CoyoteBlog has a great post on some rather dramatic global warming news from yesterday: the historical temperatures in the primary database used by climatologists have been revised downward, by a significant amount. I call that very dramatic news indeed, though my guess is that you won't see this on the evening news – most especially not the implications…
It turns out that the recent accusations of climatologist “amateur” Steve McIntyre were dead on target – the professional scientists had committed a grade-school error (a year 2000 bug, no less!) in their math. The error was hard to find for a very unscientific reason: GISS (the creators of the database) never released their methodology for peer review. Steve McIntyre did some clever reverse-engineering to make a case for their bug, and yesterday GISS admitted that he was absolutely correct.
Oops!
The reason this is a big deal is that this database is used as an input to all of the climate models that have been “proving” anthropogenic global warming. The revised historical temperatures are so much lower that they destroy one of the central premises of anthropogenic global warming: that mankind has caused global temperatures to increase through the 1900s, with the hottest years of the century coming right at the end. The “fixed” (still highly suspect in my book) shows the hottest year on record was 1934 (not 1998, as previously claimed).
Oops!
