Did you notice that RC called it “Meinshausen et al”? Barbarians :-). Anyway, they liked the paper whilst I’m less sure. As far as I can tell its not really a question of science in dispute, just what you make of it. So what M et al. do is instead of the std.ipcc “force a GCM with CO2 and see how climate changes” they try to reverse this process, and see what level of CO2 produces a given temperature change. They can’t do this with GCMs, of course, so are fitting the GCM stuff to a simpler model. And since there is uncertainty in the sensitivity, there is uncertainty in the result, but at heart it all seems simple enough and scientifically rather dull, though politically interesting. M et al. pretty well admit that they have spotted a gap in the literature and headed straight for it: there is still an important gap in the literature relating emission budgets for lower emission profiles to the probability of exceeding maximal warming levels; a gap that this study intends to fill (isn’t there an old joke about that? Can’t find it now). The other thing they find is the rather convenient result that warming depends on total CO2 emissions, rather than the CO2 trajectory. Given plausible trajectories and CO2 lifetime and climate response times this too is not exactly surprising; but its good to have it written down.
M et al. skip rather lightly over the 2 oC threshold. 2 oC is obviously somewhat arbitrary; all they say to justify it is “We focus here on 2 oC relative to pre-industrial levels, as such a warming limit has gained increasing prominence in science and policy circles as a goal to prevent dangerous climate change [25]” and 25 is Schellnhuber in Avoiding Dangerous Climate Change. I’ve commented on this before (and got a lot of comments in reply, but didn’t change my mind). RC seem to think that 2 oC is arbitrary, but on the high side: We feel compelled to note that even a “moderate” warming of 2°C stands a strong chance of provoking drought and storm responses that could challenge civilized society. My feeling is that when this stuff starts to cost real money rather than policitcal promises, and people start saying “OK, limit CO2 to keep us below 2 oC, remind me again just exactly why 2 oC is so dangerous” they are not going to find the answers very convincing. There is risk, of course, but there is risk in every course of action.
More interestingly (to me) is a comment in the Nature intro saying “Yet only a third of economically recoverable oil, gas and coal reserves can be burned before 2100 if that 2°C warming is to be avoided”, ref’ing this paper. I’ve started to wonder recently if we really do have the reserves to get much beyond 2xCO2, so picked up on this – perhaps they will provide some good sources. But what M et al. say is: We show that, for the chosen class of emission scenarios, both cumulative emissions up to 2050 and emission levels in 2050 are robust indicators of the probability that twenty-first century warming will not exceed 2 oC relative to pre-industrial temperatures. Limiting cumulative CO2 emissions over 2000-50 to 1,000 Gt CO2 yields a 25% probability of warming exceeding 2 oC–and a limit of 1,440 Gt CO2 yields a
50% probability–given a representative estimate of the distribution of climate system properties. As known 2000-06 CO2 emissions3 were 234 Gt CO2, less than half the proven economically recoverable oil, gas and coal reserves can still be emitted up to 2050 to achieve such a goal. So is the half (or a third, where ever that comes from) emissions to 2050, or to 2100? If it emissions to 2050, then we’re going to need to emit *all* of the economically viable reserves to keep on the “bad” trajectory to 2100.
M et al. (fig 3) says that total econ-viable reserves of CO2 are ~2500 Gt. It also has as x-axis the cumulative emissions 2000-2049, and that axis ends at 2500. A1F1 on that graph is past 2500, which presumably means that to get A1F1 to 2100 we have to burn *more* than the total econ-viable reserves. That seems rather unlikely. Indeed, from that graph, it would appear that most of the std IPCC/SRES scenarios are effectively impossible. It isn’t obvious to me what is wrong with my reasonning; but if it is correct I would have expected M et al. to point out that many of the SRES trajectories are impossible, and they don’t. Up to now I’ve been defending SRES (on wiki etc.) against those who say the trajectories just won’t work; but I’m now rather less sure.
Minor point: their figure 2(e) appears slightly dodgy. They show scary-looking temperature changes of 4-6 oC with central value of 5 oC to 2100 from A1F1, and that is only the inner range. It doesn’t look totally compatible with AR4 SPM.5. Partly that is because they put their zero at 1860 (which is quite defensible) and partly because they have picked the high-end A1F1… but from SPM.5 it looks to me as though A1F1 should be ~4.6 above 1900, not +5.
FWIW, I think we’re going to find out what +2 oC and above is like, because I don’t think we’ve got the will to avoid it. Studies like this one are probably politically useful less because of the new science they contain than because they push the aggregate weight of discourse in a certain direction, and give those who want to push for limits some useful ammo. But the response to the credit crunch makes it rather clear, as mt has despairingly commented, that our priority number one is a growing economy. Only once we have that will we consider other low priority matters, like, err, not growing the economy :-).
[TL points to a nice oildrum post summarising some recent literature. Which indicates that 450-600 ppm CO2 is a likely limit. It also says what I’ve been thinking – that very few people are studying this, compared to those doing other aspects of cl ch -W]
wmconnolley: scienceblogs.com/stoat archive 
