Has Global warming Stopped?, ctd.
I blame myself for what I consider to be the pretty disappointing responses to my prior post (especially the normally excellent TAS comboxes). The fact that so many people have reacted to things I wasn’t trying to say indicates that the communication failure is mine. So let me try to be clear about what I was actually trying to say.
When confronted with objections to an apparent scientific consensus, one valid approach is simply to assemble a wide variety of relevant scientists, ensure that the questions posed to them are technical questions within their scope of competence, and rely on their findings. This is the basic idea behind the UN IPCC, and the AGW reports of various national scientific academies. This has been my approach in the case of AGW, where I have always taken the technical findings of the IPCC as the starting point for any policy analysis on this topic.
George Will (or at least the view that was reasonably imputed to him by his interlocutors), questions the validity of the scientific consensus on AGW. His interlocutors, instead of just relying on the IPCC process, tried to engage the substance of George Will’s quasi-scientific objection. They responded by saying that he has not looked at a long enough trend line.
The primary point of my post was that while I agree that George Will’s (implicit) attempted falsification of AGW theory is not compelling, neither is the logic used by his interlocutors. Both logics share a common source of failure: looking for an underlying “trend” in the temperature record independent of physical causality. There is no magic “trend”, but instead a set of causal effects based on physical interactions that drive temperature. The scientific assertion made by the global climate science community is that we have built models that allow us to understand these effects with sufficient precision to make useful forward predictions. When evaluating this assertion, then, the relevant standard is not “Is the rate of warming slowing or accelerating?”, but rather “How accurately are our models predicting the rate of warming?”. That is, we should rationally care about deviation from prediction, not deviation from trend. This is why I described George Will’s (implicit) method for addressing the certainty of our scientific knowledge as “misguided”, which in addition to explicitly disagreeing with his conclusions, seems like a funny way of defending him.
The secondary point of the post was that a component of any well-structured prediction modeling process is to have model evaluation groups separate from the model-building teams that have different incentives and reporting structures, roughly analogous to a QA team for software development or fact-checkers at a magazine. One key task of such model evaluation teams is typically to escrow copies of code used to make predictions, log forward predictions made at time X for some outcome after time X, then run the code at the time of the predicted event with actual data entered for all inputs other than the asserted causal factor, and compare the resulting model output to actual outcomes. This is done across a range of predictions to create distributions of model error. While there have been some kludgey, one-off attempts to do something like this for the Hansen 1988 testimony, and a group has tried to look at single-year predictiveness of global climate models, there is nothing like a structured program in place to do this for climate models. Such approaches are always imperfect – and I tried to point out in the post some of the reasons that this would be especially problematic in the case of global climate models – but it would still provide a far better basis for the discussion of prediction adequacy than we have now.
What’s especially ironic abut a lot of the commentary on the post is that lots of people take assertions of uncertainty in climate forecasts as undercutting the case for emissions mitigation, so those on the Right argue for uncertainty, and those on the Left argue the opposite. In the sophisticated AGW debate, the economic justification for mitigation is seen as, conceptually, an insurance premium. If the expected warming takes place with expected effects, it is very difficult to justify the economic costs of mitigation, and therefore it is a hedge against much-worse-than-expected effects. Therefore, the greater the uncertainty in climate prediction, the stronger the case for mitigation – uncertainty is not our friend. So before you accuse me of intellectual dishonesty, recognize that in pointing out limitations in the current practice of climate model validation, I am actually arguing a point that cuts against my stated policy preference.
Thanks for the clarification and the courtesy of tone that makes the American Scene one of my favorite sites.
I agree with the substantive recommendation (tracking model predictions to measure their efficacy). Indeed it seems to fall in the motherhood and apple pie category that I doubt many would argue with. So on to the less important issues!
Your claim that Will’s critics are making a mistake seems off to me. The analogy I would use is from baseball. Let’s say Will made the case that player A is better than player B because A has a better batting average. If you wanted to make a thorough case against Will you could note that batting average is a seriously incomplete approach to evaluating players. But if Will was actually wrong about the batting averages, I think it is reasonable for most critics (especially blog critics) to simply note he is wrong on his measure, rather than also make the more complicated and lengthy argument that his measure is also unsatisfactory.
best,
Tom G
— Tom G. · Jul 25, 11:19 PM · #
Jim, as a non-expert, my options for understanding scientific questions are limited. On global warming, it has been reported to me over and over again, by credible sources— both within the field of climatology and from outside of it— that the scientific consensus, in the relevant field of expertise, is overwhelmingly that it exists.
