3:00 PM - 3:15 PM
[SVC28-12] The 2019-2020 Nishinoshima activity inferred from seismic and infrasound observations in the island
Keywords:Nishino-shima, volcanic earthquakes, infrasound, satelite data transmission
On Nishinoshima Island in Ogasawara Islands, volcanic activity began in November 2013. The activity started as a small submarine eruption and gradually shifted to the activity accompanying lava flow. By October 2014, most parts of the old Nishinoshima were covered by lava flow. The volcanic activity has continued since then, and can be divided into 4 terms: the 2013-2015 activity (1st term), the 2017 activity (2nd term), the 2018 activity (3rd term), and the 2019-2020 activity (4th term).
The first landing survey after 2013 was conducted in October 2016. Geological surveys, ejecta collection, installation of seismic and infrasound sensors, and biological surveys were carried out (Takeo et al., 2018, Maeno et al., 2018). The geophysical observation equipment was buried in the lava flow during the 2nd term activity, but valuable data in the early stage of the 2nd term were obtained.
In September 2019, the "2019 Nishinoshima Comprehensive Academic Survey" was conducted by the Ministry of the Environment. We participated in the landing survey and re-installed seismic and infrasound observation equipment. The fourth phase of activity, which began in December 2019, was dominated by multiple lava runoff, and reached its peak in June-July 2020, followed by a sudden decline at the end of July. The reinstalled equipment recorded the volcanic activity from the beginning of the 4th term to the time when it was swallowed by lava flow. In this report, we introduce the outline of the observation equipment and the characteristics of the seismic and infrasound data observed during the 4th period.
In the 2019 observation, we used almost the same equipment and methods as in the 2016 observation (Watanabe et al., 2017). It is solar-powered, and data is transmitted by satellite communication. Since it is difficult to send all the 80MB of daily data in real-time due to the limitation of costs and power, we decided to send compressed running spectra only. As a result, the daily amount of transmitted data was reduced to about 500 kB. The running spectrum is created daily and sent to Tokyo so that we can get an overview of volcanic activity by the next day. Collection of the unprocessed original waveform is also possible on request.
The 4th-period activity started as following. From around 12:00 on December 4, 2019, the high-frequency components of seismic signal began to increase. Satellite data also confirmed thermal anomalies at night. At this point, the magma is approaching the shallow part, but the magma outflow has not yet started. The seismic amplitude remarkably increased after 4:00 in the early morning the next day. The air-shocks began to occur after 5:00. These observations suggest that a small eruption started at the summit around this time. After 12:00 on December 5, the amplitude and frequency of occurrence of events increased sharply, and a large-amplitude ground shaking accompanied by air-shocks occurred at 16:13. Since the thermal anomaly increased rapidly after 15:00, it is possible that the event at 16:13 was an explosive eruption accompanied by the expansion of the lava flow outlet causing full-scale outflow.
The figure shows the running spectrum of the seismic waveform from December 1st to June 21st. Immediately after the start of the activity, the dominant frequency was 2Hz, but the dominant frequency gradually decreased to 0.5 to 1Hz by early June when the eruption rate began to increase sharply. In middle to late June, the dominant frequency shifted to the high-frequency side. The temporal variation of the eruption rate estimated from satellite data does not necessarily correspond to the temporal change in the seismic amplitude. As for the particle velocity of the horizontal components of the seismic waveform, the high-frequency component of 1.5 Hz or higher consistently points in the direction of the central active vent. The particle motion of the slightly low-frequency component of 0.5 to 1.5 Hz behaves differently. It pointed slightly south of the central vent initially, and then gradually shifted to pointing to the north-south direction. These observations suggest that the high-frequency components correspond to continuous small eruptions at the central vent, while the lower frequency components may reflect the movement of fluids or pressure sources at depth. In the presentation, we will explain in more detail the transition of the 4th term activity estimated from the earthquake and infrasound waveforms.
The first landing survey after 2013 was conducted in October 2016. Geological surveys, ejecta collection, installation of seismic and infrasound sensors, and biological surveys were carried out (Takeo et al., 2018, Maeno et al., 2018). The geophysical observation equipment was buried in the lava flow during the 2nd term activity, but valuable data in the early stage of the 2nd term were obtained.
In September 2019, the "2019 Nishinoshima Comprehensive Academic Survey" was conducted by the Ministry of the Environment. We participated in the landing survey and re-installed seismic and infrasound observation equipment. The fourth phase of activity, which began in December 2019, was dominated by multiple lava runoff, and reached its peak in June-July 2020, followed by a sudden decline at the end of July. The reinstalled equipment recorded the volcanic activity from the beginning of the 4th term to the time when it was swallowed by lava flow. In this report, we introduce the outline of the observation equipment and the characteristics of the seismic and infrasound data observed during the 4th period.
In the 2019 observation, we used almost the same equipment and methods as in the 2016 observation (Watanabe et al., 2017). It is solar-powered, and data is transmitted by satellite communication. Since it is difficult to send all the 80MB of daily data in real-time due to the limitation of costs and power, we decided to send compressed running spectra only. As a result, the daily amount of transmitted data was reduced to about 500 kB. The running spectrum is created daily and sent to Tokyo so that we can get an overview of volcanic activity by the next day. Collection of the unprocessed original waveform is also possible on request.
The 4th-period activity started as following. From around 12:00 on December 4, 2019, the high-frequency components of seismic signal began to increase. Satellite data also confirmed thermal anomalies at night. At this point, the magma is approaching the shallow part, but the magma outflow has not yet started. The seismic amplitude remarkably increased after 4:00 in the early morning the next day. The air-shocks began to occur after 5:00. These observations suggest that a small eruption started at the summit around this time. After 12:00 on December 5, the amplitude and frequency of occurrence of events increased sharply, and a large-amplitude ground shaking accompanied by air-shocks occurred at 16:13. Since the thermal anomaly increased rapidly after 15:00, it is possible that the event at 16:13 was an explosive eruption accompanied by the expansion of the lava flow outlet causing full-scale outflow.
The figure shows the running spectrum of the seismic waveform from December 1st to June 21st. Immediately after the start of the activity, the dominant frequency was 2Hz, but the dominant frequency gradually decreased to 0.5 to 1Hz by early June when the eruption rate began to increase sharply. In middle to late June, the dominant frequency shifted to the high-frequency side. The temporal variation of the eruption rate estimated from satellite data does not necessarily correspond to the temporal change in the seismic amplitude. As for the particle velocity of the horizontal components of the seismic waveform, the high-frequency component of 1.5 Hz or higher consistently points in the direction of the central active vent. The particle motion of the slightly low-frequency component of 0.5 to 1.5 Hz behaves differently. It pointed slightly south of the central vent initially, and then gradually shifted to pointing to the north-south direction. These observations suggest that the high-frequency components correspond to continuous small eruptions at the central vent, while the lower frequency components may reflect the movement of fluids or pressure sources at depth. In the presentation, we will explain in more detail the transition of the 4th term activity estimated from the earthquake and infrasound waveforms.