Wednesday, 31 March 2010

A. W. Montford: The Hockey Stick Illusion

If you’ve found this review you probably know the story but basically it is this. Temperatures before the middle of the 19th century can only be estimated indirectly from proxies: physically measurable characteristics of plants or animals which lived at earlier times and which respond to temperature changes. In 1999 a young climate researcher, Michael Mann, and colleagues published a paper in Nature which suggested that temperatures at the end of the 20th century were higher than at any time in the previous 600 years. The graph of this, shaped like an ice-hockey stick, became iconic and appeared several times in the IPCC 2001 technical assessment report and elsewhere. In 2002 a retired mining engineer, Steve McIntyre, became suspicious of the fact that the paper both contradicted conventional wisdom and gave an ideal message for climate change activists that he started to investigate. His findings led to Congress setting up to two high level technical enquiries and a congressional committee held hearings.

Those of you who have seen my review of ‘Dire Predictions’ by Mann and Kump might imagine that I would automatically give a favourable review to this book. After all it is the very antithesis of the view of climate by IPCC insiders like Mann and Kump. But that is not how we work at climatedata.info. We try, as far as possible, to avoid polemics for polemics sake and let the data speak for themselves.

The author of this book, Andrew Montford, is a blogger who uses the pseudonym ‘Bishop Hill’. He started off as a political blogger and has a Bill of Rights for the UK on his site.  The Bill of Rights shows that he is of a libertarian disposition and is therefore not predisposed to accept the restrictions which would be necessary to bring about a major reduction in carbon emissions. He admits in the book that his site only took off when he started blogging on climate the ‘Hockey Stick’.

The book reads well, is well referenced, and appears to be a crushing indictment of the IPCC and the paleoclimatologists who developed the ‘Hockey Stick’. To the author, Steve McIntyre is definitely one of the ‘good guys’ and beyond reproach. Whilst it tries to present arguments from both sides there is no doubt that the book is very partisan.

In referring to the meeting to discuss the ‘Hockey Stick’ by the National Academy of Sciences in the USA he discusses the response of Michel Mann to a question concerning verification of the calculations. In a published paper Mann had reported that he had calculated a particular statistic (r squared). In the book he is quoted as telling the meeting “We didn’t calculate it. That would be silly and incorrect reasoning.” I have Googled those phrases and the only source I can find is Steve McIntyre’s blog at climateaudit.org which is the source referenced by the author. There appears to be no transcript of the NAS meeting, at least I could not find one online. I have also studied the transcript, and listened to much of the recording, of the subsequent congressional hearing and can find no evidence to support what is claimed to have been said at the NAS meeting. My attempts to get to the bottom of this do not prove or disprove what has been claimed but to make such a serious charge without independent  verification is a clear example of bias.

Throughout the book frequent reference is made to scientists withholding data and there is always the assumption that if data is withheld there are nefarious reasons for it. One of McIntyre’s complaints was that Bristlecone Pine trees were central to calculation of the ‘hockey stick’ but that the data from the trees stopped in 1980, just as the main warming started. He therefore set about visiting the site to resample the trees. This was done and the cores were re-analysed. I have tried to access the data using the links on his site including http://climateaudit.org/data/colorado/SMCD.dat but the links are dead. In a footnote in the book the author says [McIntyre] “has yet to publish the findings in a journal.” I am not assuming there is anything sinister in this but the fact that the author so glibly accepts the lack of openness from McIntyre but assumes ill-intent from others is a further demonstration of his bias.

Enough of the polemics: where does this leave the science?

Climate scientists are under pressure from politicians to present a clear picture. There is no dispute about the fact that many climate scientists felt that if temperatures had been warmer in the historic past than the present this would dilute the message they could deliver to the politicians. When a climate scientist appeared to be able to prove that current temperatures were indeed higher than in the past millennium they jumped at it as way of getting their message across. The ‘Hockey Stick’ became the star of the 2001 IPCC report.  However my feeling is that on balance the climate science community regrets the prominence given to the hockey stick. There are two reasons for this.

Firstly the aim of the IPCC is to convince the public of the necessity to take action on climate change. The use of temperature proxies has been so well discussed that many well informed citizens are aware of the need to ‘hide the decline’ or believe that everything is based on a few trees in Siberia. The ‘decline’ refers to the fact that in recent decades trees in  the northern hemisphere are not responding to temperature increases in the way they were assumed to; this brings into question the accuracy of long-term temperature estimate based on tree rings. Many people also believe that scientists have ‘cherry-picked’ the trees, the Yamal series from Siberia, that give the answer they want.  This was one of McIntyre’s first criticisms of Mann, that his method selected proxies that gave a ‘hockey stick’. Many scientists now would agree that temperatures in the medieval period were similar today but the science is not accurate enough to be able to say whether they were slightly warmer or cooler.

In an attempt to support the ‘hockey stick’ some climate scientists, including Mann and Kump in their book, say they accept that in parts of the world the medieval period was warm but the warming was less general than today. This argument is a bit disingenuous. In those parts of the world where written records of the medieval period exist there is evidence that medieval temperatures were high but of course not all parts of the world have written records of that period. So the argument comes down this; one side says “warming only occurred in parts of the world with written records”, and implies it did not elsewhere and the other side says “warming occurred in all parts of the world with written records” and implies that it did elsewhere.

Secondly, the question of past temperatures does not of itself determine whether or not anthropogenic warming is taking place. The temperature could be rising as a result of human action now even if temperature in the past were higher than today. Where it is critical is in discussing the impacts. If temperatures had been 2 °C higher in the past it is difficult to argue that a rise of that order of magnitude would be catastrophic.

From reading his blog I have always felt that Steve McIntyre viewed his role as scientific and he has made many attempts to reach out to the scientific community. Andrew Montford definitely has a political agenda and would probably be sceptical about climate change whatever the science said. I am not sure that McIntyre is well served by this book.

Author: A. W. Montford
Publisher: Stacy International, 2010
ISBN: 978 1 906768 35 5

2 comments:

Bishop Hill said...

McIntyre moved his site to Wordpress shortly after Climategate. The data files have not been transferred over, but can still be seen on the old site. The bristlecone data is here:

http://38.114.169.124/data/colorado/

Administrator said...

Thanks for the information.