Search

We assessed the seismic potential around the northeastern edge of the Longmenshan fault zone (LFZ) by analyzing the aftershocks of the 2008 Mw 7.9 Wenchuan earthquake, focusing on the right-lateral strike-slip Qingchuan fault (QCF). The data used in the present study were collected from a dense network of broadband seismographs, installed after the Mw 7.9 mainshock by the China Earthquake Administration. We analyzed 440 aftershocks that occurred around the northeastern part of the aftershock area between December 2008 and June 2009, the local magnitudes (ML) of which ranged from 2.0 to 5.2. We determined 127 well-constrained focal mechanism solutions for these aftershocks down to ML 2.5 by using P-wave polarity data in conjunction with body-wave amplitudes. An analysis of the stress tensor inversion revealed that, upon approaching the QCF from the LFZ, the stress field changes from a reverse faulting regime to a mixed regime containing strike-slip components and the orientation of the maximum compressional stress changes from a NW–SE to a WNW–ESE direction. This stress field favors a right-lateral strike-slip movement along the fault. The slip tendency analysis revealed that the QCF is regarded as being favorably oriented to the present-day stress field. The coseismic Coulomb stress changes induced by the mainshock increased along the preferred fault plane in the northeastern half of the QCF, which is further loaded by viscoelastic relaxation. This evidence suggests that the potential for a future earthquake in the QCF is high.

We re-evaluate the stress state across the Korean Peninsula by analyzing the stress tensor inversion results, strike–rake relationships, and faulting-type indicator–magnitude relationships of published earthquake focal mechanism data. Previous studies have shown that the stress state is characterized by a strike-slip faulting stress regime with ENE–WSW maximum compression, NNW–SSE minimum compression, and vertical intermediate stress. The differential stress magnitudes are estimated to be on the order of 100 MPa under the assumption of the strong-fault earthquake with a high Coulomb friction coefficient of 0.85. Conversely, other studies have shown that the weak-fault model is applicable across the Japanese Archipelago, where the differential stress level is only a few tens of MPa. We, therefore, review previous studies, evaluate the focal mechanism data across the Korean Peninsula, and infer the absolute stress level. Our comparison of the geological and tectonic histories of the Korean Peninsula and Japanese Archipelago suggests that the NNW–SSE tensional stresses in the Korean Peninsula are caused by slab rollback of the Philippine Sea Plate at around 5 Ma, west of Kyushu Island, southwestern Japan. We also find that the earthquake faulting types across the Korean Peninsula are closely related to their fault strikes, confirming a spatially uniform distribution of the strike-slip stress regime with nearly equal tensional and compressional tectonic stresses. However, this situation leads to the apparent paradox of strike-dependent fault strength. We eliminate this paradox by proposing a model whereby reverse- and normal-faulting events are caused by local stress heterogeneities around the edges of strike-slip faults under a uniform low-strength strike-slip stress state for all faulting types. We also highlight the usefulness of strike–rake and faulting-type indicator–magnitude diagrams in considering stress states.

The Asia-Pacific Region Global Earthquake and Volcanic Eruption Risk Management (G-EVER) Consortium among the Asia-Pacific geohazard research institutes was established in 2012 with the main objective of reducing the risk caused by earthquakes, tsunamis and volcanic eruptions world-wide. The first G-EVER workshop was held in Tsukuba, Japan in February 2012. The G-EVER1 accord was approved by the workshop participants which made 10 recommendations that focused on the enhancement of collaboration, sharing of resources and making information about the risks of earthquakes and volcanic eruptions freely available and understandable. The G-EVER Promotion Team in Geological Survey of Japan was organized in Nov. 2012 to coordinate the G-EVER related activities. The G-EVER Hub website was setup to promote the exchange of information and knowledge about volcanic and seismic hazards among the Asia-Pacific countries. Establishing or endorsing data interchange and analytical methods standards for geohazard institutes of the world are important to promote data sharing and comparative analyses. Several G-EVER Working Groups and projects were proposed such as the Next-generation volcanic hazard assessment WG and the Asia-Pacific region earthquake and volcanic hazard mapping project. The volcanic assessment support system is developed based on eruption history, volcanic eruption database and numerical simulations. The G-EVER hazard assessment support system is implemented with user-friendly interface, making the risk assessment system easy to use and accessible online. The Asia-Pacific region earthquake and volcanic hazard mapping project aims to make a sophisticated online hazard information system that provides past and recent earthquake and volcanic hazards information, risk assessment tools for earthquake and volcanic eruption hazards and links to global earthquake and volcanic eruption databases. The hazard mapping project plans to create the system with the cooperation of Asia-Pacific countries.

We propose a method for modeling a three-dimensional thermal structure with a particular focus on the lower crust. Our method enables high-resolution modeling without heat flow data, but with earthquake hypocenter data in the crust and seismic attenuation data in the upper mantle. In our method, the temperature at the bottom of the seismogenic layer is estimated under the assumption that the bottom depth corresponds to a brittle-ductile transition zone. Next, the temperature beneath the Moho discontinuity is estimated using seismic attenuation data of the mantle and a few point temperature data inferred from mantle xenoliths. Finally, the temperatures between the two depths are linearly interpolated. Attempts to construct an actual model for the Japanese Islands are also described.

