Aftershock Record Selection Criteria for Structural Vulnerability Assessments

Abstract

After a major earthquake, sizeable aftershocks can continue for weeks and cause considerable impact to recovery, but aftershocks are largely overlooked in engineering practice. Numerous research studies have considered the response of structures under mainshock-aftershock (MS-AS) sequences, but there is lack of validation for the artificially generated sequences used in these studies and no standardized method exists for selecting aftershock motions for consideration. This study examines three different aftershock selection criteria adopted in the literature (replicate, random, and generalized Omori’s law) and assesses the validity of artificially generated near-field MS-AS sequences, by comparing the cumulative damage potential against real sequences from Northridge (1994), Livermore (1980), and Mammoth Lakes (1980). An elastic-perfectly plastic (EPP) single-degree-of-freedom (SDOF) model is used to compare the peak ductility demand for the different selection methods. The results highlight how poorly selected aftershock ground motions can severely misrepresent the risk and offer insight on key ground motion and sequence characteristics that are crucial for accurately representing aftershock hazards. The generalized Omori’s law could potentially be used to generate representative MS-AS sequences for shallow near-field earthquakes, but additional work is needed to further investigate the influence of earthquake parameters on the inelastic response potential of aftershocks and ultimately develop a practical and reliable ground motion selection scheme or framework to support structural assessments and decision-making under aftershock hazards.