Non-ergodic probabilistic seismic hazard methodology using physics-based ground motion prediction: the case of LAquila, Italy

A non-ergodic probabilistic seismic hazard analysis (PSHA) utilizing the physics-based ground motion prediction was proposed in this study to minimize the increasing uncertainties in the use of empirical equations. The City of L’Aquila in Italy was used for illustrative purposes due to the availability of data and the historical seismicity of the site. A total of 28 seismic sources were identified in this study located within a 100 km radius from the city. Fault properties such as geometry and location were obtained from the literature, while the fault occurrence rates were obtained using the FiSH Code. A modified time-weakening friction law was proposed to model the seismic energy released by an earthquake. Uncertainties in different rupture scenarios were characterized through the Guttenberg-Richter Relations and the Characteristic Brownian Time Passage. Uncertainties in distances were characterized through probability mass functions, which were used to calculate the ground motion exceedance probabilities. The 1D elastodynamic equation coupled with the Hooke’s law was used to predict the peak ground acceleration (PGA), a measure of the ground shaking level. A hazard curve, which is a plot of PGA and its recurrence, was constructed and compared with the results of the study of Valentini, et al., AGU 100: Advancing Earth and Space Science (2019). The method proposed in this study predicts a higher hazard rates for PGAs less than 0.70 g, which implies that the ground motion was overestimated for very far sources. In contrast, lower hazard curves were observed for PGAs greater than 0.70g which can be attributed to fewer seismic sources considered in this study.