Measurements of in-situ stress orientations at relatively shallow depths were conducted at 11 sites along the Anninghe-Zemuhe fault zone, southwestern China, with a newly developed method. The Anninghe fault in the fault zone has been considered to have a potential for a large earthquake with a magnitude of about 7.5, while the Zemuhe fault shows very little such potential. The present study has mainly two objectives: one is to show new data of the spatial distribution of the stress orientations around the fault zone, which is obtained directly from the measurements; the other is to examine the possibility to detect a stress pattern in relation to the past activities of earthquakes along the fault zone. The observed principal stress orientations are distributed between the NNW-SSE and NW-SE directions, which is consistent with the earthquake focal mechanism solutions near the fault zone. On the other hand, it is unclear whether the observed stress pattern is related to past activities of the fault zone, since the data fluctuation was too large to determine this. We note that the stress orientations are favorable for slip on the Anninghe fault, while being unfavorable on the Zemuhe fault, from the present experimental data.

The 2nd G-EVER International Symposium and the 1st IUGS & SCJ International Workshop on Natural Hazards aimed to encourage extensive discussions on the present situation of natural disaster mitigation of earthquake, tsunami, volcanic eruption and landslide in the Asia and Pacific regions. The discussions included (1) prioritization of important researches to make society strong and resilient, (2) development of suitable hazard maps which are essential to society and Asia-Pacific scale hazard assessment activities, and (3) the importance of contributions to solid earth science. There were also discussions on the mitigation processes needed in the next-decade to reduce the risks of natural hazards. The 2nd G-EVER International Symposium and the 1st IUGS & SCJ International Workshop on Natural Hazards was held in Sendai, Tohoku, Japan on Oct. 19-20, 2013 (Fig. 1). The symposium was organized by the G-EVER Consortium, Geological Survey of Japan, AIST, the International Union of Geological Sciences (IUGS) and Science Council of Japan (SCJ). It was co-organized by the Geological Society of Japan and the Coordinating Committee for Geoscience Programmes in East and Southeast Asia (CCOP). Tohoku University, Disaster Prevention Research Institutes (DPRI) of Kyoto University, National Research Institute for Earth Science and Disaster Prevention (NIED), GNS Science, Philippine Institute of Volcanology & Seismology (PHIVOLCS), the Seismological Society of Japan, the Volcanological Society of Japan, Japan Association for Quaternary Research, Japanese Society for Active Fault Studies, Japan Society of Engineering Geology, and Tokyo Geographical Society supported the symposium (Fig. 2). The Asia-Pacific Region is an area with a high risk of catastrophic natural disasters such as earthquakes, tsunamis and volcanic eruptions. In today’s highly globalized economy, when a major disaster occurs, it can create unpredictable turmoil not just in the affected area but also the rest of the world. Countermeasures against these natural disasters are crucial for the sustainable development. We believe that continuous efforts to develop an effective international framework to reduce the risk of natural disasters are very important. It is expected that geological institutions will collaborate and create a system to gather and process information to formulate disaster mitigation measures in the Asia-Pacific Region. This system must be easy to use and freely accessible to the public to meet societal needs. The Sumatra earthquake of December 26, 2004 and Tohoku earthquake of March 11, 2011 clearly show the urgency of developing an information and knowledge system for infrequent natural hazards. We also have to recognize that smaller events could have adverse effect to society. For example, the magnitude of the February 2011 Christchurch earthquake was just 6.3 (ML), but caused major damage to the second largest city of New Zealand and 185 casualties. Volcanic ash ejected from the Eyjafjallajokull eruptions in Iceland on April 2000 caused more than 20,000 airline flight cancellations a day in Europe, resulting to the largest air-traffic shut-down since World War II. The IATA estimated that airline companies lost US$200 million a day during the shut-down. After the 2011 Tohoku earthquake, more efforts for the prevention and reduction of the risks of natural disasters have been made all over the world. The G-EVER Consortium promotes natural disaster risk reduction activities through the collaboration different research institutes worldwide. The consortium’s major activities are the following: 1. Establish a framework for cooperation of research institutes and organizations working on volcanic disaster prevention in the AsiaPacific region.

A multi-offset hydrophone vertical seismic profiling (VSP) experiment was done in a 747 m deep borehole at Nojima Hirabayashi, Hyogo prefecture, Japan. The borehole was drilled to penetrate the Nojima Fault, which was active in the 1995 Hyogo-ken Nanbu earthquake. The purpose of the hydrophone VSP is to detect subsurface permeable fractures and permeable zones and, in the present case, to estimate the permeability of the Nojima Fault. The analysis was based on a model by which tube waves are generated when incident P-waves compress the permeable fractures (or permeable zones) intersecting the borehole and a fluid in the fracture is injected into the borehole. Permeable fractures (or permeable zones) are detected at the depths of tube wave generation, and fracture permeability is calculated from the amplitude ratio of tube wave to incident P-wave. Several generations of tube waves were detected from the VSP sections. Distinct tube waves were generated at depths of the fault zone that are characterized by altered and deformed granodiorite with a fault gouge, suggesting that permeable fractures and permeable zones exist in the fault zone. Tube wave analysis shows that the permeability of the fault gouge from 624 m to 625 m is estimated to be approximately 2 × 10−12 m2.

