An application of vector wavefield decomposition to 3D elastic reverse time migration and field data test

Abstract Wavefield separation is a crucial step in suppressing crosstalk artifacts and improving imaging quality in elastic reverse time migration (ERTM). In isotropic elastic media, the most popular wavefield separation technique is the Helmholtz decomposition. However, the Helmholtz decomposition produces pure-mode wavefields with incorrect amplitudes, phases, and physical units. In addition, when we implement multi-shot ERTM using pure-mode wavefields, polarity changes in the converted wave can destroy reflection events because of incoherent stack. To remedy these problems, we theoretically analyze the characteristics of amplitude and phase distortion of wavefields in the decomposition of P- and S-waves based on the Helmholtz decomposition in time-space domain, and formulate accurate wavefield decomposition and recomposition equations. In this paper, we produce vector P- and S-wavefields through the following two steps. First, we correct the amplitudes and phases of the pure-mode wavefield during the Helmholtz decomposition. Second, we obtain vector P- and S-wavefields using a wavefield recomposition method. Using the separated vector wavefields, we adopt a scalar imaging condition for ERTM. These operations enable us to produce migrated images directly with the correct amplitudes, phase, and physical units. Through several 3D numerical examples and a 2D real data example, we demonstrate that the proposed ERTM provides high-quality migrated images. The method is also feasible and sufficiently robust for imaging complex subsurface structures.