How precise are U-series coral ages?

Abstract

U-series dating of fossil reef corals is a well established and widely applied technique in paleoclimate research. Many fossil corals, however, show evidence for post-depositional diagenetic alteration, and it is generally accepted that the accuracy of U-series coral ages is more limited due to coral diagenesis than analytical precision. In recent years, three models have been published that try to correct the effects of diagenesis and allow the calculation of model ages [Thompson W. G., Spiegelmann 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, 365–381; Villemant B., and Feuillet N. (2003) Dating open systems by the 238U– 234U– 230Th method: application to Quaternary reef terraces. Earth and Planetary Science Letters 210, 105–118; Scholz D., Mangini A., and Felis T. (2004) U-series dating of diagenetically altered fossil reef corals. Earth and Planetary Science Letters 218, 163–178]. Here, we assess the age variability of both conventional 230Th/U-dating and the three models by application to different sub-samples of individual coral specimens. The age variability, estimated as the 2 σ-standard deviation on the individual ages, is compared with the errors quoted by the different methods. Our results show that the errors of conventional 230Th/U-dating as well as those of the method of Thompson et al. (2003) do not account for the true age variability. The age variability of both methods is in the range of the errors given by the models of Villemant and Feuillet (2003) and Scholz et al. (2004). Furthermore, we show that the widely used reliability criteria are not sufficient to identify all diagenetically altered corals. In contrast, analysis of different sub-samples of one coral specimen allows (i) to estimate the real age variability, (ii) to test if the assumptions of the models are fulfilled, and (iii) to investigate the diagenetic processes in more detail. Thus, this method should generally be applied to obtain more reliable U-series coral ages and errors.