RESERVOIR
ENGINEERING
CHI U. IKOKU
The Pennsylvania State University
FEDERICO G. DIAZ MCRALES
INGENIERO DE PETl{()LEO
Reg. del Colegio de Ingenieros No. 48421
KRIEGER PUBLISHING COMPANY
MALABAR, FLORIDA
1992
Original Edition 1984
Reprint Edition 1992
Printed and PobJished by
KRIEGER PUBLISHING COMPANY
KRIEGER DRIVE
MALABAR, FLORIDA 32950
Copyright 1984 by John Wiley and Sons, Inc ..
Reprinted by Arrangement.
All rights reserved. No part of this book may be reproduced in any form or by any
means, electronic or mechanical. including :informationstorage and retrieval systems
without permission in writing ;, '. ',' _ 
No liability is assumed wiih respect" rhi!, use of 'information contained herein.
Printed in the United States of Atrterica. ., ' ,
Library of Congress Catalog1ngInPobllcatlon Data
Ikoku, Cbi U.
Natural gas reservoir engineering I Chi U. Ikoku.
p. em.
Reprint. Originally published: New York: Wiley, c1984.
Includes bibliographical references and index.
ISBN 0894646400 (alk. paper)
1. Gas reservoirs. I. Title.
1N880.I335 1991
622' .3385dc20
10 9 8 7 6 5 4 3 2
9121671
CD'
J
\
To Chine\o,
whose patience, understanding, and
encouragement were essential to the
completion of this book
PREFACE
This book presents concepts and applications of reservoir engineering principles
essential to optimum development of natural gas reservoirs. It is based on courses
taught at The University of Tulsa, The Pennsylvania State University, and adult
education seminars in the United States and overseas.
The development of a natural gas field always depends on the reservoir and
well characteristics as well as the equipment performance. A systems approach is
emphasized throughout the book, since change in any component of the field
production system will affect the performance of the other components. This
book is arranged so that it can be used as a text or reference work for students and
practicing engineers, geologists, and managers in the crude oil and natural gas
production industry.
Chapter 1 discusses methods of estimating nonassociated, associated, and
dissolved gas and abnormally pressured gas reserves. Reserves estimation and
performance prediction for gascondensate reservoirs are treated in Chapter 2. A
comprehensive and rigorous treatment of production decline curve analysis is
given in Chapter 3.
In Chapters 4 and 5 the theory and application of gas well testing are
discussed. Well test analysis is an important subject in reservoir engineering, since
it enables us to obtain reservoir parameters that could be used to predict future
reservoir performance. Chapter 4 considers deliverability or backpressure test
ing of gas wells. Chapter 5 discusses pressure transient analysis for gas wells. Both
the pressuresquared technique and the pseudopressure function or real gas
potential technique are treated and compared.
The systems approach is used to determine optimum gas field development
strategies in Chapter 6; examples of reservoir performance techniques and field
development patterns are presented. Chapter 7 extends some of the teChniques of
gas transmisison and gas reservoir engineering to the storage of natural gas.
Much of the material on which this book is based was drawn from the
publications of the Society of Petroleum Engineers of the American Institute of
Mining, Metallurgical and Petroleum Engineers, the American Gas Association,
the Division of Production of the American Petroleum Institute, and the Gas
Processors Suppliers Association. Tribute is due to these organizations and also to
a host of schools and authors who sponsor programs and have contributed to
petroleum literature in various other publications.
viii Preface
I am indebted to my students, whose enthusiasm for the subject has made
teaching a pleasure. To my colleagues who have adopted this material in various
petroleum and natural gas engineering departments in the United States and
overseas, I express my gratitude for their constructive criticisms and comments
that became textbook inputs. I thank Peggy Conrad for typing the manuscript.
I would like to express my appreciation to the editorial staff of John Wiley,
including Merrill Floyd and Deborah Herbert, for their patience and politeness.
I thank Cindy SteinLapidus and the memoers of Wiley's production staff for a
fine job.
Chi U. lkoku
I
CONTENTS
Nomenclature xv
1. ESTIMATION OF GAS RESERVES 1
1.1
1.2
1.3
1.4
1.5
Introduction 1
Gas in Place by Volumetric Equation 2
MaterialBalance Equation 4
1.3.1 Assumptions 4
1.3.2 Derivation 4
1.3.3 Application 6
Reserves and Reservoir Performance Predictions 6
1.4.1 Volumetric Estimates 7
1.4.2 Material Balance Estimates 10
1.4.3 Pressure Decline Curve plz Method 15
1.4.4 MBE Straightline Method (After Havlena and Odeh) 20
1.4.5 Reservoir Size 31
1.4.6 Rate versus Time 32
1.4.7 Liquid Recovery 33
1.4.8 Associated Gas Reserves 34
1.4.9 Dissolved Gas Reserves 35
General MaterialBalance Equation 36
1.5.1 Gas Reservoirs 37
1.5.2 Abnormally Pressured Gas Reservoirs (After Hammerlindl)
1.5.3 Graphical Technique for Abnormally Pressured Gas Reservoirs
References 49
Problems 50
2. GAS CONDENSATE RESERVOIRS 53
2.1 Introduction 53
2.2 VaporLiquid Equilibriums 55
2.2.1 Calculation of VaporLiquid Equilibrium 56
40
47
2.2.2 Determination of Convergence Pressure and Equilibrium Ratios 58
2.2.3 BubblePoint Pressure 60
2.2.4 DewPoint Pressure 60
x Contents
2.3 GasCondensate Testing and Sampling 60
2.3.1 Laboratory Tests of Condensate Systems 61
2.4 Condensate System Behavior in the SinglePhase Region
2.4.1 Calculation of Initial Gas in Place and Oil in Place for
GasCondensate Reservoirs 64
2.5 Condensate System Behavior in the TwoPhase Region
2.5.1 TwoPhase Gas Deviation Factor 69
2.5.2 Condensate Material Balance 72
64
67
2.5.3 Reservoir PerformanceRetrograde GasCondensate Reservoirs
2.6. Reservoir Performance Prediction 76
2.6.1 Reservoir Operation by Pressure Depletion
2.6.2 GasCondensate Reservoir Operation by Pressure Maintenance
or Cycling 87
2.6.3 Economics of GasCondensate Recovery 91
References 92
Problems 93
3. PRODUCTION DECLINE CURVES 95
3.1
3.2
3.3
Introduction 95
Economic limit 97
Classification of Decline Curves
3.3.1 Nominal and Effective Decline
3.3.2 Constant Percentage Decline
3.3.3 Harmonic Decline 113
3.3.4 Hyperbolic Decline 117
97
100
101
3.4 Fraction of Reserves Produced at a Restricted Rate
3.5 Summary 133
References 13 7
Problems 137
4. DEliVERABILITY TESTING OF GAS WELLS 141
4.1 Introduction 141
4.2 Horizontal Flow Equations 142
4.2.1 Basic Differential Equation 142
4.2.2 Steady State Flow 142
130
77
4.2.3 Gas Well Testing (According to the Steady State Theory) 151
4.3 Non Darcy Flow 152
4.4 Gas Well Deliverability Tests 156
4.4.1 Flow After Flow Tests 159
4.4.2 Isochronal Tests 162
4.4.3 Modified Isochronal Tests 166
4.4.4 Deliverability Plot 169
4.4.5 Perfonnance Coefficient C and Exponent n 170
72
r
Contents xi
4.5 SemiSteady State Equation 178
4.6 Beller Method for Analyzing Isochronal Test Data 182
4.7 JonesBlountGlaze Method 188
4.8 European Method 190
4.9 Future Inflow Performance Relationships 192
References 194
Problems 195
5. TRANSIENT TESTING OF GAS WELLS 201
5.1 Introduction 201
5.2 Transient Flow of Real Gases Through Porous Media 202
5.2.1 PressureSquared Representation 203
5.2.2 AIHussainyRameyCrawford Technique 204
5.2.3 When to Use Pseudo Pressure Approach 207
5.3 The Constant Terminal Rate Solution 209
5.3.1 Boundary Effects 214
5.4 Application of Real Gas Flow Equations 216
5.4.1 Drawdown Testing 216
5.4.2 Buildup Testing 220
5.4.3 Summary 223
5.5 Average Reservoir Pressure 233
5.5.1 Finite Reservoirs 233
5.5.2 MatthewsBronsHazebroek Method 235
5.6 Other Topics 240
5.6.1 Wellbore Storage 240
5.6.2 Fractured Wells 242
5.6.3 TypeCurve Matching 247
5.6.4 Wells Producing by Solution Gas Drive (TwoPhase Flow)
5.6.5 Restricted Entry 261
References 271
Problems 273
6. GAS FIELD DEVELOPMENT
6.1 Introduction 279
6.2 Reserves 280
6.3
6.2.1 Reservoir Performance
6.2.2 Field Development Pattern
Deliverability 283
6.3.1 Reservoir Deliverability
6.3.2 Well Spacing 287
279
280
282
284
6.3.3 Equipment Capacity Limitations 289
258
xii Contents
6.4 Predicting Reservoir Performance 293
6.4.1 Reservoir versus FlowLine Capacity 294
6.4.2 RateTime Prediction 296
6.4.3 Use of Darcy's Radial Flow Equation 298
6.5 Optimum Development Patterns (After van Dam)
6.5.1 Gas Field Development Model 307
6.5.2 PresentValue Calculations 308
6.5.3 Optimum Production Rate 309
References 311
Problems 312
7. STORAGE OF NATURAL GAS 317
7.1 Introduction 317
7.2 Natural Gas Storage in Pipelines 318
7.2.1 Storage Capacity of Simple Pipelines 319
7.3 Underground Storage of Natural Gas
7.3.1 Purpose of Underground Gas Storage
7.3.2 Segments of a Gas Storage Reservoir
7.3.3 Storage Field Reservoir Consideration
322
322
323
324
7.4 Storage in Depleted Oil Reservoirs 326
7.5 Storage in Aquifers 327
7.5.1 Exploring for Aquifer Storage Reservoirs 327
7.5.2 Growth of Storage Bubble 330
7.5.3 Operation of Aquifer Storage Reservoirs 343
7.6 Natural Gas Storage in ManMade Caverns 343
7.6.1 Storage in Salt Caverns 344
7.6.2 Storage in Conventionally Mined Caverns 349
7.6.3 S.torage in Converted Mines 352
7.6.4 Summary 352
References 353
APPENDIX A PROPERTIES OF NATURAL GASES 355
A.l
A.2
A.3
A.4
A.5
A.6
Physical Constants 355
Pseudocritical Properties of Gases
Gas Deviation Factor (zFactor)
Gas Formation Volume Factor Bg
Gas Viscosity I' 380
Gas Compressibility c
g
380
References 386
355
363
363
305
Contents xiii
APPENDIX B THE ORIFICE METER 387
8.1 Orifice Meter Equation and Constants
8.2 Orifice Meter Tables for Natural Gas
References 389
APPENDIX C BOTTOMHOLE PRESSURES
387
389
463
C.l AverageTemperature and zFactor Method
C.2 The Sukkar and Cornell Method 466
C.3 The Cullender and Smith Method 496
INDEX 501
463
,.
