Earthquake Spectra

Heavy concrete and masonry walled buildings with lightweight steel or wood flexible roof diaphragms are a common type of construction in North America. Failures of the out-of-plane wall anchorage to these roof systems and the resulting partial roof collapses during past earthquakes have led to repeated revisions to the seismic design provisions in the U.S. model building codes. However, the force levels considered in the current design provisions have remained largely unchanged since their introduction in the 1997 Uniform Building Code, and these provisions have not been fully tested by strong ground motions in the field. Using a two dimensional numerical framework, a series of nonlinear time history dynamic analyses on various building archetypes were conducted to evaluate the validity of the current wall anchorage design force levels. The results of this study reveal that the current wall anchorage design forces are generally appropriate, but with some very significant exceptions.

A performance-based methodology is presented to quantify the reduction in collapse safety of main shock–damaged reinforced concrete frame buildings with infills. After assessing their collapse safety in the intact state, the residual collapse capacity following main shock damage is evaluated by conducting incremental dynamic analysis to collapse using main shock–aftershock ground motion sequences. The median collapse capacity and conditional probability of collapse for the main shock–damaged building, normalized by that of the intact case are the metrics used to measure the reduction in collapse safety. Taller buildings with built-in soft and weak first stories have the highest reduction in collapse safety as a result of main shock damage. Among the engineering demand parameters recorded during the main shock analyses, story drift demands (peak transient and residual) and infill strut axial deformations have the highest correlation with the decline in collapse performance. The results of the main shock–aftershock incremental dynamic analysis to collapse are used to develop fragility functions for the limit state defined by the building being structurally unsafe to occupy.

This case study evaluates the vulnerability of Nevada bridges relative to earthquake hazard using two different methods. First, a distributed set of 112 realistic earthquake scenarios were processed with USGS program ShakeMap, and site-specific ground motion levels extracted for 1831 bridges in Nevada using ShakeCast. Second, using hazard curves underlying the 2014 USGS National Seismic Hazard Map (NSHM), return periods for earthquakes causing extensive damage to bridges were extracted and compared to the 1000-year design level adopted by the AASHTO. Lower capacities than those used in ShakeCast were proposed for five continuous bridge types based on a literature review. Scenarios provide points in a deterministic seismic hazard approach, with large earthquakes on known faults. NSHM hazard curves are based on a probabilistic approach. A graphical method is presented to unite the two approaches. A list of potentially vulnerable bridges was developed for Nevada Department of Transportation (NDOT) use in bridge retrofit planning. As a continuing benefit, ShakeCast now operates in Nevada to provide near-real-time inspection priorities in the event of a serious earthquake.

Recent developments in volunteered geographic information (VGI), such as the OpenStreetMap initiative, highlight the potential of these data sets as supplementary or alternative sources of spatially based building information. Its increase in usefulness is particularly evident when combined with additional open-access data, such as the CORINE initiative, which provides the geo-referenced distribution of non-residential areas in Europe. However, the systematic application of VGI in the development of exposure models for catastrophe risk assessment has been the subject of limited research. In this paper, we describe an algorithm that, based on open-access information, provides an automated tool for the development of exposure data sets of industrial buildings in Europe, at the 30 arc-second resolution. Its practical application shows that results obtained at national and regional scales are in excellent agreement with data collected from cadastral agencies in Denmark, Italy, and Portugal; which highlights the potential of the algorithm when real building information is scarce or non-existent.

This paper presents the seismic fragility assessment of the Smithsonian Institute Museum Support Center (MSC), which sustained appreciable damage during the 2011 Virginia earthquake. A three-dimensional (3-D) finite element model (FEM) for the building was created and validated using measured dynamic characteristics determined from field vibration test data. Two suites of bidirectional ground motions at different hazard levels were applied to the FEM to generate fragility curves for structural as well as nonstructural (storage cabinets) damage. The effect of brace yielding strength on structural and nonstructural damage is also investigated to provide recommendations for future retrofit. The fragility curves show that the spectral acceleration to cause structural damage to the building is not high. Due to low seismicity, however, the probability for the structure to be damaged at the design basis earthquake is small. Nevertheless, the probability for nonstructural damage is considerable, which is an important issue related to the seismic performance of the building.

