Development of a Fracture Model for Spraberry Field, Texas USA.

Abstract The development of a naturally fractured reservoir (NFR) requires not only the assessment of fracture spacing, aperture width, fracture permeability, etc. (i.e., a fracture model), but also requires the input of a representative geological model. Yet development of a fracture model is difficult in most naturally fractured reservoirs due to the inability to directly sample the fractures and the scale effects, or implications, of the modeled fractures. Integration of data is even more critical in naturally fractured reservoirs, compared to conventional matrix dominated reservoirs because high permeability fracture ‘highways’ can dominate reservoir performance. Although the Spraberry field was discovered as early as 1949, and therefore has a long production history, the nature of the fracture system in the Spraberry field is still not completely understood. Over the last seven years, interference, step rate, inter-well tracer, salt tracer, buildup tests, fall-off tests, horizontal core, discrete fracture modeling, outcrop analyses, fracture logs, production tests and profile logging data have been integrated to show the nature of the Spraberry fracture system. This paper discusses the development of a composite fracture model for Spraberry. The Spraberry fracture system has been shown to be heterogeneous and composed of three components. The first component of the fracture network is a system of very long (~1000 ft) fractures in which fluid can travel at high velocities. A second fracture system, pervasive throughout the reservoir, exists where short, relatively discontinuous fractures (~20 ft), act as a background. And finally, the third component is comprised of the low permeability matrix. In order to maximize recovery and understand production profiles, one must understand and effectively integrate all three components. The long, well connected, fracture system controls the initial flow distribution (i.e., dictates where injected fluids go first) in the reservoir. In any short-term tracer or production test (< 1 month) this system will totally dominate the response. However, with continued water injection water begins to invade the secondary (smaller discontinuous fractures) fracture system. Subsequent to the depletion of the high permeability fractures, the average fracture spacing, imbibition rate and matrix permeability become controlling elements to the production profile and recovery factors. Pressure transient tests, such as interference or buildup tests, can be used to determine the average permeability of the discontinuous fractures, the matrix, plus the more continuous fractures. The large amount of data in this case allows us to characterize naturally fractured reservoirs using geological static data (core, fracture log and outcrop) and dynamic engineering dynamic data (tracer analysis, production data, as well as pressure transient data. Both build-up tests and multi-well interference tests are presented in this paper. The information from this reservoir gives us insight to other naturally fractured reservoirs. In summary, this paper presents 1) techniques and data analyses that are relevant for other naturally fractured reservoirs, and 2) a methodology applied to maximize recovery and understand production profiles.