July 06, 2005
On The Hockey Stick
Posted to Author: Pielke Jr., R. | Climate Change
[Reminder for locals: NCAR MESA Lab, 3PM Friday, talk by Hans von Storch and subsequent panel on hockey stick issues!!]
A while back we commented that here at Prometheus we don’t do hockey sticks. Well, now we do. What follows is an unbearably long post on issues associated with the Hockey Stick. I am sure that it will be of interest only to those with a deep interest in this issue. For others, move along, nothing to see here ...
The “hockey stick” refers to a graph (see figure b here and figure 2.20 here) which shows the results of a reconstruction of global average temperature for 1,000 years. The hockey stick has been the subject of an intensifying debate that now has reach comic/tragic proportions. Last week Congressman Joe Barton (R-TX), Chair of the House Energy and Commerce Committee sent letters to three scientists who authored the hockey stick studies, as well as the head of the National Science Foundation and the head of the IPCC, asking for a range of information.
From the perspective of climate science or policy Rep. Barton’s inquiry is simply inane. There will be little insight gained on climate or how we might improve policies on climate change through his “investigation.” As Congressman Henry Waxman (D-CA) has written in response to Rep. Barton, “These letters do not appear to be a serious attempt to understand the science of global warming. Some might interpret them as a transparent effort to bully and harrass climate change experts who have reached conclusions with which you disagree.... If the Committee indeed has a genuine interest in the science of global warming, you should withdraw these letters and instead schedule a long-overdue Committee hearing on climate change.”
Of course, it is doubtful that Rep. Barton’s Committee (on Energy and Commerce, I remind you) actually has any real interest in the science of climate change, except as a tool of tactical advantage in the continuing political battle over global warming. Rep. Barton and others opposed to action on climate change will continue to gnaw at the hockey stick like a dog on a bone so long as they perceive that it confers some political benefits. The great irony here is that in many instances the supporters of the hockey stick have often been their own worst enemies and fed the flames of this debate, which now threatens the integrity of all of climate science, and to turn all of climate science into climate politics. The debate also consumes a lot of scarce attention on the climate issue – attention that would be better devoted to debates about policy options.
Can the climate science community do anything about this situation? To understand how the hockey stick issue might be de-emphasized and moved beyond requires understanding the debate and its political context.
The hockey stick is a symbol. It represents much more than the results of two studies by Mann et al.; it represents the integrity of the IPCC, claims of a human influence on climate, demands for action on climate change, and no doubt other things as well. Consider the range of emotions and issues evoked when you think of other examples of symbols that condense a great deal meaning – for example, the United States flag, a swastika, or a religious cross. Symbols serve as a sort of short cut when we try to make sense of a complicated world. The manipulation of symbols is consequently a high political art. As political scientist Murray Edelman wrote in his book The Symbolic Uses of Politics (at pp. 31-32), “It is characteristic of large numbers of people in our society that they see and think in terms of stereotypes, personalization, and oversimplifications, that they cannot recognize or tolerate ambiguous and complex situations, and that they accordingly respond to symbols that oversimplify and distort.” It is important to recognize that symbols take on meaning far beyond the “reality” referred to by the symbol. Edelman writes, “… reality can be irrelevant for persons very strongly committed to an emotionally satisfying symbol… commitment to a belief is likely to be strengthened and reaffirmed in the face of clear disproof of its validity where there is a strong prior commitment … and where there is continuing social support of the commitment by others …” This apparent disconnect helps to explain (for both sides) why the larger symbolic and political debate over the hockey stick won’t be resolved by completing claims to factual correctness made by either side.
The hockey stick became a symbol because of the IPCC. In its 2001 report of Working Group 1 the Summary for Policymakers (SPM) included only 5 graphs. The entire report has probably 100 graphs (perhaps more, I have not counted). So the IPCC was in effect saying, “We’ve looked carefully at all of the evidence of climate change and from all of those studies and reports the best example that we’d like to share with you policy makers of changes in the earth’s climate is represented with this graph.” Of course, most Prometheus readers will know that the case for a human influence on climate is well established through multiple independent lines of research. But remember, we are talking about the hockey stick as a symbol. For the uninformed outsider reading the SPM, none of this richness and context would be apparent. The IPCC offers that not only is the hockey stick the best example that it can provide of climate change, but that it has been peer-reviewed at a level more stringent than normal journal articles. Representative of such claims, the scientists at RealClimate have written, “IPCC reports undergo several additional reviews and revisions involving a large number of independent referees. Thus, the IPCC reports undergo a more stringent review process than common papers in the scientific literature.” The hockey stick is thus a powerful political symbol in the climate debate.
With the hockey stick established as a symbol it is to be expected that it attracted both support and opposition as a tactic of politics. Those advocating particular actions on climate change would, for example, prominently display the hockey stick in talks, no doubt referring to its endorsement by the IPCC. Those opposed to action on climate change (or just generally skeptical) would be drawn to the hockey stick target number one in debate over climate change. The hockey stick is debated because it is a key symbol of a an intense and meaningful political debate.
How should an outsider make sense of the hockey stick debate? The debate is technical, and includes references to all sorts of obscure statistical techniques and paleoclimate proxy methodologies, and thus may seem impenetrable. But the technical debate may be less important than the proponents on either side might claim.
The technical debate centers on the authors of the original two hockey stick studies referenced by the IPCC, Mann, Bradley and Hughes (MBH) and their primary antagonists, McIntyre and McKitrick (MM). Mann et al. often make their case at www.realclimate.org and McIntyre (and sometimes McKitrick) make their case at www.climateaudit.org.
There are several characterizations of this debate readily available in the blogosphere. One claims that MM are tools of the oil and gas industry, ready to spread misinformation and lies in order to discredit the noble truth-seeking scientists whose work is beyond reproach, all in hopes of derailing action on climate change. Another characterization holds that MM are the proverbial Davids up against the Goliaths of the IPCC and the global warming juggernaut intent on remaking the world in its environmentally-correct image.
