SMIP 1997 Seminar

The study first reviews some simple methods of calculating maximum displacement, base shear and interstory drift from acceleration response records. A record from a 17-story building in Los Angeles obtained during the Northridge earthquake is used for illustration. More detailed analyses of similar records are suggested as attractive subjects for Master's theses and senior projects. Simple analyses of records from Northridge and other earthquakes convincingly indicate base shears and drifts that are substantially in excess of those used in design. The author urges that earthquake resistant design be based increasingly on measurements of earthquake response and measured properties of materials, and less on empiricism and qualitative assessments of earthquake performance.

This paper explains the effects of rupture directivity on near-fault ground motions, describes an empirical model of these effects, provides guidelines for the specification of response spectra and time histories to represent near-fault ground motions, and provides guidelines for the selection of time histories.

The recently published 1997 Uniform Building Code incorporates two significant changes to the ground shaking criteria which apply to all structures. The first change is a revision to soil types and soil amplification factors. The second change is the incorporation of near source factors in UBC seismic Zone 4. Together these changes result in the largest increases in code ground shaking criteria which has occurred in the past 30 years. Records obtained from the strong motion Instrumentation Program (SMIP) along with USGS records were the primary sources of data used to justify these code changes.

Recent developments in near-real-time recovery of strong-motion data have been expanded to include new means to disseminate and rapidly utilize the data. The TriNet project, a cooperative project in southern California involving CDMG, Caltech and the USGS, will disseminate earthquake data and information rapidly for use by emergency responders. In another development, the release of strong-motion data by CSMIP will include simple descriptors of the shaking level (Light, Moderate, Strong and Extreme Shaking). These descriptors, designed to be easily understood by the non-specialist, will be complemented by more customary quantitative characterizations (time-history and spectral levels). To make strong-motion data widely available for use in earthquake engineering and engineering seismology, the data will be available for access by any current Web browser at the Internet site http://www.consrv.ca.gov/CGS/smip. All strong-motion data released by CSMIP over the years are or will soon be available at this location. For future significant earthquakes, strong-motion data will also be available rapidly at the site.

FEMA-2731274, Guidelines and Commentary for Seismic Rehabilitation of Buildings (ATC, 1996a., b.) represents a landmark in the practice of structural/seismic engineering. In addition to providing a long-needed consensus basis for the design of building seismic upgrades, it also is a fist major step towards the development of performance- based design procedures for seismic resistance. Under the Guidelines, design is performed with the expectation that for specified levels of ground motion intensity, building performance will remain within anticipated levels. Performance levels are defined in terms of permissible damage to individual structural and nonstructural components and are intended to provide specific defined margins of safety. In order to accommodate this performance-based approach new analytical procedures and acceptance criteria were developed. While the Guidelines represent a significant advance in the practice of earthquake engineering, calibration of the analysis procedures and acceptance criteria to real building performance is clearly needed. The use of strong motion data obtained from instrumented buildings experiencing strong earthquake ground shaking will be an essential part of this process.

Borcherdt (1994) proposed that the short- and mid-period amplification factors used to scale the estimate of site-dependent response spectra could be calculated as a continuous function of shear-wave velocity averaged in the upper 30 m for various input ground-motion levels. This proposal appears to be an improvement for estimating design response spectra over site classifications defined by soil descriptions (e.g., SEAOC, 1988) or by ranges of average shear-wave velocity (e.g., NEHRP, 1991).

However, it is not clear that the accuracy available in current measurements of shear-wave velocity is sufficient to support their use directly in design calculations of motion. For example, at Gilroy #2 there are differences of about 300 m/sec in the shear-wave velocity measurements in the upper 30 m (EPRI, 1993). In this paper, we analyze the effect of variations of shear-wave velocity in the upper 30 m for design applications including the effects of nonlinear soil response using an equivalent-linear site response formulation (Schnabel et al., 1972). The analysis uses the extensive geotechnical site characterization and shear-wave velocity measurements at Gilroy #2, a stiff soil site that has been characterized to a depth of 240 m (EPRI, 1993). Response spectral accelerations from the recorded strong motions are compared to calculated values from the equivalent-linear analyses with several shear-wave velocity profiles in the top 30 m.

The preliminary analyses suggest that the Borcherdt (1994) methodology works well at this stiff-soil site for design levels of motion near 0.4 g, appropriate for many parts of California, even though there is a difference of 60% in the measured average of the shear-wave velocities in the upper 30 m.

This paper presents a perspective on the importance of developing accurate computer models of constructed buildings and the role that measured earthquake response records play in advancing this area of structural engineering research and practice.

Highway bridges and concrete dams are critical components of California's infrastructure. Because of the difficulty devising experiments to test the complete system response of bridges and dams in earthquakes, the design and safety evaluation of these large structures are primarily based on mathematical models and numerical simulations. Strong- motion data from bridges and dams provide the real-life laboratory for verifying the models. This paper summarizes the results from recent studies of the recorded strong-motion response data from bridges and a dam. Although the strong-motion data are limited, they have provided confirmation that models are capturing the essential features of earthquake response. Limitations of the models, however, are identified and future trends for instrumentation and data utilization for bridges and dams are discussed.