Least-squares reverse-time migration with sparsity constraints

Abstract Least-squares reverse-time migration (RTM) works with an inverse operation, rather than adjoint operation in a conventional RTM, and thus produces image higher resolution more balanced amplitude the RTM image. However, least-squares introduces two side effects: sidelobes around reflectors high-wavenumber artifacts. These effects are caused mainly by limited bandwidth of seismic data, coverage receiver arrays inaccuracy modeling kernel. To mitigate these to further boost resolution, we employed sparsity constraints inversion namely Cauchy L1-norm constraints. For solving Cauchy-constrained used preconditioned nonlinear conjugate-gradient method. constrained modified iterative soft thresholding While adopting solution methods, converged faster RTM. Application examples synthetic data laboratory demonstrated that methods can promote resolution.