巻号一覧
51 巻 (1998 - 1999)
- 4 号 p. 355-
- 3 号 p. 253-
- 2 号 p. 161-
- 1 号 p. 1-
前身誌
51 巻, 1 号
選択された号の論文の14件中1～14を表示しています
- |<
- <
- 1
- >
- >|
-
都司 嘉宣, 上田 和枝, 佐竹 健治1998 年 51 巻 1 号 p. 1-17
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーFrom midnight of January 27, 1700 to noon of the next day, abnormal sea level changes were recorded at several locations on the Pacific coast of Japan. We previously interpreted these as tsunamis from a giant earthquake in the Cascadia subduction zone, off the west coast of North America. In this paper, we report our detailed examinations of the original documents recording the tsunamis, including the reliability of the documents themselves. The tsunamis were described in six independent documents, mostly local government records. No one was injured or killed at any location. In Miyako, 20 houses were burned and 13 houses were destroyed by the tsunami. At Otsuchi, a few houses and rice paddies were damaged. In Tanabe, the government storage house was inundated. The tsunamis were first noticed just before midnight of the 27th at Miyako and Otsuchi, and the next morning at the other localities. We estimated the tsunami heights from descriptions in the documents, after accounting for tide levels and vertical crustal movements of coasts since 1700. The latter correction is difficult and the uncertainty is large. The estimated tsunami heights were: 4m at Kuwagasaki, 3.2m at Tsugaruishi, both in Miyako, 3.3m at Otsuchi, 1m at Nakaminato, 1.0-1.7m at Miho, and 3.3 and 5.4m at two locations in Tanabe. We also examined the weather records for January 26-29 and found no abnormal description, ruling out the possibility of meteorological origin. Lack of regional earthquakes on these days and the tsunami height distribution indicate that the tsunami came across the Pacific Ocean. We then estimate the tsunami magnitude, Mt, for each possible tsunami source around the Pacific. The result shows that the size of earthquake that generated the tsunami was M9 wherever the source was. We examined historical data and paleoseismological results in the possible regions and confirmed that the tsunami source was the Cascadia subduction zone, and that the origin time was around 9PM on January 26, 1700 local time.抄録全体を表示PDF形式でダウンロード (4409K) -
青池 邦夫, 佐藤 魂夫, 西澤 修1998 年 51 巻 1 号 p. 19-30
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーWe have applied the Extended Information Criterion (EIC) for determining an optimum model in P-velocity tomography beneath the northern Tohoku region. Numerical simulation based on bootstrap statistics was used to calculate the EIC that measures the goodness-of-fit between true and predictive data distributions. Minimization of EIC provides an objective criterion for both the number of iteration and the damping factors that have so far been selected empirically. We can avoid the reconstruction of unreasonable model parameters resulting from an excessive reduction of travel time residuals by terminating the iterative procedure at an optimum step. The optimum damping factors determined a velocity contrast faithful to information of the data. For the present dataset, the optimum P-velocity model was obtained after the second iteration with an a priori uncertainty of 2% for fractional velocity perturbations and a given standard error of 0.1s for arrival time data. We could estimate EIC values with small errors even for fifty bootstrap samples. Thus the EIC was proved to be viable in terms of computation time, too. We also performed the error analysis of the model parameters using bootstrap statistics. Based on the error analysis, we evaluated the reliability of inverted model parameters and picked out the false images that result from poor density and ill configuration of seismic rays.抄録全体を表示PDF形式でダウンロード (9898K) -
香川 敬生, 澤田 純男, 岩崎 好規, 南荘 淳1998 年 51 巻 1 号 p. 31-40
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーFour Layered structure model with S-wave velocities of the Osaka Sedimentary Basin is presented in this paper to calculate long-period seismic response. We conducted microtremor array observation and obtained dispersion curves of Rayleigh waves at 16 sites in the basin. From the dispersion characteristics, we derived layered structure down to bedrock at each site. Compiling the results, we modeled a subsurface structure for the Osaka Sedimentary Basin. We found that the layered structure has common characteristics at all the sites in the Osaka Sedimentary Basin, and concluded that the basin structure can be modeled with four layers. Each layer has S-wave velocity of 0.35, 0.55, 1.0, and 3.2km/s, respectively.抄録全体を表示PDF形式でダウンロード (1154K) -
清水 直哉, 塚原 弘昭, 古舘 友通, 吉田 則夫1998 年 51 巻 1 号 p. 41-50