Now the existence of overwhelming consensus among credentialed experts in the relevant field isn’t something that overwhelms my critical capacity. It’s not like I turn off discrimination, or shut my mind to contrary opinion. But it is very, very hard to side against eminently qualified people from across a broad swath of a given academic field. It takes a lot of contrary evidence to change my mind in such a situation.
I put it to you that you are perhaps too hard on people who are taking the rather sensible step, when confronted with complicated scientific questions, of allowing their opinions to align with a powerful scientific consensus. And I have to ask whether you are perhaps guilty of the (entirely human) failing of having questioned the science for so long that you can’t help but question it past the point where such questioning is no longer justified by the evidence.
— Freddie · Jul 26, 12:17 AM · #
A few points:
You wrote, “There has not been a lot of measured warming for the last ten years. It’s hard to dispute this.”
This is false, or else depends on a ludicrous definition of “a lot”.
As I told you before, the problem with the denialist claim of “no/little warming in the last ten years” is that the window moves on you. I also said, “Anyone who wants to keep this canard alive has to explicitly name 1998 which is a little too transparent amongst those passing familiar with the subject.” Many of the people who called Will’s/Steyn’s claim idiocy or similar did so because Will/Steyn were trying to use 1998 without being upfront about it. That’s simply, unforgivably, fundamentally unserious.
The thrust of this post appears to be misguided. When talking about testing the models, you seem to be confusing the AR4 model projections with predictions. It’s an unfortunately common error, especially among those who don’t seem to have actually read the IPCC Assessment Reports. It would be incredible- literally unbelievable, actually- if reality matched the ensemble average.
“If the expected warming takes place with expected effects, it is very difficult to justify the economic costs of mitigation”
Perhaps you’d like to be more specific here, especially in terms of what you believe the “expected warming”, “expected effects”, and “economic costs” are. It would also be helpful if you specified over what time frame you’re talking about, making sure to include the minor issues of fossil fuel finiteness and paleoclimatic history.
— thingsbreak · Jul 26, 01:46 AM · #
I write though I think I’m one of the commenters who disappointed.
After reading this post and Freddie’s comment, I think I know what is bugging me most about Jim’s two posts on this Will imbroglio. Jim frequently says that he has ‘accepted’ the IPCC data for purposes of his many economic arguments against Waxman, etc… But these two posts show that Jim has real questions about the models.
This is not something Jim has hidden. Here’s a great March 2007 post on it: http://cc.bingj.com/cache.aspx?q=jim+manzi+national+review+climate&d=76412775436317&mkt=en-US&setlang=en-US&w=7292ab6a,8dd3da0d
(I don’t know if this subect is covered in Jim’s big NR article because that seems to be for NR Digital subscribers only.)
So Jim does have doubts on the models. But Jim cannot address that in this environment in a way that will convince people, because it will be treated as heresy. Still, he can still usefully talk about the economic issues around AGW without taking on the climate models. But Jim is really worried that the models are still just babies. And if the IPCC’s models are wrong, it may not be because true AGW believers put their thumbs on the scales. Instead they might be wrong because models like this just cannot be reliable. That means that the models may be just as likely to underestimate the risks of true disaster as to overestimate them.
If that’s right, then the other commenters may be correctly picking up on the fact the Jim is, like Will, a little bit of an AGW heretic. I know Jim’s heresy is based on a highly sophisticated distrust of this type of modeling and not on a stereotyped distrust of liberals or NGOs or committees or scientists or even NGO committees of liberal scientists. And I know Jim’s distrust of the models drives him to worry as much about things being worse than expected as their being better than expected. But Jim is not just criticizing Waxman and the IPCC on the basis of cost/benefits and the impossibility of coordinate global carbon responses. He is also saying there is good reason not to put much faith in the models.
