Assessing earthquake probability in Greater Tokyo
How likely is the repeat of a devastating earthquake in the densely populated metropolitan area of Tokyo, Japan's capital? Answering this question is one of the goals of a new, Swiss Re-sponsored scientific study designed to better assess the probability of earthquake occurrence in the Greater Tokyo area.
As an international collaborative effort between the Active Fault Research Center (AFRC) in Japan and the US Geological Survey (USGS), the joint research team will apply cutting-edge stress-transfer modelling to derive a probability assessment. The expected outcome is an improved assessment of the likelihood of the Greater Tokyo area being hit by a major earthquake. Further, this study intends to complement the ongoing preparation of a seismic hazard map undertaken by the Japanese Government.
Tokyo is one of the world's most densely populated metropolitan areas with a high concentration of residential and commercial buildings as well as industrial facilities. For any property insurer writing Japanese property exposure, a major earthquake in Tokyo is one of the key loss scenarios to be prepared for. In 1923, the Greater Tokyo region suffered one of the world's most destructive earthquakes, known as the Great Kanto earthquake, resulting in 143,000 fatalities, destroying two-thirds of Tokyo, and all of Yokohama. Today, the population of Greater Tokyo is six times larger than it was in 1923. A repeat of the Kanto earthquake (M=7.9) would cause uncounted casualties and enormous property damage.
The scientific objective of this project is the development of a comprehensive description and understanding of earthquake occurrence in Greater Tokyo which will lead to time-dependent probabilistic forecasts for large earthquakes striking the region during the coming years to decades. The achievement of this goal is supported by the unique collaborative effort bringing together specific Japanese capabilities and data resources and expertise in advanced stress-transfer calculations already successfully applied to California and other seismic-prone regions of the world such as Turkey.
An important requirement of the model will be its capability of explaining the salient features of the historical earthquake occurrence and deformation. A 400-year record of large earthquakes in and around Tokyo together with the availability of records from the most advanced instrumental seismic and geodetic networks in the world will provide an ideal testing and validation environment.
The uniqueness and originality of this study is expected to complement and contribute to a landmark project being undertaken by the Japanese government and the Japan Earthquake Research Institute to produce the comprehensive national earthquake probability map of Japan by mid-2005.
The results of this Swiss Re-sponsored study are expected to be published for public and scientific community reference before the end of 2004, with summaries of key results both in Japanese and English.
This is the second time Swiss Re has sponsored a scientific study involving Ross Stein, Shinji Toda and their team. In 1999, a similar study focused on the Greater Istanbul area in Turkey. This study determined that the devastating 1999 Marmara Sea earthquake significantly heightened the odds of a major earthquake hitting Istanbul in the near future.
Relevant downloads and links
EQ Probability Investigation of Greater Tokyo - Nihongo
EQ Probability Investigation of Greater Tokyo - Tokyo study in detail
Earthquake Conversations, by Ross S. Stein - an introduction to stress-transfer theory
Heightened odds of large earthquakes near Istanbul - scientific paper on the 1999 study
Swiss Re publication on the 1999 Istanbul study
Random Occurrence or Predictable Disaster?
