Prediction of long-period ground motions from huge subduction earthquakes in Osaka,Japan

There is a high possibility of reoccurrence of the Tonankai and Nankai earthquakes along the Nankai Trough in Japan. It is very important to predict the long-period ground motions from the next Tonankai and Nankai earthquakes with moment magnitudes of 8.1 and 8.4, respectively, to mitigate their disastrous effects. In this study, long-period (>2.5 s) ground motions were predicted using an earthquake scenario proposed by the Headquarters for Earthquake Research Promotion in Japan. The calculations were performed using a fourth-order finite difference method with a variable spacing staggered-grid in the frequency range 0.05–0.4 Hz. The attenuation characteristics (Q) in the finite difference simulations were assumed to be proportional to frequency (f) and S-wave velocity (V _s) represented by Q = f · V _s / 2. Such optimum attenuation characteristic for the sedimentary layers in the Osaka basin was obtained empirically by comparing the observed motions during the actual M5.5 event with the modeling results. We used the velocity structure model of the Osaka basin consisting of three sedimentary layers on bedrock. The characteristics of the predicted long-period ground motions from the next Tonankai and Nankai earthquakes depend significantly on the complex thickness distribution of the sediments inside the basin. The duration of the predicted long-period ground motions in the city of Osaka is more than 4 min, and the largest peak ground velocities (PGVs) exceed 80 cm/s. The predominant period is 5 to 6 s. These preliminary results indicate the possibility of earthquake damage because of future subduction earthquakes in large-scale constructions such as tall buildings, long-span bridges, and oil storage tanks in the Osaka area.     

Lessons from the 2003 Tokachi-oki, Japan, earthquake for prediction of long-period strong ground motions and sloshing damage to oil storage tanks

The 2003 Tokachi-oki earthquake (M _w 8.0) in northern Japan generated large-amplitude long-period (4–8 s) ground motions in the Yufutsu sedimentary basin, causing severe damage to seven large oil storage tanks with floating roof structures because of severe sloshing of oil. The 30,000–40,000-m³ tanks having suffered the severe damage such as fires and sinking of floating roofs experienced the sloshing with large amplitudes exceeding 3 m in which the fundamental mode was predominant. The second mode of sloshing was also excited in the 110,000-m³ tanks in which their floating roofs sank into oil, indicating that the higher modes of sloshing as well as the fundamental mode should be considered in damage prediction. The strong ground motion recordings demonstrated the earthquake dependency of predominant periods and the substantial spatial variation of the long-period shaking observed within the Yufutsu basin, meaning the necessity of source- and site-specific prediction of long-period strong ground motions. The two-dimensional numerical modeling suggested the importance of detailed structures of soft near-surface sediments as well as deep basin structure for accurate prediction of long-period strong ground motions in deep sedimentary basins.     

Validation of 3-D basin structure models for long-period ground motion simulation in the Osaka basin, western Japan

We studied the applicability of two types of existing three-dimensional (3-D) basin velocity structure models of the Osaka basin, western Japan for long-period ground motion simulations. We synthesized long-period (3–20 s) ground motions in the Osaka basin during a M6.5 earthquake that occurred near the hypothetical Tonankai earthquake source area, approximately 200 km from Osaka. The simulations were performed using a 3-D finite-difference method with nonuniform staggered grids using the two basin velocity structure models. To study the ground motion characteristics inside the basin, we evaluated the wave field inside the basin using the transfer functions derived from the synthetics at the basin and a reference rock site outside the basin. The synthetic waveforms at the basin site were obtained by a convolution of the calculated transfer function and the observed waveform at the reference rock site. First, we estimated the appropriate Q values for the sediment layers. Assuming that the Q value depends on the S wave velocity V _S and period T, it was set to Q = (1/3V _S)(T ₀/T) where V _S is in m/s and the reference period T ₀ is 3.0 s. Second, we compared the synthetics and the observations using waveforms and pseudovelocity response spectra, together with a comparison of the velocity structures of the two basin models. We also introduced a goodness-of-fit factor to the pseudovelocity response spectra as an objective index. The synthetics of both the models reproduced the observations reasonably well at most of the stations in the central part the basin. At some stations, however, especially where the bedrock depth varies sharply, there were noticeable discrepancies in the simulation results of the models, and the synthetics did not accurately reproduce the observation. Our results indicate that the superiority of one model over the other cannot be determined and that an improvement in the basin velocity structure models based on simulation studies is required, especially along the basin edges. We also conclude that our transfer function method can be used to examine the applicability of the basin velocity structure models for long-period ground motion simulations.     

