DecNet: Decomposition network for 3D gravity inversion

Three dimensional gravity inversion is an effective way to extract subsurface density distribution from gravity data. Different from the conventional geophysics-based inversions, machine-learning-based inversion is a data-driven method mapping the observed data to a 3D model. We have developed a new machine-learning-based inversion method by establishing a decomposition network (DecNet). Unlike existing machine-learning-based inversion methods, the proposed DecNet method is a mapping from 2D to 2D, which requires less training time and memory space. Instead of learning the density information of each grid point, this network learns the boundary position, vertical center, thickness, and density distribution by 2D-to-2D mapping and reconstructs the 3D model by using these predicted parameters. Furthermore, by using the highly accurate boundary information learned from this network as supplement information, the DecNet method is optimized into a DecNet(B) method. By comparing the least-squares inversion and U-Net inversion on synthetic and real survey data, the DecNet and DecNet(B) methods have shown the advantage in dealing with inverse problems for targets with boundaries.

Automated summary

Three dimensional gravity inversion is an effective method for extracting subsurface density distribution, and machine-learning-based inversion maps observed data to a 3D model. The proposed DecNet method is a mapping from 2D to 2D, requiring less training time and memory space. It learns boundary position, vertical center, thickness, and density distribution, reconstructing the 3D model using these predicted parameters. The DecNet(B) method optimizes the DecNet method by utilizing the highly accurate boundary information as supplementary data, showing advantages in solving inverse problems for targets with boundaries.