The astronomical theory and data compared - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 314

Fifth, as ice sheets expand ocean levels drop increasing land area at the expense of
the surface area of the oceans. Since land has a higher albedo than oceans this
would provide positive feedback for further cooling. In addition, the presence of
such ice sheets apparently increases world storminess as evidenced by dust and salt
content in ice cores. Hence, once an era of heavy glaciation begins there are
natural forces that propagate this trend forward in time. This raises the following
questions: (1) what triggers the origin of such ice ages and, even more baing,
(2) why do ice ages end at all and why do they end precipitously?
The Earth system is complex. The distribution and movement of water is a
key factor. Water on Earth exists in three phases (solid, liquid, and gas) and
transitions between these phases occur with large transfers of thermal energy as a
result of the high heat of the vaporization and crystallization of water. Heat input
to the Earth from the Sun is heavily weighted toward low latitudes, and there are
significant temperature gradients from the tropics to the polar areas. Were it not
for heat transfer toward polar areas (by oceans and atmosphere) along this
gradient, the polar areas would freeze over, extending glacial conditions down to
mid-latitudes via the various feedback mechanisms described above. A tenuous
balance is achieved between natural forces tending to extend glacial conditions in
polar areas toward mid-latitudes vs. the transfer of heat from lower latitudes to
counterbalance this tendency. Apparently, this unstable equilibrium can be upset
by relatively small perturbations, driving the Earth's climate either toward glacial
or interglacial conditions, amplified by significant positive feedback effects.
Data are available from ice cores, ocean sediments, and other sources which
reveal aspects of past climates dating back hundreds of thousands and even
millions of years. These data indicate that there has typically been a long-term
secular pattern of very roughly repeatable cycles whereby great ice sheets have
slowly built up over many tens of thousands of years culminating in a glacial
maximum, followed shortly by a rather abrupt end of the ice age with rapid
global warming leading to an interglacial period of perhaps 10,000 or more years.
Interglacial periods seem to end with abrupt climatic cooling to start a new ice age
that gradually expands over many tens of thousands of years. This pattern is
actually quite variable, but the outline described here seems to describe the last
four ice age cycles fairly accurately. Prior to that, the regularity of cycles was less
evident. Superimposed upon this longer term secular variation, there have been
numerous significant sudden climate changes that may be viewed as noise in the
main signal. When examined at higher resolution these short-term fluctuations
often show an extremely rapid increase in temperature over perhaps decades
followed by a slow decline back to glacial conditions over a millennium or two. It
is not clear whether or how these short-term climate fluctuations are related to
longer term secular trends.
While ice core and ocean sediment data clearly reveal the existence of past
cycles and fluctuations in the Earth's climate, these data are couched in terms of
isotope ratios and other contents of cores and sediments. Converting these data
into specific climatological variables (e.g., global average temperature, global ice
volume, etc.)

is not straightforward. The models originally developed (and