While such stories will resonate strongly with black hat/white hat types, from where I sit the reality seems a bit more complicated. Here is how I see it -- MBH conducted several studies that, by the conventional norms of the climate science community, represented excellent work and were published in leading journals in the field. But the norms of the climate science community for peer review and replication are not widely shared in other fields. So when MM were drawn to MBH (indirectly or directly by the IPCC SPM no doubt) from outside the climate science community with an eye to take a close look at the their work I’d venture that MM likely brought along with them a perspective on norms of peer review and replication quite different from MBH. This alone would be enough to generate some push back from MBH. But add the fact that McKitrick had already established himself as a collaborator with the climate skeptics, and this probably was enough to engender some serious opposition from MBH, who no doubt were used to dealing with their traditional climate skeptic opponents with loudness and bluster. MM did not go away or back down.
Now it turns out that science can be messy and scientists can be among the more disorganized people you might meet (I could show you some offices). Perhaps MBH were both unprepared to deal with an outside request to replicate the entirety of their work, and their behind-the-scenes work had enough inelegant shortcuts and ugly warts involved so as to make a full, public replication maybe uncomfortable and potentially embarrassing. To be clear -- I am not suggesting anything close to scientific misconduct or fraud, but the standard messiness and disorganization associated with the scientific research in this particular community where requests for computer codes and inside info rarely are made. For example, when asked about responding to a letter such as that sent by Rep. Barton to MBH Hans von Storch’s reply exemplified such messiness when he replied, “"If I did get such a letter, I would become desperate," he said. His colleagues often write the code for his studies, and he said, "if I asked my colleagues whether they still had the code, I'm not sure they would."” Climate scientist James Annan makes the point more generally on his blog, “I don't know of any scientist who could answer such questions. It's just not the way our work is done - there is far too much pressure for rapid and new results for us to maintain full "audit trails" and answer an unlimited number of questions from any troublemaker with too much time on their hands. By the time 5 years have passed, our work is either irrelevant and forgotten, or else superceded, either because it really was wrong, or because someone else improved on it.” (See the similar comments of Kooiti Matsuda here.) Other disciplines no doubt have different norms of conduct. Undoubtedly part of the dispute between MBH and MM is about norms of scientific inquiry, something that won’t be resolved through the technical arguments.
Bottom line – It seems reasonable to think that while MBH did not engage in scientific misconduct, that a full public replication of their work may prove potentially uncomfortable or embarrassing. (Note that no matter what it is that MBH are not revealing, it could have no consequences for either the scientific consensus on human influence on climate or perspectives on recent temperature trends in historical perspective, as these issues are supported by a larger literature.) So they resisted, and have stood fast. For their part, MM did not give up. MM publish in the peer reviewed literature, giving some credence to their critique, the Wall Street Journal weighs in, a few guys at NCAR claim to vindicate MBH and now the issue has gone all the way to the U.S. House of Representatives.
In the long run I am confident that the technical dispute between MBH and MM will be resolved in the peer-reviewed literature. But so what? Already research on paleoclimate modeling and proxies by von Storch et al. and Moberg et al. has superceded the work of MBH.
Resolving the technical dispute will do little to address the larger issues of climate science policy or the symbolic and real political implications of the hockey stick debate. Let me conclude this lengthy post with some unsolicited advice for many of the parties in the science community involved in this messy and ugly debate.
1. For the IPCC. You have a real problem here. If you do anything other than staunchly defend the hockey stick, you run the risk of being perceived as taking a step back on climate science (rightly or wrongly). Given the political agenda advanced by IPCC leadership this will be perceived as damaging. But consider also that a step back can serve to increase your credibility. It was clearly a mistake to use the MBH studies in the SPM (this RealClimate post makes this point abundantly clear enough). The case for human impacts on climate is wide and deep. In the future consider some different options for the SPM such as handing off the job of the SPM to a separate group with expertise in such things, or not even doing an SPM. You have a conflict of interest problem (real or perceptual) as well, given that M of MBH was involved with reviewing his own work as an IPCC lead author. Again, this may be acceptable given the norms of the climate science community, but those are not the norms at play in the larger political world of climate science. Consider getting some good advice on institutional design and public relations. And above all, avoid the hubris that too often characterizes climate scientists in their interactions. What works in the academy often does not in the broader world.
2. For MBH. By all means stop invoking the funding and political agenda of your opponents, which you often offer with fire and vitriol. This only serves to legitimize inquiries into your own funding and political agenda. It is easy for me to say, but I recommend not complying with Rep. Barton’s request. Be respectful, but decline or bury him with paper. Let him subpoena you if he dares (and then watch him then get buried). Add no more fuel to the fire, on RealClimate, in the press, etc. Unless you really have something serious to hide, give the world access to your original computer code and whatever else MM are asking for. Whatever benefits you get from invoking the principle of the matter seem to be dwarfed by the continuing reality of having to deal with this issue. Swallow hard, cut your losses and move on.
3. For MM. Continue to publish your work in the peer reviewed literature. Steer clear of those with political designs on your work, you’ll have more standing if you focus on the substance (M has done better than M in this regard.) Work to understand the norms of the climate science community, and don’t place blame for these norms on MBH. You might have a case to make for changing these norms, but make that case in the right venue. Focus more of your attention on the IPCC and its processes rather than MBH. There are larger issues here. Think about taking on another project. I am confident that the hockey stick issue will very soon resemble a dead horse.Posted on July 6, 2005 12:14 AM
For someone who claims to be blogging the interface between science and politics, I am aghast at the twisted ethics and broken logic you have displayed in this article.
