James Powell is wrong about the 99.99% AGW consensus

This is reposted from Critical Angle with slight modifications and updates.

In a recent article in Skeptical Inquirer, geologist and writer James Lawrence Powell, claims that there is a 99.99% scientific consensus on Anthropogenic Global Warming (AGW). You might think that after all of the harsh criticism that the 2013 Cook et al. paper (C13) has received from climate contrarians that we would be pleased to embrace the results of a critique that claims we were far too conservative in assessing the consensus. While it certainly does make a nice change from the usual rants and overblown methodological nit-picks from the contrarians, Powell is wrong to claim such a very high degree of agreement.

He makes many of the same errors that contrarian critics make: ignoring the papers self-rated by the original authors; and making unwarranted assumptions about what the “no-position” abstracts and papers mean.

Powell’s methodology was to search the Web of Science to review abstracts from 2013 and 2014. He added the search term “climate change” to the terms “global climate change” and “global warming” that were used by C13. He examined 24,210 papers co-authored by 69,406 scientists and found only five papers written by four authors that explicitly reject AGW. Assuming the rest of the abstracts endorsed AGW, this gives consensus figures of 99.98% (by abstract) and 99.99% (by author).

His definition of explicit rejection would align roughly with the seventh level of endorsement used in C13: “Explicitly states that humans are causing less than half of global warming”. In the abstracts from 1991-2011, C13 found 9 out of 11,914 that fit level 7, which using Powell’s consensus calculation assumptions, would yield 99.92%. So, there is probably not much difference between the two approaches when it comes to identifying an outright rejection paper. It’s what you assume the other abstracts say—or do not say—that is the problem.

C13 also counted as “reject AGW” abstracts that: “Implies humans have had a minimal impact on global warming without saying so explicitly, e.g., proposing a natural mechanism is the main cause of global warming”. These are more numerous than the explicit rejections and include papers by scientists who consider that natural causes are more important than human causes in recent warming, but who do not outright reject some small human contribution.

\

Competing Climate Consensus Pacmen. Cook on the left, Powell on the right.

Perhaps the simplest argument that shows that Powell is wrong is that surveys conducted in the past ten years consistently show a stubborn, small minority of scientists who dismiss the mainstream expert view on AGW. These studies include polls of scientists, analyses of the published literature and examination of the public statements of scientists (see the references below). In a new paper that is currently in press we found that consensus increases with the degree of climate science expertise of the populations studied. When the studies are limited to publishing climatologists, the consensus ranges from 84- 98%. The specific criteria chosen for what constitutes endorsement or rejection of the consensus also influences outcomes.

There are a few scientists—and even a very few who actively publish in the peer-reviewed literature on climatology—who reject or play down the human role in recent climate change. In the second part of the C13 paper, we asked the authors of the articles whose abstracts we had analyzed to rate their own papers. We received self-ratings on 2141 papers, among which 39 (1.8%) were self-rated as rejecting AGW. Of the 1189 authors who responded, 28 (2.4%) wrote papers that rejected AGW to some degree or other. The dissenters are but a small percentage of the many thousands of scientists working on climate change. They may or may not be doing good science, but it would be foolish to deny that they exist.

The no-position abstracts and papers

Powell’s main beef is that we ruled out of the calculation of consensus the two-thirds of the abstracts that did not take a position on AGW. Since the analysis of the abstracts was limited to the text, we could not guess what the non-expressed opinions of the authors were. Powell:

[James] Hansen had a total of six articles in Cook et al.’s “no position” category. A number of other prominent climate scientists show up there as well. These include (with the number of articles): R. Bradley (3), K. Briffa (2), E. Cook (5), M. Hughes (2), P. Jones (3), T. Karl (5), M. Mann (2), M. Oppenheimer (3), B. Santer (2), G. Schmidt (3), the late S. Schneider (3), S. Solomon (5), K. Trenberth (7), and T. Wigley (3). Cook et al. ruled them all out of the consensus calculation.

It is maybe worth noting that the fact that we classified abstracts by many prominent mainstream climate scientists as “no-position” rebuts the notion that we regularly cheated by looking up the authors of the abstracts and classifying them accordingly.

Powell continues: (with my emphasis)

Most of these authors, like Hansen, also have articles in one or more of the three endorsing categories. Again, we see that the Cook et al. method is about language and the subject of articles rather than whether their authors accept AGW.

Bingo. It wasn’t an opinion poll and we didn’t try to guess what the authors think about AGW generally, we just relied on their specific abstracts. 

Many papers on the impacts of global climate change did not mention a human cause. A good number of papers on impacts looked only at local or regional—not global—climatic change. The majority of the paleo-climate papers did not mention the modern era at all.

By assuming that “no-position” abstracts or papers are tacit endorsements, Powell makes the same error that contrarian critics make when they claim that the “no positions” count as rejections or don’t-knows. By making such assumptions you either end up with results that the consensus is implausibly large or absurdly small.