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーLarge amount of underground water with CO2 gas bubbles had flowed out from many spots for the period of the Matsushiro Earthquake Swarm, 1965-1967. Some of the springs are still alive. We explored possibility for detecting crustal activity, such as crustal deformation and seismicity by observing CO2 concentration in soil gas in the Matsushiro area. We measured CO2 concentration in soil gas once a week for 6 months at many places in two areas: (1) a fault zone of the Matsushiro Earthquake Swarm, and (2) floor of the tunnel of Matsushiro Seismological Observatory which is about 2.5km away from the fault zone. The measurement points are located at various geological conditions. Remarkable high concentration of CO2 in soil gas was detected at measurement points in the fault zone. In particular, high concentration of CO2 was observed at a point on a fault. In contrast, on tunnel floor, out of the fault zone, the concentration is relatively low. Temporal variation, records of CO2 concentration of all measurement points, however, resemble one another in variation pattern. This suggests that the CO2 discharge is controlled by some wide regional scale mechanism. The pattern of the concentration variation is compared with those of seismic data, crustal strain data and meteorological data observed at the observatory. Good correlation is found between temporal variation in CO2 concentration and that in crustal strain observed with north-south orientation strainmeter, that is, CO2 concentration increase with expansion of the NS70-100 component of strainmeter. We propose a model of the variation in CO2 discharge in relation to crustal strain and fracture distribution. The strainmeter data shows that the crustal strain is largely influenced by rainfall at the Matsushiro area, for example, NS70-100 component of strainmeter expands after rainfall. Since many fractures having nearly the east-west direction are observed on the surface in this area, expansion in the north-south direction is expected after rainfall. Some pre-existing cracks will open with the expansion of neighboring cracks caused by water absorption. Since large amount of CO2 gas presents under this area, CO2 discharge rate increase with increase of width of passages. We can detect strain change of the fracture zone by CO2 monitoring.抄録全体を表示PDF形式でダウンロード (1393K) -
岩田 貴樹, 中西 一郎1998 年 51 巻 1 号 p. 51-59
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーWe investigate the correlation between the occurrence of earthquake and the earth tide by analyzing the tail of the Matsushiro earthquake swarm that is recorded by the Matsushiro Seismic Array System during the period of November 1984 to December 1994. The seismic activity of the tail exhibits many bursts of earthquakes. We remove the bursts, because they affect our aim to investigate whether the correlation exists or not. We use the tidal strain data recorded by EW- and NS-component extensometers located in Matsushiro Seismological Observatory, Japan Meteorological Agency. We apply Schuster's test to the Matsushiro earthquake swarm. The hypothesis that the earthquakes in the swarm take place randomly is tested and is rejected for the EW-component. The earthquakes tend to occur when the tidal strain of the EW-component is in a compressional state. This state is consistent with the stress field at Matsushiro obtained by using focal mechanisms or in situ stress measurements.
We also investigate the spatial variation of the correlation. A significant correlation is seen in the northwestern part of Matsushiro. The largest earthquake with magnitude of 5.2 occurred on 30 December 1986 near the region. We further investigate the temporal variation of the correlation in this region. The correlation becomes high right after the occurrence of the largest earthquake, whereas the correlation is low before the occurrence of it. This suggests that the strain changes caused by the largest earthquake make the correlation high in this region.抄録全体を表示PDF形式でダウンロード (1248K) -
千釜 章, 多田 省一郎, 青沼 正光1998 年 51 巻 1 号 p. 61-73
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーSome specific legends and traces related with historical tsunamis have been handed down among the people in the northern part of Honshu, Japan. Examples are the forest of Hiba buried by a large amount of sand and the resettlement of villages damaged by tsunamis in Shimokita Peninsula, the northeast Honshu. The aim of this paper is to reexamine their reliability by compiling old documents and results from numerical simulation of tsunamis. Results from old documents combined with numerical simulation suggest that the legends at the age of Tensho and Kan-ei period would be associated with not tsunamis but floods. And the story of damage at Tanabu town by the 1611 Keicho tsunami causing resettlement to a higher place is not reliable because the simulated tsunami could not reach the locations of the village. The sequence of events of sand transport to bury Hiba forest at four different ages could be found by filed investigation and result of dating of 14C. The ages do not correspond to the tsunami events. Such process can be explained by a formation of sand dune, meaning that a large amount of sand movement should not be caused by tsunamis.抄録全体を表示PDF形式でダウンロード (1636K) -
羽鳥 徳太郎1998 年 51 巻 1 号 p. 75-82
