There are several European historical materials on Tensho Earthquake, which occurred on Jan. 18, 1586 in the south-western part of Chubu district. The through research of those materials revealed that all the information of Tensho earthquake was originated from letters written by L. Frois in October 1586 at Shimonoseki Port, Yamaguchi Prefecture as the annual report of missionary in Japan. Since a letter was hand-copied, edited, and translated to Italian or Latin from Portuguese in the late 16th century, there derived some versions of descriptions on damage of the earthquake. The letter in the Portuguese version published in Evora, Portugal in 1589, and the remained copies of the manuscript of "Japanese History" by Frois are only two reliable sources. From those credible European materials, it is apparent that there was the common talk in Kyoto after Tensho earthquake of Tsunami damage along the northwestern coast of Japan. However, it is no more than a rumor.
Precise seabottom topography on the outer slope of the Japan Trench was analyzed from swath bathymetric data. Grid bathymetric data with spacing of 0.005 degrees (555 m by 438 m at 38 degree north) was synthesized from full covered multi-beam bathymetries on the outer slope of Japan Trench. Topographic maps and profiles were drawn by the grid bathymetry. The topographic profiles are in a direction of 296 degrees which is parallel to the plate motion of the Pacific plate at the area. Fault systems characterized by horst and graben structure are developed on the outer slope of Japan Trench. They are caused on the upper surface of the Pacific plate by downward bending of the plate subduction. The fault systems developed on the western margin of the outer rise, are emphasized with evolving near by the trench axis. West throw faults are dominant in every profile. On the assumption that the Pacific plate is homogeneous in terms of physical property and moves constantly, an average displacement rate of the fault system can be estimated from the distance between the fault system and the position where the fault system is initially recognized on the profile. Strain rates close to the trench axis are estimated as 9×10-8 yr (at northern Japan Trench), 7×10-8/yr (at southern Japan Trench) and 12×10-8/yr (at northern most of Izu-Ogasawara Trench). These are almost same as the extension rate of the crustal strain in the central mountain region in the Chubu district, those are highest in Japan.
We studied the differences between the velocities of seismic waves and elastic waves in rock core samples to understand the spatial-scale effect of heterogeneous contents. The wave velocities determined from seismic observations and rock property tests of rock core sample are different in most cases. These variations are caused by the difference in the spatial scales of each measurement (i.e., the differences in wave lengths), since the waves refer to the corresponding scales and velocities of heterogeneity in the propagating medium. In order to estimate the seismic wave velocities with very short baselines (∼100 m) by "multiple blasting", we analyzed the arrival times of initial motion, and conducted waveform correlation analyses to transversal pairs for S-wave. Under the assumption that the seismic wave slowness (1/velocity) is an average value corresponding to the medium comprising homogenous fresh rocks at a small scale and the other heterogeneous components such as cracks at larger scale, we compared the seismic wave velocities with the elastic wave velocities that were measured using the rock core samples taken near the seismometers. Our results verified that parameters related to the medium heterogeneity such as crack density, crack shape, and the water content of the medium can be estimated from velocities comparisons.
The 2011 off the Pacific coast of Tohoku earthquake was the greatest earthquake in Japan with the moment magnitude of 9.0. However, the first report of JMA (Japan Meteorological Agency) magnitude estimated 3 minutes after the earthquake initiation was 7.9. Furthermore, the moment magnitude which should be calculated about 15 minutes after the earthquake initiation was not calculated due to the saturation of broad-band seismometers. As a result, the tsunami height was underestimated causing loss of many lives unfortunately. Now southwestern Japan faces the possibility of the great earthquake of Nankai Trough, so it is extremely important to develop a quick estimation system of unsaturated magnitude. In this study, we propose a method to calculate the permanent displacement accurately based on an acceleration record and a scheme to estimate moment magnitude quickly. The moment magnitude estimation method described in this study is based on the theoretical relation between coseismic displacement and hypocentral distance. By applying the method to the 2011 off the Pacific coast of Tohoku earthquake and 2003 Tokachi-Oki earthquake, the moment magnitude was estimated at 8.9 and 8.5, respectively. These values are close enough to the other estimations. In addition, by applying the method to the scenario earthquake at Nankai Trough, the correct value of moment magnitude was estimated. Finally, we propose using the equivalent hypocentral distance for large earthquake in order to consider the spatial extent of the seismic fault. Although real-time detection of the spatial extent of the fault is required, the moment magnitude can be estimated more rationally by this method.
