Seattle Fault Zone - implications for earthquake hazards
ACTIVE TECTONICS OF THE SEATTLE FAULT AND CENTRAL PUGET SOUND, WASHINGTON IMPLICATIONS FOR EARTHQUAKE HAZARDS
The Seattle fault zone and the crosscutting north-trending strike-slip zone cut through a major urban center, so an understanding of the earthquake hazard is imperative. This investigation provides new constraints on the location, lengths, slip rates, and evolution of local crustal faults that can be incorporated in new seismic hazard maps and assessments (for example, Frankel et al., 1996). Our data are most consistent with a slip rate of 0.7 to 1.1 mm/year for the Seattle fault, but provide no information about recurrence interval. If the well-documented ~900 A.D. earthquake (Bucknam et al., 1992), which produced widespread uplift of about 5 to 7 m, is characteristic with a regular recurrence interval, then the next large earthquake on the Seattle fault may not occur for several thousand years. However, there is no data that argues for or against characteristic earthquakes on the Seattle fault, and Thorson (1996) has argued that the effects of glacial loading and unloading may have significantly disturbed the earthquake cycle in the Puget Lowland. Thus, a prediction of the timing of the next large event on the Seattle fault is impossible at this time. The estimated slip rates do suggest that there is presently about 75 to 120 cm of accumulated elastic deformation on fault "A".
Our results indicate that the Seattle fault does not form a continuous trace across the Puget Sound but rather is divided into two main segments by a crosscutting fault zone. This segmentation, however, does not appear to limit the rupture area for large earthquakes. Uplift of 5-7 m about 1100 yr ago has been documented at both Restoration Point along the western fault segment (Bucknam et al., 1992) and in the Duwamish River Valley along the eastern fault segment (J. A. Boughner and B.F. Atwater, oral commun., 1996). Although radiocarbon dating does not provide the resolution to conclusively determine if this uplift occurred as one large event or two closely spaced events on different fault segments, the relationships between fault displacement and fault rupture length defined by Wells and Coppersmith (1994; their Figures 12 and 13) support a lengthy fault rupture and the single-event hypothesis. Rupture across segment boundaries is also consistent with Rubin (1996), who showed that historic earthquakes generated by intracontinental reverse faults commonly rupture across surface discontinuities such as stepovers and cross faults. Rubin (1996) concluded that making assumptions that seismic rupture will initiate and terminate at major geometric irregularities can lead to an underestimation of the magnitudes of future large earthquakes. Conversely, the possibility of smaller, more frequent earthquakes on the Seattle fault zone with rupture limited to single fault segments cannot be ruled out.
Seismic hazard assessments in western Washington must now also incorporate the north-trending fault zone in Puget Sound as a potential earthquake source. Using the length for fault "1" defined above and relationships between rupture length and moment magnitude (M) defined by Wells and Coppersmith (1994; their Fig. 9), earthquakes of M ~ 6.5 are possible on this structure. Larger events could occur if the strands of this fault zone connect at depth.