The lawsuit Michael Mann filed against people for supposedly defaming him has received quite a bit of attention in the blogosphere. In one conversation at The Blackboard, a commenter suggested it might be helpful if someone made a list of the:
top 10 or 15 of the most egregious Mann misstatements, incompetent work or deceptions. In all probability, the lawyers representing the various parties in the lawsuit (putting aside Steyn for the moment, who is unrepresented) go into the matter being unfamiliar with the long history of what has transpired with respect to Mann. It would be very time consuming to learn and understand what really happened. So, I think it would be very useful to have a list, with a compact explanation of its significance, of something like the top 10 or 15 Mann transgressions.
After giving it some thought, I decided to try making such a list. I don’t know how good my choices/descriptions will be, but it’s an interesting thing to attempt. It’s especially interesting because having a blog means I can write posts about the individual items in the list.
So that’s what I’m going to try to do. I’m copying the comment with my list here, and I’ll try to write posts discussing each item in the list (with references) over the next few weeks. I’ll also see if I can pick more topics to expand the list with.
Any thoughts, suggestions or questions are welcome. I’ll happily discuss any topic relating to Mann people might find interesting. And who knows? Maybe when I’m done, it’ll all be useful to someone.
Here are some items I’m considering for my list, along with a (relatively) short summary for each (in no particular order):
Tiljander – The Tiljander proxies were said by the data collectors to have been corrupted due to human influences, and as such, their data from ~1800 AD was worthless as a temperature proxy. Additionally, there were only two series. Not only did Michael Mann use these series with a methodology that requires proxies measure temperatures in the modern times, he used some of them upside down, and he duplicated information by using four instead of two.
r2 – Michael Mann calculated statistical verification scores for his reconstruction. He published the favorable results while hiding the adverse results. He did this to such an extent he published some scores for one test (r2) while hiding others from the same test. When challenged on this, he lied by claiming he had never calculated any of the r2 scores (despite having published some).
Additionally, the IPCC report which made him famous claimed his reconstruction passed multiple verification tests, a claim it couldn’t have made if people had known about the hidden, adverse results. Had he disclosed those adverse results, he would never have received worldwide attention, become a major public figure in the global warming debate, have written a popular book or become a player in political campaigns.
Gaspe – Michael Mann used the Gaspe tree ring data twice. In one case, he artifically extended the series further back into the past so it could reach the 1400 AD mark his paper aimed for. He did not disclose this extension nor offer any justification for it. Without it, his results would have been notably weaker. Additionally, the Gaspe tree ring data for that period was based upon a single tree. Dendrochronologists say series based upon a single tree are inappropriate for use as temperature proxies.
Temperature record as proxies – Michael Mann included modern temperature data, measured by man made instruments, in both his original reconstruction and his 2008 reconstruction. In both cases, they were treated as proxy data derived from nature.
Precipitation record as proxies – Michael Mann has repeatedly used precipitation records as proxies in his temperature reconstructions even though he co-authored a paper criticizing other authors (Soon & Balinaus) for conflating temperature and precipitation proxies.
Non-robust – In his original work, Michael Mann claimed his temperature reconstruction was robust to the removal of tree ring data. It wasn’t. He now admits (in his book) he knew it wasn’t almost immediately after publishing his original work, but he never corrected it. Instead, he wrote an another paper which built upon the earlier work.
This last one cannot be stressed enough. Michael Mann knew his temperature reconstruction was dependent upon a small amount of tree ring data from one part of the United States. Knowing this, he allowed his work to be promoted as showing what temperatures were like for the entire northern hemisphere. Had he been honest, people would have known 90% of his data was irrelevant and all that really mattered was a small number of trees in one area.
The censored directory, misused PCA and other issues tie into that last one, but really, those two paragraphs are all people need to know. That alone, or perhaps in combination with the hiding adverse results, is easily enough for a person to believe Mann’s work was fraudulent.
(Of course, there are other topics to discuss as well. This is just a beginning of a list.)
Edit: I’ll be adding links here as new posts go up. Hopefully, that’ll help people navigate.
Mann’s Screw Up #1 – “Excelgate”
Mann’s Screw Up #2 – Non-Robust
Mann’s Screw Up #2.5 – 5%
Mann’s Ego – Basis of Climate Change Evidence?
Mann’s Screw Up #3 – Statistics is Scary
Mann’s Screw Up #3 – Cherry Picking
Mann’s Screw Up #3.2 – Rotten Cherry #1
Mann’s Screw Up #3.2 – Rotten Cherry #2
Mann’s Screw Up #4 – Hiding Results
Mann’s Screw Up #4.1 – Covering Up Results
Mann’s Screw Up #5 – Inappropriate Data
Mann’s Screw Up #6 – Altering Data
Mann’s Screw Up #6.1 – “Fixing” Data
Mann’s Screw Up #7 – His “Trick”
Mann’s Screw Up #7.1 – Semantics
Mann’s Screw Up #8 – Hypocrisy
Mann’s Screw Up #9 – Overstated Conclusions
Mann’s Screw Up #9.1 – Nonsense Data