Abstract
Distribution sampling from sequences of radiocarbon ages is used to obtain confidence intervals for intermediate age estimates, sediment deposition times, and thence pollen deposition rates. Three types of age-depth model are explored by this approach, and it is demonstrated that the choice of model has implications for confidence intervals on the results. Line-fitting of polynomials by least- squares will produce narrower confidence intervals than interpolation models. It is shown that radiocarbon ages should be evenly spaced, and that results towards the ends of sequences have wider confidence intervals. Of the total width of confidence intervals for pollen deposition rates, up to half may be due to the uncertainties associated with the sediment deposition time, even in uniform sediment. The importance of including confidence intervals of all values as an essential part of the presentation of palaeoecological data is stressed.
Keywords
Get full access to this article
View all access options for this article.
