Thermohaline Circulation and Climate Change in the Southern Hemisphere: Characterizing the Spectrum of Variability Over Glacial Cycles

C D Charles (Scripps Institution of Oceanography, La Jolla, CA 92093-0220; ph. 619-534-5911; Internet: ccharles@ucsd.edu); U S Ninnemann (Scripps Graduate Department, La Jolla, CA 92093-0208; ph. 619-534-4838); P G Mortyn (Scripps Graduate Department, La Jolla, CA 92093-0208; ph. 619-534-4838)

There are several ways to approach the origin and extent of rapid climate variability throughout the Pleistocene from the deep sea sedimentary archive. For example, it is possible to determine whether the amplitude and phase of millennial scale variability is similar everywhere and whether the spectrum evolves depending on global climate state. It is also possible to look for geographic fingerprints of oceanic processes by targeting particularly sensitive regions for thermohaline circulation. Here we review the evidence for rapid change in one of these "hot spots" for circulation--the Southern Ocean.

A growing library of deep sea sediment records from the circumpolar region of the Southern Ocean allows a mappable description of ice age surface and deep water variability. The evidence gathered thus far suggests that for the surface ocean , millennial scale changes in climate and nutrient proxy variables are most dramatically expressed in the Atlantic sector, and they are greatly damped (or even non existent) in the Indian and Pacific sectors. This observation strongly suggests that the Atlantic variability is a reflection of thermohaline circulation changes, ultimately associated with the variable production of North Atlantic Deep Water. However, since the formation region of Antarctic Intermediate Water shows little evidence for millennial scale variability, it is unlikely that abrupt thermohaline changes were propagated downstream (to the North Pacific, for example) through this conduit. The explanation for the basinal difference in surface ocean properties is not obvious, but probably involves a geographically heterogeneous effect of vertical stratification. A further complication is that the surface climate changes, deduced from planktonic foraminiferal chemistry and abundance, are not in phase with deep ocean shifts. These deepwater nutrient changes, inferred from foraminiferal carbon isotopes, are also damped in the Pacific basins relative to Atlantic, but millennial scale oscillations can still be traced throughout the circumpolar region. This coherent interbasinal variability cannot be a result of changing productivity regimes --one possible overprint for carbon isotopes--and must reflect thermohaline circulation.

Typical piston cores in rapidly accumulating Southern Ocean sediments do not penetrate to the last interglacial period, precluding an assessment of multiple glacial cycles. However, recently collected (Ocean Drilling Program) sediments from high deposition rate sites in the south Atlantic span the entire Pleistocene. We intend to delineate the evolution of millennial scale variability over this interval, therefore establishing context for the ice age observations.