U-Pb zircon and titanite systematics of the Fish Canyon Tuff: an assessment of high-precision U-Pb geochronology and its application to young volcanic rocks

Abstract

A large data set of single and multi-grain zircon and titanite analyses from a sample of the Oligocene Fish Canyon Tuff (FCT), a voluminous ash flow from the San Juan Mountains of Colorado and widely used 40Ar/39Ar geochronological standard, has been used to evaluate the influence of various sources of analytical and geological uncertainty on the calculated age of this tuff by the isotope dilution U-Pb zircon method. Twenty-three single zircon grains and seven small multi-grain fractions of the FCT yield an inverse-variance weighted mean 206Pb/238U date of 28.402 ± 0.023 Ma (2σ; MSWD 0.93) and a slightly older weighted mean 207Pb/235U date of 28.529 ± 0.030 Ma (MSWD 0.74), which are insensitive to common Pb corrections. Initial 230Th disequilibria calculated from a newly measured Th/U = 2.2 for FCT pumice shards indicate Th-deficiency for these zircons; its correction brings the weighted mean 206Pb/238U date to 28.478 ± 0.024 (MSWD 0.97), minimizing the discordance of the FCT zircon analyses. These calculations culminate in Concordia ages (Ludwig, 1998) for the crystallization of the FCT zircons of 28. 476 ± 0.029 Ma (MSWD 1.50) without propagating decay constant errors, or 28.498 ± 0.035 Ma (MSWD 1.03) with propagated decay constant errors.These results are in close agreement with a previous single-grain 206Pb/238U zircon date of 28.41 ± 0.05 Ma for the FCT (Oberli et al., 1990), however they do not corroborate the discordia lower intercept date of 27.52 ± 0.09 put forth by Lanphere and Baadsgard (2001). The 28.476 ± 0.064 Ma Concordia age for FCT zircons (including systematic decay constant errors and uncertainty in Pb/U tracer calibration) is substantially older than all published 40Ar/39Ar sanidine ages, which range from 27.5 to 28.05 Ma (Lanphere and Baadsgaard, 1997; Renne et al. 1998). A lengthy period (∼400 k.y.) of zircon residence in the magma chamber before eruption is commonly called upon to explain the age discrepancy. However, FCT titanite, which may remain open to Pb diffusive exchange until quenching upon eruption, yield a minimum 230Th-disequilibrium corrected weighted mean 206Pb/238U age of 28.395 ± 0.049 Ma (± 0.078 with tracer uncertainty), weakening residence time arguments, and supporting recent calls for re-evaluation of the 40K decay constant and the accepted ages for commonly used 40Ar/39Ar mineral standards. The FCT zircon Concordia age also provides an empirical constraint on the decay constants of the U isotopes, indicating the relative accuracy of the currently accepted decay constants to within the original counting statistic errors of Jaffey et al. (1971).