Miscible and Immiscible Gas-Injection Pilots in a Middle East Offshore Environment
This paper describes a gas-injection pilot that has been implemented in offshore Middle East carbonate reservoirs to assess injectivity, productivity, macroscopic-sweep efficiency, flow assurance, and operational efficiency in a field that has a long water-injection history.
Hydrocarbon-gas injection improves microscopic-displacement efficiency and generally acts as pressure maintenance; however, unfavorable mobility ratio can affect the ultimate recovery negatively because of viscous fingering and gravity override. This paper describes a gas-injection pilot that has been implemented in offshore Middle East carbonate reservoirs (a second pilot is described in the complete paper) to assess injectivity, productivity, macroscopic-sweep efficiency, flow assurance, and operational efficiency in a field that has a long water-injection history.
The carbonate field is part of the Lower Cretaceous Lower Lekhwair formation. The field is divided into A, B, and C reservoirs, 20 to 35 ft thick individually. Each reservoir is vertically separated by nonpay tight intervals of similar thickness, composed of argillaceous limestone, isolating each reservoir from the others. Depending on the reservoir, oolitic shoal facies can be more abundant. Near the flank area, the formation is water-wet, and, as one moves toward the crestal area, the formation becomes oil-wet.
The reservoir fluid is undersaturated at initial pressure of 4,200 psig at datum depth and a temperature of 220°F. Oil gravity is 40 °API, and oil contains 1–2 mol% carbon dioxide and an insignificant amount of hydrogen sulfide. All three reservoirs have common contact and similar oil properties, with a strong compositional gradient. Production from the field started in the late 1960s through natural depletion, which indicated weak aquifer support.
Before pilot implementation, laboratory experiments were performed to determine the feasibility of hydrocarbon-gas injection. Pressure/volume/temperature experiments showed that the minimum miscibility pressure (MMP) is approximately 4,500 psia, which is higher than initial reservoir pressure, but the crude oil has a strong swelling effect. Unsteady-state coreflood experiments performed with a 200-cm-long core showed a recovery of 70% for immiscible flood and a recovery of 92% for miscible flood.
Full-field compositional simulation of gas injection incorporating a tuned equation of state indicated significant incremental oil and reasonable pressure support.
In Reservoir B, tertiary injection was performed in the transition zone, close to the oil/water contact in an area that had already experienced peripheral water injection. Injected gas was from the first-stage separator.
Pilot B: Description and Performance. Pilot B was also a line-drive pilot, but was located in the waterflooded transition area of Reservoir B. Oil was produced with the long string of producer Well P-1. The long string of observer Well Obs-1 was used to monitor the vertical gas efficiency and pressure. Gas was injected from the horizontal injector Well I-2, which is perforated at the lower part of the reservoir and is approximately 2,300 ft in length. The producer well is located 2,360 ft away from the midpoint of the horizontal section of the injector, while the observer is located 460 ft away from the injector.
A tight zone is located 5 ft above the base of the reservoir, and there are two high-permeability streaks in the middle of the formation in the producer Well P-1. The average oil saturation in the pilot area is estimated to be approximately 0.30 units, through a different set of logs. Pressure measurements showed a 0.057‑psi/ft pressure gradient toward the producer, owing to historical peripheral-water-injection field development.
The average pressure measured in the pilot area is approximately 4,900 psia, which is higher than the MMP of injected gas.
Injection/Production Performance. The gas injection was initiated in March 2002, with an average monthly injection rate of 15 MMscf/D over the first 3 months, followed by a rate increase to 25 MMscf/D, then a gradual reduction to 10 MMscf/D after gas breakthrough. 20 Bscf of gas was injected during this duration, corresponding to 1.49-pore-volume injection.
Because the pilot was located in a transition zone that had already been waterflooded, the producer showed oil production of 300–400 STB/D with 80% water cut immediately after commencement of gas injection. After 1 year of injection, the well showed an increase in oil production with decrease in water cut, indicating arrival of the oil bank and the effectiveness of gas injection. However, gas/oil ratio (GOR) then increased rapidly, indicating gas breakthrough in July 2003, after only 1.5 years of injection. The producer well was closed in April 2004 because of surface-facility constraints. Nevertheless, the well was tested biannually to monitor gas movement.
Monitoring Performance. The pressure survey of the producer and observer confirmed that, during the whole period of injection, there was good pressure support in the pilot area and the pressure always remained higher than the MMP. Three perfluorocarbons were used as gas tracers and were injected into Well I-2 in July 2002 to compare their travel time with the producer and assess areal sweep.
