July 28, 2006
Misled Again: The Hockey Stick Climate
Steve McIntyre and Ross McKitrick, Financial Post, July 12, 2006
Many people have heard the claim that the 1990s were the warmest decade of the millennium and that 1998 was the warmest year. Environment Canada headlined them on pamphlets mailed across the country a few years ago. These claims interested us in verifying exactly how scientists were able to assert so confidently that the late 20th century was warmer than when the Vikings were farming Greenland (the Medieval Warm Period). Last year, the National Post profiled our published research, which had identified major flaws in what was called the Hockey Stick -- a graph prominently featured in a report from the Intergovernmental Panel on Climate Change (IPCC) in 2001.
We knew that calling this icon into question would be controversial, but we did not expect it would spark a battle between two powerful committees of the U.S. House of Representatives and lead to the formation of a blue-chip panel of the U.S. National Academy of Sciences (NAS).
Shortly after last year's Post profile, The Wall Street Journal did a long article interviewing both us and the principal Hockey Stick author, Michael Mann. In that article, Mann was quoted as saying he would not be "intimidated" into disclosing the algorithm by which he obtained his results.
This attracted the interest of the U.S. House energy and commerce committee. Its members read our articles and became concerned about our allegations that Mann had withheld adverse statistical results and that his results depended on bristlecone pine ring widths, well known to be a questionable measure for temperature. In June, 2005, they sent questions to Mann and his co-authors about verification statistics and bristlecone pines, asked Mann for the algorithm he used and asked pro forma questions about federal funds used for their research.
This prompted a storm of controversy, ironically centring on allegations of "intimidation." Various learned societies, none of which had been offended by Mann's public refusal to provide full disclosure, were outraged that a House committee, representing the taxpayers who had paid for the results, should be trying to find out how he got them.
Then a turf war broke out. The House's science committee felt that its jurisdiction had been impinged upon and asked the energy and commerce committee to butt out. After a few months of fencing in late 2005, the science committee asked the NAS to evaluate criticisms of Mann's work, and to assess the larger issue of historical climate data reconstructions.
The NAS agreed to the science committee's request, but only under terms that precluded a direct investigation of the issues that prompted the original dispute -- whether Mann et al. had withheld adverse results and whether the data and methodological information necessary for replication were available.
In February, 2006, the NAS appointed a panel of 12 eminent academics involved in climate science but not directly involved in the temperature reconstructions of the past 1,000 years. They were not an entirely "independent" panel, as some were occasional co-authors with the Hockey Stick authors. But even this limited independence was a major departure from procedures of the IPCC, which permits authors actively involved in scientific controversy to summarize the research -- even if they end up acting as reviewers of their own work!
In March, 2006, the NAS panel held meetings in Washington at which we made a presentation (along with Mann and seven other scientists in the field).
On July 6, the panel issued a 155-page report, which managed the delicate feat of accepting virtually all the criticisms of the Hockey Stick while still saying polite things about it. A European climate scientist, who understood the balancing act, wrote us afterwards to point out it was the most severe criticism of the Hockey Stick nowadays possible.
At the NAS panel, we said that Mann's principal components were biased toward producing hockey stick-shaped series; the NAS agreed. We said that bristlecones were not a reliable temperature proxy; the NAS agreed and said they should be "avoided." We said that Mann's reconstruction failed important verification tests; the NAS agreed. We said that more than one test statistic should be reported when assessing statistical validity; the NAS agreed. We said that current methods underestimated the inherent uncertainty; the NAS agreed. On and on. On no occasion was any claim of ours refuted.
Our original articles argued that Mann's data and methodology did not permit him to claim with confidence that 1998 was the "warmest year" of the millennium or that the 1990s were the warmest decade. The NAS panel even agreed with this. After observing that little confidence could be placed in reconstructions before 1600, they stated: Even less confidence can be placed in the original conclusions by Mann et al. (1999) that "the 1990s are likely the warmest decade, and 1998 the warmest year, in at least a millennium ..."
Based on some other studies, they conceded that Mann's reconstruction was still "plausible" but, contrary to the IPCC, they said it was impossible to put confidence intervals on this opinion.
The House science committee had asked the NAS panel to report on whether paleoclimate authors were withholding data and methods. The panel chairman said this topic was "too big" for them to answer. The NAS apparently plans a new panel on the generic subject of availability of scientific data.
The NAS panel drew attention to other recent studies claiming that the 20th century was warmer than the Medieval Warm Period. We've attempted to replicate these other studies as well, only to run into one obstacle after another in identifying data and methods -- similar to the problems that led to the original congressional questions about the Mann study. In one case, the authors even refused to identify the sites from which data was collected for their study!
Despite these pointless obstacles, we know enough about the "other studies" to be confident that none of them meets the methodological standards now recommended by the panel. In fact, somewhat remarkably, two of the most recent studies even continue to use Mann's discredited principal components series.
In its press release, the NAS headlined that the present era is the warmest in 400 years. However, long before anyone had ever heard of the IPCC or the Hockey Stick, this was the prevailing view of scientists, who coined the term Little Ice Age to refer to the period leading up to modern warming. It isn't news to say the average temperature is higher now compared to the past 400 years. It was news in 2001 when the IPCC claimed with confidence that the 1990s were the warmest in 1,000 years. The real news from the NAS is that it disagreed and withdrew any claim to confidence prior to 1600.
At the NAS press conference, the panel was asked about "overselling" of the warmest-in-a-millennium claim and whether any lessons could be learned. Panel chairman Gerry North noted that the Mann paper was very recent when this claim was made and observed that it was "very dangerous to pull one paper out of the literature fresh before it's had time to season." However, the panel did not comment on IPCC procedures that invited this problem.
The IPCC lead author who selected Mann's reconstruction for prominent display in the review of millennial temperature history was none other than Mann himself. At the time, he was a fresh and ambitious PhD, an odd choice to write the "consensus" review of climate history.
The system that allows such conflicts of interest has been severely criticized by some senior climate scientists, including Hans von Storch of Germany. However, the flawed process remains unchanged for the next IPCC assessment report, due in January, 2007. As reviewers of that report, we have expressed concerns to the IPCC about prominent use of graphics and empirical results from the lead authors' own freshly published papers, which have not been in print long enough to have undergone adequate, independent review and assessment and, in some cases, not even long enough to meet IPCC publication deadlines.
In our opinion, most of the press coverage to date missed one of the biggest stories.
When asked at the press conference about lessons that could be learned, panelist Kurt Cuffey said the prominent use of the Hockey Stick graphic by the IPCC sent "a very misleading message." He said the over-selling did not come from the "science community," but from the "interaction of part of the science community with the broader public discourse and in particular with the way the [Mann et al.] reconstruction was used by the IPCC in the 2001 report."
But haven't we been told that the IPCC is the "science community?" If a knowledgeable observer such as Cuffey distinguishes the two, blaming the IPCC while defending the "science community," shouldn't we be trying to figure out exactly how the IPCC process ended up sending out a "very misleading message?" And if the process has not been fixed -- and there is no evidence that it has -- how do we know that the IPCC won't send another equally "misleading" message in the upcoming Fourth Assessment report?
Steve McIntyre is a retired mineral exploration businessman who operates www.climateaudit.org.
Ross McKitrick is an associate professor of economics at the University of Guelph.
The Frontier Centre for Public Policy
is an independent public policy think tank whose mission is "to broaden the debate on our future through public policy research and education and to explore positive changes within our public institutions that support economic growth and opportunity."