Seismic Coherence Driven Sparse Mesh for CSEM Inversion

Abstract

Many geophysical inverse problems involve large and dense coefficient matrices that often require an immense amount of computing power. Some methods can be used to reduce the processing time or physical memory required. This paper pose a significant challenge to solve large-scale inverse problems. We have developed a method that combines the adaptive mesh discretization and sparse mesh to reduce the computational complexity of CSEM inverse problems. The sparse mesh is created by Delaunay Triangulations method and constrained by seismic image. The nodes for generating triangular mesh are extracted from seismic coherence map. This sparse mesh is including all the information of geological features which are extracted from seismic image. A synthetic CSEM data are simulated for sparse mesh testing. As a result, the seismic coherence driven sparse mesh has an as high resolution inversion result as normal unstructured triangular dense mesh. Comparing with the same number unstructured triangular sparse mesh, seismic coherence driven sparse mesh has an advantage of vertical resolution.