Emissions are something we have (relatively) direct control over. Realize that these models do take into account cascading effects like triggering ocean methane

Emissions are something we have (relatively) direct control over. Realize that these models do take into account cascading effects like triggering ocean methane releases. They do take into account nonlinear effects of different GHGs. They pick possible emissions scenarios, simulate as many of these effects as they can, and look at the spectrum of possible future results. Then we look at our emissions to try to guess which simulation results our future is most likely to resemble. It's not quite as tunnel-visioned as you make it out to be. > Until we are actually removing gasses from the atmosphere permanently, the planet is fucked. I happen to agree with this, but realize that's exactly the same metric: it's just negative GHG emissions. We would do the same thing: simulate potential future outcomes given an emission / capture balance that is overall negative, estimate what might happen, and then measure what we're actually doing to see whether we might actually be on that path or not. Once we start removing gasses, it's still useful to track overall emission balance.