Distributed Slip Model for Simulating Virtual Earthquakes

We develop a physics based, generic finite fault source, which we call the Distributed Slip Model (DSM) for simulating large virtual earthquakes. This task is a necessary step towards ground motion prediction in earthquake-prone areas with limited instrumental coverage. A reliable ground motion prediction based on virtual earthquakes must account for site, path, and source effects. Assessment of site effect mainly depends on near-surface material properties which are relatively well constrained, using geotechnical site data and borehole measurements. Assessment of path effect depends on the deeper geological structure, which is also typically known to an acceptable resolution. Contrarily to these two effects, which remain constant for a given area of interest, the earthquake rupture process and geometry varies from one earthquake to the other. In this study we focus on a finite fault source representation which is both generic and physics-based, for simulating large earthquakes where limited knowledge is available. Thirteen geometric and kinematic parameters are used to describe the smooth "pseudo-Gaussian" slip distribution, such that slip decays from a point of peak slip within an elliptical rupture patch to zero at the borders of the patch. Radiation pattern and spectral charectaristics of our DSM are compared to those of commonly used finite fault models, i.e., the classical Haskell's Model (HM) and the modified HM with Radial Rupture Propagation (HM-RRP) and the Point Source Model (PSM). Ground motion prediction based on our DSM benefits from the symmetry of the PSM and the directivity of the HM while overcoming inadequacy for modeling large earthquakes of the former and the non-physical uniform slip of the latter.