SPE 74391

Natural Gas-Lift: Theory and Practice

S. Betancourt, Schlumberger, K. Dahlberg, Norsk Hydro, . Hovde, Norsk Hydro, Y. Jalali, Schlumberger
Copyright 2002, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the SPE International Petroleum Conference and
Exhibition in Mexico held in Villahermosa, Mexico, 1012 February 2002.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
This paper examines the concept of natural gas-lift or
production of oil by in situ gas. This process involves the
commingled production of an oil reservoir and a contiguous or
non-contiguous gas zone in a controlled mode as an alternative
to artificial gas-lift. Besides the normal conception of gas-lift
as a remedy to high water-cut flow, lift assistance may be
required at startup to commission production, and can also be
required at low to moderate water-cut, when the wellhead
pressure requirements are stringent. The latter is typical of
subsea installations and platform installations with highpressure processing. Contiguous gas-lift is a more complex
process, due to the interaction of the oil column with the
overriding gas-cap, but can be envisaged more easily than
non-contiguous gas-lift which requires the presence of a
suitable gas zone or depleted oil zone. We present results of
numerical modeling of the contiguous gas-lift process for
horizontal wells, for the case of a conceptual reservoir model
with characteristics similar to certain North Sea provinces.
Results show the applicability of natural gas-lift dependent
upon standoff (with respect to the initial gas-oil and water-oil
contacts) and target production rate. We also address design
considerations for natural gas-lift applications and report the
operational experience gained in the Troll field with
contiguous or gas-cap gas-lift applications. Finally we
examine a gas-lift application of the non-contiguous type.
Introduction
Artificial lift represents a major cost in the operation of oil
fields, and this cost becomes more severe as the production
environment becomes more challenging, as is the case of
offshore developments and remote areas. This paper evaluates
the alternative of using the energy of the reservoir by

controlling remotely the amount of gas that is produced, in

what is called natural gas-lift. Many fields have gas zones
that are part of the produced reservoir or separate gas or high
GOR zones. Reservoirs with thin oil zones between a gas-cap
and a bottom aquifer are natural candidates for this
application, but depleted oil reservoirs with or without
adjacent gas zones (provided that some source of gas is
accessible from the well) are also good candidates.
This paper is divided into three sections. The first section
addresses the behavior of an oil reservoir with a gas-cap and a
bottom aquifer. For this case, a conceptual reservoir model
was built to study the drivers for natural gas-lift applications.
The next section deals with the operational experience gained
in an actual field where the gas from the gas-cap is used to
produce the oil zone. Finally, the third section covers an
example of the use of a non-contiguous gas zone (lower than
the oil zone) that is used as a source of gas to enable oil
production.

Natural Gas-Lift from a Contiguous Gas Zone

This process applies mainly to reservoirs where the production
drive mechanism is dual drive (water encroachment at the
bottom and gas expansion on top). Depending upon well type
and completion, the water-cut is expected to increase with
time, and eventually artificial lift may be necessary to support
the production demand. From the recovery point of view the
displacement of oil by gas is more efficient than the
displacement by water. This is due to the lower residual
saturation of oil when gas is the displacing phase. Horizontal
wells are a very good alternative for the production of dual
drive reservoirs when the oil zone thickness is relatively small
(compared to the gas and water zones), because they provide
higher flexibility in well placement within the oil zone to
satisfy the production strategy. In fact the well placement and
the production rates play a major role defining what would be
the dominating production drive. Numerical and analytical
studies, as well as field experience, have shown that better
results are achieved by drilling the horizontal section as far as
possible from the gas-oil contact. Some studies have even
addressed the placement of the wells in the water zone.1
As the well is closer to the water zone, high water-cuts are
typical from the beginning of production. When this is the

S. BETANCOURT, K. DAHLBERG, . HOVDE, AND Y. JALALI

case the well will very likely need gas-lift to handle the
required liquid production rates, and even to restart the well
every time it is shut-in. For this application, perforations in
the gas-zone will provide the source of gas, and a surface
controlled valve will be used to control the required amount of
gas entering the tubing, as in a conventional gas-lift
installation. The main advantage of this completion scheme is
to reduce the costs of artificial lift infrastructure, especially for
offshore or remote locations.
A simulation model of a conceptual reservoir representing this
situation was built using a commercial black oil simulator.
Sensitivities on well placement and target rate were made to
observe the impact on oil recovery and the variations in gaslift requirements. This is followed by a discussion on the
selection of the flow control device.

