GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 89-5
Presentation Time: 9:00 AM

AGE OF ZIRCON: WHAT IS REALLY BEING RECORDED IN INTERMEDIATE-TO-SILICIC ARC VOLCANIC ROCKS? (Invited Presentation)


KLEMETTI, Erik W.1, KENT, Adam J.R.2, CLYNNE, Michael A.3, BERTOLETT, Elisabeth4, HERNANDEZ, Lindsey D.5 and HERROLD, Emily N.1, (1)Geosciences, Denison University, 100 W College St., Granville, OH 43023, (2)College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, OR 97331, (3)Volcano Hazards Team, USGS, Menlo Park, CA 94025, (4)Geological Sciences, University of Canterbury, Christchurch, 8041, New Zealand, (5)Geosciences, Denison University, Granville, OH 43023

Zircon are frequently found in intermediate-to-silicic arc volcanic rocks, even when zircon saturation suggests it is not crystallizing. There are a variety of zircon sources in arc volcanoes: crystallizing in active melt, inclusions in phenocrysts, crystal mush from antecedent magma or foreign cumulates and xenoliths. Each represent a different episode of the erupted magma’s evolution. Here we use temporal and composition data from zircon and zircon-hosted melt inclusions from the Cascade Range to understand the evolution of intermediate-to-silicic arc magma and what zircon is and isn’t recording.

The Lassen Volcanic Center (LVC), Mt. Hood, Mount St. Helens, Crater Lake, Medicine Lake and South Sister all contain zircon that predate the eruption age by 103-105years. Hood and St. Helens contain zircon that may have crystallized within error of the eruption age of the host lava. LVC and Hood zircon Hf, Eu/Eu* and Yb/Gd composition varies over time, reflecting changing magmatic t-xconditions, and exhibits compositional clustering that represents different crystallization conditions. However, considering the uncertainty of the analyses, interpretation of zircon U-Th age-compositional trends can be difficult (Kent and Cooper, 2017).

Zircon-hosted melt inclusions and adhered glass from the LVC are high-silica rhyolite, frequently exceeding the silica content of their host. The lone exception is zircon from the Rockland tephra, the only known large-volume explosive eruption from the LVC, where melt inclusion composition match the host magma. This implies that zircon is forming in a fractionated, cooling crystal mush that is zircon saturated. This mush (and host melt) may be sampled during the last stages before an eruption.

The zircon data suggests that intermediate-to-silicic arc magma is an amalgam of rhyolitic melt + mafic melt + intermediate phenocrysts + silicic antecrysts. In some cases, the rhyolitic mush behaves as a filter (e.g, Hood, Kent et al., 2010) creating uniform hybrid andesite. In others, it mingles with intruding magma (e.g., LVC, Klemetti and Clynne, 2014) creating abundant quenched inclusions. Zircon record evidence of high-silica mush during periods of active eruption and relative volcanic quiescence and its small size allows it to be readily transported and erupted.