The 147th (2022 FALL) SEGJ CONFERENCE



Nov.16 　　Room No.1（Large Conference Room，1F）

Session	[Methane hydrate][CO₂ I]
Chair	Hiroaki Sato (CRIEPI)

Nov.16	10:00 - 10:20　　　 The evaluation of fault stability and integrity is an important process in CO2 geological storage. Mohr-Coulomb failure criterion has been widely used for estimating the critical pore pressure for shear slip on a fault, but often only pore pressure elevation due to CO2 injection are considered and local stress field perturbations are ignored. Therefore, we proposed a method of fault stability evaluation that takes into account the stress field perturbations by utilizing a coupled fluid flow and geomechanical simulation (TOUGH-FLAC), and applied it to open available data (In Salah field). In addition, history matching with surface displacement data (InSAR) during CO2 injection was conducted, and the usefulness of the monitoring data in modeling CO2 reservoirs was discussed.	1) Fault stability analysis for CO2 injection field employing the coupled fluid flow and geomechanical simulation technique
Nov.16		*Mitsuyuki Ozawa(1),Koji Kashihara(2),Takumi Onuma(1) 1:JGI,2:JAPEX
Nov.16	10:20 - 10:40　　　 Cosmic-ray muons have been used to estimate density inside objects and its application is extending to underground due to miniaturization of detectors. Meanwhile, elastic waves, which are often used for underground exploration, are also sensitive to density, so integrated use of them is conceivable and expected to estimate density and elastic constants, which are useful for estimating physical property such as gas saturation evaluation. An integrated analysis of cosmic-ray muons and elastic waves was considered taking the correlation between the traveltime and the density into account with density and Lamé constants used as model parameters. The results suggest Lamé constants estimation reults can be improved using density and Lamé parameters as model parameters instead of density and P-wave and S-wave slowness.	2) Numerical study on optimization method for integrated analysis using cosmic ray muons and elastic waves
Nov.16		*Masashi Kodama(1),Toshiyuki Yokota(2),Jun Matsushima(3),Mohammed Y. Ali・Fateh Bouchaala(4),Hiroyuki Tanaka(5),Tadahiro Kin(6) 1:AIST, The Univ. of Tokyo, GSFS,2:AIST,3:The Univ. of Tokyo, GSFS,4:Khalifa University,5:The Univ. of Tokyo, ERI,6:Kyushu Univ., IGSES
Nov.16	10:40 - 11:00　　　 Fractured hydrate-bearing reservoirs are extensively distributed around the world. The acoustic attenuation information shows high sensitivity to the substances with different phases (i.e., gas, water, and solid phase) and their saturation filling in the pore system, and hence plays an important role in the accurate assessment of hydrate resources. However, the existence of fractures can give rise to significant anisotropy in the attenuation properties, the complex fluid effects, and the elastic scattering in hydrate-bearing reservoirs, which remain poorly understood. In this study, we studied the acoustic attenuation responses on the hydrate filling in fracture by simulation-based rock physics modeling. We showed the anisotropic attenuation properties and their differences among the varying hydrate saturation and morphology in the fracture. The results provide new insights into the effects of hydrate saturation and morphology on acoustic attenuation in fractured media and will contribute to improving the accuracy of hydrate assessment in fractured hydrate-bearing sediments.	3) Simulation-based rock physics modeling of acoustic attenuation responses in a fracture filled with hydrates
Nov.16		*Shengbiao Liu(1),Matsushima Jun・Xiaoliang Wang・Yaoming Che(2) 1:The University of Tokyo/China University of Petroleum (East China),2:The University of Tokyo
Nov.16	11:00 - 11:20　　　 Evaluating the mechanisms of S-wave attenuation in methane hydrate bearing sediments (MHBS) considering the relationship between shear waves and friction has been rarely studied. After the newly measured data shows inconsistency with predicted data according to currently accepted friction theory, dynamic friction on the contact line of the three phases is to be considered. To model and simulate the relationship between the friction and attenuation, a rock with a crack of elliptical section is constructed, which is partially saturated with a drop of liquid. Then let shear wave pass through the whole body of the rock, the drop will be stretched, and the contact line will move against dynamic friction. Also, in terms of the quantification of the contact line friction, there are several theories. And before this, the strain of each point in the rock body needs to be calculated. This work referred to Muskhelishvili’s theory and obtain the friction coefficient by obtaining the strain under arbitrarily specified far-field stress and liquid pressure, and finally integrate it into the shear wave propagation equation in MHBS.	4) Attenuation simulation of shear wave passing through partially saturated crack considering dynamic friction.
Nov.16		*NIU Zihan・Matsushima Jun(1) 1:University of Tokyo
Nov.16	11:20 - 11:40　　　 The application of machine learning and AI to geophysical exploration data have been widely discussed. Sakaguchi et al. (2022) suggested the possibility of applying AI to 3D seismic data and demonstrated its effectiveness. On the other hand, the applied data was limited. Therefore, it is desirable to evaluate to real reflection seismic processing data with more complex geological structures. In this study, we used AI processing to the real reflection seismic processing data and discussed the results. We summarized these results and discussed necessary items for social implementation.	5) Application and systematization CNN(Convolutional Neural Network) for the real reflection seismic processing data.
Nov.16		*HIRONORI SAKAGUCHI・kazuyoshi Takaichi・Satoshi Kodama・Yusuke Shimono・Hisako Shimizu(1),Shohei Ishinabe・Yoshitaka Nakayama(2) 1:CTC,2:JOGMEC


Session	[Earthquake I][Volcano]
Chair	Tsutomu Ochiai (Kanagawa Univ.)

Nov. 16	13:00 - 13:20　　　 Dense seismometer array observations were conducted at the Higashidori site of Tohoku Electric Power Co., Inc. from November 2013 to November 2015. Using three-component natural seismic wave records obtained from these observations, we estimated the subsurface structure using receiver function analysis and seismic wave interferometry (autocorrelation analysis). A total of 441 earthquakes were observed. Most of them were plate boundary earthquakes caused by Pacific slab subduction. The receiver functions showed a strong back-azimuth dependence, suggesting the possibility of the existence of dipping subsurface structures. The depth sections of the receiver functions and P- and S- wave reflections indicated the existence of a common structure at a depth of 2.5 to 3 km. It is interesting to note that the information extracted from different time windows of different components of the seismic waveform records show common structural features. From the comparison of the existing exploration results and the velocity structure model, this boundary is interpreted to correspond to a seismic basement.	6) Estimation of basement using receiver function and seismic interferometry analysis of dense seismic array records at the Higashidori site
Nov. 16		*Toshiki Watanabe(1),Shintaro Abe・Yoshihiro Sawada(2) 1:Nagoya Univ.,2:ADEP
Nov. 16	13:20 - 13:40　　　 This paper estimates the attenuation characteristics of Hachijojima Island from the twofold spectral ratio method. The twofold spectral ratio method removes the source and site amplification characteristics by using multiple earthquake records observed at multiple stations, and extracts only path characteristics, i.e., attenuation characteristics. Earthquake data observed by a dense seismic network of 46 stations in Hachijojima Island during the period from September, 2019 to March 2020 were used. Twofold spectral ratios were calculated for a total of 241 station pair data, and the Q structure was estimated by BPT as an initial model for the inverse analysis. As a result, the Q-values beneath Higashiyama volcano are higher than those of Nishiyama volcano. Besides, in Nishiyama volcano, the Q-values of the area where the dike intrusion was estimated in 2002 were relatively high.	7) Attenuation characteristics of Hachijojima Island from twofold spectral ratio method using dense seismic observation (Part 2)
Nov. 16		*Utako Watanabe・Hiroyuki Azuma・Yoshiya Oda(1),Toshiki Watanabe(2) 1:TMU,2:NU
Nov. 16	13:40 - 14:00　　　 It was reported that the volcanic activity of Hachijojima Island got active in 2002. To evaluate the risk of eruption, we will analyze underground structures by the technique using natural earthquakes. Hence, we need many accurate first arrivals of seismic waves generated by natural earthquakes to improve the accuracy of the analysis. However, the number of data is too large to detect first arrivals manually. Therefore, we applied deep learning to the data observed in Hachijojima Island. We applied EQ transformer, an AI-based first arrivals detection technique, to the observed data for 7 months. As a result, we have obtained a high value of about 94% for the recall value of the confusion matrix compared with manual first arrival detections. The difference between arrival times detected by EQ transformer and manual are 0.01s and 0.09s for P- and S-wave on average respectively.	8) Automatic Seismic Wave Detection by Deep Learning Using EQ transformer : Study of Application to Observed Data for 7 Months in Hachijojima Island
Nov. 16		*Hikaru Kunimasa・Hiroyuki Azuma・Yoshiya Oda(1),Toshiki Watanabe(2),Toshifumi Matsuoka(3) 1:Tokyo Metropolitan Univ.,2:Nagoya Univ.,3:Fukada Geological Institute
Nov. 16	14:00 - 14:20　　　 The estimated seismic intensity in the east of Kagoshima-city during the 1914 Sakurajima earthquake included 6 upper with a large spatial variation. This is thought to be influenced by the subsurface structure in this area. Therefore, we conducted array microtremor surveys for evaluating site amplification characteristics. We obtained phase velocity of Rayleigh waves and H/V spectral ratios in this area. These results suggest that the S-wave velocity of the subsurface is lower in the east than in the west of this area.	9) Field Survey for Evaluation of Site Amplification Characteristics in the East of Kagoshima-city
Nov. 16		*Misa Moriwaki・Seiji Tsuno・Masahiro Korenaga(1),Hiroaki Yamanaka(2) 1:RTRI,2:Tokyo Tech
Nov. 16	14:20 - 14:40　　　 Geological survey and modeling based on microtremor survey was conducted at KiK-net Yamamoto(MYGH10). MYGH10 has recorded maximum accelerations exceeding 1000 gal in the 2011 off the Pacific coast of Tohoku earthquake. Joint inversion analysis using the spectral ratio of borehole to surface, H/V spectral ratio, and phase velocity of microtremors was conducted to estimate the S-wave velocity structure from the surface to the seismic basement. The estimated subsurface structural model explained the observed data well, and the depth of the seismic basement was estimated to be about 600 m. Seismic response analysis of weak and strong ground motions was conducted to validate the estimated subsurface structural model. The predominant frequency of about 4-6 Hz in the strong-motion records is thought to be caused by a shift of the peak of about 6-10 Hz to a lower frequency of about 4-6 Hz due to nonlinearization of the subsurface structure.	10) Geological survey and modeling of shallow subsurface structure at KiK-net Yamamoto by Microtremor Survey
Nov. 16		*Haruhiko Suzuki・,Shunpei Manabe(1) 1:OYO
Nov. 16	14:40 - 15:00　　　 This paper describes a practical method for measuring Q-values based on ultrasonic testing data of rock core. Furthermore, the effects of a fracture, which considered the main cause of the discrepancy between in-situ and laboratory attenuation measurements, and the water-saturated conditions on the attenuation evaluation are examined by conducting ultrasonic tests on ordinary rock cores and cores with artificial fractures and immersion in water, respectively.	11) Attenuation measurements of rock core samples by using ultrasonic testing data and the effects of fracture and water saturation on attenuation evaluation
Nov. 16		*Hiroaki Sato・Tetsuji Okada(1) 1:CRIEPI


