Next Generation Attenuation Models Research Articles

Ground Motion Observation of Sabah Earthquakes on the Use of Next Generation Attenuation (NGA) Ground-Motion Models

Ground motion prediction equations (GMPEs) are being used for the estimation of the ground motion parameters which are needed for the design and evaluation of important structures. The seismic hazard may contribute greatly to the total risk; therefore the selection of appropriate GMPEs may have a substantial influence on the design and safety evaluation. For low-seismicity areas, however, the available database of strong ground motion measurements is limited, with determination of an appropriate GMPE been a rather difficult task. The objective of this study is to evaluate the next generation attenuation (NGA) ground-motion models to be applied in Sabah region. In this study, six next generation attenuation (NGA) models have been selected to be evaluated. The representation of all NGA models, are compared with the Sabah ground motion database comprises 209 two horizontal-component acceleration time series recorded within 10 to 1000 km of source to site distances for 173 earthquakes with moment magnitudes (M W ) ranging between 3.0–6.0. The comparisons are made using analyses of root of the mean square (RMS) and residuals. Two GMPEs present better residual fits than other models with smaller RMS value and indicates better estimation of the peak ground acceleration (PGA). Based on these findings, it is recommended on using the NGA relations for seismic hazard assessment of Sabah.

Equivalent Point-Source Ground-Motion Model for Subduction Earthquakes in Japan

ABSTRACTWe use an equivalent point-source ground-motion model (GMM) to characterize subduction earthquakes (interface and in-slab) in Japan. The model, which is calibrated using the newly published Next Generation Attenuation (NGA) Subduction database (Bozorgnia et al., 2020), provides a useful complement to the more traditional empirical NGA models developed from the same database. The utility of the point-source model approach lies in its ability to aid in the interpretation of observed trends in the data and to guide modifications to the GMM for application to other regions having fewer data. Key trends in the data that are parameterized with the model include: (a) the enrichment of high-frequency amplitudes for in-slab versus interface events, as modeled by a depth-dependent stress parameter, and (b) attenuation attributes that vary with event type and region, including consideration of fore-arc versus back-arc settings. The developed GMMs of this study are applicable for M 4.5–9.2 for interface events, and M 4–8.5 for in-slab earthquakes, for rupture distances (Drup) from 10 to 600 km, and for 100 m/s<VS30<1500 m/s (time-averaged shear-wave velocity in the top 30 m).

Verification of Liquefaction Potential during the Strong Earthquake at the Border of Thailand-Myanmar

ABSTRACT The study presents a liquefaction potential analysis for the Thailand-Myanmar border area due to the earthquake. Three sites reported liquefaction during Tarlay earthquake in 2011 were investigated including a microtremor to obtain the shear wave velocity profile and depth of engineering bedrock. Next-generation attenuation model was used to estimate the peak ground acceleration at the liquefied sites. A seismic ground response analysis was then conducted, and the result revealed that the shallow sand layers were very vulnerable to undergo liquefaction. The results also showed a similar tendency to the predictions based on the empirical methods.

Performance of NGA Models in Predicting Ground Motion Parameters of The Strong Earthquake

Next Generation Attenuation (NGA) West 1 and 2 models are employed to predict the ground motion parameters of strong earthquake during the 6.9 Mw Kobe Earthquake in 1995. This study is initiated by collecting the data of ground motion parameters of the earthquake. Furthermore, the ground motion prediction is performed by using the NGA models. There are three ground motion parameters observed, i.e. peak ground acceleration (PGA), spectral acceleration (SA) at 0.2 second and SA at 1 second. The performances of the models are evaluated by using the Residual Values and Root Mean Square (RMS) Error. The results show that the NGA models could predict the ground motion parameters quite appropriately. It can be seen from the correlation values of the observed and the predicted values, which is relatively consistent each other, especially for peak ground acceleration. In general, this study could recommend the procedure in selecting the attenuation model for strong earthquakes. The study framework could be implemented to predict the ground motion in other regions.

