I have a long post up at MasterResource, revisiting the issues that came up in my earlier response to Nordhaus. If you are interested in climate science, particularly the economics of climate change, I would immodestly recommend that you wait till you have a good 15 minutes and read this thing through. It’s by no means a “light” piece but I think there are some important issues that I cover.
For the economists reading, let me reproduce a large portion of one of my arguments:
In a standard economic regression analysis, we typically approach things the way one is taught in high school when learning basic statistics. Namely, you set up a null hypothesis that is the opposite of the causal relationship you (the researcher) actually think exists. Then, if there is an apparent relationship in the data (such that you get a positive value on the coefficient for a certain term in a least-squares regression, say) you can see if the result holds up at a 90 percent, 95 percent, or 99 percent confidence interval.In this normal context, the higher the confidence interval, it means the more confident you are that the apparent relationship between two measured variables isn’t spurious….Yet in charts of climate model projections, the “confidence interval” works the other way around. Here, the higher the number, the less confident we can be that an apparent match between the model and nature is due to the underlying accuracy of the model. To put it in other words, here the null hypothesis is that “this suite of climate models is accurately simulating global temperature.” Thus if we make it harder to reject the null (by ramping up the confidence level), then it gives more wiggle room for the models.Specifically, the “95% range” in the second graph above comes from looking at all of the observed “runs” of the suite of climate models, and then plotting the gray boundary that captures the realizations of 95% of the runs centered around the average. Ironically then, the less agreement there is between the individual climate models, then the wider the gray zone would be, and the harder it would be for Nature to “falsify” the suite of climate models.…Suppose for the sake of argument that one particular model accounted for 3% of the total simulated runs, and it predicted global temperature anomalies of 20 degrees Celsius from the year 2010 forward, while another particular model accounted for a different 3% of the total simulated runs, and it predicted global temperatures of minus 20 degrees C from 2010 forward.In this (absurd) situation, the RealClimate post would show a massive gray zone covering the “95% confidence interval,” and (barring an asteroid collision or a massive change in the sun), it would be inconceivable that temperature observations would fall outside of this range. Yet that would hardly shower confidence on the suite of models.