Symposium Provides Guideposts For Applying IOR Techniques

(Tech Connections Column, June 2002, American Oil and Gas Reporter)

Having participated in the SPE/DOE Improved Oil Recovery Symposium in Tulsa, improved oil recovery is a natural focus for this column. In its 2002 worldwide survey, the Oil and Gas Journal (April 15, 2002) indicates that 11 projects were planned (one carbon dioxide miscible, one combustion, six hydrocarbon miscible in Alaska, and three steam projects).

The planned CO₂ miscible project in the North Hobbs Field in New Mexico is, according to Jeff Simmon’s comments during the IOR symposium’s plenary session, the largest project of its kind in 15 years. Simmons also noted that Oxy Permian was now the largest CO₂ producer in the United States.

Working from this experience base, he described some keys to success:

• Original oil in place per acre should be 50,000 barrels or better.

• Injectivity should allow at least 5 percent of pore volume per year.

• Rock should be such that the minimum miscibility pressure can be reached without exceeding the fracture gradient.

• Vertical sweep, heterogeneity, and viscous-dominated/gravity effects must be understood.

Also during the IOR plenary session, Gordon Moore from the University of Calgary outlined how high pressure air injection (HPAI), or in situ combustion, in light oil reservoirs could be applicable. Within the United States, Continental Resources is the dominant HPAI operator with five Red River dolomite projects in the Dakotas. Four of the Continental projects are mature, and significantly, they are all classified as “successful” in the OGJ survey. Continental will also operate a large project planned in the Cedar Hills North Unit in North Dakota.

Light oil HPAI is most applicable in deep, higher temperature reservoirs where spontaneous ignition occurs. These reservoirs are often low permeability, where low injectivity would make waterflooding difficult. In this environment, low temperature oxidation is typical, rather than the high temperature oxidation associated with early combustion projects.

Moore stated emphatically that each oil must be tested for its combustion characteristics. Early combustion projects were plagued with operational problems. Advancements in compressor technology now make HPAI injection more reliable. Pumping requirements can actually be reduced since wells often flow.

Improved oil recovery has become far broader than merely processes. The term now applies to a plethora of technologies. In a DOE-supported project, researchers in New Mexico have developed an exploration “expert tool” using neural network and fuzzy logic techniques (SPE No. 75145, R.S. Balch, et. al., “Regional Data Analysis to Better Predict Drilling Success: Brushy Canyon Formation, Delaware Basin, New Mexico”) to predict production potential. Public domain data (aeromagnetic, gravity, structure and thickness) and attributes were correlated with production potential (average bopm during first year of production) using neural network techniques.

The most meaningful attributes were dip azimuth of gravity, second latitude derivative of thickness, longitude derivative of gravity, and longitude derivative of structure. The correlation coefficient for a 466-well training set was 0.90, and 0.81 for a 54-well “blind” test. Potentials for each 40-acre location were calculated and mapped. For any location, there is now an estimate of potential (poor, average or good) for Brushy Canyon locations to allow exploration efforts to quickly focus on higher potential areas.

In Oklahoma, new Hunton development has revitalized parts of the patch. In this play, wells initially produce at relatively high water-oil and gas-oil ratios. With time, as wells are pumped rapidly, oil cut increases, especially when groups of wells are produced at high rates to lower the area reservoir pressure (see Sept. 2000 case study in Petroleum Technology Digest at www.pttc.org).

Simple decline curve analysis is not appropriate for estimating reserves. Knowing early production data and using equivalent time and automatic type curve matching techniques, Marjo Operating Co. and the University of Tulsa (SPE No. 75248, Jeff Frederick, et. al., “Production Type Curves for the Hunton Formation”) have developed a technique to estimate reserves, and predict permeability and skin factor. Results are useful for economic evaluation, planning surface facilities, and future field development.

Knowing macro-scale (facies level) heterogeneity is critical when considering improved oil recovery. In early work, Tyler and Finley (SEPM “Concepts in Sedimentary and Paleontology, 1991,” V3, p. 3-7) developed a heterogeneity matrix for classifying the macro-scale heterogeneity of different depositional environments. Working from a database of 500 IOR projects in clastic reservoirs from 50 countries, Schlumberger DCS and Heriot-Watt University (SPE No. 75148, Richard Henson, et. al., “Geologically Based Screening Criteria for Improved Oil Recovery Projects”) located successful projects in the heterogeneity matrix.

Successful projects were located over most of the heterogeneity matrix, not only in favorable portions. Results indicate the effect of heterogeneity on success or failure of an IOR process depends on the type of IOR being attempted. If you are considering an IOR project, this work is an interesting read as you try to get your hands around the heterogeneity issue. 

Editor’s Note: Referenced SPE papers may be ordered at www.spe.org.