Principal indicators of past climates - Climate Change: Biological and Human Aspects - page 57

climate indicator. Deuterium-containing water, because of hydrogen's smaller size,
is not so discriminated against by living things, so the preferential evaporation and
condensing of water containing isotopes of deuterium is more solely temperature-
dependent. 1 More accurately, the deuterium concentration of ice in terrestrial ice caps
such as Antarctica and Greenland reflects the temperature difference of the water
(primarily the ocean) from which it evaporated and the air from which it froze out as
snow. Deuterium concentration in such ice caps therefore reflects part-hemispheric
more than global climate change. Yet, as we shall see in the next chapter, there is
considerable agreement between the deuterium records of the northern hemisphere's
Greenland cores and those from the southern hemisphere's Antarctic cores. However,
there are also some interesting differences that will be discussed.
The problem with the isotopic analysis of water or, to be specific, ice as a determ-
inant of palaeoclimate is that it is only possible where ice has existed undisturbed as
far back as the climate you wish to study. This restricts such palaeoclimatic studies,
both in space and time; space because the places where ice exists undisturbed with
longevity are rare, and time also because where it has lain undisturbed has not been
long: even a stable ice cap will slowly build up so that the early lower layers come
under increasing pressure, with an increasing likelihood of deformation. However,
both the Greenland and Antarctic ice sheets have enabled us to piece together the
global climate in the past couple of glacial-interglacial cycles. This has been a key
validation of the Milankovitch theory of orbital climatic forcing, although we are
aware from other evidence that Milankovitch climatic forcing has existed for many
millions of years and so presumably for as long as the Earth has had a troposphere
with discernable weather (as opposed to other factors affecting conditions on the
planet's surface such as planetesimal accretion).
The great utility of the isotopic analysis of ice is that when it does lay undisturbed
it is possible to discern annual layers of snowfall that subsequently have become
compacted to form ice. But there are only a few places where ice is so undisturbed.
As mentioned, one is at the top of the aforementioned Greenland ice cap where there
is minimal lateral movement of ice, and another is the ice cap over the lake at Vostok
in Antarctica. The cores from these places enable us to discern isotopic proportions
of ice with resolutions of a few years, yet the records cover hundreds of thousands of
years. Even so, it just takes one layer to be less clear for the counting of subsequent
layers to be out by one; there is a huge difference between much of a core having
a high resolution and the entire core being so clear. Consequently, although a core
of a thousand years of age may exhibit a high resolution, the precise dating of the
layers through physical counting of the entire length of the core to the present may
only be accurate to within a decade. Nonetheless, this means that it is possible to
discern a particularly detailed picture of the palaeoclimate going back to more than
two glacial-interglacial cycles and a highly detailed picture emerges from the end of
the last glacial maximum some 20 000 years ago. As we shall see this has provided
evidence that the hemispheric climate has 'flickered' in the past, changing rapidly in
just a few years (within a decade) or a decade or two. As we shall see in the next
chapter, climate change is not always as gradual as sometimes portrayed.

1

Those interested in physical chemistry can look up the term Clausius-Clapeyron relation.