We investigated the fault-zone structures (fault-zone width, shear-wave velocity, and Q s) of the Nojima fault, Japan, using the Love-wave-type fault-zone trapped waves (LTWs) recorded at two borehole stations—TOS2 and HRB—located along the fault. TOS2 (depth: 1673 m) is located at an end of the fault, while HRB (depth: 600 or 720 m) is located in the middle section of the fault where the greatest surface displacement of 2 m was recorded during the 1995 Kobe earthquake. The distance between the two stations is about 4 km. We found 22 records that exhibit typical LTW. We assumed the fault-zone structure to be a 2D uniform low-velocity wave guide and estimated the averaged fault-zone structure from the hypocenter to receiver by modeling the LTWs. Because we can infer the low-velocity fault zone to a depth of 8 km by observing the duration of LTWs, we obtain the average fault-zone structure from the borehole stations to a depth of approximately 8 km. The width, shear-wave velocity, and Q s of the fault zone beneath HRB are 100 m, 2.9 km/sec, and 60, respectively. On the other hand, the average width of the fault zone beneath TOS2 is 220 m, which is larger than that beneath HRB. We also conducted 3D finite-difference modeling of LTWs to confirm the spatial variations in the fault-zone structure to a depth of 8 km. The numerical simulation suggests that the Nojima fault has uniform elastic and attenuation properties along the fault zone to a depth of 8 km, while the width of the fault zone increases with the distance from the HRB to the south.

Electrical conductivity and seismic velocity are studied for plausible pore geometries in the Earth's interior for reliable quantitative analysis of experimental data such as seismic tomography and magnetotelluric explorations. Electrical conductivity of a two-phase system with equilibrium, interfacial energy-controlled phase geometry is calculated for the dihedral angles θ = 40°–100° that are typical for rock–aqueous fluid and θ = 20°–60° for rock–melt systems of lower crust and upper mantle for the case of tetrakaidecahedral grains. Electrical conductivity vs. seismic velocity correlations are acquired by combining of the simulated electrical conductivities with the seismic velocity calculated with the help of equilibrium geometry model Takei [Takei, Y., Effect of pore geometry on VP/VS: From equilibrium geometry to crack. J. Geophys. Res. 107 (2002): 10.1029/2001JB000522.] for the same pore geometries. The results show that electrical conductivity gradually decreases reaching zero when seismic velocities reach seismic velocities of intact rock for rock–melt systems, while for rock–aqueous fluid systems with θ ≥ 60° conductivity drops to zero at velocities up to 10% smaller. This can explain the seeming discrepancy of the low seismic velocity region, attributed to the high fluid fraction, and the low electrical conductivity of the same region, which is sometimes faced at collocated electromagnetic and seismic experiments.

We investigate the seismicity, fault structures, and stress field in the southern seismic gap adjacent to the source region of the 2004 Mid-Niigata earthquake based on precise earthquake locations and focal mechanism solutions. The double-difference earthquake locations reveal detailed fault structures including five discrete clusters. A stress tensor inversion method indicates that, at the southern ends of the Muikamachi and the Tokamachi faults, the stress state changes from thrust faulting to a strike-slip faulting regime. We show two possibilities for the deep-fault geometry of the southern part of the Muikamachi fault. Our preferred fault geometries of the Muikamachi fault differ from that of the mainshock, especially for the geometry of the dip direction. It is likely that the Muikamachi fault is divided into two segments whose boundary corresponds to the southern end of the source region of the 2004 event, and that the rupture area was terminated due to this fault segment boundary. The Coulomb failure stresses that are induced by the mainshock and the largest aftershock increase at the southern part of the Muikamachi faults. Furthermore, the angle between the maximum principal stress and the preferred fault planes implies that the shear stress resolving along the planes are large. This evidence suggests a high seismic risk in the seismic gap.

Quantitative analysis of collocated seismic velocity tomography and electromagnetic experiments is developed to elucidate the structure of the deep extension of the Nagamachi-Rifu fault, northeastern Japan. P and S wave seismic velocities obtained from a dense seismic network are examined and a ratio of spatial variation in P and S wave velocities d ln Vs/d ln Vp is chosen as a proxy for the influence of pore geometry. The analysis shows that the deep extension of the Nagamachi-Rifu fault reveals the d ln Vs/d ln Vp values exceeded 1.1. Such large values of d ln Vs/d ln Vp cannot correspond to equilibrium pore geometry, at which the interfacial energy is at a minimum, and indicate regions with non-equilibrium state where non-isotropic stress prevents the equilibrium pore geometry to be achieved. To specify a fine distribution of porosity and connectivity of micropore in the region, we carry out the joint analysis of the seismic velocities with the electrical resistivity data obtained by the magnetotelluric survey crossing the Nagamachi-Rifu fault. It is shown that the region at 10–17 km depths at about 20–40 km to the northwest from the hypocenter of the M5.0 earthquake occurred in 1998 exposes the highest connectivity among the adjacent areas, suggesting a strong deformation process.

