Search

A seismic waveform data acquisition system equipped with optical disk units was installed at the Abuyama Observatory, Research Center for Earthquake Prediction, Disaster Prevention Research Institute. Sixty four channels of microseisinic wavefrom data were digitized with a 12 bits AD-converter at a rate of 200 samples per second and stored on a optical disk.Before this system was operated, wavefrom data were stored on analog magnetic tapes. Those data can be restored on optical disks under the same format as the real time data.

application/pdf

application/pdf

In the northern Kinki district, we have done seismic observations using the dense seismic array stations since 2008. Total 150 temporal and permanent stations are used, and the average interval between the stations is about 5km. Using these data, we have investigated the seismic velocity structure, the shape and distribution of the reflector in the lower crust, the Moho and the PHS plate, the focal mechanisms and the stress field etc. The results of these observations and other geophysical data exhibit and characterize the strong contrast between the Tamba region and the Oumi Basin region. Difference of these two region caused to the geometry of subducting PHS plate beneath this area and the distribution and the movement of fluid in the crust.

In the northern Kinki district, we have done seismic observations using the dense seismic array stations since 2008. Total 150 temporal and permanent stations are used, and the average interval between the stations is about 5km. We get numerous mechanism data in a short time period, and the space and time resolution of the focal mechanism analyses are improved. Based on these data, we discuss about the feature of focal mechanisms, space and time variation of the regional stress field, and the detail mechanism distribution within aftershock sequence following the M4 class earthquakes.

To derive detailed crustal stress field, we tried to calculate regional stress field directly from the first motion polarity data using the algorism by Horiuchi et al, (1995). Especially, the direct calculation method in this study is suitable for the old data of Abuyama Observatory to investigate the stress field in the northern Kinki district. Manually checked P-wave polarity data for numerous micro-earthquakes during 1997-2001 are available for this study. The stress field around the Tamba region have been E-W compressional throughout the analyzed period. However, strike-slip mode and reverse-fault mode are coexisting and varying with time and location.

We installed 45 temporary seismic observation stations in the middle and northern Kinki district, where active faults are concentrated. The seismometers and data loggers used are newly developed. We perform the dense observation for about four years and analyze the data together with those of permanent stations to detect heterogeneous structures related to the strain accumulation process. The purpose of the project is to clarify the relation between the strain accumulation and the earthquake generation. In this report, we described about the observation system, the data processing system, and the database system for this project.

To derive detailed crustal stress field, we tried to calculate regional stress fielad directly from the first motion polarity data using the algorism by Horiuchi et al, (1995). Especially, the direct calculation method in this study is suitable for the Abuyama's old data to investigate the stress field in the northern Kinki district before the Hyogo-ken Nanbu earthquake in 1995. Because, the data set of the Abuyama Observatory in the last century is very precise, but number of station was too small to derive the focal mechanisms of each earthquakes.

We determined the focal mechanisms of 815 micro-earthquakes around the Biwako-Seigan fault system using the method of Maeda (1992). The average direction of P axes is around E-W, and a half of earthquakes are reverse fault type. Stress tensor inversion analyses are carried out for a lot of small blocks of 10 km square. It is apparent that a line 10 km west of the Hanaori fault system is the boundary between the regional stress field of reverse fault type around the Biwako-Seigan faults and that of strike slip type around the Tamba plateau.

In many case, the focal mechanisms of microearthquakes are derived from observed P-wave first motion polarities. However the P/S amplitude ratio is useful to improve the precision of the mechanism determination, the measurements of the P and S amplitude are not performed in routine work. We introduce a new method to determine the focal mechanisms using the information about the maximum amplitude, which is observed routinely to determine the magnitude.

Results of the seismic surveys under the Special Project for Strong Motion Evaluation in Urban Areas are reported, with special reference to the relationship between subsurface structures and seismic activities. The re-analyses of the surveys in 2004 revealed a clear image of the subducting Philippine Sea plate as well as the reflectors in the middle and lower crust as well as velocity structures along the Shingu-Maizuru Line. The shallow structure for active faults, such as the Arima-Takatsuki Tectonic Line and Uemachi fault are also obtained in the middle Kinki district.

