Review of Chapter 3 (part 2), The Hockey Stick Illusion by A. W. Montford
In an attempt to see both sides of the debate in the question of human caused climate change, I am starting to read this book: The Hockey Stick Illusion: Climategate and the Corruption of Science (Independent Minds) by Andrew Montford
As I read it, I will give a chapter by chapter review comprising quotes and commentary. So, if this is an area of interest, keep a watch.
All material quoted from the book is in italics. My comments are in plain text in brackets.
With his long experience of the
mining industry, McIntyre was well equipped to get to the underlying truth of a
compelling graphic like the Hockey Stick. In a posting on Climate Skeptics, he
pondered some similarities between the work of a mining analyst and a climate
auditor.
[A]n individual
time-series has much the same function as a drill-hole. Where there is an
ore-body (i.e. a significant ‘signal’), the information in the individual
drill-holes is not subtle. Any analyst recommending a mining stock has to look
at the drill holes – not just the compilations. The application of valid
statistical methods to invalid data can result in fiascos like Bre-X [a famous
mining scandal in which drill-hole results had been ‘improved’]. ‘Adjustments’
are always something to be suspected.
[This
is a rather unfair comparison, at best, since the Bre-X
scandal
represented intentional fraud and Mann’s climate predictions are being
discussed as a case of poor data and methodology. Of course, such a comparison
does reveal one of the underlying assumptions of McIntyre and Montford, i.e.
that Mann is part of a conspiracy of climate researchers trying to dupe the
world into believing anthropogenic global warming is real so they can keep
getting grant money to study it. This, being a central assumption of climate
contrarians, in general, deserves a little more discussion.
For a
conspiracy of this sort to be plausible, there must be at least a reasonable
motive. The stated motive by climate contrarians is that climate researchers
want to keep getting expensive grants. This is an odd motive if one understands
how scientific grants are distributed. Competition for grants is not based on
the particular outcome of past research, but rather on the quality of past
research and its relevance to furthering scientific knowledge and
understanding. Whether the results from climate research show anthropogenic
warming or not will not affect the potential to get funding, as long as the
published results stand up to scientific scrutiny.
Of
course, the argument goes something like this: since essentially all the
peer-reviewed papers on climate change are in agreement, this must be because
there is a conspiracy. This contention is further buttressed by the contention
that any paper that finds contrary evidence gets rejected by the peer-reviewed
journals. The problem with using this evidence to prove there is a conspiracy,
is that it can just as easily be interpreted that the reason most of the
published papers support anthropogenic warming is that the evidence actually
supports that interpretation, and that the papers that are rejected are
rejected on technical grounds, i.e. they are poorly done or are based on
insufficient data.
In order
to distinguish between these two possibilities it might be useful to follow the
money. Sure, climate researchers stand to get grants if they successfully carry
out and publish their research, but continuing to get grants is not dependent
on the specific outcome of the research, either pro or con anthropogenic
warming, as long as the research is well done and passes peer review. Most of
the contrarian papers have been written by scientists outside of the climate
science profession, many of which have received funding from fossil fuel linked
sources. These kinds of fund sources do have a definite stake in the
conclusions of the research they fund. The fossil fuel industry stands to lose
trillions of dollars if anthropogenic warming is taken seriously. So, of the
two sides in this controversy, which would be more likely to be a part of a
conspiracy? Who stands to gain the most from a particular outcome in the
climate controversy? If there is a conspiracy, it would seem that the fossil
fuel industry stands to lose the most, not climate scientists. Regardless of a
conspiracy or not, the case should be decided by scientific evidence.]
As he got his hands on more and
more proxy data, McIntyre became frustrated by the fact that most of the
proxies stopped at around 1980. This meant that the dramatic warming of the
1980s and 1990s, which should have vastly inflated the ring widths, couldn’t be
seen. As he tartly observed:
If the IPCC were a feasibility
study for a mere $1 billion investment in a factory or a mine, you can be sure
that the engineers would bring all this type of data up to date. The casualness
of the IPCC process in respect to not bringing the data up to date (but relying
on it for sales presentations) is really quite awe-inspiring.
[I am
not even sure how to evaluate these statements. Why would proxies be used when
actual measurements of temperature are so freely available. The very point of
proxies is that they enable estimation of historical temperatures when and
where there were no adequate, direct measurements. The closer to the present
data were gathered the more they represent direct measurements. In fact, direct
measurements are used to calibrate the proxy estimates. So, far from being a
fault, lack of proxies from the 80s and 90s, they simply reflect the
availability of more accurate temperature data.
Proxies
are also used for calibration, but there was really no need to have more recent
proxies than 1980, considering that very good temperature data and proxies were
available for much of the 20th century, more than adequate for
calibration of the older proxies.]
After a couple of weeks, and
following a few gentle reminders, an email from Scott Rutherford popped into
McIntyre’s inbox, indicating that the proxy data was now available on
Rutherford’s FTP site at the University of Virginia.
