Doomed Planet

Glaciergate meltdown

Flaws in the IPCC report writing process
The IPCC’s assessment reports are widely regarded as the ultimate references on climate matters. Behind this esteem is a belief that teams of scientists impartially evaluated a vast pool of information and together drafted and refined an impeccable document for each report. The gulf between this belief and reality is in fact huge.

The problems were highlighted by “Glaciergate”, the IPCC’s flawed comments about the Himalayan glaciers.

This section of the IPCC report opened with comments about the area of Himalayan glaciers covering 3 million hectares, which is 30,000 sq. km., and said shortly afterwards that the likelihood of them disappearing by 2035 is very high. In the next paragraph the report says that the glaciers are predicted to shrink from 500,000 sq. km. to 100,000 sq. km. by year 2035, which contradicts the earlier statements both in extent and state of retreat.

It appears that most of the IPCC’s text was a verbatim copy of passages from an article in the Indian magazine Down to Earth, not the source given in the IPCC report, with quotes in that article becoming factual statements in the IPCC report. The retreat by the year 2035 was mentioned in the article but trace the source of that information we find it was an incorrect transcription of year 2350 and the shrinkage to 100,000 sq. km. was a comment about the total of all glaciers outside the polar regions.

How could these errors and inconsistencies appear in the IPCC report, supposedly the authority on climate matters?

The answer lies in the IPCC’s procedures and in the tasks assigned to authors, reviewers and review editors of the reports.

No-one is assigned the work of policing IPCC procedures and the only people to check authors’ work are the expert reviewers, whose role is to “comment on the accuracy and completeness of the … content and … balance of the drafts … according to their own knowledge and experience.” IPCC procedures fail to mention what happens if no experts are available in a particular subject area, so presumably any review is ad hoc and by people less than expert in the subject.

Who would be a reviewer though when IPCC authors are only obliged to respond to each review comment, which is done via the review system, and are under no obligation to modify their text in response to reviewers’ comments. Reviewers complain that they must justify their comments but not the authors, besides which the review process seems only good for flagging errors in spelling, punctuation and grammar, and nominating additional information to support authors’ claims.

In the case of the Himalayan glaciers no reviewer mentioned the discrepancy in glacier size and just one IPCC reviewer commented on the different estimates of 2035 conditions. The cryptic response from the author was “Missed to clarify this one” but apparently no-one in the IPCC system policed this or followed it up.

In the 2007 report chapter 2 and 3 shows examples of authors citing their own papers to defend their arguments when reviewers objected that more published papers reached very different conclusions. In contrast to this, elsewhere some authors rejected reviewers’ comments on the grounds that more papers supported the authors’ views than the reviewers’. It seems that authors’ views dominate even when authors contradict each other.

If the IPCC reports are to have any integrity the sources used by the report authors should also be checked. Chapter 8 of the 207 report cited three scientific papers when discussing two issues related to the El Nino-Southern Oscillation but mismatched one paper to the issue it raised. In the 2001 report IPCC author James Hansen cited a graph that he had developed with two colleagues but the paper in which the graph was first published was not properly peer-reviewed because the authors refused to disclose all of their data, and yet it was featured seven times in the IPCC report.

The final step in the creation of the reports is approval of the text by a session of the Working group. This is essentially a rubber-stamping because it is simply inconceivable that at this eleventh hour the document will be thoroughly checked and corrections be demanded of authors.

IPCC authors are a law unto themselves because they can cherry-pick sources, cite material of dubious quality and even provide false references.  They also know that reviewers’ comments can be ignored and that no criticism can be expected from the IPCC Working Group.

The IPCC claims that there was a consensus but what kind of consensus was it when it is clear that the draft text was either never examined in detail or it was properly examined but no corrections were ever enforced?

It beggars belief to think that governments around the world intend spending billions, or even trillions, of dollars on the word of such a cowboy organization.

John McLean is a member of the New Zealand Climate Science Coalition

See also: John McLean’s (pdf) "’Glaciergate’ highlights IPCC flaws

Post a comment