An earthquake of magnitude 9 rocked the Pacific Northwest on January 26, 1700 around 9:00 PM. A series of independent discoveries over the past two decades have narrowed the window of occurrence of this "great quake" to a precise time. Preliminary evidence consistant with this date came from carbon dating, tree ring studies, and soil deposits. Although these lines of evidence enabled scientists to verify that a great subduction zone earthquake struck the Pacific Northwest within one or two decades of 1700, the exact date date remained unknown.

In an attempt to narrow the time frame in which this great quake could have occurred, investigators turned to historic records. Although the Cascadia Coast had been populated by Native Americans for thousands of years, written records in Washington, Oregon and British Columbia only date back as far as the late 1700s when Europeans arrived. The lack of a local written record forced the search to continue elsewhere. A major breakthrough in the hunt for a written record of the great Northwest quake was found in Japan. Japanese records described a large tsunami of non-local origin (no local shaking) striking the coast of Japan on January 27, 1700, a date that fits perfectly into the initial estimate of around 1700.

Japanese Records
Kenji Satake, a researcher from the Geological Survey of Japan, along with a team of scientists from the University of Tokyo found Japanese records to tsunami occurrences along the country's eastern coastline between January 27 and 28, 1700. Careful analysis of these historic tsunami records indicated that several coastal villages were damaged. Accounts were recorded in different villages along Japan's coastline. The following is a chronicle by the head of Miho village, 145 km southwest of Tokyo. This account tells of sea water covering land as if it were high tide.

"The water also went into the pine trees of Ego. The receding water went out very fast, like a big river. It came in about seven times before 10 a.m. of that day and gradually lost its power…Because the way the tide came in was so unusual, and was in fact unheard of, I advised the villagers to escape to Miho Shrine…It is said that when an earthquake happens, something like large swells result, but there was no earthquake in either the village or nearby."

Further records indicate that at Miyako, located on the Pacific Coast of Japan, 20 houses were damaged from tsunami heights estimated to be 2 to 3 meters high. At Otsuchi, about 30 km south of Miyako on the coast, another document describes a tsunami occurring around midnight on January 27, engulfing rice paddies and damaging two houses. Here, the height of the tsunami is estimated at about 3 meters. At Tanabe, approximately 1000 km southwest of Miyako along the coast, a storehouse floor was flooded on the morning of January 28, 1700, indicating a 2-meter tsunami. In addition to these three locations, tsunamis of unknown height were documented at two additional locations on January 28, 1700, Nakaminato and Miho.

Initially, investigators believed it was possible that these waves originated from a meteorological origin such as a storm surge. However, the uniform heights of the waves distributed along Japan's coastline indicated that the waves were characteristic of a far-field tsunami meaning they could not have been produced by a storm surge. Furthermore, typhoons cause most storm surges in Japan, and typically occur only from August to October. In fact, on January 27 and 28, 1700 the weather over central Japan was recorded as sunny or cloudy.

Other Records
Once the possibility of a link between the great Northwest earthquake of 1700 and Japanese tsunami records was discovered, scientists set out to examine all the possible earthquake sources that might have generated the 1700 Japanese tsunami. Scientists estimated that in order to cause a tsunami in Japan on January 27, 1700, an earthquake in any of these regions would have had to be a great quake (M 8.5 or greater). In addition to Cascadia, several other earthquake prone areas around the Pacific Ocean have the potential to produce a tsunami. Three of these zones produced the largest earthquakes of the 20th century: Chile in South America, the Alaska-Aleutian region in North America, and Kamchatka in the Russian region of northern Asia.

Chile: In 1960, a tsunami caused by a great (Magnitude 9+) earthquake in Chile caused flooding and damage similar to and in the same areas as the 1700 tsunami. However, Chile was occupied by Spanish colonists long before 1700, but no earthquake in January of 1700 was recorded. animation of Chilean tsunami propagation model.

Alaska/Aleutians: The 1964 Alaskan Earthquake (Magnitude 9+), caused no significant tsunami damage on the portion of the Japanese coast affected in 1700. The angle between the earthquake source and the Japanese coastline inundated in 1700 probably allows Alaska to be ruled out as a possible source.

Kamchatka: The 1952 Kamchatka earthquake (Magnitude 9) caused some tsunami damage in Japan, but not in the area where damage occurred in 1700. Sparse written records from the Kamchatka Peninsula date to 1699, but no earthquake in 1700 is known. A strong Kamchatka earthquake was recorded in 1737, but no Japanese tsunami record has been found.

The Kuril Islands, an island archepelago between Japan and Kamchatka, also produces large earthquakes. No written record from the Kurils is known for the 1700 time period. The tsunami from a magnitude 8.1 Kuril quake in 1994 did not produce damage in the area affected in 1700.

Computer generated tsunami models allowed Kenji Satake and his team to calculate how long it took the tsunami to reach Japan's coastline from the Pacific Northwest. Because tsunami travel time from the Pacific Northwest to Japan is about 10 hours, and the first recorded tsunami arrived around midnight on 27 January Japan time, the origin time of the earthquake can be calculated at about 9:30 PM local time on January 26, 1700

More on this topic - A prize-winning essay from Jeff Chapin, Princeton University `98.

Printable Versions | References