[Much
of the remaining portion of this chapter is devoted to thoroughly trashing Mann’s
data. Unfortunately, as noted in the review of the first half of the chapter,
McIntyre appears to have misunderstood the format of the data. Reading the
extended description McIntyre’s findings makes me almost wonder whether it goes
beyond a mere misunderstanding. Some of the criticisms seem over the top. The
only way to actually believe even a part of his criticisms, one needs to make
some rather astounding assumptions. First, one must assume that Mann is an abysmally
sloppy researcher, given the utter mess described in McIntyre’s analysis of the
data. If the data were as much of a mess as McIntyre claims, then other climate
researchers, presumably more careful and skilled than Mann, would have found
results contradictory to those of Mann. Instead, since the publication of MBH98
and MBH99, at least a dozen other climate papers have reconfirmed the conclusions
found in Mann’s papers. Given that these other studies, in some cases used
entirely, or substantially, different data sets, the only way to dismiss these
additional studies would be to either assume that all climate researchers play
fast and loose with their data, or that they are all co-conspirators willing to
lie outright to support Mann’s hockey stick. I realize this is what some
climate contrarians try to insist, but this is an unsubstantiated conspiracy theory,
at best.
Second,
one must assume that McIntyre, in contrast, is an extremely careful and skilled
statistician and that he has a thorough grasp of the calculations required to
perform temperature reconstructions. Considering that McIntyre has no formal
training whatsoever in climate science, this latter assumption seems
unwarranted. He can claim what he likes about himself and his skills, but the
fact that he lacks both the relevant training and publication history, suggests
he is likely a less reliable source than Mann.]
Now that the proxy database had
been cleaned up, all the mistakes corrected, and up-to-date data collected,
putting together a recalculation of the reconstructed temperature should have
been easy, but with Mann’s description of his methods being so vague, it was
still a hard task to work out exactly what he’d done.
[Given
that others who have accessed the same data have found none of the grave errors
identified by McIntyre, it is difficult to believe that his reconstitution of
the data can be trusted. Even more outrageous is his accusation here that Mann’s
description of methods is vague. This is claimed only as an untrained and
inexperienced climate researcher could, as Mann’s methods are about as standard
as they come in climate circles. When I first looked at Mann’s methods, I too
was a bit baffled, not being a climate researcher myself. After noting Mann’s
citations concerning his methods, and checking many of those sources, it
appears to me that his methods are pretty clearly outlined. That McIntyre finds
Mann so vague just belies his ignorance of the relevant methodology, which is
no fault of Mann’s. I will agree, though, that it did take a bit of digging to figure
out how some of the statistics were computed, but this, again, was due to my
own ignorance concerning the nuts and bolts of temperature reconstruction.]
With a good approximation of
Mann’s methodology at hand, McIntyre now reached the moment of truth. How would
correcting all the errors in the database affect the results? McIntyre pointed
the program at the corrected data and set the calculations in train. In a
minute he had the answer: when he saw the results, it was clear that the
hunches he’d had when looking through the graphs of the proxies were entirely
borne out. With the database corrected, the handle of the Hockey Stick was
warped – that is to say, there was a pronounced Medieval Warm Period. In fact
the temperatures of the reconstructed fifteenth century were even higher than
those reached in the twentieth century.
[The
best I can say in response to this closing salvo is to quote from The HockeyStick and the Climate Wars:
“The
paper’s dramatically different result from ours—purporting an extended warm
period during the fifteenth century that rivaled late-twentieth-century
warmth—was instead an artifact of the authors’ having inexplicably removed from
our network two-thirds of the proxy data we had used for the critical
fifteenth–sixteenth-century period.”
So,
it appears that the very thing Mann is accused of, leaving out data and
cherry-picking, is done by McIntyre. Even had McIntyre’s work had some
credence, it does not explain why every other major climate study by other
researchers came to the same general conclusions as Mann. Additionally, even if
McIntyre’s reconstruction of temperatures was able to show that the MWP was as
warm, or maybe warmer, than recent warming, it would not negate the
significance of the recent warming. Since the MWP was caused by factors other than
rising CO2, what does it matter that it was as warm as recent times?
In some sense that might even be a scarier result. If the same factors that
caused the MWP were to occur, with CO2 also rising, our predicament
would be even worse, and the need to reduce greenhouse gas emissions even
greater. At best, McIntyre’s conclusions, even if correct, are little more than
a distraction from the documented warming of the past several decades.
Just
as for the first half of Chapter 3, here are grammatically incorrect uses of
the word data in the last half of the chapter:]
So, according to Rutherford, and
somewhat contrary to what Mann had said, the data wasn’t even in one place.
After a couple of weeks, and
following a few gentle reminders, an email from Scott Rutherford popped into
McIntyre’s inbox, indicating that the proxy data was now available on
Rutherford’s FTP site at the University of Virginia.
[To
be fair, here is one case where Montford used the word data correctly:]
At the topmost rows of the file
were the data from the oldest proxies, the first starting in the year 1400.