SPE 74391

Well Placement
9.E+06
8.E+06
7.E+06

Np (bbls)

6.E+06
5.E+06
4.E+06
3.E+06
2.E+06
10000

32.5' from WOC

22.5' from WOC

12.5' from WOC

2.5' from WOC

12000

14000

16000

18000

20000

Liquid Production Rate (b/d)

Figure 1. Effect of the standoff on the total recovery after five years of
production.

Well Placement
10000

To study the production behavior of an oil zone overlaid by a

gas-cap and with a bottom aquifer, a reservoir simulation
model was constructed representing a 7 inch diameter, 4000
foot long horizontal well, placed areally in the center of the
reservoir. The reservoir was discretized with a Cartesian grid
divided in 26 gridblocks in the x direction, 15 gridblocks in
the y direction, and 27 gridblocks in the vertical direction.
The dimensions of the model were 5200 feet * 3000 feet. The
gas, oil, and water zones were 140 feet, 40 feet and 90 feet
respectively. An analytical aquifer was considered to represent
the water influx into the reservoir. The initial gas-oil contact
depth was at 5100 ft TVDSS. The oil gravity is 29 API, with
a viscosity of 1.8 cp at initial reservoir conditions (P = 2290
psia and T = 154F). The reservoir permeability is 6.5 darcies
and the anisotropy ratio is 0.5. The well considered was a
subsea well flowing to a 5000-foot subsea flow line and a
1000-foot riser to the production manifold in the platform.

Simulation Results
Sensitivity to Standoff
Although some studies have addressed the possibility of
placing the well inside the water zone and inducing inverse
coning of the oil phase for higher recovery, this was not
considered in this study due to the risks that it involves should
heterogeneities be present. In this study, the gas breakthrough
time and cumulative oil recovery (for a five-year period) were
computed for different well placements in the oil column and
different initial liquid production rates. As may be seen in
figure 1, higher recoveries are achieved when the well is
placed closer to the water-oil contact, and is produced at high
rates. The crossover of the two top curves in figure 1 shows
the interplay between standoff and rate. Figure 2 shows that
gas breakthrough time is noticeably delayed by placing the
well far from the gas-cap.

32.5' from WOC

12.5' from WOC

Gas Breakthrough Time (days)

Model Description

22.5' from WOC

2.5' from WOC

1000

100

10

1
10000

12000

14000

16000

18000

20000

Liquid Production Rate (b/d)

Figure 2. Effect of the standoff on the gas brakthrough time.

Figure 3 shows the production profile for the case when the
standoff to the water-oil contact is 2.5 ft and the initial liquid
production rate is 18000 stb/d. A production strategy is
adopted such that the well is choked at the wellhead when the
gas rate on surface exceeds 10 MMscf/day, due to gas
handling limitations. It was also considered that if the well
reached an oil production rate lower than 2000 b/d, it would
be shut-off.
Well Production Profile
GOR (Mscf/stb),
W-cut (fraction)
5

Oil Rate (stb/day)

20000
18000

Qo

4.5

16000

GOR

14000

Wcut

3.5

12000

10000

2.5

8000

6000

1.5

4000

2000

0.5

0
0

365

730

1095

1460

0
1825

Time (days)

Figure 3. Production profile for the well placed at 2.5 feet from the water-oil
contact. The liquid production rate is initially constant at 18000 b/d. After
the gas breakthrough time the liuid rate is gradually adjusted to satisfy the gas
production constraint of 10 MMscf/d.