NOMENCLATURE
QUANTITIES IN ALPHABETICAL ORDER
(*) Dimensions: L = length, m = mass, q = electrical charge, t = time, and T =
temperature.
(**) To avoid conflicting designation in some cases, use of reserve symbols and reserve
subscripts is permitted.
ReserveSPE
SPE Letter Dimeo
Quantity Standard Symbols** sions*
air requirement a
F"
angle c< alpha beta
angle e theta y gamma
angle, con tact 8(. theta 'Y(" gamma
angle of dip ad alpha 8(1 theta
area A S L'
Arrhenius reaction rate velocity constant w z L
3
/m
breadth, width, or (primarily in fracturing) thick
ness b w L
burningzone advance rate Vb Vln Ub L/t
capillary pressure P,. Pc> Pr m/LI'
charge Q q q
coefficient, convective heat transfer h hln hT m/t'T
coefficient, heat transfer, interphase convective
(use,h. or convective coefficient symbol, with
pertinent phase subscripts added) m/t
3
T
coefficient, heat transfer, overall U UT, U. m/t'T
coefficient, heat transfer, radiation I IT, 10 m/r'T
components, number of C n,.
compressibility c k, K kappa Lt'/m
compressibility factor z Z
concentration C C, n various
Courtesy of Society of Petroleum Engineers of
American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc.
xvi Nomenclature Nomenclature xvii
ReserveSPE ReserveSPE
SPE Letter SPE Letter
Quantity Standard Symbols" sions* Quantity Standard Symbols" sions*
condensate or natural gas liquid content C
L CL> fiL various fuel consumption m h
various
conductivity a sigma 'Y gamma various fuel deposition rate NR N,.. mlL't
conductivity, thermal (always with additional gas (any gas, including air)always with identify
phase or system subscripts)
k"
A lambda mLlt'T iog subscripts G g various
contact angle St theta "I .. gamma
gas in place in reservoir, total initial G g L'
damage ratio 'skin" conditions relative to forma
gasoil ratio, producing (if needed. the reserve
tion conditions unaffected by well operations) F, F,/
symbols could be applied to other gasoil ratios) R F
R
,
p rho D miL'
general and individual bed thickness h d, e L
density
gradient various
depth D y, H L
g y gamma
diameter d D L
heat flow rate Q q, <I> phi
clIp
mL'It'
diffusion coefficient D /.Lmu,8delta L'lt
heat of vaporization. latent
L"
A, lambda L';i'
dimensionless fluid influx function, linear aquifer QUI) Q'If)
i
heat or thermal diffusivity " alpha
cr, "" eta
L'lt
heat transfer coefficient. convective h h", hT mlt'T
dispersion coefficient K d L'lt
heat transfer coefficient, interphase convection
displacement s L L
(use h, or convective coefficient symbol with
displacement ratio 5 delta
F" pertinent subscripts added) mlt'T
distance between adjacent rows of injection and
heat transfer coefficient, overall U U
T
, U, mlt'T
production wells d Lib L, L
heat transfer coefficient. radiation I Jr. 10 mlt'T
distance between like wells (injection or produc
height (elevation) Z D, h L
tion) in a row a L(/> L
J
L
height (other than elevation) h d, e L
distance, length, or length of path L s, I script I L
hydraulic radius rH RH L
efficiency E " eta, e
i
index of refraction n fL mu
electrical resistivity p rho R mL'ltq'
influx (encroachment) rate e L'lt
electromotive force (voltage) E V mL2/t
2
q
influx function, fluid. linear aquifer. dimension
elevation referred to datum Z D, h L
less Q,.ff) Q,,/)
encroachment or influx rate e i L'lt
initial water saturation SWi Pwi rho, Swi
energy E U mL'lt'
injectivity index I i L'tlm
enthalpy (always with phase or system subscripts) H I mL2/r
intercept b Y various
enthalpy (net) of steam or enthalpy above reser
interfacial or surface tension (J'sigma y, "y gamma m/r
voir temperature
(. mL2/r
interstitialwater saturation in oil band Sit'/) Sn'/>
enthalpy, specific h L
2
/P
irreducible water saturation Sill' Pil\' rho. Sin'
entropy, specific s (T sigma L'Ii'T
kinematic viscosity v nu N L'lt
entropy. total S IT, sigma mL
2
/t
2
T
length L s./script I L
equilibrium ratio K k, F"1f length, path length, or distance L s./script I L
fluid influx function. linear aquifer. dimensionless Qun Q'II) mass flow rate w m mit
flow rate or flux, per unit area (volumetric ve
mobility ratio M F,
locity) u tJi psi Lit
,
mobility ratio. diffusefront approximation,
flow rate or production rate q Q L'lt
[(An + Ad)swcpt/(A,I)unswcpt}; D signifies dis
fluid (generalized) F
f
various
placing; d signifies displaced; mobilities are
flux u tJi psi various
evalaluated at average saturation conditions be
force F Q mLli'
hind and ahead of front Ms MIJi/. M.\'II
formation volume factor B F
mobility ratio, sharpfront approximation. (An/Ad) M F,
fraction gas f, F,
mobility ratio. total. [(A/)swcp/(A,)ullswcpt];
fraction liquid ff. F/., J, "swept" and "unswept" refer to invaded and
frequency
f
v nu II t
un invaded regions behind and ahead of leading
xviii Nomenclature Nomenclature xix
ReserveSPE ReserveSPE
SPE Letter Dimen SPE Letter Dimen
Quantity Standard Symbols sions* Quantity Standard Symbols sions*
edge of a displacement front M,
F"
reciprocal formation volume factor. volume at
mobility, total, of all fluids in a particular region standard conditions divided by volume at reser
of the reservoir; e.g., (Ao + At: + A".) Af lambda A lambda
cap L't/m
voir conditions b f, F
modulus, bulk K
K"
m/LI' reciprocal permeability j womega I/L'
modulus of elasticity in shear G Es m/LI'
resistance r R mL'2/tq'2
modulus of elasticity (Young's modulus) E Y m/LI' resistance r R various
mole fraction gas f, F,
resistivity, electrical P rho R mLJ/tq'
mole fraction liquid
h
F
L
,/, saturation S P rho, S
molecular refraction R N L' saturation, water, initial Swi PlI'i rho, Swi
moles, number of n N saturation, water, irreducible Sill' Pill' rho, Sill'
moles of liquid phase L nL
skin effect s S, IT sigma
moles of vapor phase V n, skin (radius of well damage or stimulation) r.v R.I L
moles, total n nf, Nf slope m A various
number (of moles, or components', or wells, etc.) n N specific gravity 'Y gamma s, Fl.
