After the 1995 Hyogo-ken-Nanbu earthquake, new national projects have stated to improve seismic observation systems in Japan. A large number of strong-motion seismographs were installed to construct dense and uniform networks covering the whole of Japan. The K-NET is one of the nationwide strong-motion observation networks and consists of about 1, 000 strong-motion seismographs installed all over Japan. The strong-motion records obtained by the K-NET are available on the Internet. The operation of the K-NET began in June 1996. Ten years have passed since the K-NET started and many strong-motion records are obtained. During these years the progress in information technologies is remarkable and new technologies become to be available. The NIED has developed the new seismographs for the K-NET using the new technologies. The main features of the new strong-motion seismograph are functions for real time data communications and processing JMA seismic intensity. The new seismographs have been installed at 910 observatories. The operation of the new K-NET system started in June 2004.

View full abstract

Nowadays seismic intensity scale has become important information for real time counter-measures soon after disastrous earthquakes. However, some investigations of damage due to recent earthquakes such as the 2003 Miyagiken Hokubu Earthquake revealed problems over how and where to set an instrument for measuring seismic intensity. The difference in seismic intensity between at the ground surface and in buildings has especially drawn attention form engineers. This paper deals with the variations of seismic intensity using records due to an array observation system for ground and building motions. The array observation system, which was installed at the campus of Tohoku Inst. of Tech. 15 years ago, has obtained some 90 motion records simultaneously at different sites of the free ground surface, the first floor and fourth floor of a 4-story building. The instrumental intensity scales were obtained from the records to compare their values at the different sites. The results made it statistically clear that the intensity at the first floor is smaller by about 0.2 than the one at the ground surface. This difference was also found in array observation of microtremors, leading to a conclusive remark so that microtremors are useful to predict a difference of seismic intensity depending on installation sites.

View full abstract

Earthquake response observation and vibration tests were examined for evaluating dynamic characteristics of soil-structure system of existing buildings. Approximately 150 buildings have been tested and observed in the past 15 years. Based on the results, new strategic response observation program is proposed, which reveals appropriate combination of limited observation in systematic way. Detailed discussion is made for some important points concerning planning of observation, new equipments and operation, data analysis and database, and future aspects of earthquake response observation.

View full abstract

In order to investigate behaviors of buildings during earthquake due to accumulate observation data, low-cost seismometers are produced and checked. Using data by shaking table test results and observed earthquake motions, the accuracy verification of low-cost seismometer is conducted through comparison with servo-type seismometer. The low-cost seismometers were installed in a detached house and earthquake motions have been observed since about three years ago. In these periods, many earthquake motions have been obtained to discuss characteristics of the earthquake motions at the site and characteristics of dynamic behavior of the detached house. The low-cost seismometer has been operational for about four years since the low-cost seismometer was produced. From the result, it is confirmed that the low-cost seismometer has enough durability.

View full abstract

Measured seismic intensity data have been observed densely during many recent earthquakes in Japan. We have constructed seismic intensity database and analyzed the characteristics of attenuation and local site correction regarding short-period ground motions in various regions of Japan. It is found that the local site correction for seismic intensity is dependent on earthquake magnitude in the case of the Kanto plain. The regression analysis indicates that there is positive correlation at lowlands such as the east of Saitama and negative correlation at hilly areas such as the south of Chiba between the intensity coefficient for local site correction and magnitude. Using the obtained attenuation formula of seismic intensity and the intensity coefficient for local site correction, the intensity inversion analyses were carried out and the short-period seismic wave radiation zones (SPRZ) were identified for the Great Kanto earthquake in 1923 and the Genroku earthquake in 1703. The result shows that there were two separate SPRZs in east-and-west both sides across the Miura peninsula during the Great Kanto earthquake. The both of energy centroids of the SPRZ correspond to the terminus part of large slip areas derived from the existing waveform inversion study. The SPRZs of the Genroku earthquake are similar to those of the Great Kanto earthquake as for the areas from Kanagawa to the south of Chiba. In addition to those, the SPRZ is deduced to spread southeastward off the Boso peninsula.

View full abstract

First we utilized Stochastic Green's functions obtained from strong motion data of K-NET, KiK-net, and the JMA Shindo-kei network in Japan to predict strong motions for a large subduction-zone earthquake. Then considering earthquake type and site characteristics, earthquake spectrum characteristics and propagation path effects were extracted. Then we used a conventional Green's function method to sum up statistical Green's functions for a moderate size earthquake and synthesize them to predict strong motions due to the expected Nankai earthquake. The resultant strong motions show similar PGA values of empirical relations and the calculated seismic intensities show similar values as observed in the previous two events.
Finally, we input these strong motions to dynamic nonlinear structural models. Our nonlinear structural models are unique because they are multiple models with different strengths with different existence ratios. From the theoretical calculations we can determine damage ratios for different types of building. For quantitative prediction we determined model parameters based on the damage statistics in Kobe after the Hyogo-ken Nanbu (Kobe) earthquake of 1995. We found that heavy damage to structures can only be found in the near-source region. We have very heavy damage ratios to low-rise steel structures, while for low-rise RC structures we have relatively small damage ratios. However, the damages in high-rise buildings were limited to those at near fault locations, and high-rise buildings in Osaka Plain did not have serious damages or plastic deformations.

