No abstract

Slip on unfavourably oriented faults with respect to a remotely applied stress is well documented and implies that faults such as the San Andreas fault and low-angle normal faults are weak when compared to laboratory-measured frictional strength. If high pore pressure within fault zones is the cause of such weakness, then stress reorientation within or close to a fault is necessary to allow sufficient fault weakening without the occurrence of hydrofracture. From field observations of a major tectonic fault, and using laboratory experiments and numerical modelling, here we show that stress rotation occurs within the fractured damage zone surrounding faults. In particular, we find that stress rotation is considerable for unfavourably oriented 'weak' faults. In the 'weak' fault case, the damage-induced change in elastic properties provides the necessary stress rotation to allow high pore pressure faulting without inducing hydrofracture.

[1] The strength and permeability of fault zones must be quantified in order to accurately predict crustal strength and subsurface fluid migration. To this end, we performed experiments on mixtures of fine-grained quartz and kaolinite incremented at 10 wt% intervals between the two end-member components (analogues for natural fault gouge) in order to establish their strength and fluid flow properties during hydrostatic and shear loading. Hydrostatically compacted samples exhibited permeability reduction on increasing effective pressures from 5 MPa to 50 MPa, with the rate of reduction displaying strong dependency on the synthetic fault gouge composition. The permeability decreases continuously with increasing kaolinite content. Porosity exhibits a distinct minimum that evolves with increasing effective pressure according to the relative compaction of the quartz and kaolinite end-members. Porosity evolution with increasing clay content is predicted satisfactorily by a simple ideal packing model. At the highest effective pressure (50 MPa), permeability reduced log-linearly over 4 orders of magnitude with increasing clay content. Mechanically, sheared gouge samples showed a continuous reduction in frictional strength with increasing clay fraction. Permeability decreased further on shear loading after initial hydrostatic compaction to 50 MPa. This was most evident for the pure quartz end-member, with two orders of magnitude additional reduction, whereas the clayrich samples were reduced only tenfold, mostly before a shear strain of 5. Variation of permeability with both clay content and shear deformation may be adequately described by previously published empirical predictors for fault zone permeability. Clay content has the largest effect on permeability, and shear deformation affects permeability of quartzrich gouges more than clay-rich gouges.Citation: Crawford, B. R., D. R. Faulkner, and E. H. Rutter (2008), Strength, porosity, and permeability development during hydrostatic and shear loading of synthetic quartz-clay fault gouge,

Abstract. Quantification of fluid transport through fault zones is critical for the understanding of fault mechanics and prediction of subsurface fluid flow. The permeability of clay-bearing fault gouge has been determined using first argon then water as pore fluids under total confining pressures ranging up to 200 MPa and pore pressures of 40 MPa at room temperature. Use of the two pore fluids allows interactions between the gouge and pore fluids to be examined. Natural clay-bearing fault gouge recovered from surface exposures of the Carboneras fault zone in southeastern Spain was used and was collected in such a way that the in situ microstructure was preserved. Cores were collected in directions relative to the well-developed planar fabrics seen in these types of fault rock. The mineralogy of the gouges included muscovite/illite, chlorite, and quartz, with minor amounts of gypsum, albite, and graphite. Glycolation of the gouge showed no discernible amounts of swelling phases. Grain size analyses revealed a bimodal grain size distribution, with the <2 •xm fraction dominant (>50 wt %). This fraction contained predominantly 17 2 22clay phases. Permeabilities in the range of 10-m to 10-m 2 were measured. Experimental results show that the previous highest in situ effective pressure to which the fault gouge had been subjected (overconsolidation pressure) could not be determined from changes in permeability. Differences between water and argon permeabilities determined on the same sample amounted to -•1 order of magnitude, even if the sample had been pressure cycled (reduction to zero and reapplication of both confining and pore pressure) using argon as pore fluid until asymptotic values for permeability had been attained. Volumetric strain measurements showed no enhanced compaction due to the introduction of water as the pore fluid, leading to the conclusion that the reduction in permeability must be due to physicochemical interactions of the water with the fault gouge. The low permeabilities measured support models invoking high fluid pressure weakening of large faults with minimal fluid loss. The stability of structured water films with varying temperature, water pressure and water chemistrv may produce a heterogeneous permeability profile with depth in fault zones.