it (the debate between AGW proponents and skeptics linked by Lubos - P) swung against them, particularly when Schmidt made the fatal debating error of dismissing the ability of the audience to judge the scientific nuances.
Gavin Schmidt told the audience that they were too stupid to understand what he had to say, when it was his job to make himself understood.
The reality is this prick is too stupid to explain AGW to a willing audience. And he is the cream of RealClimate.org!
I hope that someday the warming crowd puts Gavin on television to explain AGW science to America.

M. Simon,

Nearly all of the points made in Motl's article are the very same one's discussed at length (and refuted) in the threads here at Classical Values (http://www.classicalvalues.com/archives/2007/05/it_is_uncertain.html#comments)
(http://www.classicalvalues.com/archives/2007/05/clouds_in_chamb.html#comments)
In fact, the only thing I think hadn't been discussed is the sunspot cycle. But that's easily dismissed: Sunspots in themselves don't affect climate, but have to act through their impact on luminosity or (conceivably) cosmic-rays. But as already discussed at length, variations in luminosity and cosmic rays fail to come anywhere close to being an explanation for the GW of the last 100 years. Hence, the sunspot connection is a red herring.

Papertiger,

Gavin Schmidt is a good scientist, but not a good PR guy. Most scientists are not that good at debate, they're better at discussion, which is quite a different matter.

Neal
Have you seen Prof. Lindzen or Prof Feynman speak? It seems that a firm grasp of the topic is the main criteria for a powerful public speaker regardless of their day job.
Arrogance and condescension do not make persuasive argument, as Gavin found out (a lesson you could stand a little of yourself, Neal -ie - Refuted acording to whom?).
Actually, from his perceived reasons for failure in the debate (Micheal Creighton's height?), perhaps Gavin didn't discover anything of value from the experience.
His demeanor at Real Climate helps the skeptic cause. More power to him.

Papertiger,

- I haven't heard Lindzen, but I saw Feynman on a weekly basis for one academic year, as well as at a few classes. One was on his favorite subject, quantum electrodynamics. My point was not that NO scientists are good speakers, but that most are not. There are lots of physicists who had Feynman's respect for their knowledge and insight, but who would never have been able to debate anything effectively in a public forum: debating ability is a separate skill. That ability alone says nothing about whether one is right or wrong.

- Schmidt's better at writing than speaking. If you want to argue techie at him, try your luck at www.realclimate.org. He will blow you out of the water.

- If you don't think I refuted those arguments, go back to the threads and pick up the fight. When you and others have presented arguments, and I've answered them, and there's no further response from your side - my interpretation is that you've lost.

I've answered them, and there's no further response from your side - my interpretation is that you've lost

This particular self-serving fallacy was part of a classic Protein Wisdom thread a while back. Shame I can't find it now.

Tim Maguire,

The point remains: The purpose of a discussion is to present facts and arguments. If someone puts up an argument, and I don't respond, a reader is entitled to think that I can't dispute the point.

The reverse is also fair.

What would be the alternative? Assume that whoever leaves the argument first is the winner?

Assume that whoever says, "I have the perfect rebuttal but I can't find it" is automatically correct?

The point of discussion is to come to a resolution of differences, not through "proof by assertion" but through following out the logical conclusions; responding to new points and to new information.

Neal, remind me: what was your rebuttal to my objection that the IPCC's use of computer models is treating speculation as fact? As I recall you thought I was against statistical modelling as such; you never demonstrated an understanding of the true difficulty.

And that's the alternate explanation for a failure to respond, by the way. You may be given the last word, not because your argument was persuasive, but because your opponent sees that you can't or won't listen to his arguments. For an extreme example: I once saw a cartoon in which H. Ross Perot (remember him?) is shown as saying "What would my opponent do if government agents invaded his daughter's wedding?" Then, after a long pause: "See, they can't answer that." (Perot's actual debate performance was nearly that bad, but not quite ...)

Getting the last word doesn't mean you're right -- it may just mean you're stubborn enough not to quit when you ought to.

Michael Brazier:

- Rebuttal to your objection that the IPCC's use of computer models is treating speculation as fact: I may have missed it in dealing with your other points; if you can remember where you stated it, I can check if I said anything at the time.

However, just going with the question as you've stated it: The models are based upon well-known physical principles, like conservation of energy, radiative transfer, gas laws, etc. They have been pretty successful at predicting changes in the last few years: Hansen's model of 1988 had three scenarios, one of which he denoted as being "most likely", and it has fit pretty well (to within 10%) of the measurements through at least 2003. And this model is still being used, although it is very likely to have received several updates in 20 years. There are also other models: The IPCC consulted about 13 independent and competitive models, each with somewhat different approaches to modeling global climate, to bracket their expectations on climate behavior under the various scenarios.

Does that mean that the prediction is "fact"? No, no more than a casino's prediction of its payoff is "fact". They are reasonable statistical predictions, based on the one case on pure statistics, and on the other with an admixture of physical principles, plus projections as to the human input of C-O2. And they are a lot better than "flying blind". Or do you prefer to go without glasses, because they might not give you 20/20 vision?

