This Chapter of the DOE Report points out, uncomfortably for the “settled science” crowd, that estimates of the warming response of the atmosphere to increasing CO2, at the very core of the climate issue, are still all over the place even after 40 years of research and billions of dollars spent. It is broadly agreed that all human emissions from the 1700s to the end of this century will roughly double the CO2 level in the atmosphere. It is also broadly agreed that every doubling of the relative amount of CO2 in the atmosphere will produce a constant temperature increase (known in the trade as Equilibrium Climate Sensitivity or ECS), so going from say 400 parts per million to 800 only warms the planet by the same absolute number of degrees as going from 200 to 400 ppm did. And it is also broadly agreed, though not so widely recognized, that the estimated direct warming effect of doubling atmospheric CO2 is slightly more than 1 degree C, which is pretty trivial. The idea of a climate crisis hinges on supposed secondary warming effects triggered by the initial warming so that ECS is higher than 1 C, possibly a lot higher. And estimates of ECS have historically ranged from 1.5 C to 4.5 C with a best estimate of 3.0C, though some models now say it could be as big as 5.7C. But this hypothetical extra warming arises from hypothetical feedbacks, and climate science still can’t say whether they are huge, tiny or somewhere in between. To its credit the IPCC has now turned away from models and is looking at historical data for the answer. But that data presents uncertainties of its own. The science here is anything but settled.
The current crop of climate models have ECS values ranging from 1.8C to 5.7C. And this wide range of uncertainty, more than a factor of three, indicates even less certainty than it might seem to because in comparison to previous generations of models, the new ones generate a wider range of possibilities not a narrower one. Modelers don’t program in a specific ECS value. At least not in principle, though their expectations certainly affect how they fine-tune models after the initial runs. What they do is program in different estimates of atmospheric processes which imply different feedbacks and amplifying mechanisms and cumulatively determine the resulting ECS. If the result seems absurd to them they tweak the parameters until they get something they were expecting. But even so the big range of results means that modelers can’t agree on how the feedbacks work. And most of those feedbacks are crude approximations to begin with, not careful replications of things observed in nature, because many of the processes involve things like cloud formation which happen on too small a scale for the models to represent correctly no matter how much input data one gathers.
Given this chaos, the last IPCC report actually didn’t even bother trying to determine ECS based on the models. Instead it turned to a shiny new study in the literature that combined modern thermometer data with paleoclimate records to yield a statistical estimate of ECS that narrowed the likely range to 2.6 C to 3.9 C, with a central best estimate of 3.1 C. So a narrower range with a much higher lower bound… exactly what they were hoping for and they just happened to pick that paper. Which isn’t really how science is done.
Alas, it gets worse. Shortly after the IPCC report came out, a paper by Nic Lewis (which we reported on here) corrected some data and math errors in the earlier paper and yielded a likely range of 1.8C to 2.7C, with an ECS best estimate of 2.2C.
It gets even worse. Another move the IPCC made was to argue that ECS will get larger in the future because warming will be concentrated in a part of the tropical Pacific where the climate system is relatively inefficient at expelling heat to space. But, the DOE team noted, while models say this pattern should already be happening, the data say the opposite, namely the warming is concentrated in another part of the Pacific that is especially efficient at expelling heat, so if that continues ECS may instead decline in the future.
The bottom line: anyone who says the science is settled has no idea what they’re talking about.
There is no good reason to assume that climate feedbacks are positive rather than negative. In fact, given that the climate has remained roughly stable for hundreds, or even thousands or millions of years, depending on how one defines stable, the likelihood of dominant negative feedbacks seems more likely.