Wednesday, August 29, 2007

2 states of separation

The Wall Street's Journal's lead editorial (subs. req'd) --

The latest twist in the global warming saga is the revision in data at NASA's Goddard Institute for Space Studies, indicating that the warmest year on record for the U.S. was not 1998, but rather 1934 (by 0.02 of a degree Celsius).

Canadian and amateur climate researcher Stephen McIntyre discovered that NASA made a technical error in standardizing the weather air temperature data post-2000. These temperature mistakes were only for the U.S.; their net effect was to lower the average temperature reading from 2000-2006 by 0.15C.

You could read this and the rest of the editorial and not be told that the adjustment referred not to "U.S." but to the 48 contiguous states -- which matters because Alaska is, like, large, and some global warming implications relate to the size of the surface area affected by it. And as the blogger who found the mistake that led to adjustment acknowledges, rankings of individual year as hottest, which are affected by random variation, are less meaningful than temperature trends, which are clearly upwards.

For more and more people, believing this obscurantist stuff from the Journal is going to require ignoring what's going on outside their windows.

No comments: