I’m on the train to Cambridge for a meeting that starts tomorrow. Since I had nothing better to do, I thought I would have a quick look at the RCP8.5 data. There’s been quite a lot of criticism of the National Climate Assessment, that I discussed in this post. This is mostly because it used RCP8.5, which some claim is unrealistic. Firstly, it didn’t only consider RCP8.5, and the point about RCP8.5 is not that it is somehow likely, but that it is plausible. We want to consider a range of pathways, including those associated with increasing emissions.
Something that is often over-looked is that RCP8.5 is really a concentration/forcing pathway, not an emission pathway. There are a range of emission pathways that could lead to such a concentration pathway, including some with cumulative emissions as low as 1500 GtC [Update: based on this paper, this is the cumulative emissions from 2005. If you consider the historical emissions (up to 2005) then the lower limit (1 sigma) is about 1750 GtC, rather than 1500 GtC.]. What I wanted to do was quickly check how cumulative emissions by 2100 varied with the year in which emission reductions start, assuming that they then reduce at 3% per year.
Essentially, we could largely avoid RCP8.5 if we started reducing emissions, at 3% per year, before ~2040. Bear in mind, though, that RCP6 is associated with cumulative emissions of about 1500 GtC, so avoiding RCP8.5 doesn’t mean that there wouldn’t still be a reasonably large change in radiative forcing. After 2040, we would need to reduce emissions at a rate greater than 3% per year if we wanted to have a high chance of avoiding an RCP8.5 pathway.
So, if people really think that RCP8.5 is impossible, that would seem to suggest that they think we can reduce emissions rapidly sometime in the future, or that we’ll start doing so soon anyway. If RCP8.5 is implausible because of fossil fuel availability, or future cost, it would still seem to make sense to think about alternatives sooner, rather than later. The longer we wait, the more rapid the transition will need to be, irrespective of the reason for doing so.