GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 282-14
Presentation Time: 5:05 PM


ZHOU, Chunhui, Gemological Institute of America, 50 W 47th St., New York, NY 10036, ORLAND, Ian J., WiscSIMS, Department of Geoscience, University of Wisconsin–Madison, 1215 W Dayton Street, Madison, WI 53706, MYAGKAYA, Elina, Gemological Institute of America, 50 west 47th Street, New York City, NY 10036, HOMKRAJAE, Artitaya, Gemological Institute of America, 5345 Armada Drive, Carlsbad, CA 92008 and VALLEY, John, Department of Geoscience, University of Wisconsin, Madison, WI 53706

The oxygen isotope ratio (d18O) of biogenic carbonate, e.g. foraminiferal shells and fish otoliths, can be used to reconstruct the ambient water temperature where the carbonate formed. Here, we test this approach in pearls, which can form within the soft tissues of various freshwater and saltwater mollusks due to, for example, irritations or injuries to their mantle tissues. Pearls form concentrically and are composed of carbonate (aragonite) and the protein conchiolin, which are the same materials used to build the surrounding mollusk shell. The mineral composition of pearls and their predictable growth zoning suggest that a core-to-rim traverse of d18O analyses may reveal a time-series of environmental variation (T and d18Owater) during their formation.

For this study, d18O traverses were measured in four freshwater pearls that were cultured in Kentucky Lake, Tennessee, during the 1990s and early 2000s. To maximize the spatial-resolution of each traverse, d18O was analysed using the WiscSIMS IMS 1280 Secondary Ion Mass Spectrometer (SIMS) at UW-Madison. Ratios of d18O were measured in 10mm-diameter spots – approximately weekly temporal resolution – along the cross section of each pearl. This is the first study of d18O in pearls by SIMS and at such high spatial resolution. Each traverse has periodic d18O variations with amplitudes as large as 6‰. This range of d18O is best explained by the seasonal changes in surface temperatures observed in the lake, which typically range from 6°C in January to 30°C in August. We interpret each d18O cycle as a year of pearl growth, and thus find that the growth duration for each pearl was between 1.5 and 3.5 years. The timing and duration of winter growth cessation can also be extrapolated from the δ18O data, which provide valuable information for both the pearl culturing industry and paleoclimate scientists.

Our results suggest that pearls can be used as an environmental proxy with fidelity at seasonal or even weekly temporal resolution, and that analysis of δ18O in pearls has the potential to help with pearl identification and provenance determination.