Joint Hypocenter Determination Method Research Articles

The 1998 Papua New Guinea Earthquake and its Fault Plane Estimated from Relocated Aftershocks

The 1998 Papua New Guinea earthquake of M w 7.0 occurred near the Wewak trench where the North Bismarck plate is subducting beneath the Australian plate. Its mechanism is thrust-type, and one of the nodal planes is almost parallel to the plate interface. To determine which of the two nodal planes of the main shock is the fault plane, we relocated the main shock and aftershocks using a method of modified joint hypocenter determination. We combined and employed two types of data in this study. Firstly, we used data reported by the National Earthquake Information Center (NEIC) of the U.S. Geological Survey (USGS), which includes three stations at the northeastern edge of Irian Jaya and one station in northern Papua New Guinea, from which the epicentral distances are less than 2 degrees. Secondly, in addition to the above permanent-station data, we used data from temporary aftershock observations near the epicentral area around the Sissano Lagoon carried out by TsuJI et al. (1998). Using three-component seismometers, they carried out observations from August 2 to October 2, 1998 at three sites. Although the network did not record the main shock and immediate aftershocks, the data obtained by temporary observation sites can clearly assist in identifying their absolute locations, since it is possible to apply the joint hypocenter determination (JHD) method. Hypocenters were relocated between the coastline and the Wewak trench, distributed along a nodal plane dipping shallowly to the southwest. Therefore, we can conclude that this nodal plane is the main shock fault and that the 1998 Papua New Guinea earthquake was an interplate earthquake between the North Bismarck and Australian plates.

Multiplet-Clustering Analysis Reveals Structural Details within the Seismic Cloud at the Soultz Geothermal Field, France

Multiplet-clustering analysis is a method for precise determination of microseismic event locations and is used to identify subsurface fractures and fracture networks. A multiplet is a group of microseismic events with very similar waveforms, despite different origin times, and it is likely the expression of stress release on the same structure. The relative source locations of similar events can be determined with high resolution and accuracy by using the moving-window cross-spectrum analysis technique. Deduced seismic clusters, called multiplet clusters, are indicative of seismically activated structures, and the orientations of these structures can be estimated using the seismic clusters even though the absolute locations of the multiplet clusters cannot themselves be determined. We examine methods of determining the relative locations of multiplet clusters and introduce the concept of clustering analysis. The clustering analysis method is used to estimate the relative location of multiplet clusters by detecting phase differences between similar stacked events. We describe the procedure for multiplet-clustering analysis, estimate relative locations of multiplet clusters, and apply the method to induced microseismic data from the Soultz Hot Dry Rock field, France. We show that a fracture network can be delineated through multiplet-clustering analysis, whereas it is difficult to identify detailed structures on the basis of source locations estimated by the joint hypocenter determination method. Manuscript received 1 March 2002.

New Evidence for Segmentation of the Alaska Subduction Zone

The purpose of this article is to provide additional evidence for segmentation of the subducted plate in Alaska and to introduce a catalog of the relocated earthquakes for future studies. We used the Joint Hypocenter Determination method to relocate 14,099 subduction-zone earthquakes that occurred from July 1988 to July 1998 and were located between 58° N and 65° N latitude. The earthquake data were taken from the Alaska Earthquake Information Center catalog. The selected earthquakes were divided into 16 blocks on the basis of their hypocentral locations, and each block was relocated separately. Average epicenter shift was 3.8 km and average upward and downward depth shifts were 4.1 and 4.4 km, respectively (roughly the same number of earthquakes shifted upward [47%] and downward [53%]). The overall change with respect to the initial locations is that the seismicity became more compact, revealing details about the fine structure of the Wadati-Benioff zone. In particular, we were able to identify more precisely the boundary between the Kenai and McKinley blocks of the subducting plate. In addition, there is evidence for plate segmentation within the McKinley block. Manuscript received 15 November 2000.

