|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Voluntary Unitization of 97% of the Empire Abo Pool, Eddy County, New Mexico, went into effect October 1, 1973. A reservoir study had shown that Unitization would mean recovery of an added 30 million barrels of oil. This would result from aiding the natural gravity drainage by shutting in high gas-oil ratio wells and returning available plant residue gas to the secondary gas cap. A later study showed further improvement in rates and recoveries by drilling additional development wells in low gas-oil ratio areas. Continuing engineering analysis and an aggressive attitude by the Unit owners has resulted in substantial increases in oil recovery and productivity as compared to continuation of productivity as compared to continuation of competitive primary operations.
The Empire field is located about eight miles southeast of the city of Artesia in Eddy County, New Mexico (Fig. 1). The discovery well for the Empire Abo reservoir was the Amoco-Hondo Oil Company - Malco "A" No. 1, completed at what is now the Unit M-14 location (Fig. 2), in November 1957. Development moved rapidly to the west, east, and north from the discovery, which proved to be only one location removed from the fore-reef edge of productive limits of the reservoir. Within three years some 215 of the eventual 250 producing wells had been completed. In defining producing wells had been completed. In defining the reservoir, 29 dry holes were also drilled. Field performance before and after unitization is shown in Figures 3 and 4.
Current pool producing rate for November 1976 is 41,700 BOPD, with the ARCO-Empire Abo Unit producing 41,062 BOPD of that total. Unit gas production in November was 55,746 MCFPD. Unit injection into the Abo gas cap in November was 34,451 MCFPD.
Atlantic Richfield Company, with 34.14% interest, operates the Unit for the 112 working interest owners.
A three-layered model with vertical equlibrium pseudo-relative permeabilities and pseudo-capillary pseudo-relative permeabilities and pseudo-capillary pressures was used to match the results from a pressures was used to match the results from a 22-layered model of a portion of the Empire Abo field. Results of the two models were almost identical in both history match and prediction phases. Directionally dependent pseudo-relative permeabilities were introduced to match the drainage mechanism of the reservoir. Both horizontal displacement and drainage mechanisms can be modeled simultaneously in a multilayered system using these pseudo functions. The excellent comparison between the finely gridded model and the pseudo model allowed the simulation of alternative recovery schemes using the pseudo model with the same confidence as the pseudo model with the same confidence as the original three-dimensional model. In particular, a comparison between pressure maintenance with carbon dioxide and nitrogen was simulated, with a substantial savings in both manpower and computer time. Development of this pseudo model makes the simulation of the entire Empire Abo field feasible.
For several years vertical equilibrium pseudo-relative permeabilities and pseudo-capillary pressures have permeabilities and pseudo-capillary pressures have been used throughout the industry to collapse a three-dimensional model with many layers to a two-dimensional areal model with a single layer. The obvious advantages of pseudo models are reduced computer and manpower costs. The justification for the use of pseudo models has been the comparison of results from one-dimensional models using pseudos with those from two-dimensional, cross-sectional models. Few comparisons have been made between an areal model using pseudos and a three-dimensional simulation. The Empire Abo field offered an excellent example for such a comparison. The field had been simulated in three dimensions, using 22 layers. An analysis of the simulation results showed that conditions were such that vertical equilibrium existed in most columns of the model. By applying pseudos to the three-dimensional model, it appeared possible to reduce the number of layers and associated computer time substantially. In this paper, the techniques for obtaining a match of the three-dimensional models with a pseudo model are described. The original three-dimensional model is described along with a history of the Empire Abo field. Coning correlations and directionally dependent pseudo-relative permeabilities also are described. Results between the pseudo model and the three-dimensional model also are compared. Finally, the use of the pseudo model for predicting alternative recovery schemes is discussed.
The Empire Abo field is located in southeastern New Mexico, about 8 miles (13 km) southeast of Artesia, NM (Fig. 1). The Abo reservoir was discovered by Amoco Production Co. in Nov. 1957. Subsequent development and unitization of the Abo reservoir was documented by Christianson. The producing horizon is a carbonate reef, basal Permian (lower Leonard) in age. The productive reef is about 12 1/2 miles (20.12 km) long and 1 1/2 miles (2.47 km) wide (Fig. 2), covering about 11,339 surface acres (45.9 x 10(6) m2). The reef dips gently from southwest to northeast at about 1 degree.