- wrt to stubbornness: Are you claiming that I'm not responding to your arguments appropriately? M. Simon might claim that about one of the threads (http://www.classicalvalues.com/archives/2007/05/the_solar_conve.html), but I claim that he was missing my point. Honestly, I cannot tell whether I got through to him or whether he gave up in frustration. All I can say, for my part, is that I kept varying the argument and the wording to clarify better the issue that I was raising, and eventually he stopped objecting (and it was the same objection, again and again: residuals). There's nothing much more that I can do than that. I wasn't browbeating him, I wasn't saying he was stupid. So what other conclusion can I draw, other than that he eventually got it? If not, it's up to him to vary his language, so I can better understand what his point was. Discussion is a 2-way street.

You're welcome to go over that argument, and explain where you think I've gone wrong. Probably best to do it in that thread, however.

Neal I have no idea which points you claim to have refuted.
How about you take your own advise and go refute Molt at the refernce frame?
And if the last guy to open his mouth is the "winner", I'll just bide my time until you tire then pop in and post a quick "No it's not." and win everytime.
Of course that is stupid and pointless.
Much like your arithmatic in that previous thread regarding semi majoraxis and the effect of same on the weather of superior planets.

Papertiger,

- Basically, unless I've said that I conceded a point, I claim all of them.

- wrt the orbital variation: It's true. If the effect is bigger, the cause has to be bigger. I'm sorry you don't understand that point. M. Simons was (is?) confused because he thought I was referring to the residuals, whereas I was actually referring to the absolute values. You weren't saying very much, so I couldn't tell what you understood. Have you had freshman physics? 1st-year calculus? If so, I might be able to explain more to you; if not, it might be kind of hard.

But I'm willing to try, if you are.

- wrt Motl: I don't think it's really worthwhile talking to him. There was a report that he took a dislike to a book someone had written (it was critical of string theory) and he offered money to people to write bad reviews of the book on amazon.com. Huh? No, thanks.

Papertiger,

To clarify: If I have addressed a point, it was because I thought I had something to say about it. If what I said was contradictory to someone's point, then I claim a refutation, unless there was a defense.

I can't claim to have refuted a point that I didn't address. Naturally.

There you go again.
I'm sorry you don't understand that point. M. Simons was (is?) confused because he thought I was referring to the residuals, whereas I was actually referring to the absolute values. You weren't saying very much, so I couldn't tell what you understood. Have you had freshman physics? 1st-year calculus? If so, I might be able to explain more to you; if not, it might be kind of hard.
Your arrogance and condescension are showing.
Who the hell are you to lecture anybody? You don't even understand Kepler's second law.

Papertiger,

If you've had 1st-year physics and calculus, I can explain the point to you.

If you haven't, it might be hard.

And I am quite familiar with orbits. It is you who are over-interpreting the implications of Kepler's law. It doesn't change the fact that when you're nearer the Sun, you're going to feel hotter.

Neal: let's take a closer look at the statistical predictions of a casino, against those of the IPCC. The casino, predicting the payoff from the games it runs, knows the rules of every game, and can calculate the expected outcome of each play from that knowledge. (Indeed, the casino makes the rules for all its games, and won't run any game unless the odds favor the house.) The IPCC, by contrast, does not know the rules the Earth's climate follows -- those well-known physical principles place limits on the space of possible dynamics, but they don't force one unique possibility. So, a casino can state as a fact that, say, 1/19 of the money bet on a roulette wheel will go to the house, within some margin of error; but the IPCC cannot state any such facts about the Earth's climate. Each of the 13 models the IPCC consulted represents a guess about the rules of the game; each guess agrees with the record we have of the Earth's climate, to the IPCC's satisfaction; but uncertainties in the record are large enough that fitting a guess to it is not a strain on anyone's ingenuity.

The IPCC admits as much in this quote from the Summary for Policymakers (section 8.2.1.3): "The climate system includes a variety of physical processes, such as cloud processes, radiative processes and boundary-layer processes, which interact with each other on many temporal and spatial scales. Due to the limited resolutions of the models, many of these processes are not resolved adequately by the model grid and must therefore be parametrized. The differences between parametrizations are an important reason why climate model results differ." The authors are speaking directly of approximation errors, but there is also the question of missing physical processes -- missing from the models, that is, with their real contribution to the climate being swallowed up by parametrizations in the models. The variety of physical processes involved, and the doubt whether all the relevant ones are accounted for, is the reason why I think the IPCC's confidence in climate modelling isn't justified. Every process added to a model adds one more free variable, and one more source of statistical errors; hence every new process added makes fitting to the observations easier, which should decrease confidence in the model as a whole.