You wouldn't call what Mann has done scientific fraud? That's fine. The evidence is not conclusive one way or the other. But Mann clearly did test the Hockey Stick for the presence or absence of bristlecone pines and must have known for some considerable time that the construction was not insensitive to the use or not of "dendroclimatic indicators". Is this good ethics?
Mann inserted his own work into the TAR and nobody checked his math. Ethical?
How can you possibly advocate that scientists should not co-operate with Congress on issues of transparency and scientific probity? What sort of ethics is that?
Would you call such things ethical? I wouldn't. But then I'm not a climate scientist and so therefore cannot play the "Get Out of Scientific Ethics" card.
This article is not even wrong. It's a disgrace to scientific principles to claim that climate science, which now has such a grip on political agendas that it now tops the current G8 summit, cannot abide by the normal audit trails and verification given every other part of public policy.
Posted by: John A at July 6, 2005 06:32 AM
John A.- Thanks much for your comments. Your comment illustrates nicely the sort of issues that arise when the norms of acceptable conduct differ between different communities. It is perfectly reasonable to question the norms of the climate science community. But if challenging norms is your goal, then you probably should consider taking on a much broader focus than the hockey stick, as the norms of the climate science community shape pretty much every study published. You will find some support in such an endeavor, see for example this 2000 paper by Ron Errico in the Bulletin of the American Meteorological Society, who writes,
“Too frequently, published papers contain fundamental errors… How can a piece of work be adequately evaluated or duplicated if what was really done or meant is not adequately stated?... My paramount recommendation is that our community acknowledges that a major problem in fact exists and requires ardent attention. Unless this is acknowledged, the community will likely not even consider significant changes. I suspect that too many scientists, especially those with the authority to demand changes, will prefer the status quo.”
Posted by: Roger Pielke Jr. at July 6, 2005 06:43 AM
I'm not sure where any of you get the idea that this is particularly a "climate science" issue. In my brief career, I've worked in the fields of mathematics, agricultural engineering, electrical engineering, oceanography, and most recently climate science. I have not noticed any difference in the cultural norms. The work undertaken by the various labs I've worked at has had a wide range of policy and commercial implications, so it's not as if climate science can be considered deserving of special treatment in that respect either.
Posted by: James Annan at July 6, 2005 07:56 AM
James- Thanks for your comment, which I agree with. The norms of climate science are certainly not unique when compared to other relevant disciplines. but they are very different than, say, economics research, pharmaceutical research, public health research, and so on. There are different norms in different areas of inquiry. That seems pretty obvious.
Posted by: Roger Pielke Jr. at July 6, 2005 07:59 AM
You appear to have ignored Mr. John A's major points. MHB has not been just some normal paper in the climate community. It has become a central icon for a world-wide political agenda with public policy outcomes of enourmous consequences for nearly every human on the planet. That sets it apart for the most intense, if abnormal, scrutiny. Blathering on about differences in accepted norms is non-sense in this context. Your seeming unwillingness to acknowledge this central point remains symptomatic of a clique with serious personal and professional investment in what is widely viewed -- including now the British House of Lords -- as "end of the world" alarmism. Mass delusions eventually run their course.
Posted by: Robert Ferguson at July 6, 2005 08:02 AM
Bob- Thanks for your comment, but it appears that you may not have read my post (or maybe it was unclear). I try to explain in the post why it is that the hockey stick has become exactly the sort of all-encompassing symbol (icon, you say) that you describe. This also explains the scrutiny. But to understand the reaction and the debate, one also has to understand he norms of the climate science community.
Posted by: Roger Pielke Jr. at July 6, 2005 08:17 AM
Thanks for the clarification which still appears, to me at least, undercut by your suggestion for ignoring Congress. Intended or not, this comes off as Mann-league arrogance.
Posted by: Robert Ferguson at July 6, 2005 08:28 AM
Bob- Thanks for following up. I do agree with Henry Waxman on this issue. If Barton wants to pursue this issue, then a formal, public hearing with witnesses selected by majority and minority members of the Committee is the way to go. Ignoring the request is one way to move things in this direction (though NSF does not have such an option). I appreciate that such a move might be viewed as arrogance as you describe, but Barton's inquiry is a pretty serious precedent, and perhaps the integrity of science in policy and politics deserves some push back sometimes. I do appreciate that there will be varying, legitimate views on this. Thanks.
Posted by: Roger Pielke Jr. at July 6, 2005 08:40 AM
I started off feeling irritated that Sen. Barton's letters were labeled "inane." It seems to me that he is asking some pretty sane, reasonable questions. If the world and it's leading economies are to bend their policy to a higher purpose, supported by findings at the IPCC, surely it is reasonable to question the methods and process of the IPCC if strong evidence suggests that neither due diligence nor objectivity are foundational to its existence?
I was however, pleased with the comments to MM and MBH. I agree, MBH should come clean and move on. Harder said than done though, when coming clean will likely result in career implosion and a permanent black mark for Mr. Mann. I also agree that MM should continue their work. I have read the posts by Mcintyre at climateaudit.org and I find him to be genial and forthcoming. I resent the implications I see in many blogs, including realclimate.org, when MM are described as "one is from mining, the other an economist." This is misdirection of the most cowardly sort.
I for one do not support massive bureaucratic contortions to satisfy Kyoto in the name of "just in case." I would prefer to see the UN reformed (another topic) with a specific environmental body addressing key environmental concerns, i.e. 1) overfishing, 2) resource conservation 3) renewable energy and 4) water (and etc) -- in a targeted and soundbyte-compatible formula - with incentives possibly tied to trade.
Of course I realize that this will not make for media fodder, and given the short bonobo-like attention span of our masses today, this type of thinking is doomed to backroom parliamentary committees and underfunded research chairs. Where does that leave us? I'm not sure.
