I have finally got around to carrying out my own calculation of how much of an increase in the near-ground global average temperature (GAT) arises from direct CO2 forcing. I emphasise that this excludes feedback effects.
The full gory details are here.
If you have a technical background you should be able to follow the whole calculation, though it is not trivial. My method includes a number of simplifying fudges.
The most important of these fudges is an assumption for how the change in global average temperature varies with altitude. I had not appreciated at the outset how sensitive the near-ground GAT would be to its variation with height. Nor had I appreciated initially that the stratosphere and the mesosphere have decreased in temperature over recent decades. It turns out that the more these high-altitude layers decrease in temperature, the more the troposphere needs to increase in temperature in order to maintain energy balance.
As a result, depending upon what model I assume for the variation with altitude, the near-ground GAT increase due to a 50% increase in CO2 (i.e., from pre-industrial to 2023) ranges from 0.29 degC to 0.73 degC (based on my favoured method). The higher end of this range appears more likely as that corresponds to decreases in stratospheric / mesospheric temperatures. A broader range of results averages to 0.57 degC.
Popular or news accounts of the IPCC models are guilty of obscuring the fact that just less than half of the currently observed 1.1 to 1.2oC increase in GAT (since pre-industrial times) is, in those models, due to the direct effect of CO2 increases – namely 0.54 degC. Whilst the large spread in my predictions is unfortunate, I tentatively conclude there is reasonable agreement with the IPCC models as regards the direct effect of CO2.
The rest of the observed 1.1 to 1.2oC increase in GAT is claimed to be due to feedback effects. Whilst the occurrence of feedback effects is a very reasonable expectation, quantifying feedback effects is another matter. It’s just too difficult. Arguably it is difficult even to predict, from basic physics, whether positive or negative feedbacks would win overall. The approach taken by IPCC modellers is to include tuneable parameters to account for feedback. Generally the dominant of these is water vapour.
Water vapour will increase due to rising temperature, and hence, being a greenhouse gas, will contribute to a positive feedback. On the other hand, rising water vapour levels would – to my naïve mind – tend to increase cloud cover. The Earth’s albedo is very sensitive to cloud cover, and greater cloud cover will reflect more incoming radiation back into space. So I would expect this to provide a negative feedback. Oddly, the IPCC models appear to attribute positive feedback to cloud cover. I make no definitive statement about this. Indeed, my position is exactly that a definitive statement about the magnitude, and even the sign, of total feedback effects is extremely problematic…which is why IPCC modellers have recourse to fitting tuneable parameters instead.
That’s fair enough. What isn’t fair enough is to then claim that “the science is settled” (a claim which is intrinsically unscientific). Nor is it valid, having had recourse to fitting against uncertain physics, to place too much reliance on the future predictions of models whose extrapolation to greater CO2 concentrations is of uncertain foundation.
The delicacy of these calculations is illustrated by how little the energy balance would be upset due to a 50% increase in CO2 (if temperatures did not change to compensate). At most this is 1%, and a lot less according to my favoured models (maybe 0.35%). This isn’t surprising because 1.2oC is only ~0.4% of the near-ground absolute temperature.
These issues are why scientifically well-informed persons explore other avenues in addition to the politically obligatory “CO2 is everything”. In Section 27 of my calculation I illustrate some of these issues, specifically the variability in the insolation (the amount of energy reaching the Earth from the Sun). I make no definite claims; I only point out that there is significant doubt about the causes of the whole of the observed GAT increase, and also significant doubt about predictions.