Damping scaling factors for horizontal acceleration response spectra from subduction earthquakes in offshore and onshore regions of Japan

Abstract

Ground motion models (GMMs) are typically developed for 5% damped elastic response spectra. However, in practice, structural and non-structural engineered facilities require a range of damping ratios for seismic design, analysis, and hazard assessment. To address this need, damping scaling factor (DSF) models were developed using Bayesian inference with integrated nested Laplace approximation (INLA) for pseudo-acceleration spectra (or displacement spectra) and absolute acceleration spectra derived from subduction earthquakes in offshore (S-net) and onshore (K-NET and KiK-net) regions of Japan. The S-net stations are categorized into buried and unburied stations based on factors such as seawater depth and arrangement form. The proposed DSF model incorporated moment magnitude, rupture distance, damping ratio, and station type as predictor variables, employing a linear function of ln(β) to differentiate between onshore, offshore buried, and offshore unburied ground motions. The DSF model is applicable to subduction earthquake scenarios with moment magnitudes ranging from 5.5 to 9.1, rupture distances up to 1000 km, and 11 damping ratios between 0.5% and 30% for spectral periods of 0.01 to 10 s. It reflects significant differences among onshore, buried and unburied ground motions across most spectral periods at various damping ratios, particularly at low damping ratios and long periods. In addition, standard deviation models were developed for the proposed DSF models using a cubic function of ln(β/5). Compared to the observed standard deviation, the proposed model effectively captures the observed trends. The proposed DSF model can be effectively used to scale onshore and offshore 5% damped GMMs to other GMMs with damping ratios ranging from 0.5% to 30% for subduction earthquakes, providing a valuable tool for probabilistic seismic hazard assessment and engineering applications.