论文部分内容阅读
浅部地球的非侵害研究是众多地球科学学科的一个基本环节。这种研究通常用地球物理方法来完成，这意味着要求解地球物理反问题。遗憾的是，几乎所有的地球物理反问题都有固有的非唯一性。本文描述了一种处理非唯一性的方法，可以相信该方法比这个领域中以前的方法更直接更完善，我们不采用正则化方法来约束问题的非唯一性，而是直接针对非唯一性并描述它。基本方法是产生并描述一个模型集，该集根据观测误差在可接受的限制范围内拟合数据。为表征解的不确定性和推断所有可接受模型的共性，用统计方法来分析这个也称为总体的模型集。另外该方法也给出问题线性程度的信息和制定各参数间的折衷方案。该方法具有普遍性，能用于众多地球物理反问题。我们用两个典型的地球物理反问题说明了这个方法。第一是加利福尼亚州Ｋｅｓｔｅｒｓｏｎ的跨孔走时应用。第二个应用由地面重力测量同时估计常数密度体的密度和形状。
Non-invasive study of the shallow Earth is one of the basic aspects of many earth science disciplines. This kind of research is usually done using geophysical methods, which means solving the inverse problem of geophysics. Unfortunately, almost all of the inverse problems of geophysics are inherently non-unique. This paper describes a non-unique approach to deal with, we can believe that the method is more direct and more perfect than the previous methods in this field, we do not use the regularization method to constrain the non-uniqueness of the problem, but directly against non-unique Describe it. The basic approach is to generate and describe a model set that fits the data within acceptable limits based on observed errors. To characterize the uncertainty of the solution and extrapolate the commonalities of all the accepted models, a statistical method is used to analyze this model set, also called the population. In addition the method also gives information on the linearity of the problem and a compromise between the various parameters. This method is universal and can be used for many inverse problems of geophysics. We illustrate this method with two typical inverse problems of geophysics. The first is the cross-hole travel time application at Kesterson, California. The second application estimates the density and shape of the constant density body from the ground gravity measurements.