Climate scientist Roger Pielke Sr. took the Washington Post's Juliet Eilperin to task a few days ago for an Oct. 13 story suggesting "that 2005 is on track to be the hottest year on record."

Pielke's contention is that, while surface temperature data may support the claim, the surface thermometer record "is not ... the most appropriate metric to evaluate global warming." And from a climatological view, of course, he is absolutely correct. Various layers of the atmosphere behave quite differently, and a complete picture requires looking at all of them, which, as Pielke notes, paint a much more complex picture of warming than a simple "warmest ever" headline.

Pielke also gives an important nod to regional anomalies, which he rightly argues "are what we should be focusing on in terms of long-term climate trends, rather than surface globally-averaged temperature trends."

But this is a newspaper story for a national audience, where you really get to say one clear thing that has to communicate with people everywhere. If it were for a regional audience (like, say, mine) it could focus on a regional anomaly. But whoever is going to be reading this is living on the planet's surface, so that's the temperature number that's going to matter to them.