Nov.16 　　Room No.2（Small Conference Room，2F）

Session	[Civil Engineering I][Disaster Prevention I]
Chair	Shogo Komori (AIST)

Nov. 16	10:00 - 10:20　　　 The authors have been conducting research on the estimation of ground structure using constant microtremors observation and borehole data in the Shonai Plain along the coast of Yamagata Prefecture. In this study, borehole data were collected from open data, and microtremors array surveys were conducted at the sites to verify the accuracy of the estimation of ground structure. The S-wave structure estimated from the N-values of the borehole data and the S-wave structure estimated by the microtremor array survey showed generally good agreement in about 70% of the cases. Some deviations were observed in some areas, but this could be due to several factors, such as estimation errors in the S-wave structure from the N-values. These results confirm that the combination of borehole data and microtremor array survey is useful for estimating ground structure. In the future, we plan to develop a concrete model and validate the developed model.	28) Comparison of borehole data and microtremors array surveys for the Shonai Plain
Nov. 16		*Tsutomu Ochiai・Takahisa Enomoto(1),Yoshiya Oda・Hikaru Ota(2),Kazuya Mitsuji(3),Shigeki Senna(4) 1:Kanagawa University,2:Tokyo Metropolitan University,3:Yamagata University,4:NIED
Nov. 16	10:20 - 10:40　　　 Many landslides were induced by heavy rain within areas underlain by the Takada Rhyolite (Late Cretaceous), southern Hiroshima prefecture. Weathered rock and soil of igneous rocks on slopes have a risk of slope failure. We performed two study cases to investigate subsurface structure in the rhyolite region; first, surface wave method was applied to the slope to survey soil thickness. A boundary of S wave velocity was appeared at 2-3 m depth on the slope, which was consistent with results of simple penetration tests. Low velocity area (<170 m/s) was below the well-known value of soil (Nc=10). Second study case, Drone-electromagnetic survey (D-GREATEM) with 70 m flight height and electrical resistivity survey were conducted to investigate the distribution of reddish weathered rhyolite affected by hydrothermal alteration. The results of the D-GREATEM showed 800-1600 Ωm in entire survey lines and low resistivity value was not appeared. The weathered rhyolite was about 10 m wide between fresh rhyolite rocks. The investigation shows one of the limitation of the D-GREATEM. The electrical resistivity survey showed low resistivity values (<50 Ωm) at the alteration zone.	29) Subsurface structures survey in Cretaceous rhyolite areas, Hiroshima prefecture -Survey of soil thickness and weathering alteration zone on slope-
Nov. 16		*Yohei Morifuji・Yasuto Hirata・Kenji Kubota(1) 1:CRIEPI
Nov. 16	10:40 - 11:00　　　 At a bridge replacement site, hot spring water equivalent to the source of a nearby hot spring hotel came out in the soil boring borehole. Seven ERI surveys were conducted parallel and perpendicular around the site, to understand the geological structure and distribution of hot springs and to ensure that future construction work would not affect the source. The results of the survey revealed 3D distribution of hot springs and the geological structure of this area and provided useful data for replacement work in the future.	30) Resistivity investing at bridge replacement site near hot spring source
Nov. 16		*Keiichiro Sakanishi(1),Takeshi Fujioka・Takumi Sato(2) 1:MONY,2:Hokuyo Kensetsu
Nov. 16	11:00 - 11:20　　　 Combining trajectory data from drone flight experiments and simulations using Frequency Modulated Continuous Wave (FMCW) SAR, we created Single Look Complex (SLC) images assuming reflected waves from a single target point. Flight experiments showed that RTK has a small variation in orbit error, and the orbit data using Point Positioning (PP) and Real Time Kinematic (RTK) were incorporated into the simulation to compare their SLC images. As a result, it was confirmed that RTK is effective for SAR-equipped drones because it can achieve high positioning accuracy and hovering accuracy. We suggested that the degree of variation in orbit data is important for the creation of high-precision SLC images. Future demonstration tests under various conditions should be conducted to improve SLC images obtained by actual experiments.	31) Development and application of the simulation of FMCW SAR-equipped drones for practical observations
Nov. 16		*Yutaro Shigemitsu・Kazuya Ishitsuka・Weiren Lin(1) 1:Kyoto univ
Nov. 16	11:20 - 11:40　　　 Sequential Gaussian co-simulation (COSGSIM) is applied for evaluating an estimation error to a bedrock elevation or bedrock depth using spatially sparse sampled borehole data and spatially dense sampled geophysical data. Variation of the generated realizations by COSGSIM is regarded as an estimation error. Numerical simulation using a given true model is conducted in this study. Borehole data is created by selecting several locations from the true model, while geophysical data is produced by smoothing the true model. One hundred realizations are generated by COSGSIM to evaluate the variations. The variance of the realizations shows a larger value at distant locations from the borehole in general. In addition, the variance is large at the place where geophysical data rapidly changes in space. This result indicates that the geophysical survey result is properly implemented in the error evaluation by COSGSIM.	32) Uncertainty of the geological survey estimated by sequential Gaussian co-simulation
Nov. 16		*Chisato Konishi(1) 1:OYO Corporation


Session	[Rock][Groundwater]
Chair	Koji Kashihara (JAPEX)

