A common exercise in any intro statistic class is to split the students into two groups, one group flips a coin 100 times and records the results, another just makes 100 entries up off the top of their heads. The teacher then comes back, looks at the two lists, and usually identifies which is which with hardly a glance. How? The trick is the teacher knows that on the real list, there will be several sequences of four or five in a row of all heads or all tails, whereas on the other list students will tend to stick with a more heads-tails-heads-tails alternating approach.
Now, everyone knew what I meant just now when I wrote trick, right? Nothing deceitful, simply the method used to get an answer to a math problem. With that in mind, let's look at this 1999 email purporting to be evidence of fraud among some climate scientists:
"I've just completed Mike's Nature trick of adding in the real temps to each series for the last 20 years (i.e., from 1981 onwards) and from 1961 for Keith's to hide the decline."
The email is one of thousands sent over a period of ten years by climate researchers and other scientists, journalists, lobbyists, and the occasional flake, stolen from a university network a few days ago. DeSmogBlog has more on the theft. Obviously, emails don't change the observed reality of human assisted climate change in the cryosphere and elsewhere. Nevertheless, climate change denialists have combed through them looking for anything they can pull out of context and pass off as evidence of a global conspiracy. They're getting some media mileage out of it. Even though, so far, the best they've been able to come up with is examples like the above.
The "Mike" is Michael Mann, "hide" means to account for (See also this comment), and the trick referred to is how to resolve a question involving two sets of data. One set is the "real" actual temperature readings, the other is by proxy, tree-rings, corals, ice cores and the like. When reconstructing the temperature record going back a thousand years or more, proxies are all you get -- there were no super accurate thermometers handily placed around the globe during medieval times! But proxies only give an approximation, hence the large variance in the now familiar reconstruction graph affectionately known as the Hockey Stick represented below as shading around the blue and red lines.
But as time rolls by, and proxy data become more plentiful, the error bars (i.e. the variable shading) shrink. Eventually, thanks to the invention of modern thermometers and ships to carry them, precise temperature readings from all over the world become more widely available and increasingly reliable, and there is a relatively short interval where both the proxy reconstruction (blue) and the instrumental record (red) are used. The proxy record ends (1980), but the instrumental record continues through 1999. That was the issue being discussed in the emails: why end the plot in 1980 when there's instrumental data through the 90s? In the original 1998 paper published in Nature, Mann et al showed the instrumental data through the 1990s to complete the plot. The emailer was following suit in his own work. That's "Mike's Nature trick". It really is that simple.
Moreover, both instrumental and proxy records were clearly labeled and delineated in the original papers and many since, so there was no opportunity for any ambiguity as to what was being shown. It makes sense that "Mike" Mann would be mentioned, he has worked extensively with both kinds of data, actual and proxy, and was one of the original paleo-climatologist who developed the Hockey Stick using them. The email is just a tiny snippet of several colleagues in the midst of discussing these points and others.
How hard was it to figure this out? Anyone could have done it, assuming they wanted to. Just like any competent reporter, I asked the people involved, including Prof. Mann:
SA: What was the Mike's "trick"?
MM: All he (apparently) meant by "Mike's Nature trick" was us, in our original '98 Nature article, showing the instrumental record after the proxy record ends (1980). ... Full Text
But consider; it's taken me several grafs, and you a few minutes of reading, just to get a glimmer of what that one email was all about. The same effort would be required to untangle other stolen, out of context emails now brandished by skeptics as evidence of some kind of shadowy conspiracy. That's how easy it is to pluck something out of context and make it sounds ominous, if your goal is to misinform, prostitute yourself to the energy industry, and -- pardon the pun -- trick your readers.
