This story about climate engineering reminded me:
I strongly support climate engineering if properly analyzed, but I think that proper analysis is unlikely to be possible with most approaches. I strongly favor the development of space mirrors, because they are one of the few techniques that are both highly controllable and they also swing both ways. If it turns out some intervention is not working as we expect, we can actually stop intervening. If the global climate doesn't work the way we expect and suddenly we need to start warming the globe, we can do that by changing the orbit and reflecting sunlight that would have otherwise have shot out into space back onto Earth. Space mirrors have a flexibility and precision almost all other techniques lack, and give us the opportunity to learn how the climate works with direct experimentation and rough engineering consensus.