Summary of gUS-Japan Cooperative Research for Urban Earthquake Disaster Mitigation Workshop on Prediction of Strong Ground Motions in Urban Regionsh , July 1, 2000, Tokyo
In this workshop, we had 11 technical presentations and discussions on the methodologies of strong ground motion prediction in urban regions. Through the discussion, we were able to have an agreement in the basic framework of characterizing the earthquake source and simulating the ground motions based on the source model.
The recent two large earthquakes, the 1999 Kocaeli, Turkey, earthquake and the 1999 Chi-Chi. Taiwan, earthquake brought a great increase of the near-source ground motion records, which will be most valuable data sets for the strong motion prediction researches. In this workshop, we had the reports on the source process inversions of both earthquakes. For the Kocaeli earthquake, applicability of the simplified source model based on the source inversion result was also checked. The inversion result of the Kocaeli earthquake shows a supersonic rupture velocity. It will be an important problem whether this is exceptional or not and how we should build this in strong motion prediction procedure. The source inversion result of the Chi-Chi earthquake shows a large slip at the northern shallow part on the fault plane, which agrees with the large offset seen in the surface fault trace. We discussed effects of peculiar phenomena such as plastic deformation of soils on seismic disaster near the surface ruptured fault. We also discussed the contribution of rise time to the pulse duration of the velocity waveforms close to the fault trace both in subsurface ruptured and surface ruptured faults. It is also important to find the main control factor of the pulse period of ground motions in near source.
The source inversion results of the 4 Japanese inland earthquakes shows that the empirical relation between the seismic moment and the combined asperity area obtained mainly for the California earthquakes (Somerville et al., 1999) holds for the Japanese earthquakes, too. It was reported that the two recent large earthquakes, the 1999 Kocaeli and Chi-Chi earthquakes, also satisfy this relation. This suggests high applicability of the empirical relation to characterize the earthquake source obtained by Somerville et al. (1999).
In the discussion on the simulation ways of the ground motions based on the characterized source model, we had an agreement on the importance of computer simulations and the accurate knowledge of the underground velocity structure, especially that of the shallow part. To obtain the accurate velocity structure, use of the borehole data, microtremor measurement, reflection survey, and computer simulation of waveforms from small earthquakes are recommended depending of the target depth and the situations. The importance of the Q-values at the shallow part was also emphasized.
The followings are the important points we realized through the discussion. Our future works will be to solve these problems and to put these factors into the recipe for predicting strong ground motions.