We have had a lot to say over the years about #ECS, noting especially the way the IPCC is determined to keep pushing it up above 3° C even though there is a lot of empirical evidence showing it is more likely 2° C or less. In 2023 a study by Andrew Pauling of the University of Washington and his coauthors threw shade on one of the arguments for high ECS presented in the IPCC’s 6^th Assessment Report. The IPCC claimed ECS is likely between 2.5° C and 4.0° C, an estimate which was (in the words of Pauling et al.) “in large part achieved by consideration of emergent constraints”. Huh? Well, the idea is that instead of measuring ECS directly, you measure something else in model outputs that might have nothing to do with temperature but which appear to correlate with CO2 “forcing”, the amount of warming CO2 is causing, like (for example) difference in precipitation between two regions. Then you look at what models say is going on with that other phenomenon and use it to say what ECS likely is. Yes it’s lame and contrived. But once the idea was out there, and given a fancy name, the climate science community ran with it, proposing that CO2 forcing correlates with lots of things, including the cooling effect of volcanic eruptions. And since models identify a strong cooling effect of volcanic eruptions, they must be offsetting a lot of warming from CO2 so ECS must be high too. Pauling and his coauthors then checked the argument in the context of the Mount Pinatubo eruption of 1992 and found a wee problem: across hundreds of model runs there is no correlation between ECS and the strength of the climate response to Pinatubo. Which blew the theory right up into the stratosphere.

We are not the right people to learn the sublime science of emergent constraints from since we can’t get past the feeling that it is opportunistic and unserious. You should learn the concept from people who believe in it, although even among climate scientists themselves the whole enterprise of deriving ECS using emergent constraints has not held up well over time, which is a topic for another day. For now we will focus on the idea of measuring ECS by looking at how climate models track the post-Pinatubo cooling.

The 1992 Mount Pinatubo eruption was a large climatic event, briefly. Massive amounts of volcanic aerosols were injected into the stratosphere near the equator and from there spread north and south causing a shading and cooling of the global climate that lasted for more than a year. So it provided a perfect historical laboratory for climate models since they should be able to replicate the temperature changes given the observed aerosol data and the way in which that cooling effect should temporarily diminish CO2-driven warming.

A 2010 study had used 9 models and 23 model runs to look at whether the climatic response to volcanic forcing correlated with ECS, and they concluded it did. They then used data on the observed climatic responses to Pinatubo to argue, based on the emergent constraint theory, that the real world ECS is likely between 1.7° C and 4.1° C, which the IPCC cited in its decision to narrow the likely range to 2.5° C or higher.

Pauling and his colleagues took note of the fact that since 2010 many more models have been developed and 543 model runs have been published which means there is now a much better basis for this kind of analysis. On the larger data set they determined that there was no correlation at all between the model response to Pinatubo and the model’s ECS value. Hence the reasoning used by the IPCC was flawed and that part of the case for high ECS burned up and fell to the ground like ash from a volcano, taking a big chunk of the case for using emergent constraints with it.