But one of the graphs differed markedly from the rest, leading to a controversy in the run-up to a conference of paleo-climatologists in Tanzania in September 1999. The abnormal temperature graph was "a problem and a potential distraction/detraction from the reasonably consensus viewpoint we'd like to show," paleo-climatologist Michael Mann wrote in an e-mail, adding that he didn't want to be the one to offer "the skeptics … a field day." The lead author of the IPCC chapter, Chris Folland, wrote in another e-mail that the divergent data set "dilutes the message rather significantly."
Keith Briffa, whose team reconstructed the contradictory temperature graph, was furious, and wrote: "I know there is pressure to present a nice tidy story as regards 'apparent unprecedented warming in a thousand years or more in the proxy data.'"
For the IPCC report that was written at the time, the scientists eventually resorted to an underhanded solution to downplay the data behind Briffa's graph, which showed temperatures falling since the 1960s: the graph was simply cut off at 1960 in the IPCC report. "This sort of approach is considered problematic in science," says climate scientist Storch.
Briffa's unusually declining temperature graph points to a serious conundrum that no one has been able to explain yet: Since the 1960s, the tree-ring data no longer reflect actual temperature changes. But why, then, should tree-ring data be valid for periods before that?
At least the fourth IPCC report, published in 2007, discusses the problems with the tree-ring data at length. But even the current, valid report contains controversial passages.
Chapter 1.3.8, for example, contains a discussion of the possible relationship between climate change and the increased incidence of natural disasters, which, after Hurricane Katrina in the United States, have now become a politically charged issue.
At the IPCC report, the damage associated with such events "are very likely to increase due to increased frequencies and intensities of some extreme weather events" (italics in original). The report cites as evidence a study that supposedly demonstrates precisely this trend.
The only problem is that the study in question had not been subjected to outside peer review before the IPCC report went to press. This has since been done, and the conclusions are surprising: "We find insufficient evidence to claim a statistical relationship between global temperature increase and normalized catastrophe losses," read the report published in the compendium "Climate Extremes and Society."
Roger Pielke, a leading expert in this field, wrote in his blog: "The claims were not just wrong. The claims were based on knowledge that just doesn't exist."
Representatives of the insurance industry hold a completely different view, which presents an additional problem for the IPCC. Reinsurers, such as Munich Re, calculate their premiums on the basis of risk, so that an increase in the frequency and severity of natural disasters can translate into additional profits when new policies are concluded.