Speaking of things “hard to prove”, not to mention opportunities for self-delusion … suppose a GHG emissions control policy were implemented, with the goal of reducing global temperatures. How would you test the policy’s effectiveness, given that you have a treatment scenario and no “control” scenario? Can we agree that the GCMs are sufficiently uncertain that a definitive test is not possible?

Thinking in terms of practical politics, the emissions of GHG would be what the regulating bodies would attempt to control while temperature/climate effects (and even GHG levels in the atmosphere) would become secondary and tertiary until such times as the restrictions on emissions became an adverse issue with the voting constituents. Once the pain was felt, the voting public would demand to know the relationship between the controls and GHG levels and in turn the GHG levels and climate/temperatures. That would be time that skeptics arguments would/could come to the fore. All these reactions would be subject to some modifications based on what the “natural” trends were in the climate, and particularly, in the frequency of extreme climate events. Shear momentum of the regulation and inflexibility of government programs, in general, would tend to keep otherwise failing policy in place while the amount of pain felt by the public would have a reverse effect.

I would judge that without convincing evidence of, at least, the semi-quantitative effects of GHG on climate, the outcomes of policy would be much the same regardless of the real GHG effects on climate. Other developments, such as higher fossil fuel prices bringing market forces to bear on the use alternative energy sources, could override the regulation considerations.

When I projected the temperature based on summing the forcings of CO2, methane and nitrous oxide, the trend line went right through the predictions of all the best GCMs, when calculated under the same boundary assumption of 1% CO2 increase per year. To make that comparison, I had digitized the plotted outputs of 10 of the 15 GCM projections given in the LLNL intercomparison project. I have that comparison on my computer at work. I have to run up there this evening anyway. While there, I’ll send you a gif file of it.

Pat, you mean they respond linearly to increased ghgs and not logarithmically?

I’ve been doing a little modeling of my own, and have found that the state-of-the-art GCMs produce no more than linear responses to pure GHG warming, plus wiggles. They appear to operate as though there were no climate drivers other than atmospheric gasses.

What, then, would be the ultimate fate of such a program? And of the tone of debate surrounding the fate of the program?

If you thought climsci has become politicized, you ain’t seen nuthin’ yet. If emissions control is a “moral question”, then what Al Gore is proposing is that we treat the climate debate much as we treat the debate over abortion or capital punishment. Doesn’t that sound fun.

I agree. Having hung around my share of academics, self-delusion is almost as well practiced as
it is in Jonestown tent revivals. That said, I was mainly interested in the data landscape. If, for example,
only 20% of available ice cores are carefully picked, the whole enterprise seems absurd. However, if one
randomly samples ice cores from a population, then that is a different matter.

Thanks for the suggestions…I will read.

I like my answer because it’s at least falsifiable. I think that intent in a lot of these cases is impossible to prove or disprove, where simple self-delusion is the most parsimonious conclusion from the data.