1:45 PM - 2:00 PM
[SVC28-07] Seismic structural imaging of shallow fluid supply beneath Mt. Ontake
Keywords:Mt. Ontake, Seismic velocity structure, Reflection profile
1. Introduction
A fluid supply to shallow parts of hydrothermally active volcanoes is a key to understand phreatic eruption and preparation processes. Mt. Ontake has been extensively monitored by a dense network after the 2014 eruption, showing persistent seismicity and deflation. The magma supply system (Takagi and Onizawa 2016; Murase et al. 2016) and shallow resistivity structure (Abdallar and Mogi 2016) of this volcano were investigated, whereas previous seismic structural surveys have mainly focused on base regions (e.g., Ikami et al. 1986; Doi et al. 2013). In this study, we investigate a shallow seismic structure of Mt. Ontake using data after November 2017 when a dense summit station network was established.
2. Analysis
(1) P-wave velocity in the summit region
We first investigate a typical P-wave velocity in the summit region using intermediate-depth earthquakes which occur along the Pacific plate (~250 km deep) beneath Mt. Ontake. The first arrivals of these earthquakes are aligned from lower to higher altitude stations, and this upward propagation for ≧1500 m above sea level (asl.) is best approximated by a P-wave velocity Vp = 2600 m/s based on a semblance analysis, although the value slightly differed among events. This velocity is close to that of the southeastern flank estimated by a refraction survey of Ikami et al. (1986).
(2) A layered structure
We estimated hypocenters of 59 volcano-tectonic earthquakes in November 2017 using grid searches to minimize residuals of P- and S-phase arrival times. We examined three velocity models: (i) a uniform medium of Vp = 2600 m/s based on the analysis (1); (ii) a two-layer horizontally stratified medium of Vp = 2600 and 5900 m/s; and (iii) a three-layer horizontally stratified medium of Vp = 2600, 3900, and 5900 m/s, where we referred to the velocities of second and third layers in the southeastern flank (Ikami et al. 1986). We assumed an S-wave velocity Vs = Vp/√3, and conducted grid searches for layer boundary altitudes. The traveltime residuals were smallest when the second boundary is set at 1000 m asl., whereas two candidates for the first boundary (1900 and 2500 m asl.) showed almost equal and smallest residuals. We consider 1900 m to be more plausible because Vp = 2600 m/s is estimated for ≧1500 m asl. in (1).
(3) A reflection profile
We investigated a reflection profile of Mt. Ontake by stacking ambient noise autocorrelations of continuous 1-bit records from November 2017 to May 2020. We used a high-frequency (6-14 Hz) band to obtain high resolution images in shallow parts. The lag times were converted to depths using the velocity model obtained in (2). Results showed relatively intense reflectors down to ~4 km depths at stations close to craters, whereas most other stations show intense reflectors only to several hundred meters.
3. Discussion
The first boundary altitude estimated by (2) is close to that of old (<0.4 Ma) and new (>0.1 Ma) lavas, suggesting that Vp = 2600 and 3900 m/s represent typical velocities of them. At a similar altitude, several stations showed a possible reflector in (3).
The hypocenters estimated in (2) are distributed in 900-2100 asl., which are between deep (> 3000 m) and shallow (500 m) sources of the deflation after 2014 estimated by Narita and Murakami (2018). The relatively intense reflectors below the craters imaged by (3) are close to this depth range. A possible interpretation to these results may be a fluid migration from the deep to shallow deflation sources, associated seismicity, and resultant cracks which may be seen as the reflectors.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP19K04016, and the Ministry of Education, Culture, Sports, Science and Technology of Japan, under its The Second Earthquake and Volcano Hazards Observation and Research Program. We used data from Japan Meteorological Agency, National Research Institute for Earth Science and Disaster Resilience, Nagano and Gifu Prefectures.
A fluid supply to shallow parts of hydrothermally active volcanoes is a key to understand phreatic eruption and preparation processes. Mt. Ontake has been extensively monitored by a dense network after the 2014 eruption, showing persistent seismicity and deflation. The magma supply system (Takagi and Onizawa 2016; Murase et al. 2016) and shallow resistivity structure (Abdallar and Mogi 2016) of this volcano were investigated, whereas previous seismic structural surveys have mainly focused on base regions (e.g., Ikami et al. 1986; Doi et al. 2013). In this study, we investigate a shallow seismic structure of Mt. Ontake using data after November 2017 when a dense summit station network was established.
2. Analysis
(1) P-wave velocity in the summit region
We first investigate a typical P-wave velocity in the summit region using intermediate-depth earthquakes which occur along the Pacific plate (~250 km deep) beneath Mt. Ontake. The first arrivals of these earthquakes are aligned from lower to higher altitude stations, and this upward propagation for ≧1500 m above sea level (asl.) is best approximated by a P-wave velocity Vp = 2600 m/s based on a semblance analysis, although the value slightly differed among events. This velocity is close to that of the southeastern flank estimated by a refraction survey of Ikami et al. (1986).
(2) A layered structure
We estimated hypocenters of 59 volcano-tectonic earthquakes in November 2017 using grid searches to minimize residuals of P- and S-phase arrival times. We examined three velocity models: (i) a uniform medium of Vp = 2600 m/s based on the analysis (1); (ii) a two-layer horizontally stratified medium of Vp = 2600 and 5900 m/s; and (iii) a three-layer horizontally stratified medium of Vp = 2600, 3900, and 5900 m/s, where we referred to the velocities of second and third layers in the southeastern flank (Ikami et al. 1986). We assumed an S-wave velocity Vs = Vp/√3, and conducted grid searches for layer boundary altitudes. The traveltime residuals were smallest when the second boundary is set at 1000 m asl., whereas two candidates for the first boundary (1900 and 2500 m asl.) showed almost equal and smallest residuals. We consider 1900 m to be more plausible because Vp = 2600 m/s is estimated for ≧1500 m asl. in (1).
(3) A reflection profile
We investigated a reflection profile of Mt. Ontake by stacking ambient noise autocorrelations of continuous 1-bit records from November 2017 to May 2020. We used a high-frequency (6-14 Hz) band to obtain high resolution images in shallow parts. The lag times were converted to depths using the velocity model obtained in (2). Results showed relatively intense reflectors down to ~4 km depths at stations close to craters, whereas most other stations show intense reflectors only to several hundred meters.
3. Discussion
The first boundary altitude estimated by (2) is close to that of old (<0.4 Ma) and new (>0.1 Ma) lavas, suggesting that Vp = 2600 and 3900 m/s represent typical velocities of them. At a similar altitude, several stations showed a possible reflector in (3).
The hypocenters estimated in (2) are distributed in 900-2100 asl., which are between deep (> 3000 m) and shallow (500 m) sources of the deflation after 2014 estimated by Narita and Murakami (2018). The relatively intense reflectors below the craters imaged by (3) are close to this depth range. A possible interpretation to these results may be a fluid migration from the deep to shallow deflation sources, associated seismicity, and resultant cracks which may be seen as the reflectors.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP19K04016, and the Ministry of Education, Culture, Sports, Science and Technology of Japan, under its The Second Earthquake and Volcano Hazards Observation and Research Program. We used data from Japan Meteorological Agency, National Research Institute for Earth Science and Disaster Resilience, Nagano and Gifu Prefectures.