“Minority report: 50 year warming due to natural causes“. Anthony Watts reposts a blog article by Dr. Roy Spencer. Roy has fiddled around (from his own comments: “this was the result of a couple of hours of work on the weekend, and I didn’t mean to start a whole new research effort”) and managed to amaze himself by extracting a correlation that he uses to claim an unspecified natural cause (“changes in cloud cover”?) for the last 50 years of warming.
How does he do this? Why by mashing together the Pacific Decadal Oscillation (PDO), Atlantic Multi-decadal Oscillation (AMO) and Southern Oscillation Index (SOI) and only considering temperature change rate.
I've added the causation to Roy's correlation... After Roy Spencer, 2010/06/06.
Roy often shoots himself in the foot when he tries to use numbers and sure enough a disclaimer about a calculation error was posted within hours. Through the magic of denialist revision its gone now though. I should have grabbed the page for your entertainment.
Let’s say it together: “correlation is not causation.” Roy needs to present a clear mechanism for what he is describing. Turns out that what Roy has actually been plotting is temperature vs temperature, which obviously tracks itself very well…
Note that this computer model is getting a free pass from Anthony’s commenters as is the use of the “corrupt” CRUTem3 temperature data set from the vilified Dr. Phil Jones’ Climate Research Unit…
Wott's Up With That? 
Hi Ben,
Nice to see more traffic on your blog. Could you please elaborate more on “Roy has actually been plotting is temperature vs temperature, which obviously tracks itself very well…”
I smelled a rat when this came out, it honestly seems to me to be more fodder for the contrarians. His “findings” also contradict the results of Swanson et al. (2009, PNAS), who of course undertook a far more rigorous scientific approach.
Anyhow, if you could clarify that would be great. Thanks.
[I’m getting the “temperature vs temperature” conclusion from the comments by “Sphaerica” in the WUWT post. – Ben]
PDO and AMO are both detrended to take out any temp change, so the indices are completely divorced from any physical mechanism.
SOI is (roughly) defined as a pressure differential divided by the average of this differential. WTF this has to do with a mechanism for global warming is hard to see.
[Dr. Spencer has produced a classic example of abstract model fiddling, but because it has been done in the cause of denying man-made climate change it is embraced by the denialists. A revealing intellectual bankruptcy. – Ben]