SPE 74391

NATURAL GAS-LIFT: THEORY AND PRACTICE

It was observed from the simulations that high water-cuts are

typical from early stages of production as the well gets closer
to the water-oil contact. As will be seen in the next section the
increase in the water-cut drives the requirements for gas-lift
during the first months of production.
Gas-Lift Requirements
An initial simulation was performed to compare the
production performace of the well with and withouth gas-lift
(figure 4). If the well is produced by natural flow, the
production behaves as shown in the lower curve. The well has
to be produced at a much lower initial production rate, then
eventually, as the water-cut increases, the well will stop
flowing naturally. If gas-lift is contemplated from the onset of
production, the well is able to produce at higher rates, and
continue flowing in spite of the increasing water-cut. This
continues until the moment the gas-cap breaks through and the
gas-lift is halted.

Production Performance with and without Gas-Lift

18000

Oil Production (b/d)

16000
14000

Without Lift

12000

With Natural Gas-Lift

10000
8000
6000
4000
2000
0
0

365

730

1095

1460

1825

Time (days)

Figure 4. Comparison of the production behavior of the well with and

without gas-lift into the tubing

The optimum gas-lift rates for the well were computed over
the study time using the reservoir simulator. This was more
efficient than doing separate nodal analysis calculations for
different periods of time. The process consisted in finding the
optimum gas-lift rate at each simulation time step subject to
certain constraints, in this case the desired liquid rate and the
maximum gas production allowed. An important factor to
keep in mind is that the well should be able to deliver the
fluids at the required production manifold pressure of
500 psia.

Gas Injection Requirements

Gas Injection Rate
(Mscf/day)
10000

Oil Rate (stb/day)

20000
18000

9000

Qo

16000
14000

7000

12000

6000

10000

5000

8000

4000

6000

3000

4000

2000

2000

1000

0
0

365

730

1095

1460

1825

Time (days)

Figure 5. Gas-lift requirements to meet the production requirements.

After the breakthrough of the gas into the horizontal section of

the well, successive liquid production rate cutbacks were
applied automatically by the simulator to satisfy the condition
that the gas production rate not exceeds 10 MMscf/day. This
was based on the assumption that the surface facilities could
not handle more than that amount of free gas.
It was suspected initially that at the later stages of production,
when the water-cut has very high values, additional gas-lift
would be necessary. It was observed though that for this
particular case the well could produce until the end of the
study period (five years) without further gas-lift. In general,
as the BHP declines, there will be a moment in which the well
will not be able to meet the required pressure at the manifold.
In this case the injection of gas into the tubing by opening the
valve, would make possible higher liquid production rates at
the late stage, but the compromise exists with the allowed gas
production rate.
The well reached the end of the study with an oil production
slightly above the economic limit of 2000 stb/day.
Water Influx and Gas-cap Size
Effect of the Gascap Size
Np (stb)

GBT(days)

9.0E+06

2000

8.0E+06

1800
1600

7.0E+06

Np (GC=360')
1400
Np (GC=140')
GBT (GC=360') 1200
GBT (GC=140')

6.0E+06
5.0E+06

1000

4.0E+06

The gas-lift rate vs. time is presented in figure 5. It can be

observed that the well requires gas-lift from the beginning of
production until about 350 days. At 345 days the reservoir gas
has broken through the horizontal well and the well GOR
starts increasing sharply. After 390 days the well GOR is
enough to keep the production requirements, and further
injection is not necessary.

8000

GIR

3.0E+06

800

2.0E+06

600

1.0E+06

400

0.0E+00
10000

12000

14000

16000

18000

200
20000

Initial Liquid Rate (b/d)

Figure 6. Effect of the gas-cap size on the total recovery after five years of
production and on the gas breakthrough time.