The single biggest thing I would like to see is for President Bush to come up with a home-grown plan to reduce all toxic emissions -- not to head off global warming, but to reduce smog and other negative effects. It should be aggressive and tied to national economic goals. Yes, this would co-opt the Kyoto goals, but let's face it, Kyoto is going to fail in meeting it's own targets. India and China pollute at a per capita rate that is five times the US' and countries like Canada will simply spend taxpayer money to purchase compliance from credit-heavy nations like Russia. This is the single largest transfer of wealth from rich to poor that the world has ever seen, all in the name of the Kyoto accord. Part of me wants to label this a massive socialist conspiracy, but labels are so unproductive, aren't they?
Posted by: Guy at July 6, 2005 08:45 AM
Roger you state "The case for human impacts on climate is wide and deep." As a non-scientist, but reasonably able engineer, I find the skeptics case to be pretty wide and deep also, and perhaps better documented in terms of being more careful and less selective. Given your implication of sloppy work of climatologists would the case still be wide and deep if their work was rigorous, open and not selective?
Posted by: Murray Duffin at July 6, 2005 09:00 AM
Roger, I appreciate the fine line you've tried to walk in this article, but let me join with John A. and Robert Ferguson in calling you out on the "differing norms" argument. It is no defence whatsoever of MBH, on a matter of such profound importance, to say that their apparent negligence is justified because the norm in climate science is to push out sloppy work in a hurry. (You'll have to forgive me for paraphrasing, but that appears to be exactly what you're saying.) This is not exactly a ringing endorsement of the state of the art, and does nothing to inspire confidence in the "multiple independent lines of research" that underly the IPCC view of climate change.
I'm not sure you really want to follow your "climate science norms" argument to its logical conclusion, which is that the Hockey Stick is likely just the tip of a giant iceberg of flawed science, in which propaganda value is at least as important as technical merit. A reasonable person might ask, if such a flawed study found its way into the SPM, of all things, what else are they hiding?
Thought-provoking article, though. Thanks.
You are either naive or misinformed by Mr. Waxman and his allies that Congressman Barton’s actions represent some sort of ominous precedent. Call Senator Jeffords’ office and ask them to send you copies of their inquiries to some of Mann’s critics several years ago. Go back into the committee records and take a look at some of the letters sent out during the Chairmanship of Congressman Dingle. Placing no value judgments, Barton’s actions are anything but uncommon.
One could easily suspect Congressman Waxman and his NGO constituencies are trying to cover for Mann with misdirection. They are, after all, some of the most vocal agitators employing his work to posit the 20th century as unusual and catastrophic. In his career, Waxman has done exactly what Barton is doing, a form of discovery.
Additionally, Mann has not performed well in past public hearings, but has been openly rude to Senators and colleagues alike, and less than forthcoming.
Finally, I do not recall you ever complaining about how the McCain Senate hearings were so one-sided and publicly insulting to your absent colleagues "outside the mainstream consensus." How many times did he and others question the integrity and motives of “skeptics” and rhetorically ask, “Why do these people say these things?” without ever inviting them to appear on a panel and asking them personally.
Respectfully, your views on Congressional process appear highly selective or forgetful.
Posted by: Robert Ferguson at July 6, 2005 09:34 AM
Thanks for your comments. And yes, your perspective illustrate exactly why the hockey stick as a symbol matters and further, how this matter threatens the integrity of all of climate science. I certainly don't want to get into a debate about climate science (others have this well-covered, e.g., RealClimate), but I will just point out that most all of the skeptics accept a human influence on climate. The question that is properly debated is how significant that influence will be and what, if anything, should be done about it.
Posted by: Roger Pielke Jr. at July 6, 2005 09:43 AM
mgl- Thanks for your comment. Your paraphrase reads a bit more into my essay than I intended. By invoking the norms of climate science, I am neither excusing nor justifying MBH, but simply trying to offer a reasonable and accurate explaination for why the debate is as it is. See this earlier post for support of the notion that climate science should be more rigorous and open, for exactly the reasons that you and several others here have described:
Posted by: Roger Pielke Jr. at July 6, 2005 09:48 AM
Posted by: Jo Calder at July 6, 2005 09:49 AM
Bob- Thanks for these comments. Of course the "everybody is doing it" argument doesn't get you very far. I am pretty sure that we would have come down similarly on the Jeffords and Dingel letters, that you describe but I am unaware of, had we been aware of them and blogging at the time. I don't know much about the McCain hearings. In the future if such things come to your attention and you'd like commentary from one of us, just hit the 'Ask Prometheus" button and we'll be happy to oblige. Thanks.
Posted by: Roger Pielke Jr. at July 6, 2005 09:56 AM
BTW, Jo Calder makes an exceptional point. Mann et al. are not keeping hand-written notebooks in the 19th century. We have nearly unlimted data storage and manipulation capabilities today. Mann is not techincally incompetant, but likely covering for what Chapman et al. tagged as "just bad science."(D. S. Chapman, M. G. Bartlett, R. N. Harris, 2004. Comment on 'Ground vs. surface air temperature trends: Implications for borehole surface temperature reconstructions' by M. E. Mann and G. Schmidt. Geophysical Research Letters, 2004, 31, doi:10.1029/2003GL019054.)
Your lifeboat for them is taking serious water.
Posted by: Robert Ferguson at July 6, 2005 10:50 AM
Bob (and Jo)- I agree with these sentiments. But what you are calling for does in fact represent a change to the norms of climate science (see the quotes I cited from von Storch, Annan, and Matsuda. See also Ron Errico in BAMS.) I'll say it again -- invoking the norms of climate science is not an excuse or justification (or a lifeboat) for the actions of MBH or MM, it is simply offered as an explanation for why the debate has evolved as it has. Should the climate science community and the IPCC be more rigorous? I don't see how one can look at this case an come to any other conclusion than "yes". See this post on this subject:
Posted by: Roger Pielke Jr. at July 6, 2005 11:10 AM
"Note that no matter what it is that MBH are not revealing, it could have no consequences for either the scientific consensus on human influence on climate or perspectives on recent temperature trends in historical perspective, as these issues are supported by a larger literature."
