Inversion of a velocity model using artificial neural networks

We present a velocity model inversion approach using artificial neural networks (NN). We selected four aftershocks from the 2000 Tottori, Japan, earthquake located around station SMNH01 in order to determine a 1D nearby underground velocity model. An NN was trained independently for each earthquake-station profile. We generated many velocity models and computed their corresponding synthetic waveforms. The waveforms were presented to NN as input. Training consisted in associating each waveform to the corresponding velocity model. Once trained, the actual observed records of the four events were presented to the network to predict their velocity models. In that way, four 1D profiles were obtained individually for each of the events. Each model was tested by computing the synthetic waveforms for other events recorded at SMNH01 and at two other nearby stations: TTR007 and TTR009.

[1]  Eiichi Fukuyama,et al.  NIED seismic moment tensor catalogue for regional earthquakes around Japan: quality test and application , 2002 .

[2]  William Hung Kan Lee,et al.  International handbook of earthquake and engineering seismology , 2002 .

[3]  Stefano Gresta,et al.  Minimum 1-D velocity model in Southeastern Sicily (Italy) from local earthquake data: an improvement in location accuracy , 2003 .

[4]  Chen Ji,et al.  Basin Structure Estimation by Waveform Modeling: Forward and Inverse Methods , 2000 .

[5]  M. Bouchon.  A simple method to calculate Green's functions for elastic layered media , 1981 .

[6]  Joey Rogers.  Object-oriented neural networks in C++ , 1996 .

[7]  Nelson PULIDO.  BROADBAND FREQUENCY ASPERITY PARAMETERS OF CRUSTAL EARTHQUAKES FROM INVERSION OF NEAR-FAULT GROUND MOTION , 2002 .

[8]  Anna Esposito,et al.  Discrimination of Earthquakes and Underwater Explosions Using Neural Networks , 2003 .

[9]  Jay J. Pulli,et al.  Regional seismic event classification at the NORESS array: Seismological measurements and the use of trained neural networks , 1990 .

[10]  William R. Normark,et al.  Geology of the continental margin beneath Santa Monica Bay, Southern California, from seismic-reflection data , 2003 .

[11]  Kim B. Olsen,et al.  Nonlinear dynamic rupture inversion of the 2000 Western Tottori, Japan, earthquake , 2004 .

[12]  Peter Goldstein,et al.  85.5 SAC2000: Signal processing and analysis tools for seismologists and engineers , 2003 .

[13]  Arben Pitarka,et al.  Three-Dimensional Finite-Difference Waveform Modeling of Strong Motions Observed in the Sendai Basin, Japan , 2001 .

[14]  Jack K. Odum,et al.  High-Resolution Seismic Reflection Surveys and Modeling Across an Area of High Damage from the 1994 Northridge Earthquake, Sherman Oaks, California , 2000 .

[15]  Colin MacBeth,et al.  The application of back‐propagation neural network to automatic picking seismic arrivals from single‐component recordings , 1997 .

[16]  Fabio Roli,et al.  Application of neural networks and statistical pattern recognition algorithms to earthquake risk evaluation , 1997, Pattern Recognit. Lett..

[17]  Albert Tarantola,et al.  Neural networks and inversion of seismic data , 1994 .