Michael Brazier,

Their are two types of uncertainties that they considered:
- statistical uncertainties
- conceptual uncertainties: missing variables

In their overall assessment, they did consider both, and their "robust" conclusions are the ones that withstand both levels of assessment: the term "very likely" means that they give it 90% probability, even taking all these into account. The fact remains that these are the best models we have, done by the people who are most knowledgeable about the subject, and who are most interested in getting at the truth. The uncertainties that exist could go either way: there is no reason to expect the universe to treat us unreasonably well, either.

Neal: I'll be short here because the conversation has moved on without me--no, I'm not saying the first person to not respond wins. I'm saying you can assume nothing regarding who wins and loses based on who stops responding sooner.

There are many reasons a person will neglect to respond--they may have a life in the real world, they may think the discussion unworthy of the time it's taking, they may get bored and go do something else, etc. They have no obligation to stick around for resolution.

You can never assume you "won". Unless someone admits defeat or there is some kind of vote awarding a winner, the conversation simply ends.

tim maguire,

In these exchanges, it is often the case that there is little hope of effecting a conversion to the other side. The real point, if there is a point beyond amusing oneself, is the hope that one's presentation has had some effect on readers who may never choose to comment, but may find a point or two that gives a new insight or a new perspective.

From my point of view, the goal is to keep in mind this silent reader and to direct my argumentation thence as well as to my immediate interlocutor. My intent, at any time, is to respond in such a way that, should my response go unanswered, that the reader will have understood the key points I was trying to make.

Here here Tom! I enthusiasticaly wholeheartedly agree.

What say we establish a peer review process before someone starts claiming victory?

Neal you can even do it like the climate scientists. Go drag Froblyx back in here to vote for you.

Papertiger,

I'm not interested in votes, and in any case I don't have any contact with whoever Froblyx is. I'm just going to conduct my end of the discussion as I think appropriate.

Neal: let's take an example of conceptual error, simulating planetary orbits with Newton's laws. The accuracy of Newton's laws is about 1e-7; this means that, as long as the relative error in my measurements exceeds 1e-7, I don't need to worry about relativity effects. On the other hand, if I need more precision than that, I must use Einstein's theory instead.

When the IPCC says it's "very likely" that their climate models are accurate, what is that supposed to mean? It doesn't mean a bound on conceptual error -- that would be saying, if error in measurements of climate data exceeds 10%, we can ignore any conceptual errors in the models! For global temperature, the change we're trying to measure is less than 0.1% of the base value. If "very likely" is a bound on conceptual error, then the noise in the system is swamping any signal, and we have no clue what the future climate will be. The IPCC must have some other notion of probability in mind ... but what is it, and does it make physical sense?

"The fact remains that these are the best models we have, done by the people who are most knowledgeable about the subject, and who are most interested in getting at the truth."

This is an appeal to expertise. My objection is, however, a doubt that expertise in this subject is possible: I do not believe that a provably accurate model of the Earth's climate can be built with current technology, and I don't believe that any of the IPCC's models are provably accurate. As all the IPCC's conclusions depend on the accuracy of the climate models, none of them are proven, and none of them justify changes in policy.

To this objection, the assertion that the IPCC are the world's experts is no reply. If mermaids existed, you would expect to find them in the ocean; if experts on climate change exist, you would expect to find them at the IPCC. But pointing to the ocean doesn't prove there are mermaids, and pointing to the IPCC doesn't prove there are experts on climate change.

Michael Brazier,

- "Very likely": The 90% is a seat-of-the-pants statement. You are clearly right when you say that a sense of conceptual certainty cannot be calculated. But there is a difference between an expert saying that something is 60% certain, 80% certain, 90% certain, and 99% certain. The point of the convention chosen by the IPCC was to establish a rule for their lead authors: "When you say something is 'very likely', you better be pretty darn sure. And when you say 'virtually certain', you better be even more sure." Because they have different lead authors for the different chapters, and they need to keep the use of language reasonably consistent, in order to be understood by readers properly.

- "Experts": By any measure, the IPCC is the expression of the experts. It is staffed (part-time) by 1500 of the world's leading scientists on the atmosphere and the climate. They are respected professionals working full-time on the issue. They talk with each other, they compete with each, they know what's happening with the latest research. They are not there to sell another book, or get a check from the Competitive Enterprise Institute: they get much smaller monthly checks to do the actual work of simulations, and update them with the latest findings. They are the experts. If your life depended on knowing what was going to happen with climate, these are the guys that you would go to. If you went to Singer, or McKitrick, or whatever, it would be a bit foolish - because they don't even pretend to have a complete view of the atmosphere. They aren't even trying.

That said, does that mean that they're absolutely going to be correct? No. As Feynman once said, "Science is the belief in the ignorance of experts." But that doesn't mean that experts are not experts, or that everyone who contradicts the experts is likely to be right. Becasuse he also pointed out that, in most cases, the conventional point of view (which is to say, the view held by the experts) is probably right.

Nope.

This is going to be like IPCC Climate Science. I choose the papers. I decide who is credible. I decide what the concensus is. I have my mind made up in advance.

