| Summary: | dissertation It is well known that the inversion of geophysical data is a nonunique problem. One of the approaches to reduce the nonuniqueness is based on the joint inversion of multiple geophysical data, if the corresponding physical properties are mutually correlated. This dissertation develops a method of joint inversion of seismic and gravity gradiometry data based on the Gramian constraints. The existence of an empirical linear relationship between the logarithms of seismic velocity and density was taken into account for the construction of a Gramian stabilizer, as well as the structural Gramian constraint with the gradients of velocity and density. The developed algorithm was based on modeling the seismic field using the integral equation (IE) method and different IE-based approximations. The joint inversion results of 3D synthetic model studies indicated that the resolution of both predicted density and velocity models was significantly improved. By nature, the recovered distributions of physical properties generated from the traditional inversions of potential field data are usually diffused. This dissertation develops a multinary inversion algorithm based on multinary transformation of the model parameters. An adaptive multinary inversion approach was introduced to automatically adjust the values of corresponding standard deviation during the inversion process. This approach was tested on several model studies and a case study in the area of Nordkapp Basin of the Barents Sea, which demonstrated to be effective in improving the quality of iv the gravity inversion for geological targets with sharp density contrast. In order to recover sharp and consistent boundaries of anomalous bodies from different geophysical data for mineral exploration, this dissertation also proposes an innovative approach to the joint inversion of gravity and magnetic data based on multinary transformation of the model parameters and Gramian constraints. The novel approach not only makes it possible to provide the sharp contrasts ...
|
|---|