Mann’s Screw Up #4 – Hiding Results

Thus far in this series of posts, we’ve established Michael Mann’s original hockey stick was dependent entirely upon a small amount of data. We’ve also established Mann knew this. The obvious question is, why didn’t anyone else? Shouldn’t basic tests have caught this problem?

The answer is, they did. Mann just covered it up.

Mann performed several statistical tests to check the validity of his temperature reconstruction. His tests gave scores between -1 and 1. The higher the value, the better the score. The better the score, the better his reconstruction did on the test.

You can view it like a test in school. A score of .76 is equal to a 76%. On a difficult test, that could be a good score. On the other hand, a score of .00003 (.003%) could never be a good score. And if you got a negative score, that would probably indicate a problem.

Now then, you’ll recall Michael Mann did his reconstruction in steps because different data was available for different periods. He did tests for each of those steps. These are the results of some of those tests:

WA-R2VerificationStats

You’ll notice the values in the left column are respectable. Those are the scores done for the calibration period. Obviously, scores during calibration mean little. The scores that matter are the next ones, the verification scores. They tell us whether or not Mann’s tests verify his work’s quality.

They don’t. Prior to 1750, Mann’s reconstruction doesn’t score above a 2%. That’s terrible. Anyone who heard about that result would have been troubled. That’s why Mann didn’t publish it. He didn’t mention it in his paper, and while he published the results of many tests he perfomed, he intentionally left that test out.

Pretty bad, right? It gets worse. Michael Mann was made a lead author on the paleoclimatology (study of past climates) chapter in the Intergovernmental Panel on Climate Change’s 2001 report on global warming. This report was the most important publication on global warming, being seen by every government and media organization in the world. The chapter Mann was responsible for says:

Mann et al. (1998) reconstructed global patterns of annual surface temperature several centuries back in time. They calibrated a combined terrestrial (tree ring, ice core and historical documentary indicator) and marine (coral) multi-proxy climate network against dominant patterns of 20th century global surface temperature. Averaging the reconstructed temperature patterns over the far more data-rich Northern Hemisphere half of the global domain, they estimated the Northern Hemisphere mean temperature back to AD 1400, a reconstruction which had significant skill in independent cross-validation tests.

Mann knew his reconstruction failed a cross-validation test he performed. He hid that fact. He then described his work as having “significant skill in independent cross-validation tests” in a major, intergovernmental report, which gave his work prominent display and made him known worldwide.

Had he not covered up the unfavorable results of his tests, that likely wouldn’t have been possible. It’s likely he’d never have become remotely famous, become a central figure to the global warming debate or been able to write a popular book and make lots of money off it. In short, if he had disclosed all the results of the tests he performed, he’d have probably been a nobody for the rest of his life.


Stay tuned for the next post. In it, I’ll discuss how the attempts to cover this up continued for over a decade.

Advertisements

6 comments

  1. The table that’s included in this post (“Table 1S Pearson’s r^2 and CE scores for MBH reconstruction emulations”) has a format that’s typical of the peer-reviewed literature. Yet you say “Mann didn’t publish [the results in that table].” Does this mean that Table 1S was published as a part of somebody else’s paper?

  2. Yup. I’ll be discussing it in more detail in my next post, but that table was published by Wahl and Ammann in a paper defending Michael Mann’s work. Wahl and Ammann did not want to publish that paper, and they only did so because an academic misconduct complaint was filed over them trying to hide those results.

    That’s actually why I picked that particular image. I thought it’d provide a nice segue into the next post. Plus, I figure Mann’s supporters won’t dispute results of calculations if I cite them in publications by Mann or his supporters.

  3. Interesting that originally, it seems W&A weren’t going to publish that table at all, and it was Steven Schneider (of all people) and a reviewer who insisted they do so:

    Caspar and I are also now responding to Steve’s final requests, based on independent
    re-review. This is primarily to address publishing Pearson’s r^2 and CE calculations
    for verification, which Steve and the reviewer reason should be done to get the
    conversation off the topic of us choosing not to report these measures, and onto the
    science itself. We explain thoroughly in the appendix I mention above why we feel these
    (and other interannual-only) measures of merit are not of much use for verification in
    the MBH context, so that the fact we are reporting them is contextualized appropriately.

    http://www.ecowho.com/foia.php?file=0004.txt&search=wahl

  4. Wahl to Briffa:

    Wahl-Ammann show very clearly that there is objectively demonstrated skill at the low-frequency level of the
    verification period mean for all the MBH segments, although the earlier MBH segments do
    have really low r^2 values (indicating very little skill at the interannual level). Our
    argument that to throw out the reconstruction completely based on the fastest varying
    frequency, when it has objectively demonstrable meaning at lower frequencies, is to me
    quite reasonable. That it is some how entirely ad hoc, as McIntyre claims in one
    (more?) of his comments, is neither logical nor factual in my perspective.

    http://www.ecowho.com/foia.php?file=2829.txt&search=wahl

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s