oil (always with identifying subscripts) n n various specific heat (always with phase or system sub
oil in pi<;lce.in reservoir, initial N n L'
scripts) C c L
2
/I'T
oxygen utilization
E(l; specific heats ratio y gamma k
path length, length, or distance L s, i script I L
spec;ific injectivity index L't/m
permeability k K L' specific productivity index J.I iv L't/m
Poisson's ratio fL mu v nu, <T sigma specific volume v Vs L'/m
porosity <I> phi f, e epsilon specific weight F\I"u 'Y gamma m/L'T'
pressure p P m/LI'
stimulation radius of well (skin) rs Rs L
production rate or flow rate q Q L '/t
strain, normal and general E epsilon e, En epsilon
productivity index J j L"t/m
strain, shear y gamma Ex epsilon
quality (usually of steam) f., Q,x strain, volume e theta e
u
theta
radial distance b.r b.R L
stress, normal and general <T sigma s m/LI'
radius r R L
stress, shear T tau S.v m/LI'
radius, hydraulic rll Rtf L
surface tension IT sigma y, gamma m/I'
ratio, damage ("skin" conditions relative toJor temperature T e theta T
mati on conditions unaffected by well opera thermal conductivity (always with additional
tions) F, F,/
phase or system subscripts)
k"
A lambda mL/t'T
ratio initial reservoir free gas volume to initial thermal cubic expansion coefficient beta h I/T
reservoir oil volume m
F po' F.!to
thermal or heat diffusivity a alpha a, ...,,, eta L'/t
ratio, mobility M F,
thickness (general and individual bed) h d, e L
ratio, mobility, diffusefront approximation, time T tau
[(AD + Ad)swcr/(Ad)unswcpt]; D signifies dis
total mObility of all fluids in a particular region
placing; d signifies displaced; mobilities are of the reservoir; e.g., (Ao + At; + Aw) A, lambda A lambdacap L't/m
evaluated at average saturation conditions be total mobility ratio, [(Af)swcpt/(hf)ullswcPt];
hind and ahead of front M., MOd, M.I"II
"swept" and "unswept" refer to invaded and
ratio, mobility, sharpfront approximation, uninvaded regions behind and ahead of leading
(AO/Ad) M FA
edge of a displacement front M,
F"
ratio, mobility, total,
transfer coefficient, convective heat h
h", hT m/t'T
"swept" and "unswept" refer to invaded and transfer coefficient, heat, interphase convective
uninvaded regions behind and ahead of leading (use h, or convective coefficient symbol with
edge of a displacement front M,
F"
pertinent phase subscripts added) m/t'T
reaction rate constant k r, j L/t
transfer coefficient, heat, overall U UT, Un m/t'T
xx Nomenclature Nomenclature xxi
ReserveSPE Reserve SPE
SPE Letter Dimeo SPE Letter
Quantity Standard Symbols" sions* Subscript Standard Subscripts**
transfer coefficient, heat. radiation I I.,., In mlr'T entry e E
utilization, oxygen
velocity v V. II ILlt
viscosity fL mu 11 eta miLt
volume V v L'
equivalent eq EV
estimated E est
experimental E EX
fillup F
f
volumetric velocity (flow rate or flux. per unit
area) u <jJ psi Lit
water (always with identifying subscripts) W w various
water in place in reservoir. initial W w L'
water saturation, initial SI\'; p",; rho, SII';
finger or fingering f
F
flash separation f
F
fraction or fractional
f
r
fracture, fractured, or fracturing
f
F
free (usually with gas or gasoil ratio quantities) F
f
water saturation. irreducible S;.,. Pilt' rho. S;II'
front, front region, or interface
f
F
wave number (J'sigma fJ IlL gas g G
weight W w.G mLI!'
wetgas content Cng
clI.g fll\.g various
width, breadth, or (primarily in fracturing) thick
ness b w L
gross I T
heat or thermal h T, e theta
hole h H
horizontal H h
work W w mL2jr hydrocarbon h H
imbibition I i script i
influx (encroachment), cumulative e i
injected, cumulative I
Reserve SPE injection, injected, or injecting inj
SPE Letter inner or interior L iota, i script i
Subscript Standard Subscripts** interface, front region, or front
f
F
interference I i, i script i
air a A invaded I
atmospheric
"
A
average or mean saturation S p rho, s
band or oil band b B
invaded zone I
invasion I
irreducib Ie i script i, L iota
base b r, beta linear, lineal L I script I
boundary conditions, external e a
breakthrough BT bl
bubble point or saturation b s
burned or burning b B
calculated C calc
liquid or liquid phase L I script I
lower !. script I L
mean or average saturation S p rho, s
mixture M m
mobility A lambda M
capillary (usually with capillary pressure, Po) c
c
casing or casinghead c cg
nonwetting nw NW
normalized (fractional or relative) n r, R
contact (usually with contact angle, eJ c C oil a n
core c C outer or exterior e a
cumulative influx (encroachment) e
damage or damaged (includes "skin" conditions) s d
depleted region, depletion d 6 delta
dispersed d D
dispersion K d
displaced d s, D
displacing or displacement D S, (1 sigma
permeability k K
pore (usually with volume, V
p
) p P
production period (usually with time, t
p
) p P
radius, radial, or radial distance r R
reference r b, p rho
relative r R
reservoir R r
xxii Nomenclature
SPE
Subscript Standard
residual
~
saturation, mean or average S
saturation or bubble point b
segregation(usually with segregation rate, q,) s
shear s
skin (stimulation or damage) s
slip or slippage s
solid(s) s
stabilization (usually with time) s
steam or steam zone s
stimulation (includes "skin" conditions) s
storage or storage capacity S
strain e epsilon
surface s
swept or swept region s
system s
temperature T
thermal (heat) h
total, total system t
transmissibility T
treatment or treating
tubing or tubing head
unswept or unswept region u
upper u
vaporization, vapor, or vapor phase v
velocity v
vertical V
volumetric or volume V
water w
weight W
wellhead wh
wetting w
Reserve SPE
Letter
Subscripts**
R
p rho, S
s
S, (J" sigma
T tau
S
IT sigma
IT sigma
S
S
S
S, (J" sigma
e
(J" sigma
S, (J" sigma
(J" sigma
h, e theta
T, e theta
T
T tau
tg
U
U
V
V
v
v
W
w
th
W
I
1
ESTIMATION OF
GAS RESERVES
1.1 INTRODUCTION
Natural gas reservoirs are reservoirs in which the contained hydrocarbon fluids
exist _wholly as a vapor phase at pressure values equal to or less than the initial
value. n l i k ~ saturated crude oils and condensates, natural gases do not undergo
phase changes upon reduction in reservoir pressure. Performance predictions are
therefore relatively simple.
As it exists in the reservoir, natural gas is commonly termed wet (or raw) gas.
The total well effluent productionwhich may be a twophase mixture at the
wellheadis also called wet gas. Ordinarily, the wet gas produced is not measured
directly. Instead, the wet gas production is determined by summing separator
(residue) gas production and the vapor equivalent of separator liquid production.
For the sake of. simplicity and because the reservoir gas is a constant
composition fluid, no 'differentiation will be made between wet and residue gas in
the subsequent discussion of natural gas reservoirs. Cumulative gas produced,
G
p
, as used in this chapter means separator gas as measured plus the vapor
equivalent of the natural gas liquid (NGL) removed in the separator. Similarly,
gas formation volume factor, B
g
, and gas deviation factor, z, refer to the proper
ties of a sample of separator gas and liquid that has been recombined to represent
reservoir gas composition.
Natural gas reserves are classified according to the nature of their occur
rence. Nonassociated gas is free gas not in contact with crude oil in the reservoir.
Associated gas is free gas in contact with crude oil in the reservoir. Dissolved gas
is gas in solution with crude oil in the reservoir.
Most of this chapter will address methods of estimating nonassociated gas
reserves. Methods of estimating associated and dissolved gas will be briefly
discussed.
2 Estimation of Gas ReseITes
1.2 GAS IN PLACE BY VOLUMETRIC EQUATION
In order to make reasonable recovery predictions, estimates of the initial gas in
place in each reservoir must be made. The volumetric equation is a useful tool for
calculating the gas in place at any time; consequently, it has considerable utility in
estimating gas reserves.
Here, pore space volume in the reservoir containing gas is converted to gas
volume at standard conditions. Net volume of reservoir rock containing the gas
reserves is determined by geological information based on cores, electric or
radioactive logs, drilling records, and drill stem and production tests. Reservoir
rock volume is usually obtained by planimetering isopachous maps of productive
reservoir rock or by the polygon method for computing volumes.
The standard cubic feet of gas initially in place, G, is simply the product of
three factors: the reservoir pore volume, the initial gas saturation, and a volume
ratio (initial gas formation volume factor) that converts reservoir volumes to
volumes at standard, or base, conditions (normally 60'F and 14.7 psia). These
factors are related as follows:
where
1
G = 7758Ah<l>(1  Sw,) 
Bgi
G = initial gas in place, scf
7758 = conversion factor, bbllacreft
A = original productive area of reservoir, acres
h = net effective formation thickness, ft
<I> = porosity, fraction
Ah<l> "" pore volume of reservoir, acreft
Swi = interstitial water saturation, fraction
Bgi = initial gas formation volume factor:
Pb
Tz
res bbllscf = :'"
5.615pTbZb
P and Pb = pressure at reservoir and base conditions, psia
T and Tb = temperature at reservoir and base conditions, 'R
Z and Zb '" gas deviation factor at reservoir and base conditions
(1.1)
If B gi is in cu ft/scf, Eq. 1.1 becomes
1
G = 43,560Ah<l>(1  Sw,) Bgi (1.2)
Also
G = 43,560Ah<l>(1  Swi) pTbZb
ZPbT (1.3)
Gas in Place by Volumetric Equation 3
At any subsequent reservoir pressure, the standard cubic feet of gas in place is
given by .
