The Kofu Basin is a tectonic basin in the South Fossa Magna. On the southeast and west margins of the Basin, there develop two tectonically peculiar terrains the Sane Hills and the Ichinose Upland bordered with active faults. These boundary faults are considered to play an important role in the tectonics of the Basin, but those subsurface structures are not yet known well. We carried out a dense gravity survey in and around the Basin and contoured a detailed Bouguer anomaly map in the area concerned. On the west and southeast of the Basin, steep horizontal gravity gradient zones distribute in parallel along the active faults and they develop on the mountain side of the faults. This gravity feature suggests that the mountains are thrust over the Basin. We made two-dimensional subsurface structure analysis along two sections to cross the Sone Hills and the Ichinose Upland. It is concluded that the density boundaries are on the northern edge of the Sone Hills and on the western edge of the Ichinose Upland, respectively.

抄録全体を表示

A tsunami was generated by the 1994 Hokkaido Toho-Oki earthquake and was observed at tidal stations on the Pacific coast of the Japanese Islands. It was observed also at several tidal stations on Hokkaido coast of the Okhotsk Sea, where the arrival of the initial wave was recorded earlier than the expected time: for example, the initial motion of the tsunami was recorded at Utoro 17 minutes after the mainshock, while that the theoretically expected tsunami arrival time from the source in the south sea region of Shikotan Island is 80 minutes after it. We tried to estimate the location of the source of the early coming waves by drawing the inverse refraction diagram from each station. We suggest two probable locations of the source; one is the area 80 kilometers north of Abashiri, and the other is the area 40 kilometers north of the tip of Shiretoko Peninsula. The seismic activity at both of those areas is poor. So, we cannot judge that the early coming waves were generated by some another events. Both of those areas are situated in the continental slope regions and with steep slopes. We suppose that a land slide at either of those areas caused the early coming waves.

抄録全体を表示

The records on a disastrous earthquake of July 25, 1882, southern Okinawa Island were discovered for the first time from nine archives. The most important are: “Okinawa-nisshi” written by Hatakeyama, a secretary of prefectural governor Uesugi; “Ryukyu-iki-nisshi” written by Ozaki, high official, dispatched from the central government to inspect a political situation in the Okinawa Prefecture; the official diary of the Okinawa Prefecture; and, the official document from the Okinawa Prefecture to the central government. Synthesizing these records, the earthquake happened between 1 and 2a. m., on July 25th, 1882, and aftershocks were intermittent for seventy days following. While there were no casualties, 500 stone walls were broken in Naha City. Analysing these records, the seismic intensity of this earthquake was probably 5. The reason why this earthquake was not discovered for about one hundred years is, firstly, the records of “Kyuyo”, an official document of the Ryukyu Kingdom, had already stopped, because the Kingdom was ruined three years before the earthquake, and secondly, the Japanese earthquake observation system was insufficient at the time of the earthquake.

抄録全体を表示

An estimation process of a seismic velocity structure model from P and S wave arrival time data usually represents an iterative modification of an initial model guided by an evaluation criteria. The apparent success of the various criteria in use is partially due to the use of an initial model which restricts the investigated models, and the effectiveness of these criteria without an initial model remains unknown. In this paper, we apply Wallace's minimum message length criteria for evaluating the goodness of a seismic velocity structure model. In our criteria, the goodness-of-fit of a model is measured from the length of the data encoded by 1 and 0. The minimum message length criteria is similar to the criteria based on the maximum likelihood method. The main difference between the two approaches lies in the point that the former evaluates the goodness of a model from the viewpoint of data compression, which enables it to compute the probability of a residual with various methods. Our criteria has been applied to the northern part of Kinki district and the eastern part of Chugoku district in Japan, where several seismic velocity structure models have been determined by five experiments of explosion seismology. A set of 665 micro-earthquake events observed from 1987 to 1991 was chosen as the input data. Our result coincides with the western parts of the models derived from the results of the explosion experiments, although an empirical evaluation criteria was unstable and required a larger set of data, and the root-mean-square-error criteria completely failed to estimate a reasonable model. As the consequence of several experiments, our minimum message length criteria is superior to the other criteria when an initial model is not available, and is also effective when an initial model is available.

