The focal mechanism of the 1975 central Oita, Japan, earthquake (M=6.4) is reexamined, and the rupture process is investigated by comparing synthetic and observed P-waveform data of WWSSN long-period seismograms. The obtained focal mechanism is not a normal fault type as was previously reported, but a strike-slip fault type with dip-slip component. The tensional axis of this solution is consistent with the tectonic stress field in this region. The rupture time, the seismic moment, the fault displacement, and the stress drop are estimated to be 6.7 s, 2.2 x 1025 dyn·cm, 85cm, and 78bar, respectively. However, these only show a general or average feature of this earthquake. Based on a complicated feature of observed seismograms, we found that this event is a multiple shock and has a shallow initial rupture with short source duration. A distinct later phase of seismograms suggests a change of source mechanism during the rupture process, which is consistent with the distribution of the mapped late Quaternary faults in the focal area. The trend and sense of the fault of this earthquake and active faults around there, and strain field of this region agree with the pattern for oblique rifting, proposed by WITHJACK and JAMISON in 1986. It is suggested that oblique rifting, which appears to be related to a right-lateral displacement of the Median Tectonic Line lying east of and subparallel to the inferred rift axis, takes place in this region. The relative displacement between opposite sides of the rift is estimated to be directed in NESW, consistent with the geodetic data.

View full abstract

The regionalization method has been applied to the attenuation coefficients of the vertical components of Rayleigh waves for the Southern Atlantic-Indian Ocean to obtain the attenuation across the ridge region. The obtained ridge attenuation coefficients are similar to those for the North Atlantic ridge and much lower than the values for the East Pacific Rise. Inversion theory, in its stochastic form, shows that the low-Q zone under the ridge sems to be located between 50 km and 200 km depth, with the maximum Q^-1 value (≅15×10^-3) at a depth of about 140 km.
Comparison of the ridge model in this study with models for the North Atlantic ridge and East Pacific Rise, obtained by others, indicates that the attenuation coefficients for the study region are similar to those for the North Atlantic ridge, and much weaker than those for the East Pacific Rise. Since the spreading rate in the southern part of the Atlantic-Indian Ocean is similar to that in the North Atlantic, and lower than the corresponding rate in the Pacific, it seems that there is a close relation between spreading rates and low-Q regions under the ridges, with the Q values being higher in the region where the spreading rates are lower.
The above results suggest that the creep rate under the ridge of the Southern Atlantic-Indian Ocean may be similar to that for the Atlantic, and much lower than for the Pacific Ocean for regions of comparable ages.

View full abstract

The problem of a two-dimensional dynamical dislocation in an isotropic, homogeneous, unbounded, elastic medium is discussed at length. Beginning with the known solution for a line force, the field due to a displacement dislocation is obtained through the Volterra relation. It is shown that the field due to a dip-slip source of arbitrary dip can be expressed in terms of the field due to a vertical dip-slip source and that due to a 45° dip-slip source. The representations of the two-dimensional dynamical sources are obtained in terms of the vector solutions of the Navier equation and in terms of the source potentials. The potential representation is used to get the field due to an arbitrary line source buried in a uniform half-space.

View full abstract