Amplitude variation with incidence and azimuth stepwise inversion with coherence-attribute-constraints for anisotropic parameters

With the development of 5D seismic technology, the stable acquisition of anisotropy information from wide-azimuth seismic data has become a key scientific problem in the seismic inversion of fractured reservoirs. However, existing AVAZ inversion methods suffer from too many parameters to be estimated, large variation in contribution, and unstable inversions. Therefore, we develop an AVAZ inversion method with coherent attribute constraints to solve the problem of unstable inversion of anisotropic parameters. First, we use seismic coherent attributes to build a fracture equivalent probability model containing anisotropic information of the subsurface medium, which is used to simulate large-scale subsurface fractures and faults, thereby improving the reliability and stability of the anisotropic parameter inversion. We carry out the AVAZ inversion method in a Bayesian framework using wide-azimuth seismic data and introduce smoothing background model constraints to reduce the dependence of the inversion on the initial model and improve the stability of the inversion. Moreover, we add Gaussian and fracture probability distribution models to the objective function to improve the reasonableness and stability of the inversion. Then, through the analysis of the contribution of the estimated parameters, we found that the contribution of the isotropic parameters to the reflection coefficient is much larger than that of the anisotropic parameters to the reflection coefficient. Therefore, we developed a stepwise optimization-seeking inversion method for the isotropic and anisotropic parameters, which can reduce the number of the estimated parameters and thus improve the stability of the inversion of the anisotropic parameters. Field data show that this method produces suitable inversion results even at moderate levels of noise. Therefore, we can conclude that the proposed method has good applicability and stability in predicting the anisotropy parameters of fractured shale reservoirs.