You are on page 1of 10

SPE 75720

Severity of Water Coning in Gas Wells


M. Armenta, SPE, Louisiana State University; A. Wojtanowicz, SPE, Louisiana State University

Copyright 2002, Society of Petroleum Engineers Inc.


Trimble and DeRose2 supported Muskat theory with water
This paper was prepared for presentation at the SPE Gas Technology Symposium held in coning data and simulation for Todhunters Lake Gas field.
Calgary, Alberta, Canada, 30 April–2 May 2002.
They calculated water-free production rate using Muskat-
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
Wyckof3 model for oil wells in conjunction with the graph
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to presented by Arthurs4 for coning in homogeneous oil sand.
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 The results were comparing with a field study with a
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
commercial numerical simulator showing that the rates
for commercial purposes without the written consent of the Society of Petroleum Engineers is calculated with Muskat-Wyckof3 theory were 0.7 to 0.8 those
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 of the coning model for a 1-year period.
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.
Kabir5 used the analogy between high oil mobility well and a
typical gas well, to investigate gas well performance coning
water in bottom-water drive reservoir. He built a numerical
Abstract simulator model for a gas-water system. He concluded that
Theoretical study, reported in this paper, qualifies unique permeability and pay thickness are the most important
mechanisms of water coning in gas wells. Water coning in gas variables governing coning phenomenon. Other variables such
wells has been understood as a phenomenon similar to that in as penetration ratio, horizontal to vertical permeability, well
the oil wells. It is shown, however that both the water inflow spacing, producing rate, and the impermeable shale barrier
mechanism and its impact on well’s productivity are have very little influence on both the water-gas ratio response
substantially different. It is shown, for example, that, after and the ultimate recovery
water breakthrough, the oil-water interface at the well’s McMullan and Bassioni(6) believed that water coning behaves
completion would continue to cone, while the gas-water differently in gas wells than in the oil wells. Using a
interface reverses at the top of the cone. commercial numerical simulator they got similar results than
Analyzed in the paper are the results of a conventional Kabir(5) for the insensitivity of ultimate gas recovery with
simulation of water coning in gas wells showing that water variation of perforated interval and production rate. They
could affect productivity only at the very late stage of well’s demonstrated that a well in the bottom water-drive gas
life. However, field data, shown in the paper, evidence early reservoir would produce with small water-gas ratio until
and severe water problems. This contradiction is explained in nearly its entire completion interval is surrender by water.
the paper by including the effects of Non-Darcy flow, In this study, water problems begin when recovery factor is
perforation density and the ratio of vertical-to-horizontal less than 30%. Fig. 1 shows water-gas ratio and gas recovery
permeability in modeling of water coning in gas wells. Results factor from field data of a gas well. It shows water production
from numerical simulation combined with analytical models started after 404 days when the recovery was 22%. This well
show that an early water breakthrough and a considerable was killed for water production after 600 days of gas
increase in water production may result from combined effects production when the recovery factor was 28%. Fig. 2 shows
of increased vertical permeability, lower density of perforation gas and water production rate versus time for another gas well.
and high-velocity gas flow around the wells. It shows water production star after 119 days of production.
Gas rate was reduced from 6.0 MMSCFD to 4.0 MMSCFD
Introduction due to water production rate of 30 BPPD. These two field data
Water coning in gas well has been understood as a shows early water production in gas wells.
phenomenon similar to that in oil well. In contrast to oil wells,
relatively few studies has been reported an aspect of Analytical Comparison of Water Coning in Oil and
mechanisms of water coning in gas wells. Gas Wells Before Water Breakthrough
Muskat1 believed that physical mechanism of water coning in Two petroleum systems, oil and gas, in vertical
gas wells is identical to that for oil wells; moreover, he said equilibrium with bottom water are considered to compare
that water coning would be less serious difficulties for wells water coning in oil and gas wells before breakthrough. The
producing from gas zone than for wells producing oil. two systems have the same reservoir properties, and thickness,
2 M. ARMENTA, A. WOJTANOWICZ SPE 75720

