The Atotsugawa fault system, which is located in the northern region of Central Honshu, Japan, is composed of the Amidagawara, Atotsugawa, Mannami, Mozumi-Sukenobe, and Ushikubi faults. The Atotsugawa fault in the central part is creeping where seismicity is low. In contrast, the edges of the creeping zone are locked showing relatively high seismicity. In order to study the relationship between the distribution of earthquakes and the fault structure, we analyzed seismic survey data along and across the fault system, and relocated hypocenters using a 1-D velocity structure determined in this study. P-wave velocity structures around the Atotsugawa fault system are determined by comparing first arrival travel time data from explosion surveys with the travel times from forward calculations by a ray tracing method. On the basis of the P-wave velocity structures, we have found that the crust around the Atotsugawa fault system consists of three layers, an upper crust which includes a surface layer, a middle crust and a lower crust. Furthermore, two distinct reflectors are located at depths of about 11km and 20km below the Atotsugawa fault system. The depth of the shallower reflector is close to that of low resistivity layer. Comparing the relocated hypocenters with the depths of the reflectors, the shallower reflector is roughly coincident with the base of the seismogenic layer and the second reflector is several kilometers deeper. In addition, seismicity is concentrated in the upper crust (with velocities of 5.9-6.2km/s) and only a few earthquakes occur at the bottom of the middle crust. The difference in seismic structures between the creeping zone and the other regions is not clear. This is probably because the creeping zone does not have a significantly different velocity structure that is detectable from these data.
We analyzed shear-wave splitting beneath Hokkaido and Tohoku, Japan. Waveforms of 912 intermediate-depth earthquakes recorded at 239 seismic stations were used, and 2276 splitting parameters, the leading shear-wave polarization direction (fast direction) and delay time between two split waves, were observed. Obtained results show that most of fast directions observed at stations in the back-arc side are perpendicular to the trench-axis, whereas those at stations in the fore-arc side are sub-parallel to it. Average delay times observed at stations in the back-arc side are 0.1-0.5 s, and those at stations in the fore-arc side are smaller (0.05-0.1s). This systematic spatial variation is observed for the whole study area, from the southern part of Tohoku to the eastern part of Hokkaido. We infer that the anisotropy caused by lattice-preferred orientation of olivine, which is probably attributable to the upwelling flow portion of the mechanically-induced convection in the mantle wedge, is a likely candidate for the shear-wave splitting in the back-arc mantle wedge. If this is the case, the present observations for the east of the arc-arc junction may indicate that the upwelling flow direction is sub-parallel not to the relative plate motion direction but to themaximum dip-direction of the subductiog slab, in contrast with that in the south of the arc-arc junction where the upwelling flow sub-parallel to the relative plate motion is expected. The anisotropy structure revealed in this study plays a crucial role on understanding ongoing mantle dynamics in subduction zones.
We investigate changes in seismicity and the b-value in the Tokai region, Central Japan, based on the new magnitude catalog of JMA. We show seismicity in the subducted slab increased remarkably in October 2000 and decreased in October 2001. In contrast, seismicity in the crust decreased in October 2000 and increased in October 2001. We propose that these changes in seismicity in the slab and crust in the Tokai region were caused by the changes of the coupling state. The change of the displacement rate at locations of GPS stations indicates that the interplate coupling was weakened in late 2000 and it resumed partly in late 2001. Correlation is noted between change in seismicity and that in the b-value; the low seismicity corresponds to the high b-value. We consider the quiescence that appeared in August 1999 in the slab beneath the western coast of Suruga Bay might have been a precursor for the seismic activity including two M5-class earthquakes in April and June of 2001. Increase of seismic activities was observed in the slab and crust as well as in the Zenisu region in the beginning of 2003. We suggest that the increase might have been generatedby the acceleration of the subduction of the Izu micro-plate on the Suruga-Nankai Trough.