So Jim, are you an IPCC/AGW/Climate Model heretic? Do you think that we are foolish to put any great weight, pro- or con-, on them? If so, I’ll check out a copy of The Crucible from the library, or maybe I’ll just watch the Demi Moore version.
— tom · Jul 26, 02:15 AM · #
And if the IPCC’s models are wrong, it may not be because true AGW believers put their thumbs on the scales?
But I’m not a “true believer”, tom. I’m someone who trusts the relevant experts in their scientific fields to have the best possible current scientific knowledge. You really read my previous comment and see it as being worthy of mentioning The Crucible? Really?
I think Jim is a brilliant science writer. Far more knowledgeable on these issue than me. What I don’t believe is that I can take his analysis as more dispositive than that of brilliant and perfectly credentialed climatologists. Do you really think that means I’m calling him a “heretic”? Really?
— Freddie · Jul 26, 02:59 AM · #
Jim, I think that the biggest problem with your previous post was connecting it to George Will. If you’re trying to make a point about how to evaluate the reliability of our predictions about future global warming by testing the models, then focus on that. Will has a track record of making misguided arguments against the AGW consensus, and the “no warming since 1998” argument has a long history of being repeated and debunked. I don’t see why you’d want to put your arguments in that context, casting it as a defense of people like Will who keep making the 1998 argument.
I also don’t think that the reasons that you give for connecting your arguments to that context hold up very well. The flaws with the “no warming since 1998” argument are basic enough that it can be debunked (in a variety of ways) without invoking any sophisticated models. Drum et al aren’t bringing up climate models because there’s no need to – simpler counterarguments are sufficient.
A little bit of searching turned up this post on a study of how some climate models have fared at projecting warming since 1990. Short answer: pretty well. Is that the sort of thing that you want to see more of?
— Brad · Jul 26, 03:37 AM · #
Freddie, I’m agreeing with you. You were saying that Jim was questioning the science. That’s right. I’d been focussing on the economics and the international relations points.
I do think you’re wrong to talk about deferring to climatologists. This is a question about modelling something almost unfathomably complicated. It’s the world’s biggest math problem. And it’s very possible that no one can credibly answer it with any range of confidence that we could compare to the way we make decisions in other areas of science like medicine or electical engineering, etc…
I didn’t say or think that you would attack Jim as a heretic for making the ‘sophisticated’ case against relying on the IPCC consensus. But I’d be stunned if others didn’t.
— tom · Jul 26, 03:41 AM · #
Did you read Jim’s post, Freddie? He agreed that Will’s skepticism was misguided, and wrote:
I put it to you that you are perhaps too quick to assume that people who disagree with your preferred policy recommendations for dealing with climate change are doing so out of a misguided skepticism of the science.
— John Schwenkler · Jul 26, 05:22 AM · #
Oh, brother. He agrees Will’s skepticism is misguided! He agrees it’s misguided! I think I may faint!
Honestly, it’s just sad listening to this slobbering love for Manzi’s labored attempts to torture various models into coughing up the answers he wants. This analysis is really just a half a step more respectable than outright climate change denial, but everyone here seems to think it’s just brilliant.
The “Left” argues for certainty in climate change forecasts? That’s funny, because it doesn’t. In fact, the only person I’ve ever heard argue for certainty in climate change forecasts is Jim Manzi, and for an obvious reason: his discounted cash flow analysis of the global economy 100 years from now depends on the numbers being just so.
— Feist · Jul 26, 05:38 AM · #
QED!
— Blar · Jul 26, 12:44 PM · #
Tom G:
Fair enough. My meaning of “about the last decade” was to compare the last full calendar year (2008) to the one ten years earlier, and accept that 2008 was cooler than 1998. I accept that there are lots of other definitions of the term, which could include the last rolling 12 months to the rolling twelve months ten years earlier, and so on. In a real analysis, I would obviously have specified this. The reason I didn’t care much was that the whole point of the post was that not only considering this specific ten year trendline was misguided, but that looking for the “trend” in the data itself was misguided.
I hold the posters I cited, and especially Drum, to a higher standard than some random hack. I expect them to think through the issue with some depth, based upon their previous writing.
