Preserving Fine-Scale, Irregularly-Shaped Geological Features in Reservoir Flow Models Using Edge Properties

Lisa Stright and Jef Caers
Stanford University, Stanford, CA

Conceptual geologic reservoir models, historically qualitative in nature, are now used as training images in conjunction with multiple-point geostatistics to build detailed 3D facies architecture models from observed and interpreted outcrop, core, well and seismic. Although useful for describing the multi-scale nature of the facies and facies relationships within a depositional system, these models must be upscaled to a flow model grid to allow for decision making through reservoir simulation. During the transformation into effective property models, the intended multi-scale heterogeneity that controls reservoir flow and recovery may be lost. One of the major challenges in reservoir modeling and simulation lies in our ability to capture the information at these various scales when it impacts the recovery prediction of a reservoir model.

To preserve important fine-scale geological information, an additional modeling variable is introduced as the edge of a model cell. This cell edge is a continuous or categorical value that is associated with the cell face, rather than the cell center which is often reserved for facies types and/or petrophysical properties. Modeling with edges entails generating a 3D training image of edges and using it in conditional pattern-based multiple-point geostatistical simulations. An application is presented in which edges represent fine-scale, irregularly shaped flow barriers. The flow barriers are channel drapes that either continuously or discontinuously line channels in a turbidite channel complex. One of the great advantages of modeling with edges is their direct link with transmissibility in flow simulators. This 3D example is presented from modeling through simulation.