Recently, the seismic activity of the earthquake swarm southeast of Mt. Ontake has been studied by many researchers, especially after the volcanic eruption on October 28, 1979 because of the proximity of both events in time and space. This paper deals with the focal mechanism solutions with the aid of temporary stations installed after the eruption.
Directions of the P-axes are separated into two groups. One is in N70°W-N80°W directions though T-axes are distributed in girdle (Strike-slip, oblique-slip and dip-slip types are equally observed in the earthquake swarm). Directions of N70°W-N80°W are coincident with the pressure direction in the inland of Chubu District, central Honshu and they are supposed to be the pressure directions of the tectonic stresses in this focal region.
The other group is in N40°W direction, which is apparently inconsistent with the tectonic stresses. Earthquakes belonging to this group are strike-slip types and are occurring in the two limited regions. One is in the east of Mt. Ontake and the other is in the vicinity of Kuzoo (central part of this focal region) and is aligned across the earthquake swarm region. Such an activity indicates the existence of an active fault inconsistent with the present tectonic stresses.
It becomes clear from the present study of focal mechanisms that small amplitudes of S-waves observed at Maze (to the west of Mt. Ontake) are not accounted for the attenuation of S-waves but for the focal mechanisms of the earthquake swarm.

抄録全体を表示

Ten reservoir induced earthquakes were investigated. The relation between the dam height and the maximum magnitude of the induced earthquake was found. We show that there are the magnitude limit of earthquake related with the dam height. The Shenwo earthquake, this was only one exception of this relation, wasn't far from the 1975 Haicheng earthquake. It occurred only 43 days before the Haicheng earthquake. We think the Shenwo earthquake occurred in the precursory period of the Haicheng earthquake. So we concluded that it occurred under the special tectonic condition just before the large earthquake.
Fault plane solutions for reservoir induced earthquakes were also compared with those for natural ones. We got the result that the directions of P-axis and T-axis for induced earthquakes were not always same to those for natural ones. So, we concluded that induced earthquakes occurred in the faults which were weakened by the permeated water rather than in the faults which were easily dislocated by the tectonic force. Large aftershocks of the Xinfengjiang earthquake were investigated and we showed the examples of the shock induced by loading and unloading.
The b-values for the aftershocks of induced earthquakes were calculated, but we got only one b-value for natural one which could be compared to induced one and there was no significant difference between them.

抄録全体を表示

The relation between seismic energy and predominant period was investigated by using the recordings of earthquakes, micro-fractures in the mine and acoustic emissions (AE). The micro-fractures in the mine were observed at the depth of about 400m of the Nakatatsu mine. Acoustic emissions were observed at the triaxial test on granite under a confining pressure. Every recording used in each case was recorded at relatively short focal distance where the effect of Q factor might be disregarded. The frequency band of the observational system was designed so widely as to cover that of seismic wave in each case.
Up to the present, the frequency response of AE sensor has been scarcely known. The sensitivity and frequency response of AE sensor (PZT) were deduced theoretically and calibrated experimentally. The relation between seismic energies and predominant periods of acoustic emissions statisfies the scaling law extraporated from that of earthquakes. But the periods of the micro-fractures in the mine are considerably long and the scaling law does not hold true in this case. These results may be explained by the difference of the fracture strength of the medium in which each fracture occurs.