So now we have dueling Comittees. The IPCC only has Climate Science.

I have Climate Physics with String Theory. Much more robust.

http://powerandcontrol.blogspot.com/2007/05/lubos-motl-looks-at-sun-spots.html

Heh.

Oh, golly.

Sunspots in themselves don't affect climate

That seems fairly counterintuitive to be stated as a postulate.

However, there are many things which correlate very well to temperature. I believe it has even been proved conclusively that George Foreman has caused Global Warming.

On a slightly more serious note, I often wonder whether the working out of models covers a dirty little secret: in the math of the modeling there are of necessity sets of nonlinear differential equations, which often do not converge for given inputs. Perhaps we are only treated to papers on the ones that do.

Neal,

Go to the climate scientists. They say likely. They say that means 65% chance they are right. I wrote about this here a while back. I believe you even commented on that thread.

The esitmated 100 year signal is 2 deg F.

The estimated noise in our pre-satellite record is 1 deg F. So out of 1 deg F noise and a perfect model they estimate a 2 deg F signal. It could happen . Except that te model range is 2 deg F to 6 deg F (roughly) why the range? Well the models have different designs due to the well understood science.

Why? The truth is that when I want to model DC motors there are thousands of models to chose from with varying equations and choices of constants and interactions.

Well I lied. There is only one model for DC motors. (for AC motors there are several models depending on the reference frame).

Why is there only one model? Well you know. We actually understand the subject.

Simon
In the Cloud post you claimed that the lack of clouds due to the sun attenuating the Cosmic ray flux accounted for .5C (+/- .2C as the error margin).
If you add in the direct solance increase per decade, then there is no room left, or need for, GHG as an explanation to get a .6 degree temperature rise.

I think you found the holy grail.

Neal I like the system as it is also. Who needs votes of confidance? In science it only takes one to prove you wrong,
and "concensus" has grown distasteful to me.

M. Simon:

- Motl: His posting covers the same material and references as this one does. The rebuttals are the same. And string theory is utterly irrelevant to global warming (and some physicists are beginning to think, to real particle physics as well!)

- Models: There are many models because they do emphasize different aspects of the climate problem. The world is, after all, the most complicated thing in the world. But the robust results are the ones that survive all the different reasonable choices of parameter values, over the set of models.

(And they define a range of terms: likely, very likely, etc. Each has a specific intended meaning.)

Papertiger,

As I stated in comments to the "Clouds" posting, unfortunately there is no reason to believe that the proposed explanation by cosmic-ray flux works. However, can we leave that issue to that thread? It's not going to be much fun chasing arguments back & forth between threads.

(At this moment, as indicated there, I am in-progress with my response to your presentation there. A thoughtful comment deserves a thoughtful reply...)

Socrates, on Sunspots:

What I mean is that the Earth does not look at the Sun, see spots, and say to itself, "Omigod, sunspots! I feel so warm!" Instead, there has to be a physical connection between the sunspots and the heating in order for there to be a relationship.

One that has been proposed has been a relationship between sunspots and solar luminosity.

Another has been the possibility that sunspots affect cosmic-ray fluxes which in turn would affect clouding.

Now, both of these are possible explanations. But, for reasons discussed above and in the "Clouds" thread, they haven't panned out.

(As suggested above, I personally would prefer it if detailed discussion of the "Cloud"-related issues be held in the "Cloud" posting at: here. That will keep things a little saner.)

Neal: "The 90% is a seat-of-the-pants statement."

Yup, just as I thought: the percentages attached to the degrees of likelihood in the Summary for Policymakers have no mathematical or physical basis. If a lead author says X is "very likely", that means at most that the author is prepared to put up $9 that X will happen against my $1 that it won't. What a pity that the IPCC isn't actually making bets with its own money.

Michael Brazier:

Actually, there have been many offers to take bets on GW.

The general pattern has been:

- A skeptical scientist claims that he believes that it is more likely than not that the GW trend will reverse in X years.

- Another scientist offers to bet him even money on that claim.

- The skeptical scientist changes his mind. In the case of Lindzen, he wanted something like a 5-to-1 advantage in payoff before being willing to take the bet. Kind of makes you think that he lacks confidence in his own position, don't it? What it basically tells me is that he thinks he has a one-in-six chance of being right with his stated position.

- I have heard of one bet that has actually been accepted, at even money: Two Russian solar physicists vs. one Canadian climatologist living in Japan. I think the decision point is in 10 or 15 years.

Neal, it's nice to know the climatologists are actually prepared to back their opinions with their own money; but the point remains, that the statement "it is very likely that the climate models are accurate" is on the same intellectual level as tips on horse races. A scientific prediction it's not.

Michael Brazier:

Well, from that point of view, it's not possible to say anything about the future. Even the statement, "The sun will rise tomorrow," could be wrong: Some invisible asteroid could suddenly smash into the Earth and destroy it utterly.

Is that probable? No.

And in the same way, it is possible for the world's climatologists to come to a conclusion about what is likely (65%) or very likely (90%) or virtually certain (> 99%).

