:03 am
Sharon Friedman (docpine) — I hope it is not a drive-by shot. The problem for the regional forecasts I briefly mentioned is that extreme events are rather rare and often not well monitored. I have some objections to the computer modeling techniques as well, but I suppose I would have similar objections to aspects of climate modeling. (I have no qualms with the various regional water resource models, which do well.)
jmrsudbury — Bayesian reasoning, properly done, never disproves anything. It only confirms or disconfirms various hypotheses to varying degrees.
Unfortunately, even the LLNL model-result comparison effort does not appear to yet be using formal Bayesian methods. It would help to do so. With regard to the non-prediction of tropical temperatures, my reaction is to state that those models must not have enough of the actual phsyics yet. So new models are built. Thus attempting to discuss confirmation and disconfirmation is often, when done informally, aiming at a moving target. Nonetheless, the climate models have made many successful predictions, tending to confirm that such at least are not bad. Greater use of formal Bayesian methods could make such statements (more) quantitative. One has to ask of the climate models, sufficient for what purposes? For making global predictions, these seem to be ‘not bad’ but as the example of tropical temperatures shows, for regional predictions, ‘not so good’.
Mark Bahner — The position regarding CO2 emmisions was certainly not predetermined, but rather grew out of the evidence (and modeling) gathered over decades. The original impetus was a combination of scientific curiosity together with convincing funders (largely US congress) that reducing the uncertainty was worth the cost. It is only in this century that sufficient evidence has been gathered so that no scientific body opposed the IPCC conclusions and the most prestigious support it.
To conclude, my apologies for the double post. This occurred because the site management software stated that my post (both times) failed. It obviously lied. But my major point stands: no falsifiablity, just degrees of confirmation and dixconfirmation. Think Bayesian!
February 15th, 2008 at 8:08 am
Thank you for admitting that the models, that have been evolving for the past couple of decades, are still not sufficient. That highlights Roger’s point about the “challenge of effective policy making on climate change.” It would be useful if more of the politicians knew how the models are still insufficient. Then again, since the scientists have been saying that “next year will be the warmest year on record” for a decade, they may have caught on that those scientists are constantly wrong.
You seem confused though in that you seem to think that the tropics is not a region. The question remains though. At what degree of confirmation will an outlying observation become a disconfirmation thus suggesting an alternative to the consensus when it comes to policy making?
Then again, this is all a moot point since, as you point out, formal Bayesian methods are not being used; therefore, Roger’s original question must then still be valid.
Oh, and one last request. Could you please tell me the degree of confirmation or disconfirmation behind your ‘could’ in your sentence ending with, “… could make such statements (more) quantitative.”
John M Reynolds
February 15th, 2008 at 9:33 am
John M Reynolds said ‘Could you please tell me the degree of confirmation or disconfirmation behind your ‘could’ in your sentence ending with, “… could make such statements (more) quantitative.”‘
This makes no sense unless specific hypotheses are formulated, say two, H and K, and some evidence E is collected. I assume you were just joking. Ha.
You also said ‘Roger’s original question must then still be valid.’ Why not? It was extensively answered by several commenters on the Real Climate thread about “cold Antarctica”. Most of these answers were highly responsive, pointing out ways in which the climate models agree with, and have even predicted, most of the real world data. The only two failures are Arctic responses in 2007 (a region) and the claimed prediction of too high temperatures in the tropics (another region). I’d say that as global models the performance is quite respectable (and no modeler will claim more than that).
But Dr. Pielke, Jr. chose to ignore (or not read) most of the responses before writing his ill-thought-out opinion piece (which begins this thread).
Finally, in this post, you said ‘… since the scientists have been saying that “next year will be the warmest year on record” for a decade, …’
I’ll not take your word for it. Cite peer-reviewed literature (if you can find any).
February 15th, 2008 at 1:47 pm
David Benson–
You write, “The position regarding CO2 emissions was certainly not predetermined, but rather grew out of the evidence (and modeling) gathered over decades.”
The preponderance of evidence does not support that position.
You further write, “It is only in this century that sufficient evidence has been gathered so that no scientific body opposed the IPCC conclusions and the most prestigious support it.”
The IPCC has not made any important scientific conclusions. Let’s look at the *fundamental* scientific question that the IPCC should be addressing: “What will happen if no government action is taken?”
Has the IPCC answered that basic question? No, they have not. Have they even *tried* to answer that basic question? No, the overwhelming preponderance of evidence is that they have not even tried to answer that question. Because if they had tried to answer that question, no scientific body in the world is so incompetent that they could not come up with at least a scientific estimate, given literally more than a decade and hundreds of millions of dollars.
Look at the IPCC Third Assessment Report (TAR) and Fourth Assessment Report (AR4) “projections.” Do they meet the most fundamental requirement for projections to be scientific? Are they falsifiable? No, they are not. I have asked the scientists at “Real Climate,” Kevin Trenberth and the editors of Nature at Nature’s “Climate Feedback” blog, and others to read the IPCC AR4, and label the following assertions as “true,” “false,” or “don’t know” regarding warming (in the 21st century):
1) The IPCC thinks there is more than a 0.1 percent chance of warming more than 6.4 deg C.
2) The IPCC thinks there is more than a 10 percent chance of warming more than 6.4 deg C.
3) The IPCC thinks there is more than a 50 percent chance of warming more than 6.4 deg C.
4) The IPCC thinks there is more than a 99 percent chance of warming more than 6.4 deg C.
5) The IPCC thinks there is more than a 0.1 percent chance of warming less 1.1 deg C.
6) The IPCC thinks there is more than a 10 percent chance of warming less than 1.1 deg C.
7) The IPCC thinks there is more than a 50 percent chance of warming less than 1.1 deg C.
The IPCC thinks there is more than a 99 percent chance of warming less than 1.1 deg C.
Of all the scientists I’ve ever asked those questions, only one has provided answers. (In his own roundabout way) James Annan admitted that the correct label for every single one of those 8 assertions is, “Don’t know.”
In other words, James Annan cannot tell, from reading the AR4, whether the IPCC thinks that it’s virtually certain the warming will be less than 1.1 deg C, or virtually certain that the warming will be more than 6.4 deg C. In short, the IPCC has come to absolutely no scientific conclusion (however tentative) on what should be the most important scientific question they should be answering.
So you can say that “prestigious scientific bodies” support the IPCC “conclusions.” But that assertion is simply not supported by the evidence. The IPCC hasn’t even made a conclusion (again, even a tentative one!) regarding what the warming will be in the 21st century, in the absence of action by governments.
It’s important to recognize that this is not a new scam (yes, scam!) that the IPCC and the “climate change community” are pulling. The Limits to Growth authors have been pulling the same scam for more than 30 years. It consists of:
1) Make a bunch of “projections” of widely varying futures, most of them unrealistic (or even flat out impossible), and
2) Don’t divulge that most of the projections are unrealistic.
In the absence of an honest assessment of probabilities, most readers simply assume that all scenarios are equally probable.
February 16th, 2008 at 3:09 am
Just confirming whether or not I have been banned here.
February 16th, 2008 at 4:38 am
Mark Bahner — You are simply blathering. Go read the actual statements of the NAS, AAAS, etc., to see that what I wrote is correct. The last hold-out was the petroleum geologists, who recently rescinded their contrarian position, and now have no organizational position at all. Note that what I wrote was constructed to include even them.
The evidence for the connection between global temperature and CO2 has been known for 150 years. Ever textbook on atmospheric physics explains the details. But easier for you, I opine, is
http://www.aip.org/history/climate/index.html
Spencer Weart is knowledgable and writes well. Reading his web site will relieve you of your abysmal confusion.
Regarding your 8 questions: Since the level depends upon future human activity, no single figure can be given. However, contingent upon different kinds of human activity in the future, different levels of warming can be (approximately) predicted.
February 16th, 2008 at 7:02 am
David Benson,
Thanks for clarifying.. I think what you are saying is that you think regional weather models are equal or worse in terms of prediction than climate models? I was of course thinking of a variety of models outside the climate/weather world.