The inversion analysis using seismic intensity data is carried out to investigate the source characteristics of major damage earthquakes that have occurred in the sea off Miyagi Prefecture since 1861. It can evaluate short-period seismic wave radiation zones (SPRZ) on an earthquake fault plane. Comparison of the result of the inversion analysis for the May 26, 2003 Off-Miyagi earthquake with the source region deduced from reported aftershock activity verified sufficient accuracy of the present method. We compared the location of SPRZ among analyzed earthquakes and identified the kinds of earthquakes. The SPRZ of the June 12, 1978 earthquake was located near the seacoast of Miyagi Prefecture, adjacent to the large slip area obtained from the waveform inversion by Yamanaka and Kikuchi (2004). The SPRZ of the November 3, 1936 and July 27, 1937 earthquakes were located on the south of that of the 1978 earthquake, which may indicate that the rupture area of the 1936 and 1937 earthquakes were different from that of the 1978 earthquake. On the contrary, location of the SPRZ of the February 20, 1897 earthquake was similar to that of the 1978 earthquake. The SPRZ of the August 5, 1897 earthquake was located near the Japan Trench, while that of the April 23, 1898 earthquake was just beneath the coast of northern Miyagi Prefecture, suggesting that it was an intermediate-depth intraslab earthquake the same as the May 26, 2003 event. The location of the SPRZ of the 21 October, 1861 earthquake was mostly consistent with that of the 1978 earthquake. We consider that the 1861 earthquake should have occurred on the upper boundary of the subducting Pacific plate, not in the inland of northern Miyagi Prefecture.
The Ishii-type three-component strainmeter that can detect strain change as small as 10-10 is expected to catch a precursor of the anticipated Tokai earthquake. The instrument uses magnesensor, so that the output is also affected by magnetic fields. The response of the sensor to the geomagnetic field makes a noise that might mask signals produced by changes of the tectonic stress. However, that feature of the instrument can be utilized to get information about the underground electric conductivity, because the geomagnetic fields recorded by the magnesensor include those generated by induced electric currents. With the expectation to detect change in the electric conductivity beneath the strainmeter stations at Kakegawa and Haruno, we investigated temporal change in the pro-portionality between the outputs of the three-component strainmeters and the geomagnetic field at Kakioka situated on a stable crustal block about 250km to the northeast from the Tokai region. Using one-minute sampling data at times of geomagnetic storms during the period from April 1998 to November 2002, we calculated the response coefficient of strain to geomagnetic field at each of the storm times. It was found that the response coefficient increased with time at first, but the tendency changed in the midst of 2000 for both stations at Kakegawa and Haruno. The temporal change of the response coefficient and the long-term change of the strain are alike when the linear trend is subtracted from the latter. We think the change of the trend in the response coefficient in mid 2000 might have been related to changes in the electromagnetic conductivity beneath the Tokai region.
This paper reviews studies on the relation between the eventual earthquake size and its rupture nucleation size. Although a large number of studies have been made in order to find whether the seismic nucleation phase depends on the eventual earthquake size or not, this subject is still in controversy. Several recent papers have reported that the duration of the seismic nucleation phase scales with the final rupture size, and that the scaling relation can be explained by rupture nucleation models. These studies suggest that larger preslip occurs before a larger earthquake. On the contrary, other papers have reported that the seismic nucleation phase is independent of the earthquake size; earthquakes of all sizes initiate in a similar manner. Adding to these existing studies, we perform a numerical simulation of sliding behavior on a fault assuming two asperities of different sizes. The result shows that when the rupture of the small asperity triggers the rupture of the large asperity, short term preslip occurs at the small asperity, and the magnitude of the preslip does not depend on the eventual earthquake size. However, intermediate-term aseismic slip which occurs around the main asperity depends on the eventual earthquake size.
For earthquake prediction, we should strategically link several methods, exchange and share earthquake-related information. The hydrological method has been recently developed and not yet had a suitable theory which can adequately explain hydrological precursors to earthquakes. Therefore it is important to clarify the mechanism of earthquake-related hydrological changes and turn the hydrological information into the information which can be easily understood by seismologists. From this point of view, we propose four roles of hydrological methods for earthquake prediction; (1) providing crustal deformation data estimated from hydrological changes, which is usually useful for the short-term prediction, (2) providing the information related to displacements of ground surface caused by groundwater level changes, which will enable us to eliminate some non-tectonic deformations of ground surface and contribute to improving S/N ratio in the geodetic measurement such as GPS observations, which will be useful for the mid-term or long-term prediction, (3) providing the information related to temporal changes in permeability in and around active fault zones, which contributes much to understanding the earthquake-cycle in the active faults and the long-term prediction, and (4) providing the information of pore-pressure changes in the focal region. At present, only the first role can be applied to the practical short-term prediction. Therefore detecting more reliable hydrological precursors to earthquakes depends on how they are connected to preseismic crustal deformation. The ‘strain model’ attributes groundwater changes to volumetric strain changes, and is the only model that can quantitatively explain the precursors at present although it needs more improvement. For the improvement, comparing groundwater level changes with crustal deformation is effective.
