2011 年 63 巻 3 号 p. 107-121
Seismic reflection profiling using P-to-SV converted waves with 3-component seismometers has an advantage for examining deeper S-wave velocity profile compared with other methods such as SH-wave reflection profiling. In spite of this advantage, the reflection profiling with 3-component seismometers has not been popular because it requires three-times more channels than reflection profiling with vertical-component seismometers, which is popular for examining P-wave velocity profile. In our previous paper, we showed a possibility of examining S-wave velocity profile by analyzing P-to-SV reflected waves observed at one 3-component seismometer together with P-wave reflection profiling. Although the method in the previous paper has ambiguity to obtain S-wave velocity, the approach is attractive because one additional 3-component seismometer requires little additional cost. In this paper, we show the validity of the approach using another method. This new method is easy to apply, if P-wave velocity profile was already determined using vertical component reflection profiling. We picked P-to-SV reflected waves on observed radial-component records, and adjusted theoretical travel-time curves to the observed waves at one 3-component seismometer. P-wave velocity profile and depths of discontinuities were fixed to the result of P-wave reflection profiling. When the travel time was calculated, a ratio of P-wave velocity to S-wave velocity was assumed to be the same for all layers. The ratio for well-matched theoretical time was considered as the ratio of vertical travel times of S-wave to P-wave from the surface to the reflected layer. Shear-wave velocity for each layer was calculated from the ratio. Slowness of the P-to-SV reflected wave at large offsets depends on P-wave velocity and a depth of the reflector, but shows little dependence on S-wave velocity at large offsets. This characteristic makes it possible to separate the influence of the reflector depth and of S-wave velocity on travel-time curves of the reflected waves. Tests for simple horizontal layer models and dipping layer models showed that S-wave velocities were obtained within about 10% in error, and that obtained S-wave velocity was stable in errors of P-wave velocity and of an inclination of the layer. We applied this method to a previous seismic reflection survey with 3-component seismometers which was carried out at Fuchu-city, Tokyo. Previous study analyzed P-and S-wave velocity profiles to 2-km depth using P-and converted-wave reflection profiling. Another survey also examined P-and S-wave velocity profiles near the survey line using VSP method. We selected one station near the VSP well among the survey stations, and analyzed the data using the method. S-wave velocity profile that we obtained was consistent with the profiles of the previous studies to 2-km depth. Only one 3-component seismometer was necessary for this method with P-wave reflection profiling. Our results show that this method provides adequate shear-wave velocity profile with little additional cost.