Freddie:
As per John’s comment, I was trying to say that I think we actually have a pretty similar approach to identifying a best-estimate scientific case on this, which relies heavily on deference to scientific experts within their range of comeptence, and subject to normal intra-scientific checks-and-balances.
thingsbreak:
Please see my response to Tom G.
tom:
My take on your question has consistently been the following:
1. We should use the formal published estimates of the IPCC 4AR as the best available estimate for the expected impacts of various emissions scenarios, as well as for the estimates of the probability distributions of for these impact estimates.
2. We should recognize that there is uncertianity around each of these probability distributions, whihc is by defintion unquantifiable.
3. There are technical actions we can take to manage the modeling process better that would, to some extent, reduce this uncertainty. I believe that we should take these actions.
Brad:
In terms of your first couple of paragraphs, please see the response to Tom G.
I’m very aware of that post and other analysis like it (see Roger Pielke’s Prometheus blog archives over the past 2 years for links to all sides of this debate). Without wanting to try to adjudicate it in a combox, just note that nobody has actually escrowed model codes to allow formal analysis of prediction error. This is a basic procedure that would be done in real program of model validation.
Feist:
This is not exactly my idea of slobbering love.
— Jim Manzi · Jul 26, 12:46 PM · #
Jim, does anyone make your same argument (the models are the best we have today but they aren’t good and they are likely to be wrong in one direction as the other) on the other side?
Also, you say that the uncertainty in the models cuts in favor of spending more on insurance against AGW. Is that true if we’re looking at a basket of global disaster risks and are at least in part deciding how to allocate our ‘disaster portfolio’ spending among them? Couldn’t the lack of reliable predictions from AGW models lead us to conclude that we should do less on carbon and more on volcano-lancing, asteroid blasters and assassinating nanotech experts?
— tom · Jul 26, 02:22 PM · #
deBoer’s law: whenever people online complain about someone’s reading comprehension, they are failing to read correctly themselves. Cf. Schwenkler, John, above.
— Freddie · Jul 26, 02:24 PM · #
Is it really not clear that Jim is a) expressing skepticism towards George Will’s skepticism but also b) suggesting uncertainty at the vision of AGW that the climatology community endorses? Indeed, there is no post if not for disagreement with the climatology community.
“The scientific assertion made by the global climate science community is that we have built models that allow us to understand these effects with sufficient precision to make useful forward predictions. When evaluating this assertion, then, the relevant standard is not “Is the rate of warming slowing or accelerating?”, but rather “How accurately are our models predicting the rate of warming?”. That is, we should rationally care about deviation from prediction, not deviation from trend. “
The scientific consensus among the climatology community is that we are accurately predicting the rate of warming with our models. If Jim is disagreeing with that, he’s disagreeing with the scientific consensus. As I explicitly said, twice, that’s his right, and the fact that he is disagreeing with a scientific consensus doesn’t necessarily make him wrong. It does mean that I approach his position with additional scrutiny, which is the sensible thing for a non-expert to do.
Jesus, think, then post.
— Freddie · Jul 26, 02:34 PM · #
While this seems a reasonable suggestion in principle, I don’t understand how it could work in practice. The problem is that the timescales involved in global warming are too long to make this feasible. For this to really work, we would have to use predictions over several decades. What would that tell us? It’s not going to do us any good to wait until 2050 to find out if the climate models of 2009 are accurate or not. By then we’ll be using different models anyway, so the answer will only be of academic interest in any case.
One thing that I find interesting about this suggestion is that it’s basically proposing a regulatory structure on the climate science community. I don’t see any evidence that there’s a need for such a structure. Surely competition between different groups of researchers in the scientific marketplace is the most efficient process for producing advancements in climate science.
— Chris · Jul 26, 04:23 PM · #
Wait, Freddie, are you replying to Jesus now, or just John? I went a little bit that way with the heresy language, which I believe is a fair description of the way AGW-questioning is treated by its unwashed defenders. But if we’re talking to Christ now I have a few difference questions and a some confessions that I really don’t want to make on this thread.
That said, I think it’s pretty clear that Jim is questioning the science when he says that while the models are the best we’ve got, they aren’t even close to being tested and trustworthy in the sense that any other models used by scientists are. And I would guess that there are at least some elite defenders of AGW who agree with that. That’s why I asked Jim above if he’d dealt with anyone like that.
