Estimating Permanent Ground Displacement from Near-Fault Strong-Motion Accelerograms

Abstract

This study presents a three-phase algorithm that uses the discrete wavelet transform (dwt) to make the baseline correction to near-fault ground-motion accelerogram and to estimate permanent ground displacement. The developed algorithm is applied to the near-fault ground-motion data collected from the 1999 Chi- Chi, Taiwan, earthquake. In phase I the peak ground displacement is estimated through the decomposition of acceleration data; in phase II, the residual deformation of the ground is estimated from the decomposition of the velocity data. Combining phase I and phase II enables the near-fault ground displacement with residual deformation to be estimated. The estimated permanent ground deformation is compared with that indicated by nearby Global Positioning System (gps) data before and after the earthquake. The obtained residual ground deformation correlates well with that obtained by gps measurement.