and are perforated at the top of the producing zone. Fig. 3 the completion becomes “flooded” into water, which results in
shows the reservoir system including the properties rapid increase of water-gas ratio.
and dimensions.
A pressure drawdown needed to generate the same static water Discussion of Gas-Water and Oil-Water System’s
cone and the fluid rate for each System was calculated. For the Sensitivity to Water Inflow
system of oil and water and a cone height of 20 ft, we need a We compare water sensitivity of the two systems for two
pressure drawdown equal to 2 psi, and the oil production rate different scenarios, reservoir pressure exceeding normal
6.7 STB/D. In case of gas-water system, for the same 20 ft pressure, and smaller than the normal pressure.
height of water cone we need 8 psi pressure drop, and the gas When reservoir pressure exceeds hydrostatic pressure of
production rate of 1.25 MMSCFD. water, the oil-water system, would produce both fluids to the
From this first simple analysis we conclude as follows: surface without artificial lift. The amount of each fluid that
• It is possible to have a stable water cone of any given can be produced would depend on the well geometry (depth,
height in the two systems (oil-water and gas-water). tubing’s ID), operational parameter (tubing head pressure,
• For the same cone height in vertical equilibrium, pressure frictional pressure lost), and the fluids properties. Also, the
drop in the gas-water system is four times grater than the gas-water system would naturally produce both fluids to the
pressure drop in the oil-water system. surface. Even if the well is “loaded” in the water. However, at
• There is a big difference in the fluids production rate for high reservoir pressure, the gas rate is grater than critical rate,
gas-water and oil-water system, for the same water cone the well is unloaded from water. Thus, gas wells are less
height. It is economically possible to produce gas-water sensitive than oil well to water inflow when the reservoir
systems at the gas rate below critical. However, in most cases pressure is higher than the normal pressure.
it is not economically possible to produce oil-water systems When reservoir pressure drop below normal pressure, the oil-
without water breakthrough. water system would tolerate would tolerate water cut only
until the bottom hole flowing pressure reaches the reservoir
Analytical Comparison of Water Coning in Oil and pressure. Then the oil well would need artificial lift. Thus oil
Gas Wells After Water Breakthrough water system is capable of producing with high water cut
The objective is to compare the shape of oil-water and using artificial lift. For the gas-water system, however, it is not
gas-water interfaces at the wellbore after water breakthrough. possible to produce any more gas when bottom hole flowing
After the water breakthrough, there is a stratified inflow of oil pressure equals the reservoir pressure. The well must be
or gas with the water coveting the bottom section of the well unloaded to resume gas production. Also the gas rate is
completion. Again, two systems having the same reservoir smaller than the critical rate.
properties, and thickness, are compared – oil-water and gas- In short, under the same conditions, less amount of water is
water. Both systems are totally penetrated. An equation needed in the oil system to block the oil’s natural flow to the
describing interface shape was derived using the assumptions surface. However, production may continue using artificial
of Muskat(1). Appendix-A gives the derivation and lift. On the other hand, when water stop the natural flow of gas
mathematical computations. In reality the resulting equations to the surface, almost always it is not possible to produce
will not describe perfectly the inflow at the well. However, more gas.
they are useful to compare the coning phenomenon in oil- From this analysis we can conclude that gas wells are less
water and gas-water system. sensitive than oil wells to water when reservoir pressure is
Fig. A-2 shows the resulting profiles of the fluid interface in higher than the normal pressure. Moreover, gas wells are more
gas-water and oil-water systems. After breakthrough, the oil- sensitive than oil wells to water when reservoir pressure is
water interface at the well’s completion is horizontal, while lower than the normal pressure.
the gas-water interface tends to cone into the water. This observation has an important practical implication since
From comparison of water coning in gas-water and oil-water gas reservoirs are produced best at lowered reservoir pressure
systems, we can conclude that in gas wells, water cone is by coproducing water [Arco and Bassiouni(7)]. However, at
generated in the same way as in the oil-water system. When low-pressure even small water inflow would kill the gas well.
the water comes to the bottom of the well completion, a small
inverse gas cone is generated locally around the completion. Specific Mechanisms of Water Coning in Gas Wells
This inverse cone restricts water inflow to the completions. We selected mechanism –specific for gas wells– that
Also, the inverse gas cone inhibits upward progress of the might potentially enhance early water production in these
water cone. In the result, after the water cone breaks to the gas wells. The mechanisms investigated are vertical permeability,
completion, it cannot “take over” the completion so it remains Non-Darcy flow effect, and density of perforations.
at the completion’s bottom providing small water inflow. Effect of Vertical Permeability. It is postulated here that
Thus, the completion remains open to the gas inflow for most high vertical permeability should generate early water
of the production time. Eventually, the water cone gains so production in gas reservoir with bottom water-drive. Vertical
much body (most due advancement of gas-water contact) that permeability accelerates water coning because high vertical
permeability would reduce the time needed for a water cone
to stabilize.
SPE 75720 SEVERITY OF WATER CONING IN GAS WELLS 3