(1.4)
The volumetric equation is particularly applicable when a field is compara
tively new, that is, before sufficient quantities of gas have been produced to cause
an appreciable drop in reservoir pressure. If good data are available, then the
volumetric equation will probably be reliable. The gas formation volume factor
Bg is equal to the volume at reservoir temperature and pressure occupied by one
standard cubic foot of gas. From gas laws,
Pb
Tz
B =
g pTbZb (1.5)
If standard (or base) conditions are assumed to be 14.7 psia and 60'F, and Zb = 1
then Eq. 1.5 becomes
(14.7)(Tz) Tz
B = = 0.0283
g p(460 + 60) P (1.6)
The gas factor nroperlyJ1.ecause the omission of this
factor in gas reserve calculations may introduce errors as large as 30%.
study of core analyses. The porosity of clean sandstones may be calculated from
electric logs. In limestone reservoirs, core analyses may be supplemented by
porosity determinations from electric logging in formation. Average porosities
may be determined with a fair degree of accuracy for homogeneous sandstones;
but for limestones and heterogeneous sandstones, estimating an average porosity
is at best uncertain.
Many early estimates of gas reserves were in error because they did not
consider splfce of connate
water saturation has received considerable attention recently, such data may be
lacking entirely for many older gas reservoirs and must be estimated.onnate
water saturations rna)' be determined fro
laboratory eterminations run on cores by restored state, evaporation mercury
injection, or centrifuge methods. Connate water saturation may range as high as
50% in some gas reservoirs.
_ Reservoir pressure and temperature may be measured directl), with subsur
face
gas reserves, it is assumed that the reservoir
pressure used is completely stabilized. After a considerable quantity of gas has
been low permeability, pressures stabil
ize very slowly, often requiring several days (or longer). Consequently, estimators
should study pressure buildup characteristics of individual wells as well as pres
\
4 _ Estimation of Gas Reserves
sure distribution throughout the reservoir. Where pressure varies over the reser
voir, the average should be determined by volumetrically weighing individual
pressures.
Example 1.1. Estimate gas in place in a reservoir with an areal extent of 2550
acres, average thickness of 50 ft, average porosity of 20%, connate water satur
ation of 20%, reservoir temperature of 186F, initial reservoir pressure of 2651
psia, and reservoir gas deviation factor of 0.880 at 186F and 2651 psia.
Solution
Using Eq. 1.3
(2651)(520) .
G = (43,560)(2550)(50)(0.20)(1  0.20) (0.880)(14.7)(646)
= 146,588 MMscf
1.3 MATERIALBALANCE EQUATION
A materialbalance process is an exact accounting of the materials that enter,
accumulate in, or are depleted from a defined volume in the course of a given time
interval of operation. The material balance is, therefore, an expression of the
law of conservation of mass.
1.3.1 Assumptions
1. A reservoir may be treated as a constantvolume tank.
2. Pressure equilibrium exists throughout the reservoir, which implies that
no large pressure gradients exist across the reservoir at any given time.
3. Laboratory pressurevolumetemperature (PVT) data apply to the reservoir
gas at the average pressures used.
4. Reliable production and injection data and reservoirpressure measurements
are available. "'" ' , " ' ,
5. The change in volume of the interstitial water with pressure, the change in
porosity with pressure, and the evolution of gas dissolved in the interstitial
water with decrease in pressure are negligible.
1.3.2 Derivation
The conservation of mass may be applied to a gas reservoir to yield mass and mole
balances
mp=mjm (1.7)
and
.C _
(1.8)
where
MaterialBalance Equation 5
mp, np = cumulative gas produced in mass and mole units, respectively
m" ni = initial gas in place at initial pressure, Pi
m, n = gas remaining in reservoir at some subsequent pressure, P
Since the composition of the produced gas is constant, the gas volumes in standard
cubic feet (both produced and remaining in the reservoir) are directly propor
tional to masses and moles.
Using the constantvolume tank concept, l e ~ e the original hydrocarbon
reservoir volume (bbl) at the initial pressure Pi' Assume that at some subsequent
pressure P, Gp standard cubic feet of gas and Wp stocktank barrels of water have
been produced at the surface, We reservoir barrels of water have encroached into
the reservoir, and the remaining gas volume in the reservoir is V barrels. Since the
reservoir being considered is assumed constant, the following equation results:
(1.9)
Or
(1.10)
V" V, W" and WpBw,are in reservoir barrels; Bw is the water formation volume
factor in reservoir barrels per stocktank barrel. From the gas law,
Thus
and
where
pV
n=
zRT
p.v.
ni = 5.615 ' '
ziRT
P
v p(V  W + W B )
n = 5.615  = 5,615 ' ,e p w
zRT zRT
G
p
= cumulative gas produced from Pi to P, scf
R = gas constant, 10,732 cu ftpsi/lb molOR
6 Estimation of Gas Reserves
Substituting in Eq. 1.8 gives
[
PYi _ P(Vi  W, + WpBw)]
ziRT zRT
Or
ZbTb[PiVi cP(,V,i
G
p
= 5.615    
. . . pf Zi Z (1.11)
Therefore, expressing Vi in terms of G (scf of gas initially in place) and substitu
ting gas formation volume factors Bgi and Bg at pressures Pi and P, Eq. 1.11
becomes
G =
p B
i (1.12)
For reservoirs with no water influx and no water production, Eqs. 1.11 and 1.12
become, respectively: .
G
p
= 5.615 ZbTb Vi (Pi _l!.)
PbT Zi Z
(1.13)
and
(1.14)
1.3.3 Application
The material balance equation for a gas reservoir may be applied to estimate
initial gas in place fr()m performance data, determine the existence and estimate
effectiveness of any natural water drive, and assist in predicting performance and
reserves. It may also verify possible extensions to a partially developed reservoir
where gas in place calculated by the material balance equation is much larger than
a volumetric equation estimate and water influx is thought to be small.
1.4 RESERVES AND RESERVOIR PERFORMANCE PREDICTIONS
Efficient development and operation of a natural gas reservoir depend on knowl
edge of how the reservoir will perform in the future. To predict recovery as a
function of pressure or time, sources of energy for producing the gas from the
reservoir must be identified and their contribution to reservoir performance
evaluated. The energy required for gas production is usually derived either from
gas expansion or a combination of gas expansion and water influx.
Comparison with other fields, volumetric estimation, and decline curve are
" .
Reserves and Reservoir Performance Predictions 7
methods that may be used to estimate gas reserves in place in the reservoir; but in
actual practice, recoverable reserves are of greatest interest. Their estimation
requires predicting an abandonment pressure at which further production from
the wells will no longer be profitable. The abandonment pressure is determined
principally by economic conditions such as future market value of gas, cost of
operating and maintaining wells, and cost of compressing and transporting gas to
consumers. Since these factors are quite variable, this discussion of estimation
methods is confined to reserves in place in the reservoir and their recovery
efficiencies.
Judging the applicability of similar fields with known producing histories
depends on the experience of the estimator. When reserve estimates are required
early in the producing life of a reservoir, for example after completion of one or
two wells when practically no gas has been produced, experience in similar fields
may be the only guide to reserve estimation.
1.4.1 Volumetric Estimates
The volumetric equation (Eq. 1.1) is useful in reserve work for estimating gas in
place at any stage of depletion. During the development period before reservoir
limits have been accurately defined, it is convenient to calculate gas in place per
acrefoot of bulk reservoir rock. Multiplication of this unit figure by the best
available estimate of bulk reservoir volume then gives gas in place for the lease,
tract, or reservoir under consideration. Later in the life of the reservoir, when the
reservoir volume is defined and performance data are available, volumetric
calculations provide valuable checks on gas in place estimates obtained from
material balance methods:
Or
1
G(scflacreft) = 7758<1>(1  SWi) 
. B.i(res bbl/scf)
1
. G(scflacreft) = 43,560<1>(1  Swi)
B.i(res ft 3/scf)
pTb
= 43 560"'(1  S .)
,'t' wr ZPb
T
where bulk reservoir volume = Ah, acreft.
(1.15)
(1.16)
For natural gas reservoirs under volumetric control (no water influx or water
production), the cumulative gas produced G
p
at any pressure is the difference
between the volumetric estimates of gas in place at the initial and subsequeni
pressure conditions. Thus, for volumetric reservoirs,
. ( 1 1 )
. G
p
= 7758Ah<l>(1  Sw)   
Bgi Bg (1.17)
8 Estimation of Gas Reserves
If the gas formation volume factor Bga at the assumed abandonment pressure
(say 100 psia with compressor facilities and 500 psia without) is substituted for B
g
,
Eq. 1.17 gives G p at abandonment or the recoverable gas in place at original
conditions. Another approach often used in gas reserve estimation is to calculate
the initial gas in place from the volumetric equation and apply a recovery factor.
