Japan-France Joint Project on Strong Motion Prediction

在1995年阪神-Awaji地震后，我们强烈地重新认识到，在人口稠密的城市下方发生活动断层的地震，对强震进行精确预报的必要性。日本和法国都有大量的活动断裂，这次日法联合项目的目的是如何用理论和半经验方法预测内陆地震期间的强震。为了检验强震预报方法的有效性，我们共享了1995年神户大地震的强震数据库，该数据库不仅是关于强震记录的最完整的数据集，而且是关于震源和地下结构模型的最完整的数据集之一。我们还分享了部署在阿希加拉山谷、久路地区(日本)和沃尔维山谷(希腊EUROSEISTEST)的几个密集的强震台网。使用这些相同的数据，我们尝试了几种方法进行强运动预测(实际上是后处理)，比较了这些方法的结果，讨论了这些方法的应用能力和局限性。作为预测方法，我们采用了理论方法(波场的完全数值模拟)和半经验方法(经验格林函数法)。为了估计场地效应，我们比较了几种方法，如传统的谱比法、微震数据分析法等。1997年1月，我们在法国格勒诺布尔举行了一次关于强地面运动预测的小型研讨会。我们和许多致力于这项研究的研究人员一起出席并展示了他们的工作，共同讨论，以获得目前详细的强地面运动预测的共识，以及未来的发展方向。我们制作了包含我们两年工作的报告和与会者写下的简短报告。

After the Hanshin-Awaji disaster in 1995, we strongly re-recognized the necessity of the precise prediction of strong motions by earthquakes occurred active faults beneath the high-populated cities. There are many numbers of active faults both in Japan and France, and the aim of this Japan-France joint project is how to predict strong motions during inland earthquakes both of theorctical and semi-empirical methods. To check the validity of strong motion prediction methos, we share the strong motion database about the 1995 Hyogoken-Nanbu (Kobe) earthquake, which is one of the most complete data set about not only strong motion records but source and underground structure models. We also shared several dense strong motion networks deployed at Ashigara valley, Kushiro area (Japan), and Volvi valley (EUROSEISTEST,Greece). Using those same data, we tried strong motion predictions (actually, post-diction) by several methods, compared those results, discussed about the ability and limitations of applications of those methods. As the prediction methods, we used the theoretical method (complete numerical simulations of wavefields), and semi-empirical method (empirical Green function method). To estimate site effects, we compare several methods such as traditional spectral ratio, microtremor data analysis, etc.On January, 1997, we had a small symposium about the prediction of strong ground motions at Grenoble, France. We and many rescarchers conceming to this study attended and presented their works and discuss together to get the consensus of the strong ground motion prediction now in detail and where we will go from here in the future. We made the reports which contain our works for two years and the brief report which the attendance wrote down.