Present day crustal displacement rates are accurately observed by global navigation satellite system (GNSS). In estimating continuous displacement fields from spatially discrete GNSS data, the method of basis function expansion is a powerful tool. As basis functions, the boxcar function is the easiest to be implemented, while the cubic B-spline function has good nature in estimating continuous and smooth fields. In this study, we quantitatively compare the performance of the boxcars with that of the cubic B-splines, changing the basis-function interval L in estimating velocity fields from observed GNSS data in Japan. The result shows that the performance of the cubic B-splines with L＝50, 60, and 80 km is nearly the same or better than that of the boxcars with L＝20, 25, and 30 km, respectively. In other words, to achieve similar performance, at least about 2.5²＝6.25 times as many boxcars as cubic B-splines are needed. In the method of basis function expansion, the computation of inverse matrices is the most time-consuming process. Because its computational cost is usually proportional to the cube of the matrix size (the number of basis functions), computing inverse matrices for the boxcars takes about 250 times as much time as that for the cubic B-splines, although the difference of actual computational costs in this study was about 100 times or more. In addition, strain rates can be easily obtained by analytically differentiating the velocity field when we use the cubic B-splines. On the other hand, a main disadvantage in using the cubic B-splines is complicated computation to obtain the explicit expression of the smoothness constraint for the inversion analysis. To mitigate this problem, we show the computation results in Appendix.

Obtaining marine geophysical data, particularly from seismic surveys and observations, is difficult when using ocean-bottom seismographs (OBSs) in coastal areas without causing a conflict with fishery activities. To minimize conflict, we have designed and operated anchor-recovery-type OBSs in coastal areas shallower than 200 m. The OBSs shallower than 650 m were required to recover with their anchors to minimize such conflict during our seismic survey cruise off Yamagata Prefecture by the research vessel (R/V) Kairei in August 2019. Therefore, we improved the OBS anchor recovery system, which can operate from the continental shelf to the upper continental slope at depths shallower than 650 m. Two operation trials of the improved anchor-recovery-type OBS were conducted using a multi-purpose experiment tank at the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) Yokosuka Headquarters, to evaluate its performance before the cruise. In addition, we prepared a sweep-line with grapnel anchors in case unexpected problems occurred during the OBS recovery operation. The seismic survey cruise used a tuned air-gun array system and 39 OBSs (including eight anchor-recovery-type OBSs). The OBS recovery operation was initiated following the air-gun shooting. To prevent the propellers of R/V Kairei from tangling in the recovery rope, a two-ship operation was planned for the recovery of the anchor-recovery-type OBSs. The two-ship operation was performed as follows: First, the R/V Kairei sent an acoustic release command to the OBS. Second, the surfaced OBS was recovered by the ocean tugboat, motor vessel (M/V) Hirokai. Finally, the anchor recovery rope was wound using the winch on M/V Hirokai to recover the anchor. We successfully recovered seven OBSs (Sites 1-7) and their anchors; however, one OBS (Site 8) did not respond to acoustic communication, including the release command. We attempted to recover this unreleased OBS using the prepared sweep-line with grapnel anchors. The sweep-line was deployed from the stern of the R/V Kairei and towed around the OBS position with a radius of approximately 100 m. The OBS and its anchor were caught by the sweep-line and successfully recovered. Our improved anchor-recovery-type OBS can reduce conflict with fishery activities and enable to conduct marine seismic surveys in coastal areas from the continental shelf to the upper continental slope.

A magnitude 7.1 earthquake occurred west off Satsuma Peninsula, Japan in 2015, which was the largest earthquake ever recorded in the northern Okinawa Trough (OT). The northern OT is assumed to be in a beginning stage of a back-arc rifting that drives crustal extension, and the occurrence of the 2015 mainshock-aftershock sequence might be associated with such tectonics. In order to understand the rifting process that controls the seismicity, we precisely determined hypocenters and focal mechanisms of the aftershocks listed in the Japan Meteorological Agency catalog using both offshore and onshore seismic data. Initial hypocenters were determined using manual picks of P and S phases and a 1-D velocity structure. We then applied station correction terms to the P- and S-arrival times so as to consider low-velocity sedimentary layers beneath each station, especially beneath the offshore stations. Finally, the relative hypocenters were refined by the double-difference location method. The average root-mean-square residuals of the double differences decreased from 489 ms to 39 ms. The relocated hypocenter distribution clearly shows three linear alignments: one in the N95°E direction, consisting of left-lateral strike-slip fault events, and two in the N15-20°E direction, consisting of normal fault events. The tension axes for almost all the solutions lie in a common direction (NW-SE). The well-determined aftershocks mainly located in the upper crust with focal depths of 0-15 km. The results from our ocean-bottom seismic observation indicate that the two alignments consisting of normal fault events might be caused by crustal extension related to the back-arc rifting in the northern OT because the direction of the hypocenter distribution is consistent with the strike of the existing normal faults imaged by the multichannel seismic reflection survey. The left-lateral strike-slip fault alignment could be explained by along-axis variations in the back-arc spreading rate and presence of a transcurrent fault.

