An Eikonal‐Based Angle‐Domain Least‐Squares Kirchhoff Depth Migration in General Anisotropic Media

ABSTRACT Least‐squares migration (LSM) is an inversion‐based imaging method that seeks a reflectivity model whose predicted data best fit the recorded seismic data in a least‐squares sense. However, conventional LSM can be unreliable for anisotropic reflection data if reflection angles are neglected. First, ignoring anisotropic parameters often produces defocused images and biased kinematics. Second, conventional LSM is generally more sensitive to migration‐velocity errors than its extended‐angle counterpart. To address these limitations, we propose an angle‐domain least‐squares Kirchhoff depth migration (ADLSKDM) framework that explicitly accounts for general anisotropic effects via angle‐dependent linear inversion. Our workflow combines (i) a semi‐analytical eikonal solver that efficiently computes traveltimes in complex models under weak anisotropy, and (ii) a regularized least‐squares inversion that suppresses artefacts in angle‐domain common‐image gathers (ADCIGs). Numerical tests on anisotropic synthetic datasets demonstrate that ADLSKDM retrieves correct angle‐dependent kinematics with reduced ambiguity and fewer artefacts than conventional LSM.