The recoverable reserves are then given by
where
Ra = 43,560<1>(1  Sw;)Eg
l!gi.
Ra = gas reserves to abandonment pressure, scf/acreft
(1.1S)
Eg = recovery factor, fraction of initial gas in place to be recovered
The recovery factor from a gas reservoir is primarily a function of the
abandonment pressure and permeability. Lowering the abandonment pressure
will increase the recoverable gas. The abandonment pressure used depends on the
price of gas, the productivity indexes of the wells, the size of the field, its location
with respect to market, and the type of market. If the market is a transmission
pipeline, the <?perating pressure of the line may be a controlling factor in the
abandonment pressure for small fields; but for large fields, installation of com
pressor plants may be economically feasible, thus lowering the abandonment
pressure substanti.ally below the operating pressure of pipeline serving the area.
Some gas pipeline companies use an abandonment pressure of 100 psi/lOOO ft of
depth.
Waterdrive gas reservoirs usually have a lower recovery factor than closed
gas reservoirs because of the high abandonment pressure due to water encroach
ment into the producing wells. The reservoir permeability is also a primary factor
governing the recovery from a closed gas reservoir. Higher permeabilities result in
high flow rates for a given pressure drop. Therefore, when all other factors are the
same, the abandonment pressure is lower for a highpermeability reservoir. The
recovery factor is high when the sand is uniform and homogeneous, the perme
ability of the sand is high at low gas saturation, the percentage of the gas
containing portion of the reservoir originally underlain bYb water is relatively
small, the beds are relatively steep, and the amount of structural closure above the
gaswater contact is large.
For closed gas reservoirs, the principal factor governing recovery efficiency is
the abandonment pressure. If the abandonment pressure is known, a recovery
factor can be calculated. Expressed in percent of initial gas in place, the recovery
factor is
(1.19)
Reseryes and Reseryoir Performance Predictions 9
For waterdrive reservoirs,
E = ::1::00::(:,:S.,g,'.B...
g
"a _.::.S;;:ga"B... g"';)
g SgiBga (1.20)
where
Sgi = initial gas saturation, fraction of initial pore volume
Sga = abandonment gas saturation, fraction of initial reservoir pore
volume
B
g
, = initial gas formation volume factor (FVF), res bbllscf or cu ftlscf
Bga = abandonment gas FVF, 'res bbl/scf or cu ft/scf
Pa = abandonment pressure, psia
Za = gas deviation factor at abandonment
For strong water drives where residual gas is trapped at high pressures, Eg may
be 50 to 60% compared to 70 to SO% for partial water drives and SO to 90% for
volumetric reservoirs.
Table 1.1 provides values for residual gas saturation after water flood in core
plugs, as measured by Geffen. These may be used in Eq. 1.20 as an approximation
.. for SgQ'
Example 1.2. A proposed gas well is being evaluated. Well spacing is 640 acres
and it appears that the entire 640 acres attributed to this well will be productive.
Geological estimates indicate 30 ft of net effective pay, 15% porosity, and 30%
TABLE 1.1
Residual Gas Saturation After Water Flood as Measured on Core Plugs
Porous Material
Unconsolidated sand
Slightly consolidated
sand (synthetic)
Synthetic consolidated
materials
Consolidated sandstones
limestone
Source: After Geffen et a1.
Formation
Selas Porcelain
Norton Alundum
Wilcox
Frio
Nellie Bly
Frontier
Springer
Torpedo
Tensleep
Canyon Reef
~ ~ g , percent
16
21
17
24
2S
3038
3036
3134
33
3437
. 4050
50
10 Estimation of Gas Reseryes
interstitial water saturation. The initial pressure is 3000 psia and reservoir tem
perature is 150'F. The abandonment pressure is estimated to be 500 psia. The gas
gravity is expected to be 0.60. Base temperature and pressure are 60'F and
14.65 psia, respectively. An estimate of the gas reserve is required.
Solution
The first step is a calculation of B.
i
, which requires pseudocritical T and P,
pseudoreduced Tand P, and then z. From Fig. A.1, the pseudocritical pressure
and temperature for a 0.6gravity gas are 668 psia and 385'R, respectively.
3000 psia
Pseudoreduced pressure = = 4.5
668 psia
(150 + 460)'R
Pseudoreduced temperature = 385'R = 1.6
Referring Fig. A.2, Zi is found to be 0.83. Using Eq. 1.5,
_ (14.65)(150 + 460)(0.83) _ 00475 f 1 f
B .   O. 5 cu t sc
.' (3000)(60 + 460)(1.0)
The second step is to calculate the recovery factor, E
g
. Abandonment pressure
being 500 psia, the pseudoreduced pressure = 500/668 = 0.75. Using this value
together with the pseudoreduced temperature of 1.6 in Figure A.2, Za is found to
be 0.94. Hence from Eq. 1.19,
(500)(0.83)
Eg = 1  = 1  0.147 = 0.853 = 85%
(3000)(0.94)
The third step is to use Eq. 1.18 to calculate reserve in scflacreft:
43 560 x 0.15 x (1  0.30)
Rc = ' x 0.85 = 817,609 scflacreft
0.004755
The final step is to multiply the above figure by the net acrefeet; hence, the
estimated reserve:
G = (817,609 scf/acreft) x (640 acres)(30 ft)
= 15,698,092,800 scf
= 15.7 billion cu ft of gas
1.4.2 Material Balance Estimates
Gas in place, reserves, and water influx may be estimated from performance
history using material balance methods. This provides an independent check on
volumetric methods.
In some cases the porosity, connate water, or the effective reservoir volumes
Reserves and Reservoir Performance Predictions 11
are not known with any reasonable precision, and the volumetric method may be
used to calculate the initial gas in place (and hence the reserve). However, this
method applies only to the reservoir as a whole, because of the migration of gas
from one part of the reservoir to another.
calculatIOns. The most lIKely source of error IS estlmatmg average reserVOIr'
during the early history period when slight pressure errors
have a significant effect on results. Equations 1.12 and 1.14 may be written as
and
G = _G.J;Pe.
B
..
g
_
Bg  Bgi
(1.21)
(1.22)
Equation 1.21 or 1.22, whichever applies in a particular situation, can be used
to calculate the initial gas in place. If there is no water encroachment, the only
information required is production data, pressure data, gas specific gravity for
obtaining zfactors, and reservoir temperature. However, early in the producing
life of a reservoir, the denominator of the righthand side of the material balance
equation is very small, while the numerator is relatively large. A small change in
the denominator will result in a large discrepancy in the calculated value of initial
gas in place. Therefore. the material balance egpNjon should not be relied on
early in the producing life of the
Unfortunately, this is exactly the period in the producing life of the reservoir
when this information is very desirable. Since the necessary basic information is
available, the temptation to use the material balance equation is quite great.
However, the material balance equation should be judiciously used and with a full
understanding of its weakness. Followin is a e am e calculati . us ating,.,
both the method of using the materIal' a ance equation and its weakness early in
'flle"prodi'itmg life of1ne"reseFvaiF.
Example 1.3
(a) Calculate the initial gas in place in a closed gas reservoir if, after producing
500 MMscf, the reservoir pressure has declined to 2900 psia from an initial
pressure of 3000 psia. Reservoir temperature is 175'F, and the gas gravity is
0.60.
(b) If the reservoir pressure measurement were incorrect and should have been
2800 psia instead of 2900 psia, what would have been the true value of initial
gas in place and what is the percentage error involved?
12 Estimation of Gas Reserves
Solution
(a) Using a gas gravity of 0.60 and referring to the zfactor correlation charts
(Figs. A.l and A.2), z at 3000 psia is computed to be 0.88 and z at 2900 psia is
determined to be 0.87. The next step is to calculate the two values of B
g
:
0.00504z;T
Bg; = '
P;
bbllscf
(1.23)
(Please note Eq. 1.23 is in bbl/scf, whereas Eq. 1.6 is in cu ftlscf; the factor
that differentiates the two equations is 5.615 cu ftlbbl.)
0.005,04 x 0.88 x (175 + 460)
Bg; = 3000 = 0.000,940
0.005,04 x 0.87 x (175 + 460)
Bg2900 = 2900 = 0.000,960
Equation 1.22 is next used to compute initial gas in place:
GpBg (500,000,000) x (0.000,960)
G = "" = = 24,000,000,000 scf
(Bg  BgD 0.009,60  0.000,940 ZV'f"'1
(b) If the pressure measurements were incorrect and the true average pressure is
2800 psia, then the materialbalance equation will be solved using the true
pressure. The zfactor at 2800 psia is determined to be 0.87:
0.005,04 x 0.87 x (175 + 460)
Bg2800 = = 0.000,993
.. 2800
Next, the initial gas in place is calculated by the material balance equation:
GpBg (500,000,000) x (0.000,993) 0
G = = = 9,463,200, 00 scf
Bg  Bg; 0.000,993  0.000,940 . .
