Ice core data - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 157

Sigman and Boyle (2000) echoed this sentiment: '' ... we have not yet

identified the cause of these variations in CO 2 .''
A number of blogs on the Internet would have you believe that the
explanation for the similarity between CO 2 and T curves results simply from the
difference in solubility of CO 2 in the oceans as a function of temperature.
However, detailed analysis shows that this effect is insucient to account for the
change from about 180-200 ppm under full glacial conditions to about 280 ppm
under full interglacial conditions.
Although most of the carbon on Earth is incorporated into CaCO 3 in rocks,
this carbon pool is too stable to account for pCO 2 changes over glacial cycles.
Carbon in the terrestrial biosphere is available on shorter time frames but, in
order to deplete pCO 2 by 100 ppm, the terrestrial biosphere and soil carbon
reservoirs would have to approximately double in size over about 10,000 years.
Instead, measurements of the d 13 C from deep-sea CaCO 3 suggest that the terres-
trial biosphere released carbon during glacial times—the wrong direction to
explain lower glacial pCO 2 . The only remaining candidate driver for atmospheric
CO 2 change is the oceans. They can hold enough carbon to absorb the atmo-
spheric decrease and can change on 1,000 to 10,000-year timescales (Archer et al.,
2000).
Archer et al. (2000) described two mechanisms that have been proposed
to account for pCO 2 changes in glacial-interglacial CO 2 cycles (GICC). One pro-
posed mechanism to lower glacial pCO 2 is based on an increased rate of biological
productivity in surface waters of the oceans, leading to storage of carbon in the
deep sea due to sinking particles. Either an increase in the ocean inventory of
nutrients (PO 3 4 and NO 3 ) or a change in the ratio of nutrient to C in phyto-
plankton could have stimulated the ocean's biological pump in this way. Models
of the ocean carbon cycle indicate that the pCO 2 is extremely sensitive to the bio-
logical pump at high latitudes and relatively insensitive to low-latitude forcing.
Since iron availability limits phytoplankton growth in remote parts of the ocean, a
dustier more iron-rich glacial climate would have intensified biological product-
ivity in surface waters of the oceans. A second mechanism to lower glacial pCO 2
is to change the pH of the whole ocean, converting seawater CO 2 into HCO 3 and
CO 3 , which are unable to evaporate into the atmosphere. The pH of the ocean is
controlled by any imbalance between the influx of dissolved CaCO 3 from chemical
weathering on land and the removal of CaCO 3 by burial in the deep sea.
Skinner (2006) pointed out that the magnitude of the marine carbon reservoir
and its interaction with atmospheric CO 2 suggests a major role for the oceans in
GICC. While a simplistic model might suggest that the increased solubility of CO 2
in a colder glacial ocean would account for the reduction of CO 2 during ice ages,
detailed models indicate that this would only amount to about 30 ppm of the total
80-100 ppm reduction. Furthermore, even this moderate reduction in CO 2 would
be counteracted by the reduced solubility of CO 2 as the oceans became saltier
during ice ages, as well as by a large reduction in the terrestrial biosphere when
land is covered by ice under glacial conditions. Thus, the net reduction in CO 2
during glacial conditions due to solubility, land changes, and salinity is probably