Empirical predictive model for seismic displacement of slopes under velocity pulse-like ground motions

Based on the procedure to identify the velocity pulse for multi-component ground motions, 196 ground motions from NGA database and Wenchuan Earthquake are selected as pulse-like ground motions. The newmark decoupled model considering the nonlinear properties of soils is used to calculate the seismic displacement of slopes with different parameters (k_y, T_s) subjected to near-fault pulse-like ground motions, and the efficiency of different pulse-like ground motion parameters for predicting the displacement is also investigated. A seismic displacement model for the slopes under near-fault pulse-like ground motions is developed. The results indicate that the existing empirical models for seismic displacement of slopes significantly underestimate the displacement value induced by near-fault pulse-like ground motions, while they can be used to predict the seismic displacement caused by near-fault non-pulse ground motions. A close relationship between the seismic displacement and the velocity pulse characteristics for the pulse-like ground motions is shown, and the peak ground velocity (PGV) is the most efficient parameter in terms of the seismic displacement. The spectral acceleration at a degrade period equals to 1.5T_s (S_a(1.5T_s)) and PGV can represent the frequency content and velocity pulse characteristics of the pulse-like ground motions, respectively. Thus, an empirical predictive model for seismic displacements is developed using the vector intensity measures of S_a(1.5T_s) and PGV, which can be used in probabilistic seismic hazard analysis for the seismic displacement including the effects of near-fault pulse-like characteristics.