Overview of the various models for ice ages in the recent past (3 MYBP to present) - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 238

The cycle proposed by Bell and Eng (2007) is affected by the rise and fall of
CO 2 concentrations in interglacial-glacial cycles. At the beginning of a new glacial
cycle, the oceans remain warm and CO 2 is therefore high, providing greenhouse
resistance to further cooling. This lengthens the period over which glaciation
occurs. At the beginning of a new interglacial period, the oceans remain cold and
low CO 2 inhibits the warming process. Hence, CO 2 acts as an inertial force to
resist climate change in these cycles.
The theory of Bell and Eng has some attractive features. A cyclic climate
history naturally falls out from the fact that oceanic warming or cooling lags the
warming or cooling of the surface, and warming produces clouds that, in turn,
produce cooling via increased albedo. However, Bell and Eng did not seem to
include the greenhouse effect of water vapor, which is likely to be very significant
in opposing the putative cloud effect. Furthermore, when one traces out multiple
cycles of surface temperature, ocean temperature, cloud formation, and ice
formation, it is dicult to obtain a result with long glacial periods and
comparatively short interglacials.
8.8 MODELS BASED ON THE SOUTHERN HEMISPHERE
Because the great ice sheets that formed during past ice ages were in the NH,
almost all models based on variable solar intensity have concentrated on solar
input to the NH. However, by comparing the synchrony of climatic changes in the
NH with the SH, Blunier et al. (1998) and Blunier and Brook (2001) found that,
despite the qualitatively different temperature time series between Greenland and
Antarctica, there appears to be a correlation between major discontinuities in
Greenland and slope changes in Antarctica. This is discussed in Section 4.3 (see
Figures 4.15 and 4.16 ). The data suggest that each sudden increase in temperature
in Greenland was preceded by a rather slow moderate temperature rise in Antarc-
tica lasting a few thousand years. In addition, the EPICA Community compared
data over 125,000 years from three Antarctic sites with NGRIP data from Green-
land and the results were similar to those shown in Figures 4.15 and 4.16 . As
shown in Figures 4.15 and 4.16 , the occasional sharp rises in Greenland data also
appear to be preceded by slower more gradual rises in Antarctica (EPICA, 2006).
It has been proposed that there is a correlation between temperature patterns
in Antarctica and Greenland due to a connection between them by means of heat
transport via ocean currents known as the bipolar seesaw (Barker and Knorr,
2007). In this model, increasing solar input to the SH produces an increase in
local temperatures. This stimulates the thermohaline circulation of warm currents
to the NH. When some nonlinear threshold is exceeded, the NH undergoes rather
sudden and decisive heating. As this process proceeds, heat is drawn away from
Antarctica and starts to cool. This reduces the flow of heat to the NH. In addi-
tion, meltwater in the Greenland area interferes with North Atlantic Deep Water
(NADW) formation, reducing thermohaline circulation. Thus, the NH begins to