Log in


The Hockey Stick Illusion

The Hockey Stick Illusion: Climategate and the Corruption of Science by AW Montford is an interesting and comprehensive work on the graph that purportedly demonstrates that the 20th century had unprecedented warm temperatures. He  provides a good historical background and a detailed analysis of what the flaws are.

There are a couple of important historical bits early in the book. One is the graph of temperatures over time that appeared in the first IPCC report. That graph showed a medieval warm period, which would indicate that recent warm temperatures are not unprecedented. Indeed, that graph looked nothing like a hockey stick. The other is the email climate scientist David Deming received from another climate scientist that said "We have to get rid of the medieval warm period."

The graph known as "The Hockey Stick" was developed by Micheal Mann in 1998, and gets its name from the sharp uptick in temperatures in the 20th century which makes it look like a hockey stick. It was used in later IPCC reports to bolster it's diagnosis of unprecedented, human-caused climate change.

To the extent that a scientific non-fiction book can have a hero,  Canadian mining engineer  and founder of the Climate Audit blog Steve McIntyre is the hero of  The Hockey Stick Illusion. Much of the book deals with Steve McIntyre's problems with the hockey stick graph and his attempts to either resolve or correct them. The book contains many episodes of his attempting to get data and code, but constantly being rebuffed. I knew this happened back in 2007 when the "1998 is the warmest year on record" canard was debunked, but I had no idea how systemic it was. By the time I finished the book, I was quite impressed with McIntyre's persistence in the face of stonewalling and personal attacks.

There are some amusing stories of problems McIntyre found in Mann's data and methodology.  In one instance "79 consecutive values from a table of Eurodollar interest rates achieved an R2 [correlation}...far in excess of...[Manns temperature] proxies".  He also wrote a paper which contained 12 hockey stick graphs, all of which used Mann's methodology, but 11 of which used random data. He made presentations asking people which was Man's graph and which was based on statistical noise. The graphs are in the book, and look remarkably similar to each other.

Despite the subtitle, there are only a couple of chapters on climategate in the book. The leak of emails from East Anglia occured while the book was being written, so the climategate chapters are a last minute addition. Nevertheless, they prove to be helpful and informative. A little light is shed on the famous "get rid of the medieval warm period" remark. Although there is an email from Jonathan Overpeck (the alleged source) in which he says "I have no memory" of sending the famous email, there is evidence of a desire on the part of many climate scientists to "get rid" of the medieval warm period. For instance,  Keith Briffa in an  acknowledges that there is "pressure"  to indicate that 20th century temperatures are unprecedented, but that  "in reality the situation is not so simple."  Micheal Mann sends an email in which he says it  "it would be nice to try to 'contain'" the medieval warm period, "even if we don't yet have a hemispheric mean reconstruction available that far back."

Perhaps a better subtitle would have been "The Corruption Of Science And Climategate", because the former is covered in considerable depth. Much of the corruption comes from the politicization of the science. Particularly chilling is an excerpt of an IPCC guideline on scientific reports:

Changes...made after acceptance by the Working Group or the Panel shall be necessary to ensure consistency with the Summary for Policymakers or the Overview Chapter

In other words, when the policy recommendations aren't consistent with the scientific reports, change the reports. Very disturbing.

There are many other tales of malfeasance from the book I have not mentioned here, along with detailed analysis of the mathematics and methodology behind the hockey stick. It sometimes makes for dense reading, but is more than worth the effort.