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーA large tsunamigenic earthquake occurred off Adak Island at 4h 03m (UT), June 10, 1996. According to USGS, the epicenter is 51.564°N, 177.632°W with earthquake magnitude MS=7.6. Four hour later, small tsunami (double amplitude: 10-25cm, wave period: about 20min) was observed at the Japanese tide stations. By use of the diagram of the attenuation of wave-height with distance, tsunami magnitude on the Imamura-Iida scale is estimated to be m=2.0, which the value is similar to that of the 1986 Aleutian tsunami. The observed amplitudes in Japan for the past six Aleutian tsunamis were about 1/4 times smaller than those at the Hawaiian Islands. The directivity of tsunamis toward Japan is small, but in case of the source areas locating the East longitude region becomes effective. The average velocities of the maximum wave arriving in the Japanese Islands are scattered in the range of 240km/h to 470km/h. The velocity of large tsunamis (1946, 1957 and 1965) is rather slow, suggesting the propagation of edge waves along the continental slope.抄録全体を表示PDF形式でダウンロード (654K) -
吉岡 敏和, 苅谷 愛彦, 七山 太, 岡田 篤正, 竹村 恵二1998 年 51 巻 1 号 p. 83-97
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーThe Hanaore fault is a right-lateral strike-slip active fault about 48km long in central Japan. We carried out comprehensive surveys including trench excavations on the Hanaore fault to evaluate the seismic risk of the highly populated area, such as Kyoto City, along this fault. Three trenches were excavated on the fault. On the exposure of the northernmost Tochudani trench, a fault cutting fluvial sediments and humic soil beds appeared. The youngest age of displaced sediments is 460±60 14C yBP, and the sediments covering the fault is 360±60 14C yBP. This faulting event may be correlated to the historical 1662 Kambun earthquake. The southernmost Imadegawa trench was excavated on the road in the urban area of Kyoto City. A thrust fault cutting humic soil with pottery fragments of the Late Jomon period (about 3, 500 years ago) was observed on the trench walls. It was difficult to detect the age of the last faulting event due to lack of younger sediments and artificial modifications of the surficial materials. However, the southern part of the fault might not move during the 1662 earthquake because the damage in this area was much less than in along the northern and middle part of the fault. The historical documents recorded that the land along the Mikata fault which is located at the north of the Hanaore fault was uplifted, and the land along the western shoreline of Lake Biwa where is the east of the Hanaore fault was subsided during the 1662 earthquake. This means that the 1662 earthquake might be a multi-segment event caused by these three faults, the Mikata fault, the northern part of the Hanaore fault, and the faults along the western shoreline of Lake Biwa.抄録全体を表示PDF形式でダウンロード (2142K) -
片岡 俊一, 川瀬 博1998 年 51 巻 1 号 p. 99-112
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーSeismic ground motion at a site is strongly affected by the underground structure, especially the S-wave velocity structure below. During the 1995 Hyogo-ken Nanbu earthquake, it is found that the basin structure with the thickness of about 1km played a major role to form the damage belt in Kobe. We estimate the S-wave velocity structure at Higashinada ward, Kobe where the damage belt runs through from west to east. We deployed two different sized arrays in southeastern part of Higashinada ward. The smaller array, named KUMM array, has a diameter of 200 meters, while the larger array, named M-array, has a diameter of 2000 meters. We recorded microtremors at each array with ten three-components velocity sensors simultaneously. Independently, a seismic refraction experiment was carried out on December 12 and 14, 1995 in Hyogo and Osaka prefectures. We also recorded seismic waves from the explosion in the Osaka port by using the same KUMM array as the microtremor observation. Phase velocities of Rayleigh waves not only for microtremors but also for explosion-induced seismic waves are used to estimate the S-wave structure. Estimated S-wave velocity structure is very close to the PS logging data which is obtained at a site close to the array. Seismic refraction survey using explosions is often carried out to estimate an underground structure, in which only initial motions are used and later arrivals are considered to be noise. In this paper, we show a possibility to use phase velocities of Rayleigh wave, which is induced by an explosion, just as the case of microtremors.抄録全体を表示PDF形式でダウンロード (1553K) -
宮川 康平, 中西 一郎, 三浦 勝美, 田中 聡1998 年 51 巻 1 号 p. 113-121
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーThe Geiyo earthquake occurred on June 2 in 1905 in the western Seto Inland Sea between the Honshu and Shikoku Islands, Japan. The seismograms of the earthquake obtained at the stations of Central Meteorological Observatory were newly found at Earthquake Research Institute of the University of Tokyo. They are recorded by the Omori seismometers and tromometers, which are superior to former seismometers with respect to continuous recording. For the estimation of the magnitude and source mechanism from the seismograms, we digitize 3 records, which are Hongo EW component, Hitotsubashi EW component and Tokyo tromometer. We have to know the response of the Omori seismometer to estimate the ground motion during the earthquake. In order to know the frequency characteristics of the seismometers, we calculate their Fourier amplitude spectra. The spectra of the Hongo EW component and Hitotsubashi EW component show clear peaks which may be considered as the natural periods of the seismometers. The natural periods of Hongo EW component and Hitotsubashi EW component are about 60s and 25s, respectively. The damping constant estimated from the free oscillation record of Omori seismometer at Ishinomaki observatory is less than 0.01, and the friction is 1.7mm.抄録全体を表示PDF形式でダウンロード (4574K) -