S. BETANCOURT, K. DAHLBERG, . HOVDE, AND Y. JALALI

A simple sensitivity was made to observe the impact of the

gas-cap and aquifer size on the production performance of the
well. The size of the gas-cap was changed from 140 to 360
feet of thickness and the influence on the total recovery (Np),
and gas breakthrough time (GBT) may be seen in figure 6.
For a given size of the gas-cap, calculations were made for
different sizes of the aquifer. It was observed that as the
aquifer gets stronger, there is a delay in the breakthrough time
of the gas, also the water-cut will increase faster (not shown).
This implies that more gas from the gas-cap will be needed for
injection in the well for a longer period of time.
Other parameters known to influence the coning of the gas and
water phases into the well will have also an impact on the
lifting requirements. These are the absolute permeability, the
reservoir anisotropy, and the contrast in density and viscosity
of the oil and gas phases. The main idea is that higher the
target production rates, the faster the coning of water, and
more critical the flow-lifting requirements in the early stages
of production. When the above mentioned factors delay the
breakthrough time of the gas, the longer will be the period in
which the gas for lifting purposes needs to be provided.

SPE 74391

1%, i.e., the production rate was slightly lower as the valve
was placed higher in the completion, for the same gas-lift rate.
For the conceptual reservoir studied it is necessary to select a
valve with a range of openings that will allow the flow of 3 to
8 MMscf/day of gas. For this type of applications, where the
gas is coming from a contiguous gas zone, as the differential
pressures across the valve increases, a smaller size valve will
be required. A discrete control valve with six positions was
found to give enough flexibility for the operation of the well.

Natural Gas Lift with Continuous Flow Control Valve

Gas Injection
Rate (Mscf/d)
9000

Oil Production
(stb/d)
18000
16000
14000

Qo (Np =8.17MMstb)

12000

GIR

8000
7000
6000

10000

5000

8000

4000

6000

3000

4000

2000

2000

1000

0
0

365

730
1095
Tim e (days)

0
1825

1460

Natural Gas Lift with Discrete Flow Control Valve

Downhole Gas -Lift Valve Design

The final step is the selection of the appropriate device to
control the injection of gas as long as it is required. The gaslift rate, the upstream and the downstream pressure are
determined from the reservoir model. The options available in
flow control systems, are divided basically into two
categories, discrete and continuous control. Continuous
control refers to infinite resolution or infinite position valves
whereas discrete control refers to valves with a finite number
of positions usually 6 or 11. The sizing of the valve or
selection of the range of apertures is therefore critical for
discrete control valves.
Depending on the length of the gas zone, the design engineer
may have flexibility in the positioning of the valve. For a
contiguous natural gas-lift process, it may be desirable to
place the valve as far as possible from the oil zone (in this case
the horizontal section of the well). Doing this, the pressure
drop across the valve will be higher and this facilitates the
control of the gas zone. A differential pressure of at least 35
psi will ensure the gas-lift into the tubing. As in traditional
gas-lift, there is a compromise between the valve depth and
the liquid production rate. Deeper the valve, higher the
production rate for the same gas-lift volume. Therefore as the
valve is farther from the perforations in the oil zone, more gas
volume will be required to satisfy a certain production target.
For this case a sensitivity analysis was made for the valve
position within the completion, placing it at 80, 180 and 280
feet above the GOC. The average P across the valve ranged
from 105 to 175 psi. The impact on production was less than

Gas Injection
Rate (Mscf/d)
9000

Oil Production
(stb/d)
18000
Oil Production
(Np=8.09MMstb)

16000
14000
12000

Gas Injection Rate

8000
7000
6000

10000

5000

8000

4000

6000
4000

3000
2000
1000

2000
0
0

365

730

1095

1460

0
1825

Tim e (days)

Figure 7. Comparison of the production performance with Continuous and

Discrete Flow Control Valves for gas lift.

