Abstract

Abstract Accurate fault interpretation is key to building a robust and reliable reservoir model, and is essential for further study of fault-seal behaviour and reservoir compartmentalization. Conventional fault interpretation defines faults as planar features. Although the concept of fault damage zones (FDZ) is well established, manual delineation of FDZ from subsurface imaging data is difficult and rarely attempted. In this work we are using the concept of a seismic fault distortion zone (SFDZ). The SFDZ differs from the geological FDZ and is the 3D extent of the regions where the presence of a fault has perturbed the seismic image. The SFDZ represents an area of significant uncertainty within the seismic volume where the signal is distorted either by changes in geology associated with faulting or by imaging problems due to the presence of the fault. Delineation of the SFDZ is a key step in further analysing 3D seismic data to provide information regarding variations in seismic attributes across a fault that can be used to indicate how juxtaposition of strata influences a potential sealing fault. The SFDZ is delineated using an adaptive region growing algorithm and produces a set of detailed 3D SFDZ geobodies. The delineated SFDZ is then used as the input to a set of analysis techniques that examine 3D volumetric juxtaposition of seismic attributes, and properties relative to the fault. The techniques involve automated statistical measurement along seismic structure relative to the fault itself. The workflow has been applied and its utility assessed on an underground gas storage dataset. Initial results suggest that by taking account of the 3D geometry of the SFDZ and subsequently analysing 3D frequency and acoustic impedance volumes relative to these structures, an improved understanding of the juxtaposition of layers across the fault and the lateral variability, juxtaposition can be determined at seismic resolution, and subsequently used to assess the potential for fault seal and fluid migration.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call