FORSTER, CRAIG, University of Utah, Salt Lake City, UT, KEVIN HESTIR, Utah State University, Logan, UT, MARJORIE CHAN, University of Utah, Salt Lake City, UT, and JOE KOEBBE, Utah State University, Logan, UT

ABSTRACT: An Integrated Approach for Characterizing Reservoir-Scale Heterogeneity Using Micro- and Macroscale Petrophysical Data: Implications for Ferron Sandstone Studies

The Cretaceous Ferron Sandstone of central Utah is a popular target for detailed mapping of lithology and sedimentary structure combined with minipermeameter testing of rock permeability. Although the dataset is growing rapidly, formal techniques remain to be developed that incorporate the micro- to macroscale data in reservoir-scale fluid migration models. Our method for characterizing and modeling reservoir heterogeneities uses emerging techniques in applied mathematics. Although originally developed to deal with fractures and faults, our approach is readily applied to reservoir rocks formed by any geologic process that can be represented with a set of physically-based rules. First, a physically-based model is used to develop, or grow, a synthetic system of heterogeneities constrain d by detailed field observations, flow-test results, and the inferred geologic origins of the reservoir rocks. Heterogeneities in the Ferron Sandstone include both fluid-flow units and their bounding surfaces. In our studies of fracture systems, simple mechanical rules dictate the pattern of fracture growth. In the Ferron Sandstone, rules governing computed heterogeneity structures must recognize the successive cycles of erosion and deposition associated with each facies of this reservoir analog. Where the results of fluid-flow tests are available, the computed system of heterogeneities is constrained using a new conditional simulation method.

Once a suitable representation of the micro- to macroscale features is developed, a homogenization method is applied to scale up and average the petrophysical properties of the synthetic reservoir. This provides a formal mechanism for preserving the influence of the fine-scale features within a coarse modeling grid. An iterative sequence of heterogeneity growth and subsequent homogenization at each scale of observation is used to define input parameters for a reservoir model. Our approach differs from those of other workers because we specifically recognize the geologic origins of the heterogeneities while developing a synthetic map of the reservoir structure. Because the resulting reservoir model is physically based, our approach yields an improved framework for designing petrophysic l and geophysical sampling strategies. In addition, the computed heterogeneity structures provide a foundation for extrapolating the relatively sparse data obtained in producing reservoirs.

AAPG Search and Discovery Article #90993©1993 AAPG Rocky Mountain Section Meeting, Salt Lake City, Utah, September 12-15, 1993.