224
A. Paul· W. H. Berger
Oxygen isotope data from deep-sea sedimentary cores are commonly taken
as a direct proxy for global ice volume. It is assumed that water mass effects
and, in the case of low-latitude planktonic records, temperature effects are
small. Long records extending back through the time of the onset of Northern
Hemisphere glaciation at about 2.5 Ma before present (B. P.) (Raymo 1994) and
before show that marked changes in the character of the ice age oscillations
occurred in the late Pliocene as well as in the mid-Pleistocene. Impressive records of this type stem from North Atlantic Deep Sea Drilling Program (DSDP)
site 607 (Ruddiman et al. 1989), Ocean Drilling Program (ODP) site 677 in the
Panama Basin (Shackleton et al. 1990), ODP site 659 in the subtropical eastern
Atlantic (Tiedemann et al. 1994) and ODP site 806 on the Ontong Java Plateau
(Berger et al. 1994; Berger et al. 1995; Bickert et al. 1997). In Fig. 13a, we show
the benthic 8 18 0 record of ODP site 806B. As evidenced in this and similar
cores, the ice volume oscillations in the early Pliocene were of low amplitude
and high frequency. Then a major transition occurred from a warmer to a
cooler climate which, subsequent to about 2.5 Ma B. P., was dominated by glacial cycles of a larger amplitude with a period near 40 ka, obviously related to
obliquity variations. Data from a number of deep-sea cores indicate that another transition led to an even cooler climate witnessing the build-up of unusually large ice sheets at about 900 ka B. P. (Maasch 1988) and the onset of
high-amplitude climate cycles with near-l00-ka period at about 650 ka B. P.
(Berger and Wefer 1992; Mudelsee and Schulz 1997). This latter multiple transition phenomenon is referred to as the mid-Pleistocene climate transition
(Berger and Wefer 1992; Mudelsee and Stattegger 1997).
The climate transitions in the late Pliocene and mid-Pleistocene, as well as
the climate cycle with near-lOO-ka period, pose a twofold problem to the astronomical theory. (1) The rhythmic variations in insolation have not changed in
character over the entire Cenozoic (Berger et al. 1992; Laskar et al. 1993). (2)
The cycle with near-l 00-ka period is the dominant feature of nearly all climatic
time series of the late Pleistocene, but eccentricity-induced changes in insolation at about this period are smaller than 1 %. Hence, in order to explain the
observed behaviour, the climate system must have active internal degrees of
freedom, or there must be additional forcing mechanisms, or both.
In this chapter we consider the effect of active internal degrees of freedom
such as bedrock depression and uplift, and additional forcing mechanisms
such as the accretion of meteoroids and interplanetary dust particles (IDP) or
a slow tectonic decrease in atmospheric carbon dioxide concentration. To this
end we use a paleoclimate model (PCM) of the late Cenozoic climate based on
the work of Saltzman and Verbitsky (1993). This model is globally averaged
and zero-dimensional. It is an attempt to express the main processes which
are believed to determine paleoclimatic change on long time scales ranging
over millions of years, in a minimum form. We add two physical factors to it:
the elevation-temperature feedback and inclination forcing.
A. Paul· W. H. Berger
Oxygen isotope data from deep-sea sedimentary cores are commonly taken
as a direct proxy for global ice volume. It is assumed that water mass effects
and, in the case of low-latitude planktonic records, temperature effects are
small. Long records extending back through the time of the onset of Northern
Hemisphere glaciation at about 2.5 Ma before present (B. P.) (Raymo 1994) and
before show that marked changes in the character of the ice age oscillations
occurred in the late Pliocene as well as in the mid-Pleistocene. Impressive records of this type stem from North Atlantic Deep Sea Drilling Program (DSDP)
site 607 (Ruddiman et al. 1989), Ocean Drilling Program (ODP) site 677 in the
Panama Basin (Shackleton et al. 1990), ODP site 659 in the subtropical eastern
Atlantic (Tiedemann et al. 1994) and ODP site 806 on the Ontong Java Plateau
(Berger et al. 1994; Berger et al. 1995; Bickert et al. 1997). In Fig. 13a, we show
the benthic 8 18 0 record of ODP site 806B. As evidenced in this and similar
cores, the ice volume oscillations in the early Pliocene were of low amplitude
and high frequency. Then a major transition occurred from a warmer to a
cooler climate which, subsequent to about 2.5 Ma B. P., was dominated by glacial cycles of a larger amplitude with a period near 40 ka, obviously related to
obliquity variations. Data from a number of deep-sea cores indicate that another transition led to an even cooler climate witnessing the build-up of unusually large ice sheets at about 900 ka B. P. (Maasch 1988) and the onset of
high-amplitude climate cycles with near-l00-ka period at about 650 ka B. P.
(Berger and Wefer 1992; Mudelsee and Schulz 1997). This latter multiple transition phenomenon is referred to as the mid-Pleistocene climate transition
(Berger and Wefer 1992; Mudelsee and Stattegger 1997).
The climate transitions in the late Pliocene and mid-Pleistocene, as well as
the climate cycle with near-lOO-ka period, pose a twofold problem to the astronomical theory. (1) The rhythmic variations in insolation have not changed in
character over the entire Cenozoic (Berger et al. 1992; Laskar et al. 1993). (2)
The cycle with near-l 00-ka period is the dominant feature of nearly all climatic
time series of the late Pleistocene, but eccentricity-induced changes in insolation at about this period are smaller than 1 %. Hence, in order to explain the
observed behaviour, the climate system must have active internal degrees of
freedom, or there must be additional forcing mechanisms, or both.
In this chapter we consider the effect of active internal degrees of freedom
such as bedrock depression and uplift, and additional forcing mechanisms
such as the accretion of meteoroids and interplanetary dust particles (IDP) or
a slow tectonic decrease in atmospheric carbon dioxide concentration. To this
end we use a paleoclimate model (PCM) of the late Cenozoic climate based on
the work of Saltzman and Verbitsky (1993). This model is globally averaged
and zero-dimensional. It is an attempt to express the main processes which
are believed to determine paleoclimatic change on long time scales ranging
over millions of years, in a minimum form. We add two physical factors to it:
the elevation-temperature feedback and inclination forcing.