2012 年 65 巻 2 号 p. 205-218
The Japan Meteorological Agency has deployed strain-meters in Shizuoka, Aichi, and Nagano Prefectures, aiming to detect a slow slip phenomenon that is expected to occur just before the anticipated great earthquake of M8 class at the plate boundary. In this paper we propose a stacking method in which data at different stations are added according to the following procedure. First, the observed waveform at each station is normalized by the noise level. Then, the normalized wave form is stacked one by one in the order of the magnitude of the signal to noise ratio, by reversing the polarity considering an assumed source location, if necessary, so as the signal to be all positive. This stacking procedure is stopped when the synthesized signal to noise ratio of the stacked waveform becomes the maximum. Members in the optimum data set that gives the largest signal to noise ratio become different if the location of the assumed source is different. It is shown that the signal to noise ratio of the stacked waveform obtained by this method is enhanced 2.3 times of that of the most favorable individual data. A slow slip event as small as Mw 5.0 would be detected if it occurs within the source area of the anticipated Tokai earthquake. We think the stacking method is applicable not only to the strain-meter data, but also to the GNSS and tilt-meter data and such an extension of the method may be effective to detect rising of magma beneath volcanoes.