Orthorhombic anisotropy is considered to be the simplest realistic symmetry for many geophysical problems. Orthorhombic anisotropy exhibits both HTI (Horizontal Transverse Anisotropy) and VTI (Vertical Transverse Anisotropy) effects, making seismic velocity varying with azimuthal direction as well as polar direction. In many cases, these effects are important by themselves and can be a target of special studies.
The presence of orthorhombic anisotropy poses challenges in wide azimuth imaging which has been rapidly developed for better illumination, better imaging, and better multiple elimination. Analysis of multi-azimuth (MAZ) data often reveals noticeable fluctuations in moveout between different acquisition directions, preventing constructive summation of MAZ images due to the azimuthal dependency nature of wave propagation in orthorhombic medium. On the other hand, the co-existing VTI effects of orthorhombic anisotropy can also cause well misties and higher order moveout.
Orthorhombic anisotropy can take into account the co-existing HTI/VTI effects. We have developed an approach for imaging in the presence of orthorhombic anisotropy, including orthorhombic velocity analysis and orthorhombic migration. Following Tsvankin's work, we put forward a ray tracing approach suitable for both weak and strong anisotropy which applies to both Kirchhoff and Beam PSDM. Aimed at the challenge in orthorhombic anisotropy model building, we have developed a practical workflow which combines the co-existing HTI/VTI anisotropy estimated through multi-azimuth tomography to form the orthorhombic model.
In this paper, we first describe our approach for imaging in the presence of orthorhombic anisotropy, including the newly developed orthorhombic ray tracing method and the newly developed practical method for orthorhombic anisotropy model building. We then demonstrate with both synthetic and real data from offshore Australia that our approach can successfully take into account the co-existing HTI/VTI effects, reduce the structural discrepancies between seismic images built for different azimuths, hence produce constructive summation of MAZ dataset, resolve well mistie to match with geology, and deliver a step-change in the final seismic image quality.