A Recursive Division Stochastic Strike-Slip Seismic Source Algorithm Using Insights from Laboratory Earthquakes

There are a sparse number of credible source models available from past earthquakes and a stochastic source model generation algorithm thus becomes necessary for robust risk quantification using scenario earthquakes. We present an algorithm that combines the physics of fault rupture as imaged in laboratory earthquakes with stress estimates on the fault constrained by field observations to generate probability distributions of rise-time and rupture-speed for strike-slip earthquakes. The algorithm is validated through a statistical comparison of peak ground velocity at 636 sites in Southern California from synthetic ground motion histories simulated for 10 rupture scenarios using a stochastically generated source model against that generated using a kinematic source model from a finite source inversion. This model, selected from a set of 5 stochastically generated source models, produces ground shaking intensities in Southern California with a median that is closest to the median intensity of shaking from all 5 source models (and 10 rupture scenarios per model).