SUMMARY In order to estimate the averaged fault zone structure, i.e. the width, shear wave velocity and Qs of the Mozumi-Sukenobu fault, central Japan, we analysed the Love-wave-type fault zone trapped waves recorded by a seismic array deployed in a subsurface survey gallery. We selected the typical Love-wave-type trapped waves based on the characteristics of dispersion, amplitude, apparent velocity and the polarization of wave trains following the direct shear wave. We selected nine events showing typical Love-wave-type trapped waves from 154 earthquakes. Epicentres of the events were located along the surface trace of the Mozumi-Sukenobu fault. The nine events showing trapped waves were found to be located in a plane dipping to the south at an angle of approximately 75° to 85°. The duration of the fault zone trapped waves increase non-linearly with the increase of hypocentral distance. This result implies an increase in shear wave velocity of the fault zone with depth and/or horizontal distance from the array. In order to obtain averaged fault zone structure from hypocentre to receiver, we assumed fault zone structure to be a 2-D uniform low-velocity waveguide, and modelled the dispersion curve and the waveform of trapped waves using an analytical solution. The width, shear wave velocity and Qs of the fault zone were estimated to be 160 to 400 m, 2.9 to 3.1 km s−1 and 60 to 90, respectively. The width of the fault zone estimated in this analysis is consistent with the width of the extent of fault zone at the survey gallery. We also conducted 3-D finite difference simulation (3-D FDM) on fault zone trapped waves using depth-dependent structure models referred to the 2-D averaged structure, indicating the possible depth-dependent structures. The 2-D modelling is useful as an initial guide to the 3-D imaging of a fault zone.

Groundwater level and flow rate at 44 wells are continuously observed by the Geological Survey of Japan and the Shizuoka and Gifu Prefectural Governments for monitoring seismic and volcanic activities. The 2003 Tokachi-oki earthquake (M8.0) occurred off the south coast of Hokkaido Island, Japan on September 26, 2003. The epicentral distance to the nearest observation well is about 250 km and that to the farthest is about 1200 km. At the 22 wells, we detected changes in groundwater level or flow rate in relation to the earthquake. Most of the changes are coseismic step-like changes and/or short-period oscillations. In the nearest two observation wells, long-period oscillations with the periods of 39 and 53 minutes were also observed for several days after the earthquake, which is likely due to tsunami. In comparison between distributions of changes in groundwater level and theoretical coseismic strain by the fault model, it is clear that step-like increases were found in the contraction area of the coseismic strain. The relationship between amounts of the observed step-like groundwater-level changes and theoretical ones, calculated by the fault model using strain sensitivities of groundwater level indicates that the groundwater levels in the several wells responded to the coseismic strain.

We delineate shallow structures of the Mozumi–Sukenobu fault, central Japan, using fault zone waves generated by near-surface explosions and detected by a seismometer array. Two explosive sources, S1 and S2, were placed at a distance of about 2 km from the array, and the other two, S3 and S4, were at a distance of about 4 km. Fault zone head waves and fault zone trapped waves following direct P wave arrivals were clearly identified in the seismograms recorded by a linear seismometer array deployed across the fault in a research tunnel at a depth of 300 m. Synthetic waveforms generated by a 3-D finite-difference (3-D FD) method were compared with observed fault zone waves up to 25 Hz. The best fitting model indicates a 200-m-wide low-velocity zone extending at least to shot site S1 located 2 km east of the seismic array with a 20% decrease in the P wave velocity relative to the wall rock. The width of the low-velocity zone is consistent with the fault zone defined by direct geological observation in the research tunnel. However, the low-velocity zone should disappear just to the east of the site S1 to explain the observed fault zone waves for shot S3 and S4 located 4 km east of the seismometer array. Yet the observation and the simulation show notable trapped wave excitation even though shots S3 and S4 are outside the fault zone. These results indicate that (1) the effective waveguide for seismic waves along the fault does not exist east of source site S1 although the surface traces of the fault are observed in this region, and (2) considerable trapped waves can be excited by sources well outside the fault zone. These results highlight the along-strike variability in fault zone structure.

A prototype of a geoscience information system was designed for the treatment of seismological data obtained by an integrated system of surface and borehole arrays at the vicinity of the Nojima fault, responsible to the Kobe earthquake in 1995 (Hyogo-ken Nanbu earthquake M7.2). The information system includes a database (DB) as the core of the system, provided by the graphic user interface (GUI) and allow visualization and treatment of spatial objects and seismic waveforms. Information about earthquake hypocenters and measuring techniques are stored in the relational database, which allows quick selection of the data respond to certain criteria. It is demonstrated that the SEISMO database system development reduces the time required for time-consuming phases of selection, treatment and visualization of the data in a scientific process.

A seismic reflection survey using both explosives and vibrators was conducted in June 2001 around the Nagamachi-Rifu fault, northeastern Japan. We carried out observations of four small aperture seismic arrays in the area to reveal detailed structures of the fault. Array analysis was applied to waveform data from 15 explosives to obtain P-wave scatterer distributions in the area. The obtained P-wave scatterer distribution correlates in space with microearthquake activities and heterogeneous structures such as S-wave reflectors, a structure of caldera, and Mohorovicic discontinuity. We could also image that a sub-horizontal layer with a length of about 10 km exists in the deep extension of the Nagamachi-Rifu fault beneath the seismogenic layer.

