2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 3
Presentation Time: 2:00 PM


ATWATER, Brian F., U.S. Geol Survey, Department of Earth and Space Sciences, Box 351310, University of Washington, Seattle, WA 98195-1310 and SATAKE, Kenji, Active Fault Research Center, AIST, 1-1-1 Higashi, Site C7, Tsukuba, 305-8567, Japan, atwater@u.washington.edu

A newly authenticated account of a Japanese shipwreck on January 28, 1700, contains evidence that ordinary ocean waves met outgoing, river-mouth currents of a tsunami that may have lasted 20 hours in Japan.

The account, previously known to earthquake historians as an unattributed story in a fisheries-association book from 1943, has now been traced to a 19th-century collection of reports on 131 shipwrecks between 1670 and 1832. The 1700 tsunami plays a role in the official report from a local government’s inquiry into the loss of 470 bails of rice (=28 metric tons). The rice was seagoing cargo bound for Nakaminato, a river-mouth port 100 km northeast of Tokyo (then Edo). The rice boat approached the river mouth about 8 a.m. on January 28. At that time--equivalent to 3 p.m. January 27 at Cascadia--“high waves” prevented the boat from entering the river. The crew cast anchor offshore and stayed there until about 8 p.m., when an evening storm broke the anchor lines and drove the boat into coastal rocks. Two of the crew died and none of the rice was reported found.

The “high waves” that initiated this accident coincided with waves of the 1700 tsunami seen elsewhere in Japan. Though eight hours after the tsunami began with midnight flooding 350-380 km to the north, the “high waves” coincided with daylight flooding seen 250 km to the southwest, at Miho. There, a village headman saw the sea come in “something like a very high tide” and recede “with the speed of a big river.” He noticed “about seven” such oscillations in a four-hour interval that includes the time of the “high waves” off Nakaminato.

Those “high waves” probably represent not the tsunami by itself but instead the interaction of ocean waves with outflow from the tsunami. Eyewitnesses to the 1960 Chile tsunami at Nakaminato reported inflow of 7 knots (3.5 m/s) and outflow that went even faster. Water jetting seaward from river mouths commonly raises the heights of incoming ocean waves. A manual tells Oregon boaters, “If you are trapped outside a rough bar on an ebb tide, it is wise to lay to and wait until the flood current or inflowing current is dominant.”

Though it waned at Miho in the afternoon, the 1700 tsunami may remained large for longer at Nakaminato, as did the 1960 tsunami. If it continued to the evening storm, the 1700 tsunami lasted for 20 hours after its earliest recorded arrival in Japan.