The slip distribution of the 2003 Tokachi-oki earthquake is estimated from the 11 tsunami waveforms recorded at 9 tide gauges and 2 ocean bottom tsunami-meters (pressure gauges) off Kamaishi, Tohoku, using the waveform inversion technique. The plate interface of the source area is divided into 48 subfaults. Smoothness constrains are applied to stabilize the inversion. Large slip amounts, more than 3m, are estimated on the subfaults located east of Hiroo. The total seismic moment is 1.1×10²¹ Nm (M_w8.0). The slip distribution estimated from the tsunami waveform inversion is similar to the slip distribution estimated by Yamanaka and Kikuchi (2003) using the teleseismic body waves.

View full abstract

The 2003 Tokachi-oki, Japan, earthquake generated sloshing of liquid in large oil storage tanks with capacities over 1, 000m³ in Hokkaido and caused damage to many of them. The maximum sloshing wave heights of about 3m were observed at two tanks with natural periods of the fundamental mode of sloshing (T_s) of 7s and 5.9s in the Tomakomai western port area and at a tank (T_s=5s) in the city of Ishikari. Damage occurred in 190 large tanks in the southern part of Hokkaido. In and around the city of Tomakomai, 170 large tanks have been damaged and this corresponds to a high damage ratio of 58%. Serious damage occurred in nine tanks and eight of them concentrated in the Tomakomai western port area, where fire broke out from two tanks and floating roofs sank into the oil in seven tanks. One of the two fires was the ring fire that lasted for about 7 hours, while the other extended to the open top fire and it lasted for about 44 hours. It was long-period strong ground motion with periods from several seconds to around 10s that generated sloshing in the large tanks. In the Yufutsu plain embracing the Tomakomai area and the Ishikari plain, the shaking with a period of 7s was stronger than that in the plains in the eastern part of Hokkaido even though the latter were closer to the source region. In the plains of Yufutsu and Ishikari, the long-period strong ground motion was not confined to the Tomakomai area but distributed widely over the plains with the strongest motion observed in the area along the north-to-south axis connecting the center of the Yufutsu plain with the eastern part of the Ishikari plain. This long-period strong ground motion is presumed to be excited by the sediments reaching a thickness of 2 to 3km inside the plains. Around the Tomakomai western port, the long-lasting long-period shaking was observed with velocity responses with a damping factor of 1% keeping over 1.5m/s in a period range between 3.7 and 8.3s and with the peak value of 2.8m/s at a period of 4.8s. This peak value is about 2.5 times as large as the ground motion level that corresponds to a code established for large oil storage tanks by the Japanese fire laws. In the Tomakomai western port area, no recordings for the last three decades are larger than the ground motion level at periods of several seconds observed during the 2003 Tokachi-oki earthquake. Comparisons between the observed maximum sloshing wave heights and the observed ground motions indicate that the maximum sloshing wave heights scale with the velocity responses at T_s's. From the sloshing response analyses for the tanks in the Tomakomai area based on the observed ground motions, we found that the sloshing with the maximum wave heights of 3m and 1.3m occurred in the seriously damaged tanks with capacities of 30, 000 to 40, 000m³ (T_s-7 to 8s) and about 110, 000m³ (T_s-12s), respectively.

View full abstract

The 2003 Tokachi-oki earthquake (M_JMA=8.0) is a large inter-plate earthquake which occurred on September 26th, 2003. The strain seismograms caused by the earthquake were successfully recorded by our borehole strainmeter array about 1, 000km away from the epicenter.
Strain seismograms recorded by both borehole strainmeters and extensometers were compared at the range of frequency of seismic waves (0.001-1HZ). We obtained the result that the borehole instruments have good responses as well as extensometers. Vertical components of strain seismograms and broadband seismograms were compared, because we developed those strainmeters in order to use as strain seismometers. It was clarified that ground motions with very long-period have been exactly observed by no broadband seismometers but strainmeters. Then, we obtained the conclusion that our borehole strainmeters are more sensitive than broadband seismometers in very low frequency range (f<0.003Hz). We think that the strain seismometer is an effective instrument of recording long-period ground motion and it may bring new information in the field of seismology.

View full abstract

The Tokachi-oki earthquake (M_JMA8.0) occurred on September 26, 2003 (JST) off southeastern Hokkaido, Japan. We investigate the seismic activity before and after the mainshock using hypocenter catalogues produced by Japan Meteorological Agency and Hokkaido University. A decline of seismic activity and a seismic gap in the asperity of the 2003 earthquake had been clearly recognized from the early 1990's to just the mainshock faulting. Similar low seismicity had been observed before the former 1952 Tokachi-oki earthquake (M_JMA 8.2). These facts may reflect that the plate boundary fixed loosely has become tight coupling during one seismic cycle. A relatively low seismicity patch during the interseismic phase corresponds to the asperity. This suggests that interseismic seismic activity is strongly controlled by existence of the asperity. The aftershock region of the 2003 earthquake is approximately 160km by 160km, which is slightly smaller than that of the 1952 earthquake. This feature is in good agreement with the slip distributions of the 1952 and 2003 earthquakes estimated from tsumani waveform inversions. Large aftershocks are located out of the asperity, which may reflect the stress status in and around the asperity. Remarkable triggered seismic activity has begun just after the mainshock along the volcanic front in the eastern Hokkaido.

