Technical Contributions

This dissertation makes several significant contributions to the field of seismic migration and imaging. The core contributions of my dissertation arise from the reformulation of the standard reverse-time migration (RTM) equation as the generalized diffraction-stack migration (GDM) equation. Using the GDM equation, I successfully decompose the kernel of the RTM imaging operator into products of incoming and outgoing Green's functions, which not only gives rise to a deeper understanding of the properties of different kernel components but also leads to an imaging algorithm with less coherent noise and a higher-quality migration image. A second contribution is that, based on the GDM algorithm, an antialiasing filter is developed for RTM. It is similar to the traditional antialiasing filter used for Kirchhoff migration but now provides RTM-like images mostly free of aliasing artifacts. A third contribution is the separation of the GDM operator into primary and multiple reflection components. Consequently, a migration image with a higher resolution is achieved by the use of multiple scattering. These contributions are validated by implementations of the GDM algorithm on both synthetic and field data. A fourth contribution is the development of a phase-encoded least-squares GDM algorithm. The benefit is an increased image resolution, better signal-to-noise ratio of the migration image, and an order-of-magnitude increase in computational efficiency compared to standard least-squares GDM. Finally, I show how the storage costs of GDM can be decreased by using a wavelet-transform compression scheme.