They have a fairly strict terminology, by the way: See http://ipcc-wg1.ucar.edu/wg1/Report/AR4_UncertaintyGuidanceNote.pdf

Neal: The likelihood that an asteroid nobody has observed is less than 24 hours away from colliding with the Earth can be bounded by scientific reasoning from our observations of the sky. I can't give you the bound off the top of my head, but I'm confident it could be done, and that the bound would be extremely small. I am equally confident that similarly small bounds can be placed on any other cosmic disaster that would keep the Sun from rising tomorrow. So "the Sun will rise tomorrow" qualifies as a scientific prediction.

The point is that "likely", "very likely" and "virtually certain", as used in the IPCC Summary, are not chosen by scientific reasoning of this kind. The probabilities attached to those qualifiers have no objective meaning; the qualifiers express no more than the personal judgement of the authors, the "experts' opinion". And in this field, I repeat, I doubt that any experts exist.

I read the Note you referred to. The Summary breaks all the rules laid down in that Note; none of the judgements of likelihood in the Summary are supported by evidence of the kinds the Note calls for. So it's still on the level of tips on horse races.

MB:

a) If you asked an astronomer or an astrophysicist about the probability of a [i]known[/i] process leading to the result suggested, you could get an estimate. But it's much more likely that they would say, "Look, so many silly coincidences would have to happen at the same time that it's not even worth thinking about. Just forget about it."

b) If you ask the same folks about the possibility of an [i]unknown[/i] process leading to the end of the world, you'll get a different sort of response. That will go to the basic question: How well do we really think we understand the major forces and kinds of matter in the universe? How likely is it that we're missing something that big? And that speaks to their basic sense of where the holes may or may not be in our current understanding.

c) The numerical probabilities that can be calculated from first principles are like a), and climatologists can do that as well as astrophysicists.

d) wrt conceptual issues, like b), I think that climatologists are in a better situation than astrophysicists. The basic physics they are working with is much less subject to revision, so there should be fewer rabbits in the hat. Of course, there is the complexity; on the other hand, they know about the complexity. That doesn't mean they can predict how everything will turn out, but they should have some idea, again, of where the weak points are.

Science is ultimately a human endeavor. At a very basic level, there is some intuition involved - and some folks are just better at that than others. The folks leading the discussion at the IPCC meetings have been leaders in their fields for some time. As an aggregate, I think they deserve more trust than Singer, Stott & (especially) Crichton. If I had to bet my future on it, I'd take their advice, not S-S&C's.

And, wrt to the Summary: The Summary is a summary intended to bottom-line the entire report. The evidence in the roughly 1000 thousand pages of the technical report. Think of it as an extended abstract: You don't burden the abstract with all the reasoning, that's what the paper is for. The abstract is intended just to boil it down to the message.

So that comes down to the question.

What is the optimum temperature?

Who decides?

Suppose the temperature is "too low" due to inadequate mixing of the oceans caused by continental drift? Do we have to eliminate Central America?

Suppose the next ice age starts in 10 years. What do we do then?

BTW error bands start with measurement error. Then there is theory error. Then because we have slow computers - parameterization errors. There should be bounds for these. We know right off the bat that the cloud factor runs from about -2 to +8 depending on the model. The cloud factor estimate is probably the largest factor in variations in the models. So who has the right number?

Measurement error of course will tend to cancel out if you have enough measurements of the same thing measured under identical conditions. Don't forget we are trying to tease out a 1 deg C signal in 50 deg C of variation.

With all these unknowns and left outs you get a lot of degrees of freedom. The more freedom you have in assigning parameters the easier it is to predict the past. However, past performance is no guarentee of future results.

As to the left outs - UV radiation's effect on the upper atmosphere is too hard to model. Since it is unmodeled it will be parameterized. Was the choice of parameters correct? How sure are we?

You wouldn't trust 6 people's lives to a rocket that was so badly modeled.

And yet there seem to be a lot of folks who want to control the destinies of billions based on such models. It is going to be a hard sell.

M. Simon,

- The question of "optimum temperature" is utterly irrelevant. Whether the climate is getting hotter or cooler, the issue is the rate at which it is changing. If we look at the possible scenarios, we could be talking 5 degrees Celsius of change in under 200 years. In previous times, it would take many many thousands of years for this kind of change to be effected. Considering that it takes about a million years for a new species to develop, what all this implies is that we are setting up for a major die-out, where many of our flora & fauna disappear without leaving any descendants. Even from the utilitarian perspective of medicine, this is a rather bad thing: most of our medicines come from someone discovering that a strange plant has bizarre property (even aspirin fits into this category). And from a personal point of view, I would (and probably will) miss the fantastic variety of life in the coral reefs. But from a more theoretical - one might even say, engineering, perspective - there is the issue of system robustness. The more diversity we have in the biosphere, the more likely it is to be able to cope with the unexpected. This could be another giant asteroid, for instance. Whereas if the range of genetic diversity is impoverished, the system is more susceptible to being overwhelmed by a sudden event. An analogy could be made to having a single point of failure: always a bad idea in design.

