The latest climate contrarian meme appears to be (baseless) accusations of scientific “gatekeeping” and “censorship”. Ross McKitrick provides an example of this unmistakable trend, with a blow-by-blow account of difficulties encountered in publication of his upcoming paper on the supposed contamination of the surface temperature record. The new paper purports to debunk a single statement in the 2007 IPCC Fourth Assessment Report, one denigrating the conclusions of a previous paper by McKitrick and Patrick Michaels.
McKitrick criticizes the IPCC assertion that “locations of greatest socioeconomic development are also those that have been most warmed by atmospheric circulation“. He claims that other sections cited to support that statement do no such thing. But it turns out that McKitrick himself has it completely wrong, as he cites a passage concerning regional warming over the 21st century, instead of the actual relevant passage concerning the period 1975-2005.
Moreover a review of the relevant scientific literature reveals substantial flaws in the previous analyses of McKitrick and Michaels. That, rather than any close-mindedness or “censorship”, is the real reason why McKitrick’s analyses have become increasingly marginalized in the scientific literature, if not in the right-wing press.
McKitrick’s account of the entire saga, entitled Circling the Bandwagons:
My Adventures Correcting the IPCC goes back to 2004 when he and Patrick Michaels published their first paper on “contamination” of the surface temperature record by (a general discussion, with pointers to various papers can be found at RealClimate). That paper, entitled A Test of Corrections for Extraneous Signals in Gridded Surface Temperature Data, ran in Climate Research, with one Chris de Freitas as responsible editor.
It is perhaps unsurprising, then, that McKitrick puts forth an astonishingly spirited defence of de Freitas. De Freitas is best known as the editor whose greenlighting of an execrable paper by Willie Soon and Sallie Baliunas led to the mass resignation of four Climate Research editors later the same year, including just-installed editor-in-chief Hans von Storch. Not to mention the ties to industry-funded PR and “astro-turf” groups like Friends of Science and to the International Climate Science Coalition, as I’ve previously discussed. Certainly, McKitrick’s description of the Climate Research debacle is decidedly slanted and worth exploring in a separate post.
For now, though, let’s pass to the main event, McKitrick’s critique of a passage in the IPCC Third Assessment Report discussing his 2004 paper. Echoing and amplifying an earlier diatribe published in the National Post, McKitrick writes at p.8-9:
… [I]t was not until late 2007 that I became aware that the following paragraph had been inserted on page 244 of the Working Group I report.
McKitrick and Michaels (2004) and De Laat and Maurellis (2006) attempted to demonstrate that geographical patterns of warming trends over land are strongly correlated with geographical patterns of industrial and socioeconomic development, implying that urbanisation and related land surface changes have caused much of the observed warming. However, the locations of greatest socioeconomic development are also those that have been most warmed by atmospheric circulation changes (Sections 22.214.171.124 and 3.6.4), which exhibit large-scale coherence. Hence, the correlation of warming with industrial and socioeconomic development ceases to be statistically significant. In addition, observed warming has been, and transient greenhouse-induced warming is expected to be, greater over land than over the oceans (Chapter 10), owing to the smaller thermal capacity of the land.
(I added the underlining.)
The first point to dispense with is the reference to Sections 126.96.36.199 and 3.6.4 in support of the claim that “the locations of greatest socioeconomic development are also those that have been most warmed by atmospheric circulation changes.” There is nothing whatsoever in either section that supports the point.
In neither section is there any discussion of industrialization, socioeconomic development, urbanization or any related term.
Let’s take a closer look at McKitrick’s dissection of Section 188.8.131.52, a discussion of the spatial distribution 20th century warming trends.
Section 184.108.40.206 presents a spatial map of warming trends since 1979. In the accompanying text they state that “Warming is strongest over the continental interiors of Asia and northwestern North America and over some mid-latitude ocean regions of the [Southern Hemisphere] as well as southeastern Brazil.” These are the regions of greatest socioeconomic development? The continental interior of Asia suffered economic decline after 1990, and northwestern North America is sparsely-populated alpine forest, so the claim is rather unlikely to be true. Certainly Section 220.127.116.11 does not try to argue the point.
Now look at Figure 3.9, the “spatial map” of trends presented by the IPCC.
