Geoscience Reference
In-Depth Information
further 5,730 years, until about ten half-lives have elapsed and there are only trace
quantities of
14
C still remaining (
Figure 6.1
).
Initial methods of radiocarbon dating were based on measuring the relative pro-
portions of
14
C to the stable isotope
12
C. The mean isotopic composition of carbon
compounds in nature is around 99 per cent
12
C, around 1 per cent
13
C and 10-
10
per
cent
14
C. The stable carbon isotopic ratio
13
is expressed as the deviation in parts
per thousand from a standard, according to the expression
13
C
∂
=
(
R
/
R
0
−
1
)
×
1000
‰
(6.2)
In this expression, R is the measured ratio of
13
Cto
12
C of the sample, and R
0
is the
same ratio for the standard, much as in
Equation 6.1
. The usual standard in radiocarbon
work is once again the Cretaceous carbonate belemnite,
Belemnita americana
, from
the Pee Dee formation of South Carolina, known as PDB. Radiocarbon dates are
expressed as ages before present (BP), defined as 1950 AD, with a statistical error of
one standard deviation, and the laboratory code number. The modern standard has the
same count rate as wood grown in 1950 AD (Williams et al.,
1998
, appendix 1).
Certain qualifications to an otherwise reasonable set of assumptions now need to
be made (Faure,
1986
; Williams et al.,
1993
; Williams et al.,
1998
). First, Suess
(
1955
) found that the
14
C activity of twentieth-century wood was nearly 2 per cent
lower than that of nineteenth-century wood, as a result of 'dead' or non-radioactive
carbon emitted into the atmosphere through the burning of coal, oil and gas. This
dilution effect is termed the 'Suess effect'. Second, de Vries (
1958
) demonstrated
that the radiocarbon content of the atmosphere has not been constant but has varied
systematically in the past, with
14
C activity around 1500 and 1700 AD up to 2 per cent
greater than it was in the nineteenth century. This phenomenon is now known as the
'de Vries effect'. Third, nuclear explosions and increasing use of nuclear reactors and
particle accelerators have increased the level of
14
C activity in the atmosphere through
the input of humanly produced
14
C. Fourth, and most significantly, the production of
14
C has varied during the late Quaternary as a result of variations in the strength of the
earth's magnetic field, which acts as a shield against cosmic rays. When the earth's
magnetic field is weak, the production of
14
C in the outer atmosphere is enhanced,
and conversely. Higher values of atmospheric
14
C will reduce the age determined by
radiocarbon dating, and vice versa.
For all of these reasons, it is evident that the radiocarbon time scale is not a
calendar time scale. Radiocarbon ages are sometimes older and sometimes younger
than their equivalent calendar ages. Some form of calibration is therefore needed in
order to convert radiocarbon years into calendar years. The first successful attempt at
calibration came from radiocarbon dating of tree rings, the age of which had already
been established from ring counts (Fritts,
1976
). This approach has yielded reliable
calibration back to about 8,000 years ago, using the long-lived bristlecone pines
(
Pinus longaeva
) from the White Mountains of the United States and European oaks