A rational and quantitative shutdown criterion is required for a nuclear power plant in response to seismic shaking to determine whether the plant must be shut down for inspection. The shutdown criterion is generally defined in terms of seismic damage indicating parameters. This paper presents seismic damage indicating parameters of the recorded free-field and in-structure motions at the Onagawa, Fukushima Daiichi, Fukushima Daini, and Tokai Daini nuclear power plants during the 2011 Tohoku-oki earthquake. The observed seismic damage indicating parameters largely exceed the current U.S. and Japan shutdown thresholds for nuclear power plants, while minimal damage due to ground shaking was observed at the four Japanese plants. These observations indicate a potential for raising the current threshold without introducing any significant additional seismic risk to nuclear power plants. The insights presented in this paper can be used to guide regulation and industry methods for quickly evaluating the damage potential of future earthquakes that affect nuclear power plants. With some adjustment, a similar methodology and criterion could be applied to conventional structures and lifeline infrastructure.

When the response of a structure is influenced significantly by multiple vibration modes, the seismic demands determined by response history analyses with ground motions selected to match a single conditional mean spectrum (CMS) may be unconservative for several response quantities. An existing solution to this problem is to determine the demand as the maximum of mean demands from several CMSs. To reduce the computational effort while maintaining accuracy and precision, we introduce an alternative engineering solution that is based on a simplified version of the generalized CMS, where two specified conditioning spectral accelerations share a common hazard level. vector-valued disaggregation is developed herein to provide a single earthquake scenario for the target spectrum. The results from a realistic case study suggest that the proposed spectrum provides seismic demands that are as accurate and precise as those obtained from analyzing the structure with multiple CMSs while simultaneously reducing the computational effort by a factor of 2 or more.

This manuscript reports ground motion intensity measure (IM) correlations for a number of IM types, as measured from the NGA-West2 database. IMs considered are Spectral Accelerations with periods from 0.01 s to 10 s, Peak Ground Acceleration, Peak Ground Velocity, and Significant Duration (for 5–75% and 5–95% definitions). Results are shown for correlations of both maximum-direction and geometric mean spectral acceleration values, given the need for such maximum-direction correlations in a new ASCE 7-16 procedure. Additionally, the potential magnitude-, distance- and site-condition-dependence of IM correlations are evaluated. The results are practically important as IM correlations are increasingly used in a range of engineering and seismic hazard calculations. We find that maximum-direction spectral correlations are comparable to correlations for other spectral acceleration definitions, and that the correlations have no practically significant dependence on magnitude, distance or site conditions. These results support the collective understanding that IM correlations are stable across a range of conditions, and as a result, that existing correlation models are generally appropriate for continued use in engineering calculations.

We develop a model to predict topographic effects on ground motions using the NGA-West2 data set. Topography at a site is quantified using three parameters: smoothed curvature, smoothed slope, and relative elevation. We examine the relationship between the site residuals, which represent the average modeling error of the reference ground motion prediction equation at a station, and the three topographic parameters. Our analysis shows that there are statistically significant trends in site residuals with respect to relative elevation and smoothed curvature. We propose modification factors to the Chiou and Youngs (2014) model that account for topographic effects on both the median and the standard deviation. These factors are period dependent and are a function of the relative elevation at the site. The factors result in amplification on median by a factor as high as 1.13 at T = 0.5 s for higher values of relative elevation, and deamplification as low as 0.75 between T = 2 s to 3 s for lower values of relative elevations.

The Fourier amplitude spectrum (FAS) is widely used in seismology and earthquake engineering as it provides valuable information regarding frequency dependent amplitude of the ground motion. However, for structural design and assessment, the preferred representation of seismic hazard continues to be based on the elastic response spectrum. Therefore, conversions between these spectra are often required. In this article, the connection between FAS and the 5% damping pseudo-acceleration response spectrum (5% PSA) is explored using large data sets of spectrum-compatible time series generated from white noise. The strong dependence of the relation between FAS and 5% PSA with strong motion duration is evidenced and a duration dependent empirical relationship between the both spectra is developed. The equation is validated using recorded ground motions and spectrum-compatible time series generated from the modification of these ground motions. The equation allows simpler one-step conversions when compared to iterative approaches based on RVT theory or time-consuming methodologies that require the generation of spectrum-compatible time series.

A methodology for seismic performance assessment of onshore buried steel pipelines at fault crossings is presented. Probabilistic fault displacement hazard analysis is performed at first to determine the magnitude of the three fault displacement components in space. Next, three-dimensional (3-D) structural analysis of the pipeline via a nonlinear beam-type finite element model allows accounting for the different effect of imposed displacements in each axis. Finally, convolving of seismic hazard and structural response results in joint hazard surfaces of compressive and tensile strains that can be used to estimate the mean annual rate of exceeding any limit-state of interest under the influence of demand and capacity uncertainty.

