RADIOCARBON REDATING OF CONTAMINATED SAMPLES FROM A TROPICAL VOLCANO - THE MANSION SERIES OF ST-KITTS, WEST-INDIES

Citation
Dd. Harkness et al., RADIOCARBON REDATING OF CONTAMINATED SAMPLES FROM A TROPICAL VOLCANO - THE MANSION SERIES OF ST-KITTS, WEST-INDIES, Bulletin of volcanology, 56(5), 1994, pp. 326-334
Citations number
NO
Categorie Soggetti
Geosciences, Interdisciplinary
Journal title
ISSN journal
02588900
Volume
56
Issue
5
Year of publication
1994
Pages
326 - 334
Database
ISI
SICI code
0258-8900(1994)56:5<326:RROCSF>2.0.ZU;2-#
Abstract
Heavy rainfall and dense vegetation on tropical volcanoes produce abun dant carbonized wood in pyroclastic deposits, in addition to easy cont amination of this wood by root systems and soluble humic material. Bec ause the physical nature of the charcoal varies, some samples are more prone to contamination. Two independent studies of the same volcano, Mt Liamuiga on St Kitts in the Lesser Antilles, sometimes using sample s from the same carbonized tree, yielded a systematic difference in ra diocarbon ages. An exchange of samples and a re-investigation of three physically distinct types of charcoal yielded the following results. Rare, hard, dense charcoal, lacking contamination, which had yielded a spurious age of 2860 years BP, was redated at 1845 +/- 58 years BP. C ommon soft, friable charcoal with good cellular structure proved to be susceptible to contamination. A field decontamination technique utili zed by one group seems significant as it yields older ages than when o nly routine laboratory pre-treatment was used, indicating that the lat ter technique only partly removes the dried and hard residue produced by the decomposition of modern plant rootlets. A previous date of 2487 0 years BP obtained from powdery charcoal in a horizon beneath the Man sion 'Series' contradicted ages older than 41 000 years BP from common friable charcoal in the lower Mansion 'Series'. The soft powdery char coal was re-investigated using a sample collected a few centimeters fr om the original, although field decontamination of this sample was not possible, more extensive laboratory treatment yielded an age of ca. 4 3 000 years BP, again proving that routine laboratory pretreatments ar e inadequate. A revised geochronology for the Mansion 'Series' is desc ribed and a cautionary discussion is presented for the benefit of inve stigators using radiocarbon ages to date volcanic deposits.