Figure 7 shows the production performance of the well using

continuous and discrete flow control valves for the injection of
gas from the gas-cap. After five years a slightly higher
recovery could be achieved with a continuous flow control
device (+80 Mstb). A continuous flow control valve could
facilitate the optimization of the production, by producing just
the required amounts of gas from the gas-cap. During some
periods of production a discrete flow control valve would
make the well produce more gas than necessary with a certain
valve position, but the preceding position, will not provide
enough gas to sustain the desired levels of production.
Finally, comparing the total recovery after five years using gas
from the gas-cap for lifting versus using an unconstrained
source of gas, as may be supplied via an electric valve, it was
found that the recovery is not significantly affected. This will
be the case when the two scenarios have the same operation
efficiency. By operation efficiency, we imply the ability to

SPE 74391

NATURAL GAS-LIFT: THEORY AND PRACTICE

maintain an optimum operation of the system over time, and is

therefore related to the operator speed of response to adjust to
changes in the system. For the example presented here, the
dimensions of the gas-cap enable the extraction of the required
amounts of gas without significantly impacting the pressure of
the system. In the last section, when gas-lift from a noncontiguous zone is considered, the effects of the depletion of
the gas zone over time are also taken into consideration.
Operation of the Well
For a discrete flow control valve, an indexing mechanism that
yields increments in the flow area with each change of
position will be convenient to facilitate the operation of the
well. A continuous flow control valve has the flexibility to
change the aperture in any direction, choking or un-choking
as desired.
Pressure sensors to measure the upstream and downstream
pressures will facilitate and enable real time optimization of
the well. Knowing the differential pressure across the valve,
the upstream pressure and the gas and liquid production rates
on surface, the valve position can be adjusted to
optimize production.

Operational Experience
Gas-cap gas-lift is mainly chosen for wells located in a part of
the field where an upwards movement of the oil column is
expected. The gas-lift will then enable a high enough
production rate in order to produce oil by inverse coning.
The main objective with gas-lift completion on Troll is to
ensure production from wells in areas with upwards
movement of the oil column. In addition other benefits can be
seen. Production from wells with high water-cut can be
boosted in the period prior to gas breakthrough. In one case,
the oil rate was boosted by more than 6000 bbl/d initially with
the use of gas-lift. In addition, gas-lift can be used to optimize
oil production from a well cluster in that a poor well can be
boosted, instead of choking good wells back in order to keep
the poor well flowing. Wells with gas-cap gas-lift are restarted
more easily after shut down. Gas-lift is necessary for some
wells at the Troll field at present.

Case Study

Three wells at the Troll B platform are completed with gascap gas-lift. The wells were chosen based on expectations of
early high water-cut, as well as simulated steep rise in watercut. Two of the wells were using gas-cap gas-lift to meet the
target oil rate, and significant increases in daily oil rate were
obtained initially.

Field Description
The Troll Oil field, offshore the western coast of Norway
consists of one oil province in the west with an oil column of
ca 85 ft and a gas province in the east with a much thinner oil
column of 33-43 ft. The reservoir is overlaid by a gas-cap and
is also in contact with a vast bottom drive aquifer. The Troll
Oil development has two oil processing platforms, Troll B and
Troll C. Today, 69 wells are in production, and a total of 108
production wells are planned to be drilled.

Five wells at Troll C are completed with gas-cap gas-lift.

Three of the wells had low water-cuts at startup, and did not
require use of gas-cap gas-lift to meet target oil rate. However,
all three wells show increasing water-cut approaching the
level where gas cap-gas-lift is beneficial after shutdown of the
well. The last two wells had significant increases in oil rate
with the use of gas-cap gas-lift. For the case of the latest well,
an increase of ~3000 bbl/d was obtained using the gas-cap
gas-lift since the well had a very high initial water-cut.

The development strategy of Troll Oil contemplates the

drilling of long horizontal subsea wells, >10 000 feet long,
drilled close to the water-oil contact. Wells are placed in
clusters of two 4-well templates. The wellstreams meet at the
templates and flow to the oil-processing platform via two
gathering lines. Distances between the templates and the
platform vary between 1.9-6.2mi.

