Laboratory Measurement of Velocity vs. Effective Stress in Thrust Faults of the Oregon Accretionary Prism: Implications for Fault Zone Overpressure

Abstract

Seismic reflection profiles of many accretionary prisms, including the Oregon prism in the Leg 146 region, exhibit highamplitude, negative-polarity reflections from the decollement and other thrusts. Previous workers have suggested that these reflections image fault zones with enhanced fluid content due to dilation by very high fluid pressure. We collected samples within the frontal thrust zone and wall-rock sediments at Site 891, and from the out-of-sequence thrust and walls at Site 892 in order to evaluate this hypothesis experimentally. We present measurements of ultrasonic P-wave velocity as a function of confining pressure and fluid pressure (i.e., effective stress) on samples of the Oregon frontal thrust. These data show that (1) velocity decreases by 8% to 18% as fluid pressure rises from hydrostatic to lithostatic conditions under constant confining pressure; and (2) this effect occurs in fault zone and wall rock alike, with very similar velocity vs. effective stress relationships. We use these velocity vs. effective stress curves to calibrate the Site 891 frontal thrust reflection to fluid pressure. Synthetic waveform modeling of the frontal thrust reflection wavelet at Site 891 shows that it is the result of a thin low-velocity zone, probably less than 10 m thick, in the plane of the fault. Models employing relative amplitude scaling of the seafloor and fault zone reflectors indicate that the thin, low-impedance zone is 100-300 m/s slower than the walls, if the impedance contrast is dominated by velocity. Combining models with experimental results, we conclude that fluid pressure of 86% to 98% of lithostatic pressure reduces velocity enough to generate the reflections, if the impedance reduction of the fault zone is indeed due to fluid-pressure-induced unloading. Porosity data collected during testing of a sample from the frontal thrust zone indicates that little porosity change accompanies velocity reduction except at very high pore pressures (very low effective stresses), calling into question assumptions that low velocity necessarily implies high porosity or permeability in the fault zone.