Least-Squares Full Wavefield Migration

Conventional seismic processing uses only primary reflections for imaging, and treats high-order reflected events (multiples) as noise. Separated Wavefield IMaging (SWIM) uses the downgoing pressure wavefield to exploit the extended illumination provided by surface-multiple energy, effectively converting all receivers into virtual sources. We present a Least-Squares inversion solution for depth migration of the full reflected wavefield that includes both primary and high-order reflected energy. Standard migration of primaries and SWIM are complementary and can augment the overall imaging results when they are combined correctly. Whilst full-wavefield migration (FWM) is able to jointly image both primary and high-order reflected energy, it cannot easily balance the contribution of each component, and also involves crosstalk due to the natural blending scheme. The Least-Squares full wavefield migration (LS-FWM) described here directly computes the earth’s reflection image in an iterative manner, thereby avoiding crosstalk noise. Successful applications to both synthetic and field data examples demonstrate that LS-FWM greatly improves the imaging illumination and resolution compared to conventional migration.