Earthquake Spectra

The equivalent static eccentricities of seismic forces are usually defined by codes with simple expressions of the static eccentricity. This paper presents certain formulae for the exact calculation of these eccentricities on the basis of the dynamic response of a simplified model. From the parametric analysis of such formulae the determinative role of the torsional and lateral stiffness of the system becomes obvious for the correct evaluation of the equivalent static eccentricities. Finally, a proposal is made for the improvement of the static torsional provisions of the current codes.

Seismic-hazard assessment at small scales requires the computation of realistic broad-band accelerograms near the sources. Methods that use only natural records are often limited because of the limited data set. Among recently developed methods for source modelling, the spectral-source model, based on the k⁻² dislocation distribution seems particularly promising as it enables computation of synthetics for any source-receiver distance and directivity configuration. The spectral model takes into account important source characteristics, such as stress drop and directivity effects predicting acceptable high-frequency levels, whatever the station location (Bernard et al., 1996b). Firstly, some tests to explain the directivity properties of the model are presented, followed by the main results of a deterministic study carried out for the 1992 Erzincan earthquake, Turkey. Finally, results of a probabilistic study in the Moyenne-Durance region (France) are used to demonstrate the ability of this method to produce seismic-hazard maps and spectra linked to a specific return period.

The State of California requires local governments to pay attention to seismic safety in formulating general plans for urban development and in permitting and inspecting new construction and remodeled existing structures. The Northridge earthquake provided an opportunity to determine whether these provisions, which have been mandatory for more than two decades, actually result in lower property damages. Using data on the number of structures damaged in the Northridge event, we show that, for suburban jurisdictions, damages were lower when local governments formulated broader goals for seismic safety, developed policies to make the public more aware of seismic risks, and expended more resources on enforcing the seismic provisions of building codes. Thus, seismic safety mandates on local governments can lead to lower property damages, and these benefits are enhanced when local governments expend more effort on their implementation.

The region around Acapulco, on the Pacific coast of Mexico, is subjected to large seismic risk. This paper presents a contribution to improve microzonation of this region. We investigated site effects using three basic sources of data: strong-motion records from all of the instruments that have operated within the area; weak-motion records obtained from the installation and operation of a temporal, digital, seismograph network; and measurements of microtremors at 35 sites. We compared and evaluated different techniques of data analysis. We show that very coherent results are obtained from different kinds of measurement, and that microtremor records are very useful to interpolate sparse earthquake data. We propose two maps that reflect the fundamental characteristics of site effects in the area: dominant period and maximum relative amplification. These maps may be used to improve current microzonation of Acapulco.

Optimum isolator damping required to minimize absolute acceleration responses to stochastic ground motions in buildings with non-linear resilient-friction base isolation (R-FBI) devices is studied. Using a stochastic equivalent linearization technique, an analytical expression is developed for estimating optimum material damping values in an R-FBI system for a rigid superstructure. This optimum damping value depends on the effective time period and friction coefficient of the R-FBI system and the design earthquake intensity. The fixed base structural time period, damping, and the number of stories also affect the optimum isolator damping values, if the superstructure is flexible. However, the analytical expression mentioned above provides an upper-bound optimum isolator damping value for any superstructure with an R-FBI system subjected to a given earthquake intensity. Comparison of optimum damping values for a linear isolator and the R-FBI system, with identical lateral stiffness properties, illustrates the beneficial effect of sliding friction in reducing the optimum damping in the latter type of isolator, for a given earthquake intensity.

Local site amplification of sedimentary deposit during earthquakes is an important issue in strong ground motion analysis. The phenomenon is more obvious for sediment basin. From the strong-motion instrumentation network of Taipei basin, the ground motion characteristics of the basin effects are studied from two seismic events: the June 5, 1994 earthquake with M _L = 6.57 and the June 25, 1995 earthquake with M _L = 6.50. The objective is to investigate the effects of the basin structure on the patterns of the recorded ground motions. The analyses include: (1) response spectrum and spectral ratio analyses; (2) correlation of seismic source, PGA distribution and strong-motion duration with site amplification, (3) principal direction analysis of seismic waves in the basin. The observed variations of ground motion across the basin are different from each other because of the basin effect. It means that for the Taipei basin, the basin effects for shallow sources are going to be much more significant than for the deep sources.