New Constraints on Tectonics of Interior Alaska: Earthquake Locations, Source Mechanisms, and Stress Regime

We used the joint hypocenter determination (JHD) method to relocate 3611 crustal earthquakes that occurred from 1988 to 1999 in central Alaska. The new earthquake locations provide more details on the structure of the Kantishna cluster and better locations for the aftershock sequence of the 29 November 2000 M L 5.6 earthquake in the Minto Flats seismic zone. The JHD locations for the aftershocks of the 1995 M W 6.0 Minto Flats earthquake and 22 October 1996 M W 5.8 earthquake near the Denali fault are also available. A catalog of 196 fault-plane solutions consisting of the moment tensor solutions for the earthquakes with magnitude 4.0 or above and P -wave first-motion solutions for the earthquakes with magnitude 3.4 and above that occurred from 1988 to 2000 was composed. Moment tensor solutions were calculated using regional broadband data. This catalog was used to calculate principal stress orientations in the crust. The stress orientations change across central Alaska. In particular, the maximum principal stress orientation rotates clockwise from SE-NW to SSW-NNE direction as one moves from west to east. These stress orientations are consistent with the stress field transferred from the plate convergence in southern Alaska. In addition, we tested different velocity structures in the moment tensor inversion procedure to identify velocity models for calculating the Green's functions. The moment tensor inversion study shows that it is possible to obtain a reliable moment tensor solution for moderate-sized earthquakes ( M L ≥ 4) using three-component records from a single broadband station when the epicentral distances are between 50 and 300 km. Manuscript received 8 June 2001.

Sequence of strong intraplate earthquakes in the Kodiak Island Region, Alaska in 1999–2001

A sequence of strong earthquakes was registered in 1999–2001 in the Kodiak Island region of the Alaska‐Aleutian subduction zone. Two Mw 7 earthquakes occurred in December, 1999 and January, 2001 and an Mw 6.5 event occurred in July, 2000. These events and their aftershocks recorded by the regional seismograph network were relocated using a Joint Hypocenter Determination (JHD) method. Regional broadband data were used to obtain seismic moment tensors for the main shocks and their largest aftershocks. Relocation and moment tensor inversion results indicate that these events originated inside the subducting Pacific plate. The focal mechanisms indicate down‐dip tension with the fault planes being nearly vertical and parallel to the strike direction of the subducting plate. The 1999 and 2000 events were located down‐dip of the locked portion of the megathrust, while the 2001 event was located directly beneath it.

Fine details of the Wadati‐Benioff zone under Indonesia and its geodynamic implications

Using about 5200 handpicked P, S, pP, sP, PcP, and ScP phases from digitally recorded seismograms, together with International Seismological Centre reported phases, we obtain improved hypocentral locations for ∼1790 earthquakes, ∼983 of them having 90% confidence limits <30 km, at depths >50 km, in the period 1962 to September 1996, along the Indonesian subduction zone. We use an improved joint hypocenter determination method in which the solution is more stable than in the original one. We show that for deep earthquakes, use of the core‐reflected phases is as good as use of depth phases, and since such core phases often have larger amplitudes and sharper onsets than the depth phases, their arrival times can be read very accurately. The positions of the relocated hypocenters show that (1) a portion of the Indonesian arc between ∼110°E and 123°E longitude, and deeper than ∼500 km, is dipping southward at an angle of ∼75°, that is, in a direction opposite to the upper part of the north dipping slab, and (2) east of about 108°E, the seismic zone is wider near 670 km than near the 500 km depth. The first suggests southward lateral flow in the mantle, relative to the plate motion vector here. From the contortion of the seismic zone along the eastern portion of the Indonesian arc, we can estimate the average lateral shear strain rate in the 300 to 670 km depth range to be of the order of 10−16s−1, over the last 10–20 Myr.

