Residual moveout estimation through least squares inversion

摘要

Method for estimating residual moveout in common image gathers (51) of seismic data for use in velocity tomography (57). The method iteratively (56) flattens (55) the common image gathers against a specified reference trace through the application of conjugate-gradient least-squares inversion (53). Different from other picking methods which need to identify and track strong amplitudes, the inventive method automatically inverts for the depth residual for every grid point in the image gather. There is no need to identify and track events.

主权项

1. An automated method for estimating residual moveout in a common-image gather of seismic data for a subsurface region, comprising:
obtaining and optimizing depth residuals through computer-implemented iterative inversion of the common-image gather, using flatness of the common image gather as a criterion for the optimization, wherein the inversion is a conjugate-gradient least-squares process, and wherein the conjugate-gradient least-squares process includes,
(a) calculating horizontal offset and depth gradients of the common image gather, at all points on a selected computational grid, and(b) applying least-squares inversion to the horizontal offset and depth gradients to compute and optimize the depth residuals; using the optimized depth residuals in a tomography method to construct a model of seismic wave velocity in the subsurface region; using the model of seismic wave velocity to migrate the seismic data and thereby produce an image of estimated residual moveout, showing structural features of the subsurface region; using the image showing structural features of the subsurface region in designing a well into the subsurface region; and drilling the well and producing hydrocarbons from the subsurface region.