And pace Chris, I don’t think Jim is proposing a regulatory structure, etc… He saying HERE IS A BIG LIMITATION ON THESE MODELS (sorry for the all caps). It is what it is, and the fact that these models are the best we can do today doesn’t make the models any more reliable.
And Jesus, if you’re really on this thread, I’m sorry. Don’t make me say what for.
— tom · Jul 26, 04:33 PM · #
Oh, and one thing I meant to mention, but forgot, there was an interesting post just a couple weeks ago at realclimate.org about one way to interpret natural variation in the climate with a particular look at the last decade of temperature measurements.
— Chris · Jul 26, 04:43 PM · #
@Brad
The Tamino post is a nice analysis of how the models performance with respect to the last two decades (and as you alluded, its why the Swanson paper correctly still uses the bias from the 20 yr trend). The suggestiveness of the last decade or so and how it matches these projections are really what is at issue with respect to the science in these exchanges. Swanson and others are absolutely correct in maintaining that some of the models are somewhat accurate while still keeping a watchful eye on what is currently too short term a change to warrant a chicken little (and convenient for skeptics) reaction to many of our models. Predicting a lull or having to deal with a possible stagnancy in the trend is important however, not just for the enormous political implications, but because we’re likely to see more of these periods and need to account for them. That may mean eventually tweaking models or having to throw out certain ones wholesale (or understanding science better—are lulls due to ocean variability or aerosols?—and how much of it is in fact due to the inadequacy of models in these and other areas?) and there will be scientists/teams who will be very resistant to such things (and this is just as true if we find that low level clouds produce significant positive feedback). The political ‘sides’ that may align in what would otherwise be a commonplace battle of competing teams/models shouldn’t result in reflexive defense on either side.
And to one of the other commentators higher up—here is consensus that there is anthropogenic driven warming amongst climate scientists but there are a variety of models/varying projections. Because of the highly politicized nature of this sub field that means that they all need to be addressed because they all inevitably perk up the ears of reporters.
— nickswisher · Jul 26, 06:24 PM · #
Cf. Schwenkler, John, above.
You misused ‘cf.’ muthafucka.
(I kid with the ‘muthafucka’ part.)
— Kristoffer V. Sargent · Jul 26, 06:58 PM · #
I don’t think that Jim is disagreeing with that at all; he’s just acknowledging, as any scientist would, that that accuracy has its limits – hence his question was that of how accurate our best models’ predictions are, which is, again, a question about which there is sure to be plenty of debate among experts.
deBoer’s Law in action!
— John Schwenkler · Jul 26, 09:38 PM · #
Freddie: I’m not sure what you mean by “uncertainty” when you criticize Manzi for “suggesting uncertainty at the vision of AGW that the climatology community endorses”.
Surely, the scientific concensus is that there is a reasonable degree of uncertainty about our current understanding and predictions of climate.
I guess you could take a wholly hands-off approach, in which you argue that it’s meaningless for educated laypeople like Manzi or you (or, dare I say, me) to think about how to test the current concensus on AGW — under this strong version of expert deference, we sit back and wait, and when we are informed that the scientific consensus has concluded that the prior consensus was not exactly right, then we adopt the new consensus, again without thinking about how to test it.
This presents problems with determining exactly what opinions lie within the consensus and what opinions are outliers by individual members of the expert community, but it’s not a crazy proposal. However, even under this proposal, we all have to grant that there is some uncertainty in the accuracy of the current consensus. Under this proposal, we wholly defer to the expert community to resolve the uncertaintly, but if we deny that uncertainty even exists, we are engaging in denialism on the other side, because the uncertainty itself is part of the consensus.
Manzi is engaging, IMHO, in weak form deference, in which he accepts the consensus as consensus, but then dares do things like think about what kind of tests would be helpful to confirm the consensus over time.
— J Mann · Jul 27, 02:46 PM · #
I appreciate Manzi’s follow-up, and I realize that I was mistaken in my reply to his first commentary when I asserted that he is effectively asking, how long do we have to wait before we can say this (global warming) isn’t really happening.