Production-logging tools were used to evaluate vertical sweep efficiency. The presence of gas was indicated in the middle of the formation at Well Obs-1, which is in line with the presence of the high-permeability streak. Early breakthrough occurred through this high-permeability streak, and, afterward, vertical sweep began improving, as seen by the presence of gas at the top of the formation.
In 2010, the producer well was sidetracked for coring and openhole logs to evaluate gas-injection performance. Low values of saturation were observed in the high-permeability zone, while high values were found in the dense zone. The sampling performed at the top and bottom of the formation observed 100% gas, indicating the presence of residual oil saturation in the vicinity.
Asphaltene deposition was not observed in Pilot B producers. This may be attributed to low asphaltene content in the crude oil (0.1 wt%) as well as lower oil saturation owing to the fact that oil has already been swept by water in the zone around gas-injection Pilot B.
History Match of Pilot B. Model Description. A full-field compositional model was used to history match Pilot B, incorporating the effect of existing pressure gradients and peripheral-seawater-injection sweep. Similar local grid refinement has been carried out vertically and horizontally in the area surrounding Pilot B to reduce numerical dispersion of composition and saturation fronts.
Pressure Match. Pressure in the Pilot B area was above the MMP during the production period. It is important to replicate this phenomenon in the compositional simulation. The simulation model is able to accurately capture the pressure profile along Wells Obs-1 and P-1 before commencing the pilot and can be used to represent the fluid behavior inside the pilot area.
An acceptable history match was obtained, which captures the major features of three-phase fluid movement, and it can be used to predict the areal and vertical efficiency of this miscible pilot.
Areal and Vertical Sweep. Because of the lack of nearby wells around the pilot area, it was difficult to assess areal-sweep efficiency with high accuracy. Simulations are in agreement with water-cut observations for Well P-1. Water cut decreases after start of gas injection in both simulations and pilot results.
The gas breakthrough occurs from the high-permeability streaks that are located in the middle of the formation. The same phenomenon is seen in simulation, as illustrated in Fig. 1. The evolution of gas was also captured in the simulation model, which predicts the improvement in vertical sweep over the course of the pilot as measured in the observer well. Simulations predict that both areal and vertical sweep of the pilot were acceptable. Because the pilot lies in a transition zone that has low initial oil saturation, most of the oil has already been recovered by peripheral water injection.
The pilots demonstrate that gas injection helps improve pressure support and production, whether in miscible or near-miscible conditions. Results also indicate that gas injection can be conducted in tertiary mode to improve overall recovery in this field. However, the major challenges that affect recovery are gravity segregation and reservoir heterogeneity.
The Pilot B history-matched simulation model was used to evaluate the benefit of water-alternating-gas (WAG) -injection implementation instead of tertiary gas injection in the Pilot B area. A WAG-injection ratio of 1:1 and a 6-month cycle were used for the screening study. In this model, hysteresis and gas-trapping effects are not considered because laboratory experiments have not yet been performed to evaluate these parameters. The operational constraints of Pilot B are used in the model—that is, a GOR limit of 15 Mscf/STB and water cut of 95%.
When considering the performance comparison of tertiary Pilot B and WAG injection, WAG injection recovers significant additional oil with the same operational constraints compared with tertiary gas injection. WAG injection not only produces extra oil but also reduces by half the amount of gas required. The WAG-injection gas-usage factor, the amount of gas required to produce one unit of oil, is significantly lower than that for tertiary injection. Moreover, WAG injection can be optimized further by reducing gas volume (tapering) in later cycles.
- A gas-injection pilot was conducted in a Lower Cretaceous carbonate formation. Improvement in pressure support and production performance was observed in the pilot.
- Pilots have been carried out in the flank area, where the pressure gradient, owing to peripheral water injection, significantly affects the gas movement and behavior.
- The vertical-sweep efficiency was affected by reservoir geology as Pilot B observed early gas breakthrough through high-permeability streaks.
- A good history match of both pilots has been obtained, suggesting that miscible injection occurred in Pilot B. Tertiary miscible injection of Pilot B recovered an additional 16% of stock-tank oil in place compared with waterflooding.
- Gas injection in WAG mode is a possible solution to control mobility and improve recovery. Simulations performed on the history-matched model using a WAG-injection strategy give higher oil recovery with lower GOR.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper IPTC 18513, “A Case Study on Miscible and Immiscible Gas-Injection Pilots in a Middle East Carbonate Reservoir in an Offshore Environment,” by Jitendra Kumar, Pawan Agrawal, and Elyes Draoui, Abu Dhabi Marine Operating Company, prepared for the 2015 International Petroleum Technology Conference, Doha, Qatar, 7–9 December. The paper has not been peer reviewed. Copyright 2015 International Petroleum Technology Conference. Reproduced by permission.