Climate discounting: How do you value one tonne of CO2eq averted today versus one tonne of CO2eq averted (say) 10 or 30 or 50 years from now?

Some (potentially overlapping) factors which should be considered:

- what does warming effect as a function of CO2eq concentration look like? (presumably it's not linear, i.e. presumably the warming effect of another tonne when CO2 is 412 ppm is not the same as the marginal effect when CO2 is 500ppm or 1000pm)
- what does "badness" as a function of warming look like? (again, presumably not linear, i.e. I suspect that 10 degrees of warming is probably more than 5x worse than 2 degrees of warming)
- are there vicious circle/runaway effects?
- are there cliff-edge effects?
- how much uncertainty is there in the models? (I understand the answer to be "lots", in which case should that make us more risk averse, and therefore more prone to place greater value on immediate results)

Note: this question was also cross-posted to the Facebook effective environmentalism group https://www.facebook.com/groups/effectiveenvironmentalism/permalink/2306369839662729/