Fluid injection induced earthquakes related to resource mining and geothermal power generation are concerns among many countries. It is difficult to directly observe the spatiotemporal changes in the fracture development and the stress state due to the water injection several kilometers underground, while we can estimate some extent from borehole measurements and seismic activity. We came up with an experiment using the photoelastic effect, which allows us to directly observe the stress on complex structures. Photoelasticity is a phenomenon that enables visualization of the amount of differential stress in a transparent elastic material. This paper introduces a photoelastic experimental method to visually observe the crack propagation and stress change in the crust relating to the water injection. The utilization of a transparent and soft agar gel makes it possible to observe the spatiotemporal stress change during the water injection. A thin acrylic cuboid container filled with agar gel with a 1% concentration is prepared as a simulated crust. The gel sample is placed between the experimental apparatus consisting of polarizers, quarter-wave plates, and a white light source. Water is injected from a syringe into a plastic pipe that resembles a borehole placed in the center of the sample surface. The isochromatic fringes generated by the stress change caused by the water injection are captured on high-speed video. We prepare two models; one is a model with no boundary, and the other has a medium boundary that mimics a weak surface such as a fault. During the water injection, we clearly observe the intense stress concentration on the crack tip and changes in the stress distribution associated with crack propagation. The stress change is also observed in the point far from the crack. The crack growth rate shows quadruple differences between the two types of samples. Visualized stress state in the experiment gives us intuitive insights into the crustal condition during the water injection. The experimental method proposed in this paper will help simulate a variety of crustal conditions and improve the understanding of fluid induced earthquakes.

The Tono Research Institute of Earthquake Science, located in Mizunami-City, Gifu Prefecture, has developed a comprehensive set of borehole crustal activity observation devices and is conducting deep underground observations on earthquakes and crustal movements. A geological fault, which is called the NNW fault, runs in the vicinity of the observation points. This paper reports on pumping experiments that were carried out in February 2001 and April 2002, in the DH2 borehole (depth: 500 m) near the NNW fault. The purpose was to investigate strain behavior near the fault. Changes caused by the pumping experiments were observed by water-level gauges and strainmeters at borehole observation points TGR350 (depth: 350 m) and TGR165 (depth: 165 m), and by a three-component extensometer with a length of 30 m installed in a horizontal vault observation point, TGRvault. The horizontal distance from the DH2 borehole to the TGR350 and TGR165 boreholes is about 271 m. The distance from DH2 to the extensometer installed in the TGRvault is about 193 m. When pumping was performed in DH2, the water level in TGR350 dropped; when pumping ceased, the water level rose. As the water level went up and down, changes were observed in each strainmeter. The maximum shear strain observed in the strainmeters of TGR350 was in the NNW direction from the viewpoint of geophysics. It showed a right lateral shift variation when the water level was falling and a left lateral shift variation when the water level was rising. The NNW fault is a right lateral fault. The strainmeters of TGR165 observed a different strain variation from that of the TGR350 strainmeters. Unlike the TGR165 and TGR350 strainmeters, the TGRvault extensometer in the horizontal vault showed similar fluctuations to those of the water level in all three of its components. The NNW fault is considered to extend to a depth of about 1 km, but The TGR350 strainmeters are located in the lower sparsely fractured domain (LSFD) at the uniform base of the Toki granite at a depth of 350 m, where fractures are few. The TGR165 strainmeter is installed in the Toki granite, which is the base of the upper higher fracture domain (UHFD), at a depth of 165 m, where fractures are more significant. The TGRvault extensometers in the horizontal vault are installed in the sedimentary layer close to the surface. The results of the pumping experiments can be attributed in part to differences in the response of the medium to groundwater fluctuations, due to differences in the number and distribution characteristics of the fractures in the medium where the observation instruments are installed, resulting in the differences in strain behavior.

