[S04-P] Poster

[S04-P-02] Source area and magnitude of an aftershock following the 1854 Ansei-Nankai earthquake

^○Haruo Horikawa¹, Ichiro Nakanishi² (1.National Institute of Advanced Industrial Science and Technology, Ibaraki, Japan, 2.Kyoto University, Kyoto, Japan)

Great earthquakes repeatedly occur along the Nankai trough, where the Philippine Sea plate is subducting beneath the Eurasian (or Amurian) plate. The 1854 Ansei sequence have two great earthquakes: the eastern part (Tokai region) ruptured first and the western part (Nankai region) ruptured about 32 hours later. Historical documents show that the Ansei Nankai earthquake associated several aftershocks with strong shakings. The source region and magnitude of the aftershocks have been inferred from the descriptions of historical documents, but an exception is an aftershock that occurred on the New Year's eve on a Japanese classical calendar, which strongly shook Kochi, close to the plate boundary. In this study, we try to quantitatively estimate the magnitude and source region of this aftershock. Firstly, we inferred the seismic intensity from historical documents. The historical documents show that the ground motions were prolonged and suggest that the ground motions were rich in long-period components. Next, we estimate the source region and magnitude from the seismic intensity data, evaluating the residuals between the observed (interpreted) seismic intensities and the calculated ones with the aid of the Bayesian statistics. The seismic intensities were calculated with an empirical attenuation relationship, in which a finite fault was assumed. It is highly probable that the magnitude was larger than seven, and we found that the source region would be located to the southeast from Kochi. However, the inferred source location is not consistent with the historical descriptions that the number of felt earthquakes around Kochi greatly increased after this event, which suggests that seismicity increased around the region. Our inferred location is not closely located to Kochi. This inconsistency might be responsible for the effect of rupture directivity, which is not taken into consideration in the attenuation relationship used in this study.