Source inversion of the 1987 Whittier Narrows earthquake, California, using the isochron method

Abstract

Abstract We determine the earthquake source process from an isochron synthetic method and the recursive stochastic inverse algorithm. Using the strong-motion records from 16 stations within 15 km of the epicenter, we determined the earthquake slip and rupture time distribution on the fault with different fault geometries for the Whittier Narrows earthquake. In our study, we have chosen two best-fit fault geometries, with strike direction and dip angle of 270° and 35°, and 280° and 30°, respectively. The inversion results indicated dominantly two regions of slip on the fault plane for the Whittier Narrows mainshock, one located immediately below the hypocenter and the other near the lower east corner of the fault. In addition, there are three or four other smaller localized source areas near the hypocenter. Together, they indicate a highly heterogeneous fault slip distribution for the Whittier Narrows mainshock. The great majority of seismic moment release occurs in a 10 by 4 km region, which is consistent with the geodetic modeling results. The inversion result for the model allowing variable slip and rupture time model gave the best fit in terms of both the Euclidean norm and the weighted Euclidean norm, suggesting that the rupture velocity may be not constant. But we find no simple causal relation between these rupture velocity variations and the slip amplitude distributions. On the other hand, by comparing the mainshock slip pattern of our inversion results with the early aftershock, we found that the aftershocks tend to occur over patches on the rupture surface deficient in mainshock slip. This observation can be well explained by the barrier model of fault heterogeneity.