Masahiro Iida

Excitation of high‐frequency surface waves with long duration in the Valley of Mexico

During the 1985 Michoacan earthquake (Ms = 8.1), large-amplitude seismograms with extremely long duration were recorded in the lake bed zone of Mexico City. We interpret high-frequency seismic wave fields in the three geotechnical zones (the hill, the transition, and the lake bed zones) in the Valley of Mexico on the basis of a systematic analysis for borehole strong motion recordings. We make identification of wave types for real seismograms. First, amplitude ratios between surface and underground seismograms indicate that predominant periods of the surface seismograms are largely controlled by the wave field incident into surficial layers in the Valley of Mexico. We interpret recorded surface waves as fundamental-mode Love waves excited in the Mexican Volcanic Belt by calculating theoretical amplification for different-scale structures. Second, according to a cross-correlation analysis, the hill and transition seismograms are mostly surface waves. In the lake bed zone, while early portions are noisy body waves, late portions are mostly surface waves. Third, using two kinds of surface arrays with different station intervals, we investigate high-frequency surface-wave propagation in the lake bed zone. The wave propagation is very complicated, depending upon the time section and the frequency band. Finally, on the basis of a statistical time series model with an information criterion, we separate S- and surface-wave portions from lake bed seismograms. Surface waves are dominant and are recognized even in the early time section. Thus high-frequency surface waves with long duration in the Valley of Mexico are excited by the Mexican Volcanic Belt.

A Comprehensive Interpretation of Strong Motions in the Mexican Volcanic Belt

Large-amplitude seismograms with extremely long duration have been recorded in the Valley of Mexico. Recently, we came to recognize important effects of the Mexican Volcanic Belt (MVB) on the seismograms. We interpret strong motions observed in the MVB by analyzing the three components of surface and borehole accelerograms recorded during the 14 September 1995 earthquake ( M S 7.3). The target period range is between 0.1 and 10.0 sec, with emphasis on a 2.0- to 3.0-sec period range. In the lake-bed zone inside the valley, wave types of seismic motions are identified by calculating theoretical amplification of various seismic waves and cross-correlation functions between surface and borehole recordings. In the MVB, propagation velocities of dominant surface waves are estimated by analyses of recordings and are compared with theoretical velocities, and 2D/3D wave-propagation simulations for the MVB and the Mexico City Basin are performed. Our conclusions are the following. (1) We suggest that, while strong motions observed in the lake-bed zone are a mixture of P, S , Love, and Rayleigh waves, fundamental-mode Love waves are dominant. (2) Surface waves are found to be much more heavily amplified than S waves in the soft lake-bed deposit. (3) We confirm that the MVB is an important structure to amplify seismic motions. The lake-bed deposit, together with the basin structure, proves to remarkably amplify surface waves. (4) In the lake-bed deposit, quality factors are found to be about 3 in the long-period range of more than 1.5 sec. Manuscript received 24 March 2003.

Wave Field Estimated by Borehole Recordings in the Reclaimed Zone of Tokyo Bay

We estimate the wave field in the reclaimed zone of Tokyo Bay by analyzing the three components of surface and downhole strong-motion accelerograms recorded at two borehole stations during five medium earthquakes. The target period range is between 0.1 and 2.0 sec. First, wave types of observed seismic motions are identified by calculating theoretical amplification of body waves and surface waves and cross-correlation functions between the surface and downhole recordings. Then, the horizontal components of the surface and downhole recordings are separated into S -wave and surface-wave accelerograms. Finally, 2D wave-propagation simulations for basin structures are performed to interpret the recordings. The conclusions are as follows. (1) Main seismic motions are composed of not only S waves, but also Love waves. Surface waves are found to be more dominant than S waves in the period range of more than 1.0 sec. (2) Love waves are more heavily amplified than S waves at the theoretical predominant periods of ground in the soft surficial deposits. (3) The large surface-wave amplification is excited by the surficial deposits, together with a deep basin. (4) In the surficial deposits, quality factors are found to be about 10 around the theoretical predominant periods of ground. (5) The wave-propagation simulations suggest strong heterogeneities near the ground surface and can explain observed seismic motions qualitatively.

