Angle-domain common-image gathers for wave equation migration using regularized inversion

Abstract

Extracting angle domain common-image gathers (ADCIGs) form wave equation migration is an essential process for geophysical inversion techniques such as migration velocity analysis and amplitude-variation-to-angle inversion. Efficiency and robustness remain two major issues when developing wave equation based ADCIGs extraction methods. Among many other wave equation based ADCIGs generation methods, the extended imaging condition based approach is target-oriented and computationally efficient, makes this method promising for large-scale 3D imaging problem. However, the classical extended imaging condition based ADCIGs extraction method introduce some smearing artifacts appear as crossing events, which will decrease signal to noise level of the generated ADCIGs In this paper, we propose a modification to the conventional extended imaging condition based angle gather generation method. We reformulate the process for calculating ADCIGs as a regularized least square inversion problem. We illustrate that by replacing the traditional extended imaging condition mapping method with least square regularized inversion, the mentioned smearing artifacts in calculated ADCIGs is significantly attenuated and higher signal to noise level can be achieved. Numerical results with both synthetic and field data validate the effectiveness and efficiency of the present method.