A commercial numerical simulator, shown in Fig. 4, was Skin factor representing perforation density(11):
adopted to evaluate the effect of vertical permeability in gas  hg  k k 
wells. Two values of vertical permeability, 10 md and the S dp =  [ ]
 ln(rdp / r p )  g − g  …….………………(8)
 Lpnp   k dp k d 
other one 50 md, were considered. The results are shown in    
Figs. 5 and 6. Fig. 5 depicts water saturation in the reservoir
after 760 days of gas productions for vertical permeability 10 Skin factor due to partial penetration (12):
md. The initial water-gas contact was at 5100 ft. The top of  hg    hg k H  
the cone is at 5060 ft after 760 days of production. The water S pp =  − 1 ln  − 2 ………………………..(9)
 h per    rw kV  
cone is still below the completion and there is no water     
production. Fig. 6 shows water saturation in the reservoir after
760 days of gas productions for vertical permeability 50 md Non-Darcy skin around the well(9):
and the same depth of initial water-gas contact, 5100 ft. After
2.22 *10 −15 γ g k g β r
760 days of production, the top of the cone is at 5025 ft. The Dr = ………………..……..(10)
water cone has reached the completion, resulting in water µ g h g rw
production inflow to the well. 2.33 *1010
Non-Darcy Flow Effects on Water Coning. Non-Darcy βr = …………………….……..…...(11)
flow generates an extra pressure drop around the well bore k 1g.2
that could intensify water coning. Non-Darcy flow happens at Non-Darcy skin in the crashed rock around the perforation
high flow velocity, which is a characteristic of gas converging tunnels(11):
near the well perforations.  β dp  k g h g γ g 
The extra pressure drop is a kinetic energy component in the D p = 2.22 *10 −15  2 2   ……………..…...(12)
Forchheimer’s formula(8),  n p L p r p  µ g 
  
dp 2.6 *1010
− = βρv 2 ……………..………….……………………..(1) β dp = ……………………..…………(13)
dL k dp
The Non-Darcy effect was studied analytically for two cases
of well completion, complete penetration of the gas and water
zones, and penetration of the gas zone. In the second case the The results of the study are shown in the Figs. 8-10. Fig. 8
well perforated in only the gas zone. Fig. 7 illustrates the demonstrates the “delayed” effect of water in a gas well
completion schematic and the production system properties. completed in gas and water zone (a worst possible
The analytical model of the well inflow comprises the completion). Not only the problem occurs after 80% of gas
following components: recovered but also WGR, is independent of pressure
drawdown and production rates.
Fig. 9 indicates that combined effects of skin and Non-Darcy
Gas inflow model: flow would strong stimulate water production in gas wells.
Also, WGR increases with increasing pressure drop across the
1.422Tµ g Zq g
Pe2 − Pw2 =
k g hg
[ln(r / r ) + S + Dq ]
e w g
(9)
……….....(2)
skin.
Figs. 10 shows WGR histories for a gas well penetrating only
the gas column. Reducing well completion to the gas column
Where: S = S d + S dp + S pp …………………………….….(3) does not change WGR development; the WGR history is
and D = D r + D p ………………………………………….(4) almost identical to that for complete penetration. Interestingly,
although the completion bottom is at gas-water contact the
Water inflow model: production is practically water-free for almost half of the
recovery. This finding is in agreement with the analytical
analysis of gas-water interface and the inverse internal cone
0.00708k w hw ( Pe − Pw ) (9)
qw = …………………………….(5) mechanism presented in previous sections.
µ w Bw [ln(re / rw ) + S ] From this study we conclude as follow:
Where: S = S d + S dp + S pp …………………..……………(6) • Non-Darcy and distributed mechanical skin increase
water gas ratio (WGR) by reducing gas production rate, and
Skin factor representing mud filtrate invasion (10): increasing water inflow, and the two effects accelerate water
breakthrough to gas well.
hg  (rd − rw )  k g 
Sd = 1 − 0.2  − 1 ln(rd / rw ) …………...(7) • It does not make much difference how much of the well
h per  h per  k d  completion is covered by water as long as the completion is in
contact with water.
The above observations regarding distribute skin and Non-
Darcy effects have been based on a simple analytical
modeling. The results are partially verified with a commercial
4 M. ARMENTA, A. WOJTANOWICZ SPE 75720

