Sensitivity of Synthetic Seismograms for Different Seismic Scenarios in North Chile

Abstract eng:
This research studies the sensitivity of spectral response values to various physical earthquake scenario parameters, the latter used to generate synthetic low frequency seismograms in North Chile. Ten earthquake scenarios have been defined using seed information from the slip model of the 2010, Maule earthquake, and different physically plausible interplate locking models in the region. Firstly, the Maule 2010 finite fault rupture model was resituated in the existing seismic gap in north Chile using a curved geometry according to the Slab 1.0 model. From this seed model, one synthetic scenario with constant moment magnitude Mw 8.8 was generated with the same slip distribution as the original 2010 slip model. Three other models with variations in the slip distribution were considered. In addition, three physically plausible fault rupture models based on previous studies of interplate locking were used. Each of these scenarios was capable of generating Mw 8.2 – 8.4 earthquakes with a maximum slip of 7.5 m, approximately. Patches of major slip were located along the coast line approximately in front of the cities of Arica, Iquique, and Tocopilla, respectively. Also, three additional scenarios with moment magnitudes in the range Mw 8.5 – 8.7 were built by connecting these physical scenarios into larger rupture areas. These combined interplate locking models represented the activation of two or more asperities, similar to the experience of the 2010 Maule earthquake. Using these scenarios we built low frequency synthetic seismograms at four control sites: Arica, Iquique, Tocopilla, and Calama. The sensitivity of these results was studied by deterministic analyses on some key rupture parameters, such as mean rupture velocity and slip rise-time. Sensitivity analysis used peak ground displacement (PGD) and acceleration (PGA), pseudo-acceleration spectra, Sa, and displacement spectra, Sd. The range of values considered for mean rupture velocity was vr = 2.2-3.0 km/s. Four points were considered in the vicinity of each specified velocity to compute sensitivities. First and second order derivatives of PGD, PGA, Sd, and Sa relative to the source parameters were then used to build a Taylor series expansion to predict PGD, Sa and Sd as a function of vr. This allows to consider uncertainty in this parameter and propagate such uncertainty into spectral response values. An analogous procedure was considered for rise-time tr in the range from 2 to 10s.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket