Comparison of laser-ablation and solution-mode ICP-MS techniques for measuring speleothem 87Sr/86Sr values

Friday, 18 December 2015
Poster Hall (Moscone South)
Barbara Elaine Wortham1, Corinne I Wong1, Isabel P Montanez2, Lucas C.R. Silva3, Troy Rasbury4 and Justin J Glessner2, (1)Boston College, Chestnut Hill, MA, United States, (2)Univ of California, Davis, Earth and Planetary Sciences, Davis, CA, United States, (3)University of California Davis, Land, Air, and Water Resources, Davis, CA, United States, (4)Stony Brook University, Stony Brook, NY, United States
Reconstructing past changes in precipitation amount is critical to delineating controls on paleovegetation dynamics in South America. Although speleothem, lake, and ice core δ18O records from the region serve as a proxy of monsoon intensity, δ18O values do not serve as a direct proxy of local precipitation amount. To address this, we are developing a reconstruction of past moisture conditions using 87Sr/86Sr values measured in a fast growing speleothem (~2 mm/yr) from the central Brazilian savanna for which a late Holocene (0-1.8 ka) δ18O record has previously been developed. Speleothem 87Sr/86Sr values reflect the degree of water-rock interaction, as dictated by water residence time, and are used to interpret relative moisture conditions. In this study, we explore the potential of developing a high-resolution speleothem 87Sr/86Sr record using laser-ablation multi-collector ICP-MS by comparing analyses generated using both laser- and solution-mode techniques.

Laser-mode techniques allow for sampling of individual speleothem lamina, which provides a higher resolution record than solution-mode techniques and eliminates the potential of contamination from conventional drilling. Preliminary results from a younger portion of the speleothem yield a laser-mode (averaged) value of 0.72294 ± 0.00046 (reproduced with parallel scans) and a solution mode value of 0.72338 ± 0.00002 suggesting that these methods provide analogous 87Sr/86Sr values. However, the preliminary results on an older portion of this speleothem have solution- mode derived values of 0.72239 ± 0.00001 and 0.72166 ± 0.00001, with corresponding laser-mode values of 0.72188 ± 0.00029 and 0.72102 ± 0.00037, respectively. Suggesting that in the older portion of the speleothem, laser-mode techniques do not yield analogous solution-mode 87Sr/86Sr values. These differences in different areas of the speleothem highlight the need for method development and further testing.