Basic study for electromagnetic phenomena preceded by earthquakes in and around Tokai and Kanto regions

我们继续在东海和关东地区及其周边地区运营电磁监测站，条件良好。关于地壳形变观测，我们在研究期间部署了大约60个GPS站，这些站得到了不同预算的支持，以监测东海慢滑事件。此外，我们还继续进行了电磁波穿越的基础实验，这被认为是一种很有前途的监测地下空间时间变化的方法。在这个实验中，我们通过两周的数据叠加，在距离440Am信号源40公里的一个遥远的站点上成功地探测到了偶极矩信号。在水下电磁场实时监测方面，我们与日本科技大学合作，在滨岛岛外海部署了三分量磁强计和两分量电场仪。此时，我们成功地用高采样率(8赫兹)监测了一年多的时间，并清晰地检测到了同震电磁变化。通过观察，我们得出结论：三分量磁强计是最有希望检测DC-ULF范围内前兆变化的设备。日本岛，特别是日本中部和东京地区，大量的人工噪声主要来自直流驱动的列车的泄漏电流。因此，要获得高质量的数据是非常困难的。通过观测周期，我们发展了一些降噪方法(主成分分析；主成分分析、独立成分分析；独立成分分析；独立分量分析；对于数据泄露，我们开发了很多可以在网站上运行的初步分析程序。此时，一些观测点通过IP地址系统进行控制。

We continuously operate electromagnetic monitoring stations in and around Tokai and Kanto regions with good conditions. Concerning crustal deformation observations, we deployed about sixty GPS stations during the research period, which were supported by a different budget to monitor the Tokai slow slip event. Furthermore, we continued fundamental experiment of electromagnetic ACROSS, which is considered to be a promising method to monitor the temporal change of the underground. In this experiment, we succeeded to detect the signal by data stacking of two weeks in a remote station which was 40km away from the signal source of 440Am with the dipole moment. Concerning the underwater realtime monitoring of the electromagnetic fields, we deployed the three-component magnetometer and two-component electric field meter off the coast of Hatsushima Island in cooperation with JAMSTEC. At this moment, we succeeded to monitor more than one year with high-sampling rate (8Hz) and detected co-seismic electromagnetic changes clearly. Through the observation, we now conclude that the most promising device to detect the precursory changes in DC-ULF ranges is three-component magnetometer.Japanese Island, especially central Japan and the Tokyo areas, it is dominant that a lot of artificial noises mainly on the leakage current from DC-driven trains. Therefore, it is very difficult to obtain good quality of data. Through the observation periods, we developed some noise reduction methods (Principal Component Analysis ; PCA, Independent Component Analysis ; ICA, Weiner filter technique, Detrend Fluctuation Analysis and so on. Concerning the data disclosure, we developed a lot of preliminary analyzing programs which can operate on the website. At this moment, some observation points are controlled by using IP address system.