Constraining earthquake source inversions with GPS data 2: A two-step approach to combine seismic and geodetic datasets

Susana Custodio, Morgan T. Page, & Ralph J. Archuleta

Published 2009, SCEC Contribution #1125

We present a new method to combine static and wavefield data to image earthquake ruptures. Our combined inversion is a two-step procedure that takes into account the different resolutions of each dataset. The first step consists of an inversion of the static field, which yields a map of slip amplitude. This inversion exploits a special irregular grid that takes into account the resolution of the static data (Page et al., companion paper). The second step is an inversion of the radiated wavefield; it results in the determination of the complete time evolution of slip on the fault. In the second-step inversion, the slip amplitude is constrained to resemble the static slip amplitude map inferred from the GPS inversion. We use our two-step combined inversion to study the source process of the 2004 M6 Parkfield, California, earthquake. We conclude that slip occurred in two main regions of the fault, each of which displayed distinct rupture behaviors. Slip initiated at the hypocenter with a very strong bilateral burst of energy. Here, slip was localized in a narrow area (approximately 10~km long), the rupture velocity was very fast (>=3.5 km/s) and slip only lasted a short period of time (<1 sec). Then the rupture proceeded to a wider region 10-20 km northwest of the hypocenter. Here, the earthquake developed in a more moderated way - the rupture velocity slowed to ~3.0 km/s and slip lasted longer (1-2sec). The maximum slip amplitude in both regions was 0.4 m.

Citation
Custodio, S., Page, M. T., & Archuleta, R. J. (2009). Constraining earthquake source inversions with GPS data 2: A two-step approach to combine seismic and geodetic datasets. Journal of Geophysical Research: Solid Earth,. doi: 10.1029/2008JB005746.