Thus, an error of 100 psia, which is only 3.5% of the total reservoir wessure,
resulted in an increase in calculated gas in place of approximately 250%, or a
2Y2fold increase. Note that a similar error in reservoir pressure later in the
producing life of the reservoir will not result in an error as large as that calculated
early in the producing life of the reservoir.
Material balances on volumetric gas reservoirs are simple. Initial gas in place
may be computed from Eq. 1.22 by substituting cumulative gas produced and
appropriate gas formation volume factors at corresponding reservoir pressures
during the history period. If successive calculations at various times during the
history give consistent values for initial gas in place, the reservoir is operating
under volumetric control and computer G is reliable (Fig. 1.1). Once G has been
determined and the absence of water influx established in this fashion, the same
Reseryes and Reseryoir Performance Predictions 13
Water drive _ :.. 
. 





Volumetric reservoir


 Casingleak __



G
p
. cumulative gas produced, sef
Fig.1.1 Variation of G with G
p
equation can be used to make future predictions of cumulative gas production as a
function of reservoir pressure.
Successive application of Eq. 1.22 will normally result in increasing values of
G with time if water (or other fluid) influx is occurring. However, if there is gas
leakage to another zone due to bad cement jobs or casing leaks, the computed
value of G may decrease with time. Equation 1.22 is not applicable in these
cases, and Eq. 1.21 should be used for estimating G.
, If the gas reservoir has there will be two unknowns in the
.,. material balance equation, even though production data, pressure, temperature,
and gas gravity are known. These t",o .. lillkne.wJ1s .. aFtlodnitia asn lace and
cumulativ.e water influx. In order to use the material balance equation to calculate
initial gas in place, some independent method of estimating We> the cumulative
water influx, must be developed. '.
There are two commonly used procedures for estimating We iiiDmmI.
method is from an analysis of production data from wells drilled low on the
structure or from log or drill stem test analyses of structurally low wells drilled
after there has been some gas production and some rise in the gaswater contact. If
the change in the gaswater contact can be determined, then by volumetric
calculations it is possible to determine the water encroachment. Although this
method is often satisfactory in oil
reservoirs because of the fewer
analysis.
iilli&lSp(l"ful,!'Jllethod of estimating water influx is to rearrange the material
balance equation and determine the magnitude of the combination of initial gas in
14 Estimation of Gas Reserres
place plus cumulative water influx at several different times. The amount of gas in
place initially is a constant, regardless of the time or amount of production.
Therefore, a plot of G vs. G
p
must be a horizontal line (Fig. 1.2).
.. However, if Eq. 1.22 is used to ca'lculate G in a reservoir where there is water
influx, the calculated value of G will continue to increase as G
p
increases. This is
because an incorrect material balance equation is being used. Instead of calculat
ing G, the actual calculation is G+ f(W
e
), wheref(W
e
) is some function of water
influx. Rearranging Eq. 1.21 to solve for initial gas in place and cumulative water
influx, .
We
G+ '
Bg  Bgi
GpBg + WpBw
Bg  Bgi (1.24)
At successive' time intervals, the lefthand side of Eq. 1.24 will continue to
increase because of the We/(Bg  B
gi
) term. A plot of several of these values at
successive time intervals is illustrated in Fig. 1.2. Extrapolation of the line formed
by these points back to the point where G p = 0 shows the true value of G, because
when Gp = 0, then We/(Bg  BgD is also zero.
As a matter of economic interest, this technique can be used to estimate the
, value of We. because at any time the difference between the horizontal line (i.e.,
true value of G) and the sloping line [G + (We)/(Bg  BgD] will give the value of
We/(Bg  Bgi)' A knowledge of the magnitude of future water encroachment
into the reservoir will supply some guidance for scheduling production and
income expected from a proposed location.
Points calculated early in the producing life of the reservoir may be subject to
considerable inaccuracies; therefore, more weight should be attached to the point
calculated later in the producing life. Also, handling the different ramifications of
the material balance equation requires strict attention to the units (i.e., whether
bbl/scf or cu ft/scf, etc.) and conditions (i.e., whether reference is to reservoir
conditions or surface conditions, etc.).
,



o ~ __________________________ ___
o
Cumulative gas produced
Fig. 1.2 Effect of water influx on material balance calculations.
Reserves and Reservoir Performance Predictions 15
Material balances for waterdrive reservoirs are complex. Here, the relation
ship must be established between cumulative water influx, reservoir pressure, and
time before gas in place can be computed from performance history. This may be
done with the aid of unsteadystate flow theory. For a linear aquifer, water influx
is proportional to the square root of time and
(1.25)
For other aquifer geometries, a dimensionless time tD must be estimated, then the
dimensionless water influx QD is determined and
(1.26)
Note that the water influx constants C and C' are numerically different. Substitu
ting the unsteadystate water influx into Eq. 1.12 gives
G = G(Bg  Bgi ) + C2./lpQD  WpBw
p B
g
(1.27)
The numerical values of the term S /lp QD at various times during the
history of the reservoir can be determined (for specified aquifer geometries) using
appropriate water influx charts. The successive applications of Eq. 1.27 may be
used to estimate the most probable values of the initial gas in place and water
influx constant. The most probable values are those that, when substituted ,in Eq.
1.27, yield the minimum deviation between computed and actual performance
history. This involves a trialanderror approach wherein graphical methods are
helpful. After G and Chave been evaluated in this fashion, Eq. 1.27 may be used
for performance predictions.
1.4.3 Pressure Decline Curve p/z Method
Equation 1.13 for a volumetric (closed) reservoir may be written:
p
=
PbTGp + Pi
S.61SzbTbVi Zi (1.28) Z
Thus, a graph of P / Z vs. G p will be linear for a volumetric gas reservoir (Fig. 1.3).
The intercept at p/z = 0 gives the initial gas inplace:
The slope is given by
S.61Sz
b
T
b
ViPi
G = ''''''''
1
d
PbTzi
t
16 Estimation of Gas ReserYes
Water drive reservoir
"" .....
...........
Abandonment ......
p ';;a!; driv; I ......
1',
I '
I ............... Closed reservoir
I "
I "
Abandonmentph + .......
        
I I ,
.....
u q,a G
Cumulative gas produced
Fig. 1.3 Pressureproduction graph, gas reservoir.
The term dp / z is the remaining gas in place at any pressure p. Thus, the linear plot
may be extrapolated to give initial gas in place at zero pressure, initial gas reserves
at abandonment pressure, and cumulative gas production at any pressure of
interest.
If water is encroaching, the reservoir hydrocarbon volume is not constant
with time; consequently, a plot of p/z vs. G
p
is not a straight line. Instead, a
waterdrive reservoir normally plots as a curve that is concave upward (Fig. 1.3).
Because of water influx, the pressure drops less rapidly with production than
under volumetric control.
After a reasonable amount of gas has been produced (about 20% of the
reserve), the p/z vs. cumulative straightline plot for a volumetric (closed) reser
voir provides a satisfactory procedure for estimating recoverable gas. It must be
cautioned that if pressure alone (instead of p/z) is plotted against cumulative
gas production, the resulting graph is not linear, and extrapolations from this
pressureproduction curve may be in considerable error (Fig. 1.4).
Equation 1.28 can be written as
G = PJZi G
PJZi  p/z p (1.29)
Equation 1.29 is basis of the widely used pressurevolume and equal pound loss
methods (AGA). It merely considers two points on the pressuredecline curve to
calculate reserve initially in place and, depending on accuracy of individual data
points, may lead to conflicting results.
Taking logarithms of Eq. 1.29 and rearranging,
(p/z)
10g(pJzi  p/z) = log G
p
+ log ''
G (1.30)
Reseryes and Reseryoir Performance Predictions 17
5000rrrrrrr
Ul
'"
w
""
ct: '"
 0
'"'
N
"c
.>
U
is
'"
0>
",",
0.
. 0
.=
lJ, "'.
0_'"
:li!
z;
c
!
O .....
w
'" u..
2 3 4 5 6 7
'"
Cumulative production, MMM scf
Fig. 1.4 Comparison of theoretical values of p and plz plotted against cumulative
production from a volumetric gas reservoir.
Equation 1.30 is an example of a method in which cumulative pressure drop is
plotted against cumulative production on logarithmic coordinates. It states that
the difference in initial p / z and successive values plotted against corresponding
cumulative production on logarithmic coordinates forms a straight line at a 45'
angle with either coordinate.
Although Eqs. 1.28,1.29, and 1.30 apparently present different methods of
estimating gas reserves, they are essentially identical. Since each has individual
advantages in interpretation of results, these will be discussed in connection with
examples.
Example 1.4 (after"AGA). Use pressures and cumulative production data in
Table 1.2 to estimate recoverable gas reserves initially in place in a gas reservoir.
