The next generation of rock physics models

As more complex reservoirs are fractured and deeply buried, targeted in subsalt prospects, or found in unconventional resources such as shale, improved seismic imaging and enhanced reservoir delineation and characterization are needed. In recent years, seismic acquisition and processing have progressed as a result of advances in point-receiver measurements and complex acquisition geometries, including acquisition as a function of direction, as in wide-azimuth (WAZ) surveys. But for quantitative characterization, advanced rock physics models are required to understand reservoir complexities.

Rock physics addresses the relationship between measurements of elastic parameters made from the surface, well, lab equipment, and intrinsic rock properties such as mineralogy, porosity and pore shapes, pore fluids, pore pressures, permeability, viscosity, stresses, and overall architectural features like laminations and fractures. Rock physics provides the understanding necessary to optimize imaging and characterization solutions based on elastic data. The model-building functionality of rock physics software allows generation of rock physics models using predefined theoretical or empirical models from an extensive library or by implementing new user-defined models. Now, as the industry moves away from isotropic assumptions to heterogeneous and anisotropic solutions, rock physicists are challenged to model these complexities with ever greater accuracy.

Velocity variation with direction
Pioneering work by industry experts has shown aligned fractures or minerals give rise to an elastic anisotropy whereby wave velocities vary with their propagation direction. However, until the past decade, the seismic industry was not adequately equipped to collect and process data containing information on a large scale. This has changed over the past five years, and large-scale 3-D WAZ surveys and offset seismic data are more routinely collected, enabling more advanced and complete illumination than traditional surveys.

Attenuation of P-waves in gas-filled porous rocks is shown. (Images courtesy of WesternGeco)

However, rock physics still is in its infancy compared to seismic data acquisition techniques. Advancements are needed to use these datasets and convert them into quantitative assessments of subsurface properties. Although amplitude variation with offset and azimuth technology can be useful for determining main fracture orientations, analysis of data for quantitative assessment of fracture properties (such as orientation) requires integration of measurements. No single measurement can provide the complete picture.

Solid-fluid interactions
A number of complex interactions between fluids and solid material in a saturated rock must be considered during rock physics analyses. Interactions can be static, dynamic, mechanical, or chemical in nature.

Static fluid-solid interaction incorporates properties such as density, compressibility, and viscosity that are measured easily and can be influenced by external effects such as pressure and temperature. These effects have been measured in laboratory studies for a variety of hydrocarbons under controlled conditions. Tests resulted in the development of useful empirical relationships commonly used for rock physics modeling using either the Gassmann or Biot models.

Attenuation and dispersion of P-waves in porous rocks (global and local flow) is shown.

Conversely, dynamic fluid-solid interaction addresses change in the properties of composite media during propagation of seismic waves; Biot postulated the elastic and viscoelastic properties of these systems using fundamentals of physics. For example, when a compressional seismic wave (P-wave) propagates through media that are a coupled system, fluid flow lags behind deformation of the solid frame or skeleton of the rock. This drag causes a loss of seismic energy through generation of a second compressional wave that propagates and dissipates similar to heat.

The traditional Gassmann model commonly used by rock physics communities ignores Biot effects for simplification. However, Biot effects are becoming increasingly evident as industry experts are exposed to more heterogeneous fluids in reservoir rock (gas, oil, and brine) and exploration in deepwater clastics, beneath gas chimneys, and in geopressured reservoirs.

Fluid-solid interaction also can be a mechanical bonding. For example, in some unconsolidated heavy oil-bearing reservoirs such as those in Canada, fluid may be part of a stress-bearing rock skeleton (matrix). Changes in fluid properties as a result of variation in pressure or temperature can easily alter rock frame properties.

Fluid-solid interaction can occur as a chemical alteration of the matrix due to injection of chemically active agents such as CO2 as part of an enhanced oil recovery program. Furthermore, subtle chemical alteration of reservoir fluid can alter clay minerals. This results in general swelling or swelling associated with fracturing. Both phenomena lead to a reduction in apparent velocities as seen by various sonic tools.

Rock physics models must account for these effects, not only for better and more quantitative analysis of reservoir rocks using seismic data, but also for monitoring reservoir production through time-lapse seismic methods.

Rock physics model results for shale in a deepwater environment are shown.

Rock physics for shales
In clastics basins, shales constitute more than 80% of the rock. Yet, understanding of this type of rock often is limited. Shales not only provide a seal to hold fluids within a reservoir (sandstone), but also contain source material and high-pressured brines that frequently pose drilling challenges. From a seismic perspective, when a wavelet generated by a seismic device passes through shale in the overburden prior to reaching a reservoir target, its amplitude and phase vary significantly due to interaction between the seismic wave and the coupled solid-fluid system of the shale.

Traditional seismic processing techniques that use constant “Q” or intrinsic attenuation compensation techniques to balance amplitude losses and associated phase retardation can be inaccurate. In addition, as the industry ventures to deeper targets for hydrocarbon exploration, it is important to take account of various diagenetic changes that happen in shale as a result of compaction and various chemical changes. These changes can alter the fabric of the shale and its mechanical properties.

Rock physics for geomechanics
Advances in seismic acquisition methods have enabled reliable observation of geomechanical effects in seismic data. For example, more accurate positioning of seismic sources and receivers has improved repeatability. This, in turn, has improved signal-to-noise ratio for time-lapse seismic surveys such that signals caused by production-induced stress changes and their impact on seismic velocities now are routinely observed and interpreted.

Accurate measurements of velocity fields observed by seismic methods have potential to become a remote sensing tool for subsurface stress fields. The key to extracting stress-field information from seismic data is a stress-sensitive rock-physics model. Physical processes must include stiffening of the rock-matrix and the associated increase in seismic velocity as stress is increased, as well as the softening of the rock matrix and associated velocity decreases as the rock starts to fail during further stress increases. Dependence of seismic propagation velocity on mean- and shear-stress changes also is important to the rock physics model. Increased activity in laboratory measurements of seismic velocity as a function of stress path provides fundamental understanding of stress sensitivity of seismic velocity, taking into account other key parameters such as lithology and porosity. This is accompanied by a transfer of knowledge from civil engineering and engineering geology – where such experiments traditionally have been carried out – to the oilfield environment.

Next generation of rock physics models
Rock physics must go through significant advancement to be able to sufficiently address pore-acoustic and elastic properties of rocks as a function of azimuth. Models must be expanded to explain transition from elastic to elasto-plastic and plastic reservoirs throughout the reservoir life cycle. It is evident that reservoir rocks can change significantly during production. Fluid-solid as well as chemical interaction, reservoir compaction, and their behavior under different stress regimes add to the complexities. This information needs to be captured and incorporated into a new generation of rock physics models.

Immediate challenges faced by the rock physics community are the modeling of heavy oil and shale gas reservoirs and CO2 sequestration. This can be achieved by building upon existing models. The future is bright as rock physics becomes increasingly integrated into standard geophysical services.