Here we go again with the “settled science” doubletalk.

The science of AGW is settled . . . unless, that is, you suggest that the billions of dollars in taxpayer subsidies for climate-related research might be used in more productive ways (say, by letting taxpayers keep it). Then the story becomes: no…no, we desperately need billions more to look into the grave uncertainties.

Climate models that predict environmental catastrophe decades hence are supposedly so accurate that we must start starving ourselves of energy on their say-so — as in theClimate Change Science Program’s ”Unified Synthesis Project” on which I commented yesterday. But now we learn that the state of computer modeling is so in its infancy that billions more are necessary to make them suitable for the purposes to which they are already being put.

Reuters huffs “Scientists urge US to protect economy from climate” (which recalls IowaHawk’s climate modeler hinting that more grants might appease the weather gods). This turns out to mean insulate the climate modeler economy from their own shrieks of “settled science!” — keeping them comfortably employed and expensed while they burn billions of dollars developing models that fail every test to which they are put. “The investments would pay for satellite and ground-based instruments that observe the Earth’s climate and for computers to help make weather predictions more accurate.” Specifically, they want to make regional projections. The kind with which the CCSP USP is up to its alarmist ears.

Even Climate Wire, slavishly apologetic for alarmists, styles this as “a sharp increase in federal climate funding, roughly $9 billion above the $10 billion five-year budget projection for the multi-agency Climate Change Science Program.” Yes, that CCSP. Gack.

More surprising still is the moment of sobriety from über-alarmist Gavin Schmidt – who wastes taxpayer money by posting for hours on end on the RealClimate project of Environmental Media Services (read: the Tides Center and Fenton Communications of Alar and Mother Sheehan fame, among other campaigns). But on the squawking for new climate-modeling rent, Schmidt admits:

“When it comes to decadal predictions that explain anything more than a tiny fraction of the variability in anything more than a tiny fraction of the land area, I have yet to be shown any result that says that this is possible, even theoretically” [After the NYT posted this comment, Schmidt thought better of himself and rushed out an English-to-English translation, available at the same link].

Which brings me back to my comments for CEI on the CCSP Unified Synthesis Product: this thing does not withstand a moment of scrutiny under the Information Quality Act.