Solution
The data presented in Table 1.2 are plotted in Fig. 1.5, where line 1 has been
drawn through a plot of p / Z against cumulative production and extrapolated to
zero pressure, indicating an initial reserve of 48.3 billion cubic feet (MMMcf) of
gas in the reservoir. Line 1 also shows that there must have been an error in either
initial reservoir pressure or cumulative production to the first interval. Actually,
cumulative production includes estimated production from a wild well. A study of
Fig. 1.5 indicates this estimated production may have been too large. Curve 2, a
plot of reservoir pressure against cumulative production, shows curvature result
ing from a neglect of gas deviation factor. Curve 3 illustrates a possible erroneous
extrapolation of reservoir pressures taken during early field life, neglecting gas
deviation factor.
Use of Eq. 1.29 for estimating gas reserves initially in place is illustrated in
Table 1.3. The quantities given in Columns 1 and 2 were substituted into Eq. 1.29
18 Estimation of Gas Reseryes
TABLE 1.2
Reservoir Pressures and Cumulative Production from a Gas Reservoir
Cumulative
Reservoir Gas deviation production at
pressure, factor z, 14.4 psiaand GOF,
psia dimensionless plz MMMcf
2080 0.759 2740 0
1885 0.767 2458 6.873
1620 0.787 2058 14.002
1205 0.828 1455 23.687
888 0.866 1025 31.009
645 0.900 717 36.207
Source:' After AGA.
3000
'\.
'\
2500
N
pressure divided by
compressibility factor, ph.
2000
"
" '\I,
CD
1000
'"
500
o
o
r'\.
pressure
0
I'b.
Possible erroneous
extrapoiation of
, pressures
\ "
'\ ...... ,
10 20 30 40 50 60
Cumulative production, billion cu ft
Flg.1.5 Pressuredecline curves for a natural gas reservoir. (After AGA.)
to compute initial gas reserves in place in the reservoir and were given in Column 3
for successive values of p/z. At each successive calculation, the apparent initial
reserve in place decreased, as shown in Column 3. Previously (see Fig. 1.5), a
discrepancy was noted in these data, so initial reservoir pressure and cumulative
production to the first pressure after initial were disregarded (Column 4) to
compute an adjusted reserve shown in Column 5. Initial reserve in place is then
I
b
Reserves and Reseryoir Performance Predictions 19
TABLE 1.3
Estimatiori of Initial Gas Reserves in Place Using Eq. 1.29
Volumes in billions of cubic feet
(1 ) (2) (3) (4) (5)
Initial Adjusted Adjusted
Cumulative reserve cumulative reserve
production plz in place production in place
0 2740
6.873 2458 66.8 0
14.002 2058 56.3 7.129 43.8
23.687 1455 50.5 16.814 41.2
31.009 1025 49.5 24.136 41.4
36.207 717 49.0 29.334 41.4
Source: After AGA.
(G)
(5) + 6.9,
Initial
reserve
in place
50.7
48.1
48.3
48.3
computed by adding disregarded cumulative production of 6.9 MMMcf to each
adjusted reserve estimate in Column 5 to obtain Column 6. These reserve values
in Column 6 are consistent and identical with the value of 48.3 MMMcf indicated
by Fig. 1.5, where p/z was plotted against cumulative production. If a water drive
were present, the initial reserve estimate would apparently increase with each
successive computation.
Data necessary for use in Eq. 1.30 in estimating recoverable reserves for the
gas reservoir described above are given in Table 1.4. The actual method is
illustrated in Fig. 1.6. Values in Column 3 of Table 1.4, the change in p/z, or
(pJZi  p/z), are plotted on logarithmic coordinates with cumulative gas pro
duction from the reservoir as abscissa to give the dashed curve of Fig. 1.6. The
curvature, contrary to Eq. 1.30, results from data discrepancy previously ex
TABLE 1.4
Data for Estimation of Initial Gas Reserves in Place Using Eq. 1.30
(1 ) (2)
Cumulative
production,
MMMcf plz
0 2740
6.873 2458
14.002 2058
23.687 1455
31.009 1025
36.207 717
Source: After AGA.
(3)
Pi p
Zi z
282
682
1285
1715
2023
(4)
Adjusted
cumulative
production,
MMMcf
0
7.129
16.814
24.136
29.334
(5)
Adjusted
Pi P
Zi Z
400
1003
1433
1741
5000
N
;;.
! 1000
.
'k
200
5
/
1/
20 Estimation of Gas Reserves
l,.
I/
/
/
j;:t
vi
1(/
,
/'
/'
/'
./
Actual
1/
10 50 100
Cumulative production, MMcf
Fig.1.6 Change in plz with cumulative production for natural gas reservoir. (After
AGA.)
plained. By neglecting initial reservoir pressure and cumulative production to the
first pressure after initial, this discrepancy may be eliminated. Adjusted values of
cumulative production and change in p / z are given in Columns 4 and 5 of Table
104. These data are shown by the solid line of Fig. 1.6 that forms a 46' angk with
the abscissa. The reserve in place whenp/z was 2458 is 41.4 MMMcf as read from
the abscissa where the change in p/z is 2458. Adding the neglected production of
6.9 MMMcf, the initial reserve in place is 48.3 MMMcf. Thus, Eqs. 1.28,1.29, and
1.30 can be used to obtain identical res].!lts.
1.4.4 MBE Straightline Method (after Havlena and Odeh)
Rearranging the materialbalance equations (MBEs) as follows, volumetric MBE
(Eq. 1.14) becomes
y = a +b
\,1
1.1
(1.31) x
Waterdrive MBE (Eq. 1.12) becomes
CpB" + WpBw = C 'LQnt!.p + C
B"  Bui B<  B"i
y a x + b (1.32)
Reserves and Reservoir Performance Predictions 21
t
o Bg  Bgi ..
Fig.1.7 MBE straightline plot, volumetric gas reservoir.
I.e ' '.
where
We = t!.p
C = water influx constant
Equation 1.31 is illustrated in Fig. 1.7. A graph of CpBg vs. Bg  Bgi is a .
straight line passing through the origin with _a ,lope numerically equal to C:
The relationship in Eq. 1.32 is illustrated by Fig. 1.8. A graph of CpBg +
WpBw/ Bg  Bgi vs. 'LQD t!.p/ Bg  Bgi yields a straight line, provided the un
steadystate water influx summation, 'LQD t!.p, is accurately computed. The
resulting straight line intersects yaxis at the initial gas in has a
equal to the water influx constant, C;. .
r
o
' , , Nonlinear plots will result if the aquifer is improperly characterIzed. A
systematic upward or downward curvature suggests that th.e IS
too small or too large, respectively., while an Sshaped curve IndIcates that ahnear"
(instead of a radial) aquifer should be assumed. The points plot sequen
tially from left to right. A reversal of thIS plottIng sequence IndIcates that 'an
unaccounted aquifer boundary has been reached and that a smaller aqUIfer should
be assumed in computing the water influx term.
Example 1.5. VDlumetric Gas Reservoir
<I> = 0.13
Sg = (1  Swi) = 0.48
kg = 9.0 md
Bgi = 0.001,52 res bbl/scf
C = 0.607 (air = 1.00)
Zb = 1.00
A = 1060 acres
I.
22 Estimation of Gas Reserves
h = 54 ft
T = 164F = 624R (reservoir temperature)
T = 116F = 576R (mean flowing temperature in tubing)
p pc = 672 psia
Tpc = 365R
H = 5322 ft (well depth)
rw = 0.276 ft
re = 2980 ft
Di = 2.5 in. (internal diameter of tubing)
z vs. P, /Jg vs. p, and Bg vs. p are shown on Figs. 1.9 and 1.10.
Performance History
Time, t
years
Reservoir pressure, p
psia
Cumulative Production, G
p
MMMscf
0.0
0.5
1.0
1.5
2.0
1798
1680
1540
1428
1335
0.00
0.96
2.12
3.21
3.92
,:::::::.."'i"T., Assume
.L. Linear aquifer
'I;l1p QD Too large
:El1p QD
BgBgi
Fig.1.8 MBE straightline plot, gas reservoir with water influx.
I
r
I
Reserves and Reservoir Performance Predictions 23
1.00
'"
""
/
,./
0.98
0.96
0.017
0.016
0.015
0.94
"
h k:::'
0.0'4 ,f:o
,
"'"
I
V
i
."
o
k' /'
/'
"
"
0.92
0.013
0.90
0.88
0.86
o
/
,
V
"
""
0.012
0.011
0.010
_ _ _ _ , , , , ,
Reservoir pressure, p, psia
"
<3
Fig.1.9 'Gas deviation factor and viscosity versus reservoir pressure for Example 1.5.
Solution
(a) Volumetric estimates
1
G = 7758Ah<l>(1  Swi) 
. Bgi
= (7758)(1060)(54)(0.13)(0.48) 1 = 18.23MMMscf
0.001,52
Assuming an abandonment pressure of 200 psia, the gas formation volume
factor at abandonment Bga is read as 0.016 res bbllscf from Fig. 1.10.
G = 7758Ah"'(1  S .) __ 1 )
pa '+' WI B. B
. .
