PREDICTION OF REGIONAL STRESS AND CONSTITUTIVE RELATIONS BASED ON INVERSE ANALYSIS OF GPS DATA

本研究的目的是通过分析全国范围内GPS观测的位移和应变数据，预测区域应力和本构关系。主要结果有以下三点：1）理论上表明，位移增量对应于板块运动引起的地壳整体变形，而应变增量则代表了产生特定区域变形的区域非均匀性。2）为了预测地壳整体变形，提出了一种利用谱分解的反分析方法。该方法的主要特点是在考虑测量精度的基础上确定整体地壳形变的可测模式，并利用实测资料预测模式及其震源。3）对于区域地壳形变预测，提出了计算平均应变增量的公式，并利用应力反演方法进行了预测 ...更多信息 提出了一种新的应力增量估计方法。该公式给出了由三个GPS站组成的三角形区域上的平均应变，而无需求导。应力反演方法是利用Airy应力函数确定处于平衡状态的应力分量，利用实际GPS网数据进行的数值分析表明，利用这些方法可以预测区域应力和本构关系。区域应力增量和本构关系的预测值是合理的，但其有效性尚未得到验证。然而，它确实表明，分析方法的工作，为新的数据分析方法，在高空间分辨率测量的GPS网络数据。因此，预计目前的研究将继续下去，目前正在开发一种使用这些方法的所谓地壳变形监测仪。此外，计划通过与地壳结构和地震事件的比较来验证分析的有效性，并将该方法扩展到预测未来的地壳变形。少

This research is aimed at predicting regional stress and constitutive relations, by analyzing the displacement and strain data that are measured by the nation-wide GPS data. The following three are the major results :1) It is theoretically shown that the displacement increment corresponds to the overall crustal deformation caused by the plate movement, while the strain increment represents the regional heterogeneity that produces particular regional deformation.2) For the prediction of the overall crustal deformation, an inverse analysis method using the spectral decomposition is thus proposed. The key feature of this method is that measurable modes of the overall crustal deformation is determined by considering the accuracy of the measurement, and the modes and the associated source are predicted by using the measured data.3) For the prediction of the regional crustal deformation, the formulae to compute the average strain increment is developed and the utilization of the stress inver … More sion method is proposed to estimate the stress increment. The formulae gives an average strain taken over a triangular region formed by three GPS station without taking derivatives. The stress inversion method is to determine stress components which are in equilibrium, by making use of Airy's stress function.It is shown in numerical analysis that uses the actual GPS network data that the regional stress and constitutive relations can be predicted by using these methods. The predicted values of the regional stress increment and constitutive relations appears reasonable, even though the validity of the results are not verified. However, it is surely shown that the analysis method work for new data analysis methods of the GPS network data which are measured in high spatial resolution. Therefore, the continuation of the present research is expected, and a so-called crustal deformation monitor that uses these methods is being developed. Also, it is planned that the validity of the analysis is verified by comparing with the crust structure and the seismic event, and that the method is extended to predict future crustal deformation. Less