Analysis of seismic ground response caused during strong earthquake in Northern Thailand

This paper presents a site-specific analysis of ground response during the Tarlay Earthquake on March 24, 2011 in Northern Thailand. In this study, the NGA (Next Generation Attenuation) models were selected to predict ground motions due to the earthquake event. The equivalent linear and non-linear approaches were employed in the one-dimensional ground response analysis. Furthermore, the spectral responses produced by the equivalent linear and non-linear approaches were compared with the seismic design code of Thailand. The results showed that the ground motion from the NGA models agree with the strong motion parameters of Tarlay Earthquake. Peak ground acceleration (PGA) at ground surface obtained from both equivalent linear and non-linear approaches certainly results in the high amplification factor. In general, the study results could bring an attention to the local engineer to consider the seismic design value for Northern Thailand, particularly if the stronger earthquake happens in the future.

Applicability of different ground-motion prediction models for northern Iran

A total of 163 free-field acceleration time histories recorded at epicentral distances of up to 200 km from 32 earthquakes with moment magnitudes ranging from M w 4.9 to 7.4 have been used to investigate the predictive capabilities of the local, regional, and next generation attenuation (NGA) ground-motion prediction equations and determine their applicability for northern Iran. Two different statistical approaches, namely the likelihood method (LH) of Scherbaum et al. (Bull Seismol Soc Am 94:341–348, 2004) and the average log-likelihood method (LLH) of Scherbaum et al. (Bull Seismol Soc Am 99:3234–3247, 2009), have been applied for evaluation of these models. The best-fitting models (considering both the LH and LLH results) over the entire frequency range of interest are those of Ghasemi et al. (Seismol 13:499–515, 2009a) and Soghrat et al. (Geophys J Int 188:645–679, 2012) among the local models, Abrahamson and Silva (Earthq Spectra 24:67–97, 2008) and Chiou and Youngs (Earthq Spectra 24:173–215, 2008) among the NGA models, and finally Akkar and Bommer (Seism Res Lett 81:195–206, 2010) among the regional models.

Net earthquake hazard and elements at risk (NEaR) map creation for city of Istanbul via spatial multi-criteria decision analysis

The creation of earthquake hazard maps requires various datasets with selected attenuation relations. Based on the selected attenuation relation, the calculation time varies from half an hour to a couple of days. The length of time needed to create an earthquake hazard map also depends on the resolution of the resulting map. The time gets longer as the resolution of the resulting earthquake hazard map gets higher. The basic form of an attenuation relation requires complex calculation algorithms including geospatial information related to the region of interest. Nowadays, next-generation attenuation (NGA) models are introduced to generate more realistic earthquake hazard maps. However, the more complex the attenuation relation is, the longer time will be required to create a hazard map. This paper offers a new method to create high-resolution earthquake hazard maps, faster than using traditional attenuation relation methods, by using an analytic hierarchy process of spatial multi-criteria decision analysis and geographic information systems. This method has been generated and tested for the city of Istanbul. The resulting maps are compared with the earthquake hazard maps created for the city of Istanbul by using the NGA model of Boore and Atkinson (in Boore–Atkinson NGA ground motion relations for the geometric mean horizontal component of peak and spectral ground motion parameters (trans: Engineering Co, University of California B). Pacific Earthquake Engineering Research Center 2007). A second output of this paper is a map of the elements at risk (EaR) for the population and buildings of Istanbul, and the introduction of a new approach of net elements at risk (NEaR).

