This paper addresses the problems identified in current shale reservoir characterization practices. We also provide alternative approaches with relevant reflections on the determination of volumes in-place. Rock properties in unconventional reservoirs such as shales is of paramount importance. By comparison with conventional reservoirs, fluids are present not only in the intergranular porous media but also within the fine texture of the rock matrix (Clays, Kerogen and
Micro-Fractures) which usually are only recoverable with the aid of suitable stimulation and completion technologies. This paper questions current engineering practices related with the assumption of unrealistic cut-offs in the petrophysical
analyses which in turn may result in dangerously misleading estimates of in place volumes and thus inadequate development decisions being made.
The adsorption capacity of clays has been documented with observations on the correlations between the percentages of clay minerals in the rock and Langmuir volume (VL) determined in laboratory measurements of gas content from core
samples by means of Langmuir isotherms. Therefore it should be no surprise that clays in shale gas reservoirs are known to adsorb hydrocarbon gases and may contribute to the production when properly stimulated. We therefore recommend
that corrections for clay effects should not be arbitrarily applied in the petrophysical analysis of electric logs. The use of a total porosity-total water saturation model will help to avoid shortcomings in total gas in-place determination. Additional
reasons for the avoidance of clay porosity corrections; include the fact that there are no tools capable of differentiating between free gas and adsorbed gas.
Total porosity and water saturation methods give rise to total gas content determination with the appropiate model. Adsorbed gas content estimate, may be obtained by correlating geochemical data based on gas content from laboratory experiments and rock density measured on core and or logs.
A new method is described that efficiently and robustly simulates multiphase flows in reservoirs with modern complex well completions, such as horizontal wells with transverse hydraulic fractures. These types of completions present a difficult challenge to simulators due to the enormous range of scales in geometry and permeability. For 30+ years, simulators have been restricted to very small ratios of smallest grid block to largest and have been unable to model completions faithfully. Simplified well models with added skin, such as the Peaceman model, have been approximations to real-world completions. The new method, based on a finite element method, manages a prodigious range of scales and correctly implements the physics of multiphase fluid flow.
This paper illustrates the new method in a practical setting: the current exploitation plan of the Sierras Blancas reservoir in Argentina, which requires the solution of reservoir flows coupled to hydraulically fractured vertical and horizontal wells. The models describing hydraulic fractures and reservoir are discretized with a finite element mesh that conforms to both reservoir and fracture geometries and smoothly grades down from reservoir scale to the interior of the hydraulic fracture. Flow inside the fracture is computed along with reservoir flow and is fully coupled.
The Sierras Blancas is a gas reservoir; in the course of exploitation, the hydrocarbon in place behaves as a retrograde condensate. The multiphase aspect of this fluid is modeled with an extended black-oil model wherein the PVT properties of the fluid are derived from lab measurements. As condensate drops out of the gas at the dewpoint, the character of the equations changes from single or two phase (gas, water) to two or three phase, and properties, such as Bg, have slope discontinuities that have historically presented formidable challenges to the numerical modeling of fluid flow. The new method correctly deals with both of these numerical difficulties and the attendant scale challenges.
The behaviour of existing wells in Sierras Blancas, along with hypothetical future, more-complex well designs, is modeled faithfully. Various scenarios are being run with these models being part of a continuous-improvement programme of development of the reservoir to optimize exploitation economics.