--> Abstract: Quantification of Subjective Geologic Knowledge for Use in Reservoir Flow Models, by R. C. Faucette and L. E. Borgman; #90987 (1993).

Datapages, Inc.Print this page

FAUCETTE, R. CHRIS, and LEON E. BORGMAN, Department of Geology, University of Wyoming, Laramie, WY

ABSTRACT: Quantification of Subjective Geologic Knowledge for Use in Reservoir Flow Models

Computer simulations of fluid flow in reservoirs need values of porosity, permeability and unit thicknesses at unknown locations in order to solve for changes through time in pressure and fluid saturations. While geologists are rarely able to determine what the exact values will be, they can usually determine the range of possible values. The commonly used mathematical techniques assign a unique value at each location where the values are unknown. Therefore, they do not incorporate geologic uncertainty and subjective knowledge.

High or low values of reservoir parameters tend to persist over distances. Therefore, interpolated values at locations close to data points should have less variation than values at locations farther from data. Also, areas of high or low values often tend to be preferentially aligned (anisotropic). Covariance is a measure of the spatial persistence of a variable or the spatial correlation between two variables. Subjective geologic knowledge of anisotropy and spatial persistence of the processes that affect parameters can be used when estimating covariances, thus quantifying this subjective knowledge. Stochastic interpolation allows the values to vary within preset limits, but the variability is constrained by the covariances and measured values are honored. These values can then be us d as input for computer simulations, each of which will give a different answer. The range of answers can then be used by geologists, engineers and managers to assess risk.

An example of stochastic interpolation as input for a flow model is given using an outcrop study of a Cretaceous shallow marine sandstone in southwestern Wyoming.

AAPG Search and Discovery Article #90987©1993 AAPG Annual Convention, New Orleans, Louisiana, April 25-28, 1993.