Geoscience Reference
In-Depth Information
Figure 6.3. Effect of contamination by modern carbon on radiocarbon age. (Graph
compiled from data provided in Polach and Golson, 1966 .)
linear response. For example, 5 per cent contamination of a sample of true age 30,000
years would appear as a radiocarbon age of 21,000 years. However, sample preparation
techniques will in general remove all traces of modern carbon and are today vastly
improved on what they were even a decade ago, so this is rarely a significant issue. Far
more problematic is the detection of contamination by inert carbon, which is termed
the 'radiocarbon reservoir effect' (Bj orck and Wohlfarth, 2001 ). In most instances,
the effect is not very large, with values of only about 400 years not uncommon
in many lakes and rivers in the drier regions of Africa and Australia. However, in
some instances, the reservoir effect can be significantly large, as in certain early
Holocene and late Pleistocene lakes in arid northern Chile, where Geyh et al. ( 1999 )
have documented a reservoir effect of at least
2,000 years. Rivers flowing to the
ocean contain both old ( 14 C-depleted) and young ( 14 C-enriched) terrestrial dissolved
organic carbon, and if the young dissolved organic carbon is selectively degraded
during transit, the older carbon will preferentially enter the ocean (Raymond and
Bauer, 2001 ). Within the ocean, near-surface reservoir ages may fluctuate through
time (Bondevik et al., 2006 ), so the assumption of a single unvarying reservoir age at
any particular locality may be unjustified.
There are particular problems associated with the use of charcoal for radiocarbon
dating. We noted earlier the distinction between precision and accuracy in radiocarbon
 
Search WWH ::




Custom Search