A neural network-based ground motion model trained on ground motion simulations in NZ

Abstract

Probabilistic seismic hazard analysis (PSHA) based on physics-based simulations offers many advantages compared to more traditional empirical Ground Motion Model (GMM) based PHSA. However when a large number of seismic sources is considered, such as distributed seismicity, simulation based PHSA becomes unfeasible due to the high computational cost associated with simulations. Utilising a surrogate model, such as an artifical neural network, gives the physics-based simulation advantages without the associated computational cost. Additionally it provides an extra avenue for investigating simulation results.