All:
Here’s an idea. If you expect policy to be set and large quantities of tax dollars expended on policies based on your scientific models, I would suggest tighter QA/QC on the process. Each granting agency should require that the model is tested at least once each 5 years based on predictions of key results as determined by a panel of policy makers (and have independent review of the process and results). Those results, while possibly not publishable as papers in journals, could be compiled at the “model testing” website so that 5 years from now we could be discussing this based on data across disciplines. Wouldn’t that be interesting/productive?
February 16th, 2008 at 11:44 am
“Mark Bahner — You are simply blathering.”
I’m sorry you consider elementary science, “simply blathering.”
“Go read the actual statements of the NAS, AAAS, etc., to see that what I wrote is correct.”
I don’t have to go to the statements of the NAS, AAAS, etc., to see what you wrote is incorrect. I simply have to go to the IPCC Fourth Assessment Report.
Perhaps *you* should read it, too. Then label these 8 assertions for warming in the 21st century as “true,” “false,” or “don’t know”:
1) The IPCC thinks there is more than a 0.1 percent chance of warming more than 6.4 deg C.
2) The IPCC thinks there is more than a 10 percent chance of warming more than 6.4 deg C.
3) The IPCC thinks there is more than a 50 percent chance of warming more than 6.4 deg C.
4) The IPCC thinks there is more than a 99 percent chance of warming more than 6.4 deg C.
5) The IPCC thinks there is more than a 0.1 percent chance of warming less 1.1 deg C.
6) The IPCC thinks there is more than a 10 percent chance of warming less than 1.1 deg C.
7) The IPCC thinks there is more than a 50 percent chance of warming less than 1.1 deg C.
The IPCC thinks there is more than a 99 percent chance of warming less than 1.1 deg C.
Let me know when you have some answers. Are the answers, “true,” “false,” or “don’t know”?
“The last hold-out was the petroleum geologists, who recently rescinded their contrarian position, and now have no organizational position at all.”
You seem to be confused about what is or isn’t science. People’s opinions aren’t science. Science isn’t democratic. It’s about what is true and what is false.
Again, let me know when you decide whether the answers to those 8 assertions are “true,” “false,” or “don’t know.” Then we can talk some science.
“The evidence for the connection between global temperature and CO2 has been known for 150 years.”
Yes, for 150 years, it has been thought that, all other things equal, an increase in CO2 would increase temperature. That’s not the question, either scientifically or policy-wise. The question the IPCC should be answering is, “In the absence of intervention by governments, what will be the warming in the 21st century?”
Again, are the 8 assertions, “true,” “false,” or “don’t know”?
“Ever textbook on atmospheric physics explains the details. But easier for you, I opine, is…”
Look, David, I don’t need *you* to educate me on global warming.
“Regarding your 8 questions: Since the level depends upon future human activity, no single figure can be given. However, contingent upon different kinds of human activity in the future, different levels of warming can be (approximately) predicted.”
And you have the gall to say *I’m* “blathering!” Why don’t you answer the question? The question is NOT what the warming WILL BE in the 21st century, it’s what the IPCC Fourth Assessment Report CONCLUDED the warming will be in the 21st century. Based on the AR4, what does the IPCC THINK?
Try reading the AR4. Then answer the questions (“true,” “false,” or “don’t know”). I’ll even give you a hint on the answers…James Annan probably knows what he’s talking about.
Once you’ve done those simple tasks, then we talk further.
February 16th, 2008 at 12:03 pm
docpine — Not weather models. Regional climate models. The climate modelers tell me that their climate models do better globally than regionally. The push is on to try to make more specific predictions of the effect of future climate in various regions of the world.
Mark Bahner — You’re blathering even less coherently, if that is possible. Turns out I am a scientist, not a climate scientist nor an akin field, but a long time amateur geologist who has taken the trouble to read 3.5 books on climatology and about a dozen papers about matters climatic, mostly paleoclimate. (I’ve retired, so I’ve more time to do that, but I keep my hand in writing papers for peer-reviewed journals.)
As for your 8 silly questions, which scenario? IPCC considered several. In any case, the science of climatology proceeds rapidly these days, and the IPCC AR4 report is already a bit dated in spots. For your 8 silly questions, the answers do not lie solely in your (“true,” “false,” or “don’t know”) space. Go learn something about Bayesian reasoning. Then possibly you can write something coherent (and hopefully less repetitious). When you state ‘Science isn’t democratic. It’s about what is true and what is false.’ I conclude you have an overly naive view of science.
February 16th, 2008 at 3:05 pm
“Turns out I am a scientist, not a climate scientist nor an akin field,…”
Yes, I know you’re not a climate scientist or in any akin field. So perhaps you’d be wise to pay attention to what people who *are* climate scientists or in akin fields are saying.
For instance, James Annan *is* a climate scientist:
http://www.jamstec.go.jp/frsgc/research/d5/jdannan/
He says that the proper characterization for all 8 assertions is “don’t know.”
Or consider John Nielsen-Gammon, who is the Texas State Climatologist:
http://www.met.tamu.edu/people/faculty/nielsen-gammon.php
When I asked him to label similar assertions regarding the Third Assessment Report as “true,” “false,” or “don’t know,” he stated the following:
“With regard to your P.S. yes-no statements, none of the eight statements were specifically addressed in the reports, and because there were no probabilities assigned to the various emissions scenarios studied and not studied, the report does not provide enough information to answer other than ‘don’t know’ to each of the eight statements.”
Or consider Jesse Ausubel. He is the director of the Program for the Human Environment at The Rockefeller University, and one of the premier researchers regarding long-term trends in energy use and carbon dioxide emissions of industrial processes.
http://phe.rockefeller.edu/jesse/
Here is what he said about the scenarios in the Third Assessment Report:
“The IPCC’s 2001 Third Assessment Report uses 40 scenarios which show decarbonization and carbonization going in all different directions with no probabilities attached. Failing to provide probabilities is unscientific…”
http://phe.rockefeller.edu/PDF_FILES/CoolerHeads.pdf
“As for your 8 silly questions,…”
Oh, brother. Let’s start with the basics. They’re assertions, not questions.
But more importantly, do you really think it doesn’t matter whether, in the absence of government intervention, there is a greater than 99% chance of warming MORE than 6.4 degrees Celsius, or a greater than 99% chance of warming LESS than 1.1 degree Celsius?
Why in the world would it not matter what the warming will be in the absence of government intervention? Do you think that governments should intervene REGARDLESS of what the warming will be if they don’t intervene? If there is a 99% chance of warming LESS than 1.1 degree Celsius, why should governments intervene?
“…which scenario? IPCC considered several.”
Yes, they “considered several,” so clueless amateurs would be baffled into thinking they were actually doing science. And guess what…it worked! Clueless amateurs *have* been baffled into thinking the IPCC is actually doing science. Just like clueless amateurs think that the Limits to Growth series of books are science.
That’s why the clueless amateurs at “Real Climate” answered Roger’s question, but the authors at Real Climate didn’t. It’s much more convenient for the authors at Real Climate to rely on “useful idiots.”
http://en.wikipedia.org/wiki/Useful_idiot
I’ve made 8 assertions. All 8 of the assertions are either “true,” “false,” or “don’t know.” I’ve already given you pretty clear hints about whether the assertions should be characterized as “true,” “false,” or “don’t know.” Do you need more hints?
“Go learn something about Bayesian reasoning.”
I have a better idea. You go read something that even high school students in Batesville, Indiana are learning:
http://www.batesville.k12.in.us/Physics/PhyNet/AboutScience/Hypotheses.html
Or do you think the students at Batesville High School are being told something that isn’t true?
“When you state ‘Science isn’t democratic. It’s about what is true and what is false.’ I conclude you have an overly naive view of science.”
Ho, ho, ho! Says you! Albert Einstein begged to differ.:
“Einstein was attacked by some with anti-Jewish leanings. When a pamphlet was published entitled 100 Authors Against Einstein, Einstein retorted If I were wrong, one would be enough.’”
http://scienceworld.wolfram.com/biography/Einstein.html
Or do you think Albert Einstein also had, “An overly naive view of science”?
February 17th, 2008 at 8:04 am
Mark Bahner — Are you a high school student in Batesville? Even Wikipedia does better than the link you provided. While it would be ideal if scientific hypotheses were falsifiable, they are not. Even Sir Karl seemed to be recognizing that in his latter days. What scientific hypotheses are is
comparable.