Three‐dimensional non‐linear soil–building interaction analysis in the lakebed zone of Mexico city during the hypothetical Guerrero earthquake

The 1985 Michoacan earthquake (M = 8.1) caused very severe damage to mid-rise buildings in the lakebed zone of Mexico City, which is approximately 400 km from the epicentre in the Pacific Ocean. In the present study, we perform a three-dimensional (3-D) non-linear soil-building interaction analysis for several types of low- to high-rise buildings during the hypothetical Guerrero earthquake, and try to understand the real cause of heavy damage to mid-rise buildings in the lakebed zone during the 1985 Michoacan earthquake. We make a reasonable estimation of the input earthquake motions and the local site effects. The non-linear soil-building interaction analysis explains the damage pattern observed during the 1985 earthquake, although other analyses do not. We realize that all the factors from the earthquake source to the building superstructure must be taken into account adequately.

Three-Dimensional Finite-Element Method for Soil-Building Interaction Based on an Input Wave Field

AbstractA three-dimensional (3D) FEM for examining the soil-building interaction based on an input seismic wave field is proposed. A seismic wave field means seismic waves propagating in a 3D medium. An input seismic wave field is employed with the goal of adequately treating seismic surface waves trapped by a deep (several kilometers) underground structure in a soil-building interaction system. As the first stage of the proposed method, a simple linear method is constructed. The linear method was applied to estimate seismic responses of low- to high-rise RC model buildings during a large earthquake at a soft-soil site in Mexico City where surface waves are dominant. At the soft-soil site, all the buildings with and without piles vibrated together with the ground, probably suppressing the pile damage. The proposed method qualitatively provided us with more realistic building responses, compared with a conventional interaction analysis based on an input base motion. When a considerable amount of surface wa...

A Systematic Method for Analyzing Borehole Recordings to Estimate the Wavefield in the Lakebed Zone of Mexico City

During the 1985 Michoacan earthquake ( M S = 8.1), large-amplitude seismograms with extremely long duration (hundreds of seconds) were recorded in the lakebed zone of Mexico City. In spite of numerous studies, the lakebed seismograms are not well understood. In the present study, we develop a systematic method for analyzing borehole recordings and apply it to recordings at depths of 0 and 102 m at the Roma station to interpret the wavefield in the lakebed zone. The target frequency (period) range is between 0.2 and 3.3 Hz (0.3 and 5.0 sec). We cross-correlate the surface and borehole strong-motion seismograms and identify body- and surface-wave portions in the main motions. Then, we interpret the surface waves as fundamental-mode Love waves by calculating theoretical amplitude ratios for different-scale structures. Using a statistical time-series model with an information criterion, we separate S - and Love-wave portions from the surface recordings. Love waves are dominant, and are recognized even in the early time section. Finally, we attempt to explain S -wave transfer functions between the surface and borehole recordings by considering subsidence of soft surficial deposits for several years on the basis of a three-dimensional finite-element technique. This simulation demonstrates effects of the subsidence on amplification of S waves.

Effects of Seawater of Tokyo Bay on Short-Period Strong Ground Motion

The effects of the seawater of Tokyo Bay on strong ground motions are investigated in a short-period range of less than about 2.0 sec for engineering importance. A recent study based on observed accelerograms revealed that considerable short-period surface waves were included in strong ground motions in the reclaimed zone of Tokyo, and that those surface waves were mainly Love waves. However, it was not clarified whether short-period Rayleigh waves were present. The present study is aimed at assessing the excitation of short-period Rayleigh waves on land and at exploring the effects of the shallow (approximately 25 m) water layer of Tokyo Bay on Rayleigh waves in the reclaimed zone of Tokyo. We use a theoretical approach in which a 2D P-SV -wave field is calculated for a multilayered structural model. The model is composed of a sea zone and a land zone, and includes a water layer and soft-surface layers. To examine the effects of seawater, we use models without and with a water layer. As a result, it is concluded that in the reclaimed zone of Tokyo, stripping the water layer off of a typical model for shallow Tokyo Bay has little influence on the short-period ground motions around the theoretical predominant period of the ground of about 1.0 sec. It is not essential to incorporate the water layer in modeling shallow Tokyo Bay for short-period, ground-motion simulations.