Detailed structures of fault low velocity zones (LVZ) have been studied by analyzing fault-zone trapped waves at various active faults. These studies demonstrate that widths of the fault LVZ are ranging from an order of 10 m to a few hundred meters. In order to evaluate the effect of fault LVZ on the earthquake rupture process, a model of the LVZ related to the plastic deformation around an edge of propagating earthquake rupture is proposed. In this study, earthquake rupture processes are regarded as Mode III crack propagation. The LVZ is identical to the fault damaged zone which is related to plastic deformation in the vicinity of the crack tip. The Mode III crack analysis gives a simple relationship between the width of the low velocity zone, the breakdown stress drop at the crack tip, and characteristic slip distance d0 in friction laws. The parameters applicable to the Nojima fault producing the 1995 Hyogo-ken Nanbu earthquake show that the breakdown stress drop is 10 times larger than the static stress drop and d0 is about 10 cm. These values are consistent with the value obtained by the other study for the Nojima fault so the present model is applicable for considering earthquake rupture within the damaged zone.

SUMMARY In order to estimate the detailed orientation and density of cracks around an active fault area, we investigate the spatial distribution of shear wave anisotropy in the crust of the southern Hyogo region with high-quality waveform data recorded by a borehole network operated by the Geological Survey of Japan. We introduce the vertically aligned crack model of Hudson (1980, 1981) to explain its anisotropic structure. At stations more than 10 km away from the earthquake fault zone (i.e. Ikeda and Inagawa), crack orientation is nearly consistent with the direction of the regional maximum stress in this area (WNW–ESE). Cracks are distributed at depths below 5 km with a nearly constant crack density, e, of 0.015 at Ikeda, and below 0 km with e=0.009 at Inagawa. These upper-limit depths of crack distribution are roughly comparable with those of microearthquakes there. In contrast, cracks are aligned nearly parallel to the strike of the earthquake fault (NE–SW) at stations at both ends of the fault system (i.e. Takarazuka and Ikuha), and their intensities of anisotropy are three or four times larger than those at Ikeda and Inagawa. At Takarazuka in particular, cracks are distributed at a depth shallower than 4.5 km with a large crack density, e, of 0.06. This result suggests that major stress release by the faulting of the main shock dominates the present stress field at both ends of the fault system. In the middle of the fault system (i.e. Hirabayashi), cracks are aligned parallel to its regional maximum stress and the intensity of anisotropy is twice as large as those at Ikeda and Inagawa. This result contrasts with that of Tadokoro et al. (1999), who determined the crack orientation at a station close to Hirabayashi to be parallel to the fault direction for events in a period earlier than this study by about a year, suggesting that the stress field in this area may have changed over the timescale of a year. Around Inagawa, the distribution of anisotropy seems to have strong azimuthal variations. This result indicates small-scale (of the order of several kilometres) variations of anisotropic structure in the upper crust, probably due to strong heterogeneities in the fluid and/or constitutive materials.

Seismic coda waves have complex waveforms and are considered to be caused by 3-dimensional structural heterogeneities. We carried out 3-component small-aperture array observations of seismic coda waves from natural earthquakes in Tsukuba, central Japan, to decompose complex coda waveforms and to clarify the nature of crustal heterogeneity. Epicentral distances in the present study are less than 200 km. The observational results are summarized as follows: 1) The remarkable feature of the coda from natural earthquakes is the difference between the P-coda and the S-coda especially in the vertical component. The P-coda consists mainly of coherent P-wave phases which have almost the same slowness vector as the initial P-wave motion, whereas the S-coda is decomposed into incoherent phases with random propagation directions. 2) Coherency of waveforms among the array sensors of the vertical component rapidly becomes low from the high value of coherency of the direct S-wave. 3) In the two horizontal components, radial and transverse ones, both the P- and S-coda are decomposed into incoherent random phases. 4) Coherency in the early S-coda of the horizontal components gradually becomes low compared to the vertical component. These features indicate that S to P converted waves at horizontal boundaries of the structure are dominant in the P-coda of the vertical component, while the P-coda of the horizontal components and the S-coda consist of scattered waves from the lateral heterogeneity. This interpretation leads to a model of crustal heterogeneity in which the degree of the lateral heterogeneity is smaller than the degree of vertical heterogeneity of layered structures.

[1] After the occurrence of the 2011 Mw9.0 off the Pacific coast of Tohoku earthquake, an unusual shallow normal-faulting earthquake sequence occurred near the Pacific coast at the Ibaraki-Fukushima prefectural border. We have investigated why normal-faulting earthquakes were activated in northeast (NE) Japan, which is otherwise characterized by E–W compression. We computed the stress changes associated with the mainshock on the basis of a finite fault slip model, which showed that the amount of additional E–W tensional stresses in the study area was up to 1 MPa, which might be too small to generate normal-faulting earthquakes in the pre-shock compressional stress regime. We thus determined focal mechanisms of microearthquakes that occurred in the area before the mainshock, which indicated that the pre-shock stress field in the area showed a normal-faulting stress regime in contrast to the overall reverse-faulting regime in NE Japan. We concluded that the 2011 Tohoku earthquake triggered the normal-faulting earthquake sequence in a limited area in combination with a locally formed pre-shock normal-faulting stress regime. We also explored possible mechanisms for localization of a normal-faulting stress field at the Ibaraki-Fukushima prefectural border.

