Noise suppression in elastic reverse time migration based on optimized pseudo-stress decoupling algorithm

Elastic reverse time migration enhances subsurface imaging for oil and gas exploration by precisely delineating subsurface structures through elastic wavefield extrapolation, but challenges remain in P and S stress wavefield separation, particularly close to the source injection point where noise of energy leakage is significant. In this paper, we introduce an optimization algorithm for pseudo-stress decoupling equations to mitigate energy leakage in elastic reverse time migration. Firstly, we optimize source loading by using an exponential decay function, reducing anomalies near the source in numerical simulations. Secondly, we refine the construction process of P and S pseudo-stress wavefield decoupling equations, maintaining the amplitude, phase, and unit information of the wavefield for more accurate stress imaging. We conduct numerical experiments using H-V model and a salt model to validate the efficacy of our methods. Numerical experiments result demonstrate that our proposed methods effectively alleviate energy leakage in the decoupled stress wavefields, significantly enhance imaging precision, achieve accurate and convergent imaging of complex subsurface structures without significant noise, and improve the robustness and reliability of stress tensor based elastic reverse time migration for subsurface geological exploration in oil and gas reservoirs.