SMIP 1992 Seminar

The site response to strong and weak ground motion depends largely on the subsurface conditions at the soil site for the two rock-soil station pairs studied. The first station pair consists of a soft-soil site and a sandstone/shale site (Treasure Island and Yerba Buena Island). For strong motion, the soft-soil site is amplified by a factor of about 3 over the rock site from 0.5 to 2.0 Hz. The amplification is much higher for weak motion and suggests a dependence on signal amplitude.

A second station pair near Gilroy consisting of a stiff-soil site and a sandstone site was studied with contrasting results. Unlike the results for the soft-soil study above, the estimated stiff-soil site response is not significantly different for strong and weak motion from 0.5 to 2.0 Hz.

Empirical strong motion recordings from large magnitude (M ≥ 6) events are reprocessed with an emphasis toward preserving the long period motions. A preliminary empirical attenuation relationship for response spectral values is derived for the period range, 1-20 seconds.

The Loma Prieta earthquake sequence has provided the opportunity to compare ground motion recorded both during an magnitude 7 mainshock and numerous magnitude 2.5 to 4.5 aftershocks at 14 sites. We find two results, (1) weak motion recordings of aftershocks have predictive power to indicate the directionality of the shaking in the mainshock, and (2) non-linearity during mainshock strong motions is suggested at the sites that suffered the largest accelerations in the mainshock..

A procedure is presented for evaluating building code provisions for accidental torsion from analysis of recorded motions of nominally symmetric-plan buildings during earthquakes. This procedure is utilized to analyze the motions of three buildings recorded during recent California earthquakes. The results demonstrate that the accidental torsion specified by the Uniform Building Code is adequate in representing the torsion in the recorded motions of these three buildings. It is also concluded, although somewhat speculatively, that accidental torsion need not be considered in the design of many buildings.

To estimate the roof and story drifts occurring during severe earthquakes, the Uniform Building Code (UBC) uses 3Rw/8 as a displacement amplification factor (DAF) to amplify elastic design drifts. A comparison of several seismic design codes shows that UBC's DAF is very low. A study conducted on four instrumented buildings indicates that the maximum story drifts developed in severe earthquakes are much higher than that predicted by the UBC. This study also shows that the DAF of multi-degree-of-freedom systems is similar to that of single-degree-of-freedom systems.

A data base of 64 buildings was developed to investigate their fundamental period of vibration recorded during recent California earthquakes. These periods are compared with those calculated from the 1991 Uniform Building Code (1991 UBC) and the 1990 Structural Engineers Association of California Recommended Lateral Force Requirements (Blue Book). Steel moment frames, braced frames, concrete moment frames, and shear wall buildings are included in the data base. Also included in the study are buildings from the Gates & Foth report [12].

Ground-motion records from the California Strong Motion Instrumentation Program (CSMIP) and from other sources are used to examine relationships between strong-motion duration, elastic strength demands, structural response factors, and inelastic strength demands, as derived from analytical study.

Structural response factors R are found to have generally slight systematic dependence on strong-motion duration, but they do have a more direct, if not fundamental, relationship to characteristics of elastic response spectra. In addition to the general lack of correlation between R and duration, both elastic and inelastic strength demands (and hence, ground-motion damageability also) show little apparent consistent systematic correlations with duration.

Results of this study imply that seismic load specifications, for safety analysis or design of buildings, can generally be adequately described by appropriate ground-motion spectra without the need to explicitly specify an associated duration.

The objective of this research is to verify a newly developed method, time delay (TD) method, for earthquake analysis using data from strong motion instrumentation program (SMIP), and then applying the TD method to calculate overturning moment of structures. The TD method incorporates the attribute of finite wave speed propagating upward into the structure.

One detailed 3-dimensional mode1 and nine simplified 2-dimensional models were used for study. The effects on the overturning moment due to the use of TD method varies depending upon the predominant period of the input earthquake motions and the structures. The concept of the TD method is verified with the measured data.

This paper presents the results of a study on the responses of the San Juan Bautista 101/156 Separation Bridge instrumented by the California Division of Mines and Geology under its Strong Motion Instrumentation Program (CSMIP). The bridge experienced three significant earthquakes as follows: 1979 Coyote Lake earthquake (before seismic retrofit), 1984 Morgan Hill earthquake (after seismic retrofit), and 1989 Loma Prieta earthquake. The recorded seismic responses during the three earthquakes are analyzed and correlated with theoretically predicted responses. Adjustments of structural parameters and modeling concept to achieve satisfactory correlations are discussed. Important characteristics of relative motions between adjacent bridge segments are explained based on the records and analyses. Comments are also given regarding the performance of cable restrainers for seismic damage mitigation.

The dynamic response of Lexington Dam during the October 17, 1989 Loma Prieta eartquake and a smaller earthquake is analyzed using equivalent-linear and nonlinear procedures. The calculated motions at the dam crest are compared to the record motions during the earthquakes. The compond response of the dam using nonlinear procedures is in reasonable agreement with the recorded response and the response calculated using equivalent-linear methodes.