The 1995 off-Etorofu earthquake: Joint relocation of foreshocks, the mainshock, and aftershocks and implications for the earthquake nucleation process

Abstract Prominent foreshock activity preceded the Mw 7.9 off-Etorofu earthquake in the Kurile Islands on 3 December 1995. Using a modified joint hypocenter determination method (Hurukawa, 1995), we simultaneously relocated foreshocks, the mainshock, and aftershocks to study the foreshock activity in detail in view of the nucleation process of a large earthquake. The distribution of relocated earthquakes and their focal mechanisms suggest that the 1995 off-Etorofu earthquake was an interplate earthquake at the boundary between the North American and Pacific plates. Its source area overlaps with those of the 1958 off-Etorofu (Mw 8.3) and the 1963 off-Urup (Mw 8.5) earthquakes. Nine days before the occurrence of the mainshock, a first immediate foreshock of magnitude 6.4 occurred at the deepest point of the foreshock area. This event was followed by many foreshocks, including three additional M ≧ 6 events, which occurred east and southeast of the first foreshock. The foreshock area expanded toward the trench axis with a velocity of several to several tens centimeters per second. The number of events per day increased daily, and the final size of the foreshock area was about 80 × 30 km. Eventually, the rupture of the mainshock started at the deepest point of the foreshock area. These observational facts are consistent with recent theoretical studies and laboratory experiments, in which foreshocks are regarded as the rupture of localized asperities in a broad weak zone where the nucleation of the large earthquake started.

Relocation of shallow earthquakes in southern Alaska using Joint Hypocenter Determination method

A subset of 2660 shallow earthquakes (0–50 km) that occurred from 1988 to 1996 in south central Alaska between 155°W and 145°W and 59°N and 63°N was relocated using the joint hypocenter determination (JHD) method. Both P- and S-wave observations recorded by the regional seismic network were used. Events were relocated in twenty different groups based on their geographic location and depth using two velocity models. As a result of the relocation, the majority of the hypocenters shifted downward, while the epicenter locations did not change significantly. The distribution of the shallow subduction zone earthquakes indicates the existence of two seismically independent blocks, with one block occupying the northeastern part and the other occupying the central and western parts of the study area. The boundary between the blocks is marked by a 15 to 20 km wide seismicity gap to the southeast of 149.5°W and 62°N. The analysis of the fault plane solutions for shallow subduction zone earthquakes shows that an overwhelming majority of the solutions represent normal, oblique-normal or strike-slip faulting with predominant WNW-ESE orientation of T-axes. This indicates a down-dip extensional regime for the subducting slab at shallow depths. Very few earthquakes yielded fault plane solutions consistent with thrusting on a contact zone between the overriding and subducting plates. This result may be an indication that currently either the strain energy is not released at the contact zone or it is associated with aseismic motion.

Stress pattern in the double seismic zone beneath Cook Inlet, south-central Alaska

Relocation of south-central Alaskan earthquakes recorded by the regional seismograph network, using the joint hypocenter determination method, revealed a double seismic zone in the underthrusting Pacific plate beneath the Cook Inlet area. The lower zone is separated from the upper one by a decrease in seismic activity within a 5–10 km wide band, and extends from a depth of 50 km to about 90 km. Stress tensor inversion for earthquakes in the upper and lower zones shows different stress regimes. In the upper zone minimum principal stress is oriented in the down-dip direction and maximum principal stress is normal to the Wadati-Benioff zone (WBZ). In the lower zone both maximum and minimum principal stresses are horizontal and rotated by 40–60° from the down-dip direction of the WBZ. Below 80–90 km depth, stress tensor inversion results show down-dip orientation of the minimum principal stress and the maximum stress aligned in the direction of convergence of Pacific plate.

