There is a scientific consensus that increasing greenhouse gas concentration in the atmosphere causes surface warming, and that CO2 is a major greenhouse gas. This consensus is based on physics. We don’t need to observe the effects of human greenhouse gas pollution to know this. There is consensus that human burning of fossil fuel causes an increase in CO2 in the atmosphere. We don’t need to observe this to know it, because we know how combustion works. But it is relatively simple to measure, and it has been measured, and it is true. There is consensus that the planet’s surface has warmed. This is expected from the physics and the fact that we are increasing atmospheric CO2, but it is also relatively easy to measure, it is measured, and it is true. There are varying levels of understanding the effects of this process, and varying degrees to which the effects of surface warming are thought to cause specific effects. One could probably characterize the scientific consensus as a widespread understanding that surface warming has had and will have a range of effects, with many of those effects being changes in weather patterns or regional climatology (how warm/cool/dry/wet a region generally is across he seasons) arising from a combination of “natural variability” (what would happen without greenhouse gas pollution) and anthropogenic global warming.

It is interesting, then, to see the results of various studies of scientific consensus related to climate change. Two kinds of studies have been done. One asks scientists what they think, the other reviews the scientific literature to see what the peer reviewed papers that address climate change say. In both cases we see a number between 90 (or, really, 95) and 100 percent agreement on the stuff in the paragraph above. It is not surprising that the vast majority of scientists, and the vast majority of research papers, have very similar things to say about climate change. This is not new science, and while climate is very complex, the basics of anthropogenic global warming are well understood. The results of empirical research closely match expectations derived from the physics. It all hangs together pretty well.

What is surprising is to see that 3-6% or so disagreement. Who are those scientists, why do they disagree, what do those papers say?

I would assume that since consensus research takes time, and often looks at several years worth of papers, that some of that non-consensus reflects older thinking and older research. Also, there are climate contrarians, including some scientists, who oppose the consensus for reasons not based on the science. That sort of denial presumably comes from the simple fact that some corporations or wealthy individuals will see reduced profits as we make the inevitable shift away from fossil fuels. So some of that non-consensus may be bought and paid for self interested maneuvering.

Rasmus Benestad, Dana Nuccitelli, Stephan Lewandowsky, Katherine Hayhoe, Hans Olav Hygen, Rob van Dorland, and John Cook, in “Learning from mistakes in climate research” (Theoretical and Applied Climatology) looks at the non-consensus peer reviewed literature.

The paper is here, and author Dana Nuccitelli has a writeup on the paper here. From the abstract:

Among papers stating a position on anthropogenic global warming (AGW), 97 % endorse AGW. What is happening with the 2 % of papers that reject AGW? We examine a selection of papers rejecting AGW. An analytical tool has been developed to replicate and test the results and methods used in these studies; our replication reveals a number of methodological flaws, and a pattern of common mistakes emerges that is not visible when looking at single isolated cases. Thus, real-life scientific disputes in some cases can be resolved, and we can learn from mistakes. A common denominator seems to be missing contextual information or ignoring information that does not fit the conclusions, be it other relevant work or related geophysical data. In many cases, shortcomings are due to insufficient model evaluation, leading to results that are not universally valid but rather are an artifact of a particular experimental setup. Other typical weaknesses include false dichotomies, inappropriate statistical methods, or basing conclusions on misconceived or incomplete physics. We also argue that science is never settled and that both mainstream and contrarian papers must be subject to sustained scrutiny. The merit of replication is highlighted and we discuss how the quality of the scientific literature may benefit from
replication.

The researchers found that cherry picking was the most common explanation for the non-consensus papers contrary results. In other words, it is not the case that a small number of paper simply found the physics, or some other aspect of, global warming to be different than other researchers found, or that they were looking at a part of the system that acts differently. Rather, these papers were wrong, and for a specific reason.

We found that many contrarian research papers omitted important contextual information or ignored key data that did not fit the research conclusions. For example, in the discussion of a 2011 paper by Humlum et al. in our supplementary material, we note,

The core of the analysis carried out by [Humlum et al.] involved wavelet-based curve-fitting, with a vague idea that the moon and solar cycles somehow can affect the Earth’s climate. The most severe problem with the paper, however, was that it had discarded a large fraction of data for the Holocene which did not fit their claims.