Nov. 16	13:00 - 13:20　　　 It is necessary to understand the processes of strain accumulation and fracturing in a rock mass with complex distribution of faults. Conventionally, numerous experiments with sand boxes and numerical models have been conducted. However, there is no detailed study on the distribution of strains and fractures in the three-dimensional (3D) rock mass around a parallel two-faults system. In this study, the 3D discrete element method (3D-DEM) was used. 3D-DEM is an effective method for three-dimensionally and quantitatively understanding the internal state of objects with large-scale deformations. We carried out numerical simulations of a rock mass containing zones of vulnerability relative to their surroundings (e.g., weak zones). As a result, fractures tended to progress from bottom to top in rock mass models with two parallel weak zones. The simulated rupture distribution roughly agrees with a high fractured zone below active faults inferred from an audio-frequency magnetotelluric sounding.	33) Rupture process simulation of rock mass including weak zones
Nov. 16		*Kazuma Okada・Seiichiro Kawahara・Tada-nori Goto(1) 1:Univ. Hyogo
Nov. 16	13:20 - 13:40　　　 In an excavation of shafts and galleries in the deep underground for disposing high level radioactive waste, an excavation disturbed zone (EdZ) or excavation damaged zone (EDZ) is developed around the shafts and galleries owing to the stress redistribution. It is important to understand the behavior of the EdZ or EDZ because these zone could have an effect on the transfer behavior of radioactive nuclide. The authors performed the in situ experiment for more than 10 years to investigate the behavior of EdZ or EDZ in the 140 m and 250 m gallery in the Horonobe Underground Research Laboratory. From the results of electrical resistivity tomography and so on, the authors compared the characteristics of EdZ or EDZ between those in 140 m and 250 m gallery. The electrical resistivity increased after the gallery excavation in the 140 m gallery. This suggests that the unsaturated zone appeared near the gallery wall. However, the electrical resistivity didn’t change after 13 years from the gallery excavation. On the other hand, the electrical resistivity was almost constant after the gallery excavation in the 250 m gallery. However, electrical resistivity slightly increased after 3 years from the gallery excavation.	34) Investigation of long-term behavior of the Excavation disturbed Zone due to the excavation of horizontal galleries -Study in the Horonobe Underground Research Laboratory-
Nov. 16		*Kenji Kubota・Yohei Morifuji(1),Hiroya Matsui(2) 1:CRIEPI,2:JAEA
Nov. 16	13:40 - 14:00　　　 It is important to confirm the condition of groundwater inside the road embankment for evaluating its safety. In this study, we applied electrical measurements on the road embankment where underground water springs out from the bottom of the slope. According to the resistivity structure estimated from the electrical measurements data, we found that the area of low resistivity is distributed along the old topography. Based on the old topography and the electrical measurements, we detected that the groundwater percolates along the old topography inside the embankment and flows out to the ground surface from the boundary between cutting and filling. Electric measurements become be effective methods to confirm the condition of groundwater inside the embankment.	35) Detecting the path of groundwater seepage in the road embankment with spring using electrical resistivity survey
Nov. 16		*Yusuke Oishi・Kyosuke Onishi・Shunji Kato・Takuo Azuma・Naoya Sasaki・Nozomu Suzuki(1) 1:PWRI
Nov. 16	14:00 - 14:20　　　 This study aimed to measure the resistivity and permittivity of the ground using airborne drone-based electromagnetic (D-EM) exploration and to explore groundwater. Initially, the resistivity of soil and brick samples was measured using current at a frequency of 1.0-160 kHz, the samples were found to exhibit a frequency-dependent effect (FE). On the other hand, the D-EM exploration conducted in the present study is measured by the time domain, but it can be considered as the frequency domain by frequency-based decomposition of the waveform, the signal is used high-frequency electromagnetic waves of several tens of kHz or higher. Therefore, the FE is considered to be included in the resistivity values obtained by the D-EM. Furthermore, based on the concept of equivalent circuit, the FE was considered to be caused by ground permittivity. Therefore, one-dimensional inversion software was developed for D-EM explorations that could calculate the resistivity and permittivity. The software was verified using the data of a landslide area in Aruse, Tokushima, and steep mountains in Muroto, Kochi. In the results, D-EM explorations based on the resistivity and permittivity could identify the wet zones, thus, proving to be an effective method for groundwater exploration.	36) Groundwater exploration by airborne drone electromagnetics: Detection of underground wetting zones using permittivity
Nov. 16		*Akira Jomori・Atsuyoshi Jomori・Tetsuya Toyama(1),Syuichi Hasegawa・Atsuko Nonomura(2),Youichi Yuuki(3),Katsushi Kawato(4),Atsuhiko Kinoshita(5) 1:Neo Science,2:Kagawa University,3:OYO,4:Nippon Engineering Consultants,5:NILIM
Nov. 16	14:20 - 14:40　　　 In the past, the irregularity in the ground at the Higashidori site of Tohoku Electric Power Co. by high-density observation using a high-sensitivity velocity seismograph was examined and it was revealed that the standard variation (SD) of the spectral ratios between the reference point and the other observation points depends on frequency. Here, we examined two cases, the case where all observation points were used (All) and the case where only observation points similar to the reference point were used (Rock), and there was a clear difference in the SD characteristics between All and Rock was taken. One possible reason for this is the nonlinear behavior of the unconsolidated layer. Then, we investigated the seismic behavior of the unconsolidated layer in the small-amplitude earthquake input motion. It was clarified that a slight non-linear behavior was exhibited in the small shear strain range of 4×10-8 to 2×10-4 , though those non-linearities do not affect the predominant frequency of the unconsolidated layer.	37) Non-linear Behavior of Unconsolidated Layer during Small-amplitude Earthquake Input Motion
Nov. 16		*Yoshihiro Sawada・Yoshihiro Tazawa(1) 1:ADEP
Nov.16	14:40 - 15:00　　　 We detected the SPAC’s temporal variation by using a Surface wave excited by a microtremor signal in western Shizuoka Prefecture (Morimachi) and continuous observations were performed from October 2020 to May 2022. In this study, daily SPAC was calculated and the temporal variation of SPAC was examined using the average of SPAC over a year and a half as the reference. The results showed a difference in SPAC between weekends, weekdays, and holidays, which was assumed to be due to a change in the source of the microtremor signal. In addition, It was also found that the stream near the seismic array has a significant impact on SPAC during heavy rainfall. At the same time, seasonal changes are also found. After eliminating these effects as much as possible, we found that the SPAC in the high-frequency region decreases due to rainfall, which can be assumed to be due to a rise in the groundwater level caused by rainfall.	38) Surface wave monitoring for detection of temporal variation in microtremor signal
Nov.16		*Feng Chen・Koshun Yamaoka(1),Ryouya Ikuta(2),Shuhei Tsuji(3),Toshiki Watanabe・Haruno Koike(1),Hiromu Oba(2) 1:Nagoya University,2:Shizuoka University,3:Japan Agency for Marine Earth Science and Technology


Nov.16 　　Core Time: Room No.1 or SpatialChat, Exhibition: Exhibition Space or Online

Session	[Poster Sessions: Core Time]
Chair	Masashi Kodama (AIST)