Installation procedure of a gas-cap gas-lift valve in a well

includes perforating an interval in the gas-cap and setting
packers that allow the gas to flow through the annulus up to
the gas-lift valve. A hydraulic control line must be installed
down to the gas-lift valve, which is then operated by cycling
the pressure in the control line. One instance of setting the
production packer prematurely has been experienced during
gas-cap gas-lift completion in bad weather conditions. This
required approximately 14 days of downtime to correct. Also
one instance of prematurely opening the fluid loss valve
against the reservoir has been experienced. Gas-cap gas-lift
completion is more complicated than normal completions and
thus does give a higher risk for downtime.

To date, downhole gas-lift valves have been installed in eight

wells to utilize the production of gas from the gas-cap for
lifting purposes. All new wells drilled are prepared for later
recompletion with gas-cap gas-lift.
All new wells on Troll are drilled close to the water-oil contact
(1.6 3.3 ft) and typically produce with a significant
water-cut initially.

Timesaving changes have been made to the gas-cap gas-lift

design at Troll. Gravel packing the perforated interval has
been excluded from the gas-cap gas-lift completion procedure
due to low pressure depletion at Troll. Due to this, the risk for
collapsing the perforations is acceptable without gravel pack.
Gas-cap gas-lift completions require approximately 2 days

S. BETANCOURT, K. DAHLBERG, . HOVDE, AND Y. JALALI

All wells at Troll are subsea wells drilled from moveable

drilling rigs. No well intervention can be done from the oil
processing platform. Well interventions are thus very costly as
it will not only require rig time but also delay drilling of future
wells. Some of the wells currently producing would benefit
from gas-cap gas-lift, but due to the cost of intervention,
recompletion with gas-cap gas-lift is not economical while
drilling production wells.
The gas-cap gas-lift completion is designed to meet a required
gas-lift rate of 3.5-7 Mscft/d. The remotely controlled gas-cap
has the following positions: 0-100-80-60-40-20-0%. Online
rate calculations predict the oil and gas rate based on choke P
and pressure loss in the well. This is found to be very useful in
optimizing the oil rate.

production from the gas zone is shown. Three changes in the

position of the valve were applied at times 2, 6, and 18
months. The final oil recovery was 0.945 MMSTB. In figure
10, six changes in the valve position were simulated at times
2, 4, 12, 15, 21, and 23 months. The final oil recovery was
1.040 MMSTB.
A higher oil recovery using the natural gas-lift approach
may be achieved by optimizing the valve position changes,
and therefore better emulating the base case when an external
gas source is used. For the base case the optimum gas-lift
rates where computed. Since the gas zone is being depleted
with time, valve position adjustments are necessary to
maintain the desired injection rates.
In a more general context, for natural gas-lift from a noncontiguos zone to be feasible, the pressure of the gas zone
must be in hydrostatic equilibrium or higher than the pressure
in the oil zone.

Natural Gas-Lift from a Non-Contiguous

Gas Zone

The production behavior of the oil zone is shown in figure 8.

It was observed that from the beginning of production the well
needs artificial lift to deliver the fluids at the required
wellhead pressure of 150 psi. These requirements were
modeled as an external source of gas that was injected into the
tubing above the perforation of the oil zone. As the reservoir
pressure declines during the first three months of production,
the GOR increases and reaches a maximum value of 2000
scf/stb. When the aquifer perceives the pressure drop in the
reservoir, the aquifer influx begins yielding a stabilization of
the reservoir pressure, and the GOR, and an increase of the
water-cut. Figure 8 shows the production performance during
the first two years of production, water-cut, GOR, and the
required gas-lift rate. The recovery after two years is 1.104
MMSTB. This requires an external source of gas and artificial
gas-lift infrastructure.
In figure 9 the behavior of the well commingling the
production of the two zones and using a valve to control the

6000

2.5

5000

Qo (stb/d)

Qo, GIR

4000

GIR (Mscf/d)
GOR (Mscf/stb)

3000

WOR (frac)

2000

2
1.5
1
0.5

1000
0
0

365

0
730

Time (days)

Figure 8. Production performance of the oil zone over a two-year period.