A new simplified source model is proposed to explain strong ground motions from a huge subduction earthquake. The proposed model is simpler, and involves less model parameters, than the conventional characterized source model, which itself is a simplified expression of actual earthquake source. In the proposed model, the spacio-temporal distribution of slip within a subevent is not modeled. Instead, the source spectrum associated with the rupture of a subevent is modeled and it is assumed to follow the omega-square model. By multiplying the source spectrum with the path effect and the site amplification factor, the Fourier amplitude at a target site can be obtained. Then, combining it with Fourier phase characteristics of a smaller event, the time history of strong ground motions from the subevent can be calculated. Finally, by summing up contributions from the subevents, strong ground motions from the entire rupture can be obtained. The source model consists of six parameters for each subevent, namely, longitude, latitude, depth, rupture time, seismic moment and corner frequency of the subevent. Finite size of the subevent can be taken into account in the model, because the corner frequency of the subevent is included in the model, which is inversely proportional to the length of the subevent. Thus, the proposed model is referred to as the "pseudo point-source model". To examine the applicability of the model, a pseudo point-source model was developed for the 2011 off the Pacific coast of Tohoku earthquake. The model comprises nine subevents, located off Miyagi Prefecture through Ibaraki Prefecture. The velocity waveforms (0.2-1 Hz) and the Fourier spectra (0.2-10 Hz) at 15 sites calculated with the pseudo point-source model agree well with the observed ones, indicating the applicability of the model. Then the results were compared with the results of a super-asperity model of the same earthquake, which can be considered as an example of characterized source models. Although the pseudo point-source model involves much less model parameters than the super-asperity model, the errors associated with the former model were comparable to those for the latter model for velocity waveforms. Furthermore, the errors associated with the former model were much smaller than those for the latter model for Fourier spectra. These evidences indicate the usefulness of the pseudo point-source model.
We have investigated the inland seismic activity induced by the 2011 Off the Pacific coast of Tohoku (Tohoku-oki) Earthquake in the northern part of Tohoku district, using JMA catalog and newly determined focal mechanism solutions. The seismicity is quite high in the Akita prefecture, forming newly activated clusters. The cluster locations are complementary for the periods before and after the Tohoku-oki Earthquake. A stress tensor inversion using focal mechanism data indicates that the stress field has changed from reverse-faulting regime to strike-slip regime, with a counter-clockwise rotation of the maximum principal stress axis and the replacement of the other two principal axes. This change is qualitatively explained by the weakened E-W compressional stress due to megathrust faulting of the Tohoku-oki Earthquake. Thus the new stress field in the investigated area is unfavorable to the preexisting fault planes of reverse faulting, which brought the complementary seismic activity. Among the three active clusters in the Akita prefecture, the one to the north of Moriyoshi volcano is interesting, because the swarm-like activity forms a volumetric source with a dimension of about 3 km. Considering a possible existence of crustal fluid suggested by a reflected phase, delayed beginning of seismic activity about 2 month from the Tohoku-oki Earthquake, and the migration of seismic activity, the induced seismic activity in the area may be related to a response of crustal fluid to the coseismic stress change. Detailed investigation of the result of stress tensor inversion reveals the existence of local stress field superposed on the regional field represented by the average stress tensor. Inferred local stress field exists in the Tsugaru Strait area, southern part of Akita prefecture, and Kitakami Mountains.