I'll see your unsupported, absolutist assertion of unknown and unknowable fact, and raise you:
"The scientfic consensus of significant/severe/catastrophic anthropogenic global warming is, in fact, a strong assertion by a few with an economic or more likely philosophical interest in such, supported by the silence of the many who choose to not endanger their careers."
See? Playing without 'norms' is fun.
Posted by: Jim Carson at July 6, 2005 11:14 AM
Jim- Thanks for your comment, which if I understand correctly, suggests that you might find this paper of interest:
Posted by: Roger Pielke Jr. at July 6, 2005 11:33 AM
Well, I hope you're correct, Roger, and I also hope that this particular controversy--no matter what the outcome--provides incentives for climate scientists to be more transparent in the future.
Having read your linked article of February 14, I now have a better understanding of your position regarding the expected norms in climate science, though I am honestly shocked--even outraged--at the blase attitude exhibited by Matsuda in the comments section of your July 3 Hockey Stick article. Matsuda writes:
"...I think that McIntyre is too demanding about disclosure of program codes and input data used by scientists. Those scientists who can put all codes and data publicly available, such as Mann (even if partly due to pressure from McIntyre), are admirable, but actually exceptional. Part of the software I use may be proprietary and not readily sold as commodity either. Another part may have been written by my fellow scientists who feel free for me to use it but not releasing it publicly. Another part which was written by myself is not yet well documented, and I feel giving it away without appropriate documentation is prone to misuse. Thanks to open-source software movement, some scientists may be able to conduct all computation on open-source software, and also to make their own software open-source. I think we should encourage such movements, but I do not think we can make it norm of scientific community. "
No. No, no, no. Look: this issue is not about proprietary vs. open-source software, and it's disingenuous in the extreme to attempt to represent it as such. The MBH code, IIRC, is written in FORTRAN, and there are no restrictions whatsoever on making it publicly available. For example, my master's thesis reported the results of an energy system model, written partially by me using MATLAB code, and there's simply no way The Mathworks would come after me if my co-authors and I decided to make it publicly available (though we're not exactly being deluged with requests, if you know what I mean). The "appropriate documentation" rationale looks like another furphy, though a slightly less risible one: we're talking about what seems to be a relatively straightforward statistical algorithm here, not a GCM.
Matsuda also says:
"Usually, reproducibility as a norm of science does not mean exact reproduction with the same data and the same program ... I think that reproducibility in science means that the same general conclusion can be drawn from similar but not exactly the same data. It is true that we think such studies that inhibit access to original data less credible."
One is tempted to reply, "Well, duh," but Matsuda's argument deserves to be taken just a tiny bit more seriously than that. I think it's clear that, for the past 18 months or so, Steve McIntyre has been doing precisely what Matsuda describes: attempting to replicate the MBH results from the crumbs of information they deign to provide at irregular intervals, and he's had some quite remarkable success at zeroing in on the likely errors in the MBH algorithm. (I'd also like to ask Matsuda: What happens if you arrive at DIFFERENT general conclusions from "similar but not exactly the same data", as McIntyre has done?) Has this resulted in newfound respect for McIntyre from the climate science community? You know the answer: the closer he comes to figuring out what MBH have done, the more bizarre the defences from the climatology folks. McIntyre is intimidating Mann, he's trying to "besmirch" real scientists, the ownership details of his website are "mysterious", and so on. As to the actual substance of McIntyre's work, Mann's defenders--far from being impressed by McIntyre's bottom-up work--seem sometimes to be arguing that his reconstruction is flawed because he doesn't have access to Mann's algorithm, truly an impressive exercise in missing the point.
Besides all of this, Matsuda's argument (as well as those of von Storch and Annan) is pretty much a defence of rank amateurism in scientific practice. I know that researchers are often disorganized--I used to be one myself, and I'm pretty messy--but that is no defence against the general principle that published scientific work should be fully transparent and replicable. In cases where that's not possible (when reporting results from very complex GCMs, for example), the extent of the opacity should be disclosed up front as a caution to readers that the results are not vettable.
Too often in this debate, the climate science folks and their defenders have behaved as if their work should be exempt from scrutiny, and that laypeople should take their pronouncements on faith alone. If in the end it turns out they've done nothing more than bring this attitude out into the light, for this alone M&M should be thanked.
I've read your paper and agree that it is interesting. I do not agree, however, with your implied premise that good science has produced the somewhat uncertain result that significant AGW exists. I think sloppy, politically-motivated science has produced this.
Now that I've responded to your suggested reading, perhaps you will admit to the trivial error of overstatement of which I've accused you? Such admission would increase your credibility, unless your claim is infallibility.
As for your pleas to policymakers (U.S. Congress, in our case), how about using your considerable influence to advocate something on which even we skeptics can get on board? I propose that you lobby Congress to set up a prize--perhaps US$100 million--for the first team to develop a GCM which can accurately reconstruct the last century or so. When this GCM accurately predicts global climate for the next decade, the prize is awarded. Of course there will be plenty of arguments on parameters, measuring points, etc. But even a libertarian like me can support this idea--I see it as a chance to spend $100 million to avoid spending trillions.
Posted by: Jim Carson at July 6, 2005 12:48 PM
Jim- Thanks much for following up. I am not sure what error you are referring to (what "fact"?). Do the substantive conclusions of the IPCC depend upon MBH in any significant way? No. (Even the RealClimate folks have stated this.) This underscores the mistake that the IPCC made in so prominently emphasizing MBH in the TAR SPM. Is MBH a powerful symbol that represents, for better or worse, the corpus of climate science? Yes.
As far as influence, thanks for the kind words, but if there is one thing that we have proven in our efforts it is how little impact policy researchers actually have in this issue!
