Reproducibility of trace element profiles in a specimen of the deep-water bamboo coral Keratoisis sp.

Abstract

Bamboo corals (Order Gorgonacea, Family Isididae) are attractive prospects for deep-sea paleoceanographic reconstruction, capturing trace elements in their calcitic skeletons that may serve as environmental proxies with subdecadal resolution over multi-century timescales. We study the reproducibility and fidelity of trace-element profiles (Ba, Mg, Sr, Mn, U, Pb) in a 420-year-old specimen of the bamboo coral Keratoisis sp. from the SE USA. Using laser-ablation ICP-MS to obtain multiple replicate profiles, we use spectral techniques to distinguish noise and irreproducible variations from fully reproducible geochemical fluctuations that are candidates for environmental signals. By quantifying variability between profiles, we assess the fidelity with which the corals potentially record environmental information. Barium is the most reproducible element in the skeleton, with large fluctuations along different growth radii reproducing to within 4%. Both Mg and Sr have very uniform levels within the coral, but display low-amplitude irreproducible variations that might represent an internal biological process. In the case of Mg, which has been proposed as a paleotemperature proxy, this irreproducibility would represent an intrinsic uncertainty of ∼±0.1 to 0.4 °C. Both Mn and Pb contain some irreproducibility superimposed upon broad reproducible profiles that may be environmental signals. Some of the irreproducible Pb fluctuations correlate with cracks and dark bands in the sample suggesting detrital or surface contamination. Uranium displays large amplitude variations which are not reproducible along different radii. This suggests that uranium cannot be used for paleoenvironmental reconstruction, and may show signs of early diagenesis – a possibility that could complicate attempts to date young Keratoisis sp. samples by U-series geochemistry. The highly reproducible Ba signal allows precise alignment of profiles and thus we can show that growth rate along one radius can vary by a factor of two relative to growth along a different radius. There is no evidence that this large variation in relative growth rate affects either the Mg or Sr incorporation. In addition, geochemical anomalies in Ba and Mg indicate that the very central axis of the specimen may represent a different mode of growth. This study suggests that Keratoisis sp. corals are imperfect recorders of geochemical information, but do contain reproducible variations which are good candidates for environmental signals.