Predictive models for permanent displacement of slopes based on recorded strong-motion data of Wenchuan Earthquake

The permanent displacement induced by earthquakes can be regarded as an effective criterion for the stability estimation and anti-seismic design of slopes. Based on a large number of strong-motion records from Wenchun Earthquake, three kinds of regression models for predicting the permanent displacement in terms of (1) the critical acceleration ratio, (2) the Arias intensity and the critical acceleration ratio, and (3) the residual seismic intensity are established. Their validities are examined by the displacements converted from the measured ones of reinforced slopes by means of the energy method. These equations are well constrained and fit the data well, among which the equation in terms of the residual seismic intensity fits best, indicating that the permanent displacement has a close relationship with the residual seismic intensity. The model in terms of the critical acceleration ratio is brief but practical, and that in terms of the Arias intensity has a low value of R² and prediction precision. The results show that the models developed based on strong-motion records have zone dependence because of the site effect of stations. The proposed models can be used to generate a regional-scale seismic landslide hazard map, to provide the assessment of landslide distribution after earthquakes, and to accelerate preliminary screen of specific sites.