This paper describes a detailed field survey conducted at Lo Rojas fishermen port in Coronel, where extensive liquefaction-induced lateral spread was reported for the 2010, M_w 8.8 Maule earthquake. The survey includes SPT and SCPT soundings, as well as the use of surface-based geophysical techniques. The data was used to evaluate a multilinear regression (MLR) lateral-spread expression and to develop a detailed hydro-mechanical finite element model. Results of the MLR equation were over-conservative and proved to be very sensitive to the distance from the site to the energy source. On the other hand, the results from the numerical model agreed reasonably well with the post-event field observations.

When evaluating the liquefaction hazard within a performance-based framework, whether using simplified procedures or advanced numerical tools, the hazard and its effects on structures need to be evaluated under a range of ground motions. Choice of an optimum intensity measure (IM) in the selection and scaling of ground motions will reduce variability in the predicted response, dependence on source characteristics, and uncertainty in the prediction of the IM. This paper presents the results of a numerical parametric study, validated against centrifuge results, to evaluate the influence of different IMs on the liquefaction hazard in the far-field and near shallow-founded structures. Pore pressure redistribution and soil-structure interaction were considered in estimating the liquefaction hazard in terms of peak excess pore pressure ratio (r _u,peak ). The IMs at the base rock, far-field soil surface, and foundation with the best combination of efficiency, sufficiency, and predictability were evaluated and identified as: (1) pseudo-spectral acceleration at the site's initial fundamental period (PSA _Base [T _So ]) for predicting r _u,peak in the far-field; (2) peak ground acceleration, (PGA _Base ); and Arias intensity (AI _Base ) for predicting r _u,peak under the foundation.

A reliable mitigation of the liquefaction hazard requires an accurate estimation of the consequences of liquefaction in the context of building performance. Knowledge and use of an optimal intensity measure (IM) will reduce variability and improve accuracy of the predicted measure of performance. This paper presents the results of a three-dimensional, fully coupled, nonlinear, dynamic parametric numerical simulation of shallow-founded structures on layered, liquefiable soils, previously validated with centrifuge results. The generation and redistribution of excess pore pressures as well as soil-structure interaction effects were directly considered in the simulations. The influence of different IMs recorded at the base rock, far-field soil surface, and foundation was evaluated and compared on engineering demand parameters (EDP) that relate to structural performance and damage potential, such as foundation settlement and peak, transient, inter-story drift ratios. The IMs identified with the best combination of efficiency, sufficiency, and predictability in predicting the structural EDPs of interest were identified at the base rock as: CAV and CAV₅ for permanent settlement, PSA[T _STo ] for total and flexural drift ratios, and PGV for rocking drift ratio.

This study presents an open and transparent exposure model for the residential building stock in South America. This model captures the geographical distribution, structural characteristics (including information about construction materials, lateral load resisting system, range of number of stories), average built-up area, replacement cost, expected number of occupants, and number of dwellings and buildings. The methodology utilized to develop this model was based on national population and housing statistics and expert judgment from dozens of local researchers and practitioners. This model has been developed as part of the South America Risk Assessment (SARA) project led by the Global Earthquake Model (GEM), and it can be used to perform earthquake risk analyses. It is available at different geographical scales for seven Andean countries: Argentina, Bolivia, Chile, Colombia, Ecuador, Peru, and Venezuela (DOI: 10.13117/GEM. DATASET.EXP.ANDEAN-v1.0).

The present data paper describes an experimental campaign on five thin T-shaped reinforced concrete walls (DOI: 10.6084/m9.figshare.3490754.v2), which includes: details on the test units, materials, test setup, loading protocol, instrumentation, main features of each unit's response, organization of the provided test data, and examples of derived data. The tests aimed at assessing the influence of wall thickness on member stability, the role of lap splices on damage distribution and displacement ductility, and the effects of the simultaneous application of out-of-plane loading on the member response. A set of five companion test reports, one for each of the tested units, are included in the data set and supplement the present manuscript.

A densely instrumented system of large-scale underground structures consisting of two vertical shafts connected through a cut-and-cover tunnel and two independent shield tunnels was installed in an 8 m-diameter laminar soil box at the E-Defense shake table in Miki, Japan. The system was subjected to step-sine sweeps and scaled ground motion records of the Kobe (1995) earthquake. The underground structures were embedded in Albany Silica Sand with an average relative density of 54%. System instrumentation consisted of over 800 sensors, including strain gauges, accelerometers, displacement transducers, bender elements and pressure sensors. A U.S.-Japanese research collaboration was established to instrument the vertical shaft elements and record seismic soil pressures. Data records are archived at the NHERI DesignSafe Data Depot_and can be used to analyze the structural response, soil-structure interaction and other response parameters of individual subsurface components as well as the entire system. The DOI for the data set is 10.17603/DS21C78.