A parametric study on the scaling of natural accelerograms using spectrum intensity scales is presented. A series of nonlinear time-history analyses for a SDOF system is conducted using an ensemble of world-wide damaging earthquake records for a wide range of seismic conditions. Structural parameters considered in the study include: yield seismic coefficient, yield natural period and postyield stiffness. The effectiveness of several spectrum intensity scales is evaluated in terms of their correlation with displacement ductility demand. Results indicate that there is not a unique scale with best performance over the entire range of strength and stiffness. An alternative combined criterion to define spectrum intensity is proposed in terms of a system of spectrum intensity scales. The system optimises the correlation between spectrum intensity and displacement ductility demand. This is achieved by defining regions in the space of structural parameters for which it is possible to identify a spectrum intensity scale of best performance.

A response spectrum rule for combining the contributions from three orthogonal components of ground motion to the maximum value of a response quantity is presented. This rule, denoted CQC3, is compared to the 30% and 40% rules and the square-root-of-sum-of-squares (SRSS) rule currently specified in many design codes. It is shown that these current rules are special cases of the CQC3 rule, when certain conditions regarding the nature of the ground motion or the structural response are satisfied. Because these conditions are not always satisfied, it is argued that the CQC3 rule should be adopted as a general rule for the multicomponent combination problem. The CQC3 rule additionally offers a simple formula for determining the most critical orientation of the ground motion components for each response quantity of interest. The CQC3 rule is computationally simple and easy to implement in standard dynamic analysis codes.

A direct procedure is presented for generating a response spectrum for an arbitrary nonexceedance probability from a prescribed design mean response spectrum. An amplification factor is derived to estimate the maximum response values of an MDOF system for a nonexceedance probability from the mean maximum ones. An efficient stiffness design method for a shear building is developed with the use of its fundamental frequency and translational eigenvector as parameters for adjusting the nonexceedance probability of the seismic drifts to the specified value. The validity and accuracy of the proposed method are demonstrated by a Monte Carlo simulation together with time-history analyses.

This article explains the essential duality between logic tree analysis and sensitivity analysis in probabilistic seismic hazard assessment. The results of a logic tree analysis can be used to carry out a sensitivity analysis. More important, a preliminary sensitivity analysis can be used to focus attention on those parameters that have the largest effects on assessed hazard. This information can be of great value in the construction of a logic tree. Thus we advocate a two-stage process that begins with a sensitivity analysis and then proceeds to the construction of a logic tree.

In this study, the standard spectral ratio and the horizontal-to-vertical spectral ratio techniques are applied in order to study their effectiveness in investigating and quantifying the influence of geological conditions on strong ground motion. For this purpose, an accelerogram data set recorded at the Euroseistest array in the Mygdonia graben (lake Volvi area) near Thessaloniki, Greece, during the period April 1994 to June 1996 is used. Both experimental techniques show similar spectral ratio shapes with comparable fundamental resonant frequencies, which are well correlated with the well known geotechnical-geological conditions. Namely, the resonant frequency at the center of the valley is shifted to lower values, less than 1 Hz, while at the edge it is shifted to higher values, greater than 2 Hz. The horizontal-to-vertical spectral ratio technique is an effective method to estimate some basic characteristics of local site effects using a single accelerograph station. It reveals the fundamental resonant frequency of alluvial deposits by using only a single strong motion station, while the absolute level of the horizontal-to-vertical spectral ratio method tends to underestimate the amplification level compared to the standard spectral ratio technique.

The Rinaldi record of the 1994 Northridge earthquake is one of the most important strong motion records in earthquake engineering (largest recorded horizontal peak ground velocity in the western United States, ∼170 cm/s). Digitization of this record was not straightforward because of numerous stalls and malfunction of the half-second pulse relays. It was digitized and processed for the Los Angeles Department of Water and Power soon after the earthquake, and was recently redigitized and reprocessed. This paper presents the results of the recent digitization and processing and a comparison of the two digital versions of the record. The “new” version differs from the “old” version in the number of stalls corrected for (17 versus one stall), total length of digitized traces (∼20 s versus ∼15 s), and amplitudes of some high frequency acceleration peaks (in the “old” version, some high frequency peaks have been underestimated). The peak amplitudes of corrected acceleration, velocity and displacement, and the linear response spectrum amplitudes are not significantly different. However, the two digital interpretations have different time scales, because of the difference in the number of stalls accounted for.