Reflection-based full-waveform inversion (RFWI) can form a low-wavenumber gradient with reflection energy to update the background velocity model. It extends the conventional full-waveform inversion (FWI) beyond the penetration limits of the transmitted energy. However, conventional FWI with reflections often suffers from high-wavenumber update and cycle-skipping issues that are not amenable to building background models. To address these two challenges regarding reflection-based inversion, RFWI consists of two essential components: first, to promote kinematic updates based on reflections, the Born modeling based gradient kernel is adopted to explicitly compute lowwave number components from reflections; second, and to achieve a correct updating direction, a kinematics-oriented objective function is employed to mitigate the cycles kipping issue. We show an application of this method in a complex geological setting in the Campeche deep-water region, Gulf of Mexico. RFWI is incorporated into the earth model building workflow to help depth imaging, especially for steep dips, deep carbonates, and stratigraphy.
Presentation Date: Tuesday, September 17, 2019
Session Start Time: 1:50 PM
Presentation Time: 2:15 PM
Presentation Type: Oral