[S-SS30_28AM2] New perspective of great earthquakes along subduction zones

[SSS30-P17_PG] Seismic observations off Kii Peninsula

3-min talk in an oral session

*Yojiro YAMAMOTO¹, Tsutomu TAKAHASHI¹, Yuka KAIHO¹, Koichiro OBANA¹, Ayako NAKANISHI¹, Shuichi KODAIRA¹, Yoshiyuki KANEDA¹ (1.JAMSTEC)

Keywords:Nankai Trough, ocean bottom seismographic observation, seismicity

In the Nankai Trough subduction zone, megathrust earthquakes of M 8 class occur repeatedly. There are three main seismogenic segments (Tokai, Tonankai and Nankai earthquake regions), and these segments have ruptured sometimes simultaneously and sometimes individually. To understand the control factor of the seismic linkage among these segments and Hyuga-nada segments, Japan Agency for Marine-Earth Science and Technology has been carried out a series of wide-angle active source surveys and local seismic observations from 2008 to 2012, as a part of Research concerning Interaction Between the Tokai, Tonankai and Nankai Earthquakes' funded by Ministry of Education, Culture, Sports, Science and Technology, Japan. In this study, we show the results of two local seismic observations off Kii peninsula, the one is in the Kii channel and the other is in the Kumano-nada. The boundary of the Tonankai and Nankai segments is located in this region (Baba and Cummins, 2005), and the existence of the high velocity plutonic rock in the landward plate just beneath Shionomisaki is considered as the control factor of historical rupture variation (Kodaira et al., 2006). Japan Meteorological Agency (JMA) catalogue also indicates the spatial relationship between the seismic activity and seismogenic segments; shallow microseismicity seems to be more active in the Nankai region than in Tonankai region.The observation in the Kii channel has been performed in FY2010 and was composed of 155 short-tern (about 1.5 months) ocean bottom seismographs (OBSs) and 19 long-term (about 10 months) OBSs. First, we relocated the JMA catalogue earthquakes by using three-dimensional velocity model obtained by active source surveys and adding the first arrival time data at OBSs. As a result, the earthquakes near the trough axis were generally relocated 10-20 km shallower than JMA location. Then, we attempt to detect the earthquakes by using long-term OBS records and found the active intraslab seismicity, especially in the up-dip part of the subducted seamount (Kodaira et al., 2000). The observation in the Kumano-nada has been performed in FY2011 and was composed of 150 short-tern (about 2.5 months) OBSs and 14 long-term (about 8 months) OBSs. Now we perform the first arrival picking of these data with the seismograph data of Dense oceanfloor network system for earthquakes and Tsunamis (DONET), according to the JMA catalogue earthquake list. We will show the preliminary results of hypocenter distribution in the Kumano-nada at the presentation.