δ^18 O “stack” (Fig. 16.11). Onset times of glacial terminations are TI, TII, etc.
(After Lourens et al. 2010.)
Fig. 16.13 δD (deuterium in glacial ice, ‰, indicative of temperature) in the EPICA
core from Dome C in Antarctica (lower curve) plotted with δ^18 O from the NorthGRIP
ice core in Greenland (upper curve). The series are each on their locally determined
time scale without squeezing or stretching to emphasize the matches, but matches are
quite obvious among short-term events, mostly Dansgaard–Oeschger events in
Greenland and antarctic isotopic maxima (AIM; 8, 2, and 17 are numbered) at Dome
C.
(^) (After Wolff et al. 2009.)
(^) Finally, we come back to the interaction of glaciation with CO
2 . It is almost
proportionally removed from the atmosphere as ice piles up on land (Fig. 16.10). It
rapidly and again proportionally reappears in the atmosphere as ice ablates from the
major sheets. Its 15 μm IR absorption is clearly another of the key feedbacks that
sustains cooling during ice-sheet growth and promotes temperature rise as ice sheets
melt. The only likely reservoir of suitably labile CO 2 is the deep ocean. While it
remains difficult to document the specific timing, it is clear that ice build-up
recurrently reduces the thermohaline circulation of the oceans, an idea closely
connected with and promoted by Wallace Broecker. There must be reduced formation