Geofizički seminar: Milad Kowsari

dr. Milad Kowsari

(Icelandic Meteorological Office i University of Iceland)

Recalibration and Ranking of Ground Motions Models (GMMs) for Iceland

ABSTRACT: The probabilistic seismic hazard assessment (PSHA) is the international standard practice for seismic risk mitigation and management. One of the key elements of any PSHA is ground motion models (GMMs) that describe ground motion amplitudes, as a function of the earthquake source, seismic wave propagation path and localized site effects. However, developing GMMs is challenging, particularly in regions with sparse data. To overcome this problem, we can recalibrate GMMs to the existing data from the region under study using a Bayesian statistical approach. This methodology can be particularly useful as it allows taking prior information about the model parameters into account and adding it to the information in the likelihood that stems from the observed data. On the other hand, in the current practice of PSHA, the different estimates of the median ground motions predicted by GMMs are attributed to epistemic uncertainty. In order to reduce the epistemic uncertainty in PSHA, data-driven ranking methods can be used for selecting those GMMs that appropriately describe the characteristics of the observed ground motions in the region under study. Here, we introduce and discuss several data-driven methods that reduce subjectivity in the selection process of GMMs for application in PSHA.