Automatic Reservoir Modelling: A Sate of the Art

Sedimentary facies are a key control on the distribution of spatial heterogeneity and geological facies modelling is a critical step in the prediction of hydrocarbon reserves. Outcrop data provides analogue information that imparts greater detail on the reservoir architecture, geometries and inter- and intra-facies connectivity that can not be obtained from the subsurface alone. The Campanian Blackhawk Formation provides data of a well exposed fluvial complex system that fed wave-dominated Cretaceous shorelines. Outcrop data derived from a Unmanned Aerial Vehicle (UAV) virtual outcrop (VO) model of the Beckwith Plateau, Utah, USA was used to validate a range of traditional reservoir modelling approaches, conditioned to reservoir information from the Solider Creek Pilot Project where the Blackhawk Formation composes the reservoir interval at Soldier Creek, Utah. Results of two methodologies for generating training images for subsequent simulation with multiple-point geostatistics were benchmarked against two traditional object- and pixel-based algorithms, creating different scenarios and realizations for comparison. Impact of the simulation algorithm on the distribution of heterogeneity was investigated through a fluid-flow dynamic simulation over a 30 year field lifespan.