Nov. 16	15:20 - 17:20　　　 In this study, we extracted high-resolution reflection profiles by autocorrelation of natural earthquakes obtained from distributed acoustic sensing (DAS) installed off the coast of Sanriku. We estimated reflection traces along 50 km of the cable at every 10 m by applying autocorrelation analysis with spectral whitening for S-wave records. The resulting reflection profiles reveal the shallow subsurface structure with higher resolution than in previous studies that used ambient noise. This difference can be caused by larger energy of natural earthquakes in high frequencies than that of ambient noises. This study shows the potential to estimate the high-resolution subsurface structure using natural earthquakes obtained from DAS measurements.	P-1) Seismic interferometry for the DAS-derived natural earthquake data: Construction offshore seismic reflection profile
Nov. 16		*Yuji Sakagami(1),Takeshi Tsuji(2),Tatsunori Ikeda(1) 1:Kyushu Univ,2:Tokyo Univ
Nov. 16	15:20 - 17:20　　　 It's suggested that water ice exists in the lunar polar regions, which can be used as rocket fuel for planetary exploration. However, the exact distribution and amount of water ice haven't been clear. Ground penetrating radar (GPR) is expected to be promising methods for detecting water ice because of its ability to detect water ice in shallow areas with high resolution. In this study, we investigated the relationship between the electrical conductivity and permittivity and water contents of Lunar regolith simulant to determine the exploration depth of it using GPR. We conducted both experiments and simulations. In experiment, we observed sudden decrease in conductivity, probably caused by the freezing process. Besides in simulation, we found the water content increases with increasing heterogeneity of ice distribution up to the point of following the Archie's equation. In addition, it was found that the value of electrical conductivity increased rapidly at some points. This suggests that the value of electrical conductivity varies significantly with the degree of ice connectivity. From the above, it can be concluded that the simulation using digital rocks is useful method to investigate the difference of electrical conductivity due to the distribution of ice for Lunar regolith simulant.	P-2) The measurement and estimation of electrical conductivity of Lunar regolith simulant with different water contents
Nov. 16		*Koshi Wakura(1),Kazuki Sawayama(2),Takeshi Tsuji(3),Tatsunori Ikeda(4) 1:Kyushu university ,2:Kyoto university geothermal research laboratory,3:Tokyo university,4:Kyushu university
Nov. 16	15:20 - 17:20　　　 3D velocity structure models of sedimentary layers have been proposed in Japan, and efforts have been made to verify and modify their performance in a strong ground motion simulation. Therefore, we decided to develop a computer code for data processing and analysis in the Python to make it easier to perform the series of ground motion waveform-inversion analyses presented in Kasamatsu et al. (2021). We here present the idea for developing the code including preprocessing of earthquake ground motion records before the inversion. At this stage, we are still in the process of creating the code for the preprocessing. Our goal is to develop a general-purpose code that enables data processing and interactive operations that take advantage of the characteristics of the Python.	P-3) Sharing of the calculation code for waveform inversion analysis
Nov. 16		*Nobuyuki Yamada(1),Kentaro Kasamatsu(2),Hiroaki Yamanaka(3) 1:Kochi University,2:Kajima Corporation,3:Tokyo Institute of Technology
Nov. 16	15:20 - 17:20　　　 Heavy rainfall in July 2018 caused movements of the landslide in Aruse district in Miyoshi City, Tokushima Prefecture. These caused damage to houses, roads, etc. In order to carry out landslide countermeasure works effectively, it is necessary to clarify infiltration routes of groundwater underground three-dimensionally. In addition, it is necessary to clarify places where groundwater level tends to rise and to intensively drain groundwater there. In this study, in 2021, we conducted Drone Airborne TEM in I-3 block of Aruse district in two seasons, the flood season (July 23) and the dry season (November 26). Also, from the resistivity data obtained by Drone Airborne TEM, three-dimensional resistivity maps were created for two periods. Infiltration of groundwater during the rainy season was clarified by taking the difference between two resistivity maps and by taking the ratio of two resistivity maps. From these results, it was found that in I-3 block of Aruse district, resistivity distribution characteristics are different between east and west sides of the block. This is thought to be caused by differences in geological features such as faults, which may affect the infiltration of groundwater.	P-4) The characteristics of the resistivity of the ground in the landslide in the Aruse area, Tokushima Prefecture using Drone Airborne TEM
Nov. 16		*Atsuhiko Kinoshita・Wataru Takeshita(1),Kei Fukui(2),Takao Yamakoshi・Hiroaki Nakaya(1),Hirokazu Goto(3),Katsushi Kawato(4),Akira Jomori(5) 1:NILIM,2:Shikoku-Sanchi Sabo Office,3:NIPPON KOEI,4:NIPPON ENGINEERING CONSULTANTS,5:NEOSCIENCE
Nov. 16	15:20 - 17:20　　　 Conventional geophysical data inversion generally uses matrices of partial derivative and requires mathematical or physical relationship between observed data and models to be estimated. The use of machine learning is rapidly increasing since the method able to statistically predict models from data and require neither physical relationship between observed data and model nor the matrices of partial derivative. The machine learning may be able to shorten computation time and decrease initial model dependence in the inversion of geophysical data. It may be able to take into account discrete information with no physical relationship to geophysical data, such as soil classification or geomorphological classification. The paper summarizes neural network that is one of supervised learning, and introduces several application examples of the neural network to geophysical data analysis and interpretation.	P-5) Application of neural network to analysis and interpretation of geophysical data
Nov. 16		*Koichi Hayashi(1) 1:OYO Corporation/Geometrics
Nov. 16	15:20 - 17:20　　　 In recent years, the frequency of floods and sediment disasters caused by local heavy rains and earthquakes has been increasing. Regarding sediment disasters, it is also important to predict disasters that will occur after a while. This sediment disaster detection system has a high possibility of detecting large-scale sediment disasters. It indicates that there is a high probability that uncollapsed areas of the same scale can also be detected. In the future, we will continue to improve and implement the system with a view to the possibility of extracting uncollapsed area that will lead to early damage assessment.	P-6) Early detection of sediment disaster using satellite data
Nov. 16		*Takako Nakamura・Jun Matsushima(1),Shuichi Rokugawa(2) 1:The Univ. of Tokyo,2:NIED
Nov. 16	15:20 - 17:20　　　 I introduced cross-gradient minimization as a third constraint to clustering i nversion, which is constructed using two constraints, "smoothing" and "cluster ing", and examined its extensibility. The additional introduction of cross-gra dient minimization has led to a more compact solution, although its effect is less noticeable. Such extensions can be done simply by adding a relating term to the objective function, and the increase in computational cost is kept to a minimum. This indicates the possibility that if a new mathematical model repr esenting the distribution of underground physical properties is proposed in th e future, it will improve the reliability by additionally introducing it as a constraint condition.Here, I report the effect of the additional introduction of the cross-gradient minimization constraint and the results of applying the extended clustering i nversion with the cross-gradient constraintto the joint inversion of Gdd (Gzz) derived from airborne gravity gradient survey and aeromagnetic data.	P-7) Development of clustering inversion of potential field data for geological differentation (3) Examination of extensibility by introducing the third constraint
Nov. 16		*Eiji Nakayama(1) 1:unaffiliated


Nov.17 　　Room No.1（Large Conference Room，1F）

Session	[Civil Engineering II][Disaster Prevention II]
Chair	Masahiro Korenaga (RTRI)

Nov. 17	10:00 - 10:20　　　 We carried out comparative data acquisition of active and passive surface wave methods using three-component velocity meters (geophone) and distributed acoustic sensing (DAS) at a research field located in Richmond, California, U.S. Several different fiber optic sensor cables were deployed at several different depths in a trench of 100 m length. Three component geophones were deployed on the ground surface along the trench with 2 m spacing. A sledge hammer was used for active data acquisition. Seismic ambient noises were also recorded over several days for passive measurements. Active seismic data obtained from geophones was converted to strain time histories and quantitatively compared with DAS data. The strain time history obtained from DAS agreed very well with those obtained from the 2 Hz geophones. We compared dispersion curves of active and passive surface wave data obtained from geophones and DAS based on a multichannel analysis of surface waves (MASW) and a spatial autocorrelation (SPAC). Dispersion curves obtained from DAS are reasonably consistent with those obtained from the radial component of geophones in a frequency range between 4 and 35 Hz in active data and 3 and 20 Hz in passive data. Processing results of passive measurements demonstrated that a low sensitivity cable used for telecommunication, and a cable on the ground surface loosely covered with soil, provided consistent dispersion curves with geophones. The results imply that DAS can be applied to various types of geotechnical monitoring.	12) Quantitative comparison of active and passive surface wave data obtained from three-component geophones and distributed acoustic sensing (DAS)
Nov. 17		*Koichi Hayashi(1),Hayato Nonaka(2),Peter Hubbard(3),Yasuhiro Yokota・Kensuke Date・Kazuhiko Masumoto(2),Taka'aki Taira・Kenichi Soga(3) 1:OYO Corporation/Geometrics,2:Kajima Corporation,3:University of California Berkeley
Nov. 17	10:20 - 10:40　　　 We studied microseisms observed in the Chuetsu region of Niigata prefecture during Typhoon Hagibis in 2019, compared with microseisms during Typhoon Jebi in 2018. We recognized microseisms with a dominant period of 6 seconds only during Typhoon Hagibis in 2019. This wave component appeared when the typhoon reached the offshore area in the Pacific Ocean, and this dominant period did not correspond to half of the dominant period of significant waves observed in the Niigata port and the Naoetsu port during Typhoon Hagibis in 2019. We executed the frequency-wavenumber spectrum analysis using local array data and showed that this wave component propagated from the East side.	13) Microseisms with a period of 6 s observed in the Niigata Chuetsu area during passage of Typhoon Hagibis in 2019
Nov. 17		*Tomiichi Uetake・Hikima Kazuhito・Akihiro Shimmura(1) 1:Tokyo Electric Power Company Holdings
Nov. 17	10:40 - 11:00　　　 Bursa city is one of the large and old cities in the northwestern part of Turkey with many historical buildings. The main part of the city is located in Bursa basin covered with Quaternary deposits. Local site amplification factors at strong motion stations in the area had been estimated using earthquake ground motion data (Yamanaka et al., 2017). In this study we deduced one-dimensional S-wave profiles of shallow and deep sedimentary layers at the strong motion stations in and around Bursa basin from inversions of the amplification factors. S-wave velocity and thickness of each layer over the basement with an S-wave velocity of 2.15 km/s were determined so as to fit the amplification factors to theoretical amplifications at frequencies of 0.5 to 20 Hz. Q-values were also determined in the inversions. The derived S-wave profiles at the sites in the center of the basin have deep basement depths, while the stations at the northern and southern edges have shallow depths. The near-surface layers with low S-wave velocities are also identified at the stations in the center of the basin. We finally discussed effects of the shallow and deep parts of the sedimentary layers using the derived models.	14) Inversion of S-wave profiles from spectral amplifications in Bursa basin, Turkey
Nov. 17		*Hiroaki Yamanaka(1), Hiroe Miyake(2), Masashi Morita(3), Daiki Sato(1),Kentaro Kasamatsu(4) 1:Tokyo Inst. Tech.,2:Earthq. Res. Inst., Univ. Tokyo,3:Yokohama National Univ.,4:Kajima Tech. Res. Inst.
Nov. 17	11:00 - 11:20　　　 Almost of all conventional inversion of seismic travel-time tomography use ray-tracing techniques. In this article, we applied an adjoint state method to travel-time tomography inversion so that we can avoid to compute seismic ray-path. We first applied the eikonal equation to define the travel-time distribution in the model, where velocity (slowness) field is defined, and the mismatching between the calculated travel time at receivers with the observed first arrival time. Then we interpret that this inversion analysis is a functional optimization problem with the equational constrains of the eikonal equation, so that we extend mismatching functional by adding the eikonal equation with multiplying by the Lagrange multiplier and with integrating over the model. The gradient of mismatch functional for perturbating model velocity (slowness) is effectively calculated using an adjoint state method	15) Seismic travel-time tomography inversion using an adjoint state method
Nov. 17		*Makoto Okada(1) 1:NC Geophysical survey)
Nov. 17	11:20 - 11:40　　　 Direct SV reflection events are often observed in the near-offset “noise cones” of seismic vertical shot records. We have demonstrated the capability of reconstructing an SV reflection section from multichannel surface wave survey dataset which contained predominant noise cones and SV events. We also showed SV seismic reflection sections were concordant with SH sections along with SH-wave VSPs. However there still remained an uncertainty how SV waves generate and propagate in the near surfaces when hitting the surface vertically. We then simulated wave fields using Discrete Wave-number Method (DWM) to visco-elastic media for vertical and horizontal impulsive surface sources. Consistent results were certified between observed and synthetic waveform data when using an appropriate Vs model which could explain not only the surface wave characteristics but also SV reflection events. Accordingly, incorporation of DWM waveform synthesis is useful to estimate near-surface Vs structure in combination with ordinary MASW processing, which had a limitation in estimating Vs structure deeper than 15 m or more.	16) Characteristic fetures and utilization of SV-reflection phases buried in the noise cone of seismic survey records
Nov. 17		*Inazaki Tomio(1),Hayashi Koichi(2) 1:Independent,2:Geometrics/OYO Corporation