Relocation of earthquakes in the Cook Inlet area, south central Alaska, using the joint hypocenter determination method

AbstractThe Joint Hypocenter Determination (JHD) method was used to relocate 1604 earthquakes that occurred during 1988 to 1993 in the Wadati-Benioff zone (WBZ) beneath the Cook Inlet in south central Alaska. The study utilizes both P- and S-wave arrivals recorded by a recently improved regional seismic network, and it provides an improvement over previous studies. The JHD locations of the events allowed improved delineation of the geometry of the subducted slab and the detection of a double seismic zone within the WBZ in the 40- to 100-km-depth range. The JHD station corrections obtained correlate with the major surface geological features in the study area (slow-velocity sedimentary basins) as well as with crustal and mantle structure (high-velocity subducting slab and low-velocity anomalies beneath the active volcanoes). To investigate the dependence of the JHD relocation procedure on factors such as the hypocentral depths of the events used, the velocity model, the νp/νs ratio, and random and systematic errors in travel-time readings, synthetic travel times were calculated using an exact raytracing technique, a three-dimensional tomographic P-wave velocity model, the present network configuration, and the actual seismicity distribution. Then, the synthetic events were located with the JHD algorithm. The synthetic experiments showed that the JHD procedure is rather insensitive to the choice of plane-layered velocity model and νp/νs ratio, preserves relative hypocenter locations well, and can handle systematic and large random reading errors effectively.

Aftershock Distribution of the January 17, 1995 Hyogo-ken Nanbu Earthquake Determined by the JHD Method.

On January 17, 1995, the Hyogo-ken Nanbu (Kobe) earthquake of magnitude 7.2 occurred in the Kansai district near Kobe, Japan. After the quake, the Net-Hyogo group set temporal seismic stations in and around Kobe and Awaji Island. Using P and S arrival data from temporal observations, we determined the aftershock distribution of the Kobe earthquake. We used the JHD (joint hypocenter determination) method which eliminates mislocation of hypocenters caused by errors of the simplified model of 1-dimension velocity structure and heterogeneity near stations. From the analysis, we obtained final results of the aftershock distribution. First, the hypocenters thus obtained form a thin flat zone in the Kobe area but form a complex feature in the Awaji area. The aftershock zone strikes N51°E and the total area of the aftershock zone is 60 km wide and 15 km deep. Second, in its northeastern half (Kobe and central area), the aftershock zone forms nearly a simple plane. Rupture did not appear at the surface in the Kobe area while a number of aftershocks were located underneath the surface trace of pre-existing active faults. Third, in the southwestern half (Awaji area), it shows complicated features. The aftershock zone dips northwestward near the northern tip of Awaji Island but southeastward in the central part of the Awaji area. Although the rupture trace appeared on the surface along the Nojima Fault which runs on the western coast of the island, no aftershocks have yet been found to occur immediately beneath the surface rupture. The aftershock zone is located 3 km to the east of the Nojima Fault. It indicates that the fault geometry at the surface does not correspond to the distribution of aftershocks in the Awaji area. These results show that surface fault traces do not necessarily connect directly with the underground fault geometry estimated by aftershock distribution if the fault dips vertical. Pre-shocks occurred several hours before the main shock in the central part of the fault region where the initial rupture of the main shock started at the a depth of 17 km. Hypocenters of the pre-shocks and the initial rupture are located at the lower edge of the aftershock zone. This implicates that the precursory rupture process of the Hyogo-ken Nanbu earthquake began at the base of the seismogenic zone.

