Publications (24)29.88 Total impact
 [Show abstract] [Hide abstract]
ABSTRACT: Establishing a set of groundmotion prediction equations (GMPEs) for Japan requires earthquake source categories in the dataset. Earthquakes are typically divided into three groups: shallow crustal events that occur in the Earth’s crust, subduction interface events that occur at the interface between the crust or mantle and the subducting plate, and the subduction slab events that occur within the subducting plate. In the present study, we compared the hypocentral locations published in the catalogs of the International Seismological Centre/Engdahl–van der Hilst–Buland (ISCEHB; Engdahl et al., 1998), the Japan Meteorological Agency (JMA), and the National Earthquake Information Center (NEIC) of the U.S. Geological Survey (USGS). The hypocentral location for the same earthquake varies significantly from one catalog to another. We used the subduction interface model from the USGS, Slab 1.0, to help guide the classification. We designed four classification schemes using locations from these three catalogs. We then fitted a set of random effects models to the strongmotion dataset from these earthquakes to assess the merits of the classification schemes. Our results showed that using ISCEHB locations for events before 2005, and then using the preference order of catalogs as (1) JMA locations with high precision levels, (2) ISCEHB, and (3) NEIC (excluding the events with a fixed depth) for events since 2005, together with some conditions for subduction interface events, produced the best GMPEs in terms of the maximum log likelihood. We also found that having a separate group for the earthquakes above the subduction interface, but with a depth over 25 km, improved the goodness of fit of the GMPEs.  [Show abstract] [Hide abstract]
ABSTRACT: This paper studies the runout of earthquakeinduced landslides. There are more than 20 mechanical models to explain long runout of rapid landslides including some earthquakeinduced landslides. However, notably few of them considered the effect of seismic loading on the runout of landslides. In a previous study, we have proposed a model, which is called the multiplex acceleration model (MAM), to interpret the long runout mechanism, and a shakingtable test has been performed to verify the MAM. Because the previous MAM is a conceptual model and only a single stone has been examined under sine waves by the shakingtable test, the MAM needs to be extended and further verified. In this paper, the MAM was extended by introducing the movement change that was induced by the socalled trampoline effect of earthquake loading, and the latest practical numerical simulation program, discontinuous deformation analysis (DDA), was used to verify the mobility of earthquakeinduced landslides. After a conceptual landslide model was used to verify the effect of seismic loading on the mobility, the Donghekou landslide, which was a typical long runout landslide that was induced by the 2008 Wenchuan earthquake, was analyzed. The results show that: the seismic loading could be one of the factors that could eventually help increase the runout of landslides.  [Show abstract] [Hide abstract]
ABSTRACT: This paper extends the discontinuous deformation analysis (DDA) by using an additional evaluation of edgetoedge contact, with the aim that it can be used to accurately model the failure behaviour of joints dominated by both cohesion and interface friction angle. The original DDA can deal well with the effects of interface friction angle. However, when cohesion exists, DDA results often show an inscrutable behaviour, i.e. a slope may be unstable even if the cohesion is much greater than the theoretical value required for its critical stability. After many detailed investigations and validations, joint contact treatment was found to be the key reason why the original DDA cannot simulate the cohesive material accurately, in which every edgetoedge contact is treated as two vertextoedge contacts that may have different contact states associated with different cohesion treatments. In order to solve this problem, an additional contact type determination process for an edgetoedge contact was added into the original computer code to avoid the unreasonable situation when two contact states exist in one joint. Several examples were performed to illustrate the accuracy of the modified code and a real landslide case was analysed by using the improved DDA to estimate the shear strength on the interface. Our results show that the improved DDA can simulate the failure of cohesive frictional material accurately.  [Show abstract] [Hide abstract]
ABSTRACT: VS30, the shearwave travel time averaged soil shearwave velocity of the top 30 m, has been used to represent site effects in many recent groundmotion prediction equations (GMPEs). However, while VS30 has been found to be a reasonable parameter to represent site effects in some studies, other studies provide contradictory evidence. In the present study, a systematic comparison between the predictive capabilities of these two siteeffect parameters is carried out using a large groundmotion dataset from Japan. The basis of the adopted approach is to compare the standard deviations and amplitudes of amplification ratios in empirically modeling site effects by using either site period (TS, four times the shearwave travel time from the bedrock to the ground surface) or VS30. The site effects modeled specifically include site amplification ratios between surface and borehole records from KiKnet, in addition to the siteeffect terms from a GMPE. For KiKnet data, TS is determined to be a better predictive parameter than VS30 for soil sites with TS > 0.6 s, while the two parameters lead to a similar variability in amplification ratios for sites with TS < 0.6 s. For site effects obtained from the GMPE, VS30 and TS are statistically equal for all site classes at most periods, while VS30 leads to smaller variability than TS at some spectral periods. The conflict between the KiKnet surfaceborehole records, and the results from the GMPE is likely to be a result of large variability in the GMPE, containing source, path, and sitevariability, as compared with the reduced variability in the surfaceborehole KiKnet data pairs. Although VS30 and TS lead to statistically similar standard deviations for the data from a GMPE, TS still leads to better median amplification ratios than VS30. 
