15 Years Later: The Growing Legacy of the 1995 Kobe Earthquake

doi:10.1785/gssrl.81.1.5

Fifteen years have passed since the M 6.9 Hyogo-ken Nanbu earthquake struck the west side of Osaka Bay during the predawn hours of 17 January 1995. The earthquake started near the northeast end of Awaji Island and ruptured simultaneously to the northwest and southeast from there. To the southeast, right-lateral faulting extended about 20 km from the epicenter, and reached the Earth’s surface. To the northwest, rupture extended approximately 25 km, which brought it directly under the city of Kobe. Faulting did not quite reach the surface in Kobe, but the direction and speed of rupture exerted a strong forward directivity effect on the wavefield, and the intensity of ground motion was increased due to strong basin-edge effects. Strong shaking, widespread liquefaction, and numerous fires caused extensive damage in the city, and the earthquake came to be known as the Kobe earthquake.

More than 6,000 people lost their lives and direct economic losses exceeded 10 trillion yen (more than $100 billion). Other examples of direct hits from earthquakes on urban areas have resulted in much higher death tolls (e.g., Tokyo 1923, Tangshan 1976), but in economic terms, the Kobe earthquake is ranked as the costliest natural disaster in history. Before the earthquake, the port of Kobe was the busiest in Asia. It now ranks fourth in Japan and may never recover its pre-earthquake status.

The emergency response to the Kobe earthquake was beset with problems and has been widely criticized for making a bad situation worse. Perhaps that is so, but whenever a large city is suddenly and unexpectedly exposed to massive amounts of kinetic energy in the form of earthquake waves, it is inevitable that a lot of bad things will happen. A history of earthquake disasters demonstrates this and makes for bleak reading, but it is also true that earthquakes create opportunities for change that might otherwise not be possible.

I would like to take the opportunity of the 15th anniversary of the Kobe earthquake to draw attention to the important, and positive, role it has played in the subsequent development of earthquake preparedness and earthquake science. The Kobe earthquake changed government policies and public attitudes in Japan about earthquake preparedness. Earthquake response professionals were given greater authority to respond immediately to earthquake disasters. Substantial and sustained public education programs grew out of the disaster. It is also seen as a watershed event that instilled a spirit of individual volunteerism in Japan that had not previously existed.

The discoveries that have come out of the revolution triggered by the Kobe earthquake are leading to new insight into earthquake processes, not just in Japan but around the world.

The scientific response to the Kobe earthquake completely changed the focus of earthquake science in Japan. The discoveries that have come out of that revolution are leading to new insight into earthquake processes, not just in Japan but around the world. Perhaps this is blasphemous to write in the pages of one of the journals of the Seismological Society of America, but I believe the Kobe earthquake may come to surpass the 1906 San Francisco earthquake as the landmark earthquake in the history of our science.

Before the Kobe earthquake, the focus of concern in Japan was on a potential Tokai earthquake at the northern end of the Nankai trough south of Tokyo. After Kobe, the Japanese government embarked on a long-term program to develop a fundamental understanding of earthquake occurrence and to construct a national seismic hazard map. This was a strong shift away from the prediction-focused effort that had prevailed previously. To do this, the Japanese overcame what might be described as a tradition of territorialism in earthquake monitoring through a combination of greatly improved and unified earthquake monitoring networks. The scope of that effort is without precedent.

In earthquake engineering, it includes E-defense, one of the world’s largest shake tables, which is capable of 3-D testing of 1,200-ton structures. As a result of the Kobe earthquake the COSMOS-G2 and GRAPES GPS networks were integrated into a single nationwide network, known as GEONET (GPS Earth Observation Network), and about 400 continuously recording sites were added to it in the process. GEONET now measures crustal deformation of the Japanese islands with more than 1,200 continuously recording GPS sites. The most impressive improvements have come in seismological monitoring. These include a broadband seismic network (F-Net), extensive strong motion seismic networks (K-Net and KiK-Net), and perhaps most notable of all, a high-sensitivity borehole network (Hi-Net) that covers the entire Japanese archipelago. All of these are important, but it is Hi-Net that to me is most impressive. Developing and deploying a nationwide earthquake- monitoring network with more than 800 stations in boreholes hundreds of meters deep is exceptionally ambitious and costly. It is also an extraordinary amount of work to operate and maintain. A critically important aspect of these new networks is that most of the data are made widely available to researchers—not just across Japan, but throughout the world. The scientific payoff of this approach has been profound.

Given the combination of a rate of earthquakes at least five times higher than that of the western United States and superior earthquake monitoring networks in place in Japan (with more being deployed and others on the drawing board), I think it’s safe to assume that major discoveries in earthquake science will continue to be an important Japanese export.

Among the more spectacular results coming out of this effort are the discovery of deep, nonvolcanic tremor; the discovery of very low frequency earthquakes— both deep and in the accretionary prism; direct imaging of the Sumatra earthquake rupture; and most recently, the discovery of extreme vertical accelerations due to a “trampoline effect” in the 2008 M 6.9 Iwate-Miyagi earthquake. These are fundamental discoveries and developments that are changing the way we think about and study earthquakes. Many of these phenomena have subsequently been observed elsewhere. Given the combination of a rate of earthquakes at least five times higher than that of the western United States and superior earthquake monitoring networks in place in Japan (with more being deployed and others on the drawing board), I think it’s safe to assume that major discoveries in earthquake science will continue to be an important Japanese export.

The benefits of international exchange of ideas and insight are particularly evident in the case of episodic tremor and slip. Once tremor was documented in southwest Japan, scientists across the Pacific in Canada found tremor in Cascadia and demonstrated that it was correlated with large slow slip events. This motivated subsequent analysis, and discovery, of similar slow slip events in Japan. Since then there has been a continuous exchange of people, ideas, and insight across the Pacific and beyond. Tremor—both spontaneous tremor and tremor triggered by seismic waves from distant earthquakes—and other slow earthquakes continue to be discovered in diverse tectonic environments. It’s clear to me, as someone who is deeply involved in such studies, that we are still in the very early stages of understanding these newly discovered processes and their implications for earthquakes more generally.

Finally, it is worth noting that all of this capability and progress has drawn the attention of scientists around the world to the problems of earthquakes in Japan. Not many seismologists outside of Japan know much about the pre-Kobe earthquake history, but seismologists around the world follow the study of the more recent Niigata Chuetsu-Oki, Niigata Chuetsu, Tottori, and Iwata-Miyagi earthquakes with great interest. I believe the scientific legacy of the Kobe earthquake is still unfolding, but it is good to remind ourselves what progress can be achieved through the hard work and investment required for improved observational programs and through the important follow-up work to analyze data creatively. We should all be deeply grateful to our colleagues in Japan for making such important progress in earthquake science in the wake of the terrible tragedy of Kobe. 

Gregory C. Beroza
beroza [at] stanford [dot] edu