Comparisons among the five ground-motion models developed using RESORCE for the prediction of response spectral accelerations due to earthquakes in Europe and the Middle East

This article presents comparisons among the five ground-motion models described in other articles within this special issue, in terms of data selection criteria, characteristics of the models and predicted peak ground and response spectral accelerations. Comparisons are also made with predictions from the Next Generation Attenuation (NGA) models to which the models presented here have similarities (e.g. a common master database has been used) but also differences (e.g. some models in this issue are nonparametric). As a result of the differing data selection criteria and derivation techniques the predicted median ground motions show considerable differences (up to a factor of two for certain scenarios), particularly for magnitudes and distances close to or beyond the range of the available observations. The predicted influence of style-of-faulting shows much variation among models whereas site amplification factors are more similar, with peak amplification at around 1s. These differences are greater than those among predictions from the NGA models. The models for aleatory variability (sigma), however, are similar and suggest that ground-motion variability from this region is slightly higher than that predicted by the NGA models, based primarily on data from California and Taiwan.

Selection of Ground Motion Prediction Models for Seismic Hazard Analysis in the Zagros Region, Iran

The main objective of this article is to assess a set of candidate ground motion models in the Zagros region of Iran. The candidate models were chosen from three categories: local models that were developed based on the local data, regional models corresponding to Europe and Middle East data sets, and finally NGA (Next Generation Attenuation) models. Two different statistical approaches were applied for evaluation of these models, the first being the “likelihood” method, and the second the average “log-likelihood” method (LH and LLH). One of the most significant results of this study is that local ground motion models show more consistency with the recorded data than do NGA models.

Updated PGA, PGV, and Spectral Acceleration Attenuation Relations for Iran

We have developed updated attenuation relations for peak ground acceleration (PGA), peak ground velocity (PGV), and acceleration response spectra with 5% damping on the basis of the data (78 earthquakes and 351 records) pertaining to strong ground motion in Iran. Moment magnitude, distance, fault mechanism, site class, and zone were the model parameters considered. A term for the saturation of the acceleration amplitude was also used in the model in order to improve the estimations in near-source regions. A nonlinear regression analysis was performed to obtain the coefficients. A comparison between the data set used in the current study for Iran and two next generation attenuation (NGA) models showed good correlation between our model and the Campbell-Bozorgnia NGA model. The model described is applicable for moment magnitudes from 5.0 to 7.3, distances from 15 to 135 km, and site classes with an average shear-wave velocity at a subsurface depth of 30 m (AVS30) of more than 175 m/s.

Magnitude-Scaling Rate in Ground-Motion Prediction Equations for Response Spectra from Large, Shallow Crustal Earthquakes

Abstract We have evaluated the magnitude-scaling rates (MSRs, the rate of increase in the predicted spectrum with increasing moment magnitude) of five modern ground-motion prediction equations (GMPEs) for response spectra, including four Next Generation Attenuation (NGA) models and a model developed for data from Japan. We have found that MSRs for crustal earthquakes with a moment magnitude over 7 vary significantly among the five models, by a factor of 2–3 for some cases. The variation of MSRs among the four NGA models is alarmingly large, considering that they were derived from largely the same dataset and used the same site parameters. We have selected 641 strong-motion records from shallow crustal earthquakes with a moment magnitude over 7 from the NGA dataset and 69 from the 2008 Wenchuan, China, earthquake with a moment magnitude of 7.9, all within a short distance of 200 km from the fault rupture plane. To illustrate the extreme extent of magnitude scaling, we have fitted an attenuation model without a magnitude term to this dataset, based on an observation that an increase in one moment magnitude unit is related to an increase in fault length by a factor of 5–10, a significant increase in shaking duration, which cannot be fully accounted for by response spectra, and limited or no increase in ground-motion amplitude. The statistical analyses of the results indeed suggest that a zero magnitude scaling at spectral periods over 0.6 s may be reasonable for our dataset, while the required apparent magnitude scaling at short periods may be due to other factors such as stress drop, a parameter that is not used in any of the models considered. We will provide some plausible explanations for the possible zero magnitude scaling that can be considered as the lower limit of the uncertain magnitude-scaling rate.

