October 1, 2004

Ironies Abound in Hockey Stick Debacle

There’s new research in Science demonstrating the “hockey stick” reconstruction of earth’s temperature history over the last 1000 years suffers deficiencies that further undermine its reliability.

Why are so many researchers concerned with reconstructing a thousand years of Earth’s climate history? Some will argue it’s actually a political debate; to the winner goes the spoils — passage of or withdrawal from the Kyoto Protocol by governments worldwide.

The twentieth century indeed was warm. We know this to be so because temperatures could be measured using instruments designed for that purpose. What they indicate is that global temperature increased by about three-quarters of a degree Celsius. The question becomes: was that rate of warming unusual in a longer-term context? We’ll probably never be certain because there were no comparable instruments taking measurements in earlier centuries. Barring the unlikely discovery the Catholic Church operated a secret, well-calibrated global thermometric measurement campaign during the Crusades, for example, any comparison of contemporary measurements with those of the past will be, by definition, fraught with error.

Climatic reconstructions are tempting because they offer insight to our climatological past. Such ‘reconstructions’ are assembled using all manner of less-than-ideal climate indicators, things like boreholes, tree rings, ice cores, and historical accounts (“General Washington requisitioned five new silk blouses after the summer of 1782 in comparison to only two the previous year, from which we reconstructed the following temperature and humidity time series…”). The key to providing credence to such reconstructions is to accurately appraise the size of the error associated with each temperature estimate.

A climate reconstruction assembled by University of Virginia Assistant Professor Michael Mann is a favorite of the Kyoto-friendly Intergovernmental Panel on Climate Change (IPCC). While many other Northern Hemisphere (NH) temperature reconstructions include a Medieval Warm Period in the 11th and 12th centuries, a Little Ice Age in the 19th, and a few other short cool periods (all contributing considerable variability to the last millennium’s climate), Mann’s so-called ‘hockey stick’ indicates 900 years of approximately flat global temperature and a century on the end consisting of dramatically rising temperatures. This gives his graph its hockey-stick character with the steeply rising blade coinciding with a time when humans were adding lots of greenhouse gases to the atmosphere (Figure 1). The question arises: Were Northern Hemisphere temperatures between 1000 and 1900 really so non-variable?

Hockey Stick

Figure 1. Mann’s multi-proxy temperature reconstruction of the Northern Hemisphere for the past 1,000 years (Source: Intergovernmental Panel on Climate Change 2001).

German climatologist Hans von Storch and several colleagues cast serious doubt upon the accuracy of Mann’s reconstruction in a new paper in Science.

By way of background, climate reconstructions incorporate concurrent measurements from modern periods during which both instrumental and proxy indicators exist. Their relationship during the modern era (typically derived using statistical regression techniques) then is applied to proxy measurements from the past (when there were not instruments) to ‘hindcast’ climate.

What makes von Storch’s research so interesting is that he and his colleagues cleverly worked around the data unavailability problem using a climate model. Because their climate model is able to reasonably simulate the 20th century’s temperature record (as to mean and variability), they selected local proxies from a 1,000 year run of the model at locations where paleo-climatologists like Mann claim to have proxy samples. They then added an appropriate amount of variability (or statistical ‘noise’) to each of the local samples to simulate the non-temperature component of actual proxy data.

When von Storch recreated his modeled thousand-year Northern Hemisphere temperature history based on the relationship between his pseudo-proxies and the modeled global temperatures during the 20th century (using a technique similar to that used by Mann) he found Mann’s pseudo-proxies greatly underestimate the amount of long-term variability. The more noise that is added, the less variability is captured. For example, in Figure 2 (von Storch’s Figure 2a) shows repeated attempts to reconstruct the modeled climate history (black line) using a technique like the one employed by Mann et al. (other colored lines). It is important that our readers understand: With the statistical approach employed by von Storch et al., it is not at all necessary for their climate model to be a perfectly accurate simulation of 20th century temperature. It need only be an acceptable representation of general temperature variability over the time period.

Hockey Stick

Figure 2. Modeled temperature history for the past 1,000 years (black line) and attempts to reconstruct that history using a Mann-like multi-proxy technique when different amounts of noise are included (colored lines). The more noise, the lower the variance. (Source: von Storch et al., 2004).

In a short “Perspective” in Science that evaluates the von Storch article, East Anglia’s Timothy Osborn and Keith Briffa comment, “If the true natural variability of NH temperature is indeed greater than is currently accepted, the extent to which recent warming can be viewed as ‘unusual’ would need to be reassessed.” In the larger context of climate politics, their phraseology (“currently accepted”) becomes piquantly ironic. Their clear implication is that Mann’s temperature reconstruction (the one favored by the IPCC), is what the majority of the world’s climatologists believe to be most accurate. If that is true, then most of people studying climate must believe the Medieval Warm Period and the Little Ice Age to be minor excursions on a Northern Hemisphere temperature curve dominated by 20th century warming. The problem is, such thinking flies in the face of literally hundreds of research papers that document the existence and widespread impacts of those climatic events.

Wouldn’t it be nice if scientists would actually get together and compile evidence of those climatic events from the scientific literature? Tah dah! Guess what? Harvard scientists Willie Soon and Sallie Baliunas have done just that and published their results in the journal Climate Research, last year.

There were howls of protest by IPCC scientists, including Mann, claiming it to be necessary to change the editorial policies of Climate Research to prevent any future “mishaps” of this kind. Von Storch was appointed Editor-in-Chief. His first and, ultimately, only act was to prepare an apology for allowing the Soon and Baliunas review to see the light of day. This was quickly nixed by the journal’s publisher; von Storch resigned in protest.

Here’s the irony, the results von Storch just published in Science effectively agree with what Soon and Baliunas wrote in Climate Research in 2003: the Mann reconstruction severely underestimates past variability. So why did he protest so vociferously when he was Editor-in-Chief? And why doesn’t von Storch even reference Soon and Baliunas in his Science article, especially considering that their technique (i.e. comparative literature review) is an example of a method that is essentially free from the type of errors that von Storch identified in the work of Mann and others?

This latest research effort further bringing into question the accuracy of the Mann’s temperature reconstruction (the hockey stick) comes on the heels of another recent article by Steve McIntyre and Ross McKitrick in which they document difficulties they encountered recreating the Mann et al. temperature curve using Mann’s own data. Their paper isn’t cited by von Storch either. That one was preceded by research from Jan Esper et al. showing much greater long-term temperature variability (see http://www.co2andclimate.org/wca/2004/wca_15e.html for more information of Esper’s findings). As evidence builds in support of the long-standing paradigm that climate varies, it’s time for the IPCC to follow the lead of National Hockey League players who with their threat to strike are willing to put their hockey stick to rest in defense of a principle — in this instance not higher pay, but sound science.


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.

Intergovernmental Panel on Climate Change, 2001. Climate Change 2001: The Scientific Basis. Houghton, J.T., et al., (eds.), Cambridge University Press, Cambridge, U.K, pp 881.

Mann, M.E. R.S. Bradley, and M.K. Hughes, 1998. Global-scale temperature patterns and climate forcing over the past six centuries. Nature, 392, 779-787.

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.

Mann, M.E., and P.D. Jones, 2003. Global surface temperature over the past two millennia. Geophysical Research Letters, doi:10.1029/2003GLo17814.

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.

Osborn, T.J., Briffa, K.R., 2004. The real color of climate change. Sciencexpress, September 30, 2004.

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

Von Storch, H., et al., 2004. Reconstructing past climate from noisy data. Sciencexpress, September 30, 2004.

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