Article: Empirical models for predicting lateral spreading considering the effect of regional seismicity
[Show abstract] [Hide abstract]
ABSTRACT: A revised empirical model has been developed for predicting liquefactioninduced lateral spreading displacement (LD) as a function of both response spectral acceleration derived from strongmotion attenuation models and geotechnical parameters from Youd’s LD data set (Youd website). This revised model is different from the model of Zhang and Zhao, which overcame some drawbacks of earlier models for predicting lateral spreading and was primarily used in Japan and the western U.S. The revised model can potentially be applied anywhere if ground shaking (in terms of 5% damped acceleration or displacement response spectra) can be estimated using local strongmotion attenuation relationships. The revised model is examined using data from Japan and the western U.S. and applied to Turkey and New Zealand, where the ground shaking is estimated using appropriate strongmotion attenuation relationships for each region. The accuracy of the revised model is evaluated by comparing its predicted lateral displacements with those measured in actual earthquakes. The results show that the revised model can account for the effects of local seismicity on lateral spreading displacements and is comparable with existing prediction models.  [Show abstract] [Hide abstract]
ABSTRACT: Magnitudescaling rates (MSRs; the rates of increase in predicted response spectra with increasing moment magnitude) are evaluated for three groundmotion prediction equations for response spectra from subduction interface earthquakes, including two empirical models developed for data from Japan and a model based on synthetic records generated by using a stochastic finitefault model. MSRs vary significantly among the three models, and the difference between the two empirical models is unacceptably large. A set of 2100 strongmotion records from subduction interface events with a magnitude of 6.5 or larger from Japan, including the 11 March 2011 magnitude 9 earthquake, were compiled. The earthquakes were grouped according to magnitude, so that the magnitude spread in each group is less than 0.2 magnitude units. Each earthquake group was treated as a single event with magnitude equal to the average magnitude for the group. A simple attenuation model was fitted to the normalized and grouped data. The model has a constant term for each earthquake group to represent the effect of magnitude. Three separate functions of magnitude (a linear model for events with a magnitude greater than 7, a bilinear model, and a curved model) were then fitted to the constants, and MSRs were derived from these functions. At short periods, the derived MSRs are only a fraction of those from two of the three attenuation models. At spectral periods greater than 0.5 s, the derived MSRs are between about 1/3 and 1/2 of those of the two empirical models but are closer to those based on a set of synthetic records.  [Show abstract] [Hide abstract]
ABSTRACT: We have evaluated the magnitudescaling rates (MSRs, the rate of increase in the predicted spectrum with increasing moment magnitude) of five modern groundmotion 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 23 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 strongmotion 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 510, a significant increase in shaking duration, which cannot be fully accounted for by response spectra, and limited or no increase in groundmotion 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 magnitudescaling rate.  [Show abstract] [Hide abstract]
ABSTRACT: We compared response spectra from the M(w) 7.9 2008 Wenchuan earthquake with five modern groundmotion prediction equations (GMPEs). Ninetythree strongmotion 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 strongmotion records in moving and overlapping time windows. The average shearwave velocity of the top 30 m (V(S30)) was only available for two stations. V(S30) was extrapolated from the average of the top 20 m (V(S20)) 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 0.51.0 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 0.1s period and 2.3 at the 5.0s 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.  [Show abstract] [Hide abstract]
ABSTRACT: The great 2008 Wenchuan earthquake (12 May 2008) with a moment magnitude of 7.9 and a surfacewave magnitude of 8.0 in Shichuan, China, caused unprecedented loss of human life and widespread severe damage to many types of structures. Thirtytwo strongmotion records were obtained within a source distance of 300 km, and three nearsource records were obtained within a source distance of 20 km. We present the preliminary results on the characteristics of the nearsource records and the strongmotion aspects of this great earthquake. This earthquake may be divided into four subevents, according to the rupture time history and the final slip distribution. Three of the four subevents have large surface fault displacement, and we consider the subevents that generated the three nearsource records as surfacerupture earthquakes, supported by the comparisons made to records from other surfacerupture or buriedfault earthquakes. One station recorded strong ground motions from two subevents in two wellseparated time windows, and this allows us to examine the effect of earthquake parameters for each of the subevents. We find that, in the spectral period range of 0.52 s, the response spectra of the nearsource records from the Wenchuan earthquake are significantly less than those of buriedfault earthquakes, such as the 1989 Loma Prieta earthquake and the 1994 Northridge earthquake that have a much smaller moment magnitude than the Wenchuan earthquake. In the faultnormal direction the displacement spectra at long period for the closest station are similar to those of the Lucerne record from the 1992 Landers earthquake but significantly smaller than those of the TCU052 and TCU068 records from the 1999 ChiChi, Taiwan, earthquake. At short and intermediate period, the nearsource spectra are much larger than the design spectra in the previous version of the Chinese design code for the heavily damaged area, but they are comparable at long spectral periods.  [Show abstract] [Hide abstract]