Figure 3.9. Linear trend of annual temperatures for 1901 to 2005 (left; °C per century) and 1979 to 2005 (right; °C per decade). Areas in grey have insufficient data to produce reliable trends. The minimum number of years needed to calculate a trend value is 66 years for 1901 to 2005 and 18 years for 1979 to 2005…
So spatial trends for two periods charted and discussed. Unfortunately for McKitrick, the passage he disparages is explicitly based on the longer period from 1901, not the relevant period from 1979. The areas mentioned can clearly be seen on the lefthand map (e.g. southeastern Brazil), even without expanding to the full view. The reference is obvious in the text of the passage as well, which starts:
For the century-long period, warming is statistically significant over most of the world’s surface …
And let’s pick up the passage again at the end of McKitrick’s quote:
… as well as southeastern Brazil. In the recent period [i.e. 1979-2005], some regions have warmed substantially while a few have cooled slightly on an annual basis (Figure 3.9). … Warming in this period was strongest over western North America, northern Europe and China in winter, Europe and northern and eastern Asia in spring, Europe and North Africa in summer and northern North America, Greenland and eastern Asia in autumn (Figure 3.10).
So the list of strongly warming regions in 1979-2005 includes North America and Europe, and none at all in the southern hemisphere. Clearly the IPCC statement is not nearly as outlandish as McKitrick’s gross misinterpretation would have us believe.
This doesn’t mean that the rest of McKitrick’s critique is thereby disproven. But McKitrick does display here a shocking ignorance of climatology, and if his latest article shares the same sort of problems, then his explanation for difficulty in publication could be somewhat incomplete or selective. For example, any knowledgable reviewer would have taken exception to McKitrick’s assumptions if he made claims along these lines:
I obtained values of the effects of trends in the Arctic Oscillation, North Atlantic Oscillation, Pacific Decadal Oscillation and the El-Niño Southern Oscillation on the gridded surface trend pattern over the 1979-2002 interval. I re-did the regression results from my 2004 and 2007 papers after adding these variables into the model. I showed that the original results did not become statistically insignificant.
Again, this appears to betray a fundamental misunderstanding. The IPCC reference to locations “most warmed by atmospheric circulation changes” is not a description of the “effects of trends” in the AO or other oscillations, which are by definition pretty much trendless. Indeed, it may well be that McKitrick’s travails merely demonstrate that the peer-review system is working as it ought to.
Nevertheless, it is worth reviewing contemporary reaction to McKitrick and Michaels (2004). At the time, Rasmus Benestad at Real Climate pointed out:
[T]here are a number of issues that they did not address that logically must must be addressed for their conclusions to be tenable. MM04 failed to acknowledge other independent data supporting the instrumental thermometer-based land surface temperature observations, such as satellite-derived temperature trend estimates over land areas in the Northern Hemisphere (Intergovernmental Intergovernmental Panel on Climate Change, Third Assessment Report, Chapter 2, Box 2.1, p. 106) that cannot conceivably be subject to the non-climatic sources of bias considered by them. Furthermore, they fail to reconcile their hypothesis with the established large-scale warming evident from global sea surface temperature data that, again, cannot be influenced by the local, non-climatic factors they argue contaminate evidence for surface warming. By focusing on thermometer-based land observations only, and ignoring other evidence conflicting with their hypothesis, MM04 failed to address basic flaws in their arguments.
Benestad also noted a post from climate blogger Tim Lambert, who pointed out an error in the calculation of area weighting (a mixup of radians and degrees) that almost doubled the portion of warming trend attributed by McKitrick to economic factors, despite McKitrick’s claim that the correction “hardly changed the results” (hat tip to Frank O’Dwyer).
But more relevant for the current discussion is more direct evidence of spurious correlation. In his published comment on the McKitrick and Michaels paper, Benestad demonstrated that its statistical significance was “vastly overstated”.
However, when a validation was performed on a similar analysis for
which the regression model was calibrated with a subset of the data, and the remaining data were used for validation, it became apparent that models based on the factors that McKitrick & Michaels used had no skill (i.e. were not able to reproduce the independent data). The negligence to account for inter-station dependencies in the analysis resulted in spurious results and inflated confidence levels in the analysis of McKitrick & Michaels.
McKitrick and Michaels replied to Benestad’s comment, but Benestad made a strong case that the reply misses the mark:
… McKitrick and Michaels (2004b, or “MM04b”) argue that such validation experiments (i.e, splitting up the data to test the validity of statistical modelling) is not common in the refereed climatological literature. That argument is puzzling indeed, as such tests are standard in statistical modeling exercises, and have been used and documented in many peer-reviewed articles in the meteorological and climatalogical literature (see this list of publications by just one researcher alone or even the introductory textbook by Wilks, 1995).