基于汶川地震远场强震动记录的厚覆盖土层对长周期地震动影响分析

利用2008年5月12日汶川8．0级地震的远场强震动记录,研究太原盆地厚覆盖土层对长周期地震动的影响,并将其结果与地震安全性评价工作中的土层反应分析结果进行比较。结果表明：厚覆盖土层对长周期地震动有很强的放大作用,覆盖层厚度越厚,峰值出现的周期越大,而相对于基岩最大放大可达9倍左右。另外,对于厚覆盖层的场地,观测结果与场地地震安全性评价计算结果在一定的长周期范围内差别相当大,相对于基岩,观测结果的放大系数至少高于场地地震安全性评价计算结果的2倍以上,最大达到近9倍。     

山地与平原地形对地震动影响的对比分析

山地和平原地形在破坏性地震中对震害和地震动有很大影响,不同地形对地震动有不同的地形效应。为研究山地和平原地形对地震动的影响规律,基于集集地震后续M L6.6余震的强震观测记录,对位于不同地形且与震中位于同一直线上的四个强震台记录处理计算,从波形特征、频谱分析、加速度反应谱计算、与规范谱比较等方面进行对比分析,完成长周期地震动反应谱分布的计算。研究表明,山地地形对地震动的高频成分有地形放大效应,沉积平原对地震动的低频长周期部分有放大作用,并且长周期反应谱的峰值区域始终位于沉积平原中部。平原上强震记录产生的长周期放大系数谱已超出规范谱,沉积平原上长周期建筑的抗震设防应引起重视。     

Validation of the Deep Velocity Structure of the Tokachi Basin Based on 3-D Simulation of Long-Period Ground Motions

The construction of 3-D basin velocity structures is ongoing in many regions of Japan. The structure models are constructed mainly for the prediction of long-period ground motions from future large earthquakes. In this paper, we validate the 3-D velocity structure model of the Tokachi basin, a deep sedimentary basin located in eastern Hokkaido, Japan, based on 3-D simulation of long-period (2–20 s) ground motions from three nearby intermediate-depth earthquakes; this model was constructed by the National Research Institute for Earth Science and Disaster Prevention (NIED). We make comparisons between the observed and synthetic long-period ground motions for the basin-induced surface waves as well as the direct S-wave. We also try to revise the 3-D velocity structure in the western part of the Tokachi basin based on 1-D velocity structures estimated using long-period S-wave modeling and the microtremor survey method. We then perform the 3-D simulation again to validate the revised model. Based on quantitative comparisons of the long-period ground motions from these simulations with those observed, we conclude that the NIED and revised velocity structure models are generally good at the central basin sites, but that both models require modification at the basin edges to explain the details of the observed basin-induced surface waves.     

Estimation of strong ground motions from hypothetical earthquakes on the Cascadia subduction zone,Pacific Northwest

Strong ground motions are estimated for the Pacific Northwest assuming that large shallow earthquakes, similar to those experienced in southern Chile, southwestern Japan, and Colombia, may also occur on the Cascadia subduction zone. Fifty-six strong motion recordings for twenty-five subduction earthquakes ofM _s7.0 are used to estimate the response spectra that may result from earthquakesM _w<81/4. Large variations in observed ground motion levels are noted for a given site distance and earthquake magnitude. When compared with motions that have been observed in the western United States, large subduction zone earthquakes produce relatively large ground motions at surprisingly large distances. An earthquake similar to the 22 May 1960 Chilean earthquake (M _w 9.5) is the largest event that is considered to be plausible for the Cascadia subduction zone. This event has a moment which is two orders of magnitude larger than the largest earthquake for which we have strong motion records. The empirical Green's function technique is used to synthesize strong ground motions for such giant earthquakes. Observed teleseismicP-waveforms from giant earthquakes are also modeled using the empirical Green's function technique in order to constrain model parameters. The teleseismic modeling in the period range of 1.0 to 50 sec strongly suggests that fewer Green's functions should be randomly summed than is required to match the long-period moments of giant earthquakes. It appears that a large portion of the moment associated with giant earthquakes occurs at very long periods that are outside the frequency band of interest for strong ground motions. Nevertheless, the occurrence of a giant earthquake in the Pacific Northwest may produce quite strong shaking over a very large region.     

