RCRL Studies of Thief Zones in SACROC Field

Access full-text files



Journal Title

Journal ISSN

Volume Title



Geologic models of the northern platform of Sacroc have been constructed over several years (2001–present) by the Reservoir Characterization Research Laboratory (RCRL). These models show a complex stratigraphy of deposition, exposure (diagenesis), and erosion. Canyon units were deposited as depositional cycles, and the latest Canyon cycles have been heavily eroded. The Cisco units are composed of debris flows, biohermal buildups, and grain-dominated units representing many significant sea-level changes, culminating in a major exposure event before being engulfed in carbonate muds of Wolfcamp age.

Geologic models have been converted to petrophysical models by RCRL using the rock fabric method. Porosity-permeability transforms have been defined for each stratigraphic unit using core data and thin-section analyses. Geologic models have been converted to porosity and permeability models using these transforms.

Kinder Morgan has initiated a CO2 flood in the northern platform area. The company's injection program has uncovered numerous intervals in which the rate of injection is significantly higher than expected from the petrophysical model. These intervals are referred to as void space conduits by Kinder Morgan and as thief zones in this report.

The problem facing Kinder Morgan is that the large volume of injected fluid taken by thief zones significantly decreases the volume of remaining-oil saturation contacted by injected fluids, resulting in poor recovery. This report summarizes efforts to explain the geological and petrophysical nature of these thief zones through an exhaustive study of core and log data from well 37-11.

Thief zones are defined as having significantly higher flow rates than expected from matrix properties. At Sacroc, these zones are located by injection profiles, and the injection profile from 37-11 shows high variability in injection volume. Using core data, we calculated kh values for each perforated interval, as well for each injection and no-injection interval within each set of perforations. A positive relationship between kh and water injection is assumed. Unfortunately, in 37-11 only a polymer injection profile was available for this study. A positive relationship between kh and polymer injection is probably true as well.


LCSH Subject Headings