Skip to content

There is word circulating that a paper soon to appear in Science magazine concludes that the climate sensitivity—how much the earth’s average temperature will rise as a result of a doubling of the atmospheric concentration of carbon dioxide—likely (that is, with a 66% probability) lies in the range 1.7°C to 2.6°C, with a median value of 2.3°C. This is a sizeable contraction and reduction from the estimates of the climate sensitivity given by the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4), in which the likely range is given as 2.0°C to 4.5°C, with a best estimate of 3.0°C.

Further, the results from the new analysis largely eliminate the “fat tail” of the distribution of possible values of the climate sensitivity (that the IPCC AR4 report was fond of) which included the possibility that very large climate sensitivities are a realistic possibility. In the new paper, the authors find only “vanishing probabilities” for a climate sensitivity value greater than 3.2°C and that values greater than 6.0°C are “implausible.” Contrast that with the IPCC assessment of the literature (summarized in our Figure 1) which routinely includes studies concluding there is a greater than a 10% possibility that the true climate sensitivity exceeds 6°C and some which find that there is a greater than 5% possibility that it exceeds 10°C.

Figure 1. Climate sensitivity distributions retained (and in some cases recast) by the IPCC from their assessment of the literature. Note the “fat tail” towards the right which indicates the possibilities of the climate sensitivity having a very large positive value (that is, a huge degree of global temperature rise for a doubling of the atmospheric carbon dioxide concentration) (source: IPCC AR4).

The new paper, from a team of researchers led by Andreas Schmittner of Oregon State University, throws cold water on the IPCC’s tails. Here is its rather provocative abstract:

Assessing impacts of future anthropogenic carbon emissions is currently impeded by uncertainties in our knowledge of equilibrium climate sensitivity to atmospheric carbon dioxide doubling. Previous studies suggest 3 K as best estimate, 2–4.5 K as the 66% probability range, and non-zero probabilities for much higher values, the latter implying a small but significant chance of high-impact climate changes that would be difficult to avoid. Here, combining extensive sea and land surface temperature reconstructions from the Last Glacial Maximum with climate model simulations we estimate a lower median (2.3 K) and reduced uncertainty (1.7–2.6 K 66% probability). Assuming paleoclimatic constraints apply to the future as predicted by our model, these results imply lower probability of imminent extreme climatic change than previously thought.

Figure 2 shows the distribution of the range of the earth’s probable climate sensitivity as determined by Schmittner et al. Note the rapid drop-off in the probability that the climate sensitivity is much greater than 3°C (the IPCC “best estimate” for the sensitivity), and that the distribution falls off less slowly towards the left (towards lower sensitivity) than towards the right (higher sensitivities). The “fat right-hand tail” of the distribution is gone and the possibility that the climate sensitivity is in the 1°C to 2°C range is not minimal.

Figure 2. Distribution of the land/ocean climate sensitivity as determined by Schmittner et al. (adapted from Schmittner et al., 2011).

The Schmittner et al. results join a growing number of papers published in recent years which, by employing investigations of the earth’s paleoclimate behavior (that is, how the earth’s temperature changes in the past when subject to changing climate forcings) have come to somewhat similar conclusions, especially regarding the (lack of) evidence to support the existence of the fat right-hand tail.

For example, researchers James Annan and Julia Hargreaves published a paper in 2009 that concluded many of the assumptions underlying the possibilities of very high climate sensitivities were unjustified. They wrote:

When instead reasonable assumptions are made, much greater confidence in a moderate value for [the climate sensitivity] is easily justified, with an upper 95% probability limit for [the sensitivity] easily shown to lie close to 4°C, and certainly well below 6°C. These results also impact strongly on projected economic losses due to climate change.

Annan made repeated comments during the IPCC AR4 review process that the IPCC’s handling of climate sensitivity and its probability distributions were incorrect. His complaints largely fell upon deaf ears.

However, as there are appearing more and more examples in the literature, of which Schmittner et al. is one of them, making a convincing case that the very high climate sensitivities are not defendable, there will be growing pressure on the IPCC in its Fifth Assessment Report to greatly shrink the fat tail of the probability distribution for the true climate sensitivity. However, the climate “realists” very bad experience with the last IPCC process makes them chary. James Annan, writing at his blog in reference to the new Schmittner et al. paper had this to say as to what may result from it:

That said, [the Schmittner et al. paper] is a useful antidote to the exaggerated uncertainty estimates that have been prevalent over recent years, and I certainly applaud the intentions and effort underlying this substantial piece of work. In any case, I expect the merchants of doubt to do their worst on it when they cite it in the IPCC report.

But, as the evidence mounts against a high value for the climate sensitivity, and evidence grows for a low value (recall that the observed rate of global warming for the past several decades has fallen well below IPCC best estimates), the IPCC is going to be hard-pressed to retain the status quo in its Fifth Assessment Report, especially in light of the enhanced scrutiny that its AR4 misdeeds brought upon the process.

But, as James alludes to, perhaps we ought not be holding our breath.

And, for those keeping score out there, about 10 years ago, a couple of us here at WCR were part of a team which published a paper in the journal Climate Research in which we employed a variety of techniques to derive empirical estimates of the amount of temperature rise that we could expect by the end of this century—a rise that could pretty well be considered to be in-line with the climate sensitivity. We concluded that the expected temperature rise between 1990 and 2100 would be in the range 1.0°C to 3.0°C with our best guess being 1.8°C (in contrast to the IPCC estimates, which, at the time, were for a rise of between 1.4°C and 5.8°C).


Annan, J.D., and J.C. Hargreaves, 2009. On the generation and interpretation of probabilistic estimates of climate sensitivity. Climate Change104, 423-436, doi:10.1007/s10584-009-9715-y,

Michaels, P.J., P.C. Knappenberger, O.W. Frauenfeld, and R.E. Davis, 2002. Revised 21st century temperature projections. Climate Research23, 1-9.

Schmittner, A., et al., 2011. Climate sensitivity estimated from temperature reconstructions of the Last Glacial Maximum, Science, in press*,