抄録全体を表示

Uplifted Holocene marine terraces of Hatsu-shima Island (0.44km²) in the western part of Sagami Bay on the Pacific coast of central Japan have been investigated with special reference to the late Quaternary seismic crustal movement of the island. Hatsu-shima Island, which is located at the top of the western steep scarp of the northernmost part of the Sagami trough, was uplifted almost uniformly by around 2m at the time of the 1923 Kanto earthquake (M_s=8.2).
The whole surface of the island consists of a flight of marine terraces developed on Pliocene volcanics. These terraces are clearly classified into three groups; I, II, and III. Terrace I, the highest (about 50m asl), is distributed only in the southeastern small part of the island. Terrace II, 25-40m asl, is the widest in the island and tilting, generally, northwestward. Terraces I and II have been correlated to Obaradai Terrace of about 80, 000yrs B. P. in age and to Misaki Terrace of about 60, 000yrs B. P. in age, respectively, by SUGIHARA (1980) based on marker tephras covering the terraces.
Terraces III, rather narrow but well defined especially in the northern and western parts of the island, is distributed along the present shoreline at the foot of a former sea cliff as high as about 20m surrounding Terrace II (or I in the southeastern part). In this paper they are subdivided into three; IIIa, IIIb, and IIIc. Terrace IIIa, the highest and widest among group III, is rather flat with about 10m asl inner margin. Planned excavation at three localities on this terrace has revealed that terrace deposits composed of marine gravel and sand of 1.5-2m thickness rest unconformably on rather weathered bed rock. Fossil shells that were obtained at the base of terrace deposits at Excavation B, which are judged to be in situ and to represent the terrace's age, have been dated at 6, 730±190 ¹⁴C yrs B. P. (GaK-9080). From this ¹⁴C date, as well as from the topographical features, Terrace IIIa can be interpreted as the highest Holocene terrace formed about 6, 000 years ago (so-called Numa Terrace). The height of former shoreline is estimated at around 9m asl. Terraces IIIb and IIIc are narrow emergent shingle beach, whose inner margins are 6-7m asl and about 4m asl, respectively. IIIc includes the emergent beach of 1923 in its lowest part. Although the topographical separation between IIIb and IIIc is not necessarily clear, they can be distinguished from each other as IIIb surface is covered by dense vegetation including many pine trees which have been growing since before the 1923 coseismic uplift, whereas IIIc surface has no vegetation. Therefore, it is almost certain that at least two times of intermittent emergence had occurred between the formation of Terrace IIIa and the 1923 coseismic uplift. However, the dates of emergence have remained undefined, because ¹⁴C dates of fossil shells which were collected from Terraces IIIb and IIIc are so much scattered implying that the materials are not in situ. In the northern part of the island there is a prominent flat surface higher than Terrace IIIa, 10-15m asl, where Hatsu-shima village is situated. In this paper it is named IIIa' Surface and considered a secondary surface modified from Terrace IIIa by landslides of a back scarp and artificial alteration based on its sedimentological features, surrounding topographical features, and rather young ¹⁴C dates of materials obtained from this surface.

抄録全体を表示

Electric field on the sea-floor have been observed by making use of the electric powerfeeding arrangement for the submarine system which was laid off the Tokai-oki, south coast of Chubu district, Japan. The submarine system consists of four seismographs and a tsunami-meter, and each instrument is in series through a submarine co-axial cable with the shore station (Omaezaki weather station). The direct current transmitted from the power supplier goes through the inner conducter of the submarine cable, and is released from the sea earth attached near the terminal apparatus, and sends back to the shore earth through the sea water and floor. As the direct current remains constant, the change in the supply voltage is equal and opposite in polarity to the extraneous voltage between the sea earth and the shore earth in distance of 110km.
It has been preliminary reported that the supply voltage had repeated the daily change with the amplitude of about 0.1 volt and the pattern was similar to the change in the geomagnetic field. In this paper, at first, the fluctuations in the supply voltage are compared with the geomagnetic and geoelectric fluctuations on land in detail. It is confirmed that the changes in the supply voltage express the charge in the submarine electric field which are mainly caused by geomagnetic fluctuations.
Anomalous changes of the supply voltage were recognized before and just after the Izuhanto-tohooki earthquake on June 29, 1980 with magnitude 6.7. One of the anomalous changes is the cluster of pulse-like changes from about 17 hours to about 20 minutes before the occurrence of the earthquake. Another is a concurrent bay-like change with the earthquake. These anomalous changes seem to be related with the earthquake.

抄録全体を表示

The acoustic emission that occurs during tri-axial compression test of anisotropic rock samples stratified was studied under confining pressure up to 300MPa. The m value of Ishimoto-Iida's formula was determined by the comparator with the seven channels of different threshold levels. Occurrence pattern of acoustic emissions and variation of m value for Kamioka gneiss were strongly affected by the direction of layers of rock. On the specimen with layers of rock oriented at 45° to the principlal stess axis, many axial stress drop were observed, which is caused by the existence of weak zone of the anisotropic rocks stratified. The m value changed before and after these axial stress drops. Frequency-amplitude relation of gneiss was represented by two straight lines before axial stress drop, but represented by a single straight line after axial stress drop. The m value of Miyama sandstone had decreased untile axial stress drop, and recovered after axial stress drop. Associated with small axial stress drop caused by local fracture in sample, the activity of forshocks was weak and the vigorous aftershock activity decayed rapidly. When the large axial stress drop was produced by the frictional sliding after several local fractures, the foreshock activity became higher immediately before axial stress drop, and then the activity of aftershock decayed slowly. The consideration from investigetions of acoustic emission are summarized with respect to the various types of axial stress drop occurring in anisotropic rock samples stratified.

