Hierarchical Approach to Facies and Property Distribution in a Basin-Floor Fan Model, Scarborough Gas Field, North West Shelf, Australia

OGIP of very dry gas within a large, very low-relief faulted anticline covering about 800 sq. km. It has been appraised by four additional wells and a 3D seismic survey, and is being evaluated for development. The Scarborough reservoir consists of Early Cretaceous deepwater turbidite sands deposited in a basin-floor fan setting. These sands were sourced from the expansive, northward-prograding Barrow Group fluvio-deltaic system located some 50 km to the south of Scarborough. The reservoir interval is a three-tiered fan sequence with variable sand content and quality: a highquality, high net-to-gross Lower Fan unit which contains the majority of the gas-in-place, overlain by lower net-to-gross and lower quality Middle and Upper fans. The dominant reservoir lithofacies are quartzose medium- and fine-grained sandstones which are largely unlithified and uncemented, with average porosities of greater than 30% and permeabilities of 100’s to 1000’s of millidarcies. The background lithofacies are mudstones and siltstones which straddle the silt-mud boundary. Static geological models and dynamic flow simulation models have been used to integrate seismic and well data, petrophysical analysis, and sedimentologic and stratigraphic interpretation. Outcrop analogues for deepwater basin-floor fan systems include the extensive, well-exposed and well-studied Permian Ecca Formation of the southern Karoo Basin, South Africa, the Carboniferous Ross Formation of western Ireland, and the Eocene Ainsa Formation of northern Spain. These outcrop analogues suggest that siltstone facies are bottom-loaded within genetically related depositional packages. Reservoir models were used to investigate the effects of stratigraphic organisation and lithofacies distribution on reservoir performance predictions. By reference to outcrop analogues, an hierarchical approach was developed to systematically distribute depositional facies and lithofacies within the models. This permitted investigation of the effect of stratigraphic features of different scales on predicted production rates, reservoir performance and individual well performance. Three hierarchical levels of facies models were incorporated into the stratigraphic zones. Two levels define deterministic and stochastic depositional facies geometries, and the third and finest level is lithofacies, or reservoir rock type. Seismic data were used to map large-scale depositional facies elements, and lithofacies were interpreted from well logs calibrated to conventional cores. The use of lithofacies distributed within depositional facies provides flexibility in the modelling workflow, provides the template for distribution of the rock properties of porosity, horizontal and vertical permeability, and water saturation, and allows systematic investigation of the effect of siltstone baffles on predicted flow streams, particularly on the timing of water arrival. The field is predicted to have a strong flank- and bottom-water drive in the Lower Fan and flank-water drive in the Middle and Upper fans. The siltstone baffles at various scales are expected to affect ultimate gas recovery, reservoir flow streams, and individual well performance, especially because of the limited gas column height and limited ability to handle water production. These baffles are expected to provide some protection from early water arrival.