The advantage of working at a university, is that you get the opportunity to attend interesting seminars, especially if Kit Carruthers reminds you (on which note, I also discovered that Kit submitted his PhD thesis last month). The talk today was by Nathan Gillett and discussed what would happen if we emitted, in total, 5000GtC. This would be a really large amout of CO2; a bit less than 10 times as much as we’ve emitted already and about twice as much as would be emitted along RCP8.5 by 2100. However, it is possible to recover and burn as much as 5000GtC, so it’s not impossible that we would do so. I suspect, however, that it is highly unlikely, one reason for which being that it would be extremely stupid to actually do so.
The bottom line from the talk was that if we were to emit 5000 GtC the world would warm by between 6.6 and 11oC, with the Arctic warming by between 15.3 and 19.7oC. However, as I said above, it seems highly unlikely that we would actually be stupid enough to emit anything close to 5000GtC. However, there were a couple of other aspects that I found interesting. Something I’ve discussed a few times is the idea that how much we warm probably depends, largely, on our cumulative emissions. Essentially the transient climate response to cumulative carbon emissions (TCRE) is thought to be between 0.8oC and 2.5oC per 1000GtC. However, this is thought to only hold till around 2000GtC, at which point the warming was thought to decreases slightly. This more recent work suggests, however, that it continues linearly out to 5000GtC.
In a sense this linear dependence on cumulative emissions is surprising, given that warming depends logarthimically on atmospheric CO2 concentration. The reason, however, is largely because as we continue to emit more and more CO2, the airborne fraction is expected to increase (as the efficacy of the natural sinks decreases) and this compensates for the logarithmic dependence. Recently, however, it was suggested that the sensitivity to cumulative emissions was not only not linear, but was also lower than the other models suggested.
This appeared to be largely because the model assumptions lead to the airborne fraction remaining roughly constant and that, hence, the logarithmic depdence dominated (and, also, using a model with a low climate sensitivity). However, from the talk today it seems that all major models suggest that the airborne fraction should increase as we continue to emit CO2. Furthermore, these models – apparently – don’t include all possible carbon cycle feedbacks (such as permafrost release).
Anyway, that’s all I was going to say. It was an interesting talk and seemed to further strengthen the idea that warming is likely linear with respect to cumulative emissions. Essentially, if we have some warming target, and we want to give ourselves some chance of staying below that without requiring negative emissions, then we have a carbon budget. For our current targets (staying well below 2oC), we’ll use up the carbon budget within a couple of decades.