Underground structures in the Kinki district were surveyed from 2002 to 2006 under the Special Project for Earthquake Disaster Mitigation in Urban Areas. We carried out seismic observations with deploying 35 temporary stations across the Kinki district from the Pacific side to the Japan Sea side, in order to reveal subsurface structures, such as subducting Philippine Sea slab in southern Kinki and the continental Moho in northern Kinki. About 50, 000 event data were selected from continuously observed seismic records of both the temporary stations and nearby permanent stations during May, 2005 and Nov, 2006. We also try to develop a data communication system with a mobile telephone for a BREW application to improve the observation system.

Several dynamite explosions were performed along the E-W survey line from Kyoto to Gifu under the "DAIDAITOKU project" on OCT 16, 2006. We observed these explosions along another NNE-SSW trending survey line from Kameoka to Minoo to investigate the deep seismic reflector beneath the Tamba Plateau. Total 100 temporal stations were installed along the 30km long survey line. We can identify the evident later phases in the seismic records. These later phases are PxP reflection from the deep crustal reflector.

Disaster Prevention Research Institute (DPRI), Kyoto University has conducted surveys of underground structure in the Kinki district under the Special Project for Earthquake Disaster Mitigation in Urban Areas since 2002. We carried out observations of earthquakes Kinki since 2004 deploying 28 observation stations along a control-sourced seismic survey line from Shingu City, Wakayama Prefecture to Kyoutango City, Kyoto Prefecture. This observation system worked well for obtaining high quality earthquake records with high sensitivity. In this report we describe the improvement of observation systems and the measures against snow.

We determined newly focal mechanisms of about 900 microearthquakes in the Tanba region, adjacent area of the 1995 Hyogo-ken Nanbu earthquake, using up/down sense of P-wave onset, and also analyzed regional stress field using the stress tensor inversion method. Average feature of focal mechanisms and the stress field was not changed by the Hyogo earthquake. But, southwestern areas in the Tanba region near the rupture of the Hyogo earthquake show the changes on P-axis azimuth and the stress field. These results are well corsistent with the theoretical stress pattern caused by the Hyogo earthquake.

Since beginning of 2003, micro-earthquake activity in the Tamba Plateau, Central Japan, have been lowered. This seismic quiecence is similar to that before the 1995 Hyogo-ken Nanbu earthquake and some moderate-size earthquakes in the Tamba region.

Focal mechanisms of microearthquakes in Tamba plateau and surrounding area are investigated. Most P-axes areoriented E-W or ESE-NWN directions. Both strike-slip and reverse fault type earthquakes exist. By the 1995 HyogokenNanbu earthquake, however, it is possible that P-axes in the Tamba plateau changed to ESE-NWN from E-W.

Recently, several large and middle earthquakes were occurred, and seismicity became active in southwesternJapan. The Yamasaki fault system is typical in-land active fault in southwestern Japan. The last large earthquakearound the Yamasaki fault occurred more than 1100 years ago.We investigated the seismicity along the Yamasaki fault, and its surrounding areas. Seismic activity of theseareas shows periodicity of 5-6 years. Much seismic quiescence induced activation, migration were observedbefore/after large earthquakes (M>5). After 2001, active area along the Yamasaki fault migrated toward southeastdirection.

September 15, 2002, the Middle-western Tottori earthquake (Mj5.3) occurred. The seismic activities showed foreshock-mainshock-aftershockpattern. The rupture of the mainshock is ENE-WSW direction. Although, following aftershock activities are complicatedshowing Z shaped distribution. In this paper, we review the seismicity in the San-in area, and describe seismic activities around theMiddle-western Tottori earthquake precisely.

In the Tamba plateau, microearthquake activities show sometimes linear distribution. After the M5.1 earthquake in 2001, linearactivity with 20km length was observed. We developed a simple method to detect such linear trend of seismicity from theearthquake catalogue. In the middle of the Tamba region, NE-SW linearity is detected almost continuously. Along the Mitokefault, many microearthquakes occurred simultaneously.

