Globally Lukewarm Research

Letters by Congressman Joe Barton (R-TX), chair of the House Committee on
Energy and Commerce, to Drs. Michael Mann, Ray Bradley and Malcolm Hughes
(MBH) have raised a firestorm of responses. Mr. Barton is accused of
intimidating scientists by requesting information related to their 1998
publication [in the science journal Nature] of the so-called Hockeystick
graph; it shows temperatures over the past 1000 years, as obtained from a
statistical analysis of various proxy (non-thermometer) data, like
tree-rings, and has been used as evidence for the existence of
anthropogenic global warming (AGW).

But Mr. Bartons requests are entirely proper and legitimate. It is an
embarrassment to science that the matter had to be raised by a
non-scientist. Consider:

1. Is the 20th century really the warmest in the past 1000 years as
suggested by the Hockeystick? Probably not. This scientific issue
appears to be settled, with even supporters of AGW now admitting that the
Hockeystick is not really needed to support their conclusion of substantial
AGW.

2. Is the MBH analysis correct or should it be withdrawn from
publication? The initial referees never conducted a detailed review of the
MBH paper in spite of the fact that it disagreed strongly with other work
that clearly showed a Medieval Warm Period, followed by a Little Ice Age
(from about 1400 to 1850 AD). Two Canadian researchers, Stephen McIntyre
and Ross McKitrick (MM), tackled the forbidding task of conducting a
detailed audit. Their first publication in 2003 found that the underlying
data had been manipulated in a number of ways. Full disclosure: I was a
reviewer of this MM paper and recommended its publication. As far as I
know, their conclusion has not been successfully challenged.

In a follow-on audit, MM found that the statistical methodology used by MBH
was faulty and strongly biased toward detecting hockey-stick shapes even
with random data (WSJ Feb 14, page 1). This result, published in a leading
journal, has been widely supported. MBH have instead tried to claim it
“doesn’t matter,” but then withheld the essential computer code involved in
their calculations. (It was released only after the recent Barton letter.).

3. This episode bears on both the practice of science and the role of the
Congress: Is publicly funded research freely available to the scientific
community for legitimate purposes? After all, the essence of science is to
allow results of one group to be reproduced independently by others.

I believe that in the long run the scientific community is best qualified
to police its own affairs. Unfortunately, the climate issue, because of
its vast public-policy consequences, has become polarized; hard positions
taken on the Hockeystick controversy preclude its resolution by normal
scientific procedures. This has affected even such respected journals like
Science and Nature, which are no longer willing to provide a forum for
conflicting views.

Congress can help the scientific community in this instance by providing
such a forum through hearings where information can be exchanged and
conflicting views freely aired.

###

Atmospheric physicist S FRED SINGER (pictured, before attending Accuracy in Academia’s 20th anniversy dinner) is professor emeritus of environmental
sciences at the University of Virginia and former director of the US
Weather Satellite Service (now NESDIS-NOAA). He has authored several books
and research papers on climate science. He is a Fellow of the American
Geophysical Union, the American Physical Society, and the American
Association for the Advancement of Science. This article is excerpted from his weekly e-mailed report.