3D Simulations of M 7 Earthquakes on the Wasatch Fault, Utah, Part I: Long-Period (0-1 Hz) Ground Motion

Abstract We predict ground motions in the Salt Lake basin (SLB) during M 7 earthquakes on the Salt Lake City segment of the Wasatch fault (WFSLC). First we generate a suite of realistic source representations by simulating the spontaneous rupture process on a planar, vertical fault with the staggered-grid split-node finite-difference (FD) method. The initial distribution of shear stress is the sum of both a regional depth-dependent shear stress appropriate for a dipping, normal fault and a stochastically generated residual shear stress field associated with previous ruptures. The slip-rate histories from the spontaneous rupture scenarios are projected onto a detailed 3D model geometry of the WFSLC that we developed based on geological observations. Next, we simulate 0- to 1-Hz wave propagation from six source models with a 3DFD code, using the most recent version of the Wasatch Front Community Velocity model. Horizontal spectral accelerations at two seconds (2-s SAs) reveal strong along-strike rupture direction effects for unilateral ruptures, as well as significant amplifications by the low-velocity sediments on the hanging-wall side of the fault. For ruptures nucleating near the southern end of the segment, we obtain 2-s SAs of up to 1.4 g near downtown SLC, caused by a combination of rupture-direction and basin-edge effects. Average 3-s SAs and 2-s SAs from the six scenarios are generally consistent with values predicted by four next-generation attenuation models.

Modeling the Joint Probability of Earthquake, Site, and Ground-Motion Parameters Using Bayesian Networks

Abstract Bayesian networks are a powerful and increasingly popular tool for reasoning under uncertainty, offering intuitive insight into (probabilistic) data-generating processes. They have been successfully applied to many different fields, including bioinformatics. In this paper, Bayesian networks are used to model the joint-probability distribution of selected earthquake, site, and ground-motion parameters. This provides a probabilistic representation of the independencies and dependencies between these variables. In particular, contrary to classical regression, Bayesian networks do not distinguish between target and predictors, treating each variable as random variable. The capability of Bayesian networks to model the ground-motion domain in probabilistic seismic hazard analysis is shown for a generic situation. A Bayesian network is learned based on a subset of the Next Generation Attenuation (NGA) dataset, using 3342 records from 154 earthquakes. Because no prior assumptions about dependencies between particular parameters are made, the learned network displays the most probable model given the data. The learned network shows that the ground-motion parameter (horizontal peak ground acceleration, PGA) is directly connected only to the moment magnitude, Joyner–Boore distance, fault mechanism, source-to-site azimuth, and depth to a shear-wave horizon of 2.5 km/s (Z2.5). In particular, the effect of V S 30 is mediated by Z2.5. Comparisons of the PGA distributions based on the Bayesian networks with the NGA model of Boore and Atkinson (2008) show a reasonable agreement in ranges of good data coverage.

Ground-Motion Modeling of Hayward Fault Scenario Earthquakes, Part II: Simulation of Long-Period and Broadband Ground Motions

Abstract We simulate long-period ( T >1.0–2.0 s) and broadband ( T >0.1 s) ground motions for 39 scenario earthquakes ( M w 6.7–7.2) involving the Hayward, Calaveras, and Rodgers Creek faults. For rupture on the Hayward fault, we consider the effects of creep on coseismic slip using two different approaches, both of which reduce the ground motions, compared with neglecting the influence of creep. Nevertheless, the scenario earthquakes generate strong shaking throughout the San Francisco Bay area, with about 50% of the urban area experiencing modified Mercalli intensity VII or greater for the magnitude 7.0 scenario events. Long-period simulations of the 2007 M w 4.18 Oakland earthquake and the 2007 M w 5.45 Alum Rock earthquake show that the U.S. Geological Survey’s Bay Area Velocity Model version 08.3.0 permits simulation of the amplitude and duration of shaking throughout the San Francisco Bay area for Hayward fault earthquakes, with the greatest accuracy in the Santa Clara Valley (San Jose area). The ground motions for the suite of scenarios exhibit a strong sensitivity to the rupture length (or magnitude), hypocenter (or rupture directivity), and slip distribution. The ground motions display a much weaker sensitivity to the rise time and rupture speed. Peak velocities, peak accelerations, and spectral accelerations from the synthetic broadband ground motions are, on average, slightly higher than the Next Generation Attenuation (NGA) ground-motion prediction equations. We attribute much of this difference to the seismic velocity structure in the San Francisco Bay area and how the NGA models account for basin amplification; the NGA relations may underpredict amplification in shallow sedimentary basins. The simulations also suggest that the Spudich and Chiou (2008) directivity corrections to the NGA relations could be improved by increasing the areal extent of rupture directivity with period.