The most authoritative catalog of Japanese historical damaging earthquakes lists a M~6 earthquake of Jan. 11, 1859 near Iwatsuki, Saitama Prefecture, in the Kanto district of central Japan. The earthquake is said to have damaged Iwatsuki Castle considerably, and the seismic intensity there is evaluated as 5-upper on the Japan Meteorological Agency (JMA) scale. However, the only historical source of the damage is a Tokugawa shogunate official document of Mar. 9, 1859 describing that it lent the lord of the Iwatsuki Castle money for repairing the castle damage caused by a “previous earthquake.” Although the date of the “previous earthquake” is not written explicitly, the editor of the catalog considered the earthquake to be the Jan. 11 event felt strongly in a wide area of the Kanto district. Since there is no other record of damage caused by this earthquake despite that many documents were written in those days, we examined whether the “previous earthquake” is truly the 1859 Jan. 11 event or not. We discovered the historical document entitled “Sho-ammon-utsushi”, a collection of transcripts of official documents of the Iwatsuki clan, which has been unknown in Japanese historical seismology. It contains a transcript of the letter asking the Tokugawa shogunate for lending money of reconstruction following the 1855 earthquake disaster. The letter was written on Dec. 4, 1855, 23 days after the 1855 Nov. 11 Edo (present Tokyo) earthquake (M 7.0-7.1) that killed about 10,000 people mainly in Edo, about 30 km south of Iwatsuki. In the letter, severe damage in and around the castle town of Iwatsuki caused by the Edo earthquake is described in detail, which is new valuable information regarding the Edo earthquake. On the other hand, no record related to the 1859 earthquake is found in the “Sho-ammon-utsushi.” Therefore, we conclude that the “previous earthquake” is the 1855 Edo earthquake and that the 1859 earthquake was not a damaging one. By comparing the intensity distribution of ground shakings of the 1859 event with the seismic intensity distributions of various earthquakes in the Kanto district during 1919-2021, we infer that the 1859 earthquake was an M 4-class event in the depth range 40-80 km beneath the area around 36.1˚N and 139.8˚E. We evaluate the seismic intensity at Iwatsuki due to the 1855 Edo earthquake as 5-upper or 6-lower on the JMA scale, which has been unknown so far.

The Tohoku-Oki earthquake (M_w 9.0) occurred off the Pacific coast of Japan, on March 11, 2011. The seismic activity in Japan changed after the Tohoku-Oki earthquake. Especially around the border between Ibaraki and Fukushima prefectures, so many earthquakes within the crust have occurred. The largest earthquake in this region is the Fukushima Hamadori earthquake (M_w 6.6) on April 11, 2011, which caused severe damage. In this region, the seismic activity had been very low before the Tohoku-Oki earthquake, but the seismic activity became significantly active after the Tohoku-Oki earthquake. There are few previous studies about this region that investigated the crustal structure in detail because the seismic activity was low before the Tohoku-Oki earthquake. Thus, to investigate the crustal structure in this region is very important from the perspective of elucidating the mechanism of large inland earthquakes such as the Fukushima Hamadori earthquake. Earthquake Research Institute, The University of Tokyo, and other organizations deployed a temporary dense seismic array in this region after the Fukushima Hamadori earthquake. We examined the waveforms obtained from the seismic array and found two reflected S waves from the deep part of crust. We estimated the locations of the reflection points in the deep part of the crust by applying Reverse VSP (Vertical Seismic Profile) method which has been used for seismic exploration. As a result, we found two S wave reflection boundaries at depths of (1) 15～23 km and (2) 26～34 km in the crust. The reflection boundary (2) was found to be the Moho discontinuity by comparing with previous studies in this region. It can be interpreted that the reflection boundary (1) indicates the upper boundary of the layer with crustal fluids in terms of the large amplitudes of the reflected S waves. We compared the locations of the reflection boundary (1) with seismicity in this region and found that the location of the reflection boundary (1) was consistent with that of the Fukushima Hamadori earthquake and the focal area at the depth of 15～20 km which was one of the characteristics of seismic activity in the region. Thus, the crustal fluids may be related to the seismic activity in the region triggered after the Tohoku-Oki earthquake.