ABSTRACT: Our previous studies show that site effects (amplification of rock motions), source and path effects are coupled when response spectra are used to characterize the amplification ratios for a soil site modelled as nonlinear or elastic. The coupling is referred to as a “side effect” of using response spectral amplification ratios. In the present study we use a suite of rock site records, well distributed with respect to magnitude and source distance, from crustal, subduction interface and slab earthquakes to evaluate the response spectral amplification ratio for soft soil sites. We compare these sideeffects for ground motions generated by three types of earthquakes, and we find that, at periods much shorter or much longer than the natural period of a soil site modelled as elastic, the average amplification ratios with respect to rock site ground motions from three types of earthquakes are moderately different and are very similar for other spectral periods. These differences are not statistically significant because of the moderately large scatter of the amplification ratios. However, the extent of magnitude and sourcedistancedependence of amplification ratios differs significantly. After the effects of magnitude and source distance on the amplification ratios are accounted for, the differences in amplification ratios between crustal and subduction earthquake records are very large in some particular combinations of source distance and magnitude range. These findings may have potential impact in establishing design spectra for soft soil sites using strong motion attenuation models or numerical modelling.  [Show abstract] [Hide abstract]
ABSTRACT: Attenuation models derived from recorded ground motions are still important elements of probabilistic seismic hazard studies. Engineers use empirical attenuation models to derive the displacement demand for a site of interest from an earthquake at a given location. Many attenuation models have been published for different parts of the world and for different types of earthquakes. Most models have a simple function of constant or magnitudedependent geometric spreading, and seldom consider wellknown seismological effects such as Moho reflection for shallow crustal earthquakes, multiple travel paths and constructive interference for subduction earthquakes, and special characteristics of volcano zones. The reason for not accounting for such effects may be the desire for simplicity in the attenuation functional forms for engineering applications and a lack of records from which to reliably identify these effects quantitatively. In this article, a large set of strongmotion records obtained from dense recording networks in Japan is used to derive geometric attenuation functional form and a possible manner to model the effect of volcanic zones. A liberal approach is taken to introduce a relatively large number of parameters that can account for known seismological effects while retaining a fairly simple attenuation functional form, based on analyses of residuals from simple models similar to those published previously. Preliminary results are reported here, together with the proposed geometric attenuation function forms and plausible explanation of the physical process that leads to the proposed geometric attenuation functions. The proposed model shows a large increase in the maximum likelihood from the random effects methodology, the elimination of bias in the distribution of residuals with respect to source distance, and much improved fitting for wellrecorded earthquakes.  [Show abstract] [Hide abstract]
ABSTRACT: We investigate a special type of variability in response spectral amplification ratios computed from numerical “engineering” models for a soft soil site. The engineering models are defined by shallow soil layers over “engineering” bedrock with a shearwave velocity over 600–700m/s and the model is subjected to vertical propagating shear waves. The variability, perhaps unique in earthquake engineering, is a result of the “perfectly accurate” computational procedure. For example, an engineering soil site model, subjected to two rock site records or the two horizontal components of a rock site record, produces different response spectral amplification ratios. We use a large number of strongmotion records from “engineering” rock sites, with a reasonably balanced distribution with respect to magnitude and source distance, generated by subduction earthquakes in Japan, to investigate the nature of the variability. In order to avoid any approximation in removing the effect of soil nonlinear response, we use a simple model, a single horizontal soil layer over a bedrock, modelled as elastic. We then demonstrate that a similar type of variability observed in the one or twodimensional nonlinear soil models is caused by the nature of response spectral amplification ratios, not a direct result of soil nonlinear response. Examination of variability reveals that the average of response spectral amplification ratios systematically depends on both earthquake magnitude and source distance. We find that, at periods much longer than the site natural periods of the soil sites, the scatter of the amplification ratios decreases with increasing magnitude and source distance. These findings may have a potential impact in establishing design spectra for soft soil sites using strongmotion attenuation models or dynamic numerical modelling.  [Show abstract] [Hide abstract]