Comparison of Strong Ground Motion from the Wenchuan, China, Earthquake of 12 May 2008 with the Next Generation Attenuation (NGA) Ground-Motion Models

A total of 72 free-field accelerograms recorded at rupture distances of less than 200 km during the Wenchuan earthquake were used to compare the Wenchuan earthquake ground motions with those predicted by the recent Next Generation Attenuation (NGA) ground-motion models developed using global data from shallow crustal earthquakes, but few have recordings from earthquakes with magnitudes as large as 7.9. Overall, the Wenchuan earthquake produced smaller than expected long-period ( T >1 s) ground motions, but higher than expected short-period ( T <0.5 s) ground motions compared with the predicted motions by the NGA models. This general trend is observed at all distances, but it is stronger at distances greater than 50 km. The scaling with V S 30 from the Wenchuan earthquake is not consistent with the scaling from the NGA models: the Wenchuan data show a much stronger V S 30 dependence at short spectral periods and a weaker V S 30 dependence at long spectral periods. The short-period hanging wall effects on rock sites are consistent with the NGA hanging wall scaling, but the short-period ground motions on soil sites on the footwall are greater than those predicted by the NGA models.

A Comparison of Recorded Response Spectra from the 2008 Wenchuan, China, Earthquake with Modern Ground-Motion Prediction Models

We compared response spectra from the M w 7.9 2008 Wenchuan earthquake with five modern ground-motion prediction equations (GMPEs). Ninety-three strong-motion records within 300 km of the fault plane were selected for comparison with the GMPE models of Zhao, Zhang et al. (2006), Abrahamson and Silva (2008), Boore and Atkinson (2008), Campbell and Bozorgnia (2008) and Chiou and Youngs (2008) for spectral periods up to 5.0 s. The site class of the recording stations used for the Zhao, Zhang et al. (2006) model was inferred from response spectral ratios of the horizontal and vertical components (H/V) computed from the strong-motion records in moving and overlapping time windows. The average shear-wave velocity of the top 30 m ( V S 30) was only available for two stations. V S 30 was extrapolated from the average of the top 20 m ( V S 20) when possible and inferred from the H/V response spectral ratios when necessary. The average predictions of all models were acceptable. The Zhao, Zhang et al. (2006) model gave the best predictions for peak ground acceleration and short spectral periods, especially up to 100 km of the source distance. All Next Generation Attenuation (NGA) models predicted the recorded spectra very well for periods of 0.5–1.0 s and at 5.0 s. The Chiou and Youngs (2008) model gave the best overall predictions. The standard deviations of all attenuation models were similar at a 5% significance level. However, differences between spectra estimated by various NGA models were statistically and practically significant, with the largest difference between the average predictions being nearly a factor of 1.4 at the 0.1-s period and 2.3 at the 5.0-s period for data within a source distance of 100 km. Although one earthquake did not produce median ground motions that the GMPEs are designed to predict, such a large difference represents a challenge for empirical models when estimating spectra from very large crustal earthquakes.

