5:15 PM - 6:45 PM
[SVC26-P11] Infrasonic and seismic observations of rockfalls at Hoei crater, Fuji volcano
Keywords:Rockfall, Infrasound, Seismology, Volcanic Disaster Mitigation
In a volcanic disaster mitigation of Mt. Fuji, there is a demand for observation techniques to identify the unpredictable erupting vent regardless of weather conditions. As a possible technique, we are focusing on the infrasound observations. However, practical evaluation of new observation sites is difficult using low-frequent eruptions. To address this gap, this study focuses on rockfalls occurring in Hoei crater on Mt. Fuji, which have have been witnessed by many climbers. As a first step toward using a rockfall signal as a phenomenon testing new observation sites, we conducted infrasonic and seismic campaign observation in Hoei crater to understand rockfall signals.
The campaign observations held over two periods: from July 23 to October 1 on 2022 and from August 5 to October 2 on 2023. A station was set on the top of the pyroclastic cone at the center of Hoei crater (red symbols in Figs. 1b and 1c). Apart from it by ~50 m, an additional station was set from August to September in 2023 (blue symbols in Figs. 1b and 1c). Each station consists of a UD-component seismometer (L-22D Sercel Inc.) and a three-element very-small-aperture infrasound array, which places three microphones (Type 7744N, ACO Co.) with a span of ~10 m. In addition to these campaign stations, we analyzed the data from the seismic networks surrounding Fuji volcano (V-net, NIED and JMA, and the seismic network of ERI, the University of Tokyo; Fig. 1a). In particular, we focused on V.FJ8G and V.FUJ2, which is 1st and 2nd nearest to Hoei crater (~1 km and ~2 km northwest from Hoei crater, respectively).
The campaign stations in Hoei crater observed various signals, including rockfalls, earthquakes, airplane noise, and military exercises. These signals could be identified by the signal characteristics such as the waveform shape, the amplitude ratio of infrasonic and seismic sensors, and the apparent velocity of the infrasound signal in the infrasound array plane. Although the analysis of the campaign of 2023 is still ongoing, we found at least 14 and 2 rockfall signals in the campaign of 2022 and 2023, respectively. Relatively large rockfalls could be observed at stations 8 km away from Hoei crater, while small rockfalls were difficult to detect even at station V.FUJ2. The rockfall signals had duration of several tens of seconds and power in a broad band (Fig. 2). Infrasound array estimated the direction of the rockfall signal arrival, indicating the direction of the collapsed wall. Some rockfall signals showed the change of the direction of the infrasound arrival by ~20 degrees in ~20 s, implying the movement of the rockfall mass.
The seismic waveforms recorded at the network stations V.FJ8G and V.FUJ2, located on the crater’s far side, exhibited maximum amplitude during the initial phase (red and orange lines in Fig. 2b) in contrast to the spindle-shaped waveforms at our stations within the crater (green lines in Fig. 2b). Though understanding the causes of this waveform difference remains a future work, we are considering the radiation characteristics or the change of the source-receiver distance as a potential factor.
In conclusion, our study demonstrates that we could observe and identify rockfalls in Hoei crater. The infrasonic array contributed to identifying the rockfall signals from the other signals and understanding the movement of a rockfall mass. However, several issues remain, including infrasound wind noise reduction, detailed analysis of the infrasonic array and seismic network, comparison with other observations such as visual images or meteorological data, and exploring the causes of rockfalls. Addressing these issues is crucial for better understanding and utilization of rockfall signals.
We used the seismic data provided by the National Research Institute for Earth Science and Disaster Resilience, the Japan Meteorological Agency, and Earthquake Research Institute, the University of Tokyo.
The campaign observations held over two periods: from July 23 to October 1 on 2022 and from August 5 to October 2 on 2023. A station was set on the top of the pyroclastic cone at the center of Hoei crater (red symbols in Figs. 1b and 1c). Apart from it by ~50 m, an additional station was set from August to September in 2023 (blue symbols in Figs. 1b and 1c). Each station consists of a UD-component seismometer (L-22D Sercel Inc.) and a three-element very-small-aperture infrasound array, which places three microphones (Type 7744N, ACO Co.) with a span of ~10 m. In addition to these campaign stations, we analyzed the data from the seismic networks surrounding Fuji volcano (V-net, NIED and JMA, and the seismic network of ERI, the University of Tokyo; Fig. 1a). In particular, we focused on V.FJ8G and V.FUJ2, which is 1st and 2nd nearest to Hoei crater (~1 km and ~2 km northwest from Hoei crater, respectively).
The campaign stations in Hoei crater observed various signals, including rockfalls, earthquakes, airplane noise, and military exercises. These signals could be identified by the signal characteristics such as the waveform shape, the amplitude ratio of infrasonic and seismic sensors, and the apparent velocity of the infrasound signal in the infrasound array plane. Although the analysis of the campaign of 2023 is still ongoing, we found at least 14 and 2 rockfall signals in the campaign of 2022 and 2023, respectively. Relatively large rockfalls could be observed at stations 8 km away from Hoei crater, while small rockfalls were difficult to detect even at station V.FUJ2. The rockfall signals had duration of several tens of seconds and power in a broad band (Fig. 2). Infrasound array estimated the direction of the rockfall signal arrival, indicating the direction of the collapsed wall. Some rockfall signals showed the change of the direction of the infrasound arrival by ~20 degrees in ~20 s, implying the movement of the rockfall mass.
The seismic waveforms recorded at the network stations V.FJ8G and V.FUJ2, located on the crater’s far side, exhibited maximum amplitude during the initial phase (red and orange lines in Fig. 2b) in contrast to the spindle-shaped waveforms at our stations within the crater (green lines in Fig. 2b). Though understanding the causes of this waveform difference remains a future work, we are considering the radiation characteristics or the change of the source-receiver distance as a potential factor.
In conclusion, our study demonstrates that we could observe and identify rockfalls in Hoei crater. The infrasonic array contributed to identifying the rockfall signals from the other signals and understanding the movement of a rockfall mass. However, several issues remain, including infrasound wind noise reduction, detailed analysis of the infrasonic array and seismic network, comparison with other observations such as visual images or meteorological data, and exploring the causes of rockfalls. Addressing these issues is crucial for better understanding and utilization of rockfall signals.
We used the seismic data provided by the National Research Institute for Earth Science and Disaster Resilience, the Japan Meteorological Agency, and Earthquake Research Institute, the University of Tokyo.