HR: 1340h
AN: S43B-1303    [Abstracts]
TI: The estimation of source position of long period seismic wave by Complex Meyer matching pursuit
AU: * Matsubayashi, H
EM: hiro_m@bosai.go.jp
AF: National Research Institute for Earth Science and Disaster Prevention, 3-1 Ten'nodai, Tsukuba, 3050006, Japan
AU: Matsumoto, T
EM: mtakumi@bosai.go.jp
AF: National Research Institute for Earth Science and Disaster Prevention, 3-1 Ten'nodai, Tsukuba, 3050006, Japan
AB: We have developed the method of long period source position estimation which uses Complex Meyer matching pursuit (CMMP). Our method can estimate the source position from seismic array data, and the estimation can avoid the influence of source mechanism. We generally analyze the seismic array data by semblance method. The method is useful for the detection of wave propagating direction and the wave source position estimation. But if you analyze array data by the method, the seismic wave has to be coherent wave and longer wavelength than array span, because the method calculates the index of seismic wave coherency. Therefore, the estimation has the effect of the non-isotropical earthquake mechanism. Our method also can estimate wave propagating direction and the wave source position. Furthermore, our method can avoid the influence of earthquake mechanisms. First, we use CMMP for seismic signal picking from array data. Then, we get the reciprocal number distribution of variance of the picking time difference, and estimate the largest number position as the source position. The CMMP picking has 3 merits. The picking can be automatic. Our method can avoid the influence of source mechanism, because Complex Meyer wavelet function varies as seismic signals. And then, the analysis time window of our method is defined by Complex Meyer wavelet function, automatically. We can expect that the analysis results by our method are more equality than semblance. We applied our method to estimation of the source position of the long period (16sec) tremor at Aso volcano. We used the Hi-net tiltmeters data. The distances between tiltmeters are 8 – 27km. We could get the estimated epicenter position of the tremor which was 2.8km southwest from the active crater of Aso volcano. And the depth of the estimated source was 1km under sea level. The large number distribution was on small region. The north-south length of the region was about 1km, and east-west length of the region was about 3km. We could not apply the semblance method to the estimation, because of incoherency of the array data. But we could use our method for the estimation. Therefore, we suggest that our method is a useful method for array data analysis.
DE: 7280 Volcano seismology (8419)
SC: Seismology [S]
MN: 2007 Fall Meeting