ABSTRACT: In current groundmotion models, the uncertainty in predicted ground motion is usually modeled with a lognormal distribution. One consequence of this is that predicted ground motions do not have an upper limit. In reality, however, there probably exist physical conditions that limit the ground motion. Applying the usual uncertainty distribution in probabilistic seismic hazard analysis may lead to groundmotion estimates that are unrealistically large, especially at the low annual probabilities considered for important structures, such as dams or nuclear reactors. A recently proposed statistical procedure to compare the actual and expected numbers of predicted spectral accelerations exceeding a given value gives clear results when applied to a groundmotion model developed for Japan from a very large strongmotion data set. It shows that, for increasingly large spectral accelerations, the actual number of exceedances becomes progressively less than the expected number of exceedances. The pattern of this discrepancy depends on the site class and the earthquake tectonic category. These results suggest that assuming a normal distribution for the prediction errors of an attenuation model (empirical groundmotion prediction equation) is likely to result in overestimation of the extreme values of spectral accelerations. 
Article: Response spectral amplification ratios from 1 and 2dimensional nonlinear soil site models
[Show abstract] [Hide abstract]
ABSTRACT: In order to determine the effect of geometry on the ground response of 2dimensional (2D) basins filled with soils that can develop nonlinear response, we use three basin models with width/depth ratios 3, 6 and 10. The three basins are subjected to a suite of rock site records with various magnitudes and source distances. We compute response spectral amplification ratios at four locations on the surface of the 2D basins, and determine the average variation of the amplification ratios with respect to excitation spectra, for peak ground acceleration (PGA) and 3 spectral periods of 0.2, 0.5, 1s. Similarly, we compute the average response spectral amplification ratios for two 1dimensional (1D) nonlinear models, one having the soil profile at the basin centre and the other having a soil profile at half the depth of the basin. From the relationship between the average amplification ratios and excitation spectra, we determine the crossover point in terms of excitation spectral values that separate the amplification range from the deamplification range. Our results show that the crossover point varies significantly from one location to another on the ground surface and from one basin to another, in a range of 0.3–1.1g for PGA. The effects of basin geometry are very strong at weak and moderate excitation, but decrease with increasing excitation spectra in a significant portion around the basin centre. Our results provide some justification for using 1D models for 2D basins with a width/depth ratio ⩾6 if the soil site is subjected to strong ground shaking.  [Show abstract] [Hide abstract]
ABSTRACT: As part of the "It"s Our Fault" project, we are working on estimating ground motions from large plate boundary earthquakes at specified locations in the Wellington region in terms of response spectra and acceleration time histories. These motions may provide synthetic strongground time histories for a future major earthquake. For engineering applications in NZ, considering the high frequency content of a synthetic accelerogram is a vital part of any dynamic loading analysis. To do this we need to produce broadband accelerograms for which we use an empirical Green"s function technique that was developed by overseas researchers. First we characterize the fault parameters using waveform inversion. Then, we define a suitable set of source parameters, such as the area of fault plane, moment magnitude, slip distribution (fault heterogeneity) within the fault plane and the propagation pattern of the rupture. With these parameters, we can generate synthetic accelerograms that contain the signature of all parameters for the specific fault. The method can also be used for a future earthquake using fault model parameters derived from empirical scaling functions. The synthetic records will not only contain the required response spectra but also appropriate duration of strong ground shaking specifically for a given fault. We will present preliminary results from our initial trial using the strong motion dataset from the 2003 Fiordland earthquake. The work presented here may have farreaching effect for selecting accelerograms for a particular site.  [Show abstract] [Hide abstract]
ABSTRACT: A New empirical model has been developed for predicting liquefactioninduced lateral spreading displacement and is a function of response spectral displacements and geotechnical parameters. Different from the earlier model of Zhang and Zhao (2005), the application of which was limited to Japan and California, the new model can potentially be applied anywhere if ground shaking can be estimated (by using local strongmotion attenuation relations). The new model is applied in New Zealand where the response spectral displacement is estimated using New Zealand strongmotion attenuation relations (McVerry et al. 2006). The accuracy of the new model is evaluated by comparing predicted lateral displacements with those which have been measured from aerial photos or the width of ground cracks at the Landing Road bridge, the James Street loop, the Whakatane Pony Club and the Edgecumbe road and rail bridges sites after the 1987 Edgecumbe earthquake. Results show that most predicted errors (defined as the ratio of the difference between the measured and predicted lateral displacements to the measured one) from the new model are less than 40%. When compared with earlier models (Youd et al. 2002, Zhang and Zhao 2005), the new model provides the lowest mean errors.  [Show abstract] [Hide abstract]
