The Oligocene to the Quaternary: climate and biology - Climate Change: Biological and Human Aspects - page 130

although the last glacial-interglacial cycle is fairly typical of the series for the past
700 000 years (in the 100 000-year Milankovitch-dominated Earth; see the preceding
section), the fourth cycle ago features an interglacial that may be the best analogy
to the current interglacial in its natural state (without anthropogenic carbon dioxide).
Milankovitch forcing changes that took place four cycles ago were similar to those
taking place now. However, as the last cycle is the easiest to study, and indeed in the
history of palaeoclimatic research it was the one that was first studied closely, it is
appropriate to look at it.
The big difference in global climate between glacial and interglacial times relates
to global precipitation and, obviously, temperature. In glacial times there is less
rainfall globally simply due to the fact that there is less evaporation from the oceans.
With regards to temperature, the glacial Earth is broadly about 4 ◦ C cooler than its
interglacial counterpart and possibly at least 5 ◦ C cooler during the LGM, 18 000 years
ago. However, this represents a global average which, as we shall see, is the same
order of magnitude (i.e. single-figure degrees Celsius) as is anticipated for global
average anthropogenic warming over the next century. However, at high latitudes,
near and on the ice sheets, temperatures during glacials were far cooler and more like
12-14 ◦ C cooler than now.
The tropics were less affected by glacials but they did not entirely escape the cooling
effects. Analysis of forams from the Pacific, and other palaeoclimatic indicators,
suggests that the tropical Pacific sea surface was about 3 ◦ C cooler during the LGM
than today. The Pacific, being such a vast reservoir of heat, is an important component
of the Earth's climatic system. If the high latitudes were to experience a greater
glacial-interglacial temperature difference than the 5 ◦ C global average then one
might expect that major parts of the tropical zone would see a smaller deviation to
maintain the average. However, this is less true on land than in the sea. On land the
effects of mountains ranges on climate and other factors come into play. Some parts
of the tropics were clearly less affected by the glacial as tropical species survived
not just the last glacial but the previous several that took place in the Quaternary. Yet
this was not so everywhere. For example, some parts of Brazil were over 5 ◦ C cooler
during the LGM than today (Stute et al., 1995).
There is much evidence for our understanding of at least the broad glacial-
interglacial temperature changes and timing. There is even a considerable body
of evidence exploring many detailed aspects; alas there is not the space to review
it all here. But the work of Timothy Herbert and his team reported in Science in
2010 serves to illustrate one of the sorts of evidence available. They determined the
timing and amplitude of tropical sea-surface temperature (or SST) change as it is an
important part of solving the puzzle of the Plio-Pleistocene ice ages. They used alken-
one-based tropical sea-surface temperature records (see section 2.2.2) from the major
ocean basins to show coherent glacial-interglacial tropical temperature changes of
1-3 ◦ C that align with (but slightly lead) global changes in ice volume and deep-ocean
temperature over the past 3.5 million years. They showed that tropical temperatures
became tightly coupled with benthic

18 O (an indicator of global ice volume in a time
of ice ages, such as the Quaternary) and Milankovitch orbital forcing after 2.7 mil-
lion years. In addition to providing a chart of temperature and timing, they concluded

δ