3D pre-stack inversion constrained by spatial reflection features using the Sylvester's equation

Abstract

Reflection feature-constrained inversion (RCI) is a multichannel strategy that can effectively improve the stability and continuity of seismic inversion in complex geology, especially for pre-stack inversion with higher uncertainty. To address the limitations of conventional RCI to 2D profiles and high computational costs, a 3D pre-stack seismic inversion constrained by spatial reflection features was developed. First, a new approach to characterize seismic reflection features by decomposing seismic events along the vertical and lateral directions was proposed, which significantly reduces the size of the reflection feature operators. Second, we constructed reflection feature regularization terms and designed a fast reflection feature-constrained inversion (FRCI) for 2D seismic profiles. Finally, the proposed FRCI was extended to 3D by considering the spatial reflection features in two mutually perpendicular directions. Fast spatial reflection feature-constrained inversion (FSRCI) was proposed by constructing spatial reflection feature operators in a 3D coordinate system. The key to this method is that the parameter estimation is transformed into solving the 3D Sylvester equation such that the model parameters can be estimated directly from 3D seismic data. The inversion results of the overthrust model and field data showed that the proposed method can improve efficiency while ensuring spatial continuity. In summary, this study developed a fast pre-stack seismic inversion constrained by spatial reflection features based on the characterization of stratum spatial features, which is effective in complex geology, such as high-dipping and faults.