We did make the case for prizes in space flight a while back here, and I generally think that it represents an innovative approach to certain aspects of science policy. On GCMs in particular, I have in the past been pretty critical of prediction-based decision making, and like a professor I am going to point you to some more reading:
Sarewitz, D., R. A. Pielke, Jr., 2000: Breaking the Global-Warming Gridlock. The Atlantic Monthly, 286(1), 55-64.
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.
Posted by: Roger Pielke Jr. at July 6, 2005 01:12 PM
Some other prominent scientists are even worse at archiving data and methods:
1) Briffa refuses to identify the 387 sites used in Briffa et al . I've tried on a number of occasions and Briffa does not respond.
2) Esper refuses to provide or archive tree ring data used in Esper et al . Again I've had no response.
3) Crowley has "misplaced" the original data for Crowley and Lowery 2000.
3) the tree ring data at S.O. & P imput to the IPCC AR4 is password-protected. Why should tree ring data be confidential?
4) Jones has refused to archive the station data used in the CRU temperature index
5) Jacoby has selectively archived "temperature sensitive" tree ring sites and has archived hardly any North American measurements taken in the past 15 years.
6) Thompson had not archived any information on Himalayan ice cores until very recently, when he did so recently but very inadequately as a result of my requests to Climatic Change. What if his lab burned down and his unarchived data were destroyed.
As I understand U.S. policy on data archiving, where U.S. federal funding is involved, data archiving is an obligation, not an option. Why isn't the NSF ensuring compliance? Why isn't anyone talking about this?
Posted by: Stephen McIntyre at July 6, 2005 01:26 PM
Steve- Thanks for your comment. You have good evidence here for a systematic problem with respect to NSF policies. As you probably well know, oversight is typically not a strong function of government, and NSF is no different. Why not write up in a short essay all of these examples, cite the relevant NSF policies, describe the disconnect that you have found, and then submit as a brief commentary to a journal like the Bulletin of the AMS or AGU's EOS? It won't change the world, but it will force some sort of a response to your claims, and get you on the record within the community. (For better or worse, journals are the media of communication that matter most in science.) It would be a small marginal effort compared to what you have already done. By the way, if Barton had only written his letter to Bement, then he would be on considerably more solid ground, as his job is to oversee federal agencies (even if NSF is the jurisdiction of House Science). Here is the Bement letter:
Posted by: Roger Pielke Jr. at July 6, 2005 01:50 PM
In academic parlance, let's leave the validity of your statement to the reader as an exercise. I am unable to argue that the sky is blue.
If you discount GCM's, is there any test of the AGW hypothesis that you could proffer that might satisfy us skeptics? One that tests significant, not trivial, AGW of course.
Posted by: Jim Carson at July 6, 2005 02:04 PM
Stephen – even though I've argued previously that researchers should be willing submit to review of their work, what you're describing is a fishing expedition. Let's turn it around: if I were to write to every researcher in the oil exploration industry, from government to private to academic, and ask all of them for their data and methods, code, well logs, maps, etc. for the past ten years in order to see for myself if Hubbert's Peak is really upon us (which I think has bigger implications for future energy policy than climate change does), what response do you think I should expect? Immediate and unfettered access? Not a chance.
But I appreciate that you have left it at "some other prominent scientists are even worse at archiving data and methods" without directly accusing them of hiding something.
Everybody else: Geoscience research, especially the analysis of multi-component data sets with thousands – to – millions of samples, is very messy. From experience, what goes into a final published paper is a half-full harddrive of files ending in .m .c .o .ps .eps .dat that represent the potentially hundreds of avenues you traveled to get to your final paper. In dismissing the significance of the norms question that Roger poses, you're asking for a scientist to simultaneously be both a creative researcher and a detailed archivist. There is a reason that those two jobs exist in our socioeconomic system, and there is a reason that people who choose either have very different personalities.
For those of you who were in Congress let me give what I think is an entirely appropriate analogy: your boss was the main sponsor of a law that passed in 1998. You wrote the original bill starting in 1996 and were the main shepherd all the way through conference committee. Based on an article I just read in the American Prospect, it turns out the premise on which this law is based has been questioned. Please go back and reconstruct the entire bill-writing process, providing your notes of all conversations you had from any stakeholder while you were drafting the bill, all notes from any meetings with lobbyists and constituents, all notes from all meetings with other staffers from the authorizing committee, all conversations that led to the three hearings on your bill, how you chose the witnesses who testified at those hearings, the process that went into the Chairman's Substitute at markup and how much influence any stakeholder community had in altering that mark, the information you provided to other staffers prior to committee markup, etc.... Even though we realize that it took 2 years and 3 months from initial draft to President's signature, we expect all this information by July 16th. Thanks!
This is not to apologize for MBH, this is to ask those of you who aren't geoscientists but are finding it very easy to criticize to step back and have some understanding here.
Posted by: kevin at July 6, 2005 02:05 PM
Jim- Thanks, you ask "... is there any test of the AGW hypothesis that you could proffer that might satisfy us skeptics?" I doubt it. You either find the evidence compelling or you don't. But warts and all, the work reported in the IPCC (which is a great single source to much of the literature) makes a case for a human impact on the climate system. The US NRC recently made the case for a broader perspective on human influences, here -- http://books.nap.edu/catalog/11175.html. When energy balances in a system are influenced it seems almost trivial to suggest that there will be effects. Whether or not such effects will be significant or not is of course an important question, not least because the notion of "significant" is a value judgment.
But here we go getting sucked back into the maw of a scientific debate which is, from where I sit, largely irrelevant to action (or worse yet, a distraction from actions that make sense independent of the science). The work that I have focused on for the past decade emphasizes policy actions that are robust to disagreements on such scientific questions. Have a look at the articles I suggested, start with the Atlantic Monthly piece and browse through our climate archives on this blog. The great tragedy of the climate issue is that many adaptation and mitigation actions make sense on their own merits, regardless of one's views about how the future will pan out.
