4:30 PM - 4:45 PM
[SVC31-21] Underwater acoustic signals associated with the eruption of Fukutoku-Oka-No-Ba and Hunga Tonga - Hunga Ha'apai volcanoes observed with the OBSs of the submarine cabled observatory S-net
Underwater acoustic signals generated by the eruption of submarine volcanoes at Fukutoku-Oka-No-Ba and Hunga Tonga - Hunga Ha'apai, may propagate through deep-sea sound channels (Sound Fixing and Ranging, SOFAR channel) in the sea and reach far away (Metz, 2021).
This time, using the ocean bottom seismometers (OBSs) of the submarine cabled observatory “S-net” consisting of 150 stations, the underwater acoustic signals associated with these submarine volcanic eruptions are detected.
In S-net, a velocimeter and an accelerometer are attached to each station in the same way for each of the three components in arbitrary direction. This time, the velocimeter is used. The vertical component is calculated based on the direction of gravity measured by the accelerometer by the same method as Iwase (2016) and used for analysis. The sampling frequency is 100 Hz.
The original waveform data contains background noise such as earthquake signals and ship noise, and it is difficult to identify eruption-related signals at first glance. The ship noise has a sound source near the stations, and it is considered that there is no correlation between each station over a wide range. In addition, the signals associated with earthquakes are mainly composed of seismic waves such as P waves and S waves. On the other hand, the underwater acoustic signal related to the submarine volcanic eruption propagates through the SOFAR channel at the sound speed and reaches the station, but its propagation velocity is slower than the above-mentioned seismic waves. Therefore, as a means to distinguish the acoustic signal propagating through the SOFAR channel from the seismic waves, the horizontal distance between each volcano and each station is divided by the sound speed (1500 m/s) to obtain the propagation time, and the observation waveform at each station is obtained. , The spectrogram (running spectrum) was drawn by shifting the time axis by the propagation time corresponding to each station, and the differences among the stations were compared. As a result, the acoustic signal associated with the eruption should be displayed at the same position in the drawn spectrogram regardless of the station.
As a result, acoustic signals associated with each eruption were detected at more than 40 stations of the S-net. For the detection examples of Fukutoku-Oka-No-Ba eruption, the spectrograms of stations S1N08 and S4N13 of S-net are shown in Fig. 1 and Fig. 2, respectively. For Hunga Tonga - Hunga Ha'apai eruption, the spectrograms of stations S1N12 and S2N15 of S-net are shown in Fig. 3 and Fig. 4, respectively. The horizontal axis is the elapsed time (seconds) from each stated time. The filled triangles indicate the events associated with the eruption, and the hollow triangles indicate the earthquake that is not related to the eruption.
There are some signals that are difficult to distinguish from T-phase of the earthquake, and I will check them in detail.
Acknowledgments
I would like to thank National Research Institute for Earth Science and Disaster Prevention (NIED), for providing data.
References
D.Metz (2021): http://www.jamstec.go.jp/j/jamstec_news/fukutokuokanoba/column03/.
R. Iwase (2016): JJAP, 55, 07KG01.
This time, using the ocean bottom seismometers (OBSs) of the submarine cabled observatory “S-net” consisting of 150 stations, the underwater acoustic signals associated with these submarine volcanic eruptions are detected.
In S-net, a velocimeter and an accelerometer are attached to each station in the same way for each of the three components in arbitrary direction. This time, the velocimeter is used. The vertical component is calculated based on the direction of gravity measured by the accelerometer by the same method as Iwase (2016) and used for analysis. The sampling frequency is 100 Hz.
The original waveform data contains background noise such as earthquake signals and ship noise, and it is difficult to identify eruption-related signals at first glance. The ship noise has a sound source near the stations, and it is considered that there is no correlation between each station over a wide range. In addition, the signals associated with earthquakes are mainly composed of seismic waves such as P waves and S waves. On the other hand, the underwater acoustic signal related to the submarine volcanic eruption propagates through the SOFAR channel at the sound speed and reaches the station, but its propagation velocity is slower than the above-mentioned seismic waves. Therefore, as a means to distinguish the acoustic signal propagating through the SOFAR channel from the seismic waves, the horizontal distance between each volcano and each station is divided by the sound speed (1500 m/s) to obtain the propagation time, and the observation waveform at each station is obtained. , The spectrogram (running spectrum) was drawn by shifting the time axis by the propagation time corresponding to each station, and the differences among the stations were compared. As a result, the acoustic signal associated with the eruption should be displayed at the same position in the drawn spectrogram regardless of the station.
As a result, acoustic signals associated with each eruption were detected at more than 40 stations of the S-net. For the detection examples of Fukutoku-Oka-No-Ba eruption, the spectrograms of stations S1N08 and S4N13 of S-net are shown in Fig. 1 and Fig. 2, respectively. For Hunga Tonga - Hunga Ha'apai eruption, the spectrograms of stations S1N12 and S2N15 of S-net are shown in Fig. 3 and Fig. 4, respectively. The horizontal axis is the elapsed time (seconds) from each stated time. The filled triangles indicate the events associated with the eruption, and the hollow triangles indicate the earthquake that is not related to the eruption.
There are some signals that are difficult to distinguish from T-phase of the earthquake, and I will check them in detail.
Acknowledgments
I would like to thank National Research Institute for Earth Science and Disaster Prevention (NIED), for providing data.
References
D.Metz (2021): http://www.jamstec.go.jp/j/jamstec_news/fukutokuokanoba/column03/.
R. Iwase (2016): JJAP, 55, 07KG01.