Chapter 19 of Climate Change: The Facts is written by Dr John Abbot and Dr John Nicol on the contribution of CO2 to global warming. The authors describe recent theoretical investigations suggesting that the warming effect of CO2 may be an order of magnitude (a factor of ten) smaller than the numbers used by the IPCC.
Amazingly the assumptions about the forcing effect of CO2 that are fed into the most advanced models used by the IPCC are based on speculations advanced over a century ago. The authors found one recent paper reporting an experimental investigation of the critical parameters compared with over 12,000 scientific papers on the topic of General Circulation Climate Models. They conclude that there is a place for modelling but it must be informed by observational data.
The main sources in the paper are work on modern spectroscopy, that is the way the various gases in the atmosphere handle the incoming and outgoing radiation. This is a somewhat technical matter because there is a wide spectrum of radiation and the way that the various wavelengths are absorbed and radiated depends on the atomic configuration of the various gases. In brief the greenhouse gasses have complex molecules that permit bending, stretching and vibration while the molecules of Nitrogen and Oxygen do not.
It seems that the hard core of human-induced warming theory derives from the mathematical work of Professor Svante Arrhenius that was published in 1896 suggesting that doubling atmospheric CO2 would in increase the global temperature by 5 or 6 degrees C. This was not based on the spectroscopy of CO2 but on the theoretical extrapolation of some attempts to measure the surface temperature of the Moon by the American Samuel Langley. This was based on estimates of the infrared radiation leaving the Moon and reaching the Earth. Strangely the authors did not provide any critical appraisal of the accuracy and soundness of that work and the way that Arrhenius used it.
Arrhenius wrote that it was necessary to obtain experimental evidence for the results that he predicated but the sophistication and the cost of the experimental effort was beyond him. His speculations became public and important when James Hansen picked up the numbers that were coming out of the computer models using versions of the original Arrhenius calculations. In his testimony to the US congressional committee in 1988 be used the number of 4.2C from doubling CO2. The authors note that the IPCC has been adjusting the number downward in small steps from 3.8 in 1995 to 3.5 in 2001 and 3.26 in 2007.
Remarkably they report a statement by Hansen that it might need to be reduced further to 2.5. Other investigators have been bolder in downward adjustments, the references are not at my fingertips but some are as low as 1C and others claim that the suppressing effect of CO2 could outweigh the forcing effect and so there is probably no net effect at all.
This is a fun chapter for nerds, especially the part about the bending and stretching of the molecules.
Bonus reading for nerds. An early (1991) warning on the potential abuse of models in rural research from a leading soil physicist in the CSIRO. And a paper on the pernicious influence of mathematics on science noting the way that mathematicians have to simplify and that often misses the whole scientific point.