ARCO Oil and Gas Company has drilled two horizontal drainhole wells in the Empire Abo Unit. A horizontal drainhole well is one in which the wellbore is turned from vertical to horizontal in a short radius and the horizontal hole is then drilled out some distance into the formation. These wells were drilled to evaluate the mechanical feasibility of the drilling process and to examine the effect producing through the drainholes would have on the well's tendencies to form gas cones. Although several problems were encountered while drilling the problems were encountered while drilling the drainholes, the drilling technique used does seem to be mechanically sound. The wells have not been on production long enough to fully evaluate their gas coning performance as compared to conventionally performance as compared to conventionally completed wells.
This paper will briefly examine the gas coning problem in the Empire Abo Unit, discuss some of the techniques used to limit gas coning in the Unit, and review ARCO's experience with horizontal drainholes.
ARCO Oil and Gas Company operates the Empire Abo Unit, located in the Empire Abo Pool of Eddy County, New Mexico (Figure 1). The Unit consists of approximately 11,000 acres and represents about 97% of the entire pool.
Production is from the Permian (Lower Leonard) Abo Reef dolomite at a depth of approximately 6200'. The productive reef development is productive reef development is approximately 12.5 miles long and 1.5 miles wide (Figure 2). The cross-sectional view in Figure 3 illustrates the massive reef development. The main producing mechanism is gravity drainage which is now supplemented by the injection of the residue gas into the gas cap.
The pool was discovered in 1957. Competitive development of the field on 40-acre spacing was rapid. The pool was operated on a competitive basis until 1973 at which time the Empire Abo Unit was formed. S.H. Christianson's paper gives a detailed description of the reservoir and discusses factors involved in the formation of the Unit.
Prior to unitization, approximately 96 MMBO were produced from the Abo Reef. An additional 91 MMBO have been produced from the date of unitization through April, 1980. The Unit's current production rate is about 25,000 BOPD. production rate is about 25,000 BOPD. GAS CONING IN THE UNIT
At the time of discovery, the Abo reservoir pressure was above the crude's bubble point. Competitive production of the pool resulted in a drop in pressure to below the bubble point and the formation of a secondary gas cap. By the early 1970's several wells in the up-dip area of the pool began to produce at gas/oil ratios (GOR's) in excess of the solution GOR indicating the production of free gas from the gas cap. Drill stem test information indicated that these high GOR wells were perforated below the level of the regional perforated below the level of the regional gas/oil contact. Localized depressions in the gas/oil contact around the wellbores (gas cones) were causing the high GOR production. production. Producing Empire Abo wells at high GOR's is undesirable for two reasons.
This paper discusses the objectives of full field simulation and the important considerations in full field model development. Major ARCO fieldwide models are described to illustrate these considerations and show how the use of full field simulation has developed. It is concluded that the reservoir management plans and investment decisions are made around full field modeling. Also, reservoir simulation has become more routine as computer costs have dropped. The trend of faster and larger computers should continue, increasing their use in reservoir engineering in the future.
There are a number of reasons to conduct full field simulations.
1) Full field simulation can give us insight into recovery efficiency under various drive mechanisms and can show us the variations throughout a complex reservoir. 2) It is a must for field development planning including new well drilling (location and timing). 3) Full field simulation allows us to evaluate artificial lift constraints and optimize separation and plant processing design. 4) Simulation studies support our points of view when dealing with partners, regulatory agencies, and foreign governments.
Full field simulations at a minimum maintain an overall material balance and yield regional performance characteristics of complex reservoirs. This alone can keep our problems in focus and prevent faulty judgments from prevailing. Coarse grid simulations are accurate many times in predicting recovery efficiency depending on the reservoir description and dominant drive mechanism. For example, a coarsely gridded areal model can predict a dominant gravity drainage mechanism quite accurately.
An important consideration in full field modeling is what type of simulation will be sufficient to achieve our objectives of the full field study. Is a dual-porosity simulator necessary? Is a compositional model necessary or will a black oil model suffice? Are thermal properties important? Is tracking important?
The most important consideration in designing a full field model is to understand the major description and producing characteristics, and the status of the reservoir. What is the Kv/Kh ratio? What will the gas cap effect be? Does the oil zone have a tar mat barrier? Are variable PVT properties essential for proper description? Will there be an active water drive? Will faults serve as complete or partial horizontal flow barriers? Will the dip require a special grid design? Is the study in the early or later part of the reservoir's producing life? The answers to these and other questions determine whether the development of the full field model will be straight forward or complex.