There is a very thick (>10 km) sedimentary basin in the Ishikari lowland, the central Hokkaido. This thick sedimentary basin affects strong ground motions observed from earthquakes, e.g., the 2018 Hokkaido Eastern Iburi Earthquake occurred in the southeastern part of the Ishikari lowland. Such a thick sedimentary basin structure is difficult to explore by using microtremor array or seismic reflection. This study investigated the sedimentary basin structures at 16 strong motion stations in the southeastern part of the Ishikari lowland by using the borehole-to-surface spectral ratio and the receiver function method based on the observed strong-motion records. First, at the KiK-net stations, the surface-to-borehole spectral ratio was obtained from observed strong-motion records. The S-wave velocity structure models between the surface and the borehole seismograph were estimated by fitting theoretical spectral ratios to the observed one. Next, the velocity structure models between the surface and the seismic bedrock were derived from the receiver function calculated from the observed records. We used the downhill simplex method to obtain the velocity structure model from the receiver function. The inverse analysis with three different initial models resulted in variations in the estimated velocity structure models. Synthetic velocity waveforms of the aftershock of the 2018 Eastern Iburi earthquake were calculated based on the estimated velocity structure models. The synthetic velocities (0.1-1 Hz) agree with the S waves of the observed records. The synthetic waveforms obtained from three models at each site were very similar. We compare the resulted velocity models with the microtremor array surveys carried out by previous studies. The fundamental mode phase velocities calculated from the velocity structure models obtained from the receiver functions correspond to the observed phase velocity from the microtremor array survey in higher (>0.4-0.8 Hz) frequencies. However, in lower frequencies (<0.4 Hz), the observed phase velocities at HKD184, IKRH03, and IBUH03 agree with the higher-mode phase velocities than the fundamental ones. This agreement suggests that the higher-mode Rayleigh waves dominate the observed phase velocities in the lower frequencies. The velocity structure models estimated from the receiver function analysis indicated that the low Vs (<1.1 km/s) layer deposited thickly in the western part of the Ishikari lowland and the middle Vs (1.1<Vs<2.7 km/s) layers were thick in the eastern part of the lowland. The depth of the estimated seismic bedrock was 12 km at HDKH04, located in the southeastern part of the lowland, which was the deepest between the study sites.

Near the Tono Research Institute of Earthquake Science (TRIES), Mizunami City, Gifu, two 500 m- deep shafts with diameters of 4.5 m and 6.5 m were excavated by the Japan Atomic Energy Agency (JAEA). The 6.5 m shaft is crossed by an NNW-trending geological fault (NNW fault), which has a right-lateral strike-slip component and a dip-slip component. Since before the excavation TRIES has been performing observations by a comprehensive set of borehole crustal activity observation devices in deep boreholes and extensometers in a vault near the shafts to observe variations in, for example, strain, tilt, and groundwater level continuously. During the shaft excavation, groundwater inflow to the shafts was encountered. The water was twice pumped out. During the pumping and non-pumping periods, TRIES’s observation data such as the groundwater level, the strain, and the tilt showed anomalous variations. The analysis of them revealed relationships among the groundwater level, the fault, the strain and the tilt. Strain analysis of TGR350 borehole station showed that the direction of the maximum shear strain was parallel to strike of the NNW fault. In addition, when the water in the shafts was being pumped out, the groundwater level at TGR350 declined and the maximum shear strain showed a right-lateral strike-slip movement in the NNW direction. When pumping was paused, the groundwater level increased, and the maximum shear strain showed a left-lateral strike-slip movement in the NNW direction. On the other hand, strain analysis of TGR165 borehole station showed different behavior from that of TGR350 borehole station deeper than TGR165. Tilt analysis of TGR165 borehole station showed tilt changes nearly perpendicular to strike of the NNW fault corresponding to the pumping and the non-pumping periods. Tilt analysis of TGR350 borehole station showed tilt changes of E-W direction with much smaller amplitude than that of TGR165. Behavior of the strain and the tilt during the pumping and non-pumping periods can be attributed in part to differences in the response of the medium to groundwater fluctuations, due to differences in the number and distribution characteristics of the fractures in the medium where the observation instruments are installed. We also discussed the observation results from the viewpoint of poroelasticity.

Distributions, characteristics, and activities of offshore fault-related folds which comprise the southern part of the Eastern Boundary Fault Zone of Ishikari Lowland (EBFIL) are clarified by conducting a high-resolution multi-channel seismic reflection survey. To determine termination point of the fault zone, we placed seven survey lines perpendicular to the fault-related folds with a total line length of 107.5 km. We choose a compact water gun (15 cubic inch) as the seismic source to obtain high-resolution imagery of the Late Quaternary and investigate activity of the fault zone. Based on seismic facies, we distinguished packages of offshore sediment (unit 2, 3 and 4, in descending order), which separated by downlap surfaces and correlated to the Middle to the Late Pleistocene. They are progressively deformed by the active fault and compose series of fault-related folds (Fo1, Fo2 and Fo3). Traces of these folds are figured out from the seismic sections obtained by this study and existing surveys. Deformation related to these folds are clearly recognized in the topography of the unit 4 base. The reliefs of strata indicate that the Fo2 fold should have uplifted seafloor at a rate of 0.09-0.20 m/ky since 300 ka. The structures, spatial distributions and growth-rates of the folds suggest that Fo1 and Fo2 should compose series of anticline developing at the front of EBFIL. Besides, Fo2 fold would compose the southern termination of the fault zone. This study has successfully demonstrated that densely operated high-resolution seismic survey can figure out distribution and activity of offshore active faults in the Late Quaternary, even at the outside of shelf.