Registro de resúmenes

Reunión Anual UGM 2023


SIS-35

 Resumen número: 0060  |  Resumen aceptado  
Presentación en cartel

Título:

FINITE-FAULT STOCHASTIC GROUND MOTION SIMULATIONS OF THE 12 APRIL 2012 MW 7.0 GULF OF CALIFORNIA MEXICO EARTHQUAKE

Autores:

1 Juan Manuel Azua Flores EDPonente
Centro de Investigación Científica y de Educación Superior de Ensenada, CICESE
juanmazuaf@gmail.com

2 Raul Ramon Castro
Centro de Investigación Científica y de Educación Superior de Ensenada, CICESE
raul@cicese.mx

Sesión:

SIS Sismología Sesión regular

Resumen:

On April 12, 2012, an earthquake with magnitude Mw7.0 occurred at the central part of the Gulf of California (GoC) Mexico, where the North American (NA) and Pacific (PA) tectonic plates form a dextral trans-tensional tectonic plate boundary. Finite fault stochastic simulation algorithms simulate strong ground motions from large earthquakes by summing simulated point-source stochastic ground motion contributions of smaller subdivided cells. This procedure allows the simulation of source-related effects such as rupture directivity. However, finite-fault stochastic simulation algorithms require site-specific calibration to adequately simulate ground motions. Furthermore, only a few recordings of the 12 April 2012 Mw7.0 central GoC earthquake are available at local and regional distances. Therefore, the aim of this study is to calibrate the source parameters using the finite-fault stochastic ground motion simulation algorithms FINSIM (Beresnev and Atkinson, 1998) and EXSIM (Motazedian and Atkinson, 2005) to local and regional S-wave ground motion observations of this large earthquake. To calibrate the algorithms, we used recordings from two stations with the smallest source-site distance (i.e., stations SRIG and NE80, located at 190.6 and 203.1 km, respectively). We used a grid search approach to look for the source parameters that minimized the peak ground acceleration, acceleration Fourier amplitude spectra and acceleration response spectra residuals between observations and simulations. We also visually compared the ground motions duration. We found that a stress drop of 22 bars and a strength factor value of 0.9 best calibrated FINSIM while a stress drop of 27.5 bars, slip percentage of 15%, subfault length of 3.4 km and subfault width of 3.8 km minimized the EXSIM residuals. Then we estimated the S-wave quality factor Qs for ray paths between the studied event and stations with the next smaller source-site distances (i.e., HSIG, SQX, PPXB and SPIG, located at a distance range between 216.6 to 331.7 km), using the calibrated source-parameters and both FINSIM and EXSIM codes. We found both simulation codes produce similar Qs-values, with smaller Qs-values for ray paths that cross the tectonic limit between NA and PA plates, compared with ray paths that cross mainly continental crust. We conclude that both FINSIM and EXSIM codes may be useful on finite-fault stochastic ground motion simulation since they produce similar results. Our results advance on the understanding of the seismic risk and hazard of the areas surrounding the central-north GoC since calibrated EXSIM and FINSIM finite-fault algorithms may be used to estimate strong-ground motion at areas where no recordings from similar large earthquakes exist while the Qs factors also advance on our understanding of S-wave attenuation in the area.





Reunión Anual UGM 2023
29 de Octubre al 3 de Noviembre
Puerto Vallarta, Jalisco, México