Near-lithostatic pore pressure at seismogenic depths: A thermoporoelastic model

Abstract

A model is presented for pore pressure migration through a transition layer separating a meteoric aquifer at hydrostatic pressure from a deeper reservoir at lithostatic pressure. This configuration is thought to be pertinent to the South Iceland seismic zone (SISZ) and to other tectonically active regions of recent volcanism, where volatiles are continuously released by ascending magma below the brittle-ductile transition. Poroelastic parameters are computed for basaltic rock. The model is 1-D, the fluid viscosity is temperature dependent and rock permeability is assumed to be pressure dependent according to a dislocation model of a fractured medium. Environment conditions are considered, pertinent to basalt saturated with water at shallow depth (case I) and at mid-crustal depth (case II). If the intrinsic permeability of the rock is high, no significant effects are observed in the pressure field but advective heat transfer shifts the brittle-ductile transition to shallower depths. If the intrinsic permeability is low, the pressure-dependent permeability can propagate near-lithostatic pore pressures throughout most of the transition layer, while the temperature is practically unaffected by advective contributions so that the rock in the transition layer remains in brittle condition. Geometrical parameters characterizing the fracture distribution are important in determining the effective permeability: in particular, if an interconnected system of fractures develops within the transition layer, the effective permeability may increase by several orders of magnitude and near-lithostatic pore pressure propagates upwards. These modelling results have important bearings on our understanding of seismogenic processes in geothermal areas and are consistent with several geophysical observations in the SISZ, in connection with the two 2000 June M = 6.5 earthquakes, including: (i) fluid pressure pulses in deep wells, (ii) low resistivity at the base of the seismogenic layer, (iii) low VP/VS ratio and time-dependent seismic tomography, (iv) heterogeneity of focal mechanisms, (v) shear wave splitting, (vi) high b-value of deep foreshocks, (vii) triggered seismicity and (viii) Radon anomalies. © 2006 The Authors Journal compilation © 2006 RAS.