An example of a recent fieldwide grid design of our Bima Field study is illustrated in Figs. 1 and 2. The horizontal gridding was more dense on the northern ham because only this portion of the field has been developed. The vertical zonation scheme reflects geologic flow units. Layer 1 is the upper poorly-consolidated zone which mantles the structure and has the highest permeabilities. Layer 2 is a gross consolidation of permeable permeabilities. Layer 2 is a gross consolidation of permeable oil-stained burrows and tight cemented zones. A net-to-gross ratio is required in this zone to honor the true "pay" thickness and pore volume. Layer 3 is a low permeability zone between the pore volume. Layer 3 is a low permeability zone between the upper pay intervals (Layers 1 and 2) and the lower pay zone in Layer 4. Layer 3 could have been removed from the model if the upper and lower pay zones were isolated from each other. However, pressure communication and coning of water and gas across Layer 3 has been observed. The lowest interval (Layer 5) provides the aquifer support for bottom water drive observed in provides the aquifer support for bottom water drive observed in some areas of the field. The Bima full field model will require special simulator treatments because of unique reservoir problems. This includes depth-dependent fluid properties, problems. This includes depth-dependent fluid properties, pressure-dependent permeability, pseudo horizontal well pressure-dependent permeability, pseudo horizontal well productivities, gas and water coning and tonguing correlations, productivities, gas and water coning and tonguing correlations, and well management routines to honor electrical submersible pump characteristics. pump characteristics. EARLY STUDIES
Full field simulation studies by ARCO have been project- or need-driven and designed around description and objective considerations. Several previously published ARCO studies illustrate this point. One of the early full field simulations was an update study of the performance of the high pressure miscible recovery project at the University Block 31 Middle Devonian Reservoir in Crane County, Texas. The objective of this study was to develop a reservoir management tool that could be used for both field and plant optimization in the future.
Renewed interest in horizontal drainhole drilling has occurred in the last
few years.(l-g) During the early 1950's interest in the technique was high and reportedly many drainholes were drilled. Numerous patents and several articles appeared at that time.(lO-l3) In the late 1960's Sinclair Oil and Gas Company, prior to its merger with Atlantic Richfield Company, advanced horizontal drainhole drilling technology in connection with experiments to recover shale oil. Sinclair found early 1950's techniques unreliable in obtaining a horizontal drainhole and proceeded to develop methods to drill off a high-angle whipstock, continue building angle to near horizontal (90°), and then stabilize the drilling assembly to limit further angle change. A patent was issued covering the key points of this method. (14)
Since mid-1979 ARCO Oil and Gas Company has drilled and produced four horizontal drainholes in the Empire Abo Unit, Lea County, New Mexico. These horizontal holes were drilled into the lower part of the oil column to evaluate their effectiveness in reducing gas coning during production. Studies had indicated that reducing coning would lead to higher oil recovery per well. Drilling improvements achieved during these four operations will be discussed. The production history for the first well will be compared with that of conventional off set wells.
DRILLING THE HORIZONTAL DRAINHOLES
Details of the horizontal drainhole drilling operations for the Empire Abo wells and a description of the reservoir, have been previously presented. (1-8 ) The goal of drilling nearly 209 feet of drainhole from the well configuration shown in Figure 1 was not achieved in the first two wells due to reaching torque limits. Table 1 shows the footage drilled for each of the four drainholes. Reduced torque was achieved on the last two drainholes. In each of these cases drilling stopped when all available wiggly drill collars had been run through the drainhole curve. Using current techniques and additional flexible collars, it appears feasible to drill drainholes 300 to 500 feet long.
Shown in Tables 2-5 are t.4.- survey data for each drainhole. Figures 2-5 arc, isometric drawings which illustrate the path of each well. Figure 6 compares costs in constant 1981 dollars per foot for drilling the drainhole in each well. This cost represents the cost of drilling only the drainhole portion of the well. Included is the cost from the time at which the whipstock is ready to run until the drainhole is finished and surveyed. The fat age drilling cost was reduced by more than half for the last two wells compared with the first two wells. For each situation the total well cost would include drilling the vertical portion of the well, casing it, locating the gas-oil contact, and preparation for whipstock setting.
The cost of preparing for whipstock setting is minimal for an open hole completion. When a drainhole is initiated from inside casing, milling the casing in advance of drainhole drilling can be expensive compared with the open hole completion.