9:30 AM - 9:45 AM
[STT41-03] Comparison of Green's Functions of Shallow Ground by Surface Wave Survey and Seismic Interferometry.
Keywords:Surface wave survey, Artificial excitation, Subsurface structure, Wave propagation, Microtremor
Introduction
Knowledge of the subsurface structure is important for earthquake disaster prevention and building seismic safety. There are two methods of investigation that do not involve excavation: artificial excitation and passive microtremor record processing. One of the former methods is surface wave survey, e.g. Hayashi et al. (2001), and the latter method is to obtain pseudo-shot records by applying seismic wave interferometry to microtremor records, as described by Takahashi et al (2022). While the theory of shallow Rayleigh wave survey is well established and widely used, the method of obtaining pseudo-shot records from microtremor records is still in the experimental stage and has not been compared with surface wave survey. In this study, the Green's function and dispersion curves were calculated by applying both methods to the same survey line in order to compare and verify the two methods.
Observation and Experiment
As shown in Fig. 1, observations were conducted on the premises of the Higashiyama Campus of Nagoya University (Chikusa-ku, Nagoya City). P-S logging has been conducted in the vicinity of the survey line in the past. A total of 41 geophones and McSEIS-ATs (1ch) and McSEIS-ATs (3ch) were used for microtremor observations, and observations were conducted on three survey lines at different sensor intervals. For the surface wave survey, 24 geophones were connected to the data logger McSEISE-SW. The west end was placed at -40 m, and the 0.5 m west side of the west end was excited, and the sensor array and excitation position were moved eastward by 1 m with each shot until the east end reaches 40 m. After the east end reached the 40 m point, only the excitation position was shifted by 1 m.
Results
Fig. 2 shows the waveforms of the Green's function with a sensor spacing of 4 m and a source at -40 m. Fig. 3 shows the waveforms with the same spacing and a source at 0 m. Fig. 4 shows the waveforms with a sensor spacing of 1 m and a source at 0 m. Wave propagation from the source was observed in the pseudo-Green functions obtained by applying seismic wave interferometry to microtremor records. Similarly, wave propagation from the source was also observed in the result of surface wave survey as shown in Fig. 5. As shown in Fig. 6, the dispersion curves at the -2 m geodetic coordinate show a transition from a wide sensor interval of constant microtremors to a narrow sensor interval, and then to a shallow Rayleigh wave survey, with a higher resolution range toward the high-frequency side. When the dispersion curve was converted into a relationship between S-wave velocity and depth using the method by Ballard (1964), as shown in Fig. 7, a subsurface structure was obtained in which S-wave velocity increases with depth.
Discussion
The wave propagation velocities of the pseudo-shot records based on microtremor records and the surface wave survey records are generally consistent. The wave propagation velocities seem to be different around at 7-16 m of survey line. This suggests that the subsurface structure differs from east to west around this area. However, since these are visual measurements, it is necessary to look at the dispersion curves by surface wave multi-channel analysis in order to investigate them accurately. The results of the dispersion curves obtained by Ballard's method are generally consistent with previous findings.
Knowledge of the subsurface structure is important for earthquake disaster prevention and building seismic safety. There are two methods of investigation that do not involve excavation: artificial excitation and passive microtremor record processing. One of the former methods is surface wave survey, e.g. Hayashi et al. (2001), and the latter method is to obtain pseudo-shot records by applying seismic wave interferometry to microtremor records, as described by Takahashi et al (2022). While the theory of shallow Rayleigh wave survey is well established and widely used, the method of obtaining pseudo-shot records from microtremor records is still in the experimental stage and has not been compared with surface wave survey. In this study, the Green's function and dispersion curves were calculated by applying both methods to the same survey line in order to compare and verify the two methods.
Observation and Experiment
As shown in Fig. 1, observations were conducted on the premises of the Higashiyama Campus of Nagoya University (Chikusa-ku, Nagoya City). P-S logging has been conducted in the vicinity of the survey line in the past. A total of 41 geophones and McSEIS-ATs (1ch) and McSEIS-ATs (3ch) were used for microtremor observations, and observations were conducted on three survey lines at different sensor intervals. For the surface wave survey, 24 geophones were connected to the data logger McSEISE-SW. The west end was placed at -40 m, and the 0.5 m west side of the west end was excited, and the sensor array and excitation position were moved eastward by 1 m with each shot until the east end reaches 40 m. After the east end reached the 40 m point, only the excitation position was shifted by 1 m.
Results
Fig. 2 shows the waveforms of the Green's function with a sensor spacing of 4 m and a source at -40 m. Fig. 3 shows the waveforms with the same spacing and a source at 0 m. Fig. 4 shows the waveforms with a sensor spacing of 1 m and a source at 0 m. Wave propagation from the source was observed in the pseudo-Green functions obtained by applying seismic wave interferometry to microtremor records. Similarly, wave propagation from the source was also observed in the result of surface wave survey as shown in Fig. 5. As shown in Fig. 6, the dispersion curves at the -2 m geodetic coordinate show a transition from a wide sensor interval of constant microtremors to a narrow sensor interval, and then to a shallow Rayleigh wave survey, with a higher resolution range toward the high-frequency side. When the dispersion curve was converted into a relationship between S-wave velocity and depth using the method by Ballard (1964), as shown in Fig. 7, a subsurface structure was obtained in which S-wave velocity increases with depth.
Discussion
The wave propagation velocities of the pseudo-shot records based on microtremor records and the surface wave survey records are generally consistent. The wave propagation velocities seem to be different around at 7-16 m of survey line. This suggests that the subsurface structure differs from east to west around this area. However, since these are visual measurements, it is necessary to look at the dispersion curves by surface wave multi-channel analysis in order to investigate them accurately. The results of the dispersion curves obtained by Ballard's method are generally consistent with previous findings.