SMIP 2025 Seminar

Most duration metrics used to characterize design ground motions are based on the acceleration tie series which reflects the duration of the short-period ground motion but may not reflect the duration of the long-period ground motion. A new period-dependent duration based on the cumulative-squared response of a 50%-damped single-degree-of-freedom oscillator is presented. The period-dependent duration metric shows that for T>1 sec, the duration is systematically larger than the traditional accelerogram-based duration. There is also a large variability of the ratio of the long-period duration to the acceleration-based duration. To develop design time histories for structure sensitive to the long-period ground motion, the long-period duration of the time histories should be considered in the selection process. 

Traditional ergodic ground motion models (GMMs), typically conditioned on the time-averaged shear-wave velocity in the upper 30 m of soil (VS30), exhibit high site-to-site variability due to their inability to capture site-specific features such as resonances caused by impedance contrasts. Microtremor-based horizontal-to-vertical spectral ratios (mHVSR), which can contain peaks linked to site resonances, offer a useful supplement to VS30 for site response prediction. We use peak features from mHVSR spectra from 685 sites across California and neighboring regions (e.g., Nevada, Oregon) to develop mHVSR-conditioned site response models, which reduce site-to-site variability compared to a baseline linear site amplification model. 

Horizontal spectral response at a site can change considerably with orientation, in what is referred to as directionality. This paper summarizes recent and ongoing studies on ground motion directionality in strike-slip earthquakes. Ground motions from strike-slip earthquakes exhibit systematic directional patterns, with the maximum spectral response occurring near the transverse orientation. Ground motions from physics-based simulations of strike-slip earthquakes are found to exhibit directionality characteristics consistent with recorded motions, although they tend to have greater polarization at shorter periods, presumably due to smaller wave scattering than the one occurring in real earthquake ground motions. Lastly, new approaches to quantifying component-to-component variability of response spectral ordinates are introduced, and it is shown that this source of variability can be significant for seismic hazard assessment. These findings support the development of orientation-dependent ground motion models and emphasize the importance of accurately representing azimuthal dependence and variability in hazard analysis and seismic design, especially in regions dominated by strike-slip faulting, such as California. 

We examine the applicability of the Bayless and Abrahamson (2019) effective amplitude spectrum ground motion model (BA18) for the Sierras region of Eastern California. An overall tendency to overpredict low-frequency ground motions, and to underpredict high-frequency ground motions is identified. The over- and underpredictions are due to site condition discrepancies between the stations considered in the BA18 model development, and the Sierras region, namely the velocity structure used in estimating earthquake magnitudes, and the high-frequency attenuation and site amplification patterns. An analysis of site terms shows the potential for including site-specific predictors beyond the time shear-wave velocity (VS) in the top 30 m (VS30) for BA18 applications in the Sierras. 

The performance of University of Southern California (USC) hospital and Los Angeles County Fire Command and Control (FCC) buildings during the 1994 Northridge earthquake are two important case studies of how base-isolation can prevent significant structural damage during seismic activity. Both structures have experienced several ground motions of lesser intensity in the subsequent years. Analysis of the structural response to the later ground motions offers valuable insights into the structural parameters and bearing characteristics. This paper presents a comparative study of the evolving dynamic properties of the USC hospital and the FCC buildings using system identification based on vibration records. 

Following experimental testing of large scale pier walls, a simple numerical model is implemented to model the interaction effects of biaxial flexure, shear, and sliding. Based on test observations, a phenomenological hysteretic sliding model is suggested to capture flexure induced sliding behavior. Such a model should have both strength and stiffness degradation with increasing sliding displacements and cyclic loading, as well as a method to transfer load back into the flexural mechanism during the rocking phase of sliding motion. A provisional model is implemented in OpenSees using currently available formulations and applied to a model of the experimental walls. The results indicate promising behavior in matching the fundamental period with the measured data, but further refinement is needed to fully capture sliding and behavior at large displacements. 

This paper focuses on assessing the vibration characteristics of Briones Dam in Northern California. First, earthquake-based Horizontal-to-Vertical Spectral Ratio (eHVSR) was estimated by dividing the horizontal records by the vertical components, and the SSR was determined by comparing crest recordings with those from the abutment. Additionally, a microtremor-based HVSR was computed based on field measurements. The fundamental frequency was estimated using three empirical methods: mHVSR (0.7 to 1 Hz), eHVSR (0.9 to 1.1 Hz), and SSR (1.2 Hz). The slight variations among these three methods suggest the need for further investigations that consider the geological and geotechnical conditions of the dam.