Quick aftershock relocation of the 1994 Shikotan Earthquake and its fault planes

The 1994 Shikotan earthquake of Mw 8.2 occurred near the western Kurile trench where the Pacific plate is subducting beneath the North American plate. Although its mechanism is thrust‐type, it is not a typical low‐angle thrust event with a nodal plane parallel to the plate boundary that occurs at the plate boundary. In order to determine which of the two nodal planes of the main shock was the fault plane, we relocated the main shock and all aftershocks with M > 5.3 using a modified joint hypocenter determination method. Earthquakes on the northwestern side of the aftershock area including the main shock are located along a plane dipping east‐southeastward with a dip of 70°. This plane is about 80 to 150 km long and 70 km wide. The main shock occurred at the north‐northeastern bottom of this plane, which is almost parallel to one of the two nodal planes of the main shock determined by Harvard University, which has a strike of N51°E and a dip of 76°. Thus we conclude that this plane is the fault plane, that the fault broke at the northeastern bottom of the fault plane and that the rupture propagated to the surface and also to the southeast. In addition to the aftershocks aligned with this fault plane, there is a secondary alignment, which is subparallel to the main fault plane and about 50 km east, where the largest aftershock of Mw 7.3 occurred on Oct. 9. Since several aftershocks occurred on this secondary fault plane on the same day as the main shock, it seems reasonable to suppose that this secondary fault plane was generated together with the main fault plane at the time of the main shock.

Implications of spatial and temporal development of the aftershock sequence for the Mw 8.3 June 9, 1994 Deep Bolivian Earthquake

On June 9, 1994 the Mw 8.3 Bolivia earthquake (636 km depth) occurred in a region which had not experienced significant, deep seismicity for at least 30 years. The mainshock and aftershocks were recorded in Bolivia on the BANJO and SEDA broadband seismic arrays and on the San Calixto Network. We used the joint hypocenter determination method to determine the relative location of the aftershocks. We have identified no foreshocks and 89 aftershocks (m > 2.2) for the 20‐day period following the mainshock. The frequency of aftershock occurrence decreased rapidly, with only one or two aftershocks per day occuring after day two. The temporal decay of aftershock activity is similar to shallow aftershock sequences, but the number of aftershocks is two orders of magnitude less. Additionally, a mb ∼6, apparently triggered earthquake occurred just 10 minutes after the mainshock about 330 km east‐southeast of the mainshock at a depth of 671 km.The aftershock sequence occurred north and east of the mainshock and extends to a depth of 665 km. The aftershocks define a slab striking N68°W and dipping 45°NE. The strike, dip, and location of the aftershock zone are consistent with this seismicity being confined within the downward extension of the subducted Nazca plate. The location and orientation of the aftershock sequence indicate that the subducted Nazca plate bends between the NNW striking zone of deep seismicity in western Brazil and the N‐S striking zone of seismicity in central Bolivia. A tear in the deep slab is not necessitated by the data. A subset of the aftershock hypocenters cluster along a subhorizontal plane near the depth of the mainshock, favoring a horizontal fault plane. The horizontal dimensions of the mainshock [Beck et al., this issue; Silver et al., 1995] and slab defined by the aftershocks are approximately equal, indicating that the mainshock ruptured through the slab.

P-wave residuals at Fiji from deep earthquakes in the Tonga subduction zone

A total of 47 deep earthquakes occurring between 1982 and 1985 in the Tonga subduction zone were used to derive station corrections for the Fiji Seismograph Network (FSN ). The Joint Hypocenter Determination (JHD) method was employed to relocate hypocenters using P -wave arrival times at selected teleseismic and FSN stations. P -wave residuals for Fiji show variations of more than 1 sec between southern Viti Levu on the one hand and northern Viti Levu and Vanua Levu on the other hand. The smallest residuals are observed in southern Viti Levu where the oldest volcanic rocks in Fiji are mapped and on the island of Koro. Station residuals in northern Viti Levu, Vanua Levu, and the smaller Fijian islands except for Koro are on average 0.4 to 0.7 sec higher than in southern Viti Levu. The effect of hypocentral mislocation on the observed residual variations is negligible at stations in Viti Levu, Ovalau, and Beqa, but it may be important for stations in Vanua Levu, Koro, Taveuni, and Yasawa. Large P -wave delays observed at some stations are probably caused by low velocity sedimentary surface layers. The variations in crustal thickness that are probably not larger than 5 km would explain only a minor part of the residual differences. Without a better knowledge of crustal structure the contribution of lateral variations in upper mantle P velocities is difficult to assess.

