Eastern Oyster (Crassostrea virginica) shell geochemistry archives geographically unique multi-metal fingerprints at known contaminated sites

Oysters archive information about the environment in which they lived within their shells. Elemental metal to Ca ratios (Me/Ca) of their shells can vary due to drivers related to climate, environment, or an individual's biological processes. However, it is not clear how effective Me/Ca are for assessing population relatedness given the diversity of factors that may influence shell geochemistry. In this study, we use dimension reduction and clustering analyses to evaluate whether a multi-metal geochemical fingerprint can identify unique groupings of Eastern oysters (Crassostrea virginica) collected from the Eastern coast of the United States (Virginia, Maryland, and Delaware). We made bulk shell Me/Ca measurements (Na/Ca, Mg/Ca, K/Ca, Ti/Ca, Mn/Ca, Fe/Ca, Sr/Ca, Ba/Ca, and Pb/Ca) via ICP-MS. Our intra-state comparisons, using Welch's t-tests, did not yield consistent differences, however, when values were statistically different between sites, a given Me/Ca value was more likely to be higher in Superfund site-proximal samples. Here we show Principal Component Analysis coupled with K-Means Clustering can subdivide groups of samples into unique populations, in state- and site-specific clusters. More broadly, we provide a valuable geochemical and statistical approach to assessing population relatedness for C. virginica, which can be applied to evaluate relationships between groupings of oysters based on geography, climate, and/or temporal variations.