On the other hand

One claim made by climate change naysayers involves the so-called "heat island effect." That's where temperature-measuring stations are affected by sources of nearby heat - such by being placed too close to an air conditioner or on an asphalt surface. The result, the deniers claim, is temperature readings that are too warm and thus unsuited to analyzing climate change.

Noted naysayer Anthony Watts has made something of a cottage industry of that, with volunteers taking photographs of poorly-situated stations and making up slide shows of the results.

I was never quite sure what their scientific point was in terms of dismissing global warming, both because it wasn't like no one knew about "heat islands" until these folks brought it up - it is, after all, why temperatures in city centers tend to be a bit higher than those in outlying areas - and even if the temperatures recorded were too high in absolute terms, if they're going up over time it's still evidence of warming.

But, um, it develops that there is a more fundamental problem with the enterprise, as reported by DeSmog Blog:

A recent peer-reviewed paper in the Journal of Geophysical Research looked at data from 114 weather stations from across the US over the last twenty years and compared measurements from locations that were well sited and those that weren’t.

They did find an overall bias, but it was towards cooling rather warming.

According to the authors,

“the bias is counter intuitive to photographic documentation of poor exposure because associated instrument changes have led to an artificial negative (“cool”) bias in maximum temperatures and only a slight positive (“warm”) bias in minimum temperatures.”

[Emphasis in original.]

That is, the "bias" was toward measuring minimum temperatures as slightly higher than they should have - but measuring maximum temperatures below where they should have, and to a greater extent. The net result was an average temperature slightly below where - slightly cooler than - it should have been.

This is latest in an expanding body of science that has looked at the urban heat island effect in excruciating detail and found nothing to undermine the observed and disturbing warming in the US over the last several decades,

DeSmog Blog concludes, citing studies that found, among other things, that "any urban-related trend is an order of magnitude smaller" - that is, about one-tenth as much - "than decadal and longer time-scale trends" and thus would be swamped by those longer-term trends.

But I doubt that will stop the naysayers from pushing the same line as they have. As someone noted in comments to the post,

Either way - the damage is done because optics count.

Which serves to drive home the point that, as I said just below, the naysayers are not engaged in science. PR, yes. "Optics," yes. Science, no.