4:30 PM - 4:45 PM
[SVC34-15] Detection of small volcanic tsunami signals from Kita-Ioto Caldera using dense ocean-bottom pressure data of DONET
Keywords:Submarine volcano, Caldera, Tsunami, Volcanic earthquakes, Waveform analysis
Kita-Ioto Caldera is an active submarine volcano in the Izu-Bonin arc, where non-double-couple (NDC) earthquakes (Mw 5.2–5.3) were observed every 2–5 years. Our previous study (Sandanbata & Saito, 2022, JpGU) reported milli-meter tsunami signals following the 2008 and 2015 earthquakes recorded by a single ocean bottom pressure (OBP) gauge of the DART (Deep-ocean Assessment and Reporting of Tsunami) system far off the coast of Japan. The tsunami data enabled us to find that an intra-caldera fault rupture occurred in the caldera due to its pressurized magma reservoir; the mechanism is called trapdoor faulting. Yet, the OBP gauge has lost high temporal resolution data for periods including the timings of recent NDC earthquakes in 2017 (Mw 5.2) and 2019 (Mw 5.3) and was unavailable for investigation of these earthquakes.
We here use multiple OBP data from dense array stations of DONET (Dense Oceanfloor Network system for Earthquakes and Tsunamis), southwest of Japan, to reveal whether the 2017 and 2019 earthquakes caused tsunamis, like the 2008 event. After checking that tsunami signals are not clearly identified in each OBP record due to poor signal-to-noise (S/N) ratios, we perform a tsunami waveform stacking to extract tsunami signals from Kita-Ioto Caldera. We first simulate the tsunami generation and propagation from a trapdoor faulting model proposed for the 2008 earthquake in the caldera. We next estimate tsunami travel times at each OBP station by reading the first peaks of the synthetic waveforms. We then shift waveform traces of the OBP data by the travel time differences and stacked them. For comparison, this waveform stacking method is applied to the synthetic waveforms from the 2008 earthquake model.
Consequently, we find clear tsunami-like oscillations in the stacked waveforms from the OBP data following the two earthquakes. The oscillations in the stacked OBP waveforms are overall similar to those in the stacked synthetic waveforms in terms of wave period and phase. Therefore, we interpret the OBP oscillations as stacked small-amplitude tsunami signals from Kita-Ioto Caldera and suggest that the two earthquakes occurred with the trapdoor faulting mechanism, similar to the 2008 event. Note that the stacked waveform shapes for the 2017 and 2019 earthquakes are slightly different, implying differences in detailed source properties of the two earthquakes, such as location or length of the intra-caldera fault system. Thus, our detection technique of small volcanic tsunami events will enable us to widen our study targets of submarine volcanoes and to investigate volcanic activity under water that have been overlooked by conventional monitoring systems.
We here use multiple OBP data from dense array stations of DONET (Dense Oceanfloor Network system for Earthquakes and Tsunamis), southwest of Japan, to reveal whether the 2017 and 2019 earthquakes caused tsunamis, like the 2008 event. After checking that tsunami signals are not clearly identified in each OBP record due to poor signal-to-noise (S/N) ratios, we perform a tsunami waveform stacking to extract tsunami signals from Kita-Ioto Caldera. We first simulate the tsunami generation and propagation from a trapdoor faulting model proposed for the 2008 earthquake in the caldera. We next estimate tsunami travel times at each OBP station by reading the first peaks of the synthetic waveforms. We then shift waveform traces of the OBP data by the travel time differences and stacked them. For comparison, this waveform stacking method is applied to the synthetic waveforms from the 2008 earthquake model.
Consequently, we find clear tsunami-like oscillations in the stacked waveforms from the OBP data following the two earthquakes. The oscillations in the stacked OBP waveforms are overall similar to those in the stacked synthetic waveforms in terms of wave period and phase. Therefore, we interpret the OBP oscillations as stacked small-amplitude tsunami signals from Kita-Ioto Caldera and suggest that the two earthquakes occurred with the trapdoor faulting mechanism, similar to the 2008 event. Note that the stacked waveform shapes for the 2017 and 2019 earthquakes are slightly different, implying differences in detailed source properties of the two earthquakes, such as location or length of the intra-caldera fault system. Thus, our detection technique of small volcanic tsunami events will enable us to widen our study targets of submarine volcanoes and to investigate volcanic activity under water that have been overlooked by conventional monitoring systems.