A one-step Newmark displacement model for probabilistic seismic slope displacement hazard analysis

Estimating the earthquake-induced sliding displacement is important in assessing the stability of slopes during earthquakes. Current Newmark displacement models generally use ground motion intensity measures (IMs) as predictors, and the uncertainties of predicting IM values need to be accounted for in probabilistic seismic slope displacement hazard analysis. This paper aims at developing a simple one-step predictive model for the Newmark displacement based on only four seismological parameters (moment magnitude, rupture distance, fault categories and shear wave velocity at top 30 m V-s30) rather than any IMs. The predictive model is suitable for critical acceleration from 0.02 g to 0.25 g, covering most of the susceptible earthquake-induced landslide cases. First, the proposed one-step model is compared with some recently developed IM-based Newmark displacement models. It is found that both the median predictions and the variabilities of the proposed one-step model are in reasonable agreement with those obtained by the IM-based models. Second, several hypothetical slopes are used to compare the slope displacement hazard curves between the one-step model and the IM based models. The new model can be used as an alternative method for a fully probabilistic analysis of the earthquake-induced slope displacements. In addition, spatial correlations of the Newmark displacement residuals are also investigated using strong-motion data from the Northridge and Chi-Chi earthquakes. (C) 2016 Elsevier B.V. All rights reserved.