Artificial lift requirements were identified from the beginning of production.
Non-Contiguous Gas Lift
Qo (stb/d)
GIR (Mscf/d)

7000
6000

Qo, GIR

The following example illustrates the use of a non-neighbor

gas bearing formation for the production of an oil reservoir. A
conceptual reservoir model was built to study the drainage of
an oil zone that requires the assistance of a deeper gas zone to
produce. Both zones are commingled through a vertical well.
The gas zone (96 ft thickness) is deeper, and has a higher
pressure (3500 psi). The oil zone (64 ft thickness) has an
initial pressure of 2800 psi, and a bubble point pressure of
2560 psi. The top of the oil zone is at 10000 ft and the
distance between the two zones is 120 ft. The permeabilities
of the oil and gas zones are 500 md and 350 md respectively
with an anisotropy ratio of 0.1. The oil gravity is 35API and
the gas has a specific gravity of 0.65. The oil zone is
associated with an aquifer, whereas the gas zone relies only on
the expansion of the gas as production drive mechanism.

Non-Contiguous Gas Lift

GOR, WOR

additional operation time. Any additional downtime should be

added to this estimate.

SPE 74391

0.7
0.6

WOR (frac)

5000

0.5

4000

0.4

3000

0.3

2000

0.2

1000

0.1

0
0

365

WOR

0
730

Time (days)

Figure 9. Production performance by commingling the oil and gas zones.

Three changes in the valve position were simulated at times 2, 6, and 18
months.

SPE 74391

NATURAL GAS-LIFT: THEORY AND PRACTICE

the most suitable completion, and the computation of the

optimum amount of gas required for injection. Normally, the
gas-lift requirements for the case of contiguous oil-gas zones
cease soon after the breakthrough of the gas into the well.
Soon after breakthrough occurs the GOR will be high enough
for the well to flow naturally. Eventually as the water-cut
increases, gas production from the gas-cap may become
necessary again.

Non-Contiguous Gas Lift

9000

Qo (stb/d)

8000

GIR (Mscf/d)

7000
6000
Qo, GIR

5000
4000
3000
2000
1000
0
0

365

730

Time (days)

Figure 10. Production behavior by commingling of the oil and gas zones.
Five changes in the valve position were simulated.

Conclusions
Lifting by in-situ gas is a viable alternative for the exploitation
of oil reserves. The use of surface-operated downhole flow
control valves for this purpose enables the operator to regulate
the amount of gas produced and therefore optimize
production, as in any other artificial lift situation. The main
advantage of a Natural Gas-Lift process is the reduction in
costs in artificial lift infrastructure, especially for offshore or
remote locations.
For the case of gas prodution from a contiguous gas zone, the
lifting requirements are strongly dependent on the
breakthrough of the gas-cap gas into the well. A simulation
model of the well and reservoir permits the identification of

For the case of gas production from a non-contiguous gas

zone, natural gas lift is feasible provided that the pressure in
the gas zone is in hydrostatic equilibrium or higher than the
pressure in the oil zone.

Acknowledgements
The authors would like to thank the Troll unit members (A/S
Norske Shell, TotalFinaElf Petroleum, Conoco Norge,
Statoil), Norsk Hydro and Schlumberger for the permission to
publish this paper.

References
1.

2.
3.

Haug, B., Ferguson, W., Kydland, T., Horizontal Wells in the Water
Zone: The Most Effective Way of Tapping Oil from Thin Oil Zones?,
paper SPE 22929, presented at the 66th ATCE, Dallas, TX, October
1991.
North Sea Fields Atlas, Energy Resources Consultants, UK 1996.
Sinha, S., Kumar, R., Vega, L., Jalali, Y., Flow Equilibration Toward
Horizontal Wells Using Downhole Valves, Paper SPE 68635, presented
at the Asia Pacific Oil and Gas Conference and Exhibition, 17-19 April
2001, Jakarta, Indonesia.