Induced earthquakes accompanied with the filling action of the Hiyoshi-dam, in the Northern Kinki district, Japan are studied from the data of the small temporary seismological observation network near the dam-site and the routinely operated network.The filling action of the dam was executed on Mar. 17, 1997, about three years after the 1995 Hyogo-ken Nanbu earthquake of M 7.3, which is about 80km southwest from the dam-site. Seismic activity is activated by the big earthquake even in the dam-site.Therefore the area is thought to be sensitive to the changes in the stress or strain.However, the change in the seismicity is not found during and after the filling action of the dam, from the detailed analyses of the data.

We determined focal mechanism solutions of 19 earthquakes in the eastern part of Shikoku and its vicinity from P first motions. After Kobe earthquake, very advantageous environment for the determination of focal mechanism solutions was prepared.We picked up the data of P first motion not only from high sensitivity seismograms but also from ones of instrumental seismic intensity systems.The results support the orthogonal tectonic stress systems generating earthquakes in the upper crust and the upper most mantle.

To examine the effect of station distribution on precision of hypocenter determination, we carried out a simple simulation.Theoretical travel times are calcurated from assumed hypocenter to surface stations. These travel time data are given to hypocanter determination program which we usualy use. Then, we compare the results of relocation and the assumed hypocanters. At area appart from nearest stations more than 20km, depth of the relocated hypocenters are deeper than assumed ones. These depth shift are caused by station distribution, velocity structure, algorithm of program and initial value.

The Northern Hyogo earthquake (Mj5.4) occurred on JAN12 in 2001. Following activities expanded 6km along EW direction and 7km aiong NS direction.From space distribution and focal mechanisms, the hypocenter distribution can be divided into 4 or 5 clusters.In this study, we compare the seismicity in past and the activities in 2001, and discuss about the feature of seismicity and focal machanisms.

The urgent aftershock observations were carried out just after the 2000 Tottori-ken Seibu Earthquake (Mj7.3). Five temporal stations were installed around the aftershock area. A huge number of small aftershocks were recorded. At the southern part of the area, aftershocks are distributed linerly along NNW-SSE direction. Depth range of aftershocks are 3 - 12km. In the northern part of the area, aftershocks spread wider zone, and make some clusters. Depth of aftershocks are mainly 5km. The feature of the distribution of aftershocks are well correlated with the slip distribution of the mainshock.

We compare the three dimensional structure of b-value to that of seismic velocity in the Tamba region. No correlation is recognized between b-value and P-wave velocity. However, we can recognize a weak positive correlation between b-value and S-wave velocity and a weak negative correlation between b-value and Poisson's ratio.These correlations suggest that b-value can be affected by the existence of fluid in the Tamba region.

The earthquake with Mj 7.3 occurred in the western part of Tottori prefecture.At epicentral area, the higher level of intensity 6 were recorded. No one was killed, but more than 100 peoples were injured.The seismicity has been high on the border area of Tottori-Shimane prefecture. In the epicentral area, the swarm earthquakes have been occurred from 1989.Most of these swarm earthquakes distribute on the low slip portion of the main shock.Four low frequency earthquake which occurred at 30 km depth were observed after October 1997. About 12 hours before the main shock, small event of Ml .7 occurred at 3 km distance from the rupture start point of the main shock.Aftershock distribution was obtained from the routine network and the urgent temporary observations.On the south part, aftershocks distribute within the narrow zone, but they disperse in the north part. The induced earthquake occurred in two area where ware south-east of Mt. Daisenand east of Shimane prefecture. Most of aftershocks related with the rupture of main shock distribute on the low slip portion of the main shock.

