ClimateGate: Maybe Models Really Are Dumb

You may have seen the Washington Times’ lead story reporting that, when Obama’s Department of Agriculture computer model assessments of cap-and-trade’s impact revealed that it would encourage farmers to plant trees for carbon credits instead of food, the administration told the modelers to change the assumptions to get a different result. You see, this would drive the cost of food even higher still — a very regressive tax, hitting seniors and the poor first and worst — as would the other hits brought about by cap-and-trade energy rationing making your energy prices “necessarily skyrocket”, and the spike in fertilizer costs from massive fuel-switching from coal to the necessary feedstock natural gas. Big problem, this modeling of the bill’s impact.

So the administration rushed to tell us there were problems with the models, wrong assumptions, etc. The modelers, without the intervention of politicians, had gotten it wrong, you see. But now that the pols have told them what to do, why, all will be well.

This episode is reminiscent of the well-known problem with “climate” computer models, though the climate models are even worse. And the response to their deficiencies has been very different.

General circulation models are not useful anywhere there are large bodies of water or mountains nearby — which is most of the earth’s surface — and cannot adequately account for the sun, clouds or the oceans — major climate drivers all. Among other deficiencies.

And yet, when it comes to their computer models being so wrong that they tell us it has continued to warm even as we experience observed cooling, they tell us there must be something wrong with the observations — the old “me or yer old lyin’ eyes” argument — because the models must be respected (tampering with the assumptions would, here too, produce another result, and they can’t have that).

But of course, everyone knows that models predicting the weather are very reliable. Trillion-dollar-policy reliable, apparently.