Session	[Infrastructure Maintenance][Buried Objects]
Chair	Keiichiro Sakanishi (MONY)

Nov. 17	13:00 - 13:20　　　 Frequency-domain electromagnetic methods can be used to obtain information at different depths by changing frequencies. One of the systems is a multi-frequency electromagnetic survey system that can measure several different frequencies at once. One such system is the GEM-2 (Geophex, USA), which can transmit and receive electromagnetic fields at more than 10 frequencies at a time. However, an increase in the number of frequencies results in a decrease in the signal-to-noise ratio, while a decrease in the number of frequencies measured at a time results in the need to measure multiple times. In order to obtain a wide range of depth information using a large number of frequencies, it is important to maintain data quality and perform efficient measurements with a small number of measurements. Therefore, in this study, to investigate an efficient measurement procedure, measurements were performed by changing the number and combination of frequencies measured at a time, and the obtained data were quantitatively evaluated. As a result, it was found that the resultant quadrature data did not change much when the combination of frequencies was changed, but the standard deviation depended on the number of measured frequencies.	17) Evaluation of calibration and measurement data by combination of different frequencies for multi-frequency electromagnetic survey system
Nov. 17		*Ryosuke Umezawa・Yuji Mitsuhata・Toshiyuki Yokota(1) 1:AIST
Nov. 17	13:20 - 13:40　　　 We carried out experiments to discuss about direction of antenna and survey line density for detecting buried pipes parallel to the roadway using a vehicle mounted GPR. The results of the experiments suggested that changing the direction of the antenna was not cost-effective means for a vehicle mounted GPR. We confirmed that the detectability of longitudinal buried pipes was improved by increasing the density of survey lines. However, there is a limit to measure many survey lines at once by a vehicle. Thus, the radar data acquired with multiple runs must be synthesized for the detection of longitudinal buried pipes. For precise positioning, we mounted MMS (Mobile Mapping System) on the vehicle, and we obtained the correction value for shifting to overlap a laser point cloud of a run to the other run. Consequently, we could successfully synthesize the cross section of the GPR image. In the future, we plan to improve the data synthesis technology for the advanced detectability of buried pipes.	18) A detection method of buried pipes parallel to the roadway using a vehicle mounted GPR
Nov. 17		*Kunio Aoike・Yoshihiro Yamashita・Fumiya Nakayama・Satoshi Okada・Hideto Yonezawa(1) 1:OYO Corporation
Nov. 17	13:40 - 14:00　　　 We show the way to estimate the moisture distribution in bridge deck slabs using multichannel ground-penetrating radar. The method which uses propagation velocity estimated from traveltime reflected from the bottom face of deck slabs and the thickness of the deck slab to estimate dielectric constant and moisture content, fits well to reference data. The reflection wave from the bottom face of deck slabs is clearly identified from other signals and it enable us to analyze automatic processing with high accuracy. In this study, we also organize typical signals of ground-penetrating radar applying for bridge deck slabs and show essential points for analysis of the deck slab.	19) Mapping the moisture distribution of reinforced concrete deck slabs using ground-penetrating radar
Nov. 17		*Kyosuke Onishi(1),Takayuki Kobayashi(2),Yusuke Oishi(1) 1:Public Works Research Institute,2:Oyo Corporation
Nov. 17	14:00 - 14:20　　　 In rockfill dams, differential subsidence is likely to occur near protruding bedrock, and shear deformation caused by large-scale earthquakes may occur. Through this shear deformation zone , further erosion of the embankment due to water storage is concerned. In this process, fine grains are gradually washed out, and high permeable zone is generated. Under such background, it was expected that the location of the high permeable zone in the embankment could be detected by the DC electrical method when the reservoir water level was lower due to the periodic inspection of the rockfill dam. As a results, we detected a remarkable resistivity change zone near the expected right bank abutment. Next, assuming that the high permeable zone is formed by the outflow of fine particles, using the Kozeny-Carman equation and the Glover equation for soil consisting of sand and clay, we estimated the hydraulic conductivity, electrical resistivity, and cross-sectional area of the high permeable zone. Based on these estimates, we performed a numerical simulation using a resistivity model assuming the high permeability zone. As a result, the remarkable resistivity change zone with the reservoir level fluctuation was almost consistent with the analyzed results using the measurement data at the site. In conclusion, it is likely to exist high permeable zone where fine grains flowed out due to shear deformation in the upper part of the steep change in bedrock slope near the abutment.	20) Monitoring of high permeable zone in embankment of rock fill dam using the DC electrical method
Nov. 17		*Koichi Suzuki(1),Hideaki Ohnishi・Tetsuro Mampo・Go Orukawa・Masaharu Takabatake(2) 1:Hokkaido University,2:J-Power
Nov. 17	14:20 - 14:40　　　 The noncontact acoustic inspection method using acoustic irradiation induced vibration and laser Doppler vibrometer has already been verified in various actual concrete structures, and its effectiveness has been confirmed. On the other hand, due to the improvement of the method itself, it has become possible to explore a wide area in a shorter time than at the beginning of the research. Therefore, this time, we conducted a re-verification experiment on the existing concrete specimens using the current noncontact acoustic inspection method.	21) Research on noncontact acoustic inspection method for non-destructive inspection, -Example of re-verification of concrete specimens -
Nov. 17		*Tsuneyoshi Sugimoto・Kazuko Sugimoto(1),Noriyuki Utagawa・Chitose Kuroda(2) 1:Toin Univ. of Yokohama,2:Sato Kogyo Co., Ltd.


Session	Executive Sessions
Chair	Yuji Mitsuhata (AIST)

Nov.17	15:00 - 16:00	Kouji Umeda (Hirosaki Univ.)
Nov.17	15:00 - 16:00
Nov.17	16:00 - 17:00	Makiko Hashiba (Hirosaki City)
Nov.17	16:00 - 17:00


Nov.17 　　Room No.2（Small Conference Room，2F）

Session	[Geothermal]
Chair	Toshiki Watanabe (Nagoya Univ.)