On February 12, 1999, the earthquake with magnitude 4.0 occurred near the Kameoka City, Kyoto Prefecture. Before the earthquake, micro-earthquake activity around the epicenter was lowered since November, 1998. Similer phenomenon was also observed before the M5.0 earthquake on May 28, 1987 near Kameoka. These seismic quiecence can be detected clearly on the cumulative curve of the earthquake occurrence in 20 km square area including epicenter of the mainshock. Many exsamples of seismic quiecence before great earthquakes have been reported. We confirmed that seismic quiecence occurred before some moderate-size earthquakes. Analyses about such quiecence are important for investigation of earthquake prediction

To reveal the detailed stress field after the 1995 Hyogo-ken Nanbu earthquake, we relocate hypocenter for aftershocks (M>2.3) and determine these focal mechanisms using seismic records obtained by GROUPS-95.Some aftershocks show nearly N-S compression or normal fault type events whose mechanism aren't consistent with the regional stress field of approximately E-W compression. We define these aftershocks as the unusual aftershocks. These are mainly located at the upper boundary between slipped and unslipped zone in the mainshock. We suggest that the unusual aftershocks are caused by the disturbance of the local stress field due to the heterogeneities of the slip and the stress distribution of the main rupture.

The SATARN system is a unified microseismic network operated in D.P.R.I., which consists of five conventional microseismic sub-networks managed by Kamitakara, Hokuriku, Abuyama, Tottori, and Tokushima observatories. Seismic wave data obtained in each sub-network are transmitted to D.P.R.I. Using either satellite communication link or NTT public phone lines. SATARN is configured with WIN system, which is developed on the UNIX operating system using internet technology. Data processing systems are in operation both in D.P.R.I. And observatories. Obtained seismic wave form data are also broadcasted to the Osaka District Meteorological Observatory, Japan Meteorological Agency.

Most of plate boundaries are located beneath the sea. It is very important to observe crustal deformation around these area. Now, we are developing a seafloor crustal deformation observation system which combines the kinematic positioning by differential GPS and the precise acoustic ranging.The precise acoustic ranging is very important technique for this system. In this study, we perform experiments to confirm the precision of acoustic ranging. According to these results, observation results are affected by following signal after true direct signal. It is possible to improve this observation system by removing these affected data.

In order to reveal the change in the crust, from the view point of the crustal heterogeneity, we investigate the temporal variation of coda Q-1 and b-value. We analyze the waveform data of shallow micro-earthquakes(M 1.5-3.0) in Tainbaregion recordedin a period from 1987 to 1996. At 3.0 and 4.0 Hz bands the temporal variation in coda Q-1 correlates significantly with the Hyogo-kenNanbu earthquake. The variation in coda Q-1 inversely correlates with the temporal variation of b-value. The fracture dimensions of micro-earthquakes which contribute to the variation in b-value are estimated to be 300-400 m. This scale length corresponds to the wavelengths of 3-4 Hz waves, showing significant variation in coda Q-1. The crustal activity in Tamba region is controlled by the heterogeneity with dominant scale of 10^2-l0^3 m.

Hypocenters of earthquakes determined from the data recorded by Kamitakara, Hokuriku, Abuyama and Tottori Observatories, Kyoto University during 1976-1991 are collected to define the distribution of the thickness of the seismogenic layer of the crust in the inner zone of southwest Japan. Total of 63, 961 well-located earthquakes are selected and used to determine the cutoff depths in the crust. As a result, the thickness of the seismogenic layer varies 5-25 km from place to place and is closely related to the thermal structure of the crust. Further, active faults and large inland earthquakes are likely to occur at the steepchange in the cutoff depth. Thus, the thickness of the seismogenic layer is important to examine the rheological characteristics of the crust and eventually to reveal the relationship beteen the crustal heterogeneity and large inland earthquakes.

We have newly developed a public-line telemetry system, which is composed of transmitters and a data acquisition system.The transmitters at seismic stations store waveform data of triggered events. The data acquisition system at DPRI, Kyoto Univ., once a day, gathers trigger information from all the stations, and discriminates seismic data from noise-triggered ones by comparing the trigger time among neighboring stations.After this, only the seismic wave data are transmitted to DPRI from each station.We have installed ten temporary stations around Lake Biwa region by using the telemetry system stated above. Following analyses will be done in this region: detailed seismic activity, 3-Din-homogeneous structure of P-wave velocity and of coda scatterers in the crust, depth distribution of earthquakes, and focal mechanism by waveform inversion.