From Geomechanical Modeling to Seismic Imaging of 3-D Faults
Although typically interpreted as 2D surfaces, faults are 3D narrow zones of highly and heterogeneously deformed rocks, with petrophysical properties differing from the host rock. 3D fault structure and properties are primary controls on fluid flow in faulted reservoirs. Even though seismic data are one of the main ways of subsurface investigation, fault zones are often at the limit of seismic resolution and not fully explored. We propose a synthetic workflow to assess the potential of seismic data for imaging fault structure and properties. The workflow is based on a discrete element method (DEM) to simulate fault formation, simple relationships to modify the initial elastic properties (e.g. density, P- and shear wave velocities) based on the volumetric strain calculated from the DEM, and a ray-based modeling (pre-stack depth migration or PSDM simulator). The PSDM simulator handles 3D effects in resolution and illumination as function of various parameters such as velocity model, survey geometry, wavelet, etc. We illustrate the application of the workflow to a 3D large displacement normal fault in an interlayered sandstone-shale sequence for two models, one with constant fault slip and the second with linearly variable fault slip along the strike. Although the DEM does not target processes at the grain scale, but rather meter size bulk strain, it produces realistic fault geometries and strain fields. Seismic cubes of these models are generated for an homogeneous overburden and several wave frequencies. High frequencies show the large impact of the fault on the reflectors, which are offset but also laterally distorted. In the variable fault slip model, the fault has a larger impact on the seismic as the displacement increases, and the fault tip can be interpreted in map view. As wave frequency decreases, the fault is displayed as a simpler structure. We do a more quantitative analysis of the seismic by extracting the fault damage zone geobody from seismic attributes. This allows a direct comparison between the fault zone identified on the seismic and the fault volume in the initial geomechanical model, giving guidelines on how to better deal with the seismic for fault interpretation. Our modeling provides ways to fully understand how faulting impact seismic, and therefore to tune acquisition and processing parameters for fault characterization.
AAPG Datapages/Search and Discovery Article #90194 © 2014 International Conference & Exhibition, Istanbul, Turkey, September 14-17, 2014