Beginning with the End in Mind: Exploration-Phase Reservoir Modeling of Multi-Story Channel/Levee Complexes, Deep-Water Nigeria
Published:December 01, 2006
Douglas S. Moore, David T. McGee, David W. Burge, Nathan A. Geier, Gary Wu, 2006. "Beginning with the End in Mind: Exploration-Phase Reservoir Modeling of Multi-Story Channel/Levee Complexes, Deep-Water Nigeria", Reservoir Characterization: Integrating Technology and Business Practices, Roger M. Slatt, Norman c. Rosen, Michael Bowman, John Castagna, Timothy Good, Robert Loucks, Rebecca Latimer, Mark Scheihing, Hu Smith
Download citation file:
Geoscientists commonly employ high-quality three-dimensional seismic data to reduce deep-water prospect risk. For years, direct hydrocarbon Indicators (DHIs) have addressed hydrocarbon presence, however only more recently have images from three-dimensional datasets been employed to interpret the subsurface reservoir architecture of deep-water sinuous channel systems. Such architectural understanding, particularly in multi-story deep-water channel complexes, is key to predicting well count and spacing, thus ultimately prospect value.
By combining seismic profile channel interpretation, volume interpretation and analog studies, detailed geocellular models were created for offshore Nigeria isolated and stacked multi-story channel complexes. Gross rock volume was defined by interpreting the channel complex top and base horizons that define the reservoir ‘container.’ Seismic facies analysis was applied to amplitude extractions representative of interpreted discrete architectural zones within the reservoir container. In turn, interpreted facies maps were used for conditioning layers within the geocellular model.
Dimensional information for facies objects was recorded from the three-dimensional seismic volume and from both subsurface and outcrop analogs. Individual facies objects were stochastically placed, heavily conditioned to seismic control, having an overall aim of replicating analog end members. The key identified uncertainties of net-to-gross and sand object connectivity were characterized quantitatively and qualitatively through the application of volume sculpting, opacity rendering, and multi-body detection techniques. Geocellular modeling-ranged results were assigned within the flow simulation process to condition well performance, and ultimately, risked pre-drill prospect value was calculated. Subsequent drilling of one of the modeled prospects provided calibration for the assumptions entered into the geocellular models, enhancing the value of the models as predictive tools for analogous leads/prospects.