An earthquake of M7.4 occurred off Tokaido region of southeast off the Kii Peninsula, Japan, on September 5, 2004, accompanied with a foreshock of M7.1 which proceeded about 5 hours and was located off the Kii Peninsula about 40km west from the main shock. Aftershock distribution were divided into 4 groups; the area of the M7.1 foreshock along the axis of the Nankai trough (group I), the area of the M7.4 main shock along the axis of the trough (group II), the area perpendicular to the trough axis containing the main shock (group III), and the area apart from the main shock in NE direction (group IV). The CMT solutions of groups I and II are thrust fault type that has a pressure axis in almost N-S direction. The focal mechanisms of group III are strike-slip fault type that has a nodal plane with the strike in NW-SE direction. The strike direction of the NW-SE direction is consistent with the aftershock distribution. The focal mechanisms of group IV are strike-slip fault type that has a nodal plane with similar strike but with an opposite slip direction to the group III. Micro-earthquakes with magnitude 3 began to occur in the main shock area, corresponding to occurrence of the foreshock of M7.1. The focal mechanisms of these earthquakes show the type of the group II. Aftershock activity of the main shock in both areas of group II and group III began almost at the same time. The aftershock activity shows ordinary decay with time, however, the number of earthquakes in the NW area of the group III decreased rapidly with time. In addition, slightly low frequency earthquakes occurred in the NW part of group III. Thus the group III is suggested to have different characters from groups I and II.
This article reviews studies on ground motion characteristics during the 2004 earthquake sequence off the Kii peninsula and the Tokai district occurred on September 5. Long-period ground motions are observed at stations in the large sedimentary basins such as Kanto, Nobi, and Osaka. Typical predominant periods of ground motions at each basin are approximately 8-10s for Kanto basin, 3s for Nobi, and 6s for Osaka, respectively, which corresponds to thickness of sedimentary layers of each basin. Long-period ground motion modelings for the event were done by several authors for confirming and improving underground velocity structure models. The reliable underground structure model confirmed by this waveform modeling shall be used for strong ground motion simulation of the anticipated Tokai, Tonankai and Nankai earthquakes.
We obtained detailed aftershock distribution and three-dimensional seismic velocity structure in and around the focal area of the 2004 Mid Niigata Prefecture (Niigata-Chuetsu) earthquake (M6.8) by inverting travel time data from a dense observation network and surrounding permanent stations. We adopted double-difference tomography method for the inversions. The fault planes of the main shock and the largest aftershock, which are delineated by two parallel aftershock alignments, took place where seismic velocity changes abruptly at the boundary between the lower velocity hanging wall and the higher velocity footwall for both P- and S-waves. The seismic velocity structure changes along the strike of the faults. The major velocity boundary zone between the lower velocity hanging wall and the higher velocity footwall migrates to the west at a location near the central part of the focal area, where the main shock hypocenter is located. Some parts of the fault planes, e. g. around the hypocenter of the M6 aftershock on Oct. 27, are imaged as low velocity zones. Asperity (large coseismic slip area) of the main shock seems to correspond with a higher velocity portion along the fault plane.
The fault model for the 2004 Mid-Niigata Prefecture Earthquake is examined based on the nature of surface fault, the results of leveling survey and the aftershock activity. The surface faults of the Mid Niigata Prefecture Earthquake in 2004 were found along the pre-existing active fault traces; the Obiro fault and the northern part of the western marginal fault of the Muikamachi basin (WFM). The amount of vertical displacement was 10 to 30cm. The elevation change of benchmarks indicates that the southwestward extension of the Obiro Fault (Suwa-toge Flexure) corresponds to the source fault. The slip on the fault was ca. 1.8m. The hypocenter distribution shows that the Obiro Fault is dipping toward WNW with about 50 degrees, and that the WFM extends to the north with the same dip as the Obiro Fault. In the central part of epicentral area, there is a double planar structure composed of these faults. Both the slip on the Obiro Fault and the small displacement of the surface earthquake fault (WFM) have proved indispensable in matching the calculated deformation with the observed one. There are two fault models for the earthquake; 1) low-angle thrust (WFM) branches off from the tip of the high-angle Obiro Fault at a depth of 2.8km (upward-flattening structure), 2) WFM is a high-angle reverse fault parallel to the Obiro Fault (double planar structure). Geomorphologically, the double planar structure is the most probable one. The Obiro Fault is responsible for the main shock. The WFM can be correlated with the source fault of the largest aftershock with a small amount of slip.