Relocation of teleseismically recorded earthquakes near Tennant Creek, Australia: Implications for midplate seismogenesis

The three MS > 6 Tennant Creek, Australia, earthquakes of January 22, 1988, were preceded by small and moderate earthquakes in 1986 and 1987. We have used the method of joint hypocenter determination to estimate the positions of teleseismically recorded 1986–1987 earthquakes, the 1988 main shocks, and aftershocks with respect to a system of surface fault scarps. The 1986–1987 shocks and the 1988 main shocks nucleated near the center of the zone of surface scarps, where fault‐segment boundaries at depth are implied by complexities in the distribution of scarps at the surface. This suggests that the fault segments that ruptured in 1988 were already in existence in 1986–1987, which is consistent with the hypothesis that strong midplate earthquakes occur on preexisting faults. The redetermined 1986–1987 hypocenters are, however, also consistent with the hypothesis that midplate seismicity is localized by stress concentration due to bulk rheological heterogeneity of the crust, because they are situated on a regionally prominent gravity anomaly. The teleseismically recorded seismicity of the Tennant Creek region prior to the 1988 main shocks has the temporal pattern of a swarm followed by a lull. The concentration of swarm earthquakes between two scarps is consistent with models in which precursory‐swarm earthquakes correspond to faulting that is spatially distinct from the site of primary main‐shock faulting. The time interval between swarm and main shocks is similar to intervals between intermediate‐term precursory swarms and main shocks in regions that have much higher rates of tectonic loading; this similarity suggests that the time intervals between precursory swarms and subsequent main shocks are not strongly influenced by the rate of tectonic loading, but are determined primarily by time dependence of the failure process. The spatial distribution of teleseismically recorded aftershocks is in most respects like that of aftershocks recorded by a local network of portable stations in the half year following the main shocks. The set of teleseismically recorded aftershocks, like the set of locally recorded aftershocks, includes some events that occurred well away from the causative faults of the main shocks. At the length scale of the 1988 main‐shock rupture, the distribution of aftershocks occurring more than one year after the main shocks is not representative of the distribution of earlier aftershocks.

1964年新潟地震の余震分布再調査結果について

The focal process of the 1964 Niigata Earthquake was reinvestigated on the basis of hypocentral distribution of its aftershocks. This study indicates that the aftershocks are distributed on a fault plane dipping westward.Although it has been clear that the fault strike of the mainshock was in N20°E direction, the dip of the fault was not still clear due to a poor resolution of hypocenter of aftershocks. To resolve the difficulty, we reexamined seismological data obtained by the Japan Meteorological Agency (JMA).Reexamination of seismograms of nearby stations enabled us to supplement more than 1200 new P and S arrivals of aftershocks. We also dentified a number of P and S arrivals from the data which were previously reported as unidentified phases. The Joint Hypocenter Determination method was used to get a more reliable aftershocks distribution. The number of located aftershocks much increased, as about 380 aftershocks are well located by this study.Aftershocks on the vertical cross section which is normal to the fault strike shows that aftershocks are on a westward dipping plane. The dip of the plane is estimated as 50 degrees which is consistent with the focal mechanisms reported by several studies. Although the dip angle depends on the velocity model used in hypocenter location, westward dipping of aftershocks is valid, independent of several different velocity models. Therefore we estimate that the subduction of the Japan Sea under the north-east Honshu does not occur in the southern part of the eastern margin of the Japan Sea.The aftershock activity is found to be low around the hypocenter of mainshock which is located near the bottom of aftershock region, suggesting a large strain release around the nucleation point of mainshock. Relative position of forerunning seismic activity which preceded the mainshock by two years seems to be within the shallow part of the aftershock region east of Awashima-island which is located in the western middle of the focal region. The epicentral distribution of aftershocks indicates that aftershock occurrence is scarce around Awashima-island. A similar relation was reported in the case of the 1983 Nihonkai-chubu earthquake between its aftershocks and Kyurokujima-island, which is situated east of the middle of the aftershock region. Few aftershocks occurred in the area around Kyurokujima-island. In spite of the difference in relative location, that is, Awashima is situated west of the aftershock region while Kyurokujima is in the east, this suggests possibilities that crust around the islands cannot sustain enough strain to generate aftershocks or it behave as an earthquake barrier.

