Throughout the modern history of exploration geophysics, the evaluation of conventional reservoirs (i.e., those driven by porosity and fluid saturation) primarily involved the imaging of stacked seismic data to identify trapping mechanisms such as anticlines and faults from which hydrocarbon accumulation can occur. This concept started to evolve in the early 1990s when prestack time migration (PSTM) became commonplace and the concepts and pitfalls of amplitude vs. offset (AVO) started to become widely known. Shortly thereafter, exploration for Class II AVO anomalies became in vogue after the easy Class III “bright spots” were discovered and drilled, some of which only turned up high porosity sands or fizz gas.

These experiences demonstrated that additional diagnostic information was needed. In quantitative interpretation (QI) the objective is to quantify reservoir conditions, including internal and external geometries, elastic and geomechanical properties, lithology, and fluid content. This can be achieved through the development of reservoir-specific rock physics models and their use in the inversion of seismic gathers. These inversion products are then calibrated using other sources of rock property data such as well logs, core and drilling, and completion engineering data.

A new way of doing business

Beginning in the last decade, these quantitative tools were modified so they could be applied to unconventional resources such as tight oil and gas. These reservoirs have low porosity and permeability and often require enhanced recovery techniques such as hydraulic fracture stimulation to achieve economic recovery levels. Thus, QI techniques designed for porosity-driven plays are not suitable for unconventional plays.

This is conceptually easy to understand: Conventional plays consist of a matrix rock with pores. The matrix can be granular sandstones or porous limestones; it is not important as long as rock properties such as bulk moduli are known. Complexities may occur if clays or calcites are present in siliceous granular rocks. Fluids such as water, gas or oil inhabit the pore space. Models can be generated for any combination of porosity and fluid content. More complex models allow for clay variations, calcite variations or specific geometries such as laminated reservoirs.

In unconventional plays this paradigm does not exist. The rocks are typically mudstones with some degree of lamination where the lamina can consist of shales, silts or limestones. (True pure shale formations are usually not appropriate reservoirs because they are typically high in clay content and have difficulty fracturing and maintaining the fracture after fluid pressure is removed). In this scenario, the rock properties are an average over some vertical interval, which will vary between log resolution and seismic resolution. The properties of interest are porosity, total organic carbon and volume of clay because these three properties typically have the greatest impact on acoustic and density readings from logs or seismic.

Fractures: to be or not to be

Because fine-grained reservoirs often have single-digit porosities and nanodarcy permeabilities, some other means of fluid transport from matrix to wellbore must be found. These pathways are fractures. Creating fractures is, of course, the purpose of hydraulic fracturing, but what about natural fractures? Natural fractures can be excellent connectivity pathways, but under some circumstances they can be too good. If there are water-bearing formations in the vicinity of the organic-rich shale and either natural or induced fractures connect the wellbore with these formations, the well will be watered out. An additional danger is losing pumped fluid pressure via high-conductivity fracture systems.

Realizing the importance of natural fracture networks, seismic acquisition has advanced considerably in an effort to detect and characterize these fractures. A case in point is full-azimuth acquisition (FAZ). In offshore environments with complex structures such as subsalt, FAZ data allow the complete wavefield to be sampled and imaged. However, onshore unconventional shale plays typically exhibit directional anisotropy to some extent. This anisotropy is due to vertical fracture sets in one or more directions and is known as horizontal transverse isotropy (HTI). The principle is that seismic waves will slow down if they cross fractures, and thus the difference between seismic wavefield velocities in orthogonal directions (parallel and perpendicular to a set of fractures) acts as a proxy for the direction and intensity of fracturing. This principle applies to normal compressional (PP) seismic data as well as multicomponent converted (PS) seismic data.

Microseismic: the sounds of breaking rock

The importance of fractures in unconventional plays has led to the development of another related technology—listening for acoustic signals generated by the injection of fluids into a rock mass during well completion. These signals provide evidence of the change in stress state of rocks in the vicinity of the wellbore. With the injection of fluids into a rock mass the pore pressure increases, thus decreasing its cohesive strength and allowing movement to occur along planes of weakness in rocks that are already critically stressed. Pressure waves caused by fluid injection can travel some distance away from the wellbore, thus allowing breakage of critically stressed rock beyond the fluid-injected rock mass. It is for this reason that a direct correlation of microseismic events with fluid injection or of proppant placement remains problematic.

Nonetheless, the amount of information that can be determined from a microseismic event is remarkable. As in normal surface seismic acquisition, if proper geometries exist (meaning detection of an event from multiple locations, thus allowing sampling of the full wavefield) then not only can the magnitude of the event be determined (which is related to the area of slippage) but the sense of motion (strike, dip and rake of movement) and the type of event (opening, closing or strike-slip) also can be quantified. Armed with this treasure trove of data about the well completion, engineering and log data from the wellbore can finally be integrated with geophysical data away from the wellbore.

Bringing it all together

To accomplish the integration of diverse data types, ION developed the ResSCAN concept. These programs incorporate leading-edge multicomponent (PP and PS) survey design, acquisition, processing and analysis with the express purpose of defining the rock and fracture properties of an unconventional prospect. In the processing sequence, vertical transverse isotropy, primarily from layering effects, is removed during migration. This is followed by HTI anisotropy calculation via elliptical velocity inversion of both PP and PS data (using the AZIM and SEAC algorithms, respectively). The results of HTI anisotropy calculation are used to kinematically remove the effects of anisotropy on the azimuthal gathers, thereby flattening them, and also are used quantitatively to assess the magnitude and azimuth of fracturing and/or local and regional stress in the project area.

Rock properties are calculated via prestack joint inversion of PP and PS data, which act to constrain density estimations. These results are calibrated with logs and core, and the fracture data are calibrated with image logs, borehole breakout data and completions pressure data. The rock physics models are used to assess and interpret seismic rock property changes away from the wellbore.

As a final step, these data are fed into a geocellular grid along with overburden stress and the principle horizontal stresses to create a geomechanical static model. Simulations can be run using the completion data, attempting to reproduce the observed microseismic event cloud. Once this is achieved, simulations can then be run at proposed drilling locations to predict completion success with the goal of maximizing EUR while minimizing cost.