A Geologic-Based Reservoir Modeling Workflow from 3-D Outcrop Models of the Brushy Canyon Formation

James Borer, Michael Gardne, Mary Carr, and Safian Atan
Colorado School of Mines, Golden, CO

The combination of sparse data control, geologic uncertainty and the need to reproduce multiple scales of heterogeneity and geologic trends makes geostatistical reservoir modeling as much an art form as a science. We illustrate a decision tree for incorporating geologic information into deep-water reservoir models. An suite of 3D geologic, petrophysical, and geophysical models derived from 2.5D outcrops of the Permian Brushy Canyon Formation provide a benchmark for assessing how both geologic and modeling decisions impact modeling results.

Geologic data is inherently non-stationary. Vertical and lateral trends occur at multiple scales and statistical domains have transitional and sharp boundaries. The success of an approach depends on the ability to define model domains through the use of knowledge-based layers, regions and trend maps. The challenge is to intelligently distribute geologic information between sparse data points. Geology can help because certain geologic attributes are spatially and temporally linked through formative processes and are predictable. We outline a region-based approach for modeling deep-water reservoirs. As the principle transmission site for subaqueous flows, submarine channels determine the run-out length, gradient, topography, and degree of confinement and flow transformation. Lateral (channel, flank and interchannel) and longitudinal (proximal, distal) regions delineate confined, partially confined and unconfined flow and control lithology, correlation length, trends, body shape, arrangement and connectivity. Inside channels, bar-scale regions are used introduce a pattern of alternating reservoir quality related to channel curvature. Stratigraphic trends are modeled by varying parameters within and between individual layers with regard to hierarchical stratigraphic and sedimentologic models of submarine fan deposition. These models predict spatial change in channel type, size, branching, stacking, net-to-gross structure, facies and channel drapes.

AAPG Search and Discovery Article #90039©2005 AAPG Calgary, Alberta, June 16-19, 2005