HR: 16:15h
AN: V34B-02    [Abstracts]
TI: Gas Flow in Volcanic Rocks: Lattice-Boltzmann Simulations of Porosity-Permeability Relationships
AU: * Llewellin, E W
EM: ed.llewellin@bristol.ac.uk
AF: BP Institute for Multiphase Flow, Department of Earth Sciences, University of Cambridge, Madingley Road, Cambridge, CB3 0EZ United Kingdom
AU: Blower, J D
EM: jdb@mail.nerc-essc.ac.uk
AF: Environmental Systems Science Centre, University of Reading, 3 Earley Gate, Reading, RG6 6AL United Kingdom
AU: Mantle, M D
EM: mdm20@cam.ac.uk
AF: Department of Chemical Engineering, University of Cambridge, Pembroke Street, Cambridge, CB2 3RA United Kingdom
AB: The flow of exsolved gases through the volcanic system exerts a fundamental control on the nature of a volcano's activity. Gas flow determines the internal pressure of magmatic bubbles and so strongly influences bubble growth kinetics and the process of magma fragmentation, both of which exert a first-order control on the magnitude and intensity of energy release during an eruption. Gas flows through connected networks of bubbles and cracks which may change dynamically. The complex and highly-variable nature of the network geometries has hitherto prevented the development of a general, quantitative description of the permeability of volcanic materials. We present a combined experimental--numerical investigation of the permeability of these networks comprising: 1) direct numerical simulation of gas flow through permeable networks using the lattice-Boltzmann method (LBM) and the finite element method (FEM); 2) direct imaging of fluid flow through permeable real and analogue volcanic rocks using magnetic resonance imaging (MRI). The permeability of a network is controlled by its connectivity and the permeability of its components. The direct numerical simulation techniques are used to determine the permeability of the network's textural components (e.g. cracks, ellipsoidal and spherical bubbles, bubble--bubble throats and bubble--crack intersections). The permeability of the whole network is then calculated using a network model analagous to those used to determine the conductance of complex electrical circuits. The numerical results are validated against experimental MRI velocimetric data. Our approach allows the permeability of magma and volcanic rocks of arbitrary texture to be determined rapidly and accurately. The texture may be derived from tomographic imaging of natural samples or may be generated artificially from theoretical considerations. This latter approach allows the permeability of pre-fragmentation magma to be predicted even when no natural samples exist.
DE: 8414 Eruption mechanisms and flow emplacement
DE: 8428 Explosive volcanism
DE: 8430 Volcanic gases
DE: 8434 Magma migration and fragmentation
SC: Volcanology, Geochemistry, Petrology [V]
MN: Fall Meeting 2005