numerical simulator for the well-reservoir model shown production and is 8 feet taller than the one with no skin effect.
in Fig. 4. From the Non-Darcy flow's study we can conclude as follow.
Unfortunately the commercial simulator used for this study
does not simulate the distributed skin and wall Non-Darcy Effect of Perforations on Water Production. The next step
flow effects around the well. Instead, it applies these effects for this study was to investigate how density of perforation
merely at the wellbore wall. In the results, simulated values of affects water production in gas well. Specifically, we studied
pressure in the reservoir at the bottom of well completion behavior of the water-gas ratio. Perforations concentrate gas
increases with increasing input value of skin (Non-Darcy plus inflow around the well, increase flow velocity and further
mechanical). This leads to incorrect physical behavior in amplify the effect of Non-Darcy flow. The effect is examined
where water coning reduces with increasing skin here using the modified analytical model from the previous
pressure drop. section (Figs. 7). We used similar calculation procedure
We demonstrate the correct physical behavior analytically. An including skin and Non-Darcy effect, and two values of
additional pressure drop around the well caused by Non-Darcy density, 4 shoots per foot to 12 shoots per foot. The results are
flow is modeled by modifying the Forchheimer equation. The shown in Fig. 14.
solution to Forchheimer equation for steady state, radial flow, There is a 40 percent reduction in water-gas ratio resulting
isotropic formation, constant density and viscosity is(13): from a three-fold increase in perforation density. Shows in Fig
14 is the effect of decreased pressure drawdown that
3.161 * 10−12 βγ g zTq g2  1 1 
1424µzTq g ln(r1 / r2 ) significantly reduces WGR. Thus, well perforations enhance
p12 − p22 = +  −  water production due Non-Darcy flow effect; the smaller the
hk h2  r2 r1  perforation density the higher the water-gas ratio.
.…………………………………………………..(14)
CONCLUSIONS
The model is modified replacing h by hper in the second right 1. Water coning in gas wells is physically different than oil
term of Eq. 14 to include the Non-Darcy due to limited wells. Water cone does not tend “to flood” the well
entry(14), (15) (We have not found any information on how to completion thus leaving larger section of well completion
include Non-Darcy effect due to crashed zone around the open to gas inflow.
perforation tunnels). 2. Gas wells are more sensitive to water production than oil
Using the modified model, we assumed gas rate equal to 20 wells when the reservoir pressure is lower than the normal
MMscfpd, and three cases of skin effect: no Non-Darcy effect, pressure; at high reservoir pressure, as more gas recovered by
including Non-Darcy effect without limited entry effect, and lowering reservoir pressure using technique of water co-
the combined effects of Non-Darcy flow and limited entry. production, or “out-running” the aquifer there is a need
The results are shown in Fig. 11. for water inflow control at the well to maintain
Analysis of the plots in Fig. 1 indicates that when Non-Darcy well’s productivity.
flow occurs around the well, pressure drawdown will increases 3. Distributed skin effect around the well comprising
in the well’s vicinity (20-30 ft). Al semi-log pressure permeability damage, Non-Darcy flow effect, and density of
distribution plots has two different trends, one away from the perforation promotes water coning in gas wells resulting in
well where Darcy flow is in control (m2), and another near the early breakthrough and elevated values of water-gas ratio.
well bore where Non-Darcy flow is predominant (m1). For this
example case, the values of m2 and m1 are: m1= 159, and m2= Nomenclature
19. The result demonstrates dramatic effect at Non-Darcy flow Bw = water formation volume factor, reservoir
and its potential to enhance water coning. barrels per surface barrels.
The effect of distributed skin is also modeling by modifying D = Non-Darcy flow coefficient, day/MSCF
the input data to the commercial numerical simulator through dp = pressure derivative, psia
permeability reduction zone around the well completion. Size dL = length derivative, feet
of the zone and its permeability is estimated using results in hg = thickness of gas, feet
Fig. 11. We assume a size of 30 ft. Also, using the two slopes hpre = perforated interval, feet
(m1, and m2) we compute permeability as, hw = thickness of water, feet
m   19  k = peremeability, millidarcies
k1 =  2  * k 2 =   *100 = 11.9md ≅ 10md kd = altered reservoir peremeability, millidarcies
 m1   159 
kdp = crashed zone peremeability, millidarcies
• Fig. 12 shows water saturation in the reservoir after 1124 kH = horizontal peremeability, millidarcies
days of gas productions without the effect of distributed skin. kg = gas peremeability, millidarcies
A 32 feet water cone developed from the initial water-gas kV = vertical peremeability, millidarcies
contact at 5100 ft, to 5068 ft after three years of gas kw = water peremeability, millidarcies
production. Fig. 13 shows water saturation in the reservoir L = length, feet
after 1124 days of gas productions whit inclusion of the skin. Lp = length of perforation, feet
The water cone extends up to 5054 ft after 1124 days of np = number of perforations
SPE 75720 SEVERITY OF WATER CONING IN GAS WELLS 5