Posted by: Roger Pielke Jr. at July 6, 2005 02:24 PM
So, we have a set of policy prescriptions, and some science (bad or not) will be used to have these shoved down our throats?
Is that the correct understanding?
Posted by: Hinheckle Jones at July 6, 2005 04:31 PM
Just some comments on snips from the above:
"Do the substantive conclusions of the IPCC depend upon MBH in any significant way? No."
"Let's turn it around: if I were to write to every researcher in the oil exploration industry, from government to private to academic, and ask all of them for their data and methods, code, well logs, maps, etc. for the past ten years in order to see for myself if Hubbert's Peak is really upon us"
"But warts and all, the work reported in the IPCC (which is a great single source to much of the literature) makes a case for a human impact on the climate system."
Posted by: Murray Duffin at July 6, 2005 07:47 PM
"The great irony here is that in many instances the supporters of the hockey stick have often been their own worst enemies and fed the flames of this debate, which now threatens the integrity of all of climate science, and to turn all of climate science into climate politics."
The "integrity of climate science" is only *now* being "threatened?"
What about the IPCC Third Assessment Report's (TAR's) projections for methane atmospheric concentrations, CO2 emissions and atmospheric concentrations, and resultant temperature increases, which were released to the public way back in 2001?
Specifically what about the facts that:
1) The most outrageously high projections were apparently added AFTER scientific peer review?
2) The methane atmospheric concentrations are projected to increase at pre-1990 rates, even though the IPCC TAR was published in 2001, when it was quite clear that atmospheric methane concentrations were plateauing?
3) The "scenarios" and "projections" come with explicit IPCC disclaimers against ANY associated estimations of probability. (Without any associated estimates of probability, the "projections" are simply pseudoscientific rubbish,…similar to the "projections" in "Limits to Growth" and "Beyond the Limits".)
In my opinion, the IPCC TAR's projections for atmospheric methane concentrations, CO2 emissions and atmospheric concentrations, and resultant temperature increases constitute the greatest fraud in the history of environmental science. I hardly think that the "Hockey Stick" is a NEW threat to the integrity of climate science. Given that the IPCC TAR's projections have been now been out for ***4 years*** without being thoroughly denounced, I question whether the IPCC has any scientific integrity left to threaten.
Posted by: Mark Bahner at July 6, 2005 07:59 PM
Roger Pielke wrote, "Do the substantive conclusions of the IPCC depend upon MBH in any significant way? No."
Murray Duffin responded, "What are the substantive conclusions?"
The ***most*** substantial conclusion, bar none, is that the earth will warm by "1.4 to 5.8 degrees Celsius" unless governments act.
And by far the most important number in that range is the UPPER number of "5.8 degrees Celsius"...because almost everyone would agree that, if the warming is at the lower value of 1.4 degrees Celsius, it's not a terribly pressing problem.
So that "5.8 degree Celsius" projection is absolutely critical. And the 5.8 degrees Celsius value is also absolutely fraudulent, pseudoscientific rubbish.
Posted by: Mark Bahner at July 6, 2005 08:13 PM
Kevins comment "For those of you who were in Congress let me give what I think is an entirely appropriate analogy: your boss was the main sponsor of a law that passed in 1998...(snip)... we expect all this information by July 16th. Thanks!"
If you wish to imply Kevin that the MBH paper was primarily a political exercise, then your analogy is reasonable. Since it is being treated as a scientific paper, one is entitled to expect a different approach are we not ?
Better example, I have a neat experiment using a special form of palladium electrodes in a weak H2O2 solution that produces an excess of energy and some thermal neutrons. You cannot reproduce the result, but that is, I claim, because you are not using exactly the same sort of palladium electrodes, and your solution is not quite right. However I refuse to say exactly what the electrodes are, or what the solution is, except that they are "conventional". If I was funded by public money to carry out the original experiments, would the government be entitled to the appropriate details of the experiment ?
Posted by: Ed Snack at July 6, 2005 09:36 PM
Murray wrote "Most AGW supporters, modellers and the IPCC selectively ignore possible warming causes like solar magnetic activity reducing cosmic ray penetration of the atmosphere, referring to solar forcing only and always only in terms of irradiance. "
IPCC does include a page on this hypothesis (188.8.131.52)
Lacking data in support for the cosmic ray hypothesis that has been floating around for at least three decades, what more can IPCC do?
Another thing is worth mentioning because so many get it wrong: MBH 98 was published *after* the Kyoto treaty was drafted and thus can't have influenced it. Those who wish to attack the Kyoto treaty will have to find another target.
Posted by: Thomas Palm at July 7, 2005 03:43 AM
Posted by: Jeff Norman at July 7, 2005 08:11 AM
Thanks for your comments. In reply to your question, yes, "discerible influence" is the language that the IPCC uses and is based on a literature that I would characterize as wide and deep. As far as paragraphs, I'll defer to Shep or Genevieve, but I think it has somethng to do with controlling spam.
I am going to more rigorously follow my own advice around here and not enter into scientific debates. My case is that action on climate change (and for what I mean by action, read my papers, start here http://sciencepolicy.colorado.edu/admin/publication_files/resourse-1761-2005.32.pdf) can and should be independent of all of these diversionary debates on climate science, enjoyed a great deal by both sides. In other words, if you think the IPCC is right or wrong, I don't care, I can make a case for actions that do not depend on this difference. This is the proverbial third way.
Posted by: Roger Pielke, Jr. at July 7, 2005 08:29 AM
Nature has an interview with the IPCC's rajendra Pachauri.
Here are relevant excerpts:
"Do you feel obliged to respond?
I will first consult my colleagues in the IPCC. Over the next days we will decide whether and how to react. We might not do anything at all.