View full abstract

We inferred minutely the geometry of the upper surface of the seismic Philippine Sea (PHS) slab beneath the region from Ise Bay to western Shikoku, southwest Japan, based on hypocentral distribution. Despite the importance of the slab geometry for the subduction zone seismotectonics as well as for the assessment and forecasting of destructive interplate and slab earthquakes, the shape of the PHS slab has been obscure and controversial. We selected well-determined hypocenters during the period from Oct. 1997 through Dec. 2002 from the integrated hypocenter database prepared by the Japan Meteorological Agency. At first, we composed depth data set of the slab upper surface from vertical cross sections of hypocentral distribution along longitude and latitude of every 0.1 degree, and drew first-trial isodepth contours of the upper surface of the PHS slab assuming that the upper boundary of slab earthquake distribution roughly coincides with the slab upper surface. Then, in many areas we made vertical cross sections of hypocenter distribution in many directions including inferred dip-direction and strike of the slab, and revised isodepth contours. The main results are as follows: (1) In the region to the east of Lake Biwa, northwestward gently-dipping PHS slab reaches around the depth of 35km, in which disastrous slab earthquakes presumably took place. (2) To the northeast it continues to the slab beneath the central Tokai district, but to the southwest it is obscure whether it continues to the slab beneath the eastern Kii Peninsula or it is separated from the latter. (3) The slab beneath the middle part of the Kii Peninsula may be torn into two parts, with the southwestern part underlying the northeastern part. (4) There are double seismic zones beneath the southern Kii Peninsula. (5) In the west part of the Kii Peninsula no slab earthquake occurs in deeper part, whereas just in this area a remarkable shallow crustal swarm earthquake activity exists. (6) Beneath Shikoku the slab is being subducted with gentle dip angle and forms a broad ridge. The slab looks continuous from Shikoku to Kyusyu. Recently the PHS slab geometry has been investigated by seismic refraction/reflection profilings and receiver function analyses. Those results are shallower than our result in eastern Shikoku suggesting that slab earthquakes in this region occur in the oceanic mantle. We relocated slab earthquakes in this region using a velocity structure model consistent with the result of seismic profiling and found that the shallower main activity lays in the slab mantle of the assumed model with deeper remarkable events beneath it. In order to clarify the true slab geometry it is essentially important to know in which part of the slab, oceanic crust or mantle, slab earthquakes are actually taking place in each region.

View full abstract

An inversion analysis has been developed to evaluate short-period seismic wave radiation zones on an earthquake fault plane using seismic intensity data. It is a useful method for historical earthquakes, for which neither strong motion nor tsunami data have been observed by instruments. Since the accuracy of the present method is verified by using ground motion waveforms synthesized by the stochastic Green's function method, it is concluded that the effects of directivity of the fault rupture process and the radiation pattern on seismic intensity distribution can be neglected in the frequency range that is effective for seismic intensity. The present method is applied to great earthquakes that have occurred since 300 years ago in the Nankai Trough seismogenic zone in southwestern Japan, where the Philippine Sea plate is subducting beneath the Eurasian plate. The reliability of their solutions is discussed based on a sensitivity analysis. It is noted that short-period seismic wave radiation zones are often adjacent to large slip areas, termed asperities, but don't always overlap them. Furthermore, they are sometimes located where fault rupture has stopped. Short-period seismic waves are not always radiated from the same zone. For example, fault zones in the Enshu-Nada Sea, the Kumano-Nada Sea, the Kii Channel and the Kochi seacoast have radiated short-period seismic waves at every event, but whether or not short-period seismic waves were radiated in the interior of Suruga Bay, off the Shiono Cape and off the Muroto Cape varied with the events. It is also concluded that crustal structures such as subducted seamounts and subducted ridges take an important role of radiation of short-period seismic waves.

View full abstract

A large intraslab earthquake (M_JMA 7.1) with the focal depth of about 70km occurred in the Pacific slab below off Miyagi Prefecture, northeastern Japan on May 26, 2003. This earthquake source brought strong high-frequency ground motion to the near-source region. At first, a source model is estimated by the forward modeling through strong motion simulations using the empirical Green's function method. Strong ground motions are simulated in the broadband frequency range (0.3-10Hz). Four borehole station data of KiK-net are used. A simple source model having several rectangular “strong motion generation areas” (Miyake et al., 2003) is assumed. The fittings of acceleration, velocity, and displacement waveforms are fairly well. Next, strong ground motion simulation is conducted at all the stations of K-NET and KiK-net (surface) within about 100km distance from the epicenter to validate the obtained source model. The spatial distributions of peak horizontal velocity and acceleration are well reproduced, which gives an evidence that our obtained source model can describe principal characteristics of the actual source process of this earthquake. However, synthetic peak horizontal accelerations are much larger than observed ones at several stations, which may be caused by nonlinear effect of the superficial layers during the mainshock. The obtained source model contains three strong motion generation areas. The spatial distribution of the strong motion generation areas is almost corresponding to the large slip regions from the kinematic waveform inversion analysis. The total size of the strong motion generation areas against a given seismic moment follows the empirical relationship with source-depth for intraslab earthquakes by Asano et al. (2003). Each strong motion generation area of this source model has 105MPa of stress drop. The varieties and uncertainties in estimation of the stress drop value on strong motion generation areas are also discussed.

