In 1855, a powerful earthquake struck the Japanese city of Edo (today’s Tokyo), killing thousands. The region sits atop multiple tectonic plates that have caused innumerable quakes over the centuries, and because the greater metropolitan area is now home to more than 30 million people, it’s critical to mitigate the threat. Japanese scientists have been examining historical records to better understand past quakes and have found that the autobiography of a Kabuki actor can shed light on the 1855 temblor.
A Time of Turmoil
The 1855 Ansei Edo quake, named for the Ansei imperial era of 1854–1860, came at a time of upheaval in Japan, both literally and figuratively. There were three great Ansei earthquakes: the Tokai and Nankai quakes, both in 1854 and both magnitude 8.4, and the Edo quake the following year, magnitude 7.0. Meanwhile, Japanese society was facing its greatest challenge in centuries. Having been under the hegemony of the Tokugawa shogunate, which implemented a policy of national seclusion for over 230 years, Japan was finally forced to open its doors to ships and trade by American gunboat diplomacy in 1854.
When Edo was hit on 11 November 1855, as many as 10,000 people lost their lives, and over 50,000 structures were destroyed in the temblor and subsequent fires. Some of the devastation can be seen in woodblock prints of the day that depict a giant underground catfish (Namazu) that was believed to have caused earthquakes when it thrashed about.
The Forgotten Manuscript
Fast-forward to 2020, and researchers at the University of Tokyo have found another way to use art to scientifically evaluate the 1855 calamity. Scientists analyzed a manuscript written by Kabuki actor Nakamura Nakazo III to infer the depth of the earthquake. In a poster presented at a joint conference of the Japan Geoscience Union and AGU (JpGU-AGU Joint Meeting 2020) in July, they noted that later editions of the manuscript had already been the basis for varying estimates of the quake’s hypocenter from relatively shallow in the crust to deep in the Philippine Sea plate. However, when the team analyzed Nakamura’s original handwritten manuscript of the autobiographical work Temae Miso (Self-Praise), recently acquired by Tokyo’s National Diet Library, it found a significant difference compared with later editions.
“A strong rumble occurred,” Nakamura wrote. “The women and children were surprised and screamed. I said, ‘Calm down, it’s a big earthquake.’ Omitu Bando said to me, ‘You should stand up rather than sit.’ I stood up. Then the strong shaking started, and I could not walk normally.” Instead of the first sentence, one later edition has “a strong upward movement came from the ground,” and where the writer describes standing, the later edition reads, “I stood up and walked. Then the strong shaking started….”
Researchers concluded that because the shaking began when Nakamura stood up instead of after he began walking, there was a relatively short period between the arrival of different seismic waves from the quake—in this case, the rumble and the shaking. Longitudinal, or P, waves are fastest and correspond to the rumble described by Nakamura. Transverse, or S, waves travel about half the speed and correspond to the shaking. Just as the distance to a thunderstorm can be estimated by the lag between a lightning flash and the sound of thunder, the S–P interval can suggest the distance to an earthquake’s epicenter.
The team concluded that the 1855 quake had an S–P time of 5–10 seconds and, because of the thick sedimentary layers of the Kanto region surrounding Tokyo, a relatively shallow depth of about 20 kilometers, which would place the rupture in the subducting Philippine Sea plate. Many researchers have estimated the depth at over 30 kilometers.
Such details are critical because the Japanese government believes there’s a 70% chance of another 1855-type quake in the next 30 years with as many as 23,000 casualties, according to poster coauthor Kenji Satake, director of the University of Tokyo’s Earthquake Research Institute. “Since the typical recurrence interval of large earthquakes is several decades to centuries,” Satake added, “we have to use other methods and data to study such large earthquakes in the past and the potential for the future.”
“Ground shaking and earthquake damage is larger for shorter hypocentral distances,” said coauthor Ryoichi Nakamura, another member of the institute. “Because the 1855 earthquake occurred right beneath Tokyo, the depth strongly affects ground shaking and damage.”
William Ellsworth, a professor of geophysics at Stanford University who was not involved in the research, believes the poster agrees with other findings.
“Seismologists have debated [the quake’s depth] for more than a century,” said Ellsworth. “The plates colliding beneath Tokyo provide a wide range of possibilities, both deep and shallow. The recent paper by Nakamura et al.…makes clever use of reports of the shaking to argue for a relatively shallow depth. Their work supports the conclusion of William Bakun, who used other historical accounts of the earthquake shaking to determine its magnitude, location, and depth.”
The poster is part of a greater interdisciplinary effort at the University of Tokyo. Seismologists teamed up with historians from the Historiographical Institute in an effort called the Collaborative Research Organization for Historical Materials on Earthquakes and Volcanoes. Inaugurated years after the catastrophic magnitude 9 Great East Japan Earthquake of 2011, its aim is to improve seismic understanding by compiling a long-term database of events based on historical materials. That means analyzing obscure records like Nakamura’s manuscript, written in a highly cursive hand that only experts can decipher.
“Different kinds of materials provide different kinds of information on earthquakes,” said poster coauthor Reiko Sugimori, an associate professor in the Historiographical Institute and the only team member who was able to read the Kabuki actor’s manuscript. “Earthquake casualties or damage in each village were summarized as reports, which are useful to estimate the distribution of seismic intensity, from which earthquake location and size can be estimated. On the other hand, daily records or personal diaries, written by the same person in the same location, can provide homogeneous daily records of seismicity, including foreshocks or aftershocks. Pictures are also useful because they provide visual records of earthquake damage.”
Researchers plan to continue adding details from historical materials to their historical seismic event database, and their work highlights the importance of long-term seismic knowledge. Later this month, Seismological Research Letters will publish a focus section coauthored by Satake on historical seismology.
—Tim Hornyak (@robotopia), Science Writer