Reservoir Characterization in the San Andres Formation of Vacuum Field, Lea County, New Mexico: Another Use of the San Andres Algerita Outcrop Model for Improved Reservoir Description
Emily L. Stoudt, Michael A. Raines, 2004. "Reservoir Characterization in the San Andres Formation of Vacuum Field, Lea County, New Mexico: Another Use of the San Andres Algerita Outcrop Model for Improved Reservoir Description", Integration of Outcrop and Modern Analogs in Reservoir Modeling, G. Michael Grammer, Paul M. “Mitch” Harris, Gregor P. Eberli
Download citation file:
Since its discovery in 1929, Vacuum field, Lea County, New Mexico has produced 361 million bbl of oil and 298 bcf of gas from the Leonardian–Guadalupian (Permian) San Andres Formation. Despite this impressive performance, 65–70% of the estimated original oil in place remains in the ground. Complex variations in the porosity and permeability of reservoir lithologies are the principal causes for incomplete hydrocarbon recovery. Dolomitized oolitic-peloidal and fusulinid-peloidal packstones and grainstones constitute the productive reservoir facies. Interbedded in these porous units are tight, anhydritic or quartzose dolomudstones to mud-rich dolopackstones and dolomitic sandstones. Cores recovered from the nonporous intervals reveal textures, fabrics, and grain types indicative of (1) deposition in tidal-flat (peritidal) environments or (2) diagenetic modification, including collapse breccias, sinkholes, caves, and vertical fractures plugged with quartz sand or anhydrite cement, suggestive of exposure (karst) overprinting.
Early geologic models of the Vacuum field were primarily generated from petro-physical data, resulting in lithostratigraphic correlations that crossed time lines and flow units. Application of a chronostratigraphic framework for the San Andres based on outcrop studies results in an updated reservoir model for Vacuum field. Improvements include (1) recognition of localized, tight, tidal-flat cycles in separate horizons in the youngest San Andres high-frequency sequence (HFS) instead of correlating all nonporous, high gamma-ray intervals as a single, regional quartzose sandstone (Lovington sand), (2) identification of bypassed pay in porous, strike-parallel dolopackstones that are stratigraphically equivalent to, but downdip from, the tidal flats, and (3) recognition of the presence of tight karst intervals in older HFSs that compartmentalize the most continuous San Andres pay interval. Deeper, uncored San Andres lithologies can be more accurately characterized by comparing petrophysical information with facies attributes from older San Andres outcrops. An untested reservoir may occur in basal San Andres skeletal dolopackstones that are sealed by dolomudstones.
Figures & Tables
Integration of Outcrop and Modern Analogs in Reservoir Modeling
Building robust 3-D reservoir models is a major challenge that requires incorporation of geologically defined input parameters. This publication provides an overview of current approaches used in the development of geologically constrained and integrated reservoir models. Each of the 18 papers addresses various stages in the process of creating a reservoir model through the development and incorporation of an analog, extracting the quantitative input parameters on lateral and vertical variability, and the development and modification of a 3-D reservoir model based upon geologically constrained data. This applied volume is divided into two sections. The first is a set of papers illustrating the value and methodology of acquiring geometrical data on the lateral and vertical distribution of reservoir facies, within a sequence stratigraphic framework, using both outcrop analogs and detailed study of modern depositional systems. The second section includes both case studies where outcrop and modern analog data have been incorporated into subsurface reservoir models, as well as papers that illustrate recent advances in simulation and geostatistical methodologies. Together, the two sections provide a comprehensive look at integrated reservoir modeling.