Major constraints on the use of radiocarbon dating for tephrochronology

It is well recognized that radiocarbon dating is a powerful tool capable of providing a detailed chronostratigraphy for areas that have been subjected to recurrent volcanic activity during Late Pleistocene and Holocene time. In such geological contexts carbon-bearing materials suitable for 14C dating are, for the most part, refractory componetns of pedosphere (e.g. palaesol humic matter) and organic detritus (e.g. charred/uncharred remains of higher plants) which occur within the tephra suites. Since such materials derive initially from atmospheric carbon dioxide, as well as record significant lapses in volcanic activity, it is then often assumed that these conditions comply ideallly with the fundamental requirements of the radiocarbon dating theory. However, two recent 14C dating programmes based on materials from the volcanic districts of the Phlegrean Fields and the island of Procida (Campania, southern Italy) yielded evidence to the contrary and, in turn, gave rise to concern over the validity of the 14C dating method when applied in volcanic regions. In particular, it has been found that the conventional radiocarbon ages recorded by modern tree-leaves from the caldera of Solfatra volcano range from 'modern' to ca. 5000 yr BP. Furthermore, the magnitude of the age anomaly is determined by the pattern of localised dilution of the mean atmospheric 14C concentration by juvenile (14C-free and 13C-enriched) carbon dioxide issuing from nearby volcanic vents. In the case of paleosols, a series of 14C measurements on different chemical fractions isolated from the humic matter in a suite of four paleosol horizons entrained in tephra deposits at Procida island yielded, for each paleosol horizon, a systematic ca. 2000 yr scatter of 14C ages. It is shown that such intrasoil age differences, which reflect the origin and history of humic constituents, can be used to afford a better definition of the true paleosol age.