5:15 PM - 7:15 PM
[STT41-P06] Determining the Long-period ground motion level using a smartphone
Keywords:Long-period ground motion level, smartphone, Sva, Earthquake Zabuton
The measurement of seismic motion and seismic intensity using smartphones and relatively inexpensive MEMS sensors has become widely used. While there have been many evaluations of the accuracy of these observations, few reports have focused on Long-period ground motions. Regarding long-period ground motion, the Japan Meteorological Agency (JMA) has added the Long-period ground motion level to the announcement criteria for earthquake early warnings since February 1, 2023. The Long-period ground motion level is an index that classifies the magnitude of shaking into four levels based on the degree of damage caused by earthquakes in high-rise buildings of 14 to 15 stories or more.
In this study, we conducted an experiment with the aim of verifying the extent to which smartphones and relatively inexpensive MEMS sensors can be used to determine the level of long-period ground motion. In the experiment, an acceleration sensor was installed on a shaking table that reproduces long-period ground motion. The shaking table used was the Earthquake Zabuton (Hakusan Corporation), an earthquake simulator that can reproduce long-period ground motion with an amplitude of several meters. The smartphone used for evaluation was the iPhone 12 mini (Apple). The accelerometers used for comparison were the iPod touch (Apple), WT901BLECL (WitMotion), and JU-410 (Hakusan Corporation). The long-period ground motions reproduced by the Earthquake Zabuton were selected from the Noto Peninsula Earthquake (M7.6) on January 1, 2024, the Fukushima Offshore Earthquake (M7.4) on March 16, 2022, and the Kumamoto Earthquake (M7.3) on April 16, 2016, which recorded Long-period ground motion levels 4. For each earthquake, we selected one observation data set for each of the Long-period ground motion levels 1, 2, 3, and 4, and conducted experiments on a total of 12 long-period ground motions. The Japan Meteorological Agency's determination of long-period ground motion levels uses the maximum value of the absolute velocity response spectrum (Sva) obtained from the observation waveform of the seismograph on the ground. In this study, we also used the same method as the JMA to calculate the Sva in the range from 1.6 to 7.8 seconds from the observation waveform of the acceleration sensor installed on the shaking table, and determined the long-period ground motion level from the maximum value. We compared the results obtained from the experimental data with the Sva and long-period ground motion level published by the JMA on the web. We also compared the differences depending on the type of acceleration sensor installed on the shaking table. For the determination of the Long-period ground motion level, regardless of the magnitude of the Long-period ground motion level, the results of the determination using any of the accelerometers were almost the same as the JMA's determination. For the Sva, the results of the iPhone, iPod touch, and JU-410 were close to the Sva published by the JMA for any earthquake and Long-period ground motion level. For the relatively inexpensive MEMS sensor WT901BLECL, there was some variation from event to event due to the fact that the frequency measurement range was up to 5 seconds, but the difference in numerical values and the shape of the spectrum from the Sva published by the JMA was relatively large. As a result of this research, it was found that it is possible to measure the Sva and determine the Long-period ground motion level using the performance of the accelerometer built into the iPhone and iPod touch.
In this study, we conducted an experiment with the aim of verifying the extent to which smartphones and relatively inexpensive MEMS sensors can be used to determine the level of long-period ground motion. In the experiment, an acceleration sensor was installed on a shaking table that reproduces long-period ground motion. The shaking table used was the Earthquake Zabuton (Hakusan Corporation), an earthquake simulator that can reproduce long-period ground motion with an amplitude of several meters. The smartphone used for evaluation was the iPhone 12 mini (Apple). The accelerometers used for comparison were the iPod touch (Apple), WT901BLECL (WitMotion), and JU-410 (Hakusan Corporation). The long-period ground motions reproduced by the Earthquake Zabuton were selected from the Noto Peninsula Earthquake (M7.6) on January 1, 2024, the Fukushima Offshore Earthquake (M7.4) on March 16, 2022, and the Kumamoto Earthquake (M7.3) on April 16, 2016, which recorded Long-period ground motion levels 4. For each earthquake, we selected one observation data set for each of the Long-period ground motion levels 1, 2, 3, and 4, and conducted experiments on a total of 12 long-period ground motions. The Japan Meteorological Agency's determination of long-period ground motion levels uses the maximum value of the absolute velocity response spectrum (Sva) obtained from the observation waveform of the seismograph on the ground. In this study, we also used the same method as the JMA to calculate the Sva in the range from 1.6 to 7.8 seconds from the observation waveform of the acceleration sensor installed on the shaking table, and determined the long-period ground motion level from the maximum value. We compared the results obtained from the experimental data with the Sva and long-period ground motion level published by the JMA on the web. We also compared the differences depending on the type of acceleration sensor installed on the shaking table. For the determination of the Long-period ground motion level, regardless of the magnitude of the Long-period ground motion level, the results of the determination using any of the accelerometers were almost the same as the JMA's determination. For the Sva, the results of the iPhone, iPod touch, and JU-410 were close to the Sva published by the JMA for any earthquake and Long-period ground motion level. For the relatively inexpensive MEMS sensor WT901BLECL, there was some variation from event to event due to the fact that the frequency measurement range was up to 5 seconds, but the difference in numerical values and the shape of the spectrum from the Sva published by the JMA was relatively large. As a result of this research, it was found that it is possible to measure the Sva and determine the Long-period ground motion level using the performance of the accelerometer built into the iPhone and iPod touch.