H2O-CO2 DEGASSING IN RHYOLITIC ERUPTIONS AND IMPLICATION OF MAGMA ASCENT HISTORY

Magma ascent rate is a key parameter in characterizing volcanic eruption style, tephra dispersion, and volcanic atmospheric impact. Many methods, which involve seismic activity, phenocrysts breakdown, volatile exsolution, H₂O diffusion profiles in melt inclusions, xenolith transport, and U-Th-Ra isotopic disequilibrium, have been employed to investigate volcanic eruptions rate. Liu et al. (2005) developed an empirical model of mixed H₂O and CO₂ solubility for natural rhyolitic melt at a condition of 700—1200 ^oC and 0--500 MPa. Combining this solubility model with diffusivities of both H₂O and CO₂ from Behrens and Zhang (2009) and Zhang et al. (2007), we conducted thousands of numerical simulations on H₂O and CO₂ diffusion in a bubble/melt system during rhyolitic syn-eruptive eruptions. Using these homogeneous bubble growth model based on the model for H₂O of Forestier-Coste et al. (2012), we investigate the effects of the integrated decompression time, various non-linear decompression paths, and different initial CO₂/H₂O content on the H₂O and CO₂ concentration profiles around bubbles within the melt after the eruption. Our results show that for all conditions, H₂O diffusion profile is much more flat than that of CO₂, which is consistent with the fact that CO₂ diffusivity is lower than H₂O. The curvatures of CO₂ diffusion profiles reach a maximum at moderate decompression rates. For a given average decompression rate but a different decompression paths, CO₂ diffusion profile can present widely different average curvatures. Based on our simulations, we propose that the evaluation of post-eruption H₂O and CO₂ diffusion profiles in the melt of different samples could provide constraints beyond the average decompression rate and better reconstruct the eruption dynamics.