HR: 11:15h
AN: T42B-04 INVITED     [Abstracts]
TI: Some Recent Laboratory Measurements of Fault Zone Permeability
AU: * Morrow, C A
EM: cmorrow@usgs.gov
AF: U.S. Geological Survey, 345 Middlefield Rd., Menlo Park, CA 94025 United States
AU: Lockner, D A
EM: dlockner@usgs.gov
AF: U.S. Geological Survey, 345 Middlefield Rd., Menlo Park, CA 94025 United States
AB: The permeability of fault zone material is key to understanding fluid circulation and the role of pore fluids in earthquake generation and rupture dynamics. Permeability results of core samples from several scientific drillholes are presented, including new results from the SAFOD drillsite in California and the Chelungpu Fault in Taiwan. Permeability values at simulated in situ pressures range from 10^-18 to 10^-23 m², a broad range reflecting differences in rock type, proximity to the fault (i.e., fault core, damage zone or country rock), and degree of interseismic healing and sealing. In addition to these natural characteristics, stress-relief and thermal cracking damage resulting from core retrieval will tend to increase the permeability of some of the deepest crystalline rock samples, although testing under in situ conditions can reduce these errors. Recently active fault rocks, with an interconnected network of fractures, tend toward the higher end of the permeability range, whereas fault rocks that have had time to heal through hydrothermal processes tend to have lower permeabilities. In addition, the permeability of borehole-derived core samples was found to be more sensitive to applied pressure than equivalent rocks obtained from surface outcrops because of weathering and other processes. Thus, permeability values of surface samples can not be adequately extrapolated to depth, highlighting the importance of deep drilling studies in determining in situ transport properties. Permeability studies also reveal the storage capacity of the fault rocks, an important parameter in the determination of excess fluid pressure potential. Storage capacity was found to be 10^-10 to 10^-11/Pa in the Chelungpu Fault cores. Typical down-hole permeability measurements are generally 1-2 orders of magnitude higher than laboratory-derived values because they sample joints and fractures in the damage zone that are larger in scale than the core samples. Consequently, most fluid flow at depth is likely to occur through these larger-scale features rather than through the matrix of the rock. Then a `typical' fault structure might consist of a narrow, low-permeability fault zone surrounded by a higher permeability damage zone, in which the damage zone acts as a conduit for fluids parallel to the fault. The low-permeability, often clay-rich shear zone will impede flow across the fault. Quartzo-feldspathic lithologies such as those of the Nojima Fault in Japan, fit this scenario. However, such a fault model may not be appropriate in many cases, particularly in sedimentary sequences. For instance, there may be little difference in the permeability of the fault-rock and surrounding rock in clay-rich sandstones or siltstones, such as those of the Chelungpu Fault in Taiwan. For interbedded sandstones and siltstones sequences such as the Moab Fault in Utah, clay-rich layers may be entrained in the fault zone, forming a continuous barrier to fluid flow (fault gouge) across the fault, but with very different damage zone/country rock evolution and hydrologic properties.
DE: 8034 Rheology and friction of fault zones (8163)
SC: Tectonophysics [T]
MN: Fall Meeting 2005