High-precision U-series measurements of more than 500,000 year old fossil corals

Robust, independent age constraints on the absolute timing of climate events based on the U-series dating of fossil coral are sparse before the last glacial cycle. Using multiple-collector inductively coupled plasma mass spectrometry with multiple-Faraday protocols, we are able to date similar to 600 ka samples with an uncertainty of better than +/- 15 ka (2 sigma), representing a three-fold improvement in precision compared with previous techniques. Using these methods, we report U-series measurements for a suite of >500 thousand year old (ka) corals from Henderson Island, an emergent atoll in the south-central Pacific Ocean. The fossil corals show extraordinarily little diagenetic alteration for their age and the best-preserved sample yields a U-series age of 600 15 ka (2 sigma), which overlaps with the timing of the warm Marine Isotope Stage (MIS) 15 interglacial. The open-system model of Villemant and Feuillet [Villemant B. and Feuillet N. (2003) Dating open systems by the U-238-U-234-Th-230 method: application to Quaternary reef terraces. Earth and Planetary Science Letters 210(1-2), 105-118.] and the linear regression (or open-system isochron) is clearly limited for such old samples. However, the open-system model developed by Thompson et al. [Thompson W.G., Spiegelman M.W., Goldstein S.L., and Speed R.C. (2003) An open-system model for U-series age determinations of fossil corals. Earth and Planetary Science Letters 210(1-2), 365-381.] appears to reliably correct for open-system effects in roughly half of the corals, giving a MIS 15 origin for these. Thus the data provide evidence that the systematic addition of Th-230 and U-234 through alpha-recoil is a dominant open-system process occurring in the Henderson Island fossil reef system. Several coral samples yield significantly older Thompson et al. open-system ages between 650 and 750 ka. The uncertainty on these ages (typically +/- 30 kyrs) is too great for precise assignment but most data overlap with the MIS 17 interglacial. The reliability of these ages is currently unclear. It is shown that separate aliquots of the same coral can yield different Thompson model ages. Therefore, there appear to be additional diagenctic mechanisms that create further anomalous excursions in the U-series systematics, limiting the reliability of the Thompson et al. open-system model. (c) 2007 Elsevier B.V. All rights reserved.