Near-Source Strong Ground Motion Characteristics of the 2008 Wenchuan Earthquake

The free-field strong ground motions recorded at 64 sites within 200 km from the ruptured Longmen Shan fault during the 12 May 2008 M w 7.9 Wenchuan earthquake were used to examine the near-source strong ground motioncharacteristics. The magnitude of vertical and horizontal peak ground accelerations (PGAs), frequency content, and duration of acceleration histories are analyzed. The observed peak ground accelerations at distances exceeding 60 km from the ruptured fault were significantly higher than those predicted by the Next Generation Attenuation (NGA) model, whereas, for periods of larger than 1 s, the spectral accelerations are much smaller than those of the NGA model. The same is also true for the observed peak ground velocities (PGVs). The duration of ground shaking is longer than for most earthquakes of comparable similar magnitude. For sites within 60 km from the ruptured fault, the accelerations observed on the hanging wall are larger than those observed on the footwall. The duration of ground shaking along the rupture direction (or the forward direction) is shorter than that opposite to the rupture direction (or the backward direction). There is also a systematic increase of PGA and PGV at sites located in the direction of the rupture propagation, whereas no such clear trend was found in the stations located along the backward direction. From the observed acceleration time histories at various locations, the rupture process of the 2008 Wenchuan appears to be more complicated than the focal mechanism interpreted from far-field seismic records. For example, the generation mechanism of the two distinct wave trains observed at Wolong and other stations need more detailed studies.

Evaluation of Evidence for Inhibition of Very Strong Ground Motions in the Abrahamson and Silva Next Generation Attenuation Ground-Motion Model

Abstract A modification to the Rhoades et al. (2008) procedure for evaluating inhibition of very strong ground motions in strong-motion models is developed that accounts for the correlation in the total residuals. The modified method treats the intraevent and interevent residuals separately. The alternative method is applied to the Abrahamson and Silva (2008) Next Generation Attenuation (NGA) ground-motion model and strong-motion data set. We find that there is no statistical basis to indicate that the Abrahamson and Silva (2008) NGA model overpredicts the occurrence of very large ground motions.

Directivity Correction for the Next Generation Attenuation (NGA) Relations

Using ground motion residuals of four NGA models, directivity effect and the dependence of directivity on various seismological and geometrical parameters were studied and correction terms for the four models were obtained. Directivity was clearly seen in the ground motion residuals of most earthquakes in the NGA database. Specifically, this phenomenon was found to be significant for large earthquakes, but it was not consistently observed for small events. It was seen to strongly depend on period of ground motion and on source distance, having its largest values in the very near-source region and for long periods. This study further demonstrated that the intensity of directivity depends on the geometry of the fault, orientation of fault surface with respect to the site, and the location of hypocenter. Applying the correction terms to the NGA models resulted in considerable reductions in the uncertainty term sigma, especially at long periods and at short distances.

A Test of the Applicability of NGA Models to the Strong Ground-Motion Data in the Iranian Plateau

The Next Generation Attenuation (NGA) project has now published several new sets of empirical ground-motion prediction equations (GMPEs) for PGA, PGV, and response spectral ordinates. These models significantly advance the state-of-the-art empirical ground-motion modeling and account for many effects that have not been directly accounted for in the existing Iranian GMPEs. Assuming that the present strong-motion database in Iran is unlikely to drastically change in the near future, the question we ask in this study is: Can the NGA models be applied in Iran? In order to answer this question, the NGA models of Boore and Atkinson [2008], Campbell and Bozorgnia, [2008], and Chiou and Youngs [2008], which are shown to be representative of all NGA models, are compared with the Iranian strong-motion database. The database used in this study comprises 863 two-component horizontal acceleration time series recorded within 100 km of epicentral distances for 166 earthquakes in Iran with magnitudes ranging from 4.0–7.4. The comparisons are made using analyses of residuals. The analysis indicates that the NGA models may confidently be applied within the Iranian plateau. To provide more reliable constraint on finite-fault effects and nonlinear site response in the Iranian equations, it would be useful to drive new GMPEs based on a merger of the NGA and Iranian databases.