Nov.17	10:00 - 10:20　　　 We report a method for estimating subsurface thermal conductivity, heat flow supplied from deep and past ground surface temperature (GST) change (GST history) from deep borehole temperature data with subsurface structure information. For the method, GST history is considered as “heat source” for estimating subsurface thermal conductivity. We applied the method to borehole temperature data obtained at a carbon dioxide capture and storage (CCS) demonstration project site in Tomakomai, Hokkaido. To apply the method to borehole temperature data without subsurface structure information, we developed a simple method for inferring subsurface structure related to layered thermal conductivity from borehole temperature data only. To verify the usefulness of this method, we assumed that borehole temperature data obtained at Tomakomai CCS demonstration project site has no information on subsurface structure and estimated GST history, subsurface thermal conductivity, and heat flow successfully. Acknowledgements: We used borehole temperature and lithologic data obtained from observation well of the CCS demonstration project conducted by the Ministry of Economy, Trade and Industry in Tomakomai, Hokkaido, Japan.	39) Estimation of subsurface thermal conductivity and heat flow from deep borehole temperatures using ground surface temperature change due to recent warming
Nov.17		*Shusaku Goto(1),Makoto Yamano(2) 1:AIST,2:ERI, Univ. Tokyo
Nov. 17	10:20 - 10:40　　　 The various geophysical surveys are conducted by Japan Oil, Gas and Metals National Corporation (JOGMEC). The reflection seismic survey reveals the geological history of the formation of the Onikobe caldera, and shows not only the series of principal fractures and faults but also the horizontal intrusions of magma at the base of caldera. Currently, airborne magnetic survey by drone is planned to conduct in such area as around the former Onikobe geothermal power plant where the HELITEM survey by JOGMEC did not cover. Prior to the survey, the series of analysis including the inversion of total geomagnetic anomaly data to the magnetization of five (5) subsurface layers, the comprehensive moving window correlation analysis between geomagnetic and gravity-gradient data sets, in addition to the conventional analysis, in order to assess the sites of the buried dikes and volcanic conduits and the distributions of lava, and to verify the geomagnetic surveys to be effective to understand past volcanic activities. The results of the analyses show the distributions of the buried prominent and stable lava flows and the sites of volcanic conduits which supplied the dacite lava. These sites are distributed linearly and are considered to relate one of the principal fractures and faults shown in the seismic survey. And they are the left lateral strike-slip faults considering the areas are under east-west compressional stresses by the Pacific plate.	40) The airborne magnetic survey in the Onikobe geothermal area
Nov. 17		*Shigeki Mizutani(1) 1:KGE
Nov. 17	10:40 - 11:00　　　 In geothermal areas, the distribution of hydrothermally altered layers has been clearly visualized by magnetotelluric (MT) exploration. Based on the MT results, the distribution of geothermal reservoirs can be discussed. However, the spatial interval of MT sites is generally several hundred meters or more, and the detailed distribution of the hydrothermally altered layers was not clearly identified. In this study, audio-frequency MT (AMT) survey was carried out with the site intervals of 50-200 m, and the detailed distribution of low-resistivity zones (considered to be the hydrothermally altered layers) was imaged. By using such detailed subsurface images, we succeeded in inferring the ascending paths of geothermal fluid and the descending ones of rainwater. Such information will greatly contribute to the development of geothermal energy.	41) Migration paths of geothermal fluid inferred from super-high-density audio-frequency magnetotelluric survey
Nov. 17		*Tada-nori Goto・Nagi Yamashita(1) 1:Univ. Hyogo
Nov. 17	11:00 - 11:20　　　 Magnitude estimation from seismic data is one of the essential analyses when we perform seismic observations, but sometimes we have a record with only one component or a limited number of stations. In this study, we evaluate the magnitude estimation ability with a small number of station data, the induced micro-earthquakes recorded at the Fenton Hill Hot Dry Rock geothermal test site in 1984. We compare the moment magnitude estimates of the events from the corner frequency of a single component seismic data. We also estimate the low frequency flat level of the seismic spectrum that is proportional to the seismic magnitude. The amplitude level obtained from the two stations shows good agreement between two stations, while the seismic moment obtained from the corner frequencies show different values on each station. The results indicate amplitude level is more reliable for estimating event size than using corner frequency when we have a limited number of seismograms.	42) Magnitude estimation for micro-earthquakes at Fenton Hill hot dry rock geothermal test site
Nov. 17		*Nana YOSHIMITSU(1),Michael Fehler(2) 1:Kyoto Univ.,2:MIT
Nov. 17	11:20 - 11:40　　　 In 2022, we carried out a DAS seismic study in the Takigami geothermal field in Kokonoe town, Oita prefecture. The survey was carried out by installing optical fiber cable down to 2,000 m into the TP-2 geothermal well with 12 seismic sources in the vicinity of Takigami field. Based on the DAS data and synthetic seismic waveforms, the E-W Vp depth profile determined a distinct vertical fault with an N-S strike concordant with the regional geology. Migrated reflection phases were obtained by F-K filtering for a better reflection image of the deeper part. The S-to-P converted phases were fortuitously observed by surface geophones when natural earthquakes occurred during the survey. The conversions depths are estimated at around 1 km and 2.5 km depth.	43) Geothermal exploration study in the Takigami geothermal field using the DAS system in the geothermal well
Nov. 17		*Junzo Kasahara・Yoko Hasada(1),Shigetsugu Furuya・Masahiro Shimazaki(2),Hiroshi Ohnuma(1),Hitoshi Mikada(3),Yoshihiro Fujise(1) 1:ENAA,2:IdemitsuKosan Co.,3:Kyoto Univ.


Session	[Resource Exploration I][Oil and Natural Gas I]
Chair	Mitsuyuki Ozawa (JGI)

Nov. 17	13:00 - 13:20　　　 For the majority of oil and gas fields, it is significant predicting the spatial distribution of hydrocarbon source rocks. Lishui Sag is one of the most promising offshore oil and gas exploration areas in China. However, the knowledge of source rocks is limited to a few locations of previous drillings, so it is necessary to determine the distribution of source rock in the whole area. We describe an integrated method involving non-standard processing of seismic reflection data with sedimentary facies analyses to control the prediction accuracy of source rocks. And we applied the method on our study area, Lishui Sag. The results show that the seismic reflection of source rocks in this area is characterized by low frequency and high amplitude. Source rocks in the East Subsag and South Subsag are few and thin because these two places are far away from the source of sediment, the thickness of source rocks is less than 200m. The thickness of source rocks in the West Subsag is the largest in the whole area, reaching 500m. By comparing the results in the literature, we believe that the method can provide reliable information on the spatial distribution of potential source rocks.	44) A case study: Combining acoustic correlation and seismic attributes to calculate source rocks in Lishui Sag, China
Nov. 17		*Na Li(1), Jun Matsushima(2),Zhang Jinliang・Xuwei Luan・Tao Chen(3) 1:BNU & UTokyo,2:UTokyo,3:BNU
Nov. 17	13:20 - 13:40　　　 Modulation of the common vibroseis signal is introduced. Linear vibroseis and common nonlinear vibroseis signals are compared. Utilizing analysis of amplitude spectrum and correlation wavelet, detailed differences in effects caused by different signals are demonstrated and discussed. The design of nonlinear vibroseis is discussed in order to enhance the performance of vibroseis signals. Inspired by the remarkable echolocation system of bats and the potential privilege of echolocation calls of bats in retaining information, echolocation calls of bats are studied. By means of echo removal and spectral subtraction, the changing pattern of frequency modulation of echolocation calls is acquired. The bat-like vibroseis signal is mathematically derived from there.	45) Nonlinear Vibroseis Waveform Design Based on Echolocation Call of Bats
Nov. 17		*Wangjie Xu・Jun Matsushima(1) 1:Environmental system deepartment, GSFS, Univerisity of Tokyo
Nov. 17	13:40 - 14:00　　　 Seismic facies classification is a part of the seismic data interpretation. The classification of seismic facies has conventionally been performed by manual interpretation. On the other hand, recently, many seismic data have been acquired in 3D seismic, and such classification of the entire data would require huge cost. In this study, we investigate a workflow for seismic facies classification using supervised learning with CNN. Currently, our study is conducted using EfficientNet (Tan and Le (2019)), which has achieved high accuracy in image classification, and we discuss the results obtained in this study	46) A Study of Seismic Facies Classification using Convolutional Neural Network.
Nov. 17		*Yoshitaka Nakayama・Shohei Ishinabe・Kazuaki Ishikawa・Takuya Ishimaru(1) 1:JOGMEC
Nov. 17	14:00 - 14:20　　　 Our previous study proposed a coupled analysis of cosmic ray muons and seismic waves and tried to develop a joint inversion analysis method using both numerical data. To verify the applicability and performance of the developed joint inversion, we will conduct laboratory experiments of a combination of muons and seismic waves and apply the developed joint inversion to laboratory data. In the present paper, we investigate the feasibility of the laboratory experiment for the observation of muons. We demonstrate the applicability of muon laboratory measurement, and the necessity of further investigations of optimum laboratory environment to observe muons. Then, we outline the conceptual design of 1D combined laboratory measurements of muons and ultrasonic waves. We also emphasize the advantage of the 1D approach over the 3D approach.	47) Combined measurement of cosmic ray muons and elastic waves on a laboratory scale
Nov. 17		*Jun Matsushima(1),Masashi Kodama(2),Mohammed Y. Ali・ Fateh Bouchaala(3),Hiroyuki Tanaka(4),Tadahiro Kin(5),Toshiyuki Yokota(6),Makoto Suzuki(7) 1:Univ. of Tokyo, GSFS,2: AIST,3:Khalifa University,4:Univ. of Tokyo, ERI,5:Kyushu Univ., IGSES,6:AIST,7:Univ. of Tokyo, SoE
Nov. 17	14:20 - 14:40　　　 The underground rock usually contains cracks or factures which are filled with fluid, meaning that pressure difference in those cracks or fractures can be generated when seismic wave passes through it, since the shape of cracks or fractures can be deformed. Naturally, wave-induced fluid flow can be generated because of the pressure difference in the fluid. Those fluid flow in microscopic scale (squirt flow) could cause wave attenuation and velocity diffusion. Based on these energy losses, underground structure can be exposed, but most of research to date is based on idealized simple models. In this paper, 3D numerical simulations on a digital core sample which was obtained in an Abu Dhabi oil field were conducted by combining stress-strain relationship and Navier-Stokes equation, and those two equations are solved in solid and fluid domain respectively.	48) Simulation-based rock physics modeling in three-dimensional digital core rocks for wave-induced fluid flow in fractured media
Nov. 17		*Xiaoliang Wang・Jun Matsushima(1),Mohammed Ali(2),Fateh Bouchaala(2) 1:The University of Tokyo,2:Khalifa University


