---

ClimateGate: The end of credibility and the need for process control
...First, the slope of warming diminishes considerably if you exclude the adjustments. This was true in New Zealand, and it is true in lots of other local data around the world. As I understand it, a large part of the case for historical warming lies in the adjustments.

Second, it did not take the CRU document leak to know that many if not most of the scientists performing these adjustments are not only interested in determined what is, but in influencing what ought. They are not only describing a version of reality, as scientists are supposed to do, but using their description to advance a policy agenda. NASA's James Hanson is the leading exemplar of this phenomenon, but there are obviously many others. This conflation of science with policy advocacy makes it especially important that all matters of judgment, including especially the adjustment of raw data, be completely transparent.

Third, for reasons legitimate and otherwise, the data adjustments have not been transparent to scientists, journalists, and interested citizens outside the climate specialist community. The Kiwis have not released their adjustments, and actually refuse to do so in the linked article. In at least one case, the raw data has been lost or destroyed, so there is no way to examine the changes that were made. Steve McIntyre, the increasingly expert blogger who runs Climate Audit (and who earned mention in today's Wall Street Journal editorial), is repeatedly stonewalled in his efforts to get the same data (which has almost always been created using public funds) that the insiders have used to produce the "official" record.

Fourth, it is increasingly clear (especially from the computer code that drives the climate models) that neither the science behind what has been (studying the historical climate to see what happened) nor the construction of the models that purport to tell what will happen have been subject to the internal controls and quality systems that we require of, say, medical technology companies that produce and validate software that will be used to diagnose or treat an individual human being. It seems to me that we the voters ought to require the detailed validation -- and there are "internationally accepted" methods for doing this, to coin a phrase -- of any computer model cited as justification for any policy intervention. After all, lives hang in the balance....