Influence of fault plane heterogeneity on the seismic behavior in the Southern Kurile Islands Arc

The seismic behavior in the southern Kurile Islands Arc is strongly influenced by spatial heterogeneity in the mechanical properties of the plate interface. Maps showing sites of maximum seismic moment release for the great underthrusting earthquakes of 1958, 1963, 1969, and 1973 provide a first‐order image of the major regions of enhanced strength, or asperities. Hypocenters of these and an additional 83 smaller earthquakes with mb ≥ 5.3 and known underthrusting mechanisms were recomputed using the method of joint hypocenter determination in order to assess better the influence that the strength distribution exerts on the interplate seismicity patterns. Nearly all of the smaller‐magnitude earthquakes locate outside of the principal asperities of the great earthquakes. This observation is consistent with substantial release of accumulated strain energy in asperity regions during the great earthquakes and a redistribution of stress in adjacent areas following their rupture. An along‐dip, depth dependent component of strength heterogeneity appears to influence the seismicity patterns. The asperities of the 1958 and 1973 great earthquakes and hypocenters of most of the Ms ≥ 7 earthquakes occur near the downdip edge of the 100‐km‐wide seismically active plate interface. This would be expected if shear resistance on the interface increases monotonically with lithostatic pressure up to the brittle‐ductile or stick slip‐stable slip transition that defines the base of the seismogenic zone. The asperities of the 1963 and 1969 earthquakes, however, appear to be located well above the downdip edge of the coupled plate interface. The occurrence of large asperities in middepth ranges of the seismogenic interface suggests the influence of along‐dip changes in other parameters, besides lithostatic pressure, that control shear resistance. Along‐strike segmentation of the Kurile Islands thrust zone may be a consequence of along‐strike alternation in the location of the largest asperities between the deep interface and the midinterface.

A Bayesian approach to the detection of temporal changes in P wave velocity

On the basis of Akaike's Bayesian information criterion (ABIC), a new method is proposed for detecting a temporal change in a seismic velocity in a source region. The method of joint hypocenter determination was modified in order to determine a seismic velocity in a source layer as a function of time together with hypocenters and station corrections. Arrival times of initial waves of shallow earthquakes in a small area are analyzed in this method. The smoothness of the estimated temporal variation in the velocity is guaranteed by the introduction of a prior distribution of the parameter. The hyperparameter of the prior distribution of the velocity, the reading error of arrival times, and the initial velocity in the source layer are chosen to minimize ABIC. This procedure was applied to the 1983 eastern Yamanashi M = 6.0 earthquake in central Japan. We analyzed P arrival times of 374 earthquakes observed at 12 stations in the network of the National Research Center for Disaster Prevention by dividing the whole period (from October 1981 to May 1987) into 12 six‐month subperiods. Calculating ABICs for different combinations of the three parameters above, we searched for the minimum value of ABIC and found two minima. The first one corresponds to a model of a constant velocity in time, and the other corresponds to a model of a variable velocity with 5% velocity change at maximum. However, since ABIC in the former is 10 smaller than that in the latter, the former constant velocity model is statistically more suitable than the latter. Furthermore, generating artificial data with the same reading errors as the actual data, we used computer simulation to examine the lower limit of the velocity change detectable for this data set. In conclusion, the velocity in the source region is 6.24±0.18 km/s, and the velocity change exceeding 6–7% at maximum did not exist during the 6 years before and after the M = 6.0 earthquake.