p = pressure, psia 9. Beggs H.D.: Gas Production Operation, Oil and Gas
Pe = reservoir pressure at the boundary, psia. Consultants International, Inc, Tulsa, Oklahoma, (1984).
Pw = flowing bottom hole pressure, psia. 10. Jones, L.G. and Watts, J.W.: “Estimating Skin Effects in a
Qg = gas flow rate, MSCF/day Partially Completed Damaged Well,” JPT (Feb. 1971) 249-52;
Trans., AIME, 251.
Qw = water flow rate, barrel/day 11. McLeod, H.D. Jr.: “The Effect of Perforating Conditions on
qg = gas flow rate, MSCF/day Well Performance,” JPT (Jan. 1983) 31-39.
qw = water flow rate, barrel/day 12. Saidikowski, R.M.: “Numerical Simulations of the Combined
rd = altered reservoir radius, feet Effects of Wellbore Damage and Partial Penetration,” paper SPE
rdp = crashed zone radius, feet 8204 presented at the 1979 Annual Fall Technical Conference
re = outer radius, feet and Exhibition, Nevada, Sep 23-26.
rp = radius of perforation, feet 13. Katz D.: Handbook of Natural Gas Engineering, McGraw-Hill
rw = wellbore radius, feet Book Company, New York, (1959).
S = skin factor 14. Dake L.P.: Fundamentals of Reservoir Engineering, Elsevier
Scientific Publishing Company, New York, (1978).
Sd = skin factor representing mud filtrate invasion 15. Golan M. and Whitson C.: Well Performance, Prentice-Hall
Sdp = skin factor representing perforation density Inc., New Jersey (1991).
Spp = skin factor due to partial penetration 16. Craft B.C. and Hawkins M.F.: Applied Petroleum Reservoir
T = temperature, oR Engineering, Prentice Hall PTR, Englewood Cliffs, NJ (1991).
v = velocity, ft per second
y = gas-water or oil- water interface thickness, feet
ye = water thickness at the boundary, ft
Z = gas deviation factor
β = turbulent factor, 1/ feet APPENDIX - A
βr = turbulent factor for reservoir, 1/ feet Analytical development for comparison of water coning in
oil and gas wells after water breakthrough
βdp = turbulent factor for crashed zone, 1/ feet
ρ = density, lbm/ft3
wel
∂p = pressure derivative, psia
∂r = radius derivative, feet
φ = porosity
γg = specific gravity of gas
µ = viscosity, centipoises
µg = viscosity of gas, centipoises
oil / gas
µw = viscosity of water, centipoises
h Pw
Pe
References
1. Muskat, M.: Flow of Homogeneous Fluids, International Human

