Performance Review of a Miscible CO2 Tertiary Project: Rangely Weber Sand Unit, Colorado
- J.R. Hervey (Chevron U.S.A., Inc.) | A.C. Iakovakis (Chevron U.S.A., Inc.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- May 1991
- Document Type
- Journal Paper
- 163 - 168
- 1991. Society of Petroleum Engineers
- 5.4.7 Chemical Flooding Methods (e.g., Polymer, Solvent, Nitrogen, Immiscible CO2, Surfactant, Vapex), 4.6 Natural Gas, 5.2.1 Phase Behavior and PVT Measurements, 5.5 Reservoir Simulation, 5.7.2 Recovery Factors, 5.1.1 Exploration, Development, Structural Geology, 3.1.5 Plunger lift, 5.3.4 Reduction of Residual Oil Saturation, 5.4 Enhanced Recovery, 4.3.4 Scale, 5.4.1 Waterflooding, 4.3.3 Aspaltenes, 1.14 Casing and Cementing, 5.1.5 Geologic Modeling, 2.4.3 Sand/Solids Control, 5.4.9 Miscible Methods, 1.8 Formation Damage, 5.5.8 History Matching, 5.4.2 Gas Injection Methods, 4.2.3 Materials and Corrosion, 5.1 Reservoir Characterisation, 3.1.2 Electric Submersible Pumps, 1.6 Drilling Operations, 4.1.2 Separation and Treating, 5.3.2 Multiphase Flow, 4.1.5 Processing Equipment, 1.6.9 Coring, Fishing, 5.2 Reservoir Fluid Dynamics
- 4 in the last 30 days
- 507 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
A miscible CO2 flood as initiated in the Rangely Weber Sand Unit (RWSU) inColorado in Oct. 1986. The project calls for injection of a 30% slug of CO2 ata 1:1 water-alternating-gas (WAG) ratio. Ultimate incremental recovery isestimated to be 114 million STB. Current incremental recovery accounts for34.5% of production.
The RWSU is in Rio Blanco County in northwestern Colorado about 5 miles eastof the Utah/Colorado border (see Fig. 1). The producing formation for the unit,the Weber sandstone, was discovered in 1933. Initial development of the fieldon 40 acre spacing, however, was not completed until 1949. The unit currentlyis being processed primarily on 20-acre spacing. A 10-acre infill-drillingprocessed primarily on 20-acre spacing. A 10-acre infill-drilling programstared in 1983 was abandoned after marginal results. In all, program stared in1983 was abandoned after marginal results. In all, 893 wells have been drilledin the Weber formation. Oil production from Rangely peaked at 82,000 BID inmid-1956 (Fig. 2). To maximize production, the field was unitized andwaterflood operations began in 1958. Waterflooding in the RWSU has beenextremely successful. The estimated ultimate primary plus secondary oilrecovery from the unit will be 789 million STB, or 50% original oil in place(OOIP). Approximately 332 million STB, or 21.0% OOIP, can be attributed toprimary recovery. At the time of initial CO2 injection in Oct. 1986, the RWSUwas a mature waterflood with a cumulative recovery of 44.0% OOIP and aproducing WOR of 16:1. The CO2 project has extended the life producing WOR of16:1. The CO2 project has extended the life of the unit in that it accounts for34.5% of the unit's current oil production and more than 50% of the unit'sremaining production and more than 50% of the unit's remaining recoverablereserves.