Rupture process of an extended earthquake sequence: Teleseismic analysis of the Chilean Earthquake of March 3, 1985

Broadband displacement and velocity records of P waves recorded at teleseismic distances are analyzed to determine the static and dynamic source parameters of the Chilean earthquake of March 3, 1985 (Ms 7.8) and seven large (mb > 5.6) aftershocks. Besides the usual parameters of depth, moment, and focal mechanism, the analysis provides estimates of radiated energy, associated stresses, and source dimension for seven of the eight modeled earthquakes. To insure that the parameters are describing the same physical aspect of the rupture process, the parameters for each shock are computed from the time window in which most of the seismic energy is radiated. The hypocenters of an additional 149 small foreshocks and aftershocks of 4.5 ≤ mb ≤ 5.6 are computed with the method of joint hypocenter determination. The spatial and temporal variation of source parameters and earthquake locations are used to infer a description of the rupture process before, during, and after the main shock. The main shock and modeled aftershocks occurred as thrust faulting on the interface between the Nazca and South American plates. A cluster of foreshocks in the 10 days preceding the main shock also involved thrust faulting on the plate interface. The main shock itself was a complex rupture consisting of three events. The first two events, denoted as ms1 and ms2, released minor amounts of energy, and they occurred on the periphery of the small zone defined by the foreshocks. The major release of energy occurred with the third event, denoted by MS, which nucleated downdip of ms1 and ms2. Within the time window of major energy release, rupture extended approximately 90 km to the south from the point at which MS nucleated. The size of the early aftershock zone, which far exceeds our inferred dimensions for the major shock, and the strong frequency dependence of scalar moment imply that substantial slow slip occurred on the plate interface that was not associated with major energy release. The dip of the seismically active interface increases landward from about 15°, at 20 km depth, to about 35°, at 40 km depth, where the seismogenic interface reaches its maximum observed depth. The vast majority of small and moderate aftershocks occurred in the shallower part of the interface, as did the foreshocks ms1 and ms2. It appears, however, that the asperities that control the rupture of the largest earthquakes are the asperities on the deeper interface. MS and the largest 1985 aftershock, that of April 9 (Ms 7.2), occurred on the deeper interface. The Chilean earthquake of July 9, 1971 (Ms 7.5), which occurred just north of the 1985 earthquake, also involved main shock rupture on the deeper interface and was followed by aftershocks on the shallower interface. Asperities on the deeper interface appear on average to be stronger, larger, and more uniform in size than asperities on the shallow interface. A greater average strength for deeper asperities is suggested by the observation that stress drops and apparent stresses of the main shock and the after‐shock of April 9 (Ms 7.2) are higher than those of the large shallow aftershocks and is consistent with arguments that the maximum friction on a fault zone should be near the base of the seismogenic zone. One deep interface aftershock with low‐stress drop and apparent stress may have been situated below the point of maximum friction on the interface. The inferred larger and more uniform sizes of the deeper asperities are suggested by the higher proportion of large to small shocks on the deeper interface. The downdip change in the asperity size distribution may reflect increased friction on the plate interace or the coalescence of small asperities into large asperities with increased subduction. The radiated energies of the main shock and large aftershocks that are computed directly from the broad bandwidth data are smaller than those predicted by standard empirical formulas by factors ranging from 4 to 20.

浅発地震の震源と震源域地震波速度の同時決定法およびその東海地域のサブクラスタル地震への適用

Extending the method of Joint Hypocenter Determination, we devised a new method to determine hypocenters, station corrections and a seismic velocity in the source region of shallow earthquakes simultaneously. Travel times of initial P waves of earthquakes occurring in a small area are analysed in this method.We have applied this method to subcrustal earthquakes with depths of 20-40km in several small regions in the Tokai district, and P velocities of 6.8±0.3km/s were obtained. Since this value is quite similar to that in the main layer of the oceanic crust, we can conclude that these subcrustal earthquakes in the Tokai district occur within the subducted oceanic lower crust of the Philippine Sea plate.