2.
Resources Development, Boston (1982).
Timble, A.E., DeRose, W.E.: “Field Application of Water-
ye y=? water
Conning Theory to Todhunters Lake Gas Field,” paper SPE
5873 presented at the 1976 SPE-AIME 46th Annual California r
Regional Meeting, Long Beach, April 8-9. r
3. Muskat, M. and Wyckoff, R.D.: “An approximate Theory of
Water Coning in Oil Production,” Trans., AIME (1935) 114, Fig A-1 - Theoretical model used to compare water coning in oil
144-163. and gas wells after breakthrough.
4. Arthur, M.G.: “Fingering and Coning of Water and Gas in
Homogeneus Oil Sand,” AIME (1944) 155, 184-199. Assumptions: radial flow, isothermal conditions, porosity and
5. Kabir, C.S.: “Predicting Gas Well Performance Coning Water in permeability are the same in the gas and water zone, and
Bottom-Water Drive Reservoirs,” paper SPE 12068 presented at steady state conditions.
the 1983 SPE 58th Annual Technical Conference and Exhibition,
San Francisco, Oct 5-8.
6. McMullan, J.H., Bassiouni, Z.: “Optimization of Gas-Well
For gas-water system:
Completion and Production Practices,” paper SPE 58983 2πr (h − y )kφp ∂p
Qg = ……………………………….(A-1)
presented at the 2000 SPE International Petroleum Conference µg ∂r
and Exhibition, Mexico, Feb 1-3.
7. Arco, D. P. and Bassiouni, Z. A.: “The Technical And Economic
Feasibility of Enhanced Gas Recovery in The Eugene Island 2πrykφ ∂p
Qw = ………………………………………(A-2)
Field by Use a Coproduction Technique,” JPT (May 1987). µ w ∂r
8. Lee J. and Wattenbarger R.: Gas Reservoir Engineering, Society
of Petroleum Engineering, Richardson, Texas, (1996).
6 M. ARMENTA, A. WOJTANOWICZ SPE 75720

Qg µ w (h − y ) p From equation (A-4):


R= = ……………………………….(A-3) Q g µ g 1 − ( y e / h)
Qw µg y a
= pe = ………………………(A-16)
Qw µ w ( y e / h) b
µ w Pe (1 − ( y e / h))
Qg The ratio (1/b) could be found from equation (A-14) at the
At re ⇒ R= = …………...(A-4)
Qw µ g ( y e / h) well bore (r = rw ⇒ p = pw):
r 
ln e 
Rearranging equation (A-1): 1  rw 
Qg µ g ∂p ∂p = ……………(A-17)
= rhp − ryp ………………………….…(A-6) b   a   p e + (a / b)  
2πφk ∂r ∂r ( p e − p w ) −   ln  
  b   p w + (a / b)  
Rearranging equation (A-2):
Qw µ w ∂p Finally, y may be solve from equation (A-3) and (A-15):
= ry ………………………………..………(A-7) Q g µ g (h − y ) p a hp
2πφk ∂r = = ⇒ y= ………..(A-18)
Qw µ w y b [(a / b) + p]
Substituting (A-7) in (A-6):
Qg µ g ∂p Q w µ w
= rhp − p ……………………….…….(A-8) Repeating the same analysis for oil-water system:
2πφk ∂r 2πφk
2πr (h − y )kφ ∂p
Qg µ g Qo = ……………………………..…..(A-19)
Let us define some constants: a= …….….(A-9) µo ∂r
2πφkh
Qw µ w 2πrykφ ∂p
b= ……….(A-10) Qw = …………………………………..…..(A-20)
2πφkh µ w ∂r

Substituting (A-9) and (A-10) in (A-8): Qo µ w (h − y ) p


∂p R= = ………………………………...(A-21)
a = rp − bp ………………………………..………(A-11) Qw µo y
∂r
Qo µ o Q µ
Rearranging equation (A-11) gives: If: a = , and b = w w , then
2πφkh 2πφkh
1
 p
b ∂r
∂p = ………………………….……..(A-12) Qo µ o ∂p ∂p
 a   r = rh − ry ………………………..……..(A-22)
  + p  2πφk ∂r ∂r
 b  
Integrating (A-22):
 a  re ∂r 1 pe
Integrating (A-12):  + 1 ∫ = ∫ ∂p …………………………….(A-23)
pe r b  r r b p
p e
∂r
(1 / b) ∫ ∂p = ∫ ………………..………..(A-13)
[(a / b) + p] r r 1
 ( p e − p )
p

r  b
The solution for (A-13) is: ln e  = ………………………..……..(A-24)
r  a 
r 1 a  p + a / b   + 1
ln e =  p e − p − ln e  ……..……….. (A-14) b 
r b  b  p + a / b 
h
Solving for y: y = …………………………..(A-25)
The ratio (a/b) may be found dividing equations (A-9) by a 
equation (A-10):  + 1
b 
a Qg µ g
= …………………………………………(A-15)
b Qw µ w
SPE 75720 SEVERITY OF WATER CONING IN GAS WELLS 7

