IMPROVING COMBINED U-TH AND (U-TH)/HE GEOCHRONOLOGY BY MONTE CARLO STATISTICAL CORRECTIONS FOR URANIUM SERIES DISEQUILIBRIUM

Combining U-Th ion microprobe analysis with (U-Th)/He dating has strong potential for Quaternary volcano chronostratigraphy. U-Th zircon ages constrain the time of crystallization, whereas (U-Th)/He dates the eruption. Uranium-series disequilibrium of long-lived intermediate daughters ²³⁰Th and ²³¹Pa, however, requires significant corrections for Quaternary (U-Th)/He ages. For zircon, deficit in ²³⁰Th (~75.7 ka half-life) causes retarded ⁴He accumulation compared to secular equilibrium and consequently (U-Th)/He ages that appear too young. On the other hand, excesses in ²³¹Pa (~32.8 ka half-life) are expected for zircon, leading to excess He and overestimation of (U-Th)/He ages. Disequilibrium at the time of crystallization can be corrected through known partition behavior of parent and daughter isotopes (D parameters), but many volcanic rocks show evidence for protracted crystal growth and pre-eruptive residence during which radioactive decay changes initial disequilibrium. For this reason, knowing crystallization ages through direct measurement of individual crystals is essential. In order to improve disequilibrium corrections for (U-Th)/He dating, we have developed a new numerical Monte Carlo statistic routine ("MCHeCalc"). Input parameters are uncorrected (U-Th)/He ages, crystallization ages, age uncertainties, and values for D. The program output comprises individual disequilibrium corrected (U-Th)/He ages, and determines a best-fit eruption age for multiple analyses including a goodness-of-fit parameter. We tested MCHeCalc on published (U-Th)/He zircon data for Rangitawa Tephra that fall on the 0 ka crystal residence isochron, and thus are thought to be in full disequilibrium. The McHeCalc age of 330^+3.5_-4.2 ka (goodness-of-fit = 0.48; number of zircon analyses = 6) fully reproduces the published eruption age of 330±10 ka. Ongoing dating studies of Late Quaternary volcanism by combined U-Th and (U-Th)/He zircon geochronology coupled with our new Monte Carlo disequilibrium correction analysis comprise volcanism in extensional settings (Salton Buttes, Tres Virgenes, Juan de Fuca), and convergent margin volcanism (Saint Lucia). We demonstrate through comparison with published data that our method can yield accurate and precise eruption ages as young as ~2,500 years.