Seismic activity near the Yake-dake (Mt. Yake) volcano in the Hida mountain range that took place immediate after the 2011 off the Pacific Coast of Tohoku earthquake was investigated. It initiated about ten minutes after the mainshock of the Tohoku earthquake and lasted for about one month. At the beginning, two active swarms were observed. One is at the northern flank of the Yake-dake volcano and the other is located between Yake-dake and Mt. Norikura volcanoes. The latter activity decreased by March 20, and the former activity lasted until early April. It includes two M≥4.5 earthquakes and we could locate more than 9,600 events in the study area during March and April. We mainly focused on the activity near the Yake-dake volcano in this paper. Near the Yake-dake volcano, seismic activity began with M4.7 (JMA) earthquake at 14:57 JST on March 11. This M4.7 event is located 3 km north to the volcano and seismicity increased between the summit of the Yake-dake volcano and the hypocenter. On March 21, an M4.8 (JMA) event took place at 13:15 JST also at 3 km north to the volcano. After this second M≥4.5 earthquake, seismic activity migrated to the north about 1 km. Focal mechanism solutions of these swarm earthquakes show NW-SE compression stress field, which coincides with regional stress field indicated by previous studies. No temporal changes of focal mechanisms are shown during March and April, which probably indicates no magmatic activity such as dyke intrusion related to the Yake-dake volcano took place in this time period.
A great earthquake with a moment magnitude of 9.0 occurred on March 11, 2011, rupturing the plate boundary off the Pacific coast of the northeastern Japan. Large displacements induced by this great earthquake were observed by the GPS Earth Observation Network system (GEONET) over the entire Japanese Islands, and were also detected by the seafloor geodetic observation (SGO) along the Japan Trench. The maximum horizontal displacement observed by the GEONET reaches 5.4 m at the tip of the Oshika Peninsula, and exceeds 30 m at ocean bottom detected by SGO. The subsidence up to 1.1 m was observed by the land GPS along the Pacific coast, and the uplift up to 5 m was detected by the SGO near the epicenter. Based on these geodetic data, the area of coseismic slip is estimated to be ∼450 km long and ∼200 km wide along the Japan Trench with a maximum slip of larger than 50 m near the trench. Because no M≥8.5 earthquake was identified there during the historical period, how strains are accumulated and released is the most essential problem to be solved. The shallower portion of the plate interface has been considered to be decoupled, but such an idea needs to be revised after the Tohoku earthquake. The postseismic deformation exceeds 70 cm for 180 days after the main shock. The coseismic subsidence areas begin to uplift except for the area in the Iwate prefecture. Estimated afterslip exceeds 2.0 m and its area extends to the west, south and north of the coseismic slip area with a moment magnitude of 8.5 for 180 days. The area of the afterslip was extended westward and reaches a depth of approximately 90 km of the subducting plate. Northern and southern edges of the area of afterslip seem to be limited by the source region of the 1994 Sanriku-Haruka-oki earthquake and the north limit of the overriding Philippine Sea plates, respectively. Although the afterslip decayed rapidly with time, the area of afterslip is estimated to be ∼400 km long and ∼180 km wide with a maximum slip of larger than 0.15 m during the period from August 8 to September 7, 2011 that corresponds to a moment magnitude of 7.6. It is definitely important that we continue the observation of the postseismic deformation and carefully monitor the temporal and spatial change of the afterslip for the purpose of the mitigation of hazard due to incoming huge interplate earthquake and understanding the earthquake generation cycle.
The rupture of the 2011 off the Pacific coast of Tohoku Earthquake started in the source area called ‘Near the Japan Trench off southern Sanriku coast,’ where a probability of earthquake occurrence had been estimated as high, and simultaneous rupture of two source areas took place as had been forecasted as most likely. However, successively, a huge slip amounting to some 50 m took place on the shallow interplate boundary near the oceanic trench and caused rupture extensions towards south and north. The final size of the event was Mw9.0 that had not been forecasted. The failure in forecast comes from ignorance of a strong interplate coupling near the oceanic trench. Despite the failure in forecasting the size of the event, the long-term forecast provided useful information for tsunami disaster management: A tsunami earthquake as large as the 1896 one has a chance of occurring anywhere along the Japan Trench. Necessary measures against tsunami complying with this forecast would have greatly reduced the tsunami fatality and might have avoided the serious accident at Fukushima No.1 nuclear power plant. Idee fixe: a large earthquake shall not occur where no large quake has been recorded, lead to the government’s administrative misjudgment.