[1] This study evaluates a method that enhances detection of non-volcanic low frequency tremor (NVT) using VA-net, a 3-level vertical seismic array network that was recently constructed in southwest Japan. The method employs semblance analysis for continuous vertical array records and calculates semblance coefficients as a function of time and apparent vertical velocity. The sign of the best apparent velocity provides information about propagation direction, either upward or downward, allowing easy discrimination between seismic signals of natural origin and unwanted noise such as man-made vibrations. We applied the method to a NVT episode that occurred beneath the Kii Peninsula during November 2008, to demonstrate that NVT signals can be detected using only the 3-level array. Furthermore, we demonstrate that the method enables us to detect minor NVT activity which is not detected by a conventional envelope cross-correlation method, resulting in a dramatic improvement in detection capability.

[1] Focal mechanisms have been determined from P wave polarity data as well as body wave amplitudes for 154 microearthquakes that occurred around the Atotsugawa fault in central Japan between 2002 and 2004. While we found many microearthquakes with a pure strike-slip mechanism that is similar to the faulting style of the Atotsugawa fault, a considerable number of microearthquakes with reverse faulting components are also occurring. Most of the P axes are horizontal and oriented in the WNW-ESE direction, which conforms to the general tectonic trend in this area. In contrast, the T axes have a wide range of plunge, suggesting that reverse-faulting-type earthquakes as well as strike-slip ones are occurring. The most conspicuous feature in the focal mechanism distribution is the depth dependence, where shallow earthquakes are primarily reverse faulting and strike-slip earthquakes become predominant with depth. A stress tensor inversion reveals that the shallower part is characterized by a mixture of reverse and strike-slip faulting regimes and that a pure strike-slip faulting regime appears only around the bottom of the seismogenic zone. Together with other geophysical observational evidence for the fault, we suggest that the existence of a localized aseismic deformation below the Atotsugawa fault is the simplest scenario that can explain the observed stress fields. This scenario provides a stress accumulation mechanism of disastrous shallow inland earthquakes, in which the localized aseismic deformation accumulates stress onto the fault plane in the seismogenic zone during an earthquake cycle and the main shock would occur when the failure stress is reached on the fault.

A slickenside was discovered from a boring core sample 1, 880m deep in the Matsushiro earthquake swarm region in Nagano Prefecture, Japan. The slickenside has been considered to be created by earthquake faulting during the earthquake swarm period, because it accompanied fresh rock powder produced by earthquake faulting. The power spectral density of the topography of the slickenside was measured with a stylus profilometer over the wavelength range from 10-4 to 10-2m. Each profile of the surface has a "red noise" power spectrum over the wavelength range studied, with the power falling off on average between 2 and 3 orders of magnitude per decade decrease in wavelength. Meanwhile, the topography of the slickenside is found to show anisotropy in the spectral amplitude; the amplitude perpendicular to the slip direction is two to four times as large as that parallel to the slip direction. A generation mechanism for such topographic anisotropy is numerically modeled from a viewpoint of the fracture mechanics. It is supposed that the crack tip stress distribution in the plane parallel to the slip direction is represented by a model of a plane strain shear crack (Mode II crack) and that the stress distribution in the plane perpendicular to the slip direction is represented by an antiplane strain shear crack (Mode III crack). The stress field around the crack tip in the Mode III case shows a broader distribution with respect to the azimuth than that in the Mode II case. This difference of a stress field at the crack tip between the Mode II and the Mode III is a possible cause of the topographic anisotropy.

[1] We investigate how a static stress change, caused by a nearby earthquake, perturbs sliding processes on a fault and the time to the next earthquake. For this purpose, we model numerically a reverse fault in a two-dimensional, semi-infinite and elastic medium, partly stable and partly unstable under a rate- and state-dependent friction law. When no stress change intervenes, steady sliding spreads out of the deep zone of frictional stability and keeps expanding updip during the interseismic period. Aseismic slip episodes (ASEs) develop spontaneously within the region of steady sliding and repeat quasi-periodically. When steady sliding has spread beyond a certain critical length (the nucleation length) above the zone of frictional stability, one of the ASEs undergoes rapid buildup and avalanches into high-speed seismic sliding. When a step-like stress change intervenes, ASEs of larger amplitudes arise and evolve similarly to their spontaneous counterparts until one of them avalanches into an earthquake. If the stress change arrives not too early nor too late in the seismic cycle, the basic picture is as follows: delaying the stress step, be it positive or negative, delays the timing of the ASE sequence and thereby postpones the next earthquake. When the stress step is delayed to more than a certain extent, however, an earlier ASE gets to evolve into an earthquake, advancing its time discontinuously. Only in the final stage of a seismic cycle does a positive stress step systematically advance the next earthquake and a negative stress step systematically delay it.

