Multiscale Reservoir Description Models for Performance Prediction in the Kuparuk River Field, North Slope of Alaska

Mark H. Scheihing, Ron D. Thompson, Dirk Seifert

2002 · 11 citations · 1 references

Concepts

Abstract

Abstract Reservoir engineering studies of the Kuparuk River Field require reservoir description models at multiple scales. Kuparuk reservoir types range from relatively low permeability, vertically stratified and imbricated lower shoreface sandstones (A Sands) to "thiefy", glauconitic and sideritic shallow marine sandstones with complex permeability structure (C Sands). The field is also heavily faulted, having over 5200 mapped faults with throws typically between ten and several hundred feet. Geocellular models built for type curve modeling of waterflood and enhanced oil recovery (EOR) processes capture a range of A and C Sand reservoir types. These models are 2600x2600 feet with cell dimensions of 50x50x0.5 feet. Cored wells provide a robust facies and petrophysical model. Both object- and pixel-based techniques are used together with a flow unit-scale zonation to capture key heterogeneities. In the A Sands, this requires capturing the interstratification of sandy and shaly facies using object-based methods. In the C Sands, this requires capturing permeability heterogeneities caused by facies changes and siderite diagenesis using pixel-based methods. A geocellular model was built to evaluate options to improve throughput rate in the A Sands. This model is 24,000×20,000 feet with cell dimensions of 150x150x0.5 ft. It incorporates 237 faults and 11 flow units. Both object- and pixel-based techniques were used to distribute facies and petrophysical properties. A challenge in this model was in representing reservoir compartmentalization due to faulting of thinly bedded sandstones. These examples demonstrate the need for "fit-for-purpose" models that best represent key heterogeneities for a given reservoir-modeling/reservoir engineering objective.

References

1