- "What if the ice age starts in 10 years?" Actually, we are supposed to be in an ice age now. According to schedule it should get really cold - in about 60,000 years.

And that's the point: The changes that have hitherto controlled the climate have either been very slow - like the coming ice age - or catastrophic - like the asteroid that wiped out the dinosaurs.

And the other part of the problem is that we don't have the luxury to sit around and think about it. Climatic changes are happening, and the causes of more future change are building up. It is not the case that we are sitting on top of a hill wondering whether to start the toboggan down the slope. It is rather that we are already sliding down the slope, and we need to decide which direction in which to steer. A decision not to make a decision is, regrettably, still a decision: the hill doesn't care whatever way we go or where we end up. But, as riders on the sled, we probably do.

"d) wrt conceptual issues, like b), I think that climatologists are in a better situation than astrophysicists. The basic physics they are working with is much less subject to revision, so there should be fewer rabbits in the hat."

As a physics problem, the movement of bodies in the solar system was solved completely when Einstein explained the precession of Mercury's orbit, and that was ninety years ago. The basic physics of gravitation has not been revised since 1916.

The basic physics of Earth's climate, on the other hand, is the absorption spectra of the constituents of Earth's atmosphere -- and our knowledge of those spectra has been revised, significantly, within the past five years. So in fact, climatologists are much worse off than astrophysicists when it comes to the fundamentals of their respective subjects.

"The Summary is a summary intended to bottom-line the entire report. The evidence in the roughly 1000 thousand pages of the technical report."

Which is exactly why the Summary should not have been published before the technical report. (Come to think of it, has the technical report been published yet?)

"It is not the case that we are sitting on top of a hill wondering whether to start the toboggan down the slope. It is rather that we are already sliding down the slope, and we need to decide which direction in which to steer."

Wrong again -- we are on a boat in the middle of a river, flowing with a current in the dark. Mr. IPCC claims to be an expert river pilot, although when pressed he admits he's never been on this specific stretch of river and doesn't know anyone else who has. He is, however, absolutely sure the boat should go that way -- oh, and the engine should be turned off. The boat's owner is skeptical; he doesn't know the river either, but he does understand the boat, and he knows that with the engine off it steers like an elephant. And there's no light anywhere to steer by ...

In a situation like this, there's no luxury to do anything but sit still and think, or rather observe. Choosing a route when you don't have a map or a light is sheer recklessness; the prudent thing is to make a light or wait for sunrise. (We have good reason to think, after all, that the sun will rise tomorrow.)

Michael Brazier:

"Changes to our knowledge of the absorption lines in the atmosphere"
I'm not aware of any updates to this information. Would you cite a specific URL where that can be examined?

"No revision to gravitational theory since 1916"
Until quite a bit more recently than that, rivals to general relativity have been seriously entertained by professionals in the field. The Brans-Dicke theory and various of the parametrized post-Newtonian have always been in the background during the experimental tests of GR. Although GR has always been considered the leading dog, the professionals know that these alternatives have to be taken seriously.

In any case, GR has few practical implications for everyday life (small corrections needed to GPS calculations). Most of what affect us about gravity can be calculated following Newton. The area that is of more interest would be, for example, stellar structure.

There is a long story associated with the career of a star that is part of what every astronomer grows up learning. According to the story, the determinating factor for a star is how much mass it starts out with: everything hinges on that.

When I was a teaching assistant in the physics department, on loan to the astronomy department, the professor of the course admitted to the students that this entire picture was based upon a rather complex program written by a graduate student. When he got his PhD and went on to other topics, no one inherited his responsibility for this program. So, at that time, the professor confided, people pretty much believed the story, 'cause the grad student was considered pretty smart. But if you asked anyone to reproduce the results, there would be a few years of work needed before anyone would be in any position state that they'd verified it.

The Summary should not have been published without the Technical Report
- You're getting off-topic: As long as we're talking about what justifies the statements made, and their relationship to numbers, we're talking about the Technical Report.
- The actual report has been available since April. Be my guest: http://ipcc-wg1.ucar.edu/

The Climate is Changing

Even the skeptics have admitted to that: Singer, Stott and (ugh) Crichton, when asked point blank at a debate, agreed to that. As stated before, a decision not to decide is still a decision.

In the meantime, please consult the IPCC report (which is constructed from all the peer-reviewed science on climate published in the last 6 years) to see what really is known about the situation. I think the professionals know more than what you can find on a few websites.

Neal: See this article from the May 2003 issue of Physics World: "Experiments that were performed by Roland Schermaul and the late Richard Learner at Imperial College in London in 2001 have cast previous measurements of the absorption spectrum of water into considerable doubt." Granted, that's not a peer-reviewed scientific journal, but it's a report of significant changes to the basic physics of climatology.

"a decision not to decide is still a decision."

