Analysis of P-wave seismic response for fracture detection: modelling and case studies

Abstract

This thesis addresses a few specific issues in the use of wide azimuth P-wave seismic data for fracture detection based on numerical modelling and real data. These issues include the seismic response of discrete fractures, the effects of anticline and uncertainties in real data analysis. For this, I implemented the finite difference scheme for modelling the seismic response in 3D fractured media; appropriate approaches are then selected to study discrete fracture models and the effect of the anticline with 3D seismic modelling, followed by an integrate real case study. Finite difference (FD) is widely used in seismic modelling. There are three FD schemes described in this thesis, the standard staggered grid (SSG), the rotated staggered grid (RSG), and the diamond staggered grid (DSG). Both qualitative and quantitative comparison has been made to reveal their capability in modelling 3D fractured media. The SSG has shown best performance for anisotropic media with orthorhombic symmetry or higher symmetry system. For lower anisotropy symmetry, the DSG is preferred than the RSG in terms of computation efficiency. A new solution to the diamond grid issue is developed which can simplify the DSG implementation, and an optimized workflow is proposed to simulate large 3D fractured models. The SSG scheme is implemented in three dimensions and it provides a useful tool for various practical modelling studies. With the above tool, two modelling studies have been carried out, on the effects of the discrete fractures and of the presence of anticline: the Discrete Fracture Model (DFM) study provides many insights into seismic response of discrete fracture and the link between the discrete fractures and aligned micro cracks, as well as the features in scattering waves. The modelling results demonstrate that, P-wave seismic anisotropy increases with the decrease of discrete fracture spacing, and different spacing leads to different patterns in scattering waves. The study also reveals the azimuthal AVO variation on the top of discrete fracture layer, which is similar to that we find in homogenous anisotropic media. The study of the anticline structure with vertical fractures, which is built with the parameters from a real case, is to assess the anticline structure effect on fracture parameter inversion based on the Singular Value Decomposition (SVD) method. The fracture density can be resolved accurately at the top of the anticline, whilst that on the flanks tends to be over-estimated. The results also indicate that the SVD method is a reliable approach for directly estimating the fracture density. P-wave azimuthal attributes are commonly employed to invert fracture density and orientation. Many factors may affect the accuracy of the inversion results. The integrated study in this thesis shows that azimuthal coverage, offset-depth ratio, data quality and geological structures all affect the final prediction, and different attributes shows different sensitivities to these factors. Furthermore, the combined analysis of both geological observation and pre- and post-stack seismic attributes can reduce the uncertainties for fracture detection.