Although it is generally accepted that, as postulated by the Milankovitch theory, Earth’s
orbital variations play an important role in forcing glacial cycles, understanding of their
nature still remains elusive. One of the major challenges for classical Milankovitch theory is
the explanation of the 100 kyr cyclicity that has dominated global ice volume and climate
variability over the past million years. This periodicity is absent in the Milankovitch forcing:
variations of summer insolation at high latitudes of the Northern Hemisphere. The
eccentricity of the Earth’s orbit does contain periodicity close to 100 kyr, but the direct effect
of the eccentricity variations on the global Earth’s energy balance is small. In a view of this
long-standing problem, a number of alternative or complimentary mechanisms have been
proposed in recent decades to explain the 100 kyr cyclicity of the glacial cycles. Since the
number of hypotheses is growing steadily, it became increasingly clear that the
problem could not be resolved without the use of comprehensive Earth system
models that account for the major processes and feedbacks in the system. Such a
model, CLIMBER-2, was developed in Potsdam and is the principal tool for my
analysis. The CLIMBER-2 model incorporates a 3-dimensional ice sheet model
and a closed carbon cycle model. The results obtained with this model strongly
suggest the necessity for "generalization" of the classical Milankovitch theory.
The basic postulates of the Generalized Milankovitch theory can be formulated as
following: (i) the glacial cycles represent a direct, strongly nonlinear response of the
climate-cryosphere system to the orbital forcing; (ii) the strong 100 kyr cyclicity of ice
volume variations originates from phase locking of the "long" glacial cycles to the
eccentricity variations with 100 kyr periodicity. (iii) the existence of "long" glacial cycles
requires a relatively low CO2 level and large areas of the Northern Hemisphere continents to
be clear of thick layers of terrestrial sediments. Physically, the link between glacial
cycles and eccentricity is explained by the fact that the ice sheets tend to grow
monotonously during periods of low eccentricity and reach their critical size (volume)
soon after the minimum of eccentricity. When eccentricity starts to grow, the first
strong positive excursion of Milankovitch forcing leads to the rapid and irreversible
meltback of the Northern Hemisphere ice sheets. The lowering of snow and ice albedos
by the deposition of aeolian dust plays an important role in glacial termination as
well. At the same time, the CO2 concentration not only determines the regime of
glacial variability, but also strongly amplifies 100 kyr cycles. Therefore, simulation
and understanding of the glacial cycles require the use of comprehensive Earth
system models that include both physical and bio-geochemical components of the
system. |