What kind of information would you consider providing?
I would not hesitate, out of courtesy, to provide basic information about how the IPCC functions and about the manner in which we choose our authors. This is a well established and absolutely transparent process. The only criteria are scientific merit and integrity. I don't think we need to provide more information than that. I guess it will be sufficient to mention the processes and procedures of the IPCC and to refer the committee to our website."
Not so good:
"Was it unwise to give Mann's 'hockey stick' so much prominence in the IPCC's summary for policy-makers?
No. It is no exaggeration and it doesn't contradict the rest of the IPCC assessment. Of course you can always argue about details. But we assess all the available literature, and we found the hockey stick was consistent with that."
The fact that he is even being asked this question should be enough for him to offer a more nuanced reply. How about: "I am sure that at the time we would have done nothing different. but with the advantage of hindsight, it is clear that by emphasizingto a single study we opened the door for a narrow debate, when the IPCC presents a large body of literature to reach conclusions that do not depend critically upon any one single paper."
"Do you think individual scientists such as Mann need to be better protected against pressure from politicians?
The IPCC cannot do that. But Mann and his colleagues are distinguished, independent scientists who are able to explain their points of view. These letters don't curb their independence. And the recipients don't need to provide all the information requested. By and large, I don't regard this as a threat to the scientific community."
Posted by: Roger Pielke Jr. at July 7, 2005 09:28 AM
This fellow "John A". Might he be the same person who is the web administrator for the web page of Stephen Myintyre ( of M&M fame)? If so maybe he should have a neat byline, like "from the desk of Steve Mcintyre". I just hope no reference is intended in the username to our first and greatest Prime Minister, John A Macdonald. He had his own ideas.
Posted by: garhane at July 7, 2005 01:12 PM
And just a little more. Is that you, Professor Pielke, the one concerned with policy, advising a bunch of pointy heads to tell your government, you know, the place where the grants come from, the place where they all stood on the steps when the Arabs invaded singing your national anthem....that outfit, to stuff it?
For the rest one cojld recall that Mckintyre as much as said he thought the stick was like a mining scam, something he knew about from his business history. So where he says "replicate" it seems pretty obvious he wanted to get the raw original information and use it to pin Mann's hide to the barn door. Mann seems to have been smart enough to send him running all round Robin Hood's barn to find the door still locked. Now he wants "code" and there is a great confusion about what "code" is that. Everything else is pure grandstanding, and there is a great deal of that in his endless demands for information, and tedious posturing about what scientists are supposed to be doing, framed by the ludicrous standard of perfection he would use to defame them, and widening suspicions which already encompass all the other climate scientists Mann has ever shared a coffee with, so it seems, and now von Storch. The whole story is hilarious (not tragic) The researcher regression is not exactly a new phenomena is it? Soon Mcintyre will be claiming he has not received something he demanded two years ago (who is he to demand or require anything no one has a public duty to give him?)
Do you not agree when a scientist says "replicate" he probably means reproduce by other means or methods out of which new science may emerge, since all you get from true replication is a statment that the other guys did what they said they did (thank you, Robert, who posted this idea in a page I have forgotten).
So let Professor Mann respond as he may be advised by comptetent legislative counsel, and once the stage is set after a suitable campaign , wipe the floor with the dismal crowd of deniers, the distraction has gone on long enough.
Posted by: garhane at July 7, 2005 04:03 PM
Thank you for your post. It seems to me that there are three separate issues here.
1. Reproducibility of results and the "free rider" problem. You are correct that this has been a fraught issue in environmental science as opposed to other disciplines, though the situation is changing. The problem is that both climate data and models are labor and time-intensive to run. Many members of the community hold on to their results and only release them to certain collaborators- in part so that they can get credit. By contrast, in a laboratory setting, you are expected to be able to reproduce answers because you can collect your own data. Instead, investigators in climate science are confronted by people who simply want the end results of years of research. Some researchers respond to this by releasing their data to others who will look at it in a synthetic sense (a la Mann) but not scoop them on results based on the individual records. While the situation is changing- more and more of the models and the data are available online, the fact remains that reproducibility in a science where there is only one realization of the thing being studied is a different beast. I think Mann et al. should make their original programs and data available- but this will not be possible in all cases. Just as a researcher shouldn't be obliged to turn over his machine to a competitor for free just so that person can reproduce his results.
2. The nature of scientific advice. Let's say you have cancer and want the best available care. Do you go for the "community standard" or the experimental treatment recommended by the top specialist in the field. Seems to me the answer is going to depend on your personality. There is no right answer, and both answers involve separate risks. The IPCC has to balance using results that are out of date and inconclusive with using the latest results that may be wrong.
Posted by: Anand Gnanadesikan at July 7, 2005 04:21 PM
With respect to point 1 I think that is a furphy. Economics, like climate science (particularly paleoclimate science), is a non-experimental science where there is (typically) no laboratory available to replicate results. Despite this, it is standard practice in economics to make full code and data available to all others on publication of a paper. Indeed, this is a condition of publication for many of the best journals. Thus, I would argue that you are wrong to say that "reproducibility in a science where there is only one realization of the thing being studied is a different beast".
Furthermore, I would argue that if climate science is being driven by selfish motives and a quest for fame then it has substantial inherent problems. Also, talk of other scientists as competitors is a fundamentally flawed premise for science. People are people and this will always occur to a degree, but it is inimical to the proper practice of science.
Posted by: John S at July 7, 2005 04:59 PM
Sure we can -- if your position is that the IPCC has handled the HS controversy perfectly, then yes we'd disagree.
In your professional judgement, how big a part of the issue of D&A is the HS? 1%? 10% 37.3%? My understanidng of recent comments is that without the HS the IPCC consensus would not be weakened one bit ... do you agree?
Posted by: Roger Pielke, Jr. at July 27, 2006 12:38 PM