Page 291 - Earth's Climate Past and Future

CHAPTER 14 • Millennial Oscillations of Climate 267

sheets, the amount of meltwater runoff, the salinity
of the North Atlantic surface waters, and the rate of
formation of deep water. Possible links among these
responses are numerous: northward advection of warm
surface water could promote ice melting, while a low-
salinity meltwater lid on the surface ocean could stifle
deep-water formation. While various aspects of this
idea are still being explored, several problems are
apparent. For example, the major iceberg releases
occurred during times when the surface ocean was cold.
In addition, the most intensively studied oscillation—
the Younger Dryas cooling—occurred without any
apparent meltwater contribution from the ice sheets
(Chapter 13).
The evidence that millennial temperature changes
in the North Atlantic and Antarctic regions have
opposed timing points toward another possible origin
for the millennial oscillations (see Figure 14–8). This A Strong conveyor belt B Weak conveyor belt
pattern, called the bipolar seesaw, has been interpreted Warmer Cooler
as resulting from changes in the northward redistribu-
tion of heat by the Atlantic Ocean. FIGURE 14-17 Opposite hemispheric responses caused by
The typical pattern of ocean heat transport removes ocean heat transport. (A) When cross-equatorial heat flow in
excess heat from the warm tropics and carries it toward the Atlantic is strong, it warms the North Atlantic but cools
the cold poles (companion Web site, pp. 22–24). The south-polar regions. (B) When cross-equatorial flow weakens,
the temperature responses are reversed.
Indian and Pacific oceans both follow this pattern but
the Atlantic Ocean does not. Instead, heat from the
South Atlantic Ocean crosses the equator and moves
18
into the high latitudes of the North Atlantic Ocean. (in δ O values of benthic foraminifera) were unexpect-
The marine geologist Tom Crowley first proposed that edly found to have the south-polar timing. This finding
changes in the northward transport of heat through the indicates that deep-water fluctuations in the North
Atlantic Ocean accounted for the bipolar seesaw pattern Atlantic Ocean are not linked closely to processes oper-
(Figure 14–17). Large cross-equatorial transports of ating in the north, as required by the conveyor belt
heat would leave the Southern Ocean cold while warm- hypothesis, but are controlled by changes in deep flow
ing the North Atlantic Ocean. Weak transport would from the south.
leave the Southern Ocean warm while cooling the Because the largest temperature changes are
North Atlantic region. Simulations by the climate mod- centered near Greenland and the North Atlantic
eler Tom Stocker and others have provided support for Ocean, this area may still be the center of action of the
this idea. millennial-scale oscillations. One possibility is that the
The cause of this seesaw is not yet clear. The geo- atmospheric circulation in this region is unusually sen-
chemist Wally Broecker proposed that changes in the sitive to small changes in surface climate of some kind
amount of deep water formed in the North Atlantic (ice sheet elevation or some other feature). Because
18
Ocean controlled the amount of heat pulled northward millennial-scale fluctuations in ice core δ O values
into the Atlantic as part of a much larger-scale flow of tend to oscillate between similar extremes (see Figure
heat through the world ocean termed the conveyor belt 14–2), two relatively stable modes may exist in atmos-
(companion Web site p. 23). Physical oceanographers pheric circulation during glacial intervals, with abrupt
have criticized this explanation by pointing out that the switches back and forth between these modes at irregu-
large-scale circulation of the surface ocean is driven not lar (random) intervals.
by deep-water formation but by winds. Based on links already observed at orbital time
Another problem with the conveyor belt hypothesis scales (Chapter 11), large temperature changes over
has come from high-resolution analysis of North Greenland and the North Atlantic could alter tempera-
Atlantic sediments. Changes in temperature of North ture and precipitation patterns in Europe and parts of
Atlantic Ocean surface waters recorded by planktic Asia, including the high-pressure cell in Siberia. These
foraminiferal assemblages fluctuate with the northern changes could disrupt the jet stream circulation in the
millennial-scale timing described earlier, but the upper atmosphere and propagate into more distant
bottom-water fluctuations measured in the same cores regions such as the Santa Barbara Basin. Evidence from