Characteristic features of local breakdown near a propagating crack-tip during slip failure nucleation and its transition process to unstable rupture under mode II conditions are experimentally studied using a rock sample with a simulated fault, the size of which is large compared with the size of the breakdown zone. In addition, distinctive features in the nucleation zone are compared with those in the zone of steady, dynamic rupture propagation. It is found that the local breakdown stress drop, the local dynamic stress drop, the local stress increase immediately before the slip-weakening process, the dimensionless parameter S defined by Das and Aki, and the shear fracture energy are functions of crack length in the transition zone from nucleation to unstable rupture; i.e., they all increase with crack growth in the nucleation zone. This is contrasted with the observations in the zone of steady dynamic rupture propagation, where these physical parameters do not depend on crack length. An increase in the magnitudes of these parameters with crack growth in the nucleation zone results in increasing resistance to crack extension with crack growth in the nucleation zone. The reason for the increase in crack growth resistance in this zone is that a slip failure nucleus is formed at a point (or zone) where the local strength and the shear fracture energy are at a minimum. This basic physical information can provide the key to earthquake prediction, since the nucleation itself is a precursory phenomenon. The ratio of the size of the breakdown zone to the crack length decreases abruptly beyond the critical crack length. Rupture velocity v accelerates with crack growth, and reaches the shear wave velocity or super-shear, and v is found to be expressed empirically as a function of crack length in the nucleation zone. A positive correlation is found between S and v in the nucleation zone (v increases with increasing S). However, no such correlation was observed in the zone of steady, dynamic rupture propagation, where v seems to have reached the shear wave velocity of super-shear for S = 0.5. This is consistent with Das and Aki's numerical result. The interrelationships between shear stress, slip displacement, slip velocity and acceleration near the crack-tip during the breakdown in the nucleation zone are revealed, and they are compared with those in the zone of dynamic rupture propagation. The present experiments suggest that the relationship between shear stress and slip displacement is more fundamental during the breakdown process than the relationship between shear stress and slip velocity. A model is presented to describe the breakdown process in the nucleation zone

[1] This is the second paper in a two-part series on a newly developed imaging approach for small-scale heterogeneities (

[1] We analyzed shear-wave splitting of small earthquakes around the Atotsugawa fault, central Japan, to infer a spatial variation in the direction of maximum horizontal compression (Shmax). While the angle, ϕ, between the Shmax direction and fault strike ranges approximately from 55° to 80° at the stations 2 to 8 km from the fault, it approaches 45° at the stations within 1 km from the fault. The present result indicates the local stress accumulation due to the localized flow or slip along the deep extension of the fault. The Finite Element Modeling of the stress field around the fault suggests that the present existing stress field can be explained by the superposition of the stress field associated with regional deformation and the stable slip along the deep extension of the fault.

SUMMARY The feasibility of resolving a subsurface fracture system using P-wave and cross-polarized shear-wave vertical seismic profiling (VSP) was examined. In situ distributions of fractures and microcracks in a test borehole are characterized by an ultrasonic borehole televiewer (BHTV) and by measurements of core sample velocities. The BHTV images show that the fracture frequencies in the shallower part of the borehole are much higher than those in the deeper part. Two sets of dominant parallel fractures are observed by BHTV in the fracture-rich depth interval of the shallower part; the orientations of fracture strikes in the two sets are almost the same directing to NW, whereas the average dip angles of the fractures in the two sets are 73" and 46", respectively. The fracture orientations in the fracturepoor depth interval of the deeper part are found to be random. The degree of anisotropy of the core sample velocity is so small that the present VSP measurement cannot resolve the expected anisotropy from the core sample measurement. P- and two shear-wave velocities (Vp, V,, and V,) are determined from the downgoing phases of three-component VSP with a zero-offset P-wave source and with zero-offset shear-wave sources polarized in two orthogonal directions, respectively. The VSP results are as follows. (1) The shear-wave polarization anisotropy is not detected within the accuracy of 4 per cent. (2) V, and V, (average of V,, and V,) for the fracture-rich interval are 94 per cent and 86 per cent, respectively, of those for the fracture-poor interval. The VSP results are consistent with expected velocity changes due to an isotropic distribution of cracks. However, a biplanar crack model cannot satisfy the VSP results, whereas the BHTV images show two such sets of parallel fractures. It is suggested that this discrepancy is due to the lack of depth resolution in the velocity determination in the present VSP experiment.

Constitutive relations between physical parameters in the cohesive zone during stick-slip shear failure are experimentally investigated. Stick-slip was generated along a 40 cm long precut fault in Tsukuba granite samples using a servocontrolled biaxial loading apparatus. Dynamic behavior during local breakdown processes near a tip of the slipping zone is revealed; the slip velocity and acceleration are given as a function of the slip displacement and the cohesive (or breakdown) shear stress as a function of the slip velocity. A cycle of the breakdown and restrengthening process of stick-slip is composed of five phases characterized in terms of the cohesive strength and the slip velocity. The cohesive strength can degrade regardless of the slip velocity during slip instabilities. The maximum slip acceleration u max and the maximum slip velocity u max are obtained experimentally as: u max = 2 u c u max 2 and u max = Δτ b G v where uc is the critical displacement, Δτb the breakdown stress drop, G the rigidity and v the rupture velocity. These relations are consistent with Ida's theoretical estimation based on the cohesive zone model. The above formula gives good estimates for the maximum slip acceleration of actual earthquakes. The cutoff frequency ƒ max of the power spectral density of the slip acceleration increases with increasing normal stress; in particular, ƒ max is found to be directly proportional to the normal stress σn within the normal stress range less than 17 MPa as: ƒ max (kHz) = 4.0σ n (MPa) σn ƒ max increase with an increase in u max or u max . All these results lead to the conclusion that u max , u max and ƒ max increase with increasing normal stress. This is consistent with a previous observation that τb increases with increasing normal stress. The above empirical linear relation between ƒ max and σn can be explained by a linear dependence of Δτb on σn. The size-scale dependence of physical parameters is discussed, and such size-scale dependent parameters as u max and apparent fracture energy are scaled to the breakdown zone size regarded as a characteristic length, to extend the results obtained in the laboratory to an earthquake failure in the earth.

