Full-Field Simulation for Development Planning and Reservoir Management at Kuparuk River Field

Abstract

Summary The Kuparuk River field on the Alaskan North Slope produces from twostratigraphically produces from two stratigraphically independent sands of the Kuparuk River formation. A full-field reservoir model was constructed tosupport field management and development planning. The model captures planning. The model captures essential aspects of two independent producing horizons, hydraulically producing horizons, hydraulically coupled at the wellbores, andsimulates dynamic interactions between the reservoir sands and surfacefacilities. The field model is used to plan field development on the basis ofperformance ranking of drillsite performance ranking of drillsite expansions, to assess depletion performance effects of reservoir performance effects ofreservoir management strategies, and to evaluate alternative depletionprocesses and associated reservoir and facility interactions of fieldprojects. Introduction A variety of development options are under evaluation for potentialimplementation at the Kuparuk River field. Several competing depletionmechanisms currently exist in the field. Introducing new projects will add tothe complex interrelationship of these depletion processes. Full-fieldreservoir simulation is used to forecast production performance under existingand future field performance under existing and future field configurations, toevaluate reservoir and facility interactions associated with field developmentalternatives, and to support ongoing field operation and management. This paperdiscusses the application of full-field paper discusses the application offull-field simulation to development planning and reservoir management at the Kuparuk River field. Motivation for Field-Scale Simulation. Using large-scale simulation forreservoir performance appraisal and development performance appraisal anddevelopment decision making has increased steadily* since the early developmentof black-oil reservoir simulators. Full-field reservoir simulation allowsreservoir depletion processes to he fully integrated with surface-facilityoperating constraints, resulting in resolution of dynamic interactions betweenthe reservoir and facilities. The need to quantify reservoir performanceexpectations more accurately spurs the development of more rigorous and complexfull-field reservoir models. Increased understanding and certainty derived fromadvanced field-scale simulation contributes to the decisionmaking process andpartially mitigates the risk associated with the large capital expendituresrequired for reservoir development. Kuparuk Reservoir Characteristics. The Kuparuk River field is located about40 miles [64 km] west of the Prudhoe Bay Unit (Fig. 1). Discovered in 1969, the Kuparuk reservoir contained an estimated 5 billion bbl of original oil in place(OOIP) with estimated recoverable reserves of nominally 1.6 billion bbl. The reservoir is made up of two distinct sandstone members within the Kuparuk River formation, a Lower Cretaceous, shallow, marine-shelf sanddeposit. The members are separated by a major unconformity, and two units arerecognized within each member. As Fig. 2 shows, the lower member contains Units A and B (informally named), with reservoir-quality sands present primarily in Unit A. The upper present primarily in Unit A. The upper member contains Units C and D, with reservoirquality sands present only in Unit C. Fig. 3 illustratesthe lateral extent of the Kuparuk sands. The A sand extends over the entirefield; the C sand covers a smaller area. Fig. 4 shows the dominant structural aspects of the Kuparuk sands. The Kuparuk interval forms a gently dipping anticline ranging in depth from 6,500to 8,500 ft across the structure. The trapping mechanism is a combination ofstratigraphic pinchout, erosional truncation, structural closure, pinchout, erosional truncation, structural closure, and a water/oil contact. Thereservoir is highly faulted by normal faults with up to several hundred feet ofthrow, which has a significant impact on flow unit continuity and fluidmovement. The predominant north/south fault trend density is nominally 3faults/sq mile. Faulting occurred during different depositional periods of the Kuparuk sands, influencing the character of the sand accumulations. Faultingoccurred primarily postdepositional to the A sand and postdepositional to the Asand and syndepositional with the C sand. Consequently, faulting frequency andpatterns are dissimilar for the two sand bodies. The Kuparuk sands are fairly thin, averaging 50 to 100 ft of grossthickness, but laterally extensive, covering roughly 200 sq miles. The C sandcontains several highly permeable, very prolific, interbedded zones permeable, very prolific, interbedded zones having an average permeability thickness of5,000 md-ft. Reservoir quality of the C sand, expressed by the distribution ofreservoir properties, is highly influenced by diagenesis and indirectly relatedto depositional facies. The A sand is characterized by sheet-like sandstonebodies interbedded with shales and mudstones. The A sand reservoir quality iscontrolled by depositional environment, and this sand, with its averagepermeability thickness of 1,000 md-ft, is permeability thickness of 1,000md-ft, is much less productive than the C sand. JPT P. 974

Predicting Horizontal-Openhole-Completion Success on the North Slope of Alaska

“…Production and injection surveys confirm the presence of thin, high-permeability beds supporting rapid gas and water breakthrough from injector to producer. Too often, these layers are characterized by both high permeability and low pore volume, supporting high-velocity fluid migration (Starley et al 1991). The net result is that fluid migration in the C sand is uncontrollable across flood patterns and between wells.…”

Predicting Openhole Horizontal Completion Success on the North Slope of Alaska

“…Too often these layers are both high permeability and low pore volume supporting high velocity fluid migration. 10 This picture is complicated by numerous faults cutting the C Sand, with throws in excess of 100 feet. Faults through the area vary widely from sealing to highly conductive.…”

An Object-Oriented Approach to Simulator Postprocessing

“…The KFFM is a field-wide reservoir model of the Kuparuk River field containing nearly 700 wells during historic production, each completed in three model layers representing two producing horizons. 9 In predictive scenarios, the model may contain 1,200 wells and forecasts run to 30 years. The compute-server was an HP/Apollo 433 configured with 32 MB RAM and two 120-MB hard drives.…”