A robust approach to time-to-depth conversion and interval velocity estimation from time migration in the presence of lateral velocity variations

Conclusions

We have introduced a novel nonlinear inversion formulation for time-to-depth velocity conversion and coordinate mapping, which involves iterative least-squares optimization constrained by a system of partial differential equations. We turn one of the governing equations that relates seismic velocity to geometrical spreading of image rays into a cost function, and linearize the other two equations: the eikonal equation and the orthogonality condition. Regularization provides extra constraints during inversion. The proposed method appears to be robust and the inversion in practice converges fast.