The authors attempted a replication of that particular research, and found that the model they used only worked for part of the underlying data. The data that were ignored by Humlum et al contradicted their findings.

Another problem identified by Benestad et al is the lack of a consistent sensible alternative explanation for their alleged findings. “…there is no cohesive, consistent alternative theory to human-caused global warming. Some blame global warming on the sun, others on orbital cycles of other planets, others on ocean cycles, and so on. There is a 97% expert consensus on a cohesive theory that’s overwhelmingly supported by the scientific evidence, but the 2–3% of papers that reject that consensus are all over the map, even contradicting each other.”

Go read Dana Nuccitelli’s post in The Guardian for more discussion of this interesting new paper. Also, the lead author has a post on this paper at RealClimate.

In that study, the authors analyzed “the evolution of the scientific consensus on anthropogenic global warming (AGW) in the peer-reviewed scientific literature, examining 11,944 climate abstracts from 1991–2011 matching the topics ‘global climate change’ or ‘global warming’.” They learned that “66.4% of abstracts expressed no position on AGW, 32.6% endorsed AGW, 0.7% rejected AGW and 0.3% were uncertain about the cause of global warming.” Among the papers that expressed a scientific position on the topic, “97.1% endorsed the consensus position that humans are causing global warming.”

The study was actually a bit conservative, as in order to be counted as part of that ~3% not supporting the consensus position on AGW a paper did not really have to be fully against the idea. Also, since the study was done, the consensus has increased. I asked study author Dana Nuccitelli about more recent changes in consensus, and he told me, “The consensus is growing over time, and reached 98% in 2011 (the last year included in our survey). So by now the minimizers/deniers are probably in the 1-2% range in the peer-reviewed literature (contrary to the ‘crumbling consensus’ claims).”

The other day I was giving talks at a local high school, and between classes, found myself chatting with a science teacher who had just completed a module on climate change and AGW. She asked me, “Isn’t there now research that shows that the consensus isn’t really as high as previously thought? Or is that bogus? Sounds bogus to me.”

Yes. Bogus.

I’m not sure what research the teacher was referring to (it was just something she had heard about) but there is a paper just published in “Energy Policy” by economist Richard Tol, who as far as I can tell has been a naysayer of climate science for some time now. Tol’s abstract says:

A claim has been that 97% of the scientific literature endorses anthropogenic climate change… This claim, frequently repeated in debates about climate policy, does not stand. A trend in composition is mistaken for a trend in endorsement. Reported results are inconsistent and biased. The sample is not representative and contains many irrelevant papers. Overall, data quality is low. Cook’s validation test shows that the data are invalid. Data disclosure is incomplete so that key results cannot be reproduced or tested.

Nuccitelli has responded to Tol’s paper, in a post at Skeptical Science called “Richard Tol accidentally confirms the 97% global warming consensus.”

Concern Tol-ing

Tol is practicing a special kind of science denialism here, sometimes called “seeding doubt” or as I prefer it, “casting seeds of doubt on infertile ground.” In other contexts this is called “concern trolling” or the “You’re not helping” gambit. The first of two paragraphs of the Conclusion section of Tol’s paper reads (emphasis added),

The conclusions of Cook et al. are thus unfounded. There is no doubt in my mind that the literature on climate change overwhelmingly supports the hypothesis that climate change is caused by humans. I have very little reason to doubt that the consensus is indeed correct. Cook et al., however, failed to demonstrate this. Instead, they gave further cause to those who believe that climate researchers are secretive (as data were held back) and incompetent (as the analysis is flawed).

Let’s get straight that Cook et al is not flawed, despite Tol’s complaints.

Tol’s main complaint is in the coding of the abstracts. He claims that it is imperfect. Well, duh. This is, essentially, social science research, and coding of text is imperfect. Tol makes the claim that the imperfections, if corrected, might bring the consensus down to a dismal 91%. I’m pretty sure he’s wrong about that, but if he is right, we are not impressed.

