In this paper we examine the consequences of bubble nucleation mechanism on eruptive degassing of rhyolite magma. We use the results of published high temperature and pressure decompression experiments as input to a modified version of CONFLOW, the numerical model of Mastin and Ghiorso [(2000) U.S.G.S. Open-File Rep. 00-209, 53 pp.] and Mastin [(2002) Geochem. Geophys. Geosyst. 3, 10.1029/2001GC000192] for steady, two-phase flow in vertical conduits. Synthesis of the available experimental data shows that heterogeneous nucleation is triggered at ??P <5-20 MPa in water-saturated rhyolite and leads to equilibrium degassing through a discrete nucleation event. Typically 105-107 bubbles/cm3 are produced which evolve Gaussian bubble size distributions. Homogeneous nucleation requires ??P >120-150 MPa, and leads to disequilibrium degassing at extreme H2O supersaturation. In this latter case, nucleation is an ongoing process controlled by changing supersaturation conditions. Exponential bubble size distributions are often produced with number densities of 106-109 bubbles/cm3. Our numerical analysis adopts an end-member approach that specifically compares equilibrium degassing with delayed, disequilibrium degassing characteristic of homogeneously-nucleating systems. The disequilibrium simulations show that delaying nucleation until ??P =150 MPa restricts degassing to within ???1500 m of the surface. Fragmentation occurs at similar porosity in both the disequilibrium and equilibrium modes (???80 vol%), but at the distinct depths of ???500 m and ???2300 m, respectively. The vesiculation delay leads to higher pressures at equivalent depths in the conduit, and the mass flux and exit pressure are each higher by a factor of ???2.0. Residual water contents in the melt reaching the vent are between 0.5 and 1.0 wt%, roughly twice that of the equilibrium model. ?? 2003 Elsevier B.V. All rights reserved.
Additional Publication Details
Gas evolution in eruptive conduits: Combining insights from high temperature and pressure decompression experiments with steady-state flow modeling