The effects of azimuthal anisotropy on 3D and 4D seismic amplitude variation with offset responses

Abstract

Seismic reflection amplitude variation with source-receiver offset (AVO) is an important tool in hydrocarbon exploration and reservoir monitoring, due to its sensitivity to elastic rock properties that are affected by changes in pore-fluid saturation and pressure. In most cases, 4D seismic feasibility studies and interpretation analyses assume that the earth is isotropic. This assumption can be problematic because it is becoming increasingly apparent that anisotropic rocks are quite common. Furthermore, the presence of even small amounts of anisotropy can have significant effects on AVO, and in the presence of azimuthal anisotropy the AVO will vary with azimuth. We determine that if 4D seismic surveys are acquired with different survey azimuths in the presence of azimuthal anisotropy, it is likely that 4D AVO interpretations will be significantly affected, leading to incorrect or nonphysical interpretations. This possibility is especially apparent in the context of the North West Shelf, Australia, where significant stress-induced azimuthal anisotropy is prevalent in sandstone formations that form the reservoir rocks. We model 4D AVO responses with and without azimuthal anisotropy effects for a variety of pore-fluid saturation and pressure change scenarios using average reservoir properties from the Stybarrow field, Australia. We found that azimuthal anisotropy does not affect the small reflection angles of the 4D AVO response, but it has a significant effect on larger reflection angles when comparing 4D surveys acquired at different acquisition azimuths. This azimuthal behavior leads to what we call an “apparent 4D effect” when reservoir properties do not change and a “contaminated 4D effect” when reservoir properties do change. We found real data examples in which we determine that the 4D AVO response must incorporate azimuthal anisotropy to be explained correctly. Our results further emphasize the importance of repeating survey acquisition azimuths whenever possible and/or accurately accounting for azimuthal anisotropy effects.