HR: 08:30h
AN: S51B-03    [PDF]
TI: Fault Roughness and Matedness Suggest Significant Fault-Interface Dilatancy With Slip
AU: * Junger, J A
EM: Jennifer_Junger@brown.edu
AF: Brown University, Department of Geological Sciences, Providence, RI 02912-1846 United States
AU: Tullis, T E
EM: Terry_Tullis@brown.edu
AF: Brown University, Department of Geological Sciences, Providence, RI 02912-1846 United States
AB: Pore pressure in a fault zone could increase during rapid coseismic slip due to thermal pressurization of pore fluid. However, the nearly self-similar topography of natural fault surfaces combined with their high degree of matedness suggests that a considerable increase in the volume of the space between opposing fault surfaces should occur with slip. Thus, pore pressure is likely to decrease rather than increase, and dilatancy hardening rather than thermal pressurization weakening seems a likely consequence of slip. To quantitatively explore how the gap between opposing fault surfaces could depend on slip, we have analyzed profilimetry data from a large well exposed normal fault in Dixie Valley, Nevada (Power and Tullis, 1992). The difference between the topographies of the presently well-mated fault surfaces was used to determine a progression of composite topographies as a function of offset of the surfaces in the paleo-slip direction. We assume that only the points of closest contact touch at each offset and that the remainder of the composite topography would constitute a gap between the surfaces. The average of this gap corresponds to a volume per area of the fault surface. As expected, this average gap increases with displacement and represents an increasing volume with slip. For the one fault surface analyzed to date, the volume increases by a factor of 7 after 10 mm of slip, an amount that would completely overwhelm any thermal expansion of the pore fluid, and, in fact, should cause pore pressure to decrease to nearly zero, assuming no flow of fluid into the interface. It seems reasonable to assume that fault surfaces do not remain mated during coseismic slip, because the deformation processes that cause matedness appear to involve pressure solution, and therefore the matedness probably occurs during slow interseismic creep (Power and Tullis, 1989). However, assuming that the opposing walls remain rigid during coseismic slip is presumably incorrect, because cataclastic processes could occur. Thus, our calculated increase of interfacial volume is presumably an upper bound. On the other hand, although cataclasis could cause the increase in volume along the interface to be smaller than our upper bound, significant overall dilatancy is still expected since the fragments produced by cataclasis will not fit together perfectly. It is difficult to make quantitative estimates of the likely volume increase in the presence of such cataclasis, but the dilatancy we calculate due to slip of mated surfaces is so large that it brings into serious question the likelihood that thermal pressurization could weaken the fault during coseismic slip; dilatancy hardening appears more likely.
DE: 7209 Earthquake dynamics and mechanics
DE: 7215 Earthquake parameters
DE: 7260 Theory and modeling
DE: 8159 Rheology--crust and lithosphere
DE: 8164 Stresses--crust and lithosphere
SC: Seismology [S]
MN: 2003 Fall Meeting