Powell, arguing from personal incredulity:

Since it is inconceivable that any climate scientist today could have no opinion on the subject, if 97 percent accept AGW it follows that 3 percent reject it.

Studies like Doran and Zimmerman (2009) and Verheggen et al (2014) do actually get a few “don’t know” answers from samples identified as climate scientists. Some of these “don’t knows” may in fact be “can’t say” answers to questions that the scientists think were poorly framed. But it is surely likely that, in a sample of scientists that contains a minority that outright denies the major human influence on climate, there may also be a few who are genuinely uncertain about it.

Powell:

A scientist who has evidence that AGW is false will be eager to say so and to present that evidence. Who among us would not love to be that scientist!

This is a valid point. Mainstream scientists will be inclined to reserve the limited space of their abstracts for reporting novel results, therefore, statements of the obvious (endorsement of the consensus) may well go unmentioned. On the other hand, a result that goes against the grain of the consensus is, by definition, a novel result and is more likely to be explicitly reported. The C13 methodology, therefore, may well have systematically overestimated the relative level of rejection of AGW. (There was at least one exception to this, though, see the Footnote.)

Plate tectonics

Powell examined 500 recent articles on plate tectonics and found that none of them explicitly endorsed the theory and none rejected it. He claims that if you applied the C13 methodology to plate tectonics you would get a meaningless result.

I did my own little survey on a smaller, not very representative sample. I also found no explicit endorsements and no rejections at all. I was generous with what I considered to be an implicit endorsement, as I wrote: “It was sufficient for the abstract [to merit an implied endorsement rating] only to mention a plate tectonic process (e.g., displaced continents, sea-floor spreading, continental collisions etc.).” Even so, this only amounted to 5 out of the 65 articles that I looked at, about 7%. Nevertheless, if you include implicit endorsements as C13 did for AGW, the methodology works just fine for plate tectonics and produces a 100% consensus.

Anecdotally, I don’t recall reading a single paper that doubts plate tectonics that was published in the forty years since I graduated in geology. The last geologist I met who was a continental drift denier was a professor who taught my introductory geology class at the University of Sheffield in 1972. Of course, there’s no doubt any more, plate tectonics is accepted by virtually everyone and I wouldn’t quibble with anybody saying that the scientific consensus on plate tectonics approaches 100%. There were a few stubborn drift deniers who hung on for a a decade or two after the late 1960s (particularly in the Former Soviet Union), but they are nearly all dead now.

 

However, in the public domain, plate tectonics is not like climate science. There has never been any political or ideological controversy about moving continents, even as the sometimes bitter scientific debates raged. There were no newspapers running editorials talking about the lack of consensus, no politicians denying the science because they can’t or won’t accept the policy implications, and no large industries and political parties deliberately manufacturing doubt. Nobody stole the geologists’ private communications looking for damning comments that they could take out of context and use to falsely accuse them of fraud.

The plate-tectonic deniers after the late 1960s were mostly older geologists who had invested their careers in developing geological models that had quickly become obsolete. A certain stubbornness is to be expected among curmudgeons whose students suddenly claim to know more about their subjects than they do.

In contrast, some of today’s AGW contrarians are non-climatologists and were drawn to the subject because of the political implications of emissions mitigation. There are also a few real climatologists who object to the current consensus, not because it threatens the basic physical models that they are familiar with, but because they think that the AGW prognosis is overstated and less certain than is generally reported. These scientists are few. Some of them may well be politically or religiously motivated, many of them may enjoy the notoriety that taking a contrarian stance brings, a handful may be paid by industrial interests to produce contrarian “deliverables. But they exist.

No true Scotsman

There may not be many scientists who doubt the human cause of recent climate change but, because of politics, their influence is exaggerated and the public has been quite deliberately misled about the level of consensus in climate science. Still, I think it is counter-productive to claim that the number of dissenters is near-zero.

To claim that only 1 in 1000 or 10,000 experts rejects or is uncertain about AGW can only be justified by classifying the doubters as not true climate scientists. This is similar to the well-known logical fallacy of “no true Scotsman” that redefines the credentials of someone based on his views or actions.

The names of the few dissenters are relatively better known to the public (thanks to Rupert Murdoch and the Republican Party) than the many thousands of climate scientists who toil away in relative obscurity.To claim that thedoubters are an insignificant minority invites disbelief.

As a young researcher I remember the advice of an older geophysicist from the University of Glasgow, Adam C McLean (a true Scotsman), who counselled me and others never to exaggerate our arguments, because it makes them easier to contradict. (To this day, I try to screen everything I write, cutting out all of the adverbs that I can, in particular, “very”.)

Powell argues that if there were a small percentage of dissenting scientists, even as low as 3%, the public perception would be that those scientists could turn out to be a group heralding a coming paradigm shift in climate science. Indeed, that’s how some of the dissenters portray themselves, as Galileos fighting the orthodoxy. The AGW consensus is not, however, based on an immutable sacred text, but rather is a consequence of a consilience between constantly updated, multiple lines of evidence and basic physical theory.

