The most-quoted climate sensitivity range (IPCC 4AR) suggests a median temperature response to a doubling of the atmospheric CO2 concentration of 3 degrees Celsius – and a 66 percent probability range warming under CO2-doubling will be somewhere between 1.5 and 4.5 degrees.

New paleoclimatic research focusing on reconstructed temperature-CO2 relations during the Last Glacial Maximum leads to climate model runs indicating an about 23 percent lower climate sensitivity, with a median sensitivity value of 2.3 degrees Celsius (and a 66 percent range of 1.7-2.6 degrees – so an especially smaller maximum response).

As you can see in the picture above right, this is by no means the first attempt to reconstruct climate sensitivity from paleoclimatology. [Image from a 2008 climate sensitivity metastudy.]

The Last Glacial Maximum is a period when ice sheets during the last northern hemisphere ice age were at their highest extent. It lasted from somewhere between 26,500 and 19,000 years ago [and ended with the ice age cold peak and subsequent CO2-induced warming about 18,000 years ago].

In this study, which was led by Oregan State University, funded by the US National Science Foundation’s Paleoclimate Program and just published in Science, researchers used “extensive sea and land surface temperature reconstructions” of around 21,000 years ago – in stead of the (late) Holocene temperature record that is mostly used.

Climate sensitivity: an enormous pile of data – and an ongoing debate

So this sounds like some welcome good news for our climate. But which temperature-CO2 coupling would you rather use? There are good reasons why climate scientists so often stick to climate datasets of between 1850 and the present. Firstly that is the amount of available data – and secondly the relatively high reliability of it.

And than there is the chance that climate sensitivity isn’t a linear phenomenon. If a doubling of CO2 centred on values between low to pre-industrial leads to a smaller (relative) temperature response than doubling CO2 centred on the range between pre-industrial and current values – what does that tell us about future climate sensitivity – under ever-rising concentrations?

There is the possibility that the relative importance of CO2 as a climate forcer increases as it transcends the other controllers of Earth’s energy balance (some of which may be masked more in ice age studies – like uncertainties around the amount of ice age aerosol climate forcing, ice age thermohaline stability and as always insolation differences throughout the Pleistocene).

In this light paleo research is very important too – as indeed when one looks at high-CO2 warm periods (for instance in the Tertiary) some data seems to suggest a climate sensitivity that would be somewhat higher than the IPCC range.

[It is also important to note CO2 can be a triggering factor (after some climatic response threshold) and that warming can then be amplified by various fast-acting feedbacks – like clouds, which may add 50-100 percent to ‘total direct climate sensitivity’. The importance of incorporating such water feedbacks in ‘CO2’ climate sensitivity is also quantified by a 2010 NASA study.]

It would be very interesting to see whether there will be any official climate sensitivity adjustments in the forthcoming IPCC report. Meanwhile it shouldn’t hurt science to take a look at present-day temperature observations – and to remember that although we are probably (unknowingly) already at 400 ppm CO2 per somewhere in one of these months these temperatures are not the definite result of this CO2 concentration – but of that of some 30-40 years ago – thanks mainly to the inert oceans of our planet. This means on all accounts – it will get warmer still.

© Rolf Schuttenhelm | www.bitsofscience.org