3D source-growing inversion of gravity gradient data based on depth weighting

Different from the regularization inversions, the inversions using systematic search algorithms do not solve the linear systems, which is computationally efficient. However, the shape of the lower part of the source cannot be accurately provided by this method. To overcome the problem, we proposed an improved version for this class of inversion of gravity gradient data based on depth weighting in this work. A new criteria is developed for the accretion by introducing depth information to the inversion. The proposed inversion has been tested by the synthetic model with different gravity gradients. It is demonstrated that the results of our method has high space resolution, and the new criteria can guide the source growing in the right direction. In addition, the proposed inversion has been proved to have anti-noise ability.