Seismic Responses of Two RC Buildings and One Wood Building Based on an Input Wave Field

AbstractA recently proposed three-dimensional (3D) linear method for examining soil-building interactions based on an input seismic wave field is, after some improvements, applied to estimate seismic building responses in the reclaimed zone of Tokyo Bay, where ground motions include a considerable amount of surface waves, thus reconfirming the effects of the method in a different situation. A seismic wave field involves seismic waves propagating in a 3D medium. The proposed method was developed with the goal of adequately treating seismic surface waves trapped by a (several-kilometers) deep underground structure in a soil-building interaction system. Two simulations were carried out. The first simulation successfully reproduced surface, downhole, and building accelerograms that were recorded at one borehole station during two medium-sized earthquakes. In the second simulation, seismic responses of a midrise RC model building and a wood model building were favorably calculated at the other borehole station...

Estimation of the Wave Fields in the Three Geotechnical Zones of Tokyo

We estimate the wave fields in the three geotechnical zones (the hill, alluvial, and reclaimed zones) of Tokyo by analyzing the three components of surface and downhole strong-motion accelerograms recorded at six borehole stations. Although the wave field of the reclaimed zone was already estimated in a recent study, it is partly re-estimated by an improved technique. The target-period range is mainly between 0.1 and 2.0 sec. The wave types of observed ground motions are identified by calculating the theoretical vertical amplitude ratios of body waves and surface waves and the cross-correlation functions between the surface and downhole recordings. In the calculation of the theoretical amplitude ratios, the damping for wave propagation and the damping for the eigenfunctions of surface waves are separately considered. Further, the horizontal components of the surface and downhole recordings are separated into S -wave and surface-wave accelerograms. The main conclusions are summarized as follows: (1) the ground motions observed in the alluvial and reclaimed zones are mostly composed of S waves and Love waves; (2) the Love- wave amplitude ratios are larger than the S -wave ratios in a soft surficial deposit around the theoretical predominant period. The large Love-wave amplitude ratios are excited by the surficial deposit and a deep sedimentary basin; and (3) quality factors near 10 are obtained for the soft deposit around the predominant period.

Excitation of Surface Waves in the Valley of Mexico

By analyzing the three components of surface and downhole accelerograms obtained at four borehole stations located in three geotechnical zones of the Valley of Mexico, differences in the nature of seismic motions among the different zones are investigated. The target period range is mainly a narrow band including the theoretical primary predominant period (the fundamental resonant period) at each station. The wave types of seismic motions are identified by calculating the theoretical vertical amplitude ratios of various seismic waves and the cross-correlation functions between the surface and downhole recordings. In the calculation of the theoretical amplitude ratios, the damping for body-wave propagation and the damping for the eigenfunctions of surface waves are considered separately. For the two kinds of damping, period-independent and period-dependent quality factors are employed. We found that the seismic motions were dominated by surface waves rather than S waves around the predominant period at the four stations. Based on the theoretical amplitude ratios, we could explain the large vertical amplitude ratios observed at the predominant period at all stations. Our conclusions are summarized as follows: (1) The resonant seismic motions are mainly composed of Love waves and S waves. (2) The Love-wave theoretical amplitude ratios are considerably larger than the S - wave ones at the predominant period at two soft-soil stations. (3) The low values of period-dependent quality factors for the eigenfunctions of surface waves explain the large peak amplitude ratios observed at the predominant period at two soft-soil stations.