Geoscience Reference
In-Depth Information
recrystallisation of the fossil shells in the much colder waters at the bottom of the
sea (Pearson et al., 2001). Daniel Schrag (Harvard University) reached much the
same conclusion in 1999 by means of a mathematical model of the recrystallisation
of these very porous microshells buried in the sea floor (Schrag, 1999). At higher
latitudes, where there is less temperature contrast between the surface and the bot-
tom of the sea, this recrystallisation issue is less of a problem for
18 O palaeother-
mometry.
There are other problems with this form of 18 O analysis, especially with regards
to determining climates prior to the last glacial-interglacial cycle (of the order of
a hundred thousand years) and going back millions of years. These relate to dating
benthic (deep-sea) forams. First, more ancient deep-sea sediments are increasingly
disturbed due to a variety of factors over millions of years, including plate tectonic
movement and the weight of fresher sediments being superimposed on top. Second,
14 C, as used in dendrochronology (see section 2.1.1), is not applicable over such long
timescales and so radionucleotides with appropriately longer half-lives are required.
However, this reduces the resolution of sediment age determination. Consequently,
whereas we might in theory (see below) be able to use various techniques to determine
regional and global palaeoclimates down to virtually an annual level over the past
glacial-interglacial cycle (of the order of 100 000 years), when attempting to go back
tens of millions of years the palaeoclimatic resolution tends to be of the order of
several centuries, if not millennia, at best. However, it is possible to get a broad-brush
(low-resolution) picture as to how the temperature of the planet has changed over
such timescales (see Figure 2.3).
There are only a few isotopes with long half-lives, including those relating to the
uranium-235 ( 235 U) decay series, that can be used to date remains further back in
time than 14 C. And among these there are a number of radioisotopes that are not
incorporated into biotic remains but with the surrounding geological strata instead.
This obviously begs the assumption that the biotic proxy and the surrounding geo-
logical strata originated at the same time. Nonetheless, radioactive isotope dating
methods can even be used with extremely long timescales that take us back to before
life on Earth arose. Rubidium-87 ( 87 Rb), with a half-life of 4.8
10 10
×
years, is used
by geologists for dating back to the formation of the Earth.
Another problem is the resolution of change discernable from 18 O analysis. The
problem is that the climate can sometimes change faster than 18 O concentrations.
For 18 O concentration to properly reflect the climate there needs to be an appropriate
amount of mixing of the oceans' waters and new balances struck with 18 O in its
various guises in the atmosphere and ocean. This takes time. Broadly speaking, 18 O
analyses are not particularly useful in resolving climatic episodes of much less than
1000 years in duration.
In short, 18 O Foraminifera and coral palaeothermometry is of value (1) but only
for periods of time on the planet when there were virtually no ice caps, (2) if one
concentrates on such proxies from high latitudes away from the tropics so as to
minimise recrystallisation problems and (3) if the climate being elucidated is one that
lasts for roughly 1000 years or more.
However, there is another value in 18 O analysis when there were or are large ice caps
on the planet, such as now in Antarctica and Greenland, even in our warm (Holocene)
Search WWH ::




Custom Search