The IPCC Can’t Count – Author and Reviewer Numbers are Wrong

HOW many times have you heard or read words to the effect that 4000 scientists from the Intergovernmental Panel on Climate change (IPCC) supported the claims about a significant human influence on climate? I think I’ve seen it on television, radio and the Internet and I know that politicians at national levels have quoted such figures. There’s no question whatsoever. It’s utterly wrong.

In fact, once the duplicated names are removed that number falls below 2,900 and if we only want those who explicitly supported the claims it falls to only about 60. So how does 4,000 become 60? Let’s take a closer look at the real numbers.

Consider the statement “The IPCC report concluded that there is a human influence on climate. This was supported by almost 4,000 authors and reviewers.” The popular mythology – and one that IPCC never tries to correct – is that those 4,000 supported the findings but all they really supported, by their work, was the creation of the report. The IPCC is allowing, or perhaps even encouraging, an ambiguity be misinterpreted, and not for the first time either, but I digress.

But how many individuals were involved? Was it 4,000 or some other number?

The IPCC said in its flyer for its Fourth Assessment Report (4AR)

2500+ Scientific Expert Reviewers
800+ Contributing Authors
450+ Lead Authors…[note 1]

so immediately the figure drops from 4,000 to 3,750 – but a closer check shows that’s not correct either.

Each of the three IPCC Working Groups (WGs) assembles its own lists of authors and reviewers, making six lists in all. Based on those lists of reviewers and on a previous list of authors that I had already extracted from the start of each chapter of the report (and each “Summary for Policy makers”and “Technical Summary”) the total number of authors and reviewers for each working group is as follows:

WG I – 595 authors, 624 reviewers
WG II – 400 authors, 1169 reviewers
WG III – 251 authors, 469 reviewers

Based on thee figures the total number of reviewers was 2262 and the number of authors is 1246. The former is well short of the IPCC’s 2500 but the latter is near enough to the claimed 1250.

Looking first at the number of authors we must take into account that an author for one working group can also be an author for another working group.

Anil Markandya of the UK was an author of WG II chapter 8 and WG III chapter 2, and Blair Fitzharris of New Zealand was an author of WG I chapter 4 and WG II chapter 11 and “summary for policymakers”. They, along with 39 other authors who contributed to more than one working group, have been counted twice in the IPCC’s total of 1250 authors.

Other authors who have been counted twice include those whose names are sometimes shown with accents, umlauts and other marks and sometimes shown without, and sometimes even with different spelling (e.g. the umlauted ‘u’ is sometimes shown as ‘ue’). After converting the variable spelling to a single form and removing the duplicates across working groups the number of authors falls to 1189 individuals.

In a similar fashion, reviewers can comment on chapters from more than one working group and their names might have the same variable spelling. After these issues are resolved the total number of reviewers falls to 2073.

With these corrections we’ve now lost almost 500 names from the 3750 that the IPCC claimed and we’re down to 3262.

But that’s the total of the number of authors and the number of reviewers. These roles are not independent and there was nothing to prevent authors also being reviewers, even of the chapters that they had contributed to.  IPCC policies say that each author receives the draft copies of any chapter they contributed to and elsewhere we find that virtually anyone with relevant knowledge can be a reviewer, and authors often meet that requirement even for chapters that they did not work on.

Take for example Anthony Broccoli of the USA. He was an author of WG I chapter 8 and a reviewer of WG I chapter 6. All up, 383 individuals acted both as reviewers and authors. After removing the duplicate names from the composite list the number of individuals falls to 2879, which is a long way short of the claimed 4,000 and even of the IPCC’s total of 3750, which of course could be understood as individuals rather than separate lists.

At this point a few issues need to be clarified about the accuracy of the above figures because I wouldn’t like to be like the IPCC and not be precise. The manual adjustment for accents and other marks (e.g. umlauts) might not be 100% accurate. Care has been taken with the resolving of duplicate names but the results can’t be guaranteed to be correct. I’ve taken great care with identifying duplicate names but it is possible that two individuals might share the same name. On the other hand the same individual might at different times be listed under different names (e.g. women who are listed both under their maiden-name and their married name).