Tol’s key point is that the papers that are coded as not making a claim include some that do. He then incorrectly calculates how many of of those, if coded “correctly” there would be, and using this, downgrades the consensus to 91%

Nuccitelli explains in detail, in his post, how Tol’s re-analysis is badly done (see the amazing graphic at the top of this post) (go read it) and notes:

In reality, as our response to Tol’s critique (accepted by Energy Policy but not yet published) shows, there simply aren’t very many peer-reviewed papers that minimize or reject human-caused global warming. Most of the papers that were reconciled ‘towards stronger rejection’ went from explicit to implicit endorsement, or from implicit endorsement to no position. For abstracts initially rated as ‘no position,’ 98% of the changes were to endorsement categories; only 2% were changed to rejections.

Nuccitelli also notes that a separate study indicates that Tol’s method is flawed in the sense that no matter what data are used, the consensus will be decreased as an artifact of the methodology. Nuccitelli notes “…by making this mistake, Tol effectively conjured approximately 300 papers rejecting or minimizing human-caused global warming out of thin air, with no evidence that those papers exist in reality. As a result, his consensus estimate falls apart under cursory examination.”

Amazingly, when the Consensus research team fixed Tol’s methodology but applied the same question about coding papers in the no-position category, and re-calculated the percent consensus, it went up by 0.1%. Also, as Nuccitelli points out the Cook et al paper is not alone, and there have been a number of other studies that show essentially the same level of consensus among papers and/or scientists.

So, the consensus is real and isn’t going away. As is also the case with Anthropogenic Global Warming.

Had Peter Gleick obtained these documents using certain methods, and had he been a journalist, he would be up for a Pulitzer prize for investigative reporting. Had he obtained the corraborating evidence of Heartland’s unsavory strategies using a slightly different approach, he’d be fired by his editor. The thing is, Peter Gleick is not a journalist and it is absurd to hold him to “Journalistic Standards.” Peter is like the rest of us: He knows enough about the science, the politics, and the economics surrounding the issue of Anthropogenic Climate Change to have been very frustrated with the mindless zombie-like hate filled denialist movement, bought and paid for by the corporations and individuals with the most to gain from ignoring the science, to have risked falling on his sword for the benefit of the next generation. Thank you Peter. I don’t know yet if I will ultimately wish you hadn’t done this or not, but no matter what, there is a positive benefit to knowing the truth, and Heartland will never be seen quite the same way again in the future.

The documents themselves already showed a lessening of financial support for Heartland’s efforts to steer our national and international policies towards the cliff of unmitigated Anthropogenic Climate Change. Some of the donors, like Microsoft Corporation, were probably giving money to Heartland without realizing how bad an idea that was. Those donations will dry up. I’m told that some time today (maybe this has already happened) we will learn the name of the deep-pocketed “Anonymous Donor” who has focused his or her funding efforts on wrecking science education and similar activities. If that happens, regardless of that particular donor’s change of heart or lack thereof, other potentially “anonymous” donors may think twice before trusting their anonymity to Heartland, which appears overall to have certain limitations in the area of security and professional behavior.

The up side of all this is that we know more than we did before about important things. The downside is that much of the conversation in the press and the blogosphere over the last several days has been about the process and the activities of individuals and groups, not about the release of ancient Carbon into the atmosphere and its potentially dire consequences. Anthropogenic Climate Change is real, important, and needs to be addressed. Let’s do that.

Why is NCSE Now Concerned with Climate Change?

Press Release

Book excerpt: “The Denial of Global Warming” (from “Merchants of Doubt”)

Let me break it down for you

The paper, by Spencer and Braswell, was called “On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance” and it made the claim that the Earth’s atmosphere releases more heat into space than climate scientists had estimated, thus removing concern about the warming effects of fossil CO2 being released into the atmosphere. The following things were also true:

1) The paper was published in a journal, Remote Sensing, that normally does not address climate science, although there were some atmospheric scientists on the editorial board.

2) The authors, in particular Spencer, had a reputation for being “climate change denialists” which is not a kind of scientist, but rather, a politically motivated contrarian pretending to be a scientist, in this case with some scientific credentials.

3) Author Spencer was known to have made major mistakes in his research in the past.

