Seismic Imaging of Fault Zones

Faults are narrow zones in the Earth?s crust along which one side has moved relative to the other in a direction parallel to the zone. Rocks inside fault zones are highly deformed, such that their physical properties are different from those of the rocks outside the fault zone. Fault zones are important in sedimentary rocks that hold fluids (i.e. reservoirs), since the structure and properties of the fault zone in three dimensions can control how fluids move in the reservoir. This is critical in areas such as hydrocarbon exploration and production, sequestration of CO2 in subsurface reservoirs, hydrogeology, and seismic hazards. Unfortunately, we know very little about the structure and properties of fault zones: Field outcrops give a two dimensional often incomplete picture, fault zones are so narrow that they are barely imaged by seismic reflection (i.e. echo sounding), and well logs or core across fault zones are almost inexistent (in fact driller engineers are reluctant to perforate wells across fault zones). The main purpose of this project is to evaluate the potential of seismic reflection to image the structure and properties of fault zones. We do this through a synthetic workflow: We model fault zones using a mechanical method called the discrete element method (DEM). The DEM simulates the sedimentary rock as an assemblage of spherical particles. Faulting and folding arises naturally in the DEM. From a base scenario with known rock physical properties (including the acoustic properties necessary for seismic reflection), we modify the rock properties based on the deformation (more specifically the strain) in the DEM. We then use the updated acoustic properties to generate a seismic image using ray tracing methods. This seismic image is like a seismic reflection profile in 2D or a seismic cube in 3D. We can generate different images for different parameters of seismic acquisition, for example different frequencies or different survey geometry, and evaluate which parameters are better to predict the structure and properties of the fault zone. As a proof of concept, we have implemented our workflow in two-dimensions. Figure 1 shows a two dimensional DEM simulation of a normal fault (Figure 1a), the input properties for seismic imaging (Figure 1b), and two seismic images obtained with different illumination directions (Figure 1c). We have looked at the effect on the seismic image of parameters such as wave frequency and illumination direction. We have presented these results in two conferences (EAGE 2012 Fault Seal Conference and 2012 Lofoten seminar) and a scientific paper (Botter et al., 2014, Marine and Petroleum Geology 57, 187-207). We have also implemented DEM faulting simulations in three-dimensions (Figure 2), and we are currently working on the seismic imaging of these three-dimensional structures using seismic attributes based interpretation in collaboration with the company ffA (Geoteric) in the UK. Finally, all these synthetic trials will be meaningless without a real, field proof of the validity of the method. The Delicate Arch relay ramp structure in Utah, USA, is an outcrop that Atle Rotevatn (University of Bergen) has mapped in detail. We are currently working with a detailed reservoir model of this area to produce seismic images of the structure during simulated water injection and hydrocarbon production. As with the synthetic DEM model, in this real field case, we can also evaluate the effect of parameters such as wave frequency, overburden, and survey geometry on the resulting seismic images.

Faults are not discrete planes, but rather zones of deformed rock with a complex 3D geometry and internal structure. Fault zones internal structure and related permeability distribution are primary controls on reservoir connectivity over geologic and prod uction time scales. This has major implications for hydrocarbons exploration and production, as well as for geologic CO2 sequestration. The main goal of this project is to study the seismic response of fault zones through (1) Geomechanical discrete elem ent modeling (DEM) of fault zone evolution, and (2) Seismic modeling/imaging of the DEM fault zone analogues. DEM will be performed using dense, millions of particles, simulations which will allow to define the details of fault zone evolution under differ ent tectonic/stress regimes, fault boundary conditions, rock lithologies, and sedimentation/erosion. The DEM fault zone analogues will then be seismically forward modelled using a prestack depth migration simulator which, contrary to the standard 1D convo lution, acknowledges the effects of lateral continuity and illumination (survey geometry). The results of these simulations will allow us to develop principles for the seismic characterization of fault zones. Overall, this work can shed light into the se ismic response of fault zones and the acquisition and processing parameters needed to characterize these structures with seismic. This can lead to a more realistic representation of faults in reservoir models and better forecasting of reservoir behavior a t geologic and production time scales.