抄録全体を表示

A precise relocation of earthquakes using data listed in the Seismological Bulletin of the Japan Meteorological Agency reveals a double-planed structure of the deep seismic zone in the whole region of the northeastern Japan arc. The two planes are nearly parallel to each other and the distance between them is from 30km to 40km. The lower seismic plane appears at depths greater than about 60km in the Tohoku District, northeastern Japan, while that appears at depths greater than about 90km in the Kanto District.
Some interesting characteristics are found from focal mechanism study for large earthquakes and microearthquakes in the upper seismic plane. In the Tohoku District, the predominant fault-plane solution for earthquakes shallower than about 60km is low angle thrust faulting and that for deeper earthquakes is down dip compression. In the Kanto District, similar characteristics are found for earthquakes in the upper seismic plane, except that the depth at which the predominant focal mechanism changes is about 90km. This difference may be closely associated with the subduction of the Philippine Sea plate overriding the Pacific plate in the Kanto District.
In the whole region of the northeastern Japan arc, the earthquakes in the lower seismic plane is characterized by down dip extensional fault. Down dip compressional and down dip extensional earthquakes form a pair in the double-planed deep seismic zone, since the depth at which the lower seismic plane appears is consistent with the depth at which the predominant focal mechanism changes in the upper seismic plane.
The depth to the upper seismic plane just beneath the volcanic front is about 100km in the Tohoku District, whereas the depth is much deeper in the Kanto District, where the triple junction among the Philippine Sea plate, the Pacific plate and the Eurasian plate is situated.

抄録全体を表示

This paper deals with full-scale measurements of surface ground motion generated by forcing to vibrate three kinds of footings, which were each other different for the size and the situated point, resting on the surface of surface or subsurface irregular ground which was turned out to the flat surface ground from a hill composed of a steep ridge and canyon geometry. Observations indicate that the surface variations of SH wave motion are highly frequency dependent and do not necessarily decay monotonously according as the sourcereceiver distance increases. It is the measured point nearest the edge between the cutting ground and the filling ground that the surface velocity of the subsurface irregular ground is most amplified.
In order to predict the amplification and focussing effect of the subsurface irregular site condition on the surface velocity amplitude variation, we assume two-dimensional arbitrarily-shaped alluvial valley model considering the footing-soil interaction effect. We compare the measured results with the theoretical results based on the method of boundary integral equation and get the good agreement between them.

抄録全体を表示

Taking account of the recent study of Tarakanov et al. based on a new method and attributing the large-scale geoid undulations to the lateral anomalies of density in the lower part of upper mantle, global-scale geoid undulations were examined and then an excellent correlation was found between the undulations given by the GEM-8 model and the P-wave velocity anomalies found for the uppermost part of lower mantle. The negative correlation that the regions appointed to positive geoid anomalies almost coincide with those appointed to negative velocity anomalies and vice versa could be interpreted based on shock-wave experiment data showing that a Mg-rich dunite exhibits higher velocities and lower densities than Fe-rich one. Thus it was inferred that the large global-scale geoid undulations can be attributed mostly to lateral anomalies of density reflecting compositional heterogeneities presumed in the uppermost part of lower mantle.
It was noticed that deep-focus seismic zones, called the Benioff zones, seem to extend near along the ocean side of the belts characterized by steepest descent on relief of the geoid surface possibly reflecting large lateral density variations in the uppermost part of lower mantle. This spatial relationship was considered as indicative of a genetic interrelation between the Benioff zones and the large density contrasts, and then a schematic model was proposed concerning the formation of the Benioff zones as a result of differential movements induced in the laterally heterogeneous upper mantle by density currents resulting from the large density differences in the above-mentioned part of lower mantle. In the case of a two-dimensional viscous layer model the velocity of the density currents was tentatively estimated to be of the order of 10^-1-10^-2km/my, which seems sufficient for the formation of a weak zone corresponding to the Benioff zone during the Meso-Cainozoic tectonic cycle of development of active continental margins.

抄録全体を表示

R. G. A. (Residual Gravity Anomaly) was calculated along a section across the Hawaiian Ridge near Oahu Island by use of both the gravity data and the crustal velocity structure obtained by the explosion seismology. Anomalies of lithospheric thickness in this region were estimated from the result of the calculation of R. G. A. It is concluded that the thickness of lithosphere beneath the island is about 20k mthinner than that beneath the surrounding area as expected from the lithospheric thinning model.
The values of R. G. A. in the Emperor Seamount Chain are smaller than those in the Hawaiian Region obtained in this work, so the thinner lithosphere beneath the seamount approaches the normal state with increasing the age of the seamount in the Hawaiian-Emperor Seamount Chain. This agrees with the lithospheric thickening model.

抄録全体を表示