4) The research in the paper had glaring errors, discussed in more detail below.

At the time, I wrote in a Research Blogging review of the paper:

“On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance” is a big ol’ bunch of hooey. I eagerly await an explanation from the journal’s editors, Dr. Wolfgang Wagner and Mr. Elvis Wang and the editorial board as to what they are up to with this paper.

Dr. Wagner’s resignation as Editor-in-Chief, which is available in print here (pdf), is a rather startling and definitive explanation! In short, the paper should never have been published.

What was wrong with the paper?

There were two major things wrong with the paper. First, the conclusion that the Earth’s atmosphere could not heat up with extra CO2 contradicted the very important facts that the Earth’s atmosphere has heated up and this heating up correlates to increases in atmospheric CO2 very much in the manner expected if the “greenhouse model” was correct. In addition, the basic idea of a greenhouse effect is pretty simple, solid, and well understood science. If something other than the greenhouse effect was happening, that would be major news.

But that sort of “flaw” — a claim that contradicts what we are very certain of — could be a virtue. A paper contradicting what everyone knows to be true would be brilliant, an amazing discovery, the stuff of awards and accolades. But, unfortunately for the paper’s hapless authors, there were other things wrong with it as well.

The numerical results presented in the paper lack statistical significance, but this is hard to detect because error bars or estimates of statistical uncertainty are presented poorly or left out. The methods used in the paper are not described well enough to verify that they could work.

When these results were examined more closely they were found to be not replicable.

The statistical strangeness of the results are explained in part by looking at the scale at which the work is being done. Standard climate models look at climate variables over various time scales from less than a decade to centuries of time. The Spencer and Braswell research inappropriately mixed time scales in a way that seems to have given them results they were looking for rather than a valid finding.

What they did, essentially, was watch a car veering towards the curb because it was trying to avoid hitting a cat, extrapolating the direction that car was moving at that moment to predict a long term pattern (which would put the car in a neighbor’s back yard rather than grocery store, where it was actually going).

In this case, Spenser and Braswell used observational data from a short time period (veering around the cat) in a model involving long term variation (the whole drive to the grocery store averaging out all the little backs and forths one effects while driving anywhere).

Kevin Trenberth and John Fasullo exposed this aspect of the work’s flawed nature in an essay posted here on Real Climate. The flaws of the paper are also discussed here and here.

Clearly the research was flawed. Likely, it was intentionally flawed to support an unscientific politically motivated denialist view. This would make the paper a scientific fraud. As Editor-in-Chief of the journal, Dr. Wolfgang Wagner might have seen himself a little like a bank manager who accidentally left the vault open so crooks could steal the gold, thus his resignation. But how much of a failure of the peer review process was this?

There is another element at work here, I think, that should be considered. Remote Sensing is one of a new breed of journal, called “Open Access” which has a very different model for how journals should work. It is, frankly, the much preferred model over the traditional way things are published, but the fight between “closed” and “open” styles of publishing has been rather vitriolic. Indeed, the term “Open Access” uttered in some academic settings will produce sneers and disgusted looks among those who don’t understand what it is or how it works. And, to make things worse, it may even be the case that there are some commercial Open Access journals that are over-commercialized (though I have no credible evidence of this at hand … it is just something that “people say” as far as I know).

Had a major well established traditional “Closed Access” journal published this paper, it is possible that the Editor-in-Chief of that journal would not have resigned just because of a major dust-up over one paper. However, in this case, it may have been necessary because of the somewhat tenuous nature of this sort of publishing venture. Dr. Wagner does not explicitly state this in his resignation but he does make direct reference to the challenges of earning a good reputation in the scientific publishing field and the qualities of the two and a half year old journal.

In the end, the peer review processed worked because a paper clearly recognized as something that should not have been published has been rather spectacularly identified as such with this resignation, which is published in the very same journal in the form of an editorial. It could make sense to also withdraw the paper but it may be the case that there is no mechanism for this. And, this is the scientific literature after all. The paper is a testament to the efforts, worthy or not, of its authors. It should stay there amid the literature surrounding it, for posterity.