I also made corrections where the IPCC seems uncertain whether a Cuban individual was Juan Llanes-Requeiro or Juan Llanes-Requerio, or if a Swedish individual was Markku Rummukainen or Markku Rummu Ainen. Perhaps I have failed to notice other instances of IPCC incompetence.

Another problem that I discovered was that an Australian author whose name was in the composite list in the appendix to the Working Group I contribution could not be found in the lists of authors at the start of each chapter. (This was another reason for using a list of authors extracted from each chapter rather than the composite list provided by each Working Group.) I haven’t bothered to determine whether other instances of this discrepancy occur for this working group or others. I don’t know and frankly don’t care, because it looks like just another of the IPCC’s distortions and outright lies.

The IPCC’s claims of 2500+ reviewers and 1250 authors bears little relationship to the truth and its implied total of more than 3750 individuals is even worse. This is just another example of how the IPCC’s messages are ambiguous – deliberately so, in my opinion – and how the organization publishes a lobbyist’s message rather than a clear presentation of the facts.

But did the 2879 individuals all support the IPCC’s principal findings? There’s really no proof that any more than about 2% explicitly did so.

The key chapter in the IPCC’s Fourth Assessment Report of 2007 is chapter 9 of the Working Group I
contribution. No authors prior to that chapter were asked for their opinion, although perhaps we could say that some authors of two chapters, 1 and 8, gave it anyway. Authors of chapters 10 and 11 of the WG I contribution assumed the claims made in chapter 9 were correct, and so did the authors of all chapters of the contributions by Working Groups II and III.

Chapter 9 of Working Group I had 53 authors in total but more than 40 were part of a network of people who worked previously together. In direct contradiction to the IPCC’s statements that the team of authors should have a wide range of views and experiences, most were climate modellers and there were many instances where several authors were associated with the same establishment. In the latter case it was not uncommon for a contributing author to be a subordinate (academic or work) to author higher in the authoring hierarchy. [note 2]

It is impossible to say whether the findings were truly supported by all 53 authors of chapter 9 but let’s assume that they did.

The second draft of chapter 9, the last draft available for review by individuals, received comments from 552 people and 7 governments. Of these 62 only 5 reviewers explicitly endorsed the overall chapter [note 3]. None of these supporting reviewers was particularly credible because of vested interests or because this was the only review comment they made for the entire report, but let’s not quibble.

Fifty-three authors and five reviewers are all that can be said to explicitly support the claim of a significant human influence on climate. The figure of 4,000 is a myth. Even the IPCC’s total of 3750 is a myth. Not even the 2879 individuals can be shown to support the claim. The true number is about 60, or maybe if we asked each of those privately, perhaps even fewer.

It can’t be denied that other reviewers may have supported the notion of significant man-made warming but failed to explicitly state this. If that’s the case them they only have themselves to blame for not being counted with the other “true believers”. We can however be sure that support among those other IPCC authors and reviewers was not unanimous. The lists of reviewers contain the names of many people who are known to be sceptical of a significant human influence on temperature, so any implication that they supported the IPCC’s findings are totally without merit. As mentioned earlier, the support to which the IPCC referred in its flyer was towards the creation of the entire report, not the findings of a small network of climate modellers, although the “sceptical reviewers” would probably argue that their support for the report was only in trying to curb the excesses of the IPCC’s authors.

This document has shown that yet another statement by the IPCC is unsupported by any solid evidence. Not only was the IPCC dishonest about the level of support for its primary claim, either deliberately or through allowing an ambiguity to be misinterpreted, but it couldn’t even get its number of authors and reviewers correct.

If the IPCC cannot provide an accurate count of authors, reviewers and the people who acted in both roles for its report then we probably should be asking what, if any, of its statements can be believed.

***********

John McLean lives in Melbourne, Australia.  This note is republished from Mr McLean’s website with permission. 

Notes

1.  http://www.ipcc.ch/pdf/press-ar4/ipcc-flyer-low.pdf (IPCC flyer for 4AR)
2. “Prejudiced Authors, Prejudiced Findings” at
http://scienceandpublicpolicy.org/images/stories/papers/originals/McLean_IPCC_bias.pdf
3.  “An Analysis of the Review of the IPCC 4AR WG I Report” at http://mclean.ch/climate/docs/IPCC_review_updated_analysis.pdf