Variability of the Earth’s climate - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 82

Figure 2.37. CO 2 and tropical sea surface temperatures (adapted from Royer, 2010).
total negative forcing produced a global average temperature decrease of roughly
4.5 C. Hansen and Sato (2011), Chylek and Lohmann (2008), and Kohler et al.
(2009) all independently estimated the forcing at the LGM (see Table 2.2 ). The
contribution of the diminution of CO 2 at the LGM to total cooling was estimated
by these studies to be in the range 16 to 33%. While it seems likely that solar
input to higher latitudes triggered the cycles, the variability of CO 2 concentration
played a part in determining the extremity of the temperature cycle that resulted
from this trigger. The changes in CO 2 concentration between glacial maxima and
interglacials ( 180 to 280 ppm) are well documented in ice core records,
although no one seems to have a satisfactory explanation for the CO 2 concen-
tration changing this much (simple solubility in the oceans does not suce).
However, the estimates of forcings, particularly due to dust, vary considerably
from investigator to investigator and it is dicult to pin down climate sensitivity
to CO 2 change. However, changes in humidity and cloudiness are unknown and
may be very large factors. There are good estimates available of global average
temperature and the CO 2 concentration at the Last Glacial Maximum (LGM)
20,000 years ago, and if these data are compared with values in the pre-industrial
era (a few hundred years ago) one can thereby estimate the sensitivity of the
climate to CO 2 concentration over the range 180 to 280 ppm. Using this esti-
mated climate sensitivity, one can then estimate the global average temperature
rise in going from 280 to 560 ppm. The various investigators have come up with a
range of projections. It is noteworthy that this range of estimates for the real
world D T G due to doubling CO 2 from 280 to 560 ppm is from 1to 3 C.
However, as we stated above, these estimates do not take into account possible
differences in humidity and cloudiness.