Subsalt imaging by target-oriented wavefield least-squares migration: A 3-D field-data example

Conclusions

Solving wavefield least-squares migration in the image domain makes possible target-oriented application of this method, allowing reflectivity inversion at the reservoir level. The 3-D examples demonstrate that simultaneous phase-encoding in the conical-wave domain drastically reduces the computational cost of the 3-D Hessian matrix. The phase-encoded Hessian, albeit with some approximations, accurately quantifies the illumination effects on the migrated image. Since inverting the Hessian is very fast, different regularization parameters or schemes can be tried at very low cost. For the 3-D example shown in this chapter, it takes only about $6$ minutes to run $100$ iterations using $34$ CPUs ($17$ nodes with $2$ cores on each). This is a very important advantage over the conventional data-domain implementation, which requires full-domain modeling and migration at each iteration. The high efficiency of this method also makes interactive reflectivity imaging possible, where we can repeat the inversion with regularizations that incorporate different geological scenarios and obtain the results in almost real time. The 3-D reflectivity inversion results illustrate that inversion preconditioned with dip filters successfully recovers the reflectivity from the effects of uneven illumination, yielding more balanced amplitudes and higher spatial resolution in the inverted image.

Subsalt imaging by target-oriented wavefield least-squares migration: A 3-D field-data example