Tag Archives: mcintyre

More Hockey Stick Hyjinx

Update: Keith Briffa responds to the issues discussed below here.

Sorry I am a bit late with the latest hockey stick controversy, but I actually had some work at my real job.

At this point, spending much time on the effort to discredit variations of the hockey stick analysis is a bit like spending time debunking phlogiston as the key element of combustion.  But the media still seems to treat these analyses with respect, so I guess the effort is necessary.

Quick background:  For decades the consensus view was that earth was very warm during the middle ages, got cold around the 17th century, and has been steadily warming since, to a level today probably a bit short of where we were in the Middle Ages.  This was all flipped on its head by Michael Mann, who used tree ring studies to “prove” that the Medieval warm period, despite anecdotal evidence in the historic record (e.g. the name of Greenland) never existed, and that temperatures over the last 1000 years have been remarkably stable, shooting up only in the last 50 years to 1998 which he said was likely the hottest year of the last 1000 years.  This is called the hockey stick analysis, for the shape of the curve.

Since he published the study, a number of folks, most prominently Steve McIntyre, have found flaws in the analysis.  He claimed Mann used statistical techniques that would create a hockey stick from even white noise.  Further, Mann’s methodology took numerous individual “proxies” for temperatures, only a few of which had a hockey stick shape, and averaged them in a way to emphasize the data with the hockey stick.  Further, Mann has been accused of cherry-picking — leaving out proxy studies that don’t support his conclusion.  Another problem emerged as it became clear that recent updates to his proxies were showing declining temperatures, what is called “divergence.”  This did not mean that the world was not warming, but did mean that trees may not be very good thermometers.  Climate scientists like Mann and Keith Briffa scrambled for ways to hide the divergence problem, and even truncated data when necessary.  More hereMann has even flipped the physical relationship between a proxy and temperature upside down to get the result he wanted.

Since then, the climate community has tried to make itself feel better about this analysis by doing it multiple times, including some new proxies and new types of proxies (e.g. sediments vs. tree rings).  But if one looks at the studies, one is struck by the fact that its the same 10 guys over and over, either doing new versions of these studies or reviewing their buddies studies.  Scrutiny from outside of this tiny hockey stick society is not welcome.  Any posts critical of their work are scrubbed from the comment sections of RealClimate.com (in contrast to the rich discussions that occur at McIntyre’s site or even this one) — a site has even been set up independently to archive comments deleted from Real Climate.  This is a constant theme in climate.  Check this policy out — when one side of the scientific debate allows open discussion by all comers, and the other side censors all dissent, which do you trust?

Anyway, all these studies have shared a couple of traits in common:

  • They have statistical methodologies to emphasize the hockey stick
  • They cherry pick data that will support their hypothesis
  • They refuse to archive data or make it available for replication

The some extent, the recent to-do about Briffa and the Yamal data set have all the same elements.  But this one appears to have a new one — not only are the data sets cherry-picked, but there is growing evidence that the data within a data set has been cherry picked.

Yamal is important for the following reason – remember what I said above about just a few data sets driving the whole hockey stick.  These couple of data sets are the crack cocaine to which all these scientists are addicted.  They are the active ingredient.  The various hockey stick studies may vary in their choice of proxy sets, but they all include a core of the same two or three that they know with confidence will drive the result they want, as long as they are careful not to water them down with too many other proxies.

Here is McIntyre’s original post.   For some reason, the data set Briffa uses falls off to ridiculously few samples in recent years (exactly when you would expect more).  Not coincidentally, the hockey stick appears exactly as the number of data points falls towards 10 and then 5 (from 30-40).  If you want a longer, but more layman’s view, Bishop Hill blog has summarized the whole storyUpdateMore here, with lots of the links I didn’t have time this morning to find.

Postscript: When backed against the wall with no response, the Real Climate community’s ultimate response to issues like this is “Well, it doesn’t matter.”  Expect this soon.

Update: Here are the two key charts, as annotated by JoNova:


And it “matters”


The First Rule of Regression Analysis

Here is the first thing I was ever taught about regression analysis — never, ever use multi-variable regression analysis to go on a fishing expedition.  In other words, never throw in a bunch of random variables and see what turns out to have the strongest historical relationship.  Because the odds are that if you don’t understand the relationship between the variables and why you got the answer that you did, it is very likely a spurious result.

The purpose of a regression analysis is to confirm and quantify a relationship that you have a theoretical basis for believing to exist.  For example, I might think that home ownership rates might drop as interest rates rose, and vice versa, because interest rate increases effectively increase the cost of a house, and therefore should reduce the demand.  This is a perfectly valid proposition to test.  What would not be valid is to throw interest rates, population growth, regulatory levels, skirt lengths,  superbowl winners, and yogurt prices together into a regression with housing prices and see what pops up as having a correlation.   Another red flag would be, had we run our original regression between home ownership and interest rates and found the opposite result than we expected, with home ownership rising with interest rates, we need to be very very suspicious of the correlation.  If we don’t have a good theory to explain it, we should treat the result as spurious, likely the result of mutual correlation of the two variables to a third variable, or the result of time lags we have not considered correctly, etc.

Makes sense?  Well, then, what do we make of this:  Michael Mann builds temperature reconstructions from proxies.  An example is tree rings.  The theory is that warmer temperatures lead to wider tree rings, so one can correlate tree ring growth to temperature.  The same is true for a number of other proxies, such as sediment deposits.

In the particular case of the Tiljander sediments, Steve McIntyre observed that Mann had included the data upside down – meaning he had essentially reversed the sign of the proxy data.  This would be roughly equivalent to our running our interest rate – home ownership regression but plugging the changes in home ownership with the wrong sign (ie decreases shown as increases and vice versa).

You can see that the data was used upside down by comparing Mann’s own graph with the orientation of the original article, as we did last year. In the case of the Tiljander proxies, Tiljander asserted that “a definite sign could be a priori reasoned on physical grounds” – the only problem is that their sign was opposite to the one used by Mann. Mann says that multivariate regression methods don’t care about the orientation of the proxy.

The world is full of statements that are strictly true and totally wrong at the same time.  Mann’s statement in bold is such a case.  This is strictly true – the regression does not care if you get the sign right, it will still get a correlation.  But it is totally insane, because this implies that the correlation it is getting is exactly the opposite of what your physics told you to expect.  It’s like getting a positive correlation between interest rates and home ownership.  Or finding that tree rings got larger when temperatures dropped.

This is a mistake that Mann seems to make a lot — he gets buried so far down into the numbers, he forgets that they have physical meaning.  They are describing physical systems, and what they are saying in this case makes no sense.  He is essentially using a proxy that is essentially behaving exactly the opposite of what his physics tell him it should – in fact behaving exactly opposite to the whole theory of why it should be a proxy for temperature in the first place.  And this does not seem to bother him enough to toss it out.

PS-  These flawed Tiljander sediments matter.  It has been shown that the Tiljander series have an inordinate influence on Mann’s latest proxy results.  Remove them, and a couple of other flawed proxies  (and by flawed, I mean ones with manually made up data) and much of the hockey stick shape he loves so much goes away