The Weber is a Permian/Pennsylvanian Age formation. The anticlinal structurethat forms the trap has pronounced asymmetry, dipping at 15 to 300 on thesouthwestern flank and 60 to the north and northeast. The Weber depth rangesfrom 5,500 to 6,500 ft. The formation had a gas cap with an initial gas/oilcontact at 330 ft subsea (ss). The original oil/water contact was at 1,150 ftss. The Weber formation is heterogeneous and consists of a series ofinterbedded sandstones, siltstones, and shales. The seven major producinghorizons are separated by six shale breaks. Within each producing horizons areseparated by six shale breaks. Within each producing section, numerous minorshale intervals exist, which may producing section, numerous minor shaleintervals exist, which may or may not be continuous from one well to another.The average porosity of the effective producing zones is 13% (Table 1).Permeabilities range from 0.1 md to as high as 50 to 100 md. There is agenerally increasing trend in permeabilities and sand quality from east towest. The ratio of vertical to horizontal permeability varies from 0.25 to 0.5.permeability varies from 0.25 to 0.5. CO2 Project Design
As early as 1978, phase-behavior studies, corefloods, and slim-tubeexperiments were conducted with Rangely reservoir fluids and CO2 mixtures.These studies showed that a CO2 flood in the unit was technically feasible.Results led to the development of a fieldwide, fully compositional study inNov. 1983. The 1983 simulation study, used to justify the project, wasconducted with a multicomponent reservoir simulator. The model grid consistedof a 12-column x 17-layer cross section and represented one-eighth of afive-spot pattern. In addition, a 12 x 6 areal model and a 12 x 1 ID model wereused to help quantity areal (88%) and vertical (72%) sweep efficiencies. The1983 study showed that injecting a 30% slug of CO2 with a 1:1 WAG ratio wouldyield the optimum economic recovery. This injection scheme, currently beingused in the unit, incorprated injecting alternating volumes (0.015 HCPV) of CO2and water into each pattern. The total CO2 would then be succeeded by 1.0 HCPVof water. The study predicted an incremental recovery of 6.7% OOIP at theconclusion of the chase waterflood cycle. Further simulation efforts wererequired before startup in 1986 to accommodate revisions in the project design.Some of the hydrocarbon-removal facilities were eliminated from therecompression plant, changing the composition of the injected gas. Also, theactual reservoir pressure of 3,000 psia was higher than the designed minimummiscibility pressure of 2,600 psia. Fig. 3 shows the impact of injecting acontaminated CO2 stream on recovery. The composition index shown on the x axisis calculated by entering the decimal percentage of each component into theexpression. In the RWSU, the net impact of injecting CO2 with an increasedhydrocarbon concentration is an increase in the incremental oil recovery from6.7 to 7.0% OOIP. Fig. 4 shows the effect on recovery of increased reservoirpressure. Increasing the pressure from 2,600 to 3,000 psia should pressure.Increasing the pressure from 2,600 to 3,000 psia should increase the expectedrecovery from 7.0 to 7.2% OOIP. Early data from the CO2 project showed gasbreakthrough and incremental-oil response much sooner than originally forecast.At the same time, produced gas rates exceeded predictions. It therefore becameapparent that a simulation tool that described the reservoir geology in greaterdetail was needed. As a result, hybrid simulation has been used for modelingsince 1987. Hybrid simulation incorporates the use of fractal technology, amiscible-flood simulator, and an areal streamtube model. The fractalinterpolation model is a geostatistical tool used to develop a highly detailedgeologic description of the reservoir between well pairs. There are 3,000 cellsin the fractally generated cross section, 150 vertical x 20 horizontal layers.Each corner vertical cell is assigned a porosity derived from aporosity/permeability transformation obtained from core analyses. Verticalpermeability values also are entered from core-derived horizontal/verticalpermeability relationships. Fractal interpolation then assigns porositypermeability relationships. Fractal interpolation then assigns porosity andpermeability values to all cells. This high resolution allows betterrepresentation of heterogeneity and should reduce the errors resulting fromaveraging layer properties. The fractally derived cross section underinvestigation is infused into a Todd-Longstaff miscible-flood simulator togenerate fractional-flow curves for oil, gas, and water. This creates a 2Dmodel. To complete the hybrid simulation, fractional flow curves are enteredinto an areal model that relies on traditional streamtube techniques to examinebehavior between well pairs. Streamtubes have been simultaneously generated onall 750 active wells in the RWSU to minimize the boundary effects and errors.Some of the findings from using the hybrid simulator follow. 1. Crossflowoccurs between minor shale intervals as a result of large pressuredifferentials. 2. Production contributions as high as 40% can be made by layerswith porosities less than the previously used cutoff of 10% for effectivesand.
|File Size||465 KB||Number of Pages||6|