In order to demonstrate the model for gas-water and oil-water  6000 


system describing the interface between the fluids, let us solve ln 
1  4 
the following example. = = 0.031
b   2000 + 500  
(2000 − 1700) − (500) ln  
The system data is as follow:   1700 + 500  
pe = 2000 psi re = 2000 ft rw = 0.4 ft
h = 50 ft ye = 40 ft µw = 0.498 cp 7. Assuming a pressure value between pe and pw, we
Bw = 1.0 k = 100 md φ = 0.25 calculate r and the gas-water profile y, using Eqs. A-14
µg = 0.017 cp and A-18 respectively. This is the gas-water interface
profile:
The procedure is as follow:
r 1 a  p + a / b 
1. Assuming a value for the pressure drawdown (300 psi). ln e =  p e − p − ln e 
r b  b  p + a / b 
2. Calculating the flowing bottom hole pressure
( p w = p e − ∆p = 2000 − 300 = 1700) ; we assume that   2000 + 500 
6000
pw is constant along the wellbore. ln = 0.0312000 − p − 500 ln 
r   p + 500 

3. Computing the water flow rate (Qw) using Darcy’s law hp 50 * p


y= ⇒ y=
equation (16): [(a / b) + p] [500 + p]
0.00708khw ( pe − pw ) (Note that this pressure distribution does not depend on values of
Qw =
µ w Bw ln(re / rw ) flow rate but only on their ratio)
0.00708 * 100 * 40 * (2000 − 1700)
Qw = = 2000 By comparison, from equation A-25, we see that for oil-water
0.498 * 1.0 * ln(2000 / 0.4)
system y remains constant, y = 40, and independent from radius.
Fig. A-3 shows the resulting profiles of the fluid interface in gas-
4. Calculating a/b, using Eq.A-16: water and oil-water systems.

a 1 − ( y e / h)
= pe =
[1 − (40 / 50)]* 2000 = 500
b ( y e / h) (40 / 50)
fluids interface height (y) vs radii (r)
Which is constant for the system and independent for
the gas rate. 41
fluids interface height (ft)

5. Finding Qg, from Eq. A-15 in Appendix 1-A: 40

a Qg µ g Q g * 0.017 * 5.615
= ⇒ 500 = 39

b Qw µ w 2000 * 0.498
Qg = 5.22 MMscf/d 38

Note that WGR is constant for the system and depends 37

only on the system geometry (ye, h) and pressure drive 0 500 1000 1500 2000 2500

Radius (ft)
(pe).
gas-water contact oil-water contact
6. Computing 1/b, using Eq. A-17 :
Fig. A-2 - Shape of the gas-water and oil-water contact for total
r  perforation
ln e 
1  rw 
=
b   a   p e + (a / b)  
( p e − p w ) −   ln  
  b   p w + (a / b)  
8 M. ARMENTA, A. WOJTANOWICZ SPE 75720

FIGURES

9,000 140
400 30.0

8,000
350 120
Water-Gas Ratio (STB/MMSCF)

Water Production Rate (BPD)


25.0

Gas Production (MSCFPD)


7,000
300
100

Recovery Factor (%)


20.0 6,000
250
5,000 80
200 15.0
4,000 60
150
10.0 3,000
40
100
2,000
5.0
50 20
1,000

0 0.0
0 0
314 345 376 404 435 465 495 521 551 573 600
2 29 60 91 119 149 177 197 223 254
Time (days)
Time (Days)

Water-Gas Ratio Gas Recovery Factor


Gas Prodcution Rate (MSCFPD) Water Production Rate (BPD)

Fig. 1 – Water gas ratio and gas recovery factor for well CC-1
(field data) Fig. 2 – Gas and Water production rate for well CC-2 (field data)
well Well

φ= 25% Swir= 30% 2500 ft


Sgr= 20% S.G.gas=0.6
rw= 0.5 ft Pinitial= 2500 psia kr= 100 md

100 of 1 ft 50 ft
K= 100 md 20 ft layers 100 ft
φ=0.2 Oil Gas
P=2000 psi 50 ft
µ= 1.0 cp 0.017 cp Gas
T= 112 oF
ρ= 0.8 gr/cc 0.1 gr/cc

µw=0.56 cp water
9 of 10 ft, and
one 110 ft 200 ft
ρw= 1.02 gr/cc layers
Water
5000 ft
re= 1000 ft

Fig. 3 - Theoretical model used to compare water coning in oil and


gas wells before breakthrough. Fig. 4 – Base numerical model.

