1:45 PM - 3:15 PM
[MIS04-P02] Development of an LF-band broadband interferometer to identify LF-band signals associated with earthquakes
There are reports of earthquake precursor phenomena related to the LF band. Typical example is the report of an increase in the number of LF pulses at the Uji station before the Hyogo-ken Nanbu earthquake. However, Some of the increased pulses were found to be due to lightning activity. Discrimination between signals due to lightning activity and those related to earthquakes is a challenge for the identification of earthquake-related LF-band signals. On the other hand, the development of interferometry is possible nowadays because of the development of ICT and GPS technologies compared to those days. An interferometer is a system that observes changes in electric fields at multiple locations separated by several tens of kilometers and uses the phase differences in the waveforms obtained at each location to make time and space estimates of the electromagnetic wave sources. In this study, as part of the development of the "LF-band broadband interferometer," we detected earthquake-related LF-band signals by waveform survey using one interferometer element and investigated the number of pulses. We also evaluated the performance of three interferometer elements as an interferometer system.
A capacitive circular flat plate antenna (diameter 33 cm, made of aluminum, time constant 30 µs, lower frequency limit 5.3 kHz) was used as the interferometer element antenna. The circular flat-plate antenna has the advantage that it is less likely to generate corona discharges, which can be noisy for detailed waveform observation. In addition, the sensitivity of the antenna in this study is about 10 to 20 dBm higher than that of the Bolt antenna used for lightning observations. The received signal passes through a 500 kHz low-pass filter and is converted to AD with a sampling frequency of 4 MHz and a resolution of 16 bits.
Observations by an interferometer element began on October 4, 2022, on rooftop of the Faculty of Science Building, Chiba University(35.63°N, 140.10°E), and continued as of February 2022. During this period, two earthquakes of M4.9 within a radius of 100 km from the station occurred in November 2022. In this study, we attempted to detect the LF signals associated with these two earthquakes. Pulses with amplitudes in the top 15% of all pulses during the two weeks before and after the two earthquakes (October 21, 2022, to November 25, 2022) were selected for analysis, and pulses considered to be caused by cloud-to-ground flashes were excluded based on lightning discharge (cloud to ground discharge only) location estimated by blitzortung.org and pulse waveforms. As a result, an anomalous increase in the number of pulses due to very short-period pulses was observed prior to the two targeted earthquakes, which is unlikely to be caused by cloud to ground discharge.
Also, we started observations with three interferometer elements at the rooftop of the Faculty of Science, Chiba University on December 7, 2022, to evaluate the performance of the three interferometer elements as an interferometer system. The required performance as an interferometer is the accuracy of the trigger timing and the recorded waveforms. Specifically, since the observations are made at the same location, the trigger timing difference should be 30~40 μs (within 10 km target error), and the rise time of the waveforms should be same. As a result, there was no problem for the interferometer system with respect to the waveform, because the same waveform was recorded, but the time difference of triggering often exceeded 40µs between elements, and no trend was observed between elements. It was found necessary to improve the system regarding time synchronization. After the improvement, we will start the installation of the LF broadband interferometer.
A capacitive circular flat plate antenna (diameter 33 cm, made of aluminum, time constant 30 µs, lower frequency limit 5.3 kHz) was used as the interferometer element antenna. The circular flat-plate antenna has the advantage that it is less likely to generate corona discharges, which can be noisy for detailed waveform observation. In addition, the sensitivity of the antenna in this study is about 10 to 20 dBm higher than that of the Bolt antenna used for lightning observations. The received signal passes through a 500 kHz low-pass filter and is converted to AD with a sampling frequency of 4 MHz and a resolution of 16 bits.
Observations by an interferometer element began on October 4, 2022, on rooftop of the Faculty of Science Building, Chiba University(35.63°N, 140.10°E), and continued as of February 2022. During this period, two earthquakes of M4.9 within a radius of 100 km from the station occurred in November 2022. In this study, we attempted to detect the LF signals associated with these two earthquakes. Pulses with amplitudes in the top 15% of all pulses during the two weeks before and after the two earthquakes (October 21, 2022, to November 25, 2022) were selected for analysis, and pulses considered to be caused by cloud-to-ground flashes were excluded based on lightning discharge (cloud to ground discharge only) location estimated by blitzortung.org and pulse waveforms. As a result, an anomalous increase in the number of pulses due to very short-period pulses was observed prior to the two targeted earthquakes, which is unlikely to be caused by cloud to ground discharge.
Also, we started observations with three interferometer elements at the rooftop of the Faculty of Science, Chiba University on December 7, 2022, to evaluate the performance of the three interferometer elements as an interferometer system. The required performance as an interferometer is the accuracy of the trigger timing and the recorded waveforms. Specifically, since the observations are made at the same location, the trigger timing difference should be 30~40 μs (within 10 km target error), and the rise time of the waveforms should be same. As a result, there was no problem for the interferometer system with respect to the waveform, because the same waveform was recorded, but the time difference of triggering often exceeded 40µs between elements, and no trend was observed between elements. It was found necessary to improve the system regarding time synchronization. After the improvement, we will start the installation of the LF broadband interferometer.