Rising Temperatures Due to Cooking?

Writing at the Telegraph, Christopher Booker notes yet another example of cooked-up global warming data:

[Blogger Steven] Goddard shows how, in recent years, NOAA's US Historical Climatology Network (USHCN) has been "adjusting" its record by replacing real temperatures with data "fabricated" by computer models. The effect of this has been to downgrade earlier temperatures and to exaggerate those from recent decades, to give the impression that the Earth has been warming up much more than is justified by the actual data. In several posts headed "Data tampering at USHCN/GISS" [I found only this, in a cursory search. --ed], Goddard compares the currently published temperature graphs with those based only on temperatures measured at the time. These show that the US has actually been cooling since the Thirties, the hottest decade on record; whereas the latest graph, nearly half of it based on "fabricated" data, shows it to have been warming at a rate equivalent to more than 3 degrees centigrade per century.

This finding, interesting in itself, reminds me of an item I encountered a few years ago in Ian Plimer's Heaven and Earth (which I gave a mixed review) to the effect that an apparent warming trend in surface temperatures might merely be an artefact caused by development near once-isolated temperature measuring stations. I am sure that the adjustments unearthed by Goddard would be justified (or excused) as corrections, but it would seem that, if anything, the adjustment should have been made in the opposite direction. That said, it is possible that the data Goddard and I are discussing are from different sources: I am writing based on a skim of his piece and my memory of a lengthy and turgid volume I read over four years ago.

Booker concludes his piece by reiterating a past speculation of his:

Any theory needing to rely so consistently on fudging the evidence, I concluded, must be looked on not as science at all, but as simply a rather alarming case study in the aberrations of group psychology.

That's charitable, given that there is a clear incentive to massage the data, in the form of acquiring loot and prestige from the government.

-- CAV