干旱区生态环境知识资源中心

Lacustrine carbonate deposits (tufa) record variations in terrestrial hydrology and are preserved in many now-arid regions of the world, but are challenging to date with precision and accuracy. Many contain detritus and/or unsupported thorium (Th) that degrades or prevents the measurement of precise uranium-series (U-series) ages, and radiocarbon ages are frequently affected by both the reservoir (hard-water) effect and contamination with younger atmospheric carbon. The usual method of testing the accuracy of carbonate C-14 measurements, comparison with U/Th ages or organic carbon dates, does not separate the reservoir and modern contamination effects, allowing for only relatively imprecise age estimates in samples undatable by U/Th.

We have separated the modern contamination problem and the reservoir effect using a step dissolution technique on a variety of carbonate materials from Mono Lake, California, a long-lived closed-basin lake sensitive to regional precipitation variability. New dissolution experiments focus on the deglacial sediments of the Wilson Creek Formation, which preserve ostracodes and fans and mounds of thinolite, a cold-water, hydrated calcium carbonate (CaCO(3 center dot)6H(2)O). Stepped-dissolution ages of thinolite crystals and dense calcites increased by 500-1000 years over the bulk age, and produced a plateau of analytically indistinguishable ages, indicating nearly complete removal of modern carbon. Repeated experiments on ostracodes from a single sample showed an increase of >3500 years over the bulk age and >6500 years from first to last step, but ages increased up to the last similar to 5% of the CO2 evolved, without forming a plateau. This may be due to the extremely large surface area-to-volume ratio of ostracodes, and inhomogeneous dissolution of the hundreds of individual shells required for the experiments. Further experiments are planned to test the effects of modern carbon on tufa, gastropods, and other shells, with the goal of systematically testing the precision and reliability of chronologies for dramatic changes in lake level in arid regions of the world. (C) 2012 Elsevier B.V. All rights reserved.