INTER-SPECIFIC RADIOCARBON AGE DIFFERENCES IN MOLLUSC SHELLS FROM COASTAL ARCHAEOLOGICAL DEPOSITS, SOUTHEASTERN ARABIA

Marine carbonate shells are commonly radiocarbon dated for both geological and archaeological investigations due to the abundance of this material in coastal settings. The use of this material for precise dating can be complicated, however, due to potential differences between the radiocarbon ages of coeval shells of different fauna (i.e., shells of different genera that lived/died at the same time). In this study, we investigated the issue of inter-specific age variation by radiocarbon dating shells of different genera from small hearths and middens located within archaeological sites in Sharjah, United Arab Emirates. The archaeological sites were occupied during the Iron Age (1300-300BCE) and Neolithic Age (6000-3000BCE). Each sample collection targeted relatively ephemeral deposits (e.g., single use hearths or layers within small middens); these represent live-collection and immediate consumption of shellfish, thereby minimizing potential age variability within the sample. Genera compared here include gastropods (Terebralia and Hexaplexin) and bivalves (Marcia and Saccostrea). The conventional radiocarbon ages of Terebralia, Saccostrea and Hexaplexin shells from a small Neolithic shell midden were 5750+/-50, 5630+/-60, and 5480+/-50 radiocarbon years, respectively. Two separate Iron Age hearth deposits were sampled and dated. Each of the Iron Age shell groups also produced a trend of decreasing conventional radiocarbon ages, with an average of 170 years difference between each genus. Within each contemporaneous deposit we found that Marcia (an infaunal clam) and Terebralia (mangrove snail) yielded older conventional radiocarbon ages whereas Saccostrea (mangrove oyster) and Hexaplexin (a Murex-type snail) consistently produced the youngest ages. These age variations suggest that the shells of these molluscs either systematically incorporate radiocarbon from different reservoirs while precipitating, or have systematically different degrees of diagenetic contamination. These findings imply a need for caution when attempting to precisely compare ages from more than one shell type.