It is possible to compare two hypotheses, say H and K, against a body of evidence E. The hypothesis which better explains the data is then more confirmed and the other more disconfirmed. As I said, go learn something about Bayesian reasoning, a necessity for a science as complex as climatology.
So yes, A. Einstein seems to have had an overly naive view of science as a whole. His statement, “one fact would be enough”, is largely correct regarding his theory of general relativity (GR), one of the most highly internally constrained theories in existence, with only one freely adjustable parameter, the cosmological constant. However, GR has failed the test of the first attempt at a frame-dragging experiment. Perhaps the test was inadequate. Gravity Probe B tried again and the first analysis of the data found no frame-dragging. The second attempt was supposed to report last December; the report is delayed for a few more months. Note that mere facts, i.e., tests, have not (yet) brought down Einstein’s GR. Thus Popperian falsifiability is shown not to be a criterion actually used by practicing scientists. Informal (and sometimes formal) Bayesian reasoning is. So the answers to your 8 questions (posed as assertions) can only be given via a probability. Try some Bayesian reasoning. You’ll find it quite liberating.
With regard to the replies to the question on Real Climate posed by Roger Pielke, Jr., the knowledgeable amateurs provided a long list of replys. (I am especially impressed by Tamino, since I do some (Bayesian) time seris analysis myself). The replys demonstrated that the amateurs thoroughly know about the results obtained from climate models. In all cases but one (on the Real Climate thread), the models made the correct predictions. Using informal Bayesian reasoning, we have that the models are strongly confirmed (in comparison, say, to the hypothesis that more CO2 does not lead to more warming and ocean acidification).
You do youself no credit to call those capable people ‘useful idiots’. By this, and other writings of yours, my estimation of your mental prowness continues to drop.
February 18th, 2008 at 12:38 pm
“Mark Bahner — Are you a high school student in Batesville?”
No, I’m a practicing environmental engineer. I have Masters Degree in Environmental Engineering (Air Pollution Option) and more than 20 years of experience performing environmental analyses, (mostly related to air pollution).
So when people like you write things like, “It is only in this century that sufficient evidence has been gathered so that no scientific body opposed the IPCC conclusions and the most prestigious support it,” I am like an accountant who reads a heart surgeon writing in 2000 about what a wonderful job the accountants at Arthur Anderson are doing at Enron.
However, re-reading my comments here, I can see I have several times over-reacted. See below for one important particular. But it’s because I’m angry. I’m angry because environmental analyses are what I do for a living. When people do fraudulent environmental analyses–as the IPCC has been for more than a decade—they are crapping in my field. (Note: Environmental engineering actually uses the principles of environmental science to develop innovations to benefit humankind. But if environmental science is viewed as a disreputable and unscientific attempt to defend pre-existing conclusions, then environmental engineering will also be significantly adversely affected. And with a few very respect-worthy exceptions, many members of the climate change community are quickly turning the phrase “environmental science” into an oxymoron.
“Using informal Bayesian reasoning, we have that the models are strongly confirmed (in comparison, say, to the hypothesis that more CO2 does not lead to more warming and ocean acidification).”
Well, no duh! How many scientists can you name who say that–ALL OTHER THINGS EQUAL (see important note below)****– “CO2 does not lead to more warming,” or, “CO2 will not lead to the pH of the ocean declining”? I’m serious…are you really so out of it that you think such a scientist exists?
****BTW, if you know anything at all about this subject, you should know that all other things are NEVER equal in the real world. In the real world, there are:
1) Atmospheric concentrations of methane. (By the way: Which, if any, of the IPCC scenarios do you think accurately project atmospheric methane concentrations from, say, 2020-2040?)
2) Emissions of sulfates-forming sulfur dioxide, thought to cause COOLING. (By the way: Which, if any, of the IPCC scenarios to you think accurately project world SO2 emissions from, say, 2020-2040?)
3) Emissions of carbon black. (By the way: Which, if any, of the IPCC scenarios do you think accurately project world carbon black emissions from, say, 2020-2040?)
4) Changes in atmospheric water vapor, condensed water, and ice crystals. (All these can have strong warming effects, strong cooling effects, or a balanced/neutral effect.)
5) Changes in solar output. (By the way: What do you think would happen to the average world surface temperature if the sun’s activity suddenly changed to that experienced during the Maunder Mininum? Is that going to happen in the 21st century?)
6) Changes in local and regional surface albedo.
Etc., etc., etc. So any the all-other-things equal model that “predicts” increasing CO2 will lead to more warming and ocean pH to decline doesn’t mean squat. There is NO model that says that (all other things equal) CO2 does NOT lead to more warming.
“You do youself no credit to call those capable people ‘useful idiots’.”
Yes, I shouldn’t have ever used that phrase, and I apologize to the people who wrote in reply to Roger Pielke Jr.’s questions. Some of them were indeed quite good. (Though some were clearly wrong.) But more importantly, my anger was at the authors at Real Climate for not ever answering Roger’s question…which he has already asked several times. At very least, the authors at Real Climate should have added a comment that said, “Roger, we agree with responses to your question from X (or X, Y, and Z).”
“By this, and other writings of yours, my estimation of your mental prowness continues to drop.”
David, that reminds me of something Andrew Dessler wrote to me here on Prometheus in the same vane. I’ll give you an answer similar to the one I gave him. I have repeatedly asked you to perform a simple task. I have repeatedly asked you to read the IPCC Fourth Assessment Report (AR4), and properly characterize the 8 assertions I made as being, “true,” “false,” or “don’t know.” You have *still* not completed this simple task, even though I’ve told you what experts in the field (including Bayesian analysis and climate change) have written.
So one more time, I’ll ask you to read the AR4 (have you actually ever even done that…even a small part?), and properly characterize those 8 assertions. You even have the characterizations of James Annan and John Nielsen-Gammon to guide you. If you don’t have the honesty and “mental prowess” to perform that simple task—to characterize those 8 assertions as “true,” “false,” or “don’t know” based on what is in the AR4—then why in the world should I care what your opinion of my “mental prowess” is?
February 18th, 2008 at 12:43 pm
P.S. David Benson—
You write, “So yes, A. Einstein seems to have had an overly naive view of science as a whole. His statement, ‘one fact would be enough’, is largely correct regarding his theory of general relativity…”
No, David. Einstein said no such thing. If you read his quote more carefully, or the headline to which he was responding, he most emphatically did NOT say, “one fact would be enough.” He said “one (****author****) would be enough.” The headline was “100 Authors Against Einstein”…not “100 Facts Against Einstein.” The point Einstein was making was that science is not a democracy. (Which is exactly the same point I’d made earlier.) Do you agree, or disagree?
You further write, “So the answers to your 8 questions (posed as assertions)…”
No, they are not “questions (posed as assertions).” They are 8 assertions. Please consult a dictionary if you’re unclear about the difference between the two.
“…can only be given via a probability.”
Oh, really?! I can’t wait to read this!
Here (again) are 8 assertions regarding what the IPCC wrote in their Fourth Assessment Report about surface temperature rise in (approximately) the 21st century, in the absence of intervention by governments:
1) The IPCC thinks there is more than a 0.1 percent chance of warming more than 6.4 deg C.
2) The IPCC thinks there is more than a 10 percent chance of warming more than 6.4 deg C.
3) The IPCC thinks there is more than a 50 percent chance of warming more than 6.4 deg C.
4) The IPCC thinks there is more than a 99 percent chance of warming more than 6.4 deg C.
5) The IPCC thinks there is more than a 0.1 percent chance of warming less 1.1 deg C.
6) The IPCC thinks there is more than a 10 percent chance of warming less than 1.1 deg C.
7) The IPCC thinks there is more than a 50 percent chance of warming less than 1.1 deg C.
The IPCC thinks there is more than a 99 percent chance of warming less than 1.1 deg C.
Please characterize those assertions as “true,” “false,” or “don’t know.” (If you’re looking for probabilities, I’d suggest, “0 percent true, 0 percent false, and 100 percent don’t know” as the proper characterization of each assertion, based on the information in AR4. J)
P.S. Let me emphasize once again that I’m not asking you to characterize what WILL happen, but what the IPCC WROTE in their AR4. So this is not some prediction of future events. I’m asking you to simply read the AR4, and use that to document to accurately characterize those assertions, based on what the IPCC wrote.
