5:15 PM - 7:15 PM
[SCG56-P02] Analysis of seismic and rainfall responses using multi-component strainmeter
Keywords:Strain, Groundwater, Post-seismic change, Post-rainfall change
Strain is one of the most important observations to monitor fine-scale crustal deformation phenomena, such as post- and aseismic slips, and volcanic precursors. To improve the accuracy of earthquake predictions along the Nankai Trough, the national institute of Advanced Industrial Science and Technology (AIST) has been developing a comprehensive groundwater observation network since 2006, aiming to establish approximately 20 observation sites to monitor plate movements. Each station observes crustal strain, seismic waves, and groundwater levels. Among these, strain is primarily measured using the Ishii-type strainmeter. This strainmeter is equipped with a displacement detection system utilizing a displacement magnification mechanism and a magnetic sensor, enabling it to record extremely small strain changes on the order of 10-9 strain. Due to its very high sensitivity, it records strain variations not only from plate motion but also from rainfall, seismic waves, and other factors. This study investigates these variations and their characteristics.
We visually inspected observation records from the AIST multi-component strainmeter over a fixed period to examine its responses to rainfall and seismic waves. For earthquake events with strain variations, we calculated seismic waves maximum root-mean-square and hypocentral distance to explore their relationship. To examine rainfall-induced responses, we analyzed six years of records from six stations, investigating horizontal strain variations and compiling a rainfall response catalog. We also examined correlation between rainfall response occurrence and precipitation amount, compared representative variations, and discussed rainfall response characteristics.
Strain data include variations from tidal effects and atmospheric pressure fluctuations. We used the tidal analysis program BAYTAP-G [Tamura et al., 1991] to separate these components and extract gradual trend components. Since BAYTAP-G extracted data include long-term trends, we set a long-term trend removal period of about one week, ensuring no prior variations, and subtracted the linear long-term trend estimated in the period by a least square method from the data.
Our investigation of seismic waves responses revealed multiple gradual strain variations attributed to seismic waves. The number of observed responses varied among sites, with more detected closer to the Pacific Ocean. The strain responses did not always appear with seismic waves, and most earthquakes did not affect strain records. Responses from seismic wave incidence did not correlate with the maximum amplitude of seismic waves, frequency-dependent maximum amplitude, hypocentral distance, or earthquake magnitude. This suggests that factors other than seismic amplitude and earthquake scale influence strain response occurrence.
We also investigated rainfall responses and confirmed strain variations in multiple components at all stations attributed to rainfall events. We found a positive correlation between precipitation and frequency of strain variations. At multiple stations, strain variation components and the directions of principal strain exhibited common patterns across many events.
Through visual inspection of the multi-component strainmeter records, we identified strain variations in response to both rainfall and seismic waves. Note that only a few seismic waves responses were confirmed, making statistical discussion difficult. To improve the reliability of the findings, extending the analysis period and investigating additional earthquake events is necessary. Regarding rainfall responses, analyzing precipitation amount, strain variation magnitude, and variation direction may provide insights into rainfall response mechanisms.
We visually inspected observation records from the AIST multi-component strainmeter over a fixed period to examine its responses to rainfall and seismic waves. For earthquake events with strain variations, we calculated seismic waves maximum root-mean-square and hypocentral distance to explore their relationship. To examine rainfall-induced responses, we analyzed six years of records from six stations, investigating horizontal strain variations and compiling a rainfall response catalog. We also examined correlation between rainfall response occurrence and precipitation amount, compared representative variations, and discussed rainfall response characteristics.
Strain data include variations from tidal effects and atmospheric pressure fluctuations. We used the tidal analysis program BAYTAP-G [Tamura et al., 1991] to separate these components and extract gradual trend components. Since BAYTAP-G extracted data include long-term trends, we set a long-term trend removal period of about one week, ensuring no prior variations, and subtracted the linear long-term trend estimated in the period by a least square method from the data.
Our investigation of seismic waves responses revealed multiple gradual strain variations attributed to seismic waves. The number of observed responses varied among sites, with more detected closer to the Pacific Ocean. The strain responses did not always appear with seismic waves, and most earthquakes did not affect strain records. Responses from seismic wave incidence did not correlate with the maximum amplitude of seismic waves, frequency-dependent maximum amplitude, hypocentral distance, or earthquake magnitude. This suggests that factors other than seismic amplitude and earthquake scale influence strain response occurrence.
We also investigated rainfall responses and confirmed strain variations in multiple components at all stations attributed to rainfall events. We found a positive correlation between precipitation and frequency of strain variations. At multiple stations, strain variation components and the directions of principal strain exhibited common patterns across many events.
Through visual inspection of the multi-component strainmeter records, we identified strain variations in response to both rainfall and seismic waves. Note that only a few seismic waves responses were confirmed, making statistical discussion difficult. To improve the reliability of the findings, extending the analysis period and investigating additional earthquake events is necessary. Regarding rainfall responses, analyzing precipitation amount, strain variation magnitude, and variation direction may provide insights into rainfall response mechanisms.