[S-SS11] Active faults and paleoseismology

[SSS11-13] The unparalleled ability of Michizane Sugawara suggests the possibility that the August 887 Nin'na Earthquake was a Non-Nankai Earthquake

*Ritsuko S. Matsu'ura¹ (1.Earthquake Research Center, Association for the Development of Earthquake Prediction)

Keywords:Michizane Sugawara, Chronological changes in the number of events recorded in historical documents, 887 Nin'na so-called Nankai earthquake, size-predictable model

Fig. 1 shows the cumulative number of days on which at least a felt or damaging earthquake was recorded in a reliable historical document in Japan, based on the Ancient and Medieval Earthquake and Volcanic Eruption Database β-ver. (e.g. Ishibashi, 2009). The database contains earthquake data from 599 to 1607, and ghosts and any doubtful earthquakes are eliminated to the extent possible. In the original database, aftershocks within a week or a few months after a main shock were clumped in a single event of a main shock. We count days of aftershocks with one or more events, since the precise number of aftershocks is often unknown. Unless there is clear separation of events based on area and/or time of day, the date in multiple historical documents is counted as one day. Only a few earthquakes of the Tohoku district, such as 869 Jogan were recorded during the pre-Edo period. Prior to Ieyasu's relocation, too few earthquakes in Kanto district were recorded in historical records, too. Most of the earthquakes in this figure were felt in the southwestern Japan, especially in and around the Kinai district. The cumulative number of similar days in compilations by Usami (blue line in Fig. 1) resembles to the yellow line. Fig. 1 denied the commonly held belief that more earthquakes were recorded in the ancient era than in the medieval era, even if we set the start of the medieval era as 1086, when the Cloistered Emperor Shirakawa started his government. It is apparent that the recorded rate comparable to that in the end of the 16th century (Hohshoku period) was limited to the later half of the 9th century in the ancient era. That period is nothing but the time of Michizane Sugawara. "Nihon Sandai Jitsuroku" is the last of the six national histories of ancient, and it is the most detailed account of ancient Japan. The credit for it should be attributed to Michizane, who oversaw compiling it until just before he was transferred to Dazaifu. Fujino and Matsu'ura (2023) argued that the August 887 Nin'na earthquake was not a Nankai earthquake, but rather the latest activity of the Osaka Bay Fault. Fig. 1 shows that Michizane, who was in Sakaide at the time of the August 887 Nin'na earthquake, had so little chance to drop recording the tsunami damage in Shikoku and the cessation of Dogo hot springs, which would have accompanied with a Nankai earthquake, as well as the damage to the manor houses in Tosa and Awa, which would have been of great interest in Kinai. Fig. 1 reinforces the argument (Fujino and Matsu'ura, 2022) that it is highly unlikely that the authentic history of the Nankai Earthquake in the Sugawara's days would have allowed the statement that "the tsunami damage was greatest in Settsu" and omitting the damage in southern Shikoku, which would not have been missed in Kinai. Even Akitsune Imamura, who had managed to pick both the 1498 Meio and 1605 Keicho earthquakes as Nankai earthquakes to explain away his 100-year cycle belief, had not identified the 887 Nin'na as a Nankai earthquake before 1940. However, the "common sense" that Imamura developed since the 1940s has remained in Ishibashi (2023) and others. It should be noted that it has been conveyed in a rambling manner without modern criticisms.
The size-predictable model is far better suited to the size and occurrence intervals of the three recent Nankai earthquakes of Showa, Ansei, and Hoei, than the time-predictable model. Is it necessary to force ourselves to believe that the Nankai earthquake must have occurred 200 years after Hakuho and to seek the Nankai earthquake in the period of Michizane? At this point, I would like to raise this issue, because I believe that having the courage to clarify the uncertainties of historical earthquake catalogs with an open mind is a step toward a correct understanding of the real seismic hazard in Japan suitable in the 21st century.
References
Database(2020 Mar. ver.): https://historical.seismology.jp/eshiryodb/
Fujino and Matsu'ura (2022) Abstract of Fall meeting, Japanese Seismological Soc., S10-09.
Fujino and Matsu'ura (2023) JPGU2023,S-SS13-04.
Ishibashi (2009) Zisin 2, 61, S509-S517.
Ishibashi (2023) Zisin 2, 76, 55-61.