February 18th, 2008 at 11:25 pm
Mr. Benson,
I am a climate modeler. Here are my predictions
of future climate. Following the example of Dr. Hansen, and modern science, I am not going to make strict Popperian predictions but the ultra-modern Bayesian projections and scenarios.
My prediction:
The climate will warm within the next century unless something else happens in which case it will not warm.
I put a 50% probability on the scenario that global warmth will continue at its present rate. I also put a 50% probability on the scenario that something will happen and that it will not warm at its present rate and may even reverse.
Please don’t ask me to quantify what is the ‘else’ that will happen. That is so old-fashioned! There are certain parameters that we cannot just accurately identify nor quantify that will affect predictions. I am not God. Please don’t ask such silly Popperian questions!
I can assure you, that “by informal Bayesian reasoning my climate models will demonstrate quite an impressive record (in fact, perfect) in predictions made and agreement with the actual climate response.”
Given my hypothesis X and the IPCC’s hypothesis Y of future event E, I am sure that my hypothesis will prove to be far more accurate and thus bestowed with truthiness.
Just between you and me, I don’t understand why these climate scientists can get so heated up about models. My models are far far more accurate.
The IPCC actually tried to steal the algorithms for my computer model! But because you are a good friend and a fellow bayesian scientist, I can give you the secret for a trivial consideration. Just send me 1000 USD via paypal and I will send you my secret algorithm! Send money now and you get a 5% discount!
What is there to lose, eh?
February 19th, 2008 at 5:58 am
Robert Panqilinan — Ha, ha!
Mark Bahner — As one of your respondents already noted, IPCC assigned no probabilities to the various scenarios. So for each of the 8, I’d have to assign some prior (subjective) probability. As that’s not in the IPCC AR4, it’s not what they wrote. [What a silly game you are playing.]
Hmmm, you might wish to check an authoritative source regarding the quote from A. Einstein. But in the sense of ‘one person, one vote’, science is not a democracy. However, you might care to read some philosophy of science and history of science before writing further on matters which you seem not to know about.
You wrote ‘Well, no duh! How many scientists can you name who say that–ALL OTHER THINGS EQUAL (see important note below)****– “CO2 does not lead to more warming,” or, “CO2 will not lead to the pH of the ocean declining”? I’m serious…are you really so out of it that you think such a scientist exists?’
Yes, such scientists do exist. Just as there are scientists who deny the modern theory of biological evolution. In both cases these will be people speaking or writing outside their field of speciality. An example is a retired physics professor (from Yale) who is best known on the web for writing about the physics of baseball. He signed the Heartland Institute ‘petition’ which states that the global warming is natural and there is no crisis. Another, a chemist, appeared on the ‘climatechange’ blog. It is clear that he posted without understanding the first thing about CO2 in the atmosphere. Over the last two years several people claiming to be geologists have posted comments which made clear they had no clue about the role of CO2. So yes, there were lots and lots. Maybe fewer now that Real Climate has had over 5 million visits.
As for the Real Climate guys not answering Roger Pielke, Jr., there is a whole previous thread which attempts to correct his misconceptions regarding short data series. Based on that experience, and the tradition on Real Climate of leaving to the capable amateurs most questions, I found the responses quite sensible and helpful. I did not, obviously, find the main post on this thread by Roger Pielke, Jr., to be a professionally responsible response. By the way, which of the amateur responses (other than the ones Pielke quotes) do you think are wrong?
Finally, it appears that some of the model assessment questions are being addressed in the effort headed by LLNL. We can hope something useful will come out of it.
February 19th, 2008 at 6:12 am
David B. Benson said, “Greater use of formal Bayesian methods could make such statements (more) quantitative.”
I have no idea why you were not able to figure out that your ‘could’ implied these two hypotheses:
Hypothesis H: “Greater use of formal Bayesian methods WILL make such statements (more) quantitative.”
Hypothesis K: “Greater use of formal Bayesian methods WILL NOT make such statements (more) quantitative.”
All you need now is evidence in order to calculate the degree of certainty of each hypothesis. Or you could ignore this side issue and deal with the original post. Specifically, at what degree of confirmation will an outlying observation become a disconfirmation thus suggesting an alternative to the consensus when it comes to policy making?
John M Reynolds
February 19th, 2008 at 9:09 am
John M Reynolds — Go back a few comments to see what I wrote regarding Einstein’s GR and frame-dragging experiments. This example, along with many others, suggests the decision making processes actually used by scientists are not so simple as you (and others) would like such to be.
Also, you could go to even Wikipedia to find out something about Bayesian reasoning. Then you would not write so foolishly. In particular, confirmation and disconfirmation can only occur when there are (at least) two hypothesis to be compared with regard to the ability to explain the evidence. So other than an imagined generic GCM climate model, what is the competing hypothesis?
Your final question makes no sense without such.
February 19th, 2008 at 11:42 am
David Benson –
Sigh. I’m not really sure why I bother. But no one has ever accused me of lacking persistence.
Sooooo…
“As one of your respondents already noted, IPCC assigned no probabilities to the various scenarios.”
Yes, exactly. So that means the proper characterization of the 8 assertions is…? What did that one respondent say was the proper characterization for each of the 8 assertions? And what did the other respondent also was the proper characterization for each of the 8 assertions?
(Hint: The proper characterization doesn’t start with a “T,” and it doesn’t start with an “F.” So the proper characterization of each of the 8 assertions is…?)
Mark
P.S. John M. Reynolds – Best of luck to you. You’re going to need it! (Basic logic is very difficult for some.)
February 19th, 2008 at 12:48 pm
Mark Bahner — You strike me as terribly rigid. Not good for an engineer.
Haven’t you got it through your head that I don’t give a d**n for your game?
At least this post you did not repeat the infamous list of 8. I suppose that is to balance the previous post wherein you posted the list twice.
I don’t see, in any sense, how your infamous list of 8 has anything to do with either the main post by Roger Pielke, Jr., initiating this thread nor my objections to his statements therein. Are you just a concern troll?
February 20th, 2008 at 2:37 am
Wow. David, you suggested that it is not a matter of black or white, right or wrong, proven or unproven, but instead we should consider shades of grey. You said that the models cannot be proven wrong, but an observation can be shown to have a calculable degree of confirmation or disconfirmation. The stronger the confirmation, the more accurate is the model. The stronger the disconfirmation, the less accurate is the model. All my last question does is to ask you to apply your knowledge of Bayesian theory. I ask you to suggest the degree of disconfirmation that an observation would require to make the model inaccurate enough that policy makers should disregard the model. What is that level of disconfirmation? — John M Reynolds
February 20th, 2008 at 4:31 am
John M Reynolds — NO! Given evidence (observations) it is the hypotheses which are compared for the degree of confirmation. I require at least two hypotheses to compare. What is your second hypothesis, the one other than a generic GCM climate model?
Without at least two hypotheses, no comparison is possible. Got it?
February 20th, 2008 at 6:36 am
“…the one other than a generic GCM climate model”
A climate model is not a hypothesis. All you need is one hypothesis and you magically have two. Itself and its antithesis:
Hypothesis A: Climate models produce forecasts that are accurate.
Hypothesis B: Climate models produce forecasts that are NOT accurate.
Quoting John Christy again, “[t]hus what the models predicted for post 1979 tropical tropospheric warming was statistically significantly much more than observed.” At some degree of disconfirmation of Hypothesis A, we would get to the point such that policy makers should ignore the models. What is that degree of disconfirmation?
John M Reynolds
February 20th, 2008 at 7:05 am
John M Reynolds — The Bayesian factor method does not work that way. It requires a means to detrmine the probability of the evidence (data) given the hypothesis. You have not provided such. Read
http://en.wikipedia.org/wiki/Bayes_factor
and try again. (Maybe we are getting somewhere useful.)
February 20th, 2008 at 11:22 am
To illustrate what is necessary to use the Bayesian factor method, I went through the comments regarding the question raised by Roger Pielke, Jr., on the Real Climate thread to count the successes and failures of GCM climate models posted there by commenters. I found 14 successes and 2 failures (including the tropical temperature issue raised on this thread).
