Did the IPCC 2007 report chapter on Latin America "shame" itself by referencing a report written by "green activists"? The British conservative paper The Times alleged as much in a story over the weekend, calling "bogus" the following statement from the IPCC report:
Up to 40% of the Amazonian forests could react drastically to even a slight reduction in precipitation; this means that the tropical vegetation, hydrology and climate system in South America could change very rapidly to another steady state, not necessarily producing gradual changes between the current and the future situation (Rowell and Moore, 2000).
The newspaper's evidence was that the reference for the statement, "Rowell and Moore, 2000," is a non-peer-reviewed report on forest fires by the World Wildlife Fund, which cited as evidence for the claim a 1999 Nature paper. Yet, The Times says, the Nature paper lacks the 40% figure.
Now, Dan Nepstad, a forest expert, and colleagues at the Woods Hole Research Center who wrote the 1999 Nature paper say the IPCC statement is correct. He says it's a case of correct facts but footnotes that refer to the wrong papers or are at least incomplete. (Here's a statement in which Nepstad lays out his position.)
As early as 1994, Nepstad and colleagues found that as much as half the Amazonian forest periodically fell to low enough moisture levels that trees died. In a 1998 experiment they learned that the threshold was roughly a 30% level of moisture in the soil. They confirmed in a 2004 paper—3 years before the IPCC report was published—that as much as half of the forest had soil with the 30% level.
So although IPCC failed to cite studies that would back up the "40%" claim in the WWF report or the IPCC chapter, scientists had found that as much as half of the forest was as sensitive as they describe.
In an e-mail Nepstad sent to Times reporter Jonathan Leake 2 days before the story ran, he detailed those finding, writing:
At the time of the IPCC [report], there was ample evidence that a large portion of the Amazon forest is very close to the lower limit of rainfall that is necessary to sustain dense forest. We published an article in 1994 in Nature in which we estimated that approximately half of the forests of the Brazilian Amazon were periodically exposed to severe drought and soil moisture depletion, especially during El Nino events.
Today in Washington, D.C., speaking at an event hosted by the Center for American Progress, Christopher Field of the Carnegie Institution for Science dismissed the concern with the 40% figure as "inconsequential." Field will lead the portion of the 2014 IPCC report, so-called Working Group II, that will examine impacts such as forest drought. But he said that IPCC must "apply extra assessment" to reports in the "gray" literature, referring to non-peer-reviewed studies.
[Update 2/4 9am: The BBC ran a story I hadn't seen in which a scientist reviewed other evidence of a very vulnerable rainforest on January 30]