A New 1649-1884 Catalog of Destructive Earthquakes near Tokyo
and Implications for the Long-term Seismic Process

J. Geophys. Res., 111, doi:10.1029/2005JB004059., 2006
[Printable article (2.8 Mb)]

Elliot D. Grunewald and Ross S. Stein
U.S. Geological Survey, Menlo Park, CA

 

Summary: One of the most informative resources available in assessing Tokyo’s seismic hazard is the record of past earthquakes, which suggests the potential size, rate, and distribution of future earthquakes. Seismogram records are available for only the last century of seismic activity, and so the instrumental catalog does not reflect earthquakes which repeat over longer time scales. On the other hand, historical records of damage caused by earthquakes have been well documented in Japan for several centuries. These eyewitness damage descriptions have been reinterpreted as numerical intensity data and can be used to model historical earthquakes.

In this study, computational methods developed by Bakun and Wentworth were used to create location and magnitude models from intensity data for fifteen destructive earthquakes that occurred near Tokyo between 1649 and 1884. These models convey the important uncertainties in location and magnitude for each earthquake. The relative rate of small to large earthquakes in the historical catalog is consistent with rates from the instrumental catalog and geological records. This agreement suggests that the historical catalog is complete for large earthquakes. The extended catalog which combines both the historical and instrumental catalog expands the Tokyo seismic record to 350 years.

The long-term rate at which energy has been released in earthquakes was calculated from the extended catalog incorporating uncertainties in magnitude for each earthquake. This rate of energy release was then compared to the rate at which potential seismic energy accumulates, as determined by Nishimura and Sagiya using observations of slow deformation between earthquakes. The best estimates of accumulated energy and energy released in earthquakes are balanced over the past 350 years, which implies that the extended catalog is representative of the long-term seismic process. Therefore, the magnitude and frequency of these past earthquakes can be used to define probabilities of future destructive earthquakes. The catalog suggests that the probability of a M=7 or larger earthquake occurring near Tokyo during any 30-year period is 57%; the probability of a earthquake on the scale of the 1923 Great Kanto earthquake (M=7.9) is 7-11%.


Figure caption. Three versions of the 1649-1884 intensity-based catalog. (a) Intensity centers determined in this study using Bakun [2005] intensity attenuation models. (b) Epicenters inferred from intensity center-epicenter shift. (c) Usami [2003] epicenters.

 
 
 
participants
Designed and maintained by Volkan Sevilgen, USGS Disclaimer