Large Scale Aspherical Earth Structure

大量长周期数字地震数据的可用性在我们对地球大尺度非球面结构的理解方面发挥了重要作用。有许多新的模型被设计来拟合这些数据。他们在不同程度上取得了成功，但无可争辩的是，数据中的许多信号仍有待解释。其中一个原因是近似的理论公式被用来解释数据。在本提案中，我们描述了可以避免大多数近似的数据分析技术，并说明了一些近似技术能够执行得多么好。这些技术在大型数据集上的应用将导致非球面结构的许多积分约束，我们称之为结构系数。这些结构系数也可用于计算三维格林函数，以精确研究长周期源机制。

The availability of a large quantity of long-period digital seismic data has played a major role in recent advances in our understanding of the large-scale aspherical structure of the Earth. Many new models exist which have been designed to fit these data. They succeed to various degrees but it is indisputable that much of the signal in the data remains to be explained. One reason for this is that approximate theoretical formulations have been used to intepret the data. In this proposal, we describe techniques of data analysis which can avoid most approximations and we illustrate how well some approximate techniques are capable of performing. The application of these techniques to a large dataset will lead to many integral constraints on aspherical structure which we call structure coefficients. These structure coefficients can also be used to compute three-dimensional Green's functions for accurate studies of long-period source mechanisms.