We re-estimated fault parameters of the Fukushimaken-Toho-Oki earthquake occurred at 17 : 43 of November 5, 1938 to explain the re-analyzed tsunami records. In the re-analysis, we found a crustal subsidence of 3.3 cm at Onahama tide gauge station and a tsunami arrival time of 33 to 36 minutes after the origin time of the earthquake at Iwai tide gauge station. We improved the fault parameters to be consistent with the obtained results by comparing tsunami waveforms observed at six tide gauge stations with ones computed numerically. As the result, the southern limit of the fault shifted northwest by 60 km, length and dislocation decreased from 100 km to 60 km and from 230 to 69 cm, respectively, in comparison with those proposed by Abe (1977).

View full abstract

On December 14, 2004, an M 6.1 earthquake occurred in the northwestern part of Hokkaido, Japan. We installed nine temporal seismic stations around the source area immediately after the occurrence and had continued the observation for about two months after the main shock. We determined 823 hypocenters of aftershocks and one dimensional P-wave velocity structure model from the travel time data. It is found that aftershocks are clearly distributed on an eastward dipping plane with a dip angle of about 25 degrees. This plane agrees well with one of the nodal planes of the focal mechanism determined by P-wave first motions in this study. Next, we investigated a position of the main shock relative to the aftershock distribution. The station correction values for permanent stations were determined from the data during the period of the temporary observation. Estimated main shock hypocenter was confirmed to be situated on the aftershock plane. Finally, we estimated three-dimensional P-wave velocity structure based on the temporary travel time data. Relocated aftershocks using the three dimensional velocity structure were mainly distributed along the boundary between the high and low velocity zones.

View full abstract

The deformation of the brittle upper crust is primarily accompanied by displacements on faults. Nevertheless, it has often been found that continuum fluid models, usually based on a non-Newtonian viscosity, are applicable. I provide a new view of rheology for upper crustal deformation, using the fiber-bundle model, a model of damage mechanics. In my analyses, a yield stress σ_y was introduced. Below this stress, I assumed that no fiber failure occurs, and the crust behaves elastically without earthquakes. Above the yield stress, I hypothesized that the fibers begin to fail and a failed fiber is replaced by a new fiber. This replacement is analogous to an earthquake rupture. The deformation above the stress σ_y under a constant strain rate ε for a suffcient time can be modeled as an equation similar to that used for a non-Newtonian viscous flow. There is a power-law relation between mean stress in the crust σ and the strain rate ε with ε∝(σ — σ_y)ⁿ where n is constant. I derived the modified Omori’s law (Omori-Utsu’s law) for aftershock decay using a viscoelastic version of my model and got good agreement with observations taking n=5-13. I point out that further development of my model can contribute to fully understanding the upper crustal deformation.

View full abstract

A P-phase with smaller amplitude (P1)preceding another P-phase with larger amplitude (P2)has often been observed for major seismic events, which indicates a small-scale rupture, occurring around the rupture initiation point, precedes large-scale rupture. These two phases as well as the corresponding two S-phases (S1 and S2)were observed on seismograms of the 2005 West off Fukuoka Prefecture Earthquake (M_JMA=7.0) that occurred on March 20, 2005. In order to investigate the initial rupture process, we made the following analyses. We checked whether these double phases were generated from different but adjacent ruptures along the fault. We first picked the four phases (P1, P2, S1 and S2) using AIC for AR model fitting from the seismograms recorded by Digital Strong-Motion Seismograph Network and High Sensitivity Seismograph Network that deployed by NIED. We then relocated the corresponding hypocenters by means of the master event hypocenter relocation technique. The distance and time interval between the initial and main ruptures were estimated to be about 3.44 km and 3.38 s, respectively. The relationship between the duration of the initial rupture and the magnitude is consistent with the empirical scaling relation in preceding studies. The estimated apparent “average rupture propagation speed” of this initial rupture was 1.02 km/s, which is lower than the average rupture velocity in usual crustal events. The moment magnitude of the initial rupture portion was estimated to be around 4.7. We estimated the source time functions for the initial rupture of the main shock and two aftershocks of similar size using empirical Green’s functions. The moment of the initial rupture of the main shock gradually increased and had long duration, which is quite different from the pulse-shaped source time functions with shorter duration of the two aftershocks. This result, combined with the empirical scaling relation mentioned above, implies that the initial rupture is different from those of earthquakes of the same size and related to the final rupture size of the main shock.

View full abstract

A waveform inversion was conducted for the rupture process of the 2005 West Off Fukuoka Prefecture erthquake (M_J 7.0) by using aftershock records as Green’s functions. Paying attention to the similarity of the group delay time between the mainshock and aftershock records, two aftershocks were selected to be used in the inversion. As a result of the inversion, we have obtained a rupture model that can explain mainshock records in the frequency range from 0.2 to 2 Hz. The model is composed of two primary asperities; one is located in the southeast part of the fault plane, close to the Genkai Island, and the other is located in the northwest part of the fault plane, about 8 km from the hypocenter. Contributions from the former asperity is significant for the ground motions in the city of Fukuoka, which is located southeast of the source region, while contributions from the latter asperity is significant for the ground motions at Tsushima Island, which is located northwest of the source region.

View full abstract