View full abstract

The site amplification factors for peak ground acceleration (PGA) and JMA seismic intensity (I_JMA) at each observation point was evaluated from earthquake motion records of seismograph network in Ishikawa Prefecture, K-NET and KiK-net. Compared with the site amplification for PGA and I_JMA, these show positive correlation. And, the site amplification forI_JMA (F^I) shows large value in sites where the site amplification factor for PGAF^Ais small but predominant period of ground (T_g) is long. Next, assuming thatF^Iis expressed as function in terms ofF^AandT_g, the influence ofF^AandT_gonF^Iis is examined by regression analysis. As a result of analysis, it shows thatF^Ahighly correlates withF^Imore thanT_g.

View full abstract

In this study, amplification factors of seismic intensity and peak ground acceleration (PGA) are studied. These indices are obtained by K-NET, KiK-net and seismic intensity information network stations in Aomori prefecture during seven earthquakes which occurred east off Aomori prefecture. At first, regression analysis using underground data of KiK-net is performed. Then amplification factors are estimated by this empirical relation. Range of amplification factor of seismic intensity is from 0.0 to 2.5. Meanwhile, amplification factor of PGA is ranged from one to 20. As coefficient of variation of these indices is around 0.2, we think that this value is not large to estimate a site amplification. We also show validity of the amplification factor using ground motion data from the 2003 Tokachi-Oki earthquake and the 1994 Sanriku Haruka Oki earthquake.

View full abstract

Results evaluated by dynamic analysis method will be changed according to the analytical models for coupled dam-foundation-reservoir system, the values of dynamic deformation properties of dam and foundation, the dynamic analysis program used, and so forth. Therefore, in order to realize a reliable evaluation for earthquake safety of existing dams, it is necessary to verify a validity of dynamic analysis procedure. If the actual earthquake motions are obtained by the seismological observations at existing dams, it becomes possible to make the reproduction analyses for actual earthquake behaviors of existing dams, and to prove an efficiency of dynamic analysis procedure based on the real earthquake phenomena. In order to establish an accurate and reliable evaluation for seismic safety of dams, I have developed a 3-D dynamic analysis method for coupled dam-joint-foundation-reservoir system, thus far. And I have been making the 3-D reproduction analyses for the actual earthquake behaviors of several existing dams. In this paper, the 3-D reproduction analyses in regard to the Tagokura Dam during the 2004 Niigataken-chuetsu Earthquake and the Ikehara Dam during the 1995 Hyogoken-nanbu Earthquake were described. And an efficiency and a meaning of the 3-D reproduction analyses was considered from the viewpoint of an effective utilization of earthquake observation data.

View full abstract

The nonlinear soil response is examined by using KiK-net vertical array records from a large (the 2003 Tokachi-oki earthquake; M8) and small (M-6) events. First we search stations which show nonlinear soil response by comparing the S-wave spectral ratio of the Tokachi-oki earthquake with those of three small events; the S-wave spectral ratio is taken from the surface spectrum divided by the borehole one. We find three stations from 106 KiK-net stations in Hokkaido that show the nonlinear soil response. These stations' spectral ratios show amplitude decay at high frequencies and peak shift to lower frequency for the spectral ratio from the Tokachi-oki earthquake. The TKCH07 station at Toyokoro in Tokachi-district, Hokkaido, is chosen to evaluate the nonlinear soil response. This site has very soft ground different from another KiK-net stations; S-wave velocity (Vs) is less than 200m/s down to a depth of 30m. This station observed peak ground acceleration (PGA) of 403.9cm/sec² during the Tokachi-oki earthquake. Next we construct the soil model (Vs and Qs (quality factor for S-wave) structures) at TKCH07 based on the S-wave spectral ratios from 22 small earthquakes by using the genetic algorithms (GA) inversion method. Finally we evaluate the nonlinear soil response at TKCH07 by comparing the surface accelerogram observed during the Tokachi-oki earthquake with those calculated using the borehole accelerogram and the constructed soil model; the linear response analysis, equivalent linear response analyses and nonlinear response analysis are used in the calculations. The frequency dependent equivalent linear response analysis is newly proposed; this method well explains the observed surface accelerogram.

View full abstract

Mudflow type failure with gentle slope occurred at Dateshita in Tsukidate town during the 2003 Sanriku-Minami earthquake in Japan. The fluid soil was consisted pyroclastic sediment material which classified as a volcanic sandy soil with pumice. This failure phenomenon was due to loose volcanic sandy soil material with fluidization type flow by the trigger of strong ground motion of the earthquake, against of its material unsaturated condition. Unfortunately, when the main shock occurred, strong ground motion record was failed to observe at Tsukidate K-NET station, the closest observatory to Dateshita landside site. Therefore, we conducted to estimate strong ground motion seismograms at mudflow failure site, because of few study regarding the relationship between ground motion records and mudflow type failure consisted of unsaturated volcanic sandy soils. We estimated strong ground motion waveforms using empirical Green's function method using two after shocks comparing several seismic intensities surrounding the Tsukidate town. Therefore, we recognized both dissipation energy, one is computed by 1-D geotechnical response analysis and the other one is obtained by conducting laboratory testing, would be good relationship, then, we proposed new method for estimating liquefaction potential of unsaturated pyroclastic material. We finally realized mudflow type failure with gentle slope at Tukidate town is due to fluidization and liquefaction type soil collapse by the trigger of strong ground motion.