Moreover, the would-be Galileos are mutually incoherent. They do not present an alternative scientific model in the way that the fixists and mobilists did in the continental drift debate. Rather, contrarians put forward a number of different objections such as: “it’s the sun”; “it’s ocean cycles”, “it’s undersea volcanoes”, “it’s planetary cycles”, “it’s cosmic waves”; “the CO2 comes out of the ocean”; “it’s all too uncertain” and so on. They disagree as much among themselves as they do with the mainstream science.

It is certainly true that the public has been misled about the scale of scientific dissent on AGW. The most vociferous critics of C13 are those who like to portray the dissenters as a substantial and repressed minority, not as a 3% fringe group. However, whether the dissenters make up 10%, 3% or 0.001% is immaterial: that’s still a marginal proportion in the public’s eyes. The contrarian pundits insist that the dissenting scientists make up such a large minority that AGW dissent should be given equal time in the press, congressional hearings and the classroom. That’s nonsense, of course.

Powell claims that “…there is virtually no publishable evidence against AGW. That is why scientists accept the theory.” He is right, to the extent that much of the published contrarian literature is bad science, is infrequently cited and, in most cases, ought never to have been published. But he’s wrong to claim that “…99.99% of scientists publishing today accept AGW.” That is not true—although I wish it were—and it is foolish to make such a strong claim that can be so easily contradicted.

There is one positive thing to say about Powell’s work. Unlike the AGW-doubting critics of C13, he does not just talk about his problems with the C13 methodology. He backed up his criticism with replication work of his own and produced an independent estimate of the AGW consensus. Although I disagree with his result, at least he presented one.

References

Anderegg, W. R. L., Prall, J. W., Harold, J., & Schneider, S. H. (2010). Expert credibility in climate change. Proceedings of the National Academy of Sciences of the United States of America, 107, 12107-12109.

Bray, D (2010) The scientific consensus of climate change revisited. Environmental Science & Policy, 13, 340-350. doi: 10.1016/j.envsci.2010.04.001.

Carlton, J. S., Perry-Hill, R., Huber, M., & Prokopy, L. S. (2015). The climate change consensus extends beyond climate scientists. Environmental Research Letters, 10(9), 094025.

Cook, J., Nuccitelli, D., Green, S.A., Richardson, M., Winkler, B., Painting, R., Way, R., Jacobs, P., & Skuce, A. (2013). Quantifying the consensus on anthropogenic global warming in the scientific literature. Environmental Research Letters, 8(2), 024024+.

Doran, P., & Zimmerman, M. (2009). Examining the scientific consensus on climate change. Eos, Transactions American Geophysical Union, 90, 22.

Oreskes N. (2004) Beyond the ivory tower. The scientific consensus on climate change. Science, 306:1686.

Powell, J. (2015). The Consensus on Anthropogenic Global Warming. Skeptical Inquirer. Available at: 

http://www.csicop.org/si/show/the_consensus_on_anthropogenic_global_warming

Rosenberg, S., Vedlitz, A., Cowman, D. F., & Zahran, S. (2010). Climate change: a profile of US climate scientists’ perspectives. Climatic Change101(3-4), 311-329.

Verheggen, B., Strengers, B., Cook, J., van Dorland, R., Vringer, K., Peters, J., Visser, H. & Meyer, L. (2014). Scientists’ views about attribution of global warming. Environmental science & technology, 48(16), 8963-8971.

Footnote on over-reporting rejection abstracts

Astrophysicist Nir Shaviv complained (archived blogpost) that C13 misrepresented his paper, rating it as an explicit AGW endorsement (level 2), when in fact his work showed that part of the 20th Century warming was due to solar activity. He wrote (screenshot):

I couldn’t write these things more explicitly in the paper because of the refereeing, however, you don’t have to be a genius to reach these conclusions from the paper.

I’ll admit that his abstract fooled us, in the same way that his entire paper got past his reviewers. So, it is not always the case that contrarians will clearly report their novel results in an abstract.

This piece was earlier reposted on Facebook, where some readers questioned whether arguing over small differences in percentage consensus is worthwhile. To a degree, they are right. If people use a high percentage of expert agreement as a heuristic to help make up their own minds on AGW—or any other issue requiring specialized knowledge—they may not care if it is 90% or 100%. However, Powell's criticism is quite plain and, clearly for him, the difference between 97.1% and 99.99% is significant: "If there were a 3 percent minority on AGW it would matter, but there is not. The “97% consensus” is false." To date, no response has been published in Skeptical Inquirer, although there is one in the works.

This week, there will be a peer-reviewed response to a critic claiming that the 97% figure is too high (stay tuned). In the interest of balance, we need to respond not only to those who say we exaggerated the consensus on AGW, but also to those who argue we exaggerated the amount of dissent.

Posted by Andy Skuce on Tuesday, 12 April, 2016


Creative Commons License The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.