I will treat a success as good, not perfect, to assign a probablity of p = 0.8 for each success and a probability of q = 1-p = 0.2 to each failure. Then the probability of the 16 pieces of evidence, given global models hypothesis G, is
P(E|G) = p^(14)q^2.
For an alternative hypothesis, consider a ‘random’, heads-or-tails, hypothesis R which happens to come up heads as often as tails,
P(E|R) = p^8q^8.
To computer the decibans measuring just how throughly disconfirmed hypothesis R is, we have
10*log_(10)[P(E|G)/P(E|R)] = 10*[6log0.8 - 6log0.2]
= 10*[5.14854 - 1.80618] = 10*3.34236
= 33.4236 decibans,
decisive disconfirmation of ‘random’,
odds in excess of 1220 to 1. That is, the weight of the evidence is overwhelmingly in favor of hypothesis G as opposed to hypothesis R.
This does not answer to question raised by Roger Pielke, Jr., regarding regional predictions. But it does suggest a completely objective and rational approach.
February 20th, 2008 at 1:03 pm
David Benson—
You write, “You strike me as terribly rigid. Not good for an engineer.”
Well, I don’t think you have much idea about what it takes to be a good engineer. In fact, I don’t think you much idea about what it takes to be a good scientist. In my opinion, what it takes to be a good scientist is an absolutely ferocious dedication to finding and reporting the truth (the whole truth, and nothing but the truth).
“Haven’t you got it through your head that I don’t give a d**n for your game?”
It’s not a game. Not to me. Like I wrote before, I do environmental analyses for a living. It’s my business. So I take fraudulent environmental analyses very seriously. But I am at long last becoming aware that you aren’t particularly interested in the truth. So I’ll keep that in mind in future comments from you or to you.
“At least this post you did not repeat the infamous list of 8. I suppose that is to balance the previous post wherein you posted the list twice.
”
See my previous comments. In which “previous post” did I “post the list twice”?
“I don’t see, in any sense, how your infamous list of 8 has anything to do with either the main post by Roger Pielke, Jr., initiating this thread nor my objections to his statements therein.”
See, when you write something like that, I have no idea whether it’s from your cluelessness, or from your dishonesty. This is just like your response to my quote from Einstein. Anyone who knows anything about the history or philosophy of science would have recognized Einstein’s quote immediately. (See postscript.) So when you converted it to “one fact would be enough,” I figured you were simply a clueless blowhard. But now I’m not so sure. I’m not sure whether your blatant misrepresentation of Einstein’s words was from cluelessness or dishonesty.
The point of my list of 8 assertions was in trying to finally get you to admit the truth (which should not be something any scientist should have to “admit”…he/she should be proclaiming it at every opportunity!) that the proper characterization for each of those 8 assertions was, “Don’t know.” That is, the IPCC does not provide any scientifically valid (or policy-relevant) assessment of what will happen to the global surface temperature in the absence of government intervention. In short, they don’t make any conclusion about what the temperatures will be in the absence of government intervention.
So your statement that, “ It is only in this century that sufficient evidence has been gathered so that no scientific body opposed the IPCC conclusions and the most prestigious support it” completely ignored the fact that the IPCC made no conclusion about what temperatures would be in the 21st century, in the absence of government intervention. Most people would consider that to be an important question, when deciding whether or how much government intervention might be necessary.
P.S. You made the following comment (hilarious!…you’re the King of Irony, and I’ll bet you don’t even know it!) about the quote from Einstein: “Hmmm, you might wish to check an authoritative source regarding the quote from A. Einstein.”
Bwahahahaha! There’s no need for ****me**** to “check an authoritive source,” David, because I actually know what the h*ll I’m writing about. Only someone who knows absolutely nothing about the history and philosophy of science would not instantly recognize that quote as Einstein’s comment that science isn’t democratic. Why wouldn’t ***you*** “check an authoritive source,” since you obviously know nothing of the quote? Or haven’t you ever heard of Google? You know, the search engine?
But since you don’t seem to be able to handle such tasks (in addition to not having the honesty to label 8 simple assertions), here’s the quote from the Time Magazine website.
http://www.time.com/time/time100/poc/magazine/a_brief_history_of_rela6d.html
“When told of publication of the book One Hundred Authors Against Einstein, he replied, Why 100? If I were wrong, one would have been enough.”
February 20th, 2008 at 1:35 pm
Mark, David, Please keep the exchange civil. Otherwise, please exchange emails and yell at each other in private
Thanks!
February 20th, 2008 at 3:04 pm
Hi Roger,
Yes, I shouldn’t even bother.
But this whole episode has opened my eyes a bit. I’ve been thinking that, when the public becomes aware of the IPCC’s lies and exaggerations, that there would be a significant anti-science backlash. But maybe not. Maybe scientists (or “scientists”) will simply say, “Oh, we knew that all along.” And maybe most people will say, “Well, scientists know best.” And then they’ll wait for the next apocalyptical message.
I’ve thought that society has evolved tremendously from the Dark Ages. But it may be that most people need religion. Any religion. It seems like most people need to think they’re living in apocalyptical times.
Mark
P.S. But it is very sad. An awful lot of money and time has been wasted, while real problems are not being addressed.
P.P.S. Like hurricane storm surge. But that’s something for another time.
February 20th, 2008 at 5:19 pm
Hi Everyone.
I second Roger’s request to try keep the dialogue civil. Although it has been entertaining, the personal insults have gotten a bit much. But if you insist on continuing, why not go all the way? I recommend including some of these:
http://www.thejokeyard.com/your_momma_jokes/index.html
David, Mark: I think you’re speaking past each other and aren’t debating the same question. Here’s my analysis:
1. Roger made an argument about climate science and falsifiability.
2. David responded to Roger’s post, and highlighted that climate science does not abide by a strict Popperian definition. For what it’s worth, I agree.
3. Mark responded not to Roger or David’s comment, but instead wrote a meta-comment that reflects his general view on the IPCC. He made statements that appear to me to be correct (although I should say that I haven’t read the full IPCC AR4). i.e., AR4 does not in fact make predictions on warming that would occur w/o gov’t intervention. I would agree that the questions Mark raised (the infamous list of
are important…although I wouldn’t have said they are “fundamental.”
4. David is responding to Mark’s simplified view of science and neglect of Bayesian reasoning. While David’s critique appears to be valid (i.e. Mark’s view of science is overly simplified), he hasn’t adequately responded to what Mark identifies as IPCC’s shortcomings.
To respond, David would have to: a) admit that IPCC didn’t answer these important questions and that it was a shortcoming; b) show that they did answer them; or c) show that these questions aren’t important
Thoughts?
February 21st, 2008 at 1:48 am
Thanks Praj, very helpful!
David B. Benson- I’m all for Bayesian reasoning in this case, but I’m curious why your complaint is with me taking climate forecasts as presented by climate forecasters (in strict Popperian fashion), rather than with the climate forecasters, the vast majority of whom have not used Bayesian approaches (e.g., such as the IPCC).
February 21st, 2008 at 2:09 am
David. Thanks for proving Roger right. Below is my analysis of your last reply to me, but keep in mind that the key point is still the tropical troposphere prediction. The rest is just fun scientific analysis.
- – -
Perhaps I am just confused, but should the “6log0.8 – 6log0.2″ not be “8log0.8 – 8log0.2″ since you were using 16 and not 12 observations?
Out of curiosity, did you make a list of those 16 observations?
“I found 14 successes and 2 failures (including the tropical temperature issue raised on this thread).
I will treat a success as good, not perfect, to assign a probability of p = 0.8 for each success and a probability of q = 1-p = 0.2 to each failure.”
The 0.8 and 0.2 are just guesses. You are using probabilities here that have not been calculated. The end result will only be as accurate as your estimate of the success’ and failures’ probabilities.
By the way, that tropical troposphere observation is particularly important since the models predicted, as reported by the IPCC reports, that the tropical troposphere temperature would increase if the warming was caused by anthropogenic greenhouse gas emissions. Without that, the whole ‘fight global warming’ shtick is pointless.