View full abstract

A large (M7.1) intraslab earthquake occurred on May 26, 2003, on the upper plane of the double planed deep seismic zone off Miyagi Prefecture. No intraslab earthquakes with magnitudes greater than 7 have occurred since 1926 under the land area of Tohoku, northeastern (NE) Japan. We relocated hypocenters of the 2003 off Miyagi earthquake and its aftershocks to compare the background seismicity prior to this earthquake. We obtained relative earthquake locations using the Double-Difference method [Waldhauser and Ellsworth (2000)] and relative earthquake arrival times measured by waveform cross spectrum. Relocated aftershocks are distributed along a plane steeply dipping to WNW, which coincides with one of nodal planes of the focal mechanism of the main shock. Its dip angle slightly changes near the main shock hypocenter. The hypocenter of the main shock seems to be located near the Moho and aftershocks extend both into the crust and into the mantle of the slab. Many earthquakes occurred near the hypocenter of the main shock before the 2003 event. These events are distributed near the Moho and/or within the mantle of the slab. High background upper-plane seismicity within the mantle may correspond to the area to which the rupture within the crust is easier to extend, causing large intraslab events. We also relocated intraslab earthquakes in the northeastern part of Tohoku. There exist few areas where background upper-plane seismicity extends to the mantle of the slab such as the focal area of the 2003 off Miyagi earthquake. In previous studies, it has been suggested that earthquakes in the upper seismic plane of the double seismic zone in Tohoku are distributed uniformly in space, whereas those in the lower seismic plane are not. However, the results in this study suggest that the spatial distribution of upper plane events is not uniform, either. If intermediate earthquakes are caused by dehydration of hydrated crust minerals, the present result suggests that hydrated minerals are distributed unevenly in the crust.

View full abstract

The Kiotoshi fault is a NNW-trending reverse active fault of a length of 3km in south-eastern side of the Ono basin, Fukui Prefecture. This reverse fault connects NE-trending right-lateral left-stepping faults, and forms a linear boundary between mountainous land and the basin. To make clear the paleoseismic activity of the fault, the fault morphology, the feature of a fault outcrop in a quarry site, and the ¹⁴C dating of buried soil and charcoal samples from the outcrop were examined.
The fault displaced the basement rock and overlying fluvial and fan sediments, and makes a gentle step 0.8m high on the fan surface. The amount of vertical displacement of the basement rock is more than 15m. The displacements of the overlying sediments are from 0.5 to 7m or more along the main fault. Buried soil layers, presumably formed by rapid accumulation of fan deposits after faulting events, are present at three horizons in the fan sediments. The calibrated ¹⁴C ages of the buried soils are 2, 850-1, 280 cal yBP, 5, 590-4, 620 cal yBP, 11, 100-8, 810 cal yBP, and the net slips calculated from the vertical displacement of the undeformed buried soil layers are 3.6m, 6.9m and 19 to 20m from the upper layer. The average slip rate of the fault, calculated from the age and the net slip of the lowermost buried soil layer, is 1.7-2.3m/1000 yrs, i. e. the activity of the fault is classified into A class.
The Kiotoshi fault is a reverse fault accompanying the NE-trending right-lateral left-stepping faults, but the activities (average slip rate) of these strike-slip faults were considered to be low (class B to C). We need, therefore, to investigate the activities of the surrounding strike-slip faults in order to clarify the late Quaternary tectonic activities of this area.

View full abstract

To evaluate a reliability of GPS vertical data we mapped vertical crustal deformation field of Japan using continuous GPS measurements with a nationwide dense network (GEONET) of the Geographical Survey Institute during the period from 1996 to 2003. We confirmed that the GPS vertical deformation field reasonably agrees with those derived from leveling, tidal record and geomorphologic analysis. The conformity with the other data sets suggests a dependability of the GPS vertical results. The GPS data illustrate details of a spatial pattern of the vertical deformation field and manifest their usefulness when applied to constrain tectonic models. A subsidence along the pacific coast of the southeastern Hokkaido propounds a possibility of a downdip extension of a plate coupling reaching to the depth of about 80km. Uplift around Hidaka mountains in the central Hokkaido suggests a present-day mountain building process at least during the interseismic period. An apparent subsidence found in the central mountainous region of the central Honshu island contrasts sharply with the presumed uplift through Quaternary inferred from geomorphologic analyses. Vertical deformations along the Nankai trough in the southwestern Japan can be attributed to an elastic deformation due to a dragging of the subducting Philippine Sea plate. The GPS result confirms a coupling of plates and a resultant strain accumulation in the Tokai region. Those results demonstrate the usefulness of the GPS vertical data and encourage us in further applications in the studies to understand ongoing tectonic processes in Japan.

View full abstract