Irreducible Water Irreducible Water


Saturation Saturation

Swept Zone
Swept Zone

100% Water Saturation


100% Water Saturation

Fig. 5 - Distribution of water saturation after 760 days of gas


production. The well has produced at maximum gas rate. Vertical Fig. 6 - Distribution of water saturation after 760 days of gas
permeability is 10 md. production. The well has produced at maximum gas rate. Vertical
permeability is 50 md.
SPE 75720 SEVERITY OF WATER CONING IN GAS WELLS 9

2,100
rw= 0.5 ft

Water-Gas ratio (bbl/MMscfpd)


1,800

1,500

K= 100 md
P=2500 psi 1,200
o
T= 120 F Gas
40 ft
Kh / Kv = 10 900

600

300
µw=0.56 cp
ρw= 1.02 gr/cc
K= 100 md 0
Bw= 1 0 rb/STB 40 ft
Water 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Recovery Factor (%)

Pressure Drawdown = 100 psi Pressure Drawdown = 300 psi


Pressure Drawdown = 500 psi Pressure Drawdown = 1000 psi
re= 2500 ft Presure Drawdown= 1500 psi

Fig. 7 – Analytical model used to investigate the effect of Non- Fig. 8 – Water-Gas ratio versus gas recovery factor for total
Darcy in water coning. penetration of gas and water columns and without skin and Non-
Darcy effect.

3,000

2,700

Water-Gas Ratio (BLS/MMSCF)


2,400
2,100 2,100
Water-Gas Ratio (BLS/MMSCF)

1,800 1,800

1,500
1,500
1,200
1,200 900

900 600

300
600
0
300 0 0.2 0.4 0.6 0.8 1

Recovery Factor (%)


0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Pressure Drawdown = 100 psi Pressure Drawdown = 300 psi Pressure Drawdown = 500 psi
Pressure Drawdown = 100 psi
Recovery Factor (%)
Pressure Drawdown = 300 psi
Pressure Drawdown = 1000 psi Pressure Drawdown = 1500 psi
Pressure Drawdown = 500 psi Pressure Drawdown = 1000 psi
Pressure Drawdown = 1500 psi

Fig. 9 – Water-Gas ratio versus gas recovery factor for total Fig. 10 – Water-Gas ratio versus gas recovery factor for well
penetration of gas and water columns skin and Non-Darcy effect completed only through total perforation the gas column with
included. combined effects of skin and Non-Darcy.

900

850 m2

800
Pressure (psia)

750 Q= 20 M M SCFPD

700
Irreducible Water
650 Saturation

600 m1
550

500
0.1 1 10 100 1000 10000
Swept Zone
Radial Distance(ft)

DarcyFlow Non-Darcyflow ModifiedForchheiner

Fig. 11 – Pressure distribution in the reservoir using Modified Fig. 12 – Distribution of water saturation after 1124.4 days of gas
Forchheimer model of Non-Darcy flow. production; Qg = 13 MMscfpd. Low-pressure (Non-Darcy) zone
near the well bore is not included.
10 M. ARMENTA, A. WOJTANOWICZ SPE 75720

1,600

Water-Gas Ratio (BLS/MMSCF)


1,400

1,200

1,000

800

Irreducible Water 600


Saturation
400

200

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Gas Recovery Factor (%)


Swept Zone
Pressure Drawdown= 100 psi (4 spf) Pressure Drawdown = 500 psi (4 spf)
Pressure Drawdown = 1000 psi (4 spf) Pressure Drawdown = 100 psi (12 spf)
Pressure Drawdown = 500 psi (12 spf) Pressure Drawdown = 1000 psi (12 spf)

Fig. 13 – Distribution of water saturation after 1124.4 days of gas Fig. 14 – Effect of perforation density on water-gas ratio for a well
production; Qg = 13 MMscfpd. Low-pressure (Non-Darcy Effect) perforating in the gas column skin and Non-Darcy effect included.
zone near the well bore is included.

You might also like