Diffraction imaging based on Marchenko equation

Abstract

Diffractions are the primary carriers of information about small scale subsurface heterogeneities, such as fractures, faults, karsts and unconformities. However, the several orders of magnitude difference in amplitude between reflections and diffractions make diffraction imaging difficult. Many strategies have been used during processing and migration, but only few studies ever consider the influence of the survey geometry on diffraction imaging. The reason that the acquisition system can be now taken into account is because of the development of Marchenko redatuming. Given the data excited and recorded on the surface, the Marchenko equation can be used to redatum the sources and receivers to any desired position in the subsurface, with only basic prior information about the subsurface velocity and density. This makes the idea of improving diffraction imaging by adapting the survey geometry possible. This study investigates how the depth and the shape of receiver sets influence diffraction imaging. A set of forward modelling results show that different layouts of the geometry do enhance the diffraction imaging and suggest two preferred layouts, receivers located in the subsurface as a line and a semi-circle. Then the Marchenko method is summarized from the literature by theoretically deriving the 1D and 3D Marchenko equations and practically introducing the workflow of Green’s function retrieval. In the final chapter, using data generated from the surface geometry, the Marchenko redatuming is used to move the sources and receivers to the favourable location chosen during the forward modelling. The amplified diffraction imaging results indicate that the whole process we propose in this thesis is feasible.