As well, you only had two failures. I am sure we can come up with several observations that are outliers, thus reducing your estimated 1 in 1220 risk farther.
- Antarctica getting colder (This one is apparently a success according to RC)
- The edges of Antarctica are getting warmer where volcanoes exist
- fewer than normal volcanic eruptions for the past 100 or so years
- Antarctica ice getting thicker and increasing in extent
- The recent global drop in temperature of 0.64C in 12 months
- brown clouds decreasing the albedo over India and North Indian ocean
- Tropical troposphere not getting warmer
- the past 10 years saw the warming trend level off though CO2 emissions accelerated more than predicted
- The fact that CO2 warming effect is logarithmic such that as the concentration increases, the warming effect of the new molecules is diminished. This flies into the theory of the existence of a tipping point that I thought the models predicted.
- biology has increased its uptake of the extra green house gases. As a result, methane has not been increasing and plants, like trees in New York, are doing well.
- reduced solar output permitted extra cosmic rays to seed clouds reducing the albedo
- non-emissions effects by people like land use
- The estimate of the feedback warming has been exaggerated possibly due to the estimated feedback forcing of water vapour being too positive.
As well, why would you compare it to a random hypothesis? The complexity of the climate may make it seem like it is random, but the reality is there are many cycles (volcanoes being an exception) that affect climate as well as feedbacks to how those cycles overlap and combine. One of the biggest feedbacks is biology. For a useful comparison, if you won’t use its antithesis, then how about comparing it to the sun? Do the observations correlate better to the rise in man made CO2 emissions or do they correlate better with the solar cycles?
In summary, with only using two failures, a random second hypothesis, and only estimating the level of probability, your 1 in 1220 is not yet useful. The real issue would be to figure out how much of the 1980s and 1990s warming is due to the accelerating manmade CO2 emissions. Since that crucial tropical troposphere observation has been a failure, the models are now useless from a political standpoint. Thanks for proving Roger right.
John M Reynolds
February 21st, 2008 at 5:42 am
In the midst of writing a careful reply to all, I suddenly was SIGNED OUT, and the comment I was previewing was lost.
Maybe later.
February 21st, 2008 at 6:09 am
Ok, I have a little time now and then bits and drabs later today. So I’ll atttempt just a little at a time.
Regarding IPCC AR4, my comments were originally directed at their attribution of increased global warming to human causes at the 10 to 1 level. Now that the petroleum goelogists have withdrawn their statement no technical or scientific organization disagrees with the IPCC AR4 assessment, and many such organizations have supportive statements. That was my only point.
Mark Bahner — Thanks for finding the correct quotation, “If I were wrong, one would have been enough,” from A. Einstein. I had recently seen it misquoted. However, I stand by the frame-dragging experiment, as one example out of many, to show that science does not proceed in a strict Popperian fashion. In the case of frame-dragging, if the Gravity Probe B shows none, I rather expect a Gravity Probe C will be designed and flown.
February 21st, 2008 at 6:25 am
Praj — Thanks, a useful summary.
The thousands of climatologists who worked on the IPCC AR4 recognized they were not economists. So did not assign probabilites to their scenarios.
I am not a climatologist, so I’ll defer to those climate modelers who did reply to the questions about the infamous list of 8. My opinion about Mark’s questions does not matter, not even to me.
Roger Pielke, Jr. — Having sat through far to many report writing meetings, I opine that what happens in such is a form of informal Bayesian reasoning. I assume, possibly incorrectly, that the climatologists working on the IPCC AR4 did the same. That said, I of course would certainly prefer the use of some form of formal Bayesian reasoning when doing assessment. Doing such, including explict statement of (subjective) priors would, IMO, substantively inform science-based policy.
February 21st, 2008 at 8:40 am
John M Reynolds —
P(E|G)/P(E|R) = (p^(14)q^2)/(p^8q^8) = p^6/q^6
Yes, I have a list of the 16, 15 gleaned from the comments on the Real Climate thread, tropical temperatures from here.
The probabilies and the use of the ‘random’ hypthesis R are simply to illustrate the Bayes factor method. Nothing more. If you apply the method to the antithesis of G, you will discover why I did not use it for an illustration.
As for your outliers, I’ll just comment on two of them:
(1) “The recent global drop in temperature of 0.64C in 12 months.” I doubt this. Cite an authoriative reference. Even if true, climate is about a 33 year average of the weather. One year is not enough to say anything.
(2) “Brown clouds decreasing the albedo over India and North Indian ocean.” This is an exogeneous forcing, not a model prediction.
And maybe a third. A so-called tipping point occurs when tundra warms enough to express methane, which then causes further warmer, which further warms the tundra, etc. I doubt that anyone can say at what level this effect occurs, but it is known that some of the Siberian tundra has begun expressing some methane.
The physics of global warming (so-called greenhouse) gasses is well understood, starting 150 yearss ago now. Hence warming has occurred and continues, since the level of CO2 in the atmosphere has been measured continuously for 50 years now, and is known to be of fossil origin. The fact that ocean acidification is occurring (a very serious problem) is measured and known. That the warming continues for a very long time is known physics. None of these facts require climate models to measure or calculate.
The tropical temperatures are not crucial, just a matter of getting the physics of clouds right, something the climate modelers say they need to do. Regional climate forecasts, even if somewhat incorrect, provide useful guidance to planners. For example, Hadley Centre about two years ago provided such a region-by-region forecast for 2050 CE. All I can check is the forecasts for western lower U.S., British Columbia and also Chilean Patagonia and western Argentine Patagonia. In these locations the forecasts appear very sensible. I come to much the same conclusions without a computerized climate model, using solely Hadley cell expansion. In any case planners in the three westernmost U.S. states and in B.C. are already using the Hadley Centre forecasts to commence various actions. I wish I could observe the same happening in Patagonia.
I certainly have not ‘proved Roger right’. Neither have I done more than suggest that better assessment of climate model results might be done via Bayesian techniques. So far it does not appear tha LLNL is going to use a Bayesian approach, but whatever they choose to do is likely to be useful.
February 21st, 2008 at 1:44 pm
Hi Praj,
Yes, I agree completely with Roger. Thanks for your very helpful comments. I’ll address each of them, regarding my previous comments:
“3. Mark responded not to Roger or David’s comment, but instead wrote a meta-comment that reflects his general view on the IPCC.”
Well, I was responding in part to Roger’s comment that, “And out of fear that legitimate efforts at falsifiability will be used as ammunition by skeptics (and make no mistake, they will) in the politics of climate change, issues of falsification are simply ignored or avoided.”
My comment was basically, yes, skeptics will use erroneous predictions as proof that actions shouldn’t be taken based on the models. (But if a major decision is made on the basis of model predictions, such skepticism seems pretty reasonable to me.) But I also pointed out that the IPCC has never been about making accurate predictions in the first place.
“He (Mark) made statements that appear to me to be correct (although I should say that I haven’t read the full IPCC AR4). i.e., AR4 does not in fact make predictions on warming that would occur w/o gov’t intervention.”
Well, don’t take my word for it. James Annan admitted that the correct characterization for all 8 assertions was “don’t know” for AR4. John Nielsen-Gammon stated (before AR4 was released) that the correct characterization for similar assertions regarding the Third Assessment Report was “don’t know.” Kevin Trenberth wrote on the Nature Climate Feedback blog that the IPCC hasn’t made any predictions:
http://blogs.nature.com/climatefeedback/2007/06/predictions_of_climate.html
However, when I asked Kevin Trenberth to make it clear to the blog laypeople just what the implications were of what he was writing, (i.e., to label all the 8 assertions as “true,” “false,” or “don’t know”) he never did it. Even though ANOTHER reader specifically requested that someone label the assertions. I also asked Olive Heffernan (News Editor at Nature) to do the same thing. She never did. And *why* didn’t Kevin Trenberth or Olive Heffernan acknowledge that the correct characterization for all 8 assertions would be “don’t know?” I think the evidence is very clear that they simply don’t want to acknowledge that scientific fraud has occurred. See my later comments, regarding scientific integrity.
“I would agree that the questions Mark raised (the infamous list of
are important…although I wouldn’t have said they are ‘fundamental.’”