If you see ten troubles coming down the road, you can be sure that nine will run into the ditch before they reach you. -- Calvin Coolidge

Michael Brazier:

Absorption in Water Vapor
It's kind of interesting, but notice what the article DOESN'T say.
- It doesn't say that this affects the 1500 micron region. That would be very interesting, because that is the region in which C-O2 absorbs but H2O does not. That region is C-O2's "claim to fame" in the greenhouse effect: without that uniqueness, C-O2, as a trace gas, would indeed be swamped by H20.
- Unless one thinks that this newly measured absorption reflects a sudden change in the quantum mechanical characteristics of the water molecule, this has no implications for what has been seen to date of GW. Whatever H20 has been doing wrt GW, it will continue to do. It will not change its effectiveness one way or other because of an experiment in a lab. And so this new measurement cannot have any explanatory value for the increase in global average temperature over the last 100 years.
- Finally, I note that these measurements were done in 2001, and publicized in this article in 2003. Whatever implications it might have for GW will certainly have been incorporated into the IPCC report, which finalized input in mid-2006. There are teams of people who pore over that thing, looking for any hole: Fraser Institute, for example. They would have caught it. They didn't say anything about it: I looked at their critique a couple of months ago.

MB:

It's the 1 out of 10 that kills you

Murphy's Law: The buttered side of the toast is always the one that hits the carpet.

Neal King:

First, you ought to know better: any wavelength on which either the Sun or the Earth radiates is relevant to Earth's climate. Looking only at the bands where CO2 absorbs strongly, because that's the bit you understand, is like looking for lost keys only under a streetlight because that's the part you can see.

Further, if you don't know water's absorption spectrum, you can't separate its effect on the climate from the effect of CO2. The observed data includes effects from every constituent of the atmosphere -- it doesn't come with convenient labels saying "this much warming is due to water, that much to CO2"!

Finally, you've lost sight of the reason I mentioned this. I never said the IPCC left these experiments out of their last report. I said, rather, that these experiments show that our knowledge of the basic physics underlying climate studies is not stable. New and surprising events turn up all the time. To declare that now we know enough to act on -- that is the truly reckless course.

MB:

- The reason that I focus on only the C-O2 bands is not because of the streetlamp issue. It is because the other bands don't contribute to the C-O2 problem. If a band is absorbed by both C-O2 and water, then it would be true that even a substantial variation in what is, after all, only a trace gas, could have much effect on the radiative balance for that band. However, for a band that is not absorbed by water vapor but is absorbed by C-O2, C-O2 is the only player. And if, as in the case of the 1500 micron band, it gives rise to 3.8 Watts/m^2 radiative forcing, it has to be taken into account.

This does not amount to ignoring the effect of water vapor. But the only thing that affects the water vapor levels is, in fact, increasing the global average temperature. That will increase the water vapor. Virtually anything one can think of will not, because of equilibrium between the oceans and the atmosphere.

And my broader point is that, while interesting, the framework of climate science has not been shaken by this new information. Science is always in the business of discovering something new, we can't put all of our decisions to a far-off date in the distant future when, somehow, it will be known for sure that everything is known.

The part you are stubbornly failing to acknowledge is that if our top 99% best scientists on this subject are right (that's right, I mean the top 99 out of 100), we are already acting to affect the climate. The attitude of "wait & see" is analogous to leaving the car on the track even when you can hear the train whistle.

"The reason that I focus on only the C-O2 bands is not because of the streetlamp issue. It is because the other bands don't contribute to the C-O2 problem."

Neal, Neal ... CO2 is a "problem" only because it affects the Earth's temperature, and all the other stuff in the Earth's atmosphere affects the Earth's temperature, too. There is no good scientific reason to concentrate on the peaks in CO2's absorption spectrum and neglect every other wavelength; that amounts to deciding, before looking at the data, that CO2 absorption is the cause of all the climate change in the past 200 years. That is, it's assuming what you set out to prove. I'm not going to claim the IPCC made this mistake, but you certainly have.

"The part you are stubbornly failing to acknowledge is that if our top 99% best scientists on this subject are right ..."

What I am stubbornly not acknowledging is the proposition that 99% of our top scientists agreeing that something is true is a proof. I don't know if you followed the debates over string theory in the field of particle physics, but it would be fair to say that around 1990 nearly all particle physicists were convinced that string theory was leading to a complete description of the universe. It is also fair to say that string theory hasn't actually led to any such description, and that as it stands now the hopes for it back then appear vain. There was a lot of talk about "scientific consensus" in string theory's heyday, and a good bit of pressure on skeptics not to raise objections; it was very like the "consensus" of climatologists today. Yet the consensus for string theory was wrong. Why should I accept a consensus in climatology?

MB,

- It is accepted, even by today's skeptics, that GW is happening (they've given up saying it isn't). When looking for explanations for this, water vapor cannot be a candidate cause, because there are no trends for water vapor concentration: something that does not change cannot be causing a change. Therefore, IR bands that are controlled by water vapor concentration cannot be contributing to the GW we are trying to explain, because GW is a change.