The three-dimensional viscoelastic crustal structure beneath Japanese islands was modeled to simulate their crustal deformation by using a finite-element method and applying boundary conditions of the east-west horizontal compression. The result shows that there are relatively narrow zones of high strain rate at shallow depths, whose pattern is similar to that of the Niigata-Kobe tectonic zone revealed by GPS. High strain rates are not necessarily concentrated in regions where the elastic layer is relatively thin, but rather where its thickness changes abruptly.

Fault zone waves have the potential to be a powerful tool to reveal the fine structure of a fault zone down to the seismogenic depth. Seismic fault zone waves include head waves, trapped waves and direct body waves propagating in the fault zone. 3-D numerical simulation is necessary to interpret the waveforms in the presence of low-velocity zones with relatively complex fault structure. We computed finite difference (FD) synthetic seismograms to fit the seismograms of explosions, which contain frequencies up to 25 Hz, recorded by a linear seismometer array across the Mozumi-Sukenobu fault, central Japan. We find fault zone head waves, direct P waves propagating within the low-velocity zone and wave trains following the direct P waves associated with the fault for both observed and synthetic waveforms. Thus, modelling of fault zone waves is expected to determine details of complex fault zone structure.

After the occurrence of the 20 March 2005 West Off Fukuoka Prefecture Earthquake (Mj7.0), off-fault aftershocks occurred in and around Hakata bay adjacent to the main fault of the earthquake. The locations of the activity seem to coincide with the Ishido-Uminonakamichi fault (I-U fault), which is a structural boundary and not considered as an active fault. In order to elucidate relations between these off-fault aftershocks and the I-U fault, we determine the double-difference earthquake locations and focal mechanism solutions. Results of the present study show that the off-fault aftershocks are aligned in a sense of left-lateral faulting striking 120° from the north and in that of their conjugate right-lateral fault structures striking 30? from the north. Comparing the trends of these aligned structures with the orientation of the I-U fault, we conclude that earthquakes in and around the Hakata bay occurred along these aligned structures, not along a fault plane inferred from the surface trace of the I-U fault. The analysis of Coulomb failure stress due to the mainshock also supports this conclusion. Stress tensor inversion shows that the stress field in and around the Hakata bay is characterized by a strike-slip faulting. The angles between the fault trends and the maximum principal stress suggest that the aligned structures found in this study are favorably oriented, while the fault plane inferred from the surface trace of the I-U fault is unfavorably oriented.

Stress fields in the source region before and after the 2007Mw6.6 Niigataken Chuetsu-oki earthquake were investigated using aftershock focal mechanism solutions. We have attempted to determine aftershock focal mechanisms usingP-wave polarity data as well as body wave amplitudes because this earthquake occurred offshore, where observation station coverage is poor. This approach enabled us to obtain 76 well-determined aftershock focal mechanisms. Although the stress field in the studied area is known to be generally characterized by a reverse-faulting regime, the application of a stress tensor inversion method to the aftershock focal mechanisms revealed that the stress field spatially varied on a scale smaller than the fault dimension of the mainshock, with a mixture of strike-slip and reverse faulting regimes. The post-mainshock stress field estimated by the stress tensor inversion was compared with the theoretical stress field computed using an observed non-uniform slip distribution of the mainshock and variously assumed pre-shock stress fields. The results of this comparison suggest that the pre-mainshock principal stress in the WNW-ESE direction was dominant and that the magnitudes of the other two principal stresses were similar.

The Geological Survey of Japan, AIST has been monitoring groundwater using the observation network of about 40 stations in and around the Kinki and Tokai districts for earthquake prediction research. The Noto Hanto Earthquake in 2007 (MJMA6.9) occurred in the northwestern part of the Noto Hanto Peninsula, Japan on March 25, 2007. The epicentral distance to the nearest and furthest observation well is about 100 km and 400 km, respectively. Changes in groundwater level related to the Noto Hanto Earthquake were detected at many stations. Most of these were changes in the long-term trend and/or short-term oscillation; there were a few steplike changes. The coseismic static volumetric strain changes calculated from the fault model ranged between 10−9and 10−10at most of the observation stations. These changes were therefore considered to be groundwater changes caused by the ground shaking because groundwater level sensitivity to crustal volumetric strain is a few centimeters to 10−8strain at the most. We compared the coseismic groundwater level changes for the Noto Hanto Earthquake with those for five recent large earthquakes and obtained rough characteristics of the effect of the ground shaking on the groundwater level at each of the observation stations.