Impact of ground motion directionality (RotDnn) on the coupled and uncoupled inelastic response of RC circular cantilever columns

Abstract

Directionality in the representation of earthquake demand for structural seismic design has surfaced as a subject of debate in recent years. Design spectra, as defined in building and bridge codes, account for directionality by considering the peak response in all possible orientations of the motion using the RotDnn measure, where “nn” is the percentile of the peak response. Current seismic hazard maps are based on ground motion models that use the RotD50 spectra (median peak response). However, since 2009, building codes have adopted RotD100 spectra (maximum peak response), raising concern among researchers and engineers. The first objective of this study is to evaluate the impact of design spectra definition, RotD50 or RotD100, on the bidirectional inelastic response of azimuth-independent RC systems. Upcoming updates to seismic hazard maps highlight the need of understanding the implication of this choice in bridge design. While previous research focused on the relationship between RotD50 and RotD100, this work assesses the impact of each definition on the inelastic seismic response of RC circular columns designed following the Direct Displacement-Based Design approach and verified through nonlinear time history analyses (NTHA) using ground motions from shallow crustal active and subduction zone regions. The ratios of NTHA to design displacement confirm that RotD100 spectra should be used to design azimuth-independent RC systems and provide the first known verification of DDBD for bidirectional loading (second objective). These ratios are a function of ductility level, tectonic regime, effective period, and coupling (which is the third objective). Overall results show that RC circular cantilever columns designed to RotD100 spectra show less deviation from the expected design displacement than columns designed to RotD50 spectra. This study presents the first step in resolving the ongoing debate over spectra definition and provides recommendations for displacement demand estimations for response spectra-based design approaches.