Using a method suggested by Yanovskaya, we obtained Rayleigh wave group velocities with a resolution of 1.0o ×1.0o in a period range between 10 and 80 s in and around the Korean peninsula. Both regional and distant earthquake data sets were used together in analysis of group velocities. The resolution of the group velocity maps has been remarkably enhanced by the method, especially in the sparse/non-station region in the northern Korean peninsula. Some qualitative geophysical information was inferred from the group velocity maps. In the East Sea, the slow group velocities at periods longer than 40 s suggest the existence of an oceanic lithosphere at depths of 50-70 km, assuming 4 km/s of S wave velocity at a period of 40 s. On the other hand, a thick lithosphere can be inferred in the continental area from the fast group velocities at periods longer than 50 s. For most periods, the group velocities change rapidly over a short distance of about 200 km across the eastern coast of Korean peninsula, which may suggest a rapid change in the thickness of lithosphere in this area.

An attenuation profile of young uppermost oceanic crust was obtained using ocean bottom seismometers and explosive sources detonated near the ocean bottom. The quality factor (Q) increased from 6 at the sea floor to 50 at a depth of 500 m. The mechanisms of high attenuation obtained in this experiments was inferred by examining several attenuation mechanisms such as frictional dissipation, viscous shear flow, liquid squirt mechanism and scattering by small pores and large cracks. Comparison of the attenuation value in this experiment and those from other experiments was also used to clarify the mechanism of the high attenuation. Of those mechanisms mentioned above, frictional dissipation, viscous shear flow and scattering do not seem to be the cause of the high attenuation in the uppermost oceanic crust where many small pores and large scale cracks are developed. The most viable mechanism in the young uppermost oceanic crust seems to be liquid squirt mechanism.

In most marine seismic refraction studies using shots detonated near the sea surface, the location of receivers and shots are usually assumed to be located at the point of deployment and the shot times are accurately known. In the case of bottom shots and receivers, however, the above assumption can be led to big errors in the shot and receiver positions. The errors in the shot and receiver positions can cause severe distort in the interpretation of deep ocean refraction studies. In this study, an inversion algorithm has been developed to reduce the errors in the shot and receiver position. The inversion method has been used to obtain the shot and receiver positions and the shot origin times of a deep ocean refraction seismic experiment which was performed near the Juan de Fuca ridge in 1985. The comparison of two seismic record sections, one using the initial shot and receiver positions and the other using the shot and receiver positions obtained by the inversion algorithm, shows that the seismic signals emerges much more gradually in the latter. This indicated that the posotions of shots and receivers should be relocated using some inversion methods for better interpretation of deep marine seismic refraction studies.