加藤 研一, 武村 雅之, 八代 和彦1998 年 51 巻 1 号 p. 123-138
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーWe examine the excitation strengths of source spectra in high-frequency range, and discuss their regional variation. Target events are subduction earthquakes off Fukushima Prefecture, intermediate depth earthquakes in the Kanto region, and shallow earthquakes in the southern Kanto region in Japan. The analyzed data are S-wave portions of 672 accelerograms from 40 events observed at 27 stations in the southern Tohoku and the Kanto region. Source spectra are evaluated by simultaneously separating source, propagation path, and local site effects using inversion technique in spectral domain [IWATA and IRIKURA (1986)]. The magnitude of these events ranges from 3.8 to 7.0, and hypocentral distance from 20 to 200km. First of all, the relationship between the seismic moment M0 and the amplitude of source spectra in high-frequency range MHF is examined. MHF is evaluated by taking logarithmic average of acceleration source spectrum from 1 to 10Hz. MHF is nearly proportional to M1/30, which is expected from the scaling of source spectra based on ω-2 model. MHF shows regional variation even if M0 is nearly equal. To quantitatively estimate the amplitude difference in high-frequency range, BRUNE stress drop Δσ [BRUNE (1970, 1971)] is evaluated for each event using M0 and MHF. Average Δσ for shallow earthquakes in the southern Kanto, subduction earthquakes off Fukushima Prefecture, and intermediate depth earthquakes in the Kanto region is about 100 bars, 200 bars, and 370 bars, respectively. Taking into account the relation that MHF is proportional to Δσ2/3, intermediate depth earthquakes excite about 2.4 times larger amplitude than that for shallow earthquakes in high-frequency range. Thus, deeper events generally show large Δσ in comparison with shallower events, suggesting that the focal depth is a key parameter to control the excitation strengths of source spectra in high-frequency range.抄録全体を表示PDF形式でダウンロード (1846K) -
八木 勇治, 菊地 正幸, 吉田 真吾, 山中 佳子1998 年 51 巻 1 号 p. 139-148
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーWe investigate the rupture process of Hyuga-nada earthquake of April 1, 1968 (MJMA 7.5). Applying a multiple-time window inversion scheme to teleseismic body wave data, we obtained a detailed spatio-temporal distribution of moment release. The main source parameters are: the seismic moment=2.5×1020[Nm]; the rupture area=64×48[km2]; the stress drop=3.4[MPa]; the focal depth=15[km]. The rupture consists of three major asperities: the first asperity centring about 10km south and 20km west from the hypocenter and having a maximum slip of 4.0m, the second one centring about 8km north and 5km east from the hypocenter and having a maximum slip of 3.0m, and the third one centring about 50km west from the hypocenter and having a maximum slip of 3.2m. We compared the rupture area with that of a few large events (M>6.5) subsequent to the 1968 event. Then we found that the above three asperities of 1968 event coincide with the low seismicity area in the Hyuga-nada region, and do not overlap with the source area of the subsequent large events (M>6.5). This rupture pattern and the seismicity suggest that an area of slab bending as well as fracture of the slab can behave as barriers during earthquake rupture. These barriers may control the maximum size of earthquake source in this region.抄録全体を表示PDF形式でダウンロード (8562K) -
福山 英一, 石田 瑞穂, Douglas S. DREGER, 川井 啓廉1998 年 51 巻 1 号 p. 149-156
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーA system for automated moment tensor determination by inverting on-line broadband seismic waveforms operated by the FREESIA project has been developed. When an Email notice of earthquake occurrence arrives from the Japan Meteorological Agency (JMA), this system automatically retrieves the dataset for waveform inversion, examines the quality of data, and determines filtering coefficients as well as data windows. This inversion uses long period waves (from 20 seconds to 50 seconds) so that this procedure is insensitive to the onset of wave arrivals. This inversion uses at most 3 stations whose hypocentral distances are between 50km and 400km. The result of each inversion is automatically displayed on the World Wide Web within 5 minutes after the Email arrival. The obtained mechanisms are examined by comparing them with first motion solutions obtained by a microseismic network. If the variance reduction is greater than 50% and magnitude is greater than 4.0, the automatic solutions are found to be reliable.抄録全体を表示PDF形式でダウンロード (926K) -
勝俣 啓1998 年 51 巻 1 号 p. 157-159
発行日: 1998/07/03
公開日: 2010/03/11
ジャーナル フリーPDF形式でダウンロード (401K)
- |<
- <
- 1
- >
- >|