April 2, 2004

Bending the Hockey Stick

Should Michael Mann’s infamous 1,000-year temperature reconstruction look less like a hockey stick and more like a city skyline? New research on natural climate variability suggests that may in fact be the case.

New research casts doubt on paleoclimatologist Michael Mann’s influential “Hockey Stick,” a now-famous 1,000-year temperature reconstruction.

Mann’s reconstruction is easy to picture: The handle of the Hockey Stick is 900 years of climate constancy. The blade is a sharp spike representing the last 100 years, a temperature rise that many people claim is attributable to human activity.

Mann’s revolutionary temperature history culminates in 1998’s record high, attained when he splices the observed temperature history onto the end of the reconstructed history (somewhat of a dubious practice). Virtually absent from Mann’s record are the Medieval Warm Period and the Little Ice Age, two widely accepted climate phenomena that demonstrate climate’s natural variability.

An alternative reconstruction of the last millennium’s climate finds evidence of a series of warm and cold eras, with a period of high temperatures a thousand years ago rivaling the warmth of today, and a gradual descent into and recovery from lower temperatures in the intervening centuries.

Mann, a University of Virginia paleoclimatologist, created a tidal wave of debate in the scientific literature, in Congressional committees, at scientific and professional conferences, in the pages of the press, at select classrooms, lunchrooms, and boardrooms, and just about everywhere else that climate science is discussed.

The reason for the always heated (and oftentimes personal) debate is that the temperature behavior of the past century looks quite odd when compared with the nine preceding it. That is significant for two reasons. 1) The recent spike is being used to bolster claims that anthropogenic activities have caused the climate to act in ways that are “unnatural,” and 2) It discredits nearly all of the voluminous research on the climate of the past millennium—research that has clearly identified an era of high temperatures about 1,000 years ago called the Medieval Warm Period and a subsequent cool period called The Little Ice Age. Again, both are virtually absent from Mann’s record.

Yet the Mann reconstruction is held as gospel by the United Nations, the Clinton Administration scientists who produced the infamous “National Assessment” of global warming, and environmentalists. Any challenges to it are subject to swift retaliation. For instance, the publication of work by Harvard scientists Willie Soon and Sallie Baliunas in the peer-reviewed journal Climate Research—work that concluded the climate of the 20th century was not unique when viewed in the context of the past 1,000 years—was met with shocking protest. Several of the journal’s editors actually resigned, and rumors circulated of an organized boycott of Climate Research. Thankfully, the larger field of climate scientists saw that temper tantrum for what it was—ill-placed and unscientific grandstanding—and Climate Research continues to enjoy popularity among climate researchers, including editors, authors, and audience. Nevertheless, the viciousness of the attack and the petulance with which it was executed shocked the normally staid scientific community.

Despite the threat of more shenanigans by the Mann partisans, research marches on. Canadian researchers Steve McIntyre and Ross McKitrick have published a paper citing data quality issues and methodological errors in some of Mann’s analyses. They, too, preliminarily conclude that the temperatures of the 20th century are not unique. McIntyre and McKitrick continue to explore these issues and promise more results in the near future.

Their research has been criticized, as was that of Soon and Baliunas, as being performed by people who were not paleoclimatologists by training or experience, and were thus unqualified to undertake such examinations and make such conclusions. Those pejoratives were clear in the testimony of Mann himself before the Senate Environment and Public Works Committee last summer. In fact, they were so harsh that the chairman, Sen. Inhofe, publicly took exception to Mann’s language. Apparently, in Mann’s view, surveying the scientific literature (as Soon and Baliunas did) or performing mathematical analysis (as McIntyre and McKitrick did) can only be performed by a select few.