Finite-element simulations of long-period ground motions: Japanese subduction-zone earthquakes and the 1906 San Francisco earthquake

Large earthquakes at shallow depths commonly excite long-period ground motions in distant sedimentary basins, thereby inflicting damage upon large-scale structures. For example, the 2003 Tokachi-oki earthquake, Japan, damaged oil tanks in the Yufutsu Basin, located 250 km from the epicenter. Similar long-range effects were also observed during the 2004 earthquake off Kii Peninsula, Japan. In this study, we present the results of simulations of these earthquakes undertaken using the finite element method (FEM) with a voxel mesh. In addition, to examine whether the 1906 San Francisco earthquake excited long-period ground motions in the Los Angeles-area basins, we performed long-period ground motion simulations of most of the California region. The FEM simulations confirmed the importance of path effects for the development of long-period ground motions.     

Development of Long-period Ground Motions from the Nankai Trough, Japan, Earthquakes: Observations and Computer Simulation of the 1944 Tonankai (Mw 8.1) and the 2004 SE Off-Kii Peninsula (Mw 7.4) Earthquakes

Strong ground motions recorded in central Tokyo during the 1944 Tonankai Mw8.1 earthquake occurring in the Nankai Trough demonstrate significant developments of very large (>10 cm) and prolonged (>10 min) shaking of long-period (T > 10–12 s) ground motions in the basin of Tokyo located over 400 km from the epicenter. In order to understand the process by which such long-period ground motions developed in central Tokyo and to mitigate possible future disasters arising from large earthquakes in the Nankai Trough, we analyzed waveform data from a dense nation wide strong-motion network (K-NET and KiK-net) deployed across Japan for the recent SE Off-Kii Peninsula (Mw 7.4) earthquake of 5 September 2004 that occurred in the Nankai Trough. The observational data and a corresponding computer simulation for the earthquake clearly demonstrate that such long-period ground motion is primarily developed as the wave propagating along the Nankai Trough due to the amplification and directional guidance of long-period surface waves within a thick sedimentary layer overlaid upon the shallowly descending Philippine Sea Plate below the Japanese Island. Then the significant resonance of the seismic waves within the thick cover of sedimentary rocks of the Kanto Basin developed large and prolonged long-period motions in the center of Tokyo. The simulation results and observed seismograms are in good agreement in terms of the main features of the long-period ground motions. Accordingly, we consider that the simulation model is capable of predicting the long-period ground motions that are expected to occur during future Nankai Trough M 8 earthquakes.     

3D crustal structure and long-period ground motions from a M9.0 megathrust earthquake in the Pacific Northwest region

We have developed a community velocity model for the Pacific Northwest region from northern California to southern Canada and carried out the first 3D simulation of a M_w 9.0 megathrust earthquake rupturing along the Cascadia subduction zone using a parallel supercomputer. A long-period (<0.5 Hz) source model was designed by mapping the inversion results for the December 26, 2004 Sumatra–Andaman earthquake (Han et al., Science 313(5787):658–662, 2006) onto the Cascadia subduction zone. Representative peak ground velocities for the metropolitan centers of the region include 42 cm/s in the Seattle area and 8–20 cm/s in the Tacoma, Olympia, Vancouver, and Portland areas. Combined with an extended duration of the shaking up to 5 min, these long-period ground motions may inflict significant damage on the built environment, in particular on the highrises in downtown Seattle.     

