Back to the deformation and Stress Change Modeling home page

Back to the team's online recent papers pages

Non-technical summary: The Hayward Fault crosses the east side of San Francisco Bay, passes beneath San Pablo Bay and emerges again as the Rodgers Creek Fault. However, the fault line ‘steps over’ about 5-6km to the east. Due to the fault being located beneath the bay it is very difficult to map the fault structure. Parsons et al., (2003) aimed to assess how continuous the step over is and whether there are any connecting faults forming part of the step over. If the step is discontinuous, then an earthquake wouldn’t be able to rupture through, limiting the maximum magnitude of a Hayward-fault earthquake. However if an earthquake could rupture through the step easily, then in theory the entire length of the Rodgers Creek Fault could be added onto the Hayward Fault. San Pablo Bay presents several problems to Earth scientists; the faults are submerged, making them impossible to map directly. The water depth is too shallow to conduct conventional marine remote sensing techniques. Therefore an array of techniques was used to try and understand the step over. Seismic data were reanalyzed to create a velocity structure across the bay where the two faults are offset (down to a depth of 2km). Gravity and magnetic information was used so Parsons et al., (2003) could assess the structure and shape of the basin which formed as a result of the step over. The geometry of the basin could be used to indicate how continuous or discontinuous the fault is. The gravitational data could be used to compare subtle differences in the earth’s gravitational field. If the sediments of the basin are poorly consolidated, than the gravitational signal would be weaker compared to dense material such as granite. Another clue in the region was the occurrence of an earthquake in 1898 called the Mare Island earthquake, which happened somewhere under San Pablo Bay. If it could be determined which fault caused the earthquake, Parsons et al., (2003) might learn more about the mechanics of the step over. During the late 1800’s, bathymetric studies were carried out both before and after the earthquake, allowing an assessment of the change in shape of the bay floor. A small tsunami was also associated with the earthquake, which was also modeled. After considering all these parameters Parsons et al., (2004) concluded that the faults do not join. They probably do not join at depth but they do approach more closely to each other (about 4km away) than was previously thought. The conclusion of the study is that it is possible for the Rodgers Creek Fault and Hayward Fault to rupture a large earthquake if both faults were highly stressed.

Figure 3. Fault map of San Pablo Bay based on the cross sections of Wright and Smith (1992). Click on image for the whole pdf article or on the link above.