I don’t understand how the questions could not be “fundamental”…in at least three areas: 1) in the area of policy making, 2) in the area of scientific legitimacy, and 3) in the area of scientific integrity.
1) Policy Making: Suppose we all knew for a fact (direct line from the Woman Upstairs
) that all the world’s governments were going to ignore global warming from 2008 onward. Mass amnesia. They’re going to forget that CO2 can even cause global warming. Now, suppose the Woman Upstairs said that the warming to the year 2100 was going to be exactly 0.812436 deg C. (Score one for Michael Crichton…that’s exactly as he’s predicted!) Now, suppose instead that the Woman Upstairs said the warming to the year 2100 was going to be exactly 6.4 deg C.
Wouldn’t that make a big difference in deciding the appropriate policies?
Or suppose She (the Woman Upstairs) said 3.1 deg C?
In short, how can it not be “fundamental” to have an estimate of what will happen if governments *don’t* intervene, before we decide how much governments should intervene, and by what degree?
2) Scientific Legitimacy: How can a document that effectively says, “We have no idea what the temperature will be like in the 21st century, absent government involvement” be scientifically legitimate? The 8 assertions span a range of 99+% certainty that the warming will be LESS than 1.1 deg C to 99+% certainty that the warming will be MORE than 6.4 deg C. If I made the statement, “The global surface temperature may warm over the 21st century,” would that be scientifically valid (let alone useful for policy making)? I don’t see how it would be scientifically valid, because it could not be falsified.
3) Scientific Integrity: The mere fact that Kevin Trenberth and Olive Heffernan–and the authors at Real Climate, and others—have not been willing to label my assertions as, “true,” “false,” or “don’t know” is itself compelling evidence of scientific fraud. How difficult is it to simply write, “The proper characterization of each is ‘don’t know”? That’s 7 words. And it could be shortened to 2: “Don’t know.”
The only reason not to say it is because it acknowledging it acknowledges that the IPCC has been committing fraud for many years.
February 21st, 2008 at 2:42 pm
Hope this isn’t a double-post, but I missed a word in this sentence. It should have been (added word capitalized):
In short, how can it not be “fundamental” NOT to have an estimate of what will happen if governments *don’t* intervene, before we decide how much governments should intervene, and by what degree?
(Too many negatives in the sentence, but I hope the meaning is clear.)
February 22nd, 2008 at 1:43 am
“The tropical temperatures are not crucial, …” The UN 2007 climate assessment report disagrees with you.
http://scienceandpublicpolicy.org/images/stories/papers/monckton/whatgreenhouse/moncktongreenhousewarming.pdf
Since the evidence has shown that the warming is not caused by anthropogenic greenhouse gas emissions, then the models are useless from a political perspective. Your confirming that the tropical troposphere observations are a failure of the models provides us with the answer to Roger’s question. Since the IPCC should have realized this but still held their stance on the likelihood that we can even fight global warming, then we get into the claims of fraud that Mark Bahner is discussing.
And just because I am a glutton for punishment, you asked me to “Cite an authoriative reference.” I will go one better. I prefer to go to the data. Here a link to the graphed data from HadCRUT, GISS, UAH, and RSS:
http://wattsupwiththat.wordpress.com/2008/02/19/january-2008-4-sources-say-globally-cooler-in-the-past-12-months/
(If you want access to the data numbers, you can get those from links from Anthony Watt’s page linked above.)
I found it interesting that it was the GISS data set that had the largest drop in temperature of -0.75 C over 12 months. It was John Christy that had the drop of the smallest magnitude of -0.588 C over the same period. I quoted the average of the above sources in my previous comment.
Oh, and Mark, you said, “I don’t understand how the questions could not be ‘fundamental’.” I thought we already determined that they are assertions and not questions.
John M Reynolds
February 22nd, 2008 at 3:31 am
John M Reynolds — Is not Monkton a denialist? I opine you are obtaining (mis)information from wrong sources.
You write “Since the evidence has shown that the warming is not caused by anthropogenic greenhouse gas emissions, …” This is simply false, for the reasons given in my prior post addressed to you.
I doubt that Tamino, for example, would approve of your misinterpretaion and misuse of the temperature data.
February 22nd, 2008 at 4:05 am
John M Reynolds — It sems the site you linked to provides an analysis of ocean temperatures only. Scanning down the (interessting, some of them) comments one finds a link which I copy here.
http://www.cpc.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml
For the locations indicated in that data, rather substantial sea surface temperature (SST) variations have occured from 1950 onwards. So the SST record you linked to appears to be completely in line with other ENSO (La Nina-El Nino) variations. No surprises, nothing particularly special.
February 22nd, 2008 at 8:52 am
“It sems the site you linked to provides an analysis of ocean temperatures only.”
I provided two links. You did not specify which one, so I am ignoring that comment for now. That was a side issue anyway.
Let’s piece it together.
The UN’s 2007 climate assessment report showed how the atmosphere would warm or cool depending on the forcing. It compared radiative forcing from changes in solar activity, volcanic activity, anthropogenic green house gas emissions, tropospheric and stratospheric ozone, and pollutant sulphate aerosol particles. The models predicted that warming due to anthropogenic greenhouse gas emissions would have a specific fingerprint that was different than the other cases. If the warming was caused by man, then the temperature of the tropical troposphere would rise significantly. That is a ring around the globe from 30N to 30S wide. That was predicted by all five of the models used by the IPCC. That warming did not happen. Either the models used by the IPCC are wrong and cannot predict how we affect climatic temperatures, or the warming was not caused by anthropogenic greenhouse gas emissions. Either way, the models are not useful from a policy perspective.
John M Reynolds
February 22nd, 2008 at 11:03 am
John M Reynolds — That the warming is caused by anthropogenic global wamring (so-called greenhouse) gas emmisions is established beyond all reasonable doubt by
(1) atmospheric physics known for 150 years;
(2) radiocaron testing showing that the excess carbon is fossil.
End of that story.
We conclude that the models are ‘wrong’. However, I re–interate that this is admitted by climate modelers as yet another aspect of insufficient understanding of the physics of clouds.
That does not mean that the model results cannot be used to inform policy. In a previous post I gave some western side of the Americas examples.
I would not like to see policy makers treat regional climate forecasts as inviolable ‘ober dicta’. However, such forecasts, especially with ranges, ought to be of some use. For example,
http://climateprogress.org/2008/02/22/science-we-are-turning-the-west-into-a-desert/
is based on a recent article in Science. This agrees with Hadley Centre’s regional climate forecast for 2050 CE and the report in Science does not appear to depend upon it. Thus this is confirmatory.
You said “The models predicted that warming due to anthropogenic greenhouse gas emissions would have a specific fingerprint that was different than the other cases.” I doubt this. It is the sum total of all forcings, although admitedly difference ones apply at different altitudes.
February 22nd, 2008 at 1:41 pm
Hi Praj,
You wrote, “While David’s critique appears to be valid (i.e. Mark’s view of science is overly simplified),…”
In what way do you think my “view of science is overly simplified”?
I’ve written that the IPCC projections are not falsifiable, and therefore aren’t scientifically valid.
Is that what you think is “too simplified”? Could you elaborate?
Thanks,
Mark
February 22nd, 2008 at 5:35 pm
Hi Everyone.
Thanks for your detailed comments…especially yours Mark. Let me try to answer your last question. I’ll first respond generally why I think falsifiability cannot and should not be applied to all types of science. In a second post (if this one gets too long), I’ll give my very non-expert opinion on the IPCC.
So I take issue, as does David, with the idea that falsifiability can always be used to demarcate science from non-science. Falsifiability works very well when, e.g., you’re trying to determine if a given mixture is an acid or a base. If you hypothesize acid but the indicator says base, then you’ve falsified your hypothesis very easily.
But in systems science like climate science or evolution, the theories are based on cumulative evidence from many different fields. In evolution, facts from zoology, genetics, paleontology, radiometric dating, etc. add together to give a coherent picture. Because no single piece of evidence proves evolution, no single piece of evidence can disprove it. Evolution therefore does not falsify very easily. I think the same is true for climate science.
I think David hit the nail on the head when he said that science in practice does not always abide by a strict Popperian definition of falsifiability. Apparently even in general relativity, falsifiability does not necessarily apply.