If Mann’s partisans believe that climate science published by non-climate scientists is suspect, then they had better pay attention to a paper appearing in the March 23, 2004, issue of the Eos—Transactions of the American Geophysical Union, by well-published paleoclimatologists Jan Esper, David Frank, and Robert Wilson. Esper was the lead author of a paper that appeared in Science magazine back in 2002 that gave his version of the 1,000-year temperature history on the Northern Hemisphere, and it was quite a bit different than Mann’s Hockey Stick: It featured both a pronounced Medieval Warm Period and a Little Ice Age. In fact, that reconstruction shows a temperature change that was about twice that in the Mann record. Cast in the light of Esper’s reconstruction, with large inherent natural variations, the temperature swing of the last century or so doesn’t look out of the ordinary. Figure 1 shows a comparison of the Esper and the Mann 1,000-year temperature history of the Northern Hemisphere. Notice the long-term temperature variations present in the Esper reconstruction (blue line) are absent from Mann’s (red line).

Temperature Reconstructions

Figure 1. A comparison of 1,000-year temperature reconstructions. The red line is the temperature history of the Northern Hemisphere as developed by Mann and colleagues, a.k.a. “the Hockey Stick.” The blue line represents the Northern Hemispheric temperature history as constructed by Esper’s research team (source: Esper et al., 2002).

Many reasons have been put forth to try to account for the difference between these two reconstructions, by both Esper and Mann. They are fully described in Esper’s Eos paper, and include both data issues and analysis issues. Esper goes through each one and performs tests to assess their influence on the reconstructions. He basically eliminates all the possibilities except the technique used to process tree-ring data sets—the primary information relied on in the early portions of the temperature reconstructions.

The problem with tree rings is that their variations reflect more than just year-to-year climate differences (temperature and/or precipitation). As the trees age, the tree-ring production changes and introduces a spurious trend in the tree-ring series. That aging effect differs among tree species, as well as within species depending on the growing conditions (soil type, elevation, slope aspect, etc.). Therefore, it becomes very difficult to separate trends due to aging effects from trends due to climate.

Various research groups have developed different techniques to attempt to account for this problem, but since the ground truth (the true temperature) is not known, no one can be sure whose technique is the best.

Esper uses a method aimed at retaining long-period (greater than a century or so) variations in the tree-ring records, whereas Mann uses a method that virtually eliminates all long-term variation.

Though Esper goes on the show that the differences in the two reconstructions are very well explained by the difference in the way the tree-ring data are handled, he stops short of proclaiming which methodology is preferable (although that seems obvious by his choice). He instead prefers to conclude simply that “higher-frequency [decadal] climate variations are generally better understood than lower-frequency variations.”

Esper’s analysis reproduces Medieval Warm Period and Little Ice Age, making 20th-century temperatures appear much less unusual than they do in Mann’s Hockey Stick. That means the derided work of Soon and Baliunas and McIntyre and McKitrick was, to put it mildly, unfairly treated by the Mann partisans, and makes the fuss over the mere publication of the papers even more unseemly.

Once again, this new development demonstrates that the science is far from being settled on this issue. Those who deny that fact are selectively ignoring a growing body of evidence.


Esper J., D.C. Frank, and J.S. Wilson, 2004. Climate reconstructions: Low-frequency ambition and high-frequency ratification. Eos, 85, 133,120.

Esper, J., E.R. Cook, and F.H. Schweingruber, 2002. Low frequency signals in long tree-ring chronologies for reconstructing past temperature variability. Science, 295, 2250-2253.

Mann, M.E., R.S. Bradley, and M.K. Hughes, 1999. Northern Hemisphere temperatures during the past millennium: inferences, uncertainties, and limitations. Geophysical Research Letters, 26, 759–762.

McIntyre, S., and R. McKitrick, 2003. Corrections to the Mann et. al. (1998) Proxy database and Northern Hemispheric average temperature series. Energy & Environment,14, 751-771.

Soon, W., and S. Baliunas, 2003. Proxy climatic and environmental changes of the past 1,000 years. Climate Research, 23, 89–110.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Sorry, the comment form is closed at this time.

Powered by WordPress