Site Response from Ambient Vibrations in the Towns of Lod and Ramle (Israel) and Earthquake Hazard Assessment

Throughout their long history, the towns of Lod and Ramle have been severely affected by strong earthquakes. The last destructive earthquake occurred on July 11, 1927 and caused the destruction of large parts of these cities, reaching a seismic intensity of VIII–IX on the MSK scale. Such a high intensity from a relatively distant earthquake (about 70 km) of magnitude 6.2 is likely to be the result of local site effects of the sedimentary layers that may have significantly enhanced earthquake ground motions. This study is focused on estimating the seismic hazard to Lod and Ramle by implementing a three-step process (1) detailed mapping of the characteristics of the H/V spectral ratios from ambient noise, (2) incorporating geological information and well data to construct subsurface models for different sites within the investigated area and (3) estimating the seismic hazard in terms of uniform hazard site-specific accelerations. The horizontal-to-vertical spectral ratios of ambient noise were used to approximate the fundamental resonance frequencies of the subsurface and their associated amplitudes. About 360 sites in Lod and Ramle were instrumented for varying periods. The soil sites exhibits H/V peak amplitudes ranging from 4 to 6 in the frequency range 0.5–2.5 Hz. These data were used to constrain 1-D subsurface models that were developed using geological data and borehole information. H/V spectral ratio observations were checked against theoretical subsurface transfer functions at locations where borehole information is available farther constraint the range of possible V_s velocities of the different layers and thus, by means of trial an error it was possible to conclude a systematic spatial distribution of the V_s velocity and thickness in the substrata that are also consistent with the spatial distribution of the fundamental resonance frequencies of the soft sediments obtained by means of the H/V spectral ratios, and other geological and geophysical information available at different locations in the study area. The evaluated subsurface models are introduced using the SEEH procedure of Shapira and van~Eck [(1993) Natural Hazards 8, 201–205] to assess Uniform Hazard Site-Specific Acceleration Spectra for different zones within the towns of Lod and Ramle. These evaluations are very important for realistic assessment of the vulnerabilities of all types of existing and newly designed structures and for urban and land use planning.     

An evaluation of 3-D velocity models of the Kanto basin for long-period ground motion simulations

We performed three-dimensional (3-D) finite difference simulations of long-period ground motions (2–10 s) in the Kanto basin using the Japan Seismic Hazard Information Station (J-SHIS 2009), Yamada and Yamanaka (Exploration Geophysics 65(3):139–150, 2012) (YY), and Head Quarter for Earthquake Research Promotion (HERP 2012) velocity models for two intermediate depth (68–80 km) moderate earthquakes (Mw 5.8–5.9), which occurred beneath the Kanto basin. The models primarily differ in the basic data set used in the construction of the velocity models. The J-SHIS and HERP models are the results of integration of mainly geological, geophysical, and earthquake data. On the other hand, the YY model is oriented towards the microtremor-array-observation data. We obtained a goodness of fit between the observed and synthetic data based on three parameters, peak ground velocities (PGVs), smoothed Fourier spectra (FFT), and cross-correlations, using an algorithm proposed by Olsen and Mayhew (Seism Res Lett 81:715–723, 2010). We found that the three models reproduced the PGVs and FFT satisfactorily at most sites. However, the models performed poorly in terms of cross-correlations especially at the basin edges. We found that the synthetics using the YY model overestimate the observed waveforms at several sites located in the areas having V _s 0.3 km/s in the top layer; on the other hand, the J-SHIS and HERP models explain the waveforms better at the sites and perform similarly at most sites. We also found that the J-SHIS and HERP models consist of thick sediments beneath some sites, where the YY model is preferable. Thus, we have concluded that the models require revisions for the reliable prediction of long-period ground motions from future large earthquakes.     

A seismological overview of long-period ground motion

Long-period ground motion has become an increasingly important consideration because of the recent rapid increase in the number of large-scale structures, such as high-rise buildings and oil storage tanks. Large subduction-zone earthquakes and moderate to large crustal earthquakes can generate far-source long-period ground motions in distant sedimentary basins with the help of path effects. Near-fault long-period ground motions are generated, for the most part, by the source effects of forward rupture directivity. Far-source long-period ground motions consist primarily of surface waves with longer durations than near-fault long-period ground motions. They were first recognized in the seismograms of the 1968 Tokachi-oki and 1966 Parkfield earthquakes, and their identification has been applied to the 1964 Niigata earthquake and earlier earthquakes. Even if there is no seismogram, we can identify far-source long-period ground motions through the investigation of tank damage by liquid sloshing.     

