We’ve been told many things about the Intergovernmental Panel on Climate Change (IPCC) that aren’t true. For example, IPCC reports are not based entirely on peer-reviewed literature (see here and here). Nor are they necessarily written by the world’s top experts (see here and here). Nor has the IPCC sought the input of thousands of scientists regarding the crucial question of whether or not humans are responsible for global warming. Rather, that determination was made by the authors of a single chapter of the 2007 IPCC report (out of a total of 44).

So it isn’t surprising that doubt is being cast on yet another IPCC claim. According to the catechism, one of the reasons we can be assured the IPCC is a neutral and objective body is because it isn’t in the business of producing original research. In the words of an explanatory page on its website, the IPCC “does not conduct any research” of its own. It merely assesses whatever material happens to be available.

Many IPCC insiders believe this to be the case. But others allege that the IPCC isn’t as arms-length as it claims to be. Last year 232 people answered a questionnaire distributed by an external committee investigating the IPCC. All their answers were released in a 678-page PDF here after their names were removed. The remarks below are those of IPCC insiders only – authors, review editors, and bureau members. All bolding has been added by me.

The person speaking on page two expresses the IPCC party line:

The IPCC does not (and should not) do any research…

Similar comments may be found on pages 58, 98, 206, 210 and 211. Yet, according to a lead author whose remarks appear on page 188:

…on a number of occasions the IPCC has been connected to [climate] model intercomparisons/harmonization and scenario development which border on research.

This person’s concern isn’t so much that the IPCC’s no-research rule is being violated, but that when this occurs, the research the IPCC has commissioned is given more weight than alternative findings:

Such research should not be given preferential treatment compared to other sources of research – there should be a level playing field and not favored models or groups… (p. 188)

A second lead author volunteers that the IPCC:

…has at times, in my opinion, strayed into creating literature rather than assessing literature. I am thinking particularly of past special reports on emissions scenarios that generated “IPCC scenarios” rather than assessing scenarios [already available] in the literature. (p. 69)

This theme is confirmed by someone else who observes on page 322 that while the tables and graphs appearing in IPCC reports used to be copied from already-published studies (which had presumably received quality-assurance checks via the peer-review process), these exhibits are now more likely to be prepared “specifically for the IPCC reports.”

This means that some of the material the IPCC authors want to include has not, in fact, been published previously. Again, this person’s concern isn’t that the rules are being flouted flaunted. Rather, s/he tells us that the IPCC authors find it onerous to construct these exhibits themselves.

Apparently it doesn’t stop with graphs and charts. When information that IPCC authors wish to include in their chapter does not already exist in the peer-reviewed literature, some of them aren’t above arranging for it to be inserted there. Says one person:

Governments want the chapter to cover questions of current relevance for which there [is] often “grey literature” but little peer reviewed literature…An approach that has been used in such cases is that lead authors try to have material published in peer reviewed journals while they are drafting the IPCC chapter so that the published or in press article can be cited in the final draft of the IPCC chapter. (p. 68)

This is surely a no-no. If the public is being told the IPCC surveys only the currently-available literature, it’s surely cheating for IPCC authors to deliberately plant select information in journals.

This does, however, shed light on a curious discovery I made last year. The 2007 IPCC report references no less than 16 articles from a single issue of the journal Climatic Change. All told, there are 39 citations spread across four IPCC chapters.

The difficulty is that this issue wasn’t published until May 2007. Which means it didn’t exist during the time period in which the IPCC report was being written and reviewed. (The last cut-off date for IPCC expert reviewer comments for the working groups involved was July 21, 2006. But 15 of these 16 papers weren’t even accepted for publication by the journal until two months later.)

As I wrote last May, this means there are at least 39 citations in the 2007 IPCC report that:

…don’t reference research papers the wider scientific community had already digested. They don’t even reference papers that were hot off the press. Instead, in 15 of 16 cases, no expert reviewer could possibly have evaluated these papers since they hadn’t yet been accepted for publication by the journal itself.

It is difficult to avoid the conclusion that some IPCC authors have been playing fast-and-loose with the rules – and that they may have been assisted in this regard by supposedly neutral academic journals. When one considers that many IPCC authors also fill senior positions at academic journals, a theoretical loophole starts to look like a potentially serious problem. Yet to my knowledge, the IPCC has never acknowledged that this sort of bad behaviour might be a concern.

The answers insiders provided to the questionnaire also highlight the fact that 22 years of ongoing IPCC reports (including 4 large assessments, with a fifth underway) have – inadvertently or not – begun to exert an influence on the kind of climate research that is judged to be necessary, relevant, and worthy of funding by governments and research institutes.

For example, one coordinating lead author observes:

I am greatly concerned that the current model is unsustainable. It produces serious burnout in the research community and consumes valuable resources for tasks such as running the SRES scenarios with high-resolution global climate models (GCMs). It is doubtful whether this rather routine task of running the scenarios, which is undertaken only because IPCC asks for it, is an optimal use of skilled GCM scientists and massive supercomputer power. In short, the current IPCC model places a severe burden on the research community. (p. 87)

Here are a few more voices:

A relatively incoherent narrative (in my opinion) has been established by the IPCC, and scientific research in the field proceeds by embellishing this established narrative. And the end result is that we are not asking the right questions in the field of climate research, but the IPCC continues with assessing the research that has been done in response to the narrative that it has established. (p. 97)

The “IPCC community”, meaning the authors involved in IPCC reports as well as the IPCC itself, has, in effect, evolved from a panel assessing the literature on climate change to an active commissioner of research on behalf of governments. This is particularly evident as successive IPCC reports move ever closer to providing actual climate science services. (p. 325)

Much as I value what the IPCC has accomplished to date, I do question the necessity of continued end to end assessments. How much is it really meeting the needs of the policy community…and how much of it is really sustaining the interests and professional stature of the various research communities that have become intertwined with the IPCC? I wonder. (p. 366)

…the IPCC schedule is having a detrimental impact on the global base research agenda with the research activities being steered by the IPCC timing. (p. 519)

In other words, the IPCC – like any other large, influential organization – does not exist in a vacuum. It must be understood in its larger context.

And if some IPCC insiders are volunteering, in the course of airing other concerns, that people aren’t following the IPCC’s own rules, the integrity of the end product is called into question.

.