Radiocarbon dating and U Th ratios

Of the various radiometric methods available, radiocarbon dating has been by far the most useful for organic material formed within the last ^40 000 years. Radioactive 14C is produced in the atmosphere by cosmic bombardment. It is incorporated in living organisms and, on the death of the organism, provided it is excluded from exchange with the atmosphere, the 14C decays in accordance with its half-life of 5730 years. Its decay thus constitutes a radiometric clock whereby determination of the radioactivity remaining in the sample allows calculation of the sample's age. Carbon-14 dates are normally quoted as dates BP, where the present is taken as AD 1950. They are normally given with one standard error derived from the counting statistics. Several methods are available for determining the 14C activity in a sample. Increasingly, the more conventional methods using several grams of carbon extracted, then measured in gas or liquid form, are giving way to measurements using accelerator mass spectrometry (AMS), by means of which milligram-size samples can be dated. While the basic principles of radiocarbon dating are beguilingly simple, in practice establishing dates is often far from straightforward. Here it is possible to do no more than outline briefly, by way of illustration, some of the complications:

1. The atmospheric concentration of 14C has varied through time, partly as a result of changes in the rate at which the radioisotope is received at the Earth's surface, partly because of changes in the residence time of 14C in the ocean, which in turn affects 14C concentrations in the atmosphere. It follows that organisms incorporating 14C at different times will have different starting values from which decay begins. This means that 14C 'dates' are not true age determinations. Ideally, they need to be calibrated to an absolute timescale established by methods that are not subject to any similar error. For the last 9500 years, this has been done by comparing the 14C 'dates' and calendar ages of tree rings (Stuiver and Reimer, 1993).

2. Beyond the period for which direct tree-ring calibration of 14C dates is possible, there is less certainty involved in converting them to calendar ages. Both varved marine sediments (Hughen et al., 1998) and Uranium/Thorium (U/Th) dating (see below) have been used for calibration. Using a combination of these, Stuiver et al. (1998) proposed calibrations back to 24 000 BP. More recently, Hughen et al. (2004) and Bard et al. (2004) have proposed calibrations back to 50 000 years.

3. One of the consequences of the changes through time in atmospheric 14C concentrations is that, for certain periods, radiocarbon activity does not decline monotonically with age. It may plateau or even briefly increase. Quite a wide range of calibrated ages can thus be ascribed to individual 14C dates within such time intervals. In order to establish the age of materials within these periods, it is often necessary to determine a close sequence of radiocarbon 'dates' and match their pattern to the calibration wiggle. This method of 'wiggle matching' has provided a basis for some rather precise age determinations (e.g., Blaauw et al, 2003, and Figure 3.4).

4. In every environment sampled, it is essential to establish that the dated material is contemporary with the event or stratigraphic horizon to which it is intended to apply. This is not always straightforward. Sedimentation processes often include the re-deposition of older materials, thus introducing older carbon into the sample to be dated. They may also re-deposit older organisms bearing palaeoclimatic signals that are not applicable to the period of sediment accumulation (Okhouchi et al, 2002). In hard-water environments, aquatic organisms may incorporate old carbon during photosynthesis. In the ocean, the age of the carbon used by marine organisms varies greatly with water depth and location, depending on the degree to which the water column in which the organisms live has been isolated from exchange with the atmosphere. There is therefore a 'reservoir' effect that must be estimated correctly if true age is to be calculated from 14C activity. In most studies, it has been necessary to assume that in a given location, this effect remains constant over long periods. Recent work using tephra layers (see 3.4.6) deposited over both land and sea, or synchronous rapid climate shifts to correlate marine sequences with precisely dated terrestrial material (Waelbroeck et al., 2001; Bjorck et al, 2003) has shown that this assumption is not always valid.

One of the most useful dating methods for carbonate materials relies on the Uranium-238 decay series. During the course of the radioactive decay of 238U, it is replaced by 230Th. The rate of decay/replacement is known and, provided the

Calendar age (BC)

carbonate is sufficiently pure and has remained a chemically closed system, the U/ Th ratio constitutes a remarkably precise chronometer over an age range from a few years to c. 350 000 years BP. The method has been successfully applied to corals (Bard et al., 1998), speleothems and lacustrine carbonates (Edwards et al., 1987). From the mid 1980s onwards, 230Th/238U dating has been carried out increasingly by thermal ionisation mass spectrometry (TIMS). This has allowed a significant reduction in sample size along with significant improvements in the precision achievable. Some sense of its importance in pinning down late Quaternary chronologies and correlations may be obtained from the articles by Bard et al. (2004) and Shackleton et al. (2004).

0 0

Post a comment