SAMPLE SELECTION IN RADIOCARBON DATING

Archaeologists working on the island of O'ahu, Hawai'i, use radiocarbo n dating of samples of organic matter found trapped in fish-pond sedim ents to help them to learn about the chronology of the construction an d use of the aquicultural systems created by the Polynesians. At one p articular site, Loko Kuwili, 25 organic samples were obtained and fund s were available to date an initial nine. However, on calibration to t he calendar scale, the radiocarbon determinations provided date estima tes that had very large variances. As a result, major issues of chrono logy remained unresolved and the archaeologists were faced with the pr ospect of another expensive programme of radiocarbon dating. This pape r presents results of research that tackles the problems associated wi th selecting samples from those which are still available. Building on considerable recent research that utilizes Markov chain Monte Carlo m ethods to aid archaeologists in their radiocarbon calibration and inte rpretation, we adopt the standard Bayesian framework of risk functions , which allows us to assess the optimal samples to be sent for dating. Although rather computer intensive, our algorithms are simple to impl ement within the Bayesian radiocarbon framework that is already in pla ce and produce results that are capable of direct interpretation by th e archaeologists. By dating just three more samples from Loko Kuwili t he expected variance on the date of greatest interest could be substan tially reduced.