抄録全体を表示

The abrupt appearance of metazoa at the Precambrian and Cambrian boundary has been regarded as one of the most important events in the history of life on Earth. Recent progress in stratigraphical studies of the Late Proterozoic strata and their correlations has revealed that the prelude of early metazoan evolution involved several severe climate changes, the deposition of banded iron formations and phosphorites, large isotopic fluctuations of carbon, strontium and sulfur, and the rifting and collision of continents. These data imply that a significant disturbance in the biogeochemical cycle resulted in large fluctuations of climate and ocean circulation patterns, which in turn triggered biospheric evolution. The sequence of events might have been ultimately caused by mantle dynamics. Conversely, it is highly likely that the appearance of metazoa was a driving force for biogeochemical reorganization.
In this review, the current methodology of the Late Proterozoic stratigraphy is described, with the temporal correlations of major events during the Late Proterozoic to the Early Cambrian summerized. The interesting features discussed in connection with the Cambrian explosion are, (1) the close temporal correlations among ice ages, banded iron depositions, and negative excursions of the carbon isotopic curve, (2) negative excursions of the Sr isotopic curve which probably resulted from the rifting of former continents, and (3) the rapid rise of the Sr isotopic curve as correlated to the Pan-African orogeny and the deposition of massive phosphorites, which is well correlated to the Cambrian explosion.
A working stagnant ocean model is proposed to explain the above apparent correlations in a unified framework. It is postulated that the ocean was basically stagnant during the Late Proterozoic, the result of a massive supply of ferrous iron from hydrothermal activities. This stagnant ocean removed carbon dioxide as trapped organic carbon in sediments, utlimately leading to the ice ages. The anoxic ocean separated organic carbon in the deeper water and molecular oxygen at the site of photic zone bio-synthesis. This resulted in the gradual increase in the partial pressure of oxygen in the shallow water and atmosphere, as well as the storage of organic matter within sediments. Ubiquitous reducing agents, such as ferrous ion in the ocean or volcanic gases from the mantle, however, prevented a rapid increase in oxygen pressure during the Riphean and Sturtian eras. With the Earth's aging, the supply of reducing agents from the Earth's interior diminished. After the final removal of ferrous iron through oxidation, coupled with glacially induced vertical oceanic circulation at the onset of the Vendian, a rapid rise in atmospheric oxygen pressure occurred. This may be the major reason for the appearance of large body-sized metazoa. Since studies of Precambrian-Cambrian boundary events should be one of the most important topics in the Decoding Earth Evolution Program, future problems and research directions discussed here should stimulate efforts to improve the present crude model and create a deeper understanding of the causes and consequences of the appearance of metazoa.

抄録全体を表示

It is known that graben structure-the Onohara-Ai Fault system-exists sub-parallel to the Arima-Takatsuki Tectonic Line (ATL) in the northwestern part of the Kinki Triangle. In contrast, the ATL consists of east-west trending right lateral faults. Two seismic reflection surveys with a total length of 8.5km were carried out along the Yodo and Ina Rivers which cross the ATL. The results from these surveys revealed the subsurface structure along the ATL and are as follows:
1) Four seismic units were established in this area.
2) Some graben structures exist on the southern side of the ATL.
Based on the azimuthal direction of the maximum horizontal compressive stress, the strike direction of the ATL, the focal mechanisms of aftershocks of the 1995 Hyogoken-Nanbu Earthquake, and the orientation of graben structures, we interpret the ATL as an oblique fault connecting the northern Hanaore Fault and the southern Rokko Faults, and suggest that the stress field has been released through strike-slip duplexes and a pull apart structure on the northern and southern portion of the ATL, respectively.

抄録全体を表示

We calculate static stress changes for the fault model derived from geodetic data before and after the magnitude=7.2 Kobe earthquake of January 17, 1995 in order to examine the possible correlation between changes in seismicity in the surrounding regions and stress changes.
The Coulomb failure function (CFF) for right lateral slip on NE-SW or nearly E-W trending vertical faults may increase in the north of the Arima-Takatsuki Tectonic Line, where the activation of seismicity is observed. CFF for left lateral slip on NW-SE trending vertical faults may also increase around the Yamasaki fault, northwest of Kobe, which is consistent with the activation of seismicity there. In the northern Kinki district, where seismicity became a little bit lower, CFF is predicted to decrease. CFF may change a little in the Wakayama region, where significant seismicity changes are not observed.
The Kobe earthquake may have loaded the Arima-Takatsuki Tectonic Line, segments of the Median Tectonic Line in Wakayama and Tokushima and westward dipping thrusts in the Osaka plane. Therefore we should carefully monitor the crustal activity around these tectonic lines.

抄録全体を表示

We determined the maximum slip vector on the surface fault associated with the Hyogoken Nanbu, western Japan, earthquake of January 17, 1995. The surface fault is dominantly strike slip with moderate to small amounts of vertical slip. A plane table map on a scale 1:100 was made at Hirabayashi, Awaji Island, where the maximum slip was observed. Using this map, directions and offsets of three artificial fault references were measured precisely. We successfully separated the two horizontal components of slip on the fault from these offset data, using a new method. The amount and direction of horizontal slip are 2.08^+0.08_-0.26m and 45.2°^+0.6_-6.5 (clockwise from the north), respectively, and the amount of vertical slip is 1.20±0.05m. The slip vector determined here, in conjunction with varying strike of the surface fault, may explain changes along strike in the nature of faulting (i. e., contractional on some segments and extensional on others).

抄録全体を表示

I constructed the regional stress fields of northern-central-southwest Japan on the basis of the forearc differential stress determined by ridge push and slab pull forces of the subducting oceanic plate, differential stress variation across the arc due to crust/plate structural variation, and the along-arc differential stress level. Assuming a simple geometry of northern-central-southwest Japan and a high stress level in the north-south direction in central-northern Japan due to collision of the Izu-Bonin arc and the south Kuril forearc, I constructed the principal stress profiles in these regions, which are consistent with the observed E-W σ_Hmax stress fields (i. e. strike and sense of motion of the reverse fauling in northern Japan and strike-slip faulting in southwest-central Japan). On the basis of this reconstruction, I concluded that the E-W compression which caused the 1995 South Hyogo earthquake resulted from the ridge push of the Pacific plate. This force, however, cannot be distinguished from the force at Fossa Magna due to the collision between the Amurian and Okhotsk plates. Higher seismicity in southwest Japan and in the foothills of central Japan than in the high mountain areas in central Japan is due to the larger horizontal differential stress in the former areas.

抄録全体を表示