There is an explanation for why this paper was published that applies generally to all bad papers as well as to good papers. The peer review process is designed to meet several different objectives. Relevant to the present case are two of them: 1) Filtering out true drek — A zoology journal would not even consider the latest summary of bigfoot sightings from the north woods, and a medical journal would not even consider a study comparing different ways to make healing solutions from homeopathic crystals; and 2) Ensuring the quality of the research itself, methodologically, logically, substantively, and so on with carefully done and thoughtfully managed peer critique.

The first objective is sometimes summarily met by editors who simply do not consider manuscripts that are inappropriate, or by reviewers to whom the manuscripts are sent. When one receives a manuscript there is the option to return it unreviewed or with a note that it is out of range for the publication being considered. (This step is often avoided by sending potential reviewers an abstract, asking if they would be able and available to conduct a review.) The second objective is met by having appropriate reviewers … people who know the relevant specialty and literature very well … carefully go over the paper and critique it, and along with the detailed critique, provide a recommendation about publication.

In this case, according to Dr. Wagner’s resignation letter, three reviewers looked at the paper and had only minor criticisms. Given that this paper is deeply flawed, this means that either the reviewers did not really look closely at the paper (meaning, frankly, that they did not do their jobs) or they are also climate denialists and this was all some sort of conspiracy. I can think of no other alternatives to explain this pattern.

How likely is it that a given reviewer would simply glance at a paper, pretend to have read it carefully, and send back a poorly done review having ignored the details? This is not likely but I would guess that it does happen. What are the chances that three reviewers would do the same thing, by random chance? Very very unlikely, but it is also possible that all the oxygen molecules in a room could randomly migrate to one corner, suffocating everyone present. Well, OK, the latter is significantly more unlikely, but the chance of three poorly done reviews happening at once for a paper is not large.

If there happen to be three bogus reviews from slacker reviewers, one would expect the editor managing the paper to notice this. There would be signs. The editors read the papers and must have some idea of the quality of the reviewers’ critiques when they come back.

However, there is another way that this could happen, if an editor is not really on top of the game, a way that reviewers (or some subset of them) end up providing an inappropriately positive ranking for a paper on purpose. The authors could have submitted the paper to an inappropriate journal but made a reasonable argument that the journal should publish it anyway. That leaves open the possibility of the authors writing their own ticket for passage through the peer review process.

Many journals allow, or even encourage, authors to submit names of potential reviewers. For that matter, authors can submit names of people who either should not review a paper, or if they do, should be watched closely by the editors because of potential bias against the authors. This is a reasonable and even necessary part of the peer review process because there are factions and there is infighting in science, and there are historical quibbles or institutional rivalries or other similar cultural phenomena that should not stand in the way of science, and need to be worked around by sensitive and thoughtful editors.

It is quite possible that this paper was submitted to a journal that wouldn’t quite know how to handle it, along with “helpful” information of the kind that in other cases might have been, well, helpful, but in this case served to derail the normally earnest and honest process of peer review. That something like this happened was certainly on my mind when I first saw this paper in this journal. Since certain parts of the process of review are kept confidential (for good reasons) we may never know this. Ultimately, though, Dr. Wagner may have felt that the gate-keeping (in a good way) function of the editorial staff was inadequate, and thus his very powerfully symbolic resignation.

It is possible, I suppose, that the research in Spencer and Braswell’s “On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance“, way down deep beneath the trickery, the bad methodology, and the scandalous politically motivated lack of scientific rigor has in iota of scientific merit. If so, this paper is on the table and available for examination, and the hypotheses embodied there could be further considered by climate scientists.

As you know, I’ve just started blogging at a second venue, called “The X Blog.” I’m going to use this opportunity to put a list of links related to Wagner’s resignation and the demise of Spencer and Braswell’s credibility over there.

Sources:

Spencer, R., & Braswell, W. (2011). On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance Remote Sensing, 3 (8), 1603-1613 DOI: 10.3390/rs3081603

Wagner, Wolfgang. (2011). Taking Responsibility on Publishing the Controversial Paper “On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance” by Spencer and Braswell. Remote Sensing 2011, 3, 2002-2004; doi:10.3390/rs3092002