Michael Mann, together with Bradley and Hughes, produced the "Hockey Stick graph" that eliminated the Medieval Warm Period (MWP) and the subsequent Little Ice Age (LIA). This was after IPCC lead author Jonathan Overpeck has said that the MWP had to be eliminated to scare people with the Global Warming hoax. (link)
Professor Larry Bell wrote in Forbes:
The hockey stick graph at the center of this dispute was based heavily upon data taken from trees on the Yamal Peninsula in Siberia. Created by Mann and his colleagues, it supposedly proved that air temperatures had been stable for 900 years until the 20th century, and then suddenly rocketed off the charts (attributing this to human-caused greenhouse gas emissions).
But there were some problems with that graph and the research behind it. Some very big problems. One was that the Medieval Warm Period which occurred between about AD 800 and 1100 along with the Little Ice Age (not a true Ice Age) which occurred between about AD 1350-1850 somehow turned up missing. And as for those Yamal tree samples, they came from only 12 specimens of 252 in the data set… while a larger data set of 34 trees from the same vicinity that weren’t used showed no dramatic recent warming, and warmer temperatures in the Middle Ages.Interestingly Michael has made some pronouncements in a post on Peter Guest's blog:
“One side, us, the scientists, have to be true to our principles, have to be truthful to our audience, have to state our findings with appropriate caveats, and the other side sees absolutely no need to do that.”Surely he missed stating the caveats with his MBH98 paper. But the amazing piece from Michael's MANNual is almost unbelievable coming from a scientist:
“Proof is for mathematical theorems and alcoholic beverages. It’s not for science."Proof is for mathematical theorems and alcoholic beverages. Really? Is this the new Science?
"Science works in evidence through best explanations, most credible theories."The "credible" theory that carbon dioxide will cause catastrophic global warming? The "credible" theoretic model which, when phone numbers are inputted, gives a "hockey stick"output?