Investigations of the Piezomagnetic Field in the Slow Slip Event Epicentral Area in Tokai Area, Central Japan
Abstract
The University of Tokyo has conducted continuous geomagnetic observations in the Tokai area, central Japan. In this area, large interplate earthquakes have occurred repeatedly due to subduction of the Philippine Sea Plate. The main objective of our geomagnetic observations is to detect tectonomagnetic signals accompanied with the subduction process. At each station, total forces of the geomagnetic field are measured every minute using proton precession magnetometers. Observed value at the reference station in Yatsugatake, several tens of km away from the investigated area, was subtracted from those at respective stations to establish daily mean simple differences. The data at one station show a characteristic variation. Until 2000, it had decreased at a rate of 1 nT/year. However, the decrease had stopped during a period from 2000 to 2004. And it started decreasing again in 2005. The period during which the decrease had stopped corresponds to the Tokai Slow Slip Event which was detected by geodetic observations. Therefore, the variation found in the total force differences may have some relation to the Slow Slip Event. To investigate the possible relation between the changes in the geomagnetic total force and the Tokai Slow Slip Event, we performed piezomagnetic modelings. In general, piezomagnetic field is generated by the heterogeneities of the initial magnetization of rocks and/or those of stress field in the crust. To clarify which heterogeneity is dominant for explaining the observational results in Tokai area, we conducted two types of numerical modelings, one for the uniformly magnetized crust with the realistic stress field, and one for the uniform regional stress field with the highly magnetized rock bodies near the station. In the former case, the stress field variation calculated from the slip distribution estimated by Ohta et al. (2004) was substituted to the calculation. Expected changes in the total intensity amount to only less than 1/100 nT. These changes are far smaller than that can be observed. Therefore, the contribution from the uniformly magnetized crust can be ignored for explaining the observed changes. In the latter case, on the other hand, expected magnitude of the changes reaches a few nT, which is as the same order of magnitude as the observed one. Since there are intense magnetic anomalies around the station on the eastern edge of the Slow Slip Event epicentral area, the latter case is more plausible. In both of these modelings, we assumed the stress sensitivity of the rocks to be 1× 10^{-8} Pa-1. This value is by one order larger than that determined from laboratory experiments for intact rocks. However, it is consistent with the values obtained by experimental studies of some porous rocks (Hamano, 1983) and those estimated with the geomagnetic changes observed during the filling of reservoir (e.g., Davis and Stacey, 1972).
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2006
- Bibcode:
- 2006AGUFMGP43A1014Y
- Keywords:
-
- 1540 Rock and mineral magnetism;
- 1555 Time variations: diurnal to decadal;
- 7240 Subduction zones (1207;
- 1219;
- 1240)