5:15 PM - 6:45 PM
[SSS07-P09] Temporal changes in noise amplitude observed by seismometers in and around the Hakone volcano
Keywords:seismic noise, volcanic tremor
When no earthquake is occurring, seismometers record ambient noise. This includes meteorological noise, such as rain and wind, and artificial noise, such as driving cars or operating factories. Seismic noise has been increasingly focused on these days. For example, quiescence of social activities were observed by seismometers when COVID-19 was outbreak all over the world (Lecocq et al., 2020; Yabe et al., 2020). Noise amplitude was well correlated with the flow rates of a river associated with heavy rainfall near the Mogami River, Japan (Shakti and Sawazaki 2021). On the other hand, there are some cases in which seismologically important phenomena were discovered from noise-like signals. For example, non-volcanic tremors occurring in plate subduction zones, which are well known as a type of slow earthquake, were discovered from waveforms, which look like noise (Obara, 2002). It was also found that the night-time background noise gradually increased before the 2018 eruption of Kirishima volcano (Ichihara, Ohminato, et al., 2023). Since the amplitudes of background noise well corresponded to changes in volcanic activity, it is thought that the background noise consists of volcanic tremors associated with degassing events from magma (Ichihara, Kobayashi, et al., 2023). In this study, we investigated the temporal characteristics of noise observed at four stations in and around Hakone volcano and also investigated whether there is any significant background volcanic or tectonic activity.
Referring to Ichihara, Ohminato, et al. (2023), I first applied a 4-16 Hz bandpass filter to the waveforms and calculated the 1-minute RMS amplitudes of the waveforms in this study. Next, a day was divided into four 6-hour windows (0:00-5:59, 6:00-11:59, 12:00-17:59, and 18:00-23:59) to exclude short-term amplitude changes such as local earthquakes. The median of 360 RMS values in each 6-hour window was used as the representative value to obtain noise amplitude.
As a result, at the Owakudani and Owakudani-Jizoson stations located around the Owakudani fumarolic area, the amplitude of noise in the daytime (6:00-11:59 and 12:00-17:59) was about 10 times larger than that in the nighttime (0:00-5:59 and 18:00-23:59), because these seismometers are set on the ground and are affected by noise of cars and tourist facilities. However, when the eruption alert level was raised in 2015 and 2019, and when the first pandemic of COVID-19 occurred in 2020, the daytime noise amplitude was as low as that of night-time due to the closure of the facilities. In the two periods from June 2015 to May 2016 and from July 2022 to July 2023, the nighttime noise amplitude had been high. I consider that this night-time noise is likely to be unknown volcanic tremor, based on their long duration. On the other hand, at the Komagatake station in the Hakone area, the noise level was low because it is a borehole station and there is no noise source around the station. In addition, no nighttime noise amplitude changes were observed.
At the Omatazawa station in the Tanzawa mountain area, artificial noise has rarely been observed during the daytime because the station is located in the mountains, where there is no human habitation. Because this station is located along a river, when heavy rainfall of 100 mm or more per day was observed at the nearby JMA AMeDAS Tanzawa-ko, the noise level rapidly increased 10-50 times as common level, and the noise level gradually decreased over about two weeks. The noise amplitude may be corresponding to the flow rate of the river. Since there is a correspondence between rainfall and seismic noise amplitude, there is a possibility that the river flow rate can be estimated from the noise amplitude.
Referring to Ichihara, Ohminato, et al. (2023), I first applied a 4-16 Hz bandpass filter to the waveforms and calculated the 1-minute RMS amplitudes of the waveforms in this study. Next, a day was divided into four 6-hour windows (0:00-5:59, 6:00-11:59, 12:00-17:59, and 18:00-23:59) to exclude short-term amplitude changes such as local earthquakes. The median of 360 RMS values in each 6-hour window was used as the representative value to obtain noise amplitude.
As a result, at the Owakudani and Owakudani-Jizoson stations located around the Owakudani fumarolic area, the amplitude of noise in the daytime (6:00-11:59 and 12:00-17:59) was about 10 times larger than that in the nighttime (0:00-5:59 and 18:00-23:59), because these seismometers are set on the ground and are affected by noise of cars and tourist facilities. However, when the eruption alert level was raised in 2015 and 2019, and when the first pandemic of COVID-19 occurred in 2020, the daytime noise amplitude was as low as that of night-time due to the closure of the facilities. In the two periods from June 2015 to May 2016 and from July 2022 to July 2023, the nighttime noise amplitude had been high. I consider that this night-time noise is likely to be unknown volcanic tremor, based on their long duration. On the other hand, at the Komagatake station in the Hakone area, the noise level was low because it is a borehole station and there is no noise source around the station. In addition, no nighttime noise amplitude changes were observed.
At the Omatazawa station in the Tanzawa mountain area, artificial noise has rarely been observed during the daytime because the station is located in the mountains, where there is no human habitation. Because this station is located along a river, when heavy rainfall of 100 mm or more per day was observed at the nearby JMA AMeDAS Tanzawa-ko, the noise level rapidly increased 10-50 times as common level, and the noise level gradually decreased over about two weeks. The noise amplitude may be corresponding to the flow rate of the river. Since there is a correspondence between rainfall and seismic noise amplitude, there is a possibility that the river flow rate can be estimated from the noise amplitude.