View full abstract

Strong ground motion observation networks were established after the 1995 Hyogoken-Nanbu Earthquake, and many earthquakes occurred. It makes it possible for us to accumulate building damage data around the strong ground motion observation sites to investigate the relationship between characteristics of strong ground motions and damage to buildings. I investigated the destructive power of strong ground motions and the relationship between characteristics of strong ground motions and damage to buildings by using accumulated building damage data around the strong ground motion observation sites and strong ground motion records after the establishment of the strong ground motion observation sites. Additionally, I examined the correspondence of strong ground motions and actual damage to buildings by performing elasto-plastic earthquake response analyses.

View full abstract

On March 20, 2005 the West off Fukuoka earthquake of MJ7.0 occurred at about 30km northwest of Fukuoka City. Although 6- on the JMA seismic intensity scale were reported at several points in Fukuoka, structural damage was not so devastating as expected from this intensity. Still, we have clear concentration of the damaged buildings in the central part of the city. Thus, we analyze the observed strong motion record in Fukuoka. First basic characteristics were summarized, and then the structural damage potential of these strong motions (=destructive ability) was evaluated using the damage ratio prediction model, in which we use a nonlinear response analysis method for a set of structures with different strengths (so-called Nagato-Kawase model). Then, the damage ratio distribution was calculated using the strong motion waveforms reproduced from the 1-dimensional ground models reflecting the shallow soil layers down to the engineering bedrock at every small districts. It turned out that the damage ratios also becomes large, as observed, in the areas where the bedrock depths are deep, namely in the northeastern side of the Kego fault. However, the simulated damage ratios are larger than the observed damage ratios so that we have a space for improvement in our simulation model of damage.

View full abstract

The National Institute for Land and Infrastructure Management (NILIM) administers the observation of strong earthquake motion at civil engineering structures throughout Japan. This paper describes the observed strong earthquake motion at the Kaihoku Bridge station during the off-Miyagi earthquake (M7.2) on August 16, 2005 in comparison with that of the 1978 off-Miyagi earthquake (M7.4).

View full abstract

This report is a summary of an earthquake measurement result at July 23, 2005, and disaster activity that used the record of this result. We understood little damage, based on forecasting pipeline-damage. Because we collected an earthquake measurement result from 3, 800 seismograph arranged in metropolitan area. Moreover we forecasted the number of telephone-call about the Intelligent gas-meter, then we set up the system for prevention of secondary disaster and assembled the necessary personnel.

View full abstract

Site amplification factors were evaluated for strong-motion sites in Japan by using the spectral inversion technique. The seimic moment calculated from the source spectra, which were obtained through the analysis, agrees well with that of the CMT solution of the F-net, indicating the validity of the newly-introduced criteria to determine the reference site. Because the site amplification factors obtained in this study include the effects of later phases, they are suitable to be used for simulating strong ground motions by Kowada's method. Strong ground motions from the 1995 Hyogo-ken Nanbu, the 1968 Tokachi-oki and the 2003 Tokachi-oki eathquakes were simutated by the method to confirm the usefulness of the site amplification factors obtained in this study.

View full abstract

Various earthquake motion indices, maximum acceleration PGA, maximum velocity PGV, instrumental seismic intensity Ijma, SI value and RI value (maximum of real-time intensity RI), are compared each other. The PGA limited less than 5Hz (5HzPGA) is not differ form SI value that seems to be related to actual earthquake damage, form the relationship to the Ijma. Both Ijma and SI value are almost equivalent to the RI value that defined based on the power of the earthquake motion. It suggests that these indices mutually related to the actual damage. Because Ijma defined by JMA can calculate only after the earthquake and SI value must be calculated from responses for various periods, it is difficult to calculate without delay so both indices are unsuitable for the earthquake alarm. Because not only RI fits to the Ijma at the peak value, but also RI responses to the P wave arrival sensitively, RI can realize the quick and exact alarm. It is expected to beutilized at the various field because RI is assured to related physically to actual damage.

View full abstract

The strong-motion database is one of the highest priorities to be established in the field of earthquake engineering and engineering seismology. This report describes the design concepts and the data installation for the prototype of strong-motion database that will be used for the shaking table, that is the 3-D Full Scale Earthquake Testing Facility (E-Defense), established by the National Institute for Earth Science and Disaster Prevention. We focused on the scheme of the database which is easy to retrieve the data on demands, such as magnitude range, PGA, PGV, PGD, seismic intensity, distance to a source and so on.

View full abstract