Long-period ground motions from the 2003 Tokachi-oki earthquake

We study characteristics of long-period ground motions from the 2003 Tokachi-oki earthquake (Mj 8.0), a large interplate earthquake, based on spatial distribution maps and attenuation relationships for four kinds of peak ground velocity (PGV) value. The first kind (PGV(WB)) is obtained from a maximal value of vector sum of the three-component, wide-band velocity seismograms, and the other three kinds (PGV(BP10), PGV(BP20), and PGV(BP30)) are obtained from a maximal value of vector sum of the three-component, narrow band-pass filtered velocity seismograms (the central periods are 10, 20, and 30 s). The spatial distribution maps for all kinds of PGV value show azimuth dependence; the PGV values in Hokkaido, northern side of the epicenter are larger than those in Tohoku, southwestern side of the epicenter, when compared at a comparable distance. We find that the features result from the radiation pattern of long-period surface waves, that is, the source effect. The attenuation relationships show the following trends: The PGV(WB) values are larger than the sum of the PGV(BP10), PGV(BP20), and PGV(BP30) at distances (D) less than 200 km, while the PGV(WB) values are comparable to the sum of the PGV(BP20) and PGV(BP30) at D > 200 km. This indicates that the PGV(WB) values at D < 200 km are affected by ground motions with periods less than 10 s, while long-period surface waves mainly contribute to the PGV(WB) values at D > 200 km. The basin site effects generate a patchy pattern in the spatial distribution maps and a large scattering in the attenuation relationships for the PGV(WB) and PGV(BP10) values. Finally, we conclude that the PGV(WB) values from the 2003 Tokachi-oki earthquake are controlled by the radiation pattern of long-period S and surface waves and various basin site effects.     

Prediction of site factors by a non‐parametric approach

The problem addressed in this paper is the estimation of the (de)amplification of ground motion at soil sites (compared to rock sites) as a function of the intensity of the ground motion. A non‐parametric empirical approach, called the Conditional Average Estimator (CAE) method, has been used, which is different from all existing approaches. Site factors (SFs) for sites characterized with V_s30 between 180 and 360 m/s were predicted for the peak ground acceleration (PGA) and the spectral accelerations by using a combined database of recorded ground motions. Based on the results of the study, site factors for PGA and selected spectral accelerations are proposed, separately for weaker and stronger ground motions. Comparisons are made with the SFs used in two standards (Eurocode 8 and ASCE 7‐10) and with SFs proposed in the literature, including four Next Generation Attenuation (NGA) ground‐motion prediction equations. The study reveals that (i) SFs depend strongly on the ground‐motion intensity. In the case of stronger ground motions, they decrease with increasing acceleration. (ii) The SFs predicted in this study agree reasonably well with the existing SFs in the case of weak ground motion. For higher intensities of ground motion, they are generally smaller than the existing ones. Copyright © 2014 John Wiley & Sons, Ltd.     

Basin and crustal velocity structure models for the simulation of strong ground motions in the Kinki area,Japan

We constructed a prototype of the basin and crustal structure model for the Kinki area, southwest of Japan, for the simulation of strong ground motions of hypothetical crustal and subduction earthquakes. We collected results of the deep seismic velocity profiles obtained by the reflection experiments and seismic imaging results, which were conducted in the Kinki area. The obtained profiles give underground velocity structures of the crust, from the surface to the subducting slab. We also gather the basin velocity structure information of the Osaka, Kyoto, Nara, and Ohmi basins. To examine the applicability of the constructed velocity structure model to the ground motion simulation, we simulated waveforms of an intermediate size event occurred near the source area of the hypothetical subduction earthquakes. Simulated ground motions using the basin and crustal velocity structure model are fairly well reproducing the observations at most of stations, and the constructed basin and crustal velocity structure model is applicable for the long-period ground motion simulations.     

