- © 2013 by the Seismological Society of America
Intraslab earthquakes are generated by fault ruptures, which occur at shallow (20–60 km) or deep (>60 km) parts of a subducting tectonic plate slab. In both cases the rupture occurs in the vicinity of a zone with large velocity contrast and increased regional stress caused by the subducting plates. These conditions have significant effects on fault geometry, rupture dynamics, and seismic energy, which in turn affect the amplitude and duration of ground motion. Consequently, intraslab earthquakes can generate larger ground motion than shallow crustal earthquakes of same magnitude. Therefore their study is very important for strong ground motion prediction and seismic‐hazard assessment. In an effort to develop a methodology for simulating strong ground motion from intraslab earthquakes we tested the broadband ground‐motion simulation technique of Graves and Pitarka (2010) in modeling ground motion recorded from the M 6.5 2010 Ferndale, California, intraslab earthquake. The Graves and Pitarka (2010) method was originally developed for simulating crustal earthquakes. To our knowledge this is the first attempt to apply a broadband ground‐motion‐simulation method in modeling ground motion from intraslab earthquakes in the United States. A similar approach (Matsuzaki et al., 2010) was applied to studying the rupture process of the 2001 Geiyo, Japan, intraslab earthquake using the recipe developed by Irikura and Miyake (2006).
In this study we simulate and analyze ground‐motion‐acceleration time history recorded during the earthquake in a zone between Ferndale and Eureka, and Humboldt Bay. The earthquake was very well recorded in a geotechnical array located in the Humboldt Bay soft sediments. The ground‐motion acceleration recorded by the array’s instruments at several depths provided a good opportunity for testing numerical techniques and empirical models of the nonlinear seismic response of soils for input motion with moderate amplitude. A large part of our investigation is focused on modeling the observed nonlinear response …