Gravity data inversion using high-order polynomial function of density contrast varying with depth

The gravity inversion is one of the geophysical means for depicting the spatial distribution of mass bodies with density contrast. Conventional inversion methods directly invert density contrast values of each cell through both horizontal and vertical meshes. In this paper, complex density variations are approximated by polynomial functions, and a new method is proposed to determine density contrasts by inverting the coefficients of polynomial density functions. Different from the conventional inversion methods, this method can invert complex density contrasts without partitioning vertically cells. To some extent, it eases the contradiction between the quantity of mesh, memory occupancy, and inversion precision. Theoretical model tests show that polynomial coefficient inversion combined with multiple constraints can clearly highlight the position, scale and boundary information of local masses, and is superior to the conventional L2-norm inversion results. This proposed method is successfully applied to the identification of buried hills and sags in the Jiyang Depression. The boundaries of different lithologic bodies are roughly determined by density contrasts inverted, which makes up the gap of buried hill distributions showing on seismic sections. © 2019, Editorial Department OIL GEOPHYSICAL PROSPECTING. All right reserved.