The 2011 off the Pacific coast of Tohoku earthquake and response of high-rise buildings under long-period ground motions

In the afternoon of March 11, 2011, the eastern Japan was severely attacked by the 2011 off the Pacific coast of Tohoku earthquake (the Great East Japan earthquake). Nearly 30,000 people were killed or are still missing by that earthquake and the ensuing monster tsunami as of April 11, 2011. This paper reports some aspects of this devastating earthquake which hit an advanced country in seismic resistant design. It has been reported that long-period ground motions were induced in Tokyo, Nagoya and Osaka. The properties of these long-period ground motions are discussed from the viewpoint of critical excitation and the seismic behavior of two steel buildings of 40 and 60 stories subjected to the long-period ground motion recorded at Shinjuku, Tokyo is determined and discussed. This paper also reports the effectiveness of visco-elastic dampers like high-hardness rubber dampers in the reduction of responses of super high-rise buildings subjected to such long-period ground motions. The response reduction rate is investigated in detail in addition to the maximum response reduction. In December 2010 before this earthquake, simulated long-period ground motions for earthquake resistant design of high-rise buildings were provided in three large cities in Japan (Tokyo, Nagoya and Osaka) and nine areas were classified. Two 40-story steel buildings (slightly flexible and stiff) are subjected to these long-period ground motions in those nine areas for the detailed investigation of response characteristics of super high-rise buildings in various areas.     

长周期地震动强度指标与SDOF体系残余变形相关性分析

为分析两类长周期地震动作用下如何选取合理的强度指标作为结构抗震设计的输入,从2011年东日本9.0级地震、2003年十胜冲8.0级地震和2016年熊本7.3级地震数据库中选取90条可靠的远场长周期地震动。从1994年美国北岭6.7级地震和1999年台湾集集7.6级地震数据库中选取60条近场长周期地震动,以SDOF体系为研究对象,讨论了阻尼比、屈服刚度折减系数和强度折减系数对残余变形与18个地震动强度指标的相关系数的影响,对比分析了两类长周期地震作用下相关系数之间的异同。研究结果表明:在考虑近场长周期地震动作用时,建议应根据结构周期的大小来选择合适的强度指标作为地震动的输入。远场长周期地震动作用下,以PGA、PGV和PGD为代表的强度指标与残余变形的相关程度均较高,PGV稳定性略好于PGA和PGD,建议PGV作为地震动输入的控制指标。残余变形相关系数受阻尼比、屈服刚度折减系数以及强度折减系数影响不大。     

基于《建筑抗震设计规范》强度包线函数参数取值研究

利用《中国地震动参数区划图》采用的地震动参数衰减关系,以及《中国地震动参数区划图》中地震动峰值加速度和地震动加速度反应谱特征周期反推不同设防烈度和设计地震分组对应的震级和震中距,再根据《建筑抗震设计规范》中各设防水准的峰值加速度确定对应的震级和震中距,进而根据地震动强度包线参数与震级和震中距关系计算地震动强度包线参数的取值,为基于强度包线函数生成人工地震动提供参考,并讨论强度包线参数的取值规律：(1)随着设防烈度的提高,加速度时程曲线上升段持续时间t₁和平稳段持续时间t_s减小,下降段衰减指数c增大;(2)随着地震水准和设计地震分组的提高,加速度时程曲线上升段持续时间t₁和平稳段持续时间t_s增加,下降段衰减指数c减小;(3)在生成人工地震动时,除考虑峰值加速度和设计地震分组影响外,还需要考虑设防烈度影响。     

表层沉积物特性对地震地面运动的影响研究

建立包含震源、沉积盆地和表层低速沉积层的二维模型,采用交错网格有限差分/伪谱混合方法求解地震波传播,讨论沉积层厚度和速度对地震地面运动的作用。结果表明:沉积层内产生的地震波的多重反射以及转换会引起地面运动持续时间的延长,它们的相干叠加会造成地面运动峰值的放大;随着沉积层速度的增加,多重反射与转换波的能量减小,地面运动持续时间减小,但是不同速度或者不同厚度的低速层模型均显示出一致的地面运动峰值放大特征。结果说明,在包含震源、沉积盆地和沉积层的模型中,沉积层对地面运动的作用机理更复杂。在实际应用中有必要同时考虑这些因素的综合作用。