Nov.18 　　Room No.1（Large Conference Room，1F）

Session	[Earthquake II][Disaster Prevention III]
Chair	Haruhiko Suzuki (OYO)

Nov. 18	10:00 - 10:20　　　 Northern Aso Caldera was severely damaged by the 2016 Kumamoto earthquake (MJMA 7.3). KiK-net Aso (NIED) has been installed in northwestern Aso Caldera, but it has been suspended since 2012, and no strong motion records have been observed in the Kumamoto earthquake. In order to analyze earthquake damage in surrounding areas, estimating ground motion during the main shock is necessary. In order to properly evaluate the seismic motion, array observations of microtremors were performed to estimate the S-wave velocity structure up to the seismic bedrock and to examine the strong ground motion characteristics in the northern part of Aso Caldera. As a result, a highly accurate dispersion curve was obtained by the SPAC method, and the S-wave velocity structure model was estimated based on it. The ellipticity of Rayleigh wave fundamental mode estimated by the model showed relatively good correspondence to the R/V spectral ratio at the site, and also when compared to the data from previous studies around K-NET Ichinomiya, it generally corresponds to the trends in geology map and Bouguer gravity anomalies.	22) Estimation of S-wave Velocity Structure at KiK-net Aso Based on Array Observations of Microtremors
Nov. 18		*Shu Tomizaki・Hikaru Hara・Michiko Shigefuji・Tatsuo Kanno(1) 1:Kyushu Univ.
Nov. 18	10:20 - 10:40　　　 The Ebino earthquake (Mj6.1) occurred on February 21, 1968, near the border of Miyazaki, Kagoshima, and Kumamoto prefectures The damage caused by this was reported to have varied greatly from area to area, even within the relatively small area of the former Ebino Town. In this study, to investigate the factors behind this difference of the damage, we conducted microtremor array observations at two locations: the area where no damage was seen in the Ebino earthquake (EBN) and the area where the damage was severe (MSK). Then, the S-wave velocity structures were estimated, and the characteristics of the ground motion were investigated. A comparison of the estimated underground structures indicates that the EBN has a lower S-wave velocity at the surface than MSK, inferring that it is soft ground with a thicker sedimentary layer. The 1D ground amplification factor calculated using the estimated structures indicated that the underground structures at both sites amplify low frequency components.	23) Estimation of S-wave Velocity Structures around Damaged Areas by the 1968 Ebino Earthquake Based on Array Observations of Microtremors.
Nov. 18		*Hikaru Hara・Shu Tomizaki・Michiko Shigefuji・Tatsuo Kanno(1) 1:Kyushu Univ.
Nov. 18	10:40 - 11:00　　　 The Kanto Basin where the capital of Japan is located comprises thick sediments which amplify seismic waves. Constructing an accurate underground structure model is important to reveal the propagation properties of seismic waves, and provide better hazard assessment. We constructed a three-dimensional S-wave velocity model for this area using the joint inversion of multimodal dispersion curves of Rayleigh and Love waves. Our estimated model better explain the observed complex surface-wave dispersion curves that were not able to explain in the past homogeneous multilayered model. Our high-precision three-dimensional S-wave velocity model reflects underground heterogeneity and improves the earthquake simulation and hazard assessment.	24) Multimodal surface wave joint inversion for S-wave velocity structures in Kanto Basin
Nov. 18		*Hiro Nimiya(1),Tatsunori Ikeda(2),Takeshi Tsuji(3) 1:AIST,2:Kyushu University,3:University of Tokyo
Nov. 18	11:00 - 11:20　　　 In order to understand the spatial variation of surface ground S-wave velocity structure in and around Mt.Mannichi in the northern part of the Kumamoto Plain, high-density array microtremor survey, surface wave exploration, and single point microtremor measurement were conducted. The phase velocity of Rayleigh waves was large at the stations located in the mountainous area, and low phase velocities were observed at the eastern and southeastern edge part of Mt.Mannichi, and at the points on the flat areas. The estimated shallow S-wave velocity structures indicate that there is an abrupt change in the depth of the bedrock layer from the mountainous area to the edge and flat areas. The results of H/V spectra from single point microtremor measurements also show a similar trend.	25) High-density array microtremor observations and surface wave exploration in and around Mt.Mannichi in Kumamoto City for estimating shallow S-wave velocity structures
Nov. 18		*Masahiro Korenaga・Seiji Tsuno(1),Hiroaki Yamanaka(2) 1:RTRI,2:Tokyo Tech.
Nov. 18	11:20 - 11:40　　　 We tried to detect the effect of ground improvement using microbial metabolism by estimating a two-dimensional S-wave velocity structure with a waveform inversion. The two-dimensional ground along the survey line was divided into 1 m × 1 m blocks, and the optimum S-wave velocity for each block was evaluated by the inversion. Since we inverted the complicated structural model with many unknown parameters, the observed waveforms could be well reproduced. At the location where the ground improvement was conducted, the larger S-wave velocities were successfully estimated than those of the surrounding area. We also found that the surface ground has a complex S-wave velocity structure, for example, S-wave velocities near the surface differ locally.	26) Waveform Inversion for Two-dimensional S-wave Velocity Structure near Surface
Nov. 18		*Kentaro Kasamatsu(1),Hiroaki Yamanaka(2),Naohito Adachi・Tomoki Hikita(1),Takahiko Hidekawa(3) 1:Kajima Technical Research Institute,2:Tokyo Tech.,3:Kajima Design
Nov. 18	11:40 - 12:00　　　 In seismic reflection surveys targeting active faults in urban areas, three-dimensional surveys are practically difficult due to constraints such as roads and residential land, so two-dimensional surveys are exclusively applied. The authors had the opportunity to conduct a seismic reflection survey targeting the eastern edge of the Nara Basin. In this survey, in addition to the usual reflection method seismic survey, a survey line was set up for receiving only seismic waves, and lateral imaging was expanded. It was clarified that the two-dimensional geological structure extends in the direction where the existence of the fault is pointed out from the topographical information. By installing multiple seismic lines, it should be possible to understand the three-dimensional geological structure even in urban areas.	27) Application of Poor Man's 3D seismic reflection survey for active faults in urban areas
Nov. 18		*Koji Yamada・Mitsuhiro Toya・Hirofumi Kowari・Shigeru Okamoto(1),Tomotaka Iwata(2) 1:Hanshin Consultants Co., Ltd.,2:DPRI Kyoto Univ.


Nov.18 　　Room No.2（Small Conference Room，2F）

Session	[Resource Exploration II][CO₂ II][Metal]
Chair	Kyosuke Onishi (Public Works Research Institute)

