Spatial Statistics of Permeability Data from Carbonate Outcrops of West Texas and New Mexico and Implications for Fluid Flow and Reactive Transport Modeling

Sunday, February 13, 2000 - 11:00am - 12:00pm
Lind 409
Jim Jennings (The University of Texas at Austin)
Joint work with Steven L. Bryant, and Rashidul Hassan.

For 10 years the Carbonate Reservoir Characterization Research Laboratory at The University of Texas at Austin has been collecting petrophysical data from carbonate outcrops in West Texas and New Mexico to advance knowledge in the geological, petrophysical, geostatistical, and fluid flow aspects of this important class of fresh water aquifer and hydrocarbon reservoir rocks. This presentation summarizes the spatial statistics of permeability data collected from these outcrops and the implications for fluid flow and reactive transport modeling in analogous subsurface situations. In addition speculations will be offered on the role of reactive transport in the origin of some of the observed heterogeneity patterns.

The permeability data from these outcrops exhibit 2-5 orders of magnitude variability, much of it occurring within distances of a few feet or less in single rock fabric units. This short-range variability is present in every outcrop and composes most of the overall variance in each case. It has weak spatial correlation that can be modeled with semivariograms having asymptotic power-law behavior at small lags.

A variety of longer-range features are also observed including (1) vertical trends within grainstone bodies, (2) vertical average permeability contrasts between grainstone bodies, (3) 140-180 ft lateral periodicities within high frequency cycles, and (4) lateral trends at scales from several hundred to several thousand feet in high frequency cycles. The longer-range features compose a smaller fraction of the overall variability and may require careful analysis to detect.

Stochastic models explore the fluid-flow effects of these heterogeneities. Tracer and waterflood simulations demonstrate that the long-range features can control overall large-scale displacement, even though they compose much less than half of the overall variance. Special attention when modeling them is encouraged. The small-scale variability contributes to smearing of displacement fronts and enhancement of larger scale effective permeability.