5:15 PM - 7:15 PM
[SSS13-P02] Can seismic waves be detected from changes in lake water levels?
Seiches are standing waves formed in enclosed or partially enclosed bodies of water. They are related to natural oscillations that depend on the shape, size, and depth of a lake and act as an “amplifier” for seismic motions of specific wavelengths. Using Lake Suwa, a rectangular lake, as the study site, we calculated the period of seismic seiches. Lake Suwa is located in the eastern rift valley of the Itoigawa-Shizuoka Tectonic Line, where it intersects the Median Tectonic Line. Therefore, it is possible to detect the influence of seismic waves caused by earthquakes over a wide area. Accordingly, the purpose of this study was to investigate the relationship between seismic waves and water level changes by conducting high time-resolution water level observations and frequency analysis of Lake Suwa. The seismic seiches observed before and after the Noto Peninsula earthquake in January 2024 are discussed with a focus on the seismic waves of the surface wave.
Surface seiches are typically excited by wind waves, and their periods are determined by the shape and depth of the lake. In this case, the period of the strongest first mode was 1,200-1,300 s (Tanaka, 1918). The dominant wind direction during the first mode was NW to W. The first mode was oscillation, with the minor axis of the diagonal line as a node, assuming that the lake was rectangular.
The maximum amplitude of the lake level caused by the earthquake was approximately 2 cm after 3 min. The oscillations of the lake surface caused by the seismic motion continued for approximately 4 h. For normal seiches, the amplitude after the mainshock was stronger than that before the mainshock, and the second mode of the seiches (750-800 s) responded strongly. In both wind profiles, before and after the mainshock, the dominant wind direction was WNW to W; therefore, the first mode was normally dominant. The reason for this difference is that the seismic waves from the direction of the epicentre (NNW of Lake Suwa) resonated with the second mode of seiches, which oscillated in the N-S direction. Between the time of the main shock and the estimated time of arrival of the P-wave, a periodic wave group-like water level change of approximately 10-11 s was observed, and its wave group was observed at intervals of 63-68 s and was confirmed approximately three times. In addition, after the main shock, a peak of approximately 30 s was considered the period of Earth's free oscillation.
Waves that correspond to a boundary between normal and inverse dispersions (i.e., the extrema of the group velocity) are called the Airy phases. They can propagate with large amplitudes, and the group velocity of a Rayleigh wave can reach its maximum near wave packets with a period of 240 s. Therefore, we assumed that the period of approximately 240 s is the period of the surface waves. The period of approximately 240 s after the main shock was a strong response compared to the non-seismic period. Moreover, the amplitude before the main shock was stronger than that after the main shock; in particular, the response in which the amplitude of the spectrum was the strongest one day before the main shock. However, it could not be concluded from the meteorological data that wind-induced seiches were the cause of the strong response approximately 240 s before the main shock. As regards this period of 240 s, the results of the Wavelet transform will also be presented.
In summary, we confirmed the periods of forced oscillations were caused by surface waves (Rayleigh waves) with periods of 240 s before and after the earthquake. A period of 30–31 s after the earthquake was considered as the Earth's free oscillation. These results suggest that the oscillations observed in the lake can be caused by either the effect of seismic waves or the movement of crust related to an earthquake of known magnitude and location.
Surface seiches are typically excited by wind waves, and their periods are determined by the shape and depth of the lake. In this case, the period of the strongest first mode was 1,200-1,300 s (Tanaka, 1918). The dominant wind direction during the first mode was NW to W. The first mode was oscillation, with the minor axis of the diagonal line as a node, assuming that the lake was rectangular.
The maximum amplitude of the lake level caused by the earthquake was approximately 2 cm after 3 min. The oscillations of the lake surface caused by the seismic motion continued for approximately 4 h. For normal seiches, the amplitude after the mainshock was stronger than that before the mainshock, and the second mode of the seiches (750-800 s) responded strongly. In both wind profiles, before and after the mainshock, the dominant wind direction was WNW to W; therefore, the first mode was normally dominant. The reason for this difference is that the seismic waves from the direction of the epicentre (NNW of Lake Suwa) resonated with the second mode of seiches, which oscillated in the N-S direction. Between the time of the main shock and the estimated time of arrival of the P-wave, a periodic wave group-like water level change of approximately 10-11 s was observed, and its wave group was observed at intervals of 63-68 s and was confirmed approximately three times. In addition, after the main shock, a peak of approximately 30 s was considered the period of Earth's free oscillation.
Waves that correspond to a boundary between normal and inverse dispersions (i.e., the extrema of the group velocity) are called the Airy phases. They can propagate with large amplitudes, and the group velocity of a Rayleigh wave can reach its maximum near wave packets with a period of 240 s. Therefore, we assumed that the period of approximately 240 s is the period of the surface waves. The period of approximately 240 s after the main shock was a strong response compared to the non-seismic period. Moreover, the amplitude before the main shock was stronger than that after the main shock; in particular, the response in which the amplitude of the spectrum was the strongest one day before the main shock. However, it could not be concluded from the meteorological data that wind-induced seiches were the cause of the strong response approximately 240 s before the main shock. As regards this period of 240 s, the results of the Wavelet transform will also be presented.
In summary, we confirmed the periods of forced oscillations were caused by surface waves (Rayleigh waves) with periods of 240 s before and after the earthquake. A period of 30–31 s after the earthquake was considered as the Earth's free oscillation. These results suggest that the oscillations observed in the lake can be caused by either the effect of seismic waves or the movement of crust related to an earthquake of known magnitude and location.