Physical experiments were performed to gain a better understanding on the deformational mechanism of the upper Cenozoic system in the epicentral area of the 2004 Mid-Niigata Prefecture earthquake (MJMA 6.8), Japan. The present study focuses on deformation of sedimentary cover, including overpressured mudstone, caused by reverse movements along the basement fault corresponding to the source fault responsible for the main shock. Our physical models comprise dry quartz sand representing brittle sedimentary rock and viscous silicone polymer representing overpressured mudstone. Computerized X-ray tomography to analyze the kinematic evolution and geometry of folds and faults in the models indicates that minor folds and thrusts with minor displacement developed on the footwall of the major monoclinal flexure within the sedimentary cover above the steeply dipping reverse basement fault. The thrust that was diverged from the shear zone within the silicone polymer layer reached the top surface of models. These results compare well with the geometry and kinematic evolution of the major monoclinal flexures, minor folds and thrust in the epicentral area. Minor folds may have been developed during formation of the major monoclinal flexure propagating from the steeply dipping reverse basement fault, which corresponds to the source fault responsible for the main shock. The thrust beneath the surface rupture associated with the earthquake is inferred to be the out-of-syncline thrust diverging from the shear zone that runs along a certain overpressured mudstone layer. Shearing along the overpressured mudstone and gravity sliding within the sedimentary cover may play important roles for the development of these geologic structures in the epicentral area.
A waveform inversion was conducted for the rupture process of the 2004 Mid Niigata Prefecture (Niigata-ken Chuetsu) earthquake (MJ 6.8) by using aftershock records as Green's functions. Paying attention to the similarity of the group delay time between the mainshock and aftershock records, two aftershocks were selected to be used in the inversion. As a result of the inversion, a rupture model was obtained that can explain mainshock records at periods from 1 to 5s. Although only portions of 10s including S wave were used for the inversion, the later phases, which were not used for the inversion, can also be explained fairly well with the present rupture model. Another inversion, in which only one aftershock was used, was also carried out. In this case, the mainshock recordings were not explained very well, indicating the neccesity of at least two aftershocks in the analysis. The primary asperity found in the present rupture model is close to the heavily-damaged area in Tamugiyama, which was found by the investigation team of the Niigata University.
On October 23, 2004, a magnitude 6.8 earthquake occurred within the central part of Niigata prefecture; this seismic event is known as the Mid Niigata prefecture Earthquake in 2004. We investigate the subsurface structure in the area of the earthquake via receiver functions of teleseismic waveforms recorded at the Hi-net stations. The Hi-net stations located close to the aftershock region have been continuously recording ground motions since October 2000. For receiver functions estimated at stations located in areas of low relief, the arrival time of the initial pulse is clearly delayed, and its width greater than that estimated at stations in hilly and mountainous areas. These observations can be explained numerically using one- and two-dimensional model structure that contains near-surface sediments of very low velocity. To evaluate the detailed subsurface structure, we invert receiver functions for seismic velocity structure beneath each station. An extremely low velocity layer is detected beneath many stations. In the eastern part of the aftershock area, the thickness of this low velocity sediments is less than 2km, but in western areas the sediment thickness increases to more than 5km. Beneath KWNH station, on the southwestern edge of the aftershock area, the S wave velocity of the sediments is relatively low, and sediment thickness is estimated to be approximately 10km. The thickness and velocity of the sedimentary layer varies across the study area, consistent with variations in topography and surface geology. We evaluate the depth to the Moho from estimated velocity models. The depth contours of the Moho trend parallel to the coast beneath the central part of Niigata prefecture. A small ridge in the Moho is observed in the area of the boundary between the Nagano and Gunma prefectures. The gravity anomaly distribution in the study area can be explained by the variations in the Moho depths and the distributions of the low velocity sediments.
We conducted temporary GPS observations shortly after the occurrence of the 2004 Mid-Niigata Prefecture Earthquake (MJMA 6.8) at 5 sites around the southern part of its source region. These temporary sites are located between the existing continuous sites of GEONET so that we could resolve the crustal deformation pattern in more detail. We analyzed the observed data with those from surrounding GEONET as well as IGS GPS stations. The obtained daily coordinate changes revealed postseismic crustal deformation, which was characterized by the E-W shortening of the source region. Maximum postseismic displacement of about 20mm was found at Ojiya GEONET station and OGNI temporary station. In addition, the maximum uplift of about 20mm was detected at the two sites. The postseismic deformation faded out about 1 month after the main shock. It is considered that the postseismic deformation was caused by aseismic after-slip of shallow (depth<5km) fault. The fault is of a reverse type, but it is difficult to identify its dipping direction. Aseismic nature of the after-slip can be attributed to the characteristic crustal structure with thick soft sediments lying over the main shock source region. More studies are needed to resolve the generating mechanism of concentrated compressive strain in the source region of the Mid-Niigata Prefecture Earthquake.