- Yes, I'm quite aware of the situation in particle physics and string theory. It is NOT fair to say that nearly all particle theorists thought string theory was the coming thing. A large number did, and they happened to be the ones that controlled hiring for particle theory. But there were always a substantial number of particle theorists that thought the whole thing rather speculative. The seductive aspect of it was that there was so much good and exciting mathematics in the theory that people involved could get a whole lot of fun out of it, regardless of how difficult it might be to connect it to anything measurable. And the non-theorists and physicists in different fields didn't think very much about it at all.

The story with the theory of the enhanced greenhouse effect is rather different: It is a part of basic textbook studies for climatology, regardless of specialty. If it doesn't work, a whole lot of basic understanding has to be reworked. And, by the way, there is no scientific challenge to it. No one has pointed out any problems with the theory: there's no scientific crisis associated with it, everything works pretty well. There's only a political crisis, because some people don't like the implications and would rather hope that the science is wrong.

Personally, I would be over-joyed if it were wrong. But the more carefully I think about the science, the more it hangs together.

And when I read critiques, such as the one by Motl, I see how he has completely failed to understand the theory he's criticizing: He has absolutely the wrong understanding of how the enhanced greenhouse effect is supposed to be working, so how can he critique it properly? Instead, he's created his own explanation - and then attacks that. That isn't the way scientific discussion works.

(By the way, it's funny you mention string theory in this connection. Motl, of course, is a string theorist, and he's done everything he can to sabotage the reception of the books that have come out recently that point out the crisis in string theory.)

Remember what I said about the IPCC's "very likely" being no better than betting odds? Here's an item from Scott Aaronson, about the difference between examining evidence and calculating odds:

Layperson: I just heard on the radio about this new Yood d’Shnood Theory of the Universe. What do you think the odds are that it’ll turn out to be true?
Scientist: Well, so far I haven’t seen any good evidence that…
Layperson: Sure, but what’s your prediction?
Scientist: As I said, the evidence seems to be explained a lot more easily by…
Layperson: But what if you had to bet?
Scientist: Well, there are two ways to think about this. What the Yood d’Shnood proponents argue is that…
Layperson: No, don’t give me a dissertation, just give me a number!

MB:

- I'm not sure I get your point. The answer to the Layperson should clearly be: "It's not very likely. I'd bet on something else.

- With regards to how scientists approach likelihood, a pretty good article is:
"Belief and knowledge - a plea about language". Here's a brief quote: "Much is tentative, but much is well understood and unlikely to be discredited. We scientists need to convey more about the status of our knowledge than can be learned from the muddy "most scientists believe" statement. We need our listeners to know what is tentative and what is not so that they understand better the ragged but cumulative progression of science and can use current knowledge effectively, with an understanding of its inherent uncertainties, in personal and political decision making." This is by a former president of the American Physical Society, an astrophysicist. It's a very nuanced article, and is quite interesting.

- In general, the approach of the IPCC was one way to try to express the point that there are some things that they're really sure about and some things that they just think are likely, or more likely than not. Whether or not you like the 50/65/90/95/99/ >99 approach is up to you; not at the scientists liked either. But I think you should understand that there is a difference between being somewhat sure, being quite sure, and being damn sure. This is the dimension that they are attempting to express.

Neal,

You left out the possiblility of damn sure and totally wrong.

It has happened.

M. Simon,

Yes, it has.

And good scientists worry about that, all the time.

But the way you deal with that is to do the best you can on the science, without worrying about "where the chips will fall". Someone working on a complex simulation is not worrying about whether he can get the GW to increase another degree: s/he's worrying about why the program didn't compile, or why the result doesn't display the expected dependence on some parameter ("Why doesn't it wiggle here when I waggle that there?"), etc. Technical issues, that give you ways of checking that the program is, in fact, doing what you thought it was doing.

However, if you look at what the GW-denialists are doing, they are very typically coming from the point of view of not liking the social consequences of the science; and therefore impugning the science. This is why so many of them are funded by think-tanks dedicated to "free enterprise". Honestly, if there were no impact of the Kyoto Protocol on oil use, would the American Enterprise Institute be railing against the science of GW? Why would they waste their time?

Neal, that argument cuts both ways. Would the environmentalists be supporting the Kyoto Protocol if it didn't force the US to stop burning coal and oil? Would the NGOs be raising alarms about global warming, if the prospect of regulating world power generation, with all the opportunities for rent-seeking it would afford, didn't exist? Ideology and venality work on both sides of any political question; charging that those who attack a scientific theory do so because they would lose if it were true is asinine, for it's just as likely that those who accept the theory do so because they expect to gain if it were true. The motives of the arguers are irrelevant to the validity of the arguments.

MB:

What I am interested in and thinking about is the story told by the scientists, not by the environmentalists, based on their facts, reasoning, and compatibility with the rest of science.