Sparsity-promoting wide-angle least-squares one-way migration

For structural geology and geophysical exploration, it is of particular importance to get accurate images of salt domes, especially the subsalt weakly illuminated regions and the steep salt flanks. One-way wave propagators are widely used for imaging because of the high computational efficiency, low memory utilization and capability to handle models with large velocity variations. However, traditional one-way propagators cannot simulate waves propagating at large angles within 90^°, making it difficult to image steep structures. To overcome the limitation of imaging steeply dipping reflectors and weakly illuminated regions, we propose a sparsity-promoting wide-angle least-squares one-way migration method (SPWA-LSM). For details, the subsurface space is divided into several fan-shaped sub-domains, and the axis of each sub-domain is used as the local wave propagation direction. Then in each of the domains, we extrapolate the wavefields in different directions by rotating the propagation operator. Finally, we use the weight function to combine the extrapolated wavefields according to the propagation angle, thus obtaining the wide-angle propagation operator. The wide-angle propagation operator is extended to the least-squares migration framework (LSM), and then the transform domain sparsity constraint is added to the wide-angle least-squares migration (WA-LSM) to further improve imaging quality. Using the SEG/EAGE salt model and the BP2.5d model, we demonstrate the validity of the method in imaging steep subsalt structures.