In short, I (politely:)) disagree with your characterization that being unfalsifiable means something is not “scientifically valid.” Falsifiability can be a useful metric in some, but not all, areas of science.
I wrote you have an “overly simplified view of science,” because it seems you think that ALL of science can be easily described by a single idea. I think it’s really important to realize that science is not a single, monolithic concept. It changes from field to field. For certain questions, e.g., acid or base, the scientific method works as described in 8th grade textbooks. In other fields, the rules aren’t that straight forward. As a graduate student who studies space physics, I would argue that you cannot really experiment with the geophysical environment. When geophysicists use the word experiment, we mean something entirely different than someone in, e.g., particle physics.
If you accept that space physicists are scientists, you have to accept that being a scientist does not always require following the scientific method because we don’t really do controlled experiments. Just as all of science does not necessarily require the scientific method, all of science does not necessarily have to be falsifiable.
Mark: please let me know if I’ve misrepresented or misunderstood your argument.
February 23rd, 2008 at 12:04 am
“You said “The models predicted that warming due to anthropogenic greenhouse gas emissions would have a specific fingerprint that was different than the other cases.” I doubt this. It is the sum total of all forcings, although admitedly difference ones apply at different altitudes.”
So you doubt it then you admit there are differences? You seem confused. Here is the reference: (IPCC, 2007, p. 675, based on Santer et al, 2003. See also IPCC, 2007, Appendix 9C). Those charts show the warming signature of the atmosphere for 5 scenarios from 90N to 90S from ground level to about 30 km up.
John M Reynolds
February 23rd, 2008 at 5:16 am
John M Reynolds — Your citation to IPCC is insufficient for me to find what you found: The IPCC AR4 page as linked via Real Climate consists of 11 chapters as separate pdf files, some front and back matter, but no clearly identified appendices. One part of the front matter is a FAQ.
From IPCC FAQ 2.1: “Human activities contribute to climate change by causing
changes in Earth’s atmosphere in the amounts of greenhouse gas-
es, aerosols (small particles), and cloudiness. The largest known
contribution comes from the burning of fossil fuels, which releases
carbon dioxide gas to the atmosphere. Greenhouse gases and aero-
sols affect climate by altering incoming solar radiation and out-
going infrared (thermal) radiation that are part of Earth’s energy
balance. Changing the atmospheric abundance or properties of
hese gases and particles can lead to a warming or cooling of the
climate system. Since the start of the industrial era (about 1750),
he overall effect of human activities on climate has been a warm-
ng influence. The human impact on climate during this era greatly
exceeds that due to known changes in natural processes, such as
solar changes and volcanic eruptions.”
February 23rd, 2008 at 5:47 am
John M Reynolds — I happened to have the time to find your page reference in IPCC AR$ in Chapter 9. I looked on the previous and following page, but the only reference to Santer, et al., that I could find was to 1996 papers which appear to have nothing to do with the matter under discusssion.
I also note that IPCC AR4 does a climate model evaluation in Chapter 8.
February 23rd, 2008 at 7:23 am
John M Reynolds — Ok, I even found the reference to Santer et al. (2003). But I have no reference to a claim that tropical warming did not happen in approximately the manner described by IPCC AR4.
By the way, I checked the IPCC AR4 for regions I think I know something about: essential areement and the IPCC express quite a bit of uncertainty regarding their predictions. Modest of them.
I happen to know that a recent paper presented at the latest AGU desagrees with the IPCC AR4 regarding the Amazon basin. Show that the science progresses (at leat changes) faster than one can hope to summerize.
February 24th, 2008 at 12:28 pm
Hi Praj,
Thanks for all your comments. You write, “Mark: please let me know if I’ve misrepresented or misunderstood your argument.”
I think you’ve misunderstood my argument a bit. I think that’s mostly because I never got around to responding directly to the most important point of Roger’s post. Roger asked, “Are climate models falsifiable?”
And his answer was, “I am not sure.”
I don’t really agree that that’s the most important question. A more important question is, “Are the projections in the IPCC AR4 falsifiable?”
And an even more important question–for a guy like Roger, and for the public in general–is, “Are the projections in the IPCC AR4 useful for making good policies?”
I think the answer to the last question is that the projections in the IPCC AR4 clearly are not useful for making policies that have not already been decided. That’s because the projections don’t have estimated probabilities of occurrence for different scenarios. As I’ve pointed out, the IPCC projections without probabilities estimated essentially provides no estimate of the most likely warming in the 21st century, absent intervention by governments. It could be nearly 100 percent certain of more than 6 degrees Celsius warming, or nearly 100 percent certain of warming less than 1 degree Celsius.
But to go back to the less-important question of whether the IPCC AR4 projections are falsifiable…I maintain they aren’t. At least not in any practical sense. Roger has called the IPCC scenarios “conditional forecasts.” But conditional forecasts are only falsifiable if the condition(s) actually come to pass. For example, I can forecast that, if I win $10+ million in the lottery on a Tuesday, in a month starting with an “M,” the excitement will cause me to have a massive heart attack. Obviously, the conditions on which that forecast is made are almost certainly never to occur (especially since I’ve never played the lottery).
The IPCC scenarios are an analogous situation, although obviously less dramatic. Roger has compared warming since 1990 with warming under the A1F1 scenario. However, he apparently based this exclusively on CO2 emissions. The actual climate forcing of any scenario is based on emissions/atmospheric concentrations of ALL forcing agents, both positive (CO2, CH4, black carbon, N2O, etc.) and negative (SO2, organic carbon, and others). This give defenders of the models and scenarios an “out” if the CO2 emissions match the scenario, but the temperature increases do not. They can simply say, “Oh, well, if the projections for CO2 AND all the other forcings (CH4, black carbon, SO2, organic carbon, etc.) had been accurate, the actual temperatures would have been exactly as projected.”
There’s a way this could be addressed, if the IPCC was really interested in doing science. (I think it’s clear they are not.) The solution would be to isolate each of the individual forcings, and to provide estimates of the most probable (i.e., 50 percent probability the value will be lower, 50 percent probability the value will be higher) values for each of the forcings (as well as estimates of the spread). In other words, develop probability density functions for each of the forcings separately. That way, there would be a single scenario that had the most probable forcing for each of the variables. And then update the probability density functions during each six-year analysis interval. In essence, this would do away with the “scenario” situation entirely…because the scenario situation simply isn’t scientific.
Tom Wigley and Sarah Raper had a paper in Science in July 2001 that at least improved on the Third Assessment Report (and in fact was ***better*** than the Fourth Assessment Report…a clear sign of pathological science that something produced 6 years earlier was even better than the AR4). In the Wigley and Raper paper, it was assumed that all the scenarios were equally likely, and a most probable warming from 1990 to 2100 of 3.06 deg Celsius was calculated.
Mark
March 2nd, 2008 at 10:09 pm
Roger, I agree with you that the climate models themselves must be seen to be of exploratory or heuristic value, rather than predictive – since the models themselves are unable to determine their inputs. Rather, it is people who take guesses as to what various inputs might be, or are likely to be if certain trends are assumed.
But as the models are of course simply models of a very complex climate systems, they simply cannot be perfect and it must be possible to test them against actual changes to climate to see how well their understandings of the linkages between various aspects of climate correspond to actual changes in claimte conditions.
I would presume that this process of testing and improving the models goes on all the time, so I am curious how you conclude that “issues of falsification are simply ignored or avoided.”
It also seems to me that your conclusion – that the “”consistent with” game being played with climate models by activist scientists … is every bit as misleading as the worst arguments offered by climate skeptics and a distraction from the challenge of effective policy making on climate change” – is stronger than can be justified, or at least that you have not shown us the basis on which you ground such a conclusion.
Regards,
Tom
March 5th, 2008 at 8:24 am
Here is a link to an article that discusses the difficulty with falsifying:
http://rankexploits.com/musings/2008/falsifying-is-hard-to-do-β-error-and-climate-change/
John M Reynolds
March 5th, 2008 at 3:28 pm
John M. Reynolds — Thanks for posting the link. Over there I posted ‘These questions provide examples where the Bayesian factor method is likely to lead to sharper results:
E.T. Jaynes
“Probability Theory: the logic of science”‘