Dating individual zones in phenocrysts from the 2016–2017 eruption of Bogoslof volcano provides constraints on timescales of magmatic processes

We investigate the rates of magmatic processes using clinopyroxene diffusion chronometry on volcanic products erupted in August 2017 at the end of the 9-month eruption of Bogoslof volcano. The eruptive products contain plagioclase, clinopyroxene, and amphibole, all of which exhibit sharp chemical boundaries and are occasionally observed in multi-phase crystal clots with shared zoning boundaries across different mineral phases. At the shared boundaries in crystal clots, clinopyroxene and plagioclase continued to grow but abruptly changed composition from Mg# 81.7 +/- 5.8 to 72.9 +/- 3.0 and An82.5 +/- 1.4 to An61.3 +/- 5.7, respectively. Additionally, the sharp boundary marks where amphibole became unstable and began forming a reaction rim. Synthesizing these observations, we were able to determine that the shared boundaries formed as a result of rapid decompression during magma ascent, followed by storage in a shallow cryptodome, where magma accumulated prior to erupting.In order to determine the timescales of magma ascent and subsequent crystal residence times, we applied diffusion chronometry on zoned clinopyroxene phenocrysts using Mg# concentrations at 1056 degrees C determined from Fe-Ti oxide pairs. Our diffusion modeling results show that diffusion began at the stepwise boundaries in clinopyroxenes no more than180 days before the final explosive event. These results were then used to calculate crystal growth rates for shared plagioclase and amphibole rims, as shared zones in crystal clots indicate that the boundaries in all three phases formed contemporaneously. We calculate growth rates of plagioclase crystals (1.7 +/- 0.99 x 10-6 um/s) and amphibole reaction rims (2.8 +/- 0.47 x 10-6 um/s). The calculated natural growth rate of plagioclase was then used to constrain additional magmatic timescales from growth rate chronometry, results of which support our diffusion timescales.Our results indicate that the distinct boundaries in all three mineral phases formed due to ascent-driven decompression followed by shallow emplacement of mafic magma that occurred continually throughout the course of the eruption. By subtracting diffusion timescales from the date that the samples were erupted, the oldest crystal boundaries correspond to March 2017, seemingly correlating with increases in both seismicity and SO2 emissions. These observations may suggest that our petrochronometric results are supported by interdisci-plinary observations.