Understanding Extraneous Argon in Silicic Volcanic Products Using 40 Ar/ 39 Ar Laserprobe Geochronology

Abstract

The Ar-Ar dating technique is one of the most widely applied geochronological techniques to products of silicic volcanism, which represent geologically instantaneous events, and have been used to calibrate the geological timescale, correlate stratigraphy and biostratigraphy over large areas, and assess the impact of explosive volcanic eruptions. Recent advances (e.g., improved instrument precision and recalibration of the K-Ar decay scheme), are now making it possible to obtain increasingly precise and accurate ages for young volcanic eruptions, K-poor minerals, and even discrete parts of single crystals. These advances have highlighted the realisation that relatively small levels of Ar contamination (e.g., extraneous Ar, either excess (40ArE), or inherited Ar), previously assumed to be minor, may now have a considerable effect on the accuracy of ages determined using the Ar-Ar technique. To assess the issue of extraneous Ar, this study applied the Ar-Ar technique to a range of minerals (including sanidine, plagioclase and biotite), and glass separated from the products of large-volume silicic magma systems, which have undergone repeated cycles of crystallisation and rejuvenation. The in situ Ar-Ar laserprobe technique was applied to dacite of the Fish Canyon magmatic system (erupted at ~28 Ma; Colorado, USA), and the single-grain fusion Ar-Ar laser melting technique was applied to rhyolitic pumice (from explosive ignimbrite and ash fall, and effusive dome building events) of the Yellowstone Plateau Volcanic Field (< 2.1 Ma; Idaho, Wyoming, USA), and the Bishop Tuff (erupted at -0.76 Ma; California, USA). The in situ study revealed variable 40ArE contamination of feldspar (i.e., hosted in fluid and melt inclusions in plagioclase and sanidine), and biotite (incorporated 40ArE due to having a relatively high Ar mineral/fluid partition coefficient), resulting in an age range of 25.07±1.86 to 61.46±10.11 Ma. Single-grain fusion experiments revealed that in some cases the source of extraneous Ar was identifiable as partially re-set xenocrysts contaminating systems immediately prior to, and/or during eruption. In other cases, in particular some Yellowstone rhyolite domes, persistent recycling of material (crystal mixes including phenocrysts and antecrysts imparting an inherited Ar component), has resulted in a spread to older ages. This signal of inheritance is also seen in U-Pb zircon ages, but this is less evident or absent in Ar-Ar ages of co-existing glass. Ar diffusion modelling and Ar-Ar data in this study suggests sanidine is more likely to yield an eruption age. The use of sanidine feldspar, instead of anorthoclase and/or plagioclase feldspar is therefore strongly recommended. Biotite, which has shown to incorporate the largest proportion of 40ArE, should be used with caution, and successful dating of a glass phase can be a useful geochronological tool. Despite extraneous Ar contamination, the Ar-Ar dating technique can be successfully applied to the products of silicic volcanism. This work has provided the opportunity to determine new Ar-Ar eruption ages (Green Canyon Flow dome at 1.29±0.01 Ma; Sheridan Reservoir dome at 2.04±0.03 Ma, and Snake River Butte dome at 2.15±0.01 Ma), for rhyolite domes of the Yellowstone Plateau Volcanic Field previously only dated using the K-Ar technique. Finally, dating multiple phases (e.g., feldspar (sanidine); glass, and where possible zircon) is strongly recommended in order to identify cases of subtle contamination which will have a negative impact on our ability to obtain an accurate eruption age.