This is interesting. Lewis and Curry estimate equilibrium climate sensitivity (ECS) at 1.64°C. That's remarkably low. If it turns out to be true – and it seems like we won't know the true value for ECS for another decade or longer – then AGW would have been largely overstated. The estimates constantly change, which is frustrating.
I can't evaluate their methods. I'm curious to read any critiques by other climate scientists.
ECS is the long-term change in temperature from a doubling of atmospheric CO2. It's arguably the most important fact or estimate in the AGW issue. Ultimately, this is what anthropogenic warming is about. I think the baseline is the pre-industrial level, for which I've seen different estimates. I think it was 280 ppm. Now we're around 400 ppm. A doubling of the baseline would be 560 ppm. I have no idea when we're supposed to hit that.
One of the things that surprises me about AGW skeptics and lukewarmers is that they don't seem to worry about the future consequences to themselves if they're wrong. Most of them aren't elderly, so they'll likely be around for decades. If AGW turns out to be a serious problem, I expect skeptics to be pilloried. So it's ballsy to be a skeptic – I won't be surprised by violence against them in 2050 if warming becomes a problem.
It's also ballsy to be a climate scientist who offers low estimates of ECS. We haven't yet built a rational civilization when it comes to science and politics. An honest scientist who offers low estimate of ECS which turn out to be wrong will be savagely attacked. They'll be savagely attacked even before we discover whether they're right or wrong. So I think it's remarkably brave of them – being wrong on this issue is so risky, not just to themselves, but to the world. It would weigh heavily on me if I were a climate scientist. I'd have to quadruple-check the data, the analyses, and think very deeply about the validity of the methods and framework. I'd be terrified to offer estimates of ECS. It would feel like so much responsibility. At the high end, AGW could do serious harm to lots of people (well, they'd be able to see it coming decades in advance, so that might reduce the harm)
The fact that the estimates always change would give me pause, because it would imply that my estimates would change, or at least be replaced by others' future estimates using new and better methods, where the "future" could be next year. The epistemology or meta-epistemology of ECS looks very tough from the outside looking in – if the estimates are always changing, what does this mean about the methods used to generate them? Or the nature of our knowledge of climate circa 1990 - 2014? Thankfully, they've tended to go down, like from IPCC4 to IPCC5, or at least the low end of the range has gone down. It's odd. We don't keep revising Planck's Constant or our estimate of the acceleration of gravity on earth, though admittedly these aren't the best analogies. It's just tough to know what to do with ECS estimates, because we know they always change. That's the one thing we know about them. This implies that we should expect them to change in the future, until we know something that tells us to stop expecting that.