Nov. 18	10:00 - 10:20　　　 CCS reservoir monitoring is important for assuring complete sequestration and long-term stability of the CO2 injection. Reservoir expansions are sometimes large enough to be observed as the surface uplift. As the expansion and compaction of CO2 reservoirs would affect on the integrity of the sequestration system, overall geo-mechanical consideration is crucial. In this study, we start with the systematic description of four compressibility terms for poro-elastic rock materials, which affect much on the deformation due to CCS. Secondly, we explain the CCS procedure from the point of pore pressure and consider the uncertainty of compressibility. Finally, we discuss actual expansion uncertainties and point out that the amount of expansion varies in the range of five times even in the same porosity, which depends on the matrix configuration of rock materials.	49) Compressibility on Geomechanics for CCS Monitoring
Nov. 18		*Akihisa Takahashi・Takumi Onuma・Jun Matsushima(1) 1:The University of Tokyo
Nov. 18	10:20 - 10:40　　　 The integrity of seal layers against fluid leakage is essential for CO2 geological storage or CCS, but required data from the seal layers are rarely available. We investigated the method for estimating the mechanical properties of seal layers from the cutting samples. The indentation test and mineral analysis were performed on the cutting samples collected during drilling to a carbonate reservoir, and showed that the elastic moduli of the seal layer is dominantly determined by those of the clay minerals. The test results also indicated the increased elastic moduli of the samples with the decreased clay mineral content lower than 40%, which is an important finding because the intercalated stiff thin layers can increase fracture pressure partially in the seal layers and improve the seal integrity. Therefore, we focused on the elastic moduli of the low clay mineral content samples for further investigation. The comparison with the rock physics model showed that the variation in the obtained moduli could be explained by presence or absence of cementation at the contacts of mineral particles. The measurements also showed obviously lower elastic moduli than those of minerals, probably reflecting the elasticity in larger scales than mineral scales. Although further study is necessary before conclusion, the application of the indentation test to cutting samples may become a promising tool for CCS seal integrity assessment.	50) Rock Physics Modeling on Indentation Test Data for Seal Layers of CCS
Nov. 18		*Koji Kashihara(1),Yu Nagano(2) 1:JAPEX,2:JOGMEC
Nov. 18	10:40 - 11:00　　　 In the seafloor photography test using the AUV "Deep1", a fluxgate three-component magnetometer and a potentiometer for self-potential measurement were also installed, and a known sunken ship survey was conducted off the coast of Miho, Shizuoka City. At the same time, the detailed topography of the survey area was acquired by the permanently installed multi-beam echo sounder (MBES). Significant magnetic anomalies and self-potential anomalies were observed near the top of the sunken ship. The magnetic anomalies were positive on the south side of the hull and negative on the north side. The potentiometer showed a negative self-potential at the bow and a positive self-potential at the stern. These results indicate that the magnetic anomalies are mainly the magnetic field induced by the geomagnetic field and the self-potential is horizontally polarized. On the other hand, potentiometer voltage changes with Deep1 attitude are also observed, and these changes are explained by the current direction and angular velocity during course change.	51) On the Geomagnetic and Self-Potential records obtained from a Sunken ship survey by AUV Deep1
Nov. 18		*Ryuji Kubota(1),Shingo Sugimoto(2),Shinji Gotoh・Toi Tachibana(1),Kosuke Masuda・Yuka Asano・Yuki Otsuji(3) 1:KGE,2:JOGMEC/KGE,3:FSM
Nov. 18	11:00 - 11:20　　　 In order to specify parameters controlling time-domain IP (induced polarization) response, DC electrochemical measurement was conducted by employing pyrite, which is a semiconductor and shows a prominent IP effect, as a working electrode. Anodic polarization curve for pyrite obtained by scanning potential was divided into three regions, that is the Tafel region, the small potential region, and the large potential region, which is the same result as that in previous studies. For the small potential region, there is a difference in anodic polarization curves between when a working electrode (pyrite) is illuminated by fluorescent light and when the light is shut off because holes in the valence band (minority carrier) of pyrite participate in the electrochemical reaction and induce the current flow of the electrode. Exchange current density, which is one of the parameters of IP phenomenon, was calculated with the two controlling equations for the small potential region and the Tafel region.	52) Study on the electrochemical properties of pyrite showing induced polarization (IP) effect
Nov. 18		*Hajime Hishida(1),Shinichi Takakura(2),Takumi Ueda(1) 1:Waseda Univ.,2:AIST
Nov. 18	11:20 - 11:40　　　 In this study, we developed a potential receiver system for efficient multi-channel induced polarization (IP) survey on a long survey line and applied the system to a blind black ore deposit. The Dipole-Dipole method can be used for simultaneous multi-channel measurements with multiple potential electrode combinations with different electrode isolation factors (n) by using a multi-core cable system. However, the use of conventional multi-core cables in long measurement lines exceeding several hundred meters tends to increase the total weight, and the cost of the cable itself is not small. In this study, a lightweight and inexpensive multi-channel potential receiving system using a LAN cable and a small switch box was developed, which can simultaneously acquire up to n=7 data with a single current injection. This system was successfully applied to a time-domain IP survey (electrode spacing 50 m, line length 700 m) at a blind black ore deposit in Japan.	53) Development of an efficient multi-channel IP measurement system for long survey lines and application to a blind black ore deposit
Nov. 18		*Shogo Komori・Ryosuke Umezawa・Shinichi Takakura(1),Takafumi Murakita・Masaki Sugisaki・Eiichi Arai(2),Nobuhiko Shiga・Makoto Harada・Hidehiro Ishikawa(3) 1:AIST,2:JOGMEC,3:MINDECO
Nov. 18	11:40 - 12:00　　　 We propose a unified rock physics model for simulating the characteristics of elastic wave dispersion and attenuation in frozen porous media, which could have different consolidated degrees, ice particle distributions, and surrounding temperatures. In this model, we introduce the consolidation coefficient into the skeleton building step, which is divided into ice-formed and sand-formed parts, and define another distribution coefficient to differentiate ice particle suspending and supporting effects. Also, we take into account the temperature-dependent effects with three different mechanisms to achieve a better match to the observed S-wave attenuation abnormality. They are viscoelastic behaviors of ice grains due to thickness changes of water films; content changes of unfrozen water; and viscosity changes of pore fluids. In particular, the Burgers’ constitutive equation is used to establish the complex elastic modulus of ice matrices, in which a temperature-dependent steady-state creep rate is built by the Arrhenius theory with a constant-Q assumption.	54) A Temperature-dependent Rock Physics Model for Partially Frozen Sediments
Nov. 18		*Bonan Li(1),Jun Matsushima(2) 1:Nanjing Tech University, The University of Tokyo,2:The University of Tokyo

VENUE
	Hirosaki Civic Hall, AOMORI, 1 Shimo-Shirogane-Cho, Hirosaki City, AOMORI, JAPAN and ONLINE

SCHEDULE
	Wednesday, November 16, 2022
		Openning Speech	Room No.1 (Main Conference Room, 1F)	09:50-09:55
		__Speaker:	Keiichi SUZUKI (Kawasaki Geological Engineering Co. Ltd.; President of SEGJ)
		Technical Sessions	Room No.1 (Main Conference Room, 1F)	10:00-15:00
		Technical Sessions	Room No.2 (Small Conference Room, 2F)	10:00-15:00
		Poster Core Time	Room No.1 (Main Conference Room, 1F)	15:20-17:20
		Student Exchange Event	Room No.2 (Small Conference Room, 2F)	17:30-18:30

	Thursday, November 17, 2022
		Technical Sessions	Room No.1 (Main Conference Room, 1F)	10:00-14:40
		Technical Sessions	Room No.2 (Small Conference Room, 2F)	10:00-14:40
		Executive Sessions	Room No.1 (Main Conference Room, 1F)	15:00-17:00
		__Chair:	Yuji MITSUHATA (AIST; Vice-President of SEGJ)
		__Presentation(1):	Deep Structure Imaging by Geophysical Survey and Earth Science of Subduction Zone Kouji UMEDA (Hirosaki Univ.)
		__Presentation(2):	Attraction of Cherry Blossoms in Hirosaki Park Makiko HASHIBA (Hirosaki City)
	Friday, November 18, 2022
		Technical Sessions	Room No.1 (Main Conference Room, 1F)	10:00-12:00
		Technical Sessions	Room No.2 (Small Conference Room, 2F)	10:00-12:00
		Field Trip	Hirosaki Castle and Sannai Maruyama Special Historical Site	12:50-17:30

Registered and Paid		Pre-Registration (Until October 21,2022)	Onsite Registration
Conference Meeting	Member / Non-member	JPY 7,700-*	JPY 8,800-**
Conference Meeting	Student	JPY 3,850-*	JPY 4,400-**
Field Trip	Member / Non-member	JPY 3,000-*	JPY 3,000-**
Field Trip	Student	JPY 1,000-*	JPY 1,000-**

**	Pre-registrant will receive PAPER Proceedings at the registration desk of the conference.
**	PAPER Proceedings can be purchased at a price of JPY 1,650-.

DESTINATION
	Hirosaki Castle
		Most famous cherry blossom spot in Japan


	Sannai Maruyama Special Historical Site
		A Large Settlement from the Early and Middle Jomon Period (approx. 3,900 to 2,200 B.C.)


SCHEDULE
	Friday, November 18, 2022
		Departure from VENUE after the monring technical sessions around 13:00. Visit to Hirosaki Castle and Sannai Maruyama Special Historical Site. Arrival at Shin-Aomori Station around 16:50 and Aomori Airport around 17:30.


	More information about the field trip will be sent to those who have pre-registered the field trip.

CANCELLATION
	No Refund Will Be Made.

LIST OF ADVERTISEMENTS
	Geophysical Surveying Co., Ltd.(GSC) ITOCHU Techno-Solutions Corpration(CTC) JGI, Inc. Kawasaki Geological Engineering Co. Ltd.(KGE)

1) TEST for you internet connection to the online conference
2) Guide for Technical Session
3) DO NOT screenshot or record any part of the conference (including the audios).