T23C-2965
Detailed Geophysical Imaging in San Pablo Bay Reveals a New Strand of the Hayward-Rodgers Creek Fault Zone

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Janet Tilden Watt1, David A Ponce2, Patrick E Hart3, Kevin M Denton4, Tom Parsons4 and Russell Walter Graymer4, (1)U.S. Geological Survey, Santa Cruz, CA, United States, (2)USGS California Water Science Center Menlo Park, Menlo Park, CA, United States, (3)Pacific Coastal and Marine Science Center Santa Cruz, Santa Cruz, CA, United States, (4)U.S. Geological Survey, Menlo Park, CA, United States
Abstract:
High-resolution chirp seismic-reflection and marine magnetic data collected in San Pablo Bay reveal a new strand of the Hayward fault that helps constrain the geometry and connectivity of the Hayward-Rodgers Creek fault zone, one of the most hazardous faults in California. Over 1,200 km of marine magnetic data were collected in San Pablo Bay along NE-trending traverses spaced 200-m apart, and approximately 200 km of chirp data were collected along similarly oriented profiles spaced 1-km apart. Data were acquired using a 0.7-12 kHz sweep with a 20 ms length fired at 6 times per second. Due to attenuation of the acoustic signal by bay muds and persistent natural gas layers in San Pablo Bay, chirp data are only able to image the upper 2 to 5 meters of the sub-seafloor.

Offset and warping of near-surface reflections delineates a previously unrecognized NW-trending strand of the Hayward fault that extends across San Pablo Bay, from Point Pinole to Lower Tubbs Island. Vertical offset along the fault varies in both direction and magnitude, with some indication of increasing offset with depth. The fault imaged in the chirp data corresponds to gravity, magnetic, and tomographic gradients in the bay. Relocated seismicity is aligned with the surface trace of the fault and repeating earthquakes along this trend suggest this strand of the Hayward fault is creeping. A northwestward onshore projection of this fault is coincident with gravity and topographic gradients that align with a SSE-trending splay of the Rodgers Creek Fault, suggesting the Hayward and Rodgers Creek faults may connect directly rather than through a wide step-over zone. Even if the faults do not directly connect, these new data indicate that the faults are much closer together (2 km vs 4 km) than previously thought, making a through-going rupture more plausible.