I realize he is not sympathetic with George Will’s conclusion— with which he disagrees; rather, he wants to see more rigor in how AGW model projections are tested against actual climate change. And for good reason— such rigor would no doubt create a better conversation about the issue than the current one in which we battle over graph lines.
I wonder, though, at what point do we devote significant funds to address emissions in spite of the potential flaws in the models? Maybe that’s similar to the the very question Manzi raised in his first post. However, I sense he’s enormously cautious about existing projections, and I’m unsure what level and duration of scrutiny would satisfy him. At some point, perhaps, reaction— including even more significant mobilization of funds and manpower— should go forward, regardless of shortcomings in the assessment process.
Haven’t we often mobilized for war, based on threat projections that, in hindsight, left something to be desired? That isn’t to excuse such mobilizations, but to say that inevitably there’s a risk that a threat might be exaggerated. Demands for a more rigorous assessment can always be voiced, but at some point, in the case for war, we hope that our intelligence and military communities are giving us their best-faith estimate.
Sometimes the commitment has to be made before the consensus projections can be tested as thoroughly as everyone likes.
I wonder if those who consistently air skepticism and caution about warming projections are as demanding of experts in other fields— like the State department— who also make projections that affect our stability and security? I realize that war-making and emissions-curbing are not neatly analogous, but it seems the comparison could be a place to start discussion.
Why is it that reaction to global warming demands a patience and rigor we might not tolerate in the face of other dangers?
— turnbuckle · Jul 27, 05:47 PM · #
turnbuckle:
Thanks. And as I said, when 25 people misinterpret what I said, it’s rational to assume that the communication problem was mine, not that of 25 other people.
You ask:
I think that we should act today based on our best-available estimates of impacts (ie, using the IPCC 4AR). I have adovcated this in many articles and posts. The actions that I think are justified are just different than those adovcated by many people who proceed from this data. I won;t try to justify why in a combox, but will point you to my various posts on it here at TAS, for example.
— Jim Manzi · Jul 27, 06:40 PM · #
“There has not been a lot of measured warming for the last ten years.”
“If we had no warming over the past ten years (true) …”
This is a big part of the reason why Mr. Manzi can be so infuriating on this topic. Setting aside the basic question as to how important a 10 year trend is (not very), the first statement above is defensible, albeit questionable depending upon how one defines “a lot;” the second is not defensible, and is unworthy of Mr. Manzi.
The fact of the matter is that Mr. Will, aside from being misguided, is a liar. And one does not have to have any particular view on AGW or the policy implications thereof to reach that conclusion; one merely has to have a basic understanding of statistics (i.e., the ten year trend line is obviously upward, and putting ANY weight at all on a downward deviation from a one year spike in 1998 is … ignorant nonsense, to put it kindly). To elide this fact – and to (implicitly) endorse Mr. Will’s lie, is disappointing, regardless of just how relevant it may or may not be to the ultimate issues.
My point isn’t so much that this particular lie by Mr. Will and (to be charitable) mistatement by Manzi is in itself all that important (it is somewhat with regards to Mr. Will who is writing for the general public, most of whom, as he well knows, don’t have the knowledge to detect his lie). Rather, to see this kind of sloppiness from Mr. Manzi undercuts his credibility on more important other issues related to AGW.
— LarryM · Jul 27, 07:47 PM · #
If you want to get a good idea of the problem, think to yourself why wind tunnels and flight tests are necessary even though the well-developed discipline of fluid dynamics exists. If you then understand how complexity grows exponentially as dimensions multiply, you will then see why a robust skepticism is preferable to “trusting the experts.”
— Kristoffer V. Sargent · Jul 28, 06:25 PM · #
And that’s before we even get to chaos, non-abelian trajectories along the nunc fluens, nested feedback, open systems, lossy compression, and so on. The fact is, our models are much much much more likely to be wrong than right. Believe it.
(But, you say, that means we might be underestimated the impact! Yup, but it also means we might be overestimating the impact, too, and it’s a toss up which one. And, since 50/50 is the definition of maximum uncertainty, and since uncertainty is the opposite of information, we can conclude, formally, that the “might be underestimating” meme gives us precisely zero information on which to make a decision. So stop using it.)
— Kristoffer V. Sargent · Jul 28, 06:42 PM · #