One more stride forward in Thunder Horse subsalt imaging with elastic FWI

Authors:
  • BEBunmi Elebiju
    Bunmi Elebiju
  • QLQingsong Li
    Qingsong Li
  • KHKen Hartman
    Ken Hartman
  • FRFrancis Rollins
    Francis Rollins
  • YFYe Feng
    Ye Feng
  • KKKarl Kaiser
    Karl Kaiser
  • WSWerner Schinagl
    Werner Schinagl
  • CCChi Chen
    Chi Chen
  • AHAlex Yanjun Hao
    Alex Yanjun Hao
  • ZWZhiyuan Wei
    Zhiyuan Wei
  • JMJiawei Mei
    Jiawei Mei

Abstract
Subsalt reservoir imaging in the Thunder Horse field in the Gulf of Mexico remains challenging after years of tremendous efforts in both acquisition and imaging, primarily due to the massive salt canopy that causes distorted and poor illumination, leading to velocity and imaging issues at the subsalt area. In 2015, a long-offset, full-azimuth OBN survey was acquired with the aim of improving the subsalt illumination. Most recently, the application of Time-lag Full-waveform Inversion (TLFWI) using this OBN data has resulted in substantial improvements in velocity over the conventional top-down velocity model building flow, despite its acoustic assumption. Furthermore, FWI Images, normal derivatives of the FWI velocities, provide improved illumination, higher S/N, and better reservoir details near the salt stock in Thunder Horse over conventional imaging methods. However, strong elastic effects at the salt-sediment interface cause smearing of the salt boundary and a considerable salt halo in acoustic FWI results, which makes the FWI model and FWI image sub-optimal at Thunder Horse, and results in difficulty mapping and extending the seismic reflection toward the salt flank. In 2021, bp reprocessed the 2020 dense OBN with elastic TLFWI (E- TLFWI), which can model the energy at large impedance contrasts with better amplitudes and phases and thus improve the match between synthetic and recorded data, to better address this imaging challenge. E-TLFWI can greatly sharpen the salt boundaries and reduce the crosstalk noise in the Thunder Horse velocity compared to acoustic TLFWI (A-TLFWI). Evaluation of the fast-track E-TLFWI images shows a step-change improvement in the subsalt images and amplitude of Thunder Horse field. These images show improved event focusing and continuity, higher S/N, and more balanced amplitudes than its acoustic counterpart.