ABSTRACT: There is considerable interest in the credibility of probabilities of exceedance estimated by groundmotion models for very high accelerations. A common statistical approach to this problem has been to examine the uppertail shape of the distribution of residuals between recorded data and the model for evidence of suppression of high residuals. In this study, a more direct method is suggested, in which the actual number of times given accelerations are exceeded is compared to the expected numbers in strongmotion data sets. The method is illustrated by application to New Zealand and Japan models for peak ground acceleration (PGA). For the Japan model, which is based on a particularly large data set, the ratio of actual to expected number declines in a statistically significant and regular fashion from about 1 at 0.3g to about 0.15 at 1.0g. If these results are indicative of groundmotion models in general, the implications for probabilistic seismic hazard analyses may be far reaching. The method and results have particular importance for the analysis of seismic hazard at sites of critical facilities where strong ground motions with very long return periods may be of interest.  [Show abstract] [Hide abstract]
ABSTRACT: A very useful tool for the preliminary design of structures is the elastic demand spectrum that can be used in the capacity spectrum method. A pseudoacceleration relationship has to be assumed when constructing a demand spectrum. This assumption results in large errors for long period structures with large damping ratios and the conventional demand spectra require a substitute elastic structure. In the present study, the conventional demand spectra are extended to bilinear models. Pseudoacceleration is still assumed but results in acceptably small errors, when a constant viscous damping coefficient for a singledegreeoffreedom (SDF) structure is calculated from the tangent stiffness and the damping ratio is set at 5% in both elastic and yield phases. For nonlinear structures, tangent stiffness dependency of damping force could be acceptable because energy absorption is primarily the result of structural nonlinear deformation. To extend the conventional demand spectra to a bilinear model, effective period calculated from the secant stiffness has to be used. The use of effective period introduces no approximation because the peak displacement of the SDF structure is computed from nonlinear analysis in the time domain. The method presented in this study is also valid if damping coefficient proportional to initial elastic spectra is used. In this case, the pseudoacceleration is defined as the base shear coefficient that is required to produce the peak displacement of the SDF structure in a static manner. We present demand spectra of bilinear models for a number of nearsource records from large earthquakes, and spectral ratios of two horizontal components. The effects of different types of ground motion on the response reduction factor due to inelastic deformation are investigated.  [Show abstract] [Hide abstract]
ABSTRACT: Attenuation relations have been developed for 5% damped acceleration response spectra in New Zealand earthquakes. The models take account of the different tectonic types of earthquakes in New Zealand, i.e., crustal, subduction interface and dipping slab, and of the faster attenuation of highfrequency components in the volcanic region. The study used all available data from the New Zealand earthquake accelerograph network that satisfied various selection criteria, supplemented by selected data from digital seismographs. The latter provide additional records from moderateto highstrength rock sites, and of motions involving propagation paths through the volcanic region. Most of the accelerograph sites are on soil, with the few accelerograph rock sites generally being on weak rock. The data have been further augmented by seismograph records from a temporary deployment in the volcanic region. To constrain the model at short distances where New Zealand records are lacking, overseas peak ground acceleration data recorded less than 10 km from the source were included. It was found that New Zealand earthquake motions are mostly similar to those from other parts of the world for the same tectonic class, especially for crustal earthquakes, but that crustal and subduction zone earthquake motions have different spectral shapes.  [Show abstract] [Hide abstract]
ABSTRACT: A preliminary spectral attenuation model derived from JMA strong motion data is presented. A consistent record processing procedure and consistently determined focal depths were adopted, and the regression analysis followed a random effects method. The JMA data set does not have enough nearfield data to support model parameters that could reliably control the nearfield behaviour. Residuals analyses show that the attenuation model can adequately predict recorded spectra for subduction events but tends to overestimate the response spectra from interslab events in subduction zones, when the model does not explicitly account for the nature of the tectonic source.
Publication Stats
201  Citations  
29.88  Total Impact Points  
Top Journals
Institutions

20112015

Southwest Jiaotong University
 School of Civil Engineering
Huayang, Sichuan, China


20082012

GNS Science
Lower Hutt City, Wellington, New Zealand