… In their reply to Benestad(2004), McKitrick and Michaels (2004b) claim that I do not dispute their approach (i.e., multivariate regression using economic variables as potential predictors of surface temperature). That claim is both peculiar, and misses the point. A method is only valid when applied correctly. As described, above, MM04 failed egregiously in this regard. The purpose of my paper was simply to demonstrate that, whether or not one accepts the merits of their approach, a correct, and more careful, repetition of their analysis alone is sufficient to falsify their results and their conclusions.
A later analysis from McKitrick and Michaels (JGR, 2007) presumably set out to overcome all the above objections. Schmidt 2009 addressed that study (and as well as a 2006 study from de Laat and Maurellis).
[T]he basis of the results are correlations over a very restricted set of locations (predominantly western Europe, Japan and the USA) which project strongly onto naturally occurring patterns of climate variability, or are with fields with significant amounts of spatial auto-correlation. Across model simulations, the correlations vary widely due to the chaotic weather component in any short-term record. The reported correlations do not fall outside the simulated distribution, and are probably spurious (i.e. are likely to have arisen from chance alone). Thus, though this study cannot prove that the global temperature record is unbiased, there is no compelling evidence from these correlations of any large-scale contamination.
Of course, as these last two papers were published later, they are not directly relevant to IPCC AR4. But together they do provide a good view of the current state of the surface temperature record “contamination” debate in climatological literature.
McKitrick and Nicolas Nierenberg submitted a standalone rebuttal paper to Schmidt 2009 to the International Journal of Climatology. Spatial Autocorrelation and the Detection of Non-Climatic Signals in Climate Data was rejected; at least part of the problem was that the paper appeared to add little new to the discussion, as would be expected from a standalone paper. McKitrick and Nierenberg wrote a complaining letter to IJOC that obfuscates that point:
The first point alone justifies publication of a comment on Schmidt since the confusion about where the SAC correction needs to be applied goes to the heart of Schmidt’s argument… [Emphasis added]
Benestad’s 2004 article comment (and Schmidt’s 2009 article) provide clear evidence that McKitrick’s analyses yield spurious correlations; Schmidt’s latest comment summarizes the state of play:
Overall, Mckitrick’s story tells has a couple of very important messages – obviously anyone with persistence can get anything published eventually. But the reception that his results have received (very little, and what there is has been dismissive) is a reflection, not of the climate science community’s disdain for new ideas, but rather that his analysis is just not that interesting or convincing. Assuming that Barrow Alaska has the same socio-economic status as Miami, or that Moscow is the same as Siberia, makes no sense. Assuming that national educational attainment is a proxy for National Met Service competence is simply that, an assumption. Confusing real effects in surface temperature changes (local pollution, land use change) with ‘contamination’ adds no insight at all. Claiming support from de Laat and Maurellis that simply doesn’t exist is just wrong.
As a Canadian, I would have used the example of Resolute and Toronto, but otherwise it’s hard to quibble with that. And Schmidt’s conclusion is perhaps more prescient than he realized.
I am not at all surprised that statisticians like this kind of stuff, but the errors are not in the statistics (for the most part), but in the underlying assumptions – and statisticians are not necessarily going to see that.
As if to prove Schmidt’s point, here’s McKitrick’s account of how he finally achieved publication of his IPCC critique.
So I wrote to the editor of JASA [The Journal of the American Statistical Association], described what had happened at other journals, and asked if the paper might be reconsidered … [H]e pointed to a new journal that he and some colleagues had recently founded, called Statistics, Politics and Policy, which is dedicated to bringing rigorous statistical analysis to bear on important issues with policy implications. He said the paper would be a good fit, and encouraged me to submit it there. I did, and in due course my paper was accepted. It will appear in the inaugural issue this summer.
The editor of JASA from 2001 to 2009 was David Banks of Duke University, and he is now senior editor at Statistics, Politics and Policy.
But Banks also happens to have been one of the reviewers of the Wegman report, which stands as an eternal monument to the dangers of combining knowledge of statistical methods with abject subject matter ignorance (not to mention highly dubious scholarship). We may have to wait until the summer to read McKitrick’s latest, but in the mean time perhaps Banks’ review of Wegman could be made available, so we can assess this editor’s ability to understand the relevant climatological issues.
More and more, this looks like a concerted attempt to poison the well for the upcoming IPCC Fifth Assessment Report. Contrarian publications in the scientific literature remain rare, and now tend to be rebutted to devastating effect even if they manage to get through. So we can expect more whining about censorship and gatekeeping from the likes of McKitrick and others.
That’s all they’ve got left, apparently.