Assessment of the Source Characteristics of Stochastic Finite-Fault Simulations

by Emrah Yenier

Department of Earth Sciences, University of Western Ontario

The characteristics of the apparent source spectrum of stochastic finite-fault simulations are evaluated to identify if the simulated ground motions exhibit “spectral-sag” at intermediate frequencies as a result of finite fault rupture. The finite-fault simulations are performed for various magnitude, stress-drop and distance bins. The Fourier acceleration spectra of the simulations are regressed for path effects in terms hypocentral and rupture distances to determine the apparent source spectra at the center and top edge of the fault, respectively. The source spectra obtained from simulations are compared with the Brune’s single-corner point-source model. The comparisons showed that the apparent source spectra of the finite-fault simulations match with the Brune model at all frequencies and they do not show “spectral-sag” that is observed in the empirical data from large earthquakes. It is also observed that the source spectrum of finite-fault simulations at the fault center is equivalent to the Brune model with a double stress-drop of finite-fault model. The discrepancy in stress-drops vanishes when the top edge of the fault is considered.

[Full Text]

Overview of Ground-Motion Simulation Techniques

by Emrah Yenier

Department of Earth Sciences, University of Western Ontario

Prediction of ground motions that can be created by anticipated earthquakes is one of the vital issues in the mitigation of earthquake hazard. The predicted amplitudes are used for seismic design of structures and rehabilitation of existing buildings. Empirical data obtained from past earthquakes constitute the essential input for the estimation of ground motions. However, they are limited in terms of prediction of ground motions for wide ranges of magnitude, distance and site conditions in most of the tectonically active regions. Simulated ground motions are, therefore, widely used in engineering seismology and earthquake engineering related studies, especially for regions where the ground-motion data is sparse. Several approaches have been proposed for the generation of synthetic ground motions. All these methods have their own advantages and limitations. This situation makes the selection of the most suitable method for a given simulation task difficult. This paper presents an overview of stochastic, deterministic and hybrid simulation methods in terms of their modeling perspectives. It aims to highlight the advantages and limitations of these simulation methods by referring to their modeling principles to help decision makers for the selection of the appropriate method. The features of the simulation techniques presented in this paper indicate that the selection of the appropriate method is not only governed by the capabilities of the methods but also the availability of the resources (e.g. financial, time and data).

[Full Text]