= (7758)(1060)(54)(0.13)(0.48) ( 1 __ 1_) =
0.001,52 0.016
= cumulative gas produced at abandonment (recoverable gas in place
at original conditions)
0,)
Predicted recovery = J!f% of initial gas in place.
(b)
24 Estimation of Gas Reseryes
Materialbalance estimates (Eq. 1.14)
(1)
t, years
0.0
0.5
1.0
1.5
2.0
o. 1
0.08
DOG
0.04
0.02
~
j
0.0 1
5 0.008
g
o
o 0.006
. ~
E
~
"
0.004
0.002
0.00 1
(2)
psia
1798
1680
1540
1428
1335
\
,
1\
o
(3) (4) (5)
B
g
, ~ G
p
,
Resbbl/scf Bg  Bg; MMMscf
0.001,52 . r.:,.
0.00
0.001,63 14.82 0.96
0.001,79 . 6.63 2.12
0.001,96 4.45 3.21
0.002,10 3.62 3.92
"
"\.
"
'"
"....
.............
Reservoir pressure, p, psia
(6)
G[(4) x (5)],
MMMscf
14.2
14.1
14.3
14.2
r
Fig. 1.10 Gas formation volume factor versus reservoir pressure for Example 1.5.
Reserves and Reservoir Performance Predictions 25
'/' .1".
. ,
The computed values of G average 14.2 billion standard cubic feet
(MMMscf)about 22% less than the volumetric estimate. The consistency
is an indication that the reservoir is operating under volumetric control and
that the volumetric estimates are optimistic.
. 1. Pressure decline curve (p/z) method. A graph of p/z vs. G
p
is shown (Fig.
1.11). The best straight line through the history points is extrapolated to give
G = 14.2MMMscfatp/z = O. This checks the average value computed from
Eq. 1.14. The linearity suggests that water influx is negligible (volumetric
control).
At Pa = 200 psia, Za = 0.979 and Pa/za = 200/0.979 = 204. From p/z
vs. G
p
curve, G
p
at abandonment (recoverable gas in place at original
condition is 12.8 MMMscf.
2. MBE Straightline method. Graph of GpBg vs. Bg  Bgi is shown (Fig. 1.12).
The best straight line passing through the origin again gives G = 14.2
MMMscf. The slope G is computed as follows:
8.52 million res bbl
G = = 14.2 MMMscf
0.6 res bbl/Mscf
2200
\
~
\
2000
1800
1600
\
1400
1200
\
\
1000
\.
\
800
60 0
1\
0
'I
40
\.
G"' 14.2 MMMscf
0
\
)
0
20
o 4 10 16 18 20
8 12 14 2 6
Gpo MMMscf
Fig. 1.11 Pressuredecline curve gas reservoir for Example 1.5.
26 Estimation of Gas Reserves
0
8
./
V
/
6
,
V,.2MMMSCf
4
V
./'
2 V
oV
o 0.1 0.2 0.3 0.4 0.5 0.6
Bg  Bgi. res bbl / Mscf
Fig. 1.12 MBE straightline plot gas reservoir for Example 1.5.
Example 1.6. Gas Reservoir with Water Influx. Reservoir and rock properties:
Areal extent, A = 8870 acres
Formation thickness, h = 32.5 ft
Porosity, <I> = 30.8%
Interstitial water saturation, Swi = 42.5%
Gas deviation factor at standard conditions, Zb = 1.0
The performance history with corresponding values of Bg and p/z is given:
History
Cum. Gas
Time, Prod., Cp,. Res. Press. p, Gas deviation GasFVF,B
g
,
years MMMscf psia factor, z p/z Res bbl/scf
0 0 2333 0.882 2645 0.001,172
2 2,305 2321 0.883 2629 0.001,180
4 20,257 2203 0.884 2492 0.001,244
6 49,719 2028 0.888 2284 0.001,358
8 80,134 1854 0.894 2074 0.001,496
10 105,930 1711 0.899 1903 0.001,630
12 135,350 1531 0.907 1688 0.001,820
14 157,110 1418 0.912 1555 0.001,995
16 178,300 1306 0.921 1418 0.002,187
18 192,089 1227 0.922 1331 0.002,330
20 205,744 1153 0.928 1242 0.002,495
),'
i
r
i
,
I
,
V
Reserves and Reservoir Performance Predictions 27
Solution
(a) Volumetric equation
Initial gas in place:
1
G = 7753Ah<l>(1  Swi) 
Bgi
= (7758)(8870)(32.5)(0.308)(1  0.425)
= 337.9:MMMscf
(b) Material balance estimates
(1) Pressure decline curve pjz method
A plot ofp/z vs. cumulative gas produced is shown in Fig. 1.13. An upward
curvature indicates water encroachment.Thus, thep /Z method cannot be
estimate gas in place for this reservoir, Notice that the first 10 years of history plot
'as a straight line. It is only after 10""years that the graph indicates a noticeable
upward curvature to signify water influx. This plot clearly shows the insensitivity
of pressuredecline p / Z curvesor materialbalance methods in generalin de
tecting encroaching water early in the life of a natural gas reservoir. The close
approach to linearity during the first 10 years might have been incorrectly as
sumed to indicate a volumetric reservoir (no water influx), This would have given
an erroneous value of initial gas in place (G = 370 MMMscf) from an extrap
olation of the straight line to p/z = Osome 32 MMMscf higher than
metric estimate.
2800
""
K
2400
'"
2000
1600
1200
.....
"
f'. G = 370 MMMscf
Wrong extrapolation
.....
800
400
.....
.....
.I
40 80 120 160 200 240 280 320 360 400
G
p
MMMscf
Fig.1.13 Pressuredecline curve for Example 1.6.
28 Estimation of Gas Reseryes
(2) Calculation of water influx, assuming a known value of gas in place,
... cumulative
water influx W, at the end of four years is calculated by means of the following
rearranged form of Eq. 1.21:
W, = GpB.  G(Bg  B
g
,)
= (20,257)(0.001,244)  337,900(0.001,244  0.001,172)
= 0.87 million res bbl
Calculations of W, at 4year intervals throughout the production history of
the reservoir are given in Table 1.5.
(3) MBE as straight line
Since the aquifer geometry as well ..as its rock and fluid properties are
unknown an infinitely large linear aquifer is assumed for the first trial. The MBE
straightline form ofEq. 1.27 is solved and the results are plotted to determine if a
linear solution is obtained. The assumption of an infinitely large linear aquifer
leads to the use of Eq. 1.25. The calculations are summarized in Table 1.6 and the
results are plotted in Fig. 1.14. A linear plot is obtained, suggesting a satisfactory
solution. Thus, it is unnecessary to try other aquifer geometries for this example.
From the best straight line through the computed points,
G = 304 MMMscf(yintercept)
__ __ _
slope C' = :
(2.5  O)million psi(yr)ll2scf/res bbl
= 30,800 res bbUpsi(yr)1I2
TABLE 1.5
MaterialBalance Calculations of water Influx (G = 337.9 MMMscf)
(1 ) (2) (3) (4) (5) (6) (7)
GpBg G(B
g
 B
g
,) w,
Bg (2) x (3)
Bg  Bg; 337.9 x (5) (4)  (6)
t G
p Res Million Res Million Million
Years MMMscf bbllMscf res bbl bbl/Mscf res bbl res bbl
0 0 1.172
4 20.257 1.244 25.20 0.072 24.33 0.87 V
8 80.134 1.496 119.88 0.324 109.48 10.40
12 135.350 1.820 246.34 0.648 218.96 27.38
16 178.300 2.187 389.94 1.015 342.97 46.97
20 205.744 2.495 513.33 1.323 447.04 66.29
.1
I
c:
0
'';::;
oj
:J
0..
E
0
,...:U
....
Q)
..J c:
eo .
<co'
""
00
'
iJ)
w
Ct:l
Reserl'es and Reseryoir Performance Predictions 29
:;;
..0
"
' cr
tQlIIlS
I 'L:
<f "c
<1
C:Q .!.
Vol 0.
C
.
'L:
C
0:
<1"
Vol
f
%
.. 
<1"
_r:.C::!

_.
I
>
.. 
,
' 'ti
","
","
",'" ::;:

ttl l"Jo. v
t.:l
.. ::;:
E 0 ::;:
:::I e
u"

'"
N
""
'"
00
'"
'"
ON
""
'"
000 NrO
'"
()!
ON
0,,"
0,,"
Nro
N
,,"NO
l
00 OOLO'
u:i:2
u:iui..c
..,""
'" ci
N
'"
'"
co
'"
'"
o::tc5t.:o
'"
",ON
"'0,,"
.
NNrO
'1.0 V N 0
00 LOLO
..cui..coo:r
..,"""
'rCOC"lCOCOC"lr
N"<t"''I1')I1')I.OI'.
NNNNNNN
I'.NMI'.'rOO
C"ICOI1')COI1')O"<t"
ui...tMNNuiO
'rMrNNrC"l
I'.rI.OrI.OrLO
rrNNMM
0000000
0000000
0000000
.;::;
LOLOLOLOLOