A Bayesian Approach to Seismic Hazard Estmation: Maximum Values of Magnitudes and Peak Ground Accelerations

A rather simple straightforward procedure of estimating maximum values of the considered parameter (earthquake magnitude in a given region or seismic peak ground acceleration at the considered site) and quantiles of its probabilistic distribution in a future time interval of a given length is presented. To assess the peak ground acceleration using this method, the input information is the earthquake catalog and the regressive relation where the peak seismic acceleration at a given point bears the magnitude and epicentral distance of the site considered (seismic attenuation law). The method is based on the Bayesian approach, in which the influence of uncertainties of magnitudes and seismic acceleration values can be taken into account. The main assumptions for the method are the Poissonian character of the seismic event flow, a frequency-magnitude law of Gutenberg-Richter's type with a cutoff maximum value for the estimated parameter, and an earthquake catalog that has a rather large number of events. The method has been applied to seismic hazard estimation in California, the Balkans, and Japan.