Pore Analysis of Mississippian Carbonate Reservoir in Southwest Kansas Using Petrographic Image Analysis
Robert J. Dietterich, W. E. Full
Petrographic image analysis (PIA) provides pore data that can be analyzed via
vector-analytic algorithms and regression analysis. Results define various pore
types
within the reservoir, and indicate pore variables and specific pore
types
that influence permeability. This information can be used to determine reservoir
characteristics important to hydrocarbon production.
A reservoir in the St. Louis Limestone (Mississippian) in southwest Kansas
was subjected to PIA. The Damme field, located in Finney County, Kansas,
consists of an oolitic-skeletal grainstone with well
-developed intergranular
porosity. PIA, using a computer-based pore-analysis system designed at the
University of South Carolina, provided variables defining the pore complex.
Analysis of these variables via EXTENDED CABFAC/QMODEL unmixing algorithms
separated pores into distinct classes or clusters. Within the reservoir, five
clusters were defined; two clusters were from low-porosity areas represented by
small, rough pores with no interconnections.
The highly porous producing zone defined by three clusters contains medium to
large, moderately smooth pores with enhanced interconnections in two clusters
and isolated pores in the third. Regression analysis of pore types
and pore data
versus core-measured permeability indicates number and size of pore throats, and
certain pore
types
that influence permeability. Permeability can be predicted
with R2 values up to 0.98. Combining computer-based analysis with
petrophysical and geophysical data provides the information needed for a
detailed reservoir description.
AAPG Search and Discovery Article #91043©1986 AAPG Annual Convention, Atlanta, Georgia, June 15-18, 1986.