HR: 14:00h
AN: T43C-02    [Abstracts]
TI: Crustal Dehydration and Overpressure Development on the San Andreas Fault
AU: * Fulton, P M
EM: pfulton@geosc.psu.edu
AF: Dept. of Geosciences, Pennsylvania State University, 503 Deike Building, University Park, PA 16802
AU: Saffer, D M
EM: dsaffer@geosc.psu.edu
AF: Dept. of Geosciences, Pennsylvania State University, 503 Deike Building, University Park, PA 16802
AU: Bekins, B A
EM: babekins@usgs.gov
AF: U.S. Geological Survey, 345 Middlefield Rd, Menlo Park, CA 94025
AB: Previous authors have hypothesized that the apparent weakness of the San Andreas Fault may be explained by fluid overpressures resulting from the combination of crustal dehydration of the Franciscan mélange and the presence of a low-permeability serpentinite cap at its geologic contact with the Great Valley Sequence. We previously evaluated this hypothesis by calculating the spatial and temporal distribution of fluid sources and then incorporating these sources in 2-D models of fluid flow and heat transport perpendicular to the fault. We have refined our fluid source calculations using theoretical values of whole-rock H₂O content and PT histories for the Franciscan crust in the wake of northward migration of the Mendocino Triple Junction (MTJ). The sources obtained reach peak values of 10^-16 s^-1. The coupled fluid flow and heat transport model now accommodates large-scale crustal deformation in a more rigorous manner by constructing new model grids after each change in crustal thickness. In the models, we assign permeability of the crust as a function of depth. A 500-m-thick, low-permeability serpentinite body (k=10^-20 m^-2) extends across the eastern half of the 50 km-wide model domain at a depth of 2 km. In addition, various model simulations include fault structures centered in the model domain such as: a 500 m wide low permeability fault barrier (k_fault = k_crust/100), a fault conduit (k_fault = k_crust x 100), a barrier within a 1.5 km wide conduit damage zone, and a conduit plugged by a 3 km-thick and 2 km-wide barrier simulating a broad, clay-rich, low-permeability zone, at shallow depth within the fault system, which is one possible interpretation of seismic and electromagnetic data. We also test additional scenarios to evaluate sensitivity to changes in model permeability. Model results show overpressures, as large as 162% of hydrostatic (62% of lithostatic) for the model with a serpentinite cap and fault barrier, develop within 4 Ma of Mendocino Triple Junction (MTJ) passage. The high pressure region extends ~200 km south of the present day location of the MTJ, where sources are abundant. However, beyond this region sources cease due to stabilization in temperature and low H₂O content remaining in lower crustal rocks, and overpressures dissipate within thousands of years. Overall, the presence of a low-permeability cap is shown to greatly enhance regional overpressure development; however, none of the structures were capable of localizing overpressures on the fault plane. In fact, even with sources across the entire model domain, overpressures were generally larger east of the fault than on the fault plane itself. Our results suggest that metamorphic dehydration within the Franciscan crust has the potential to regionally weaken the crust for a period of ~4 Ma after MTJ passage, but is limited as a sole mechanism for long-term weakening of the San Andreas Fault along much of its length, because overpressures cannot be sustained.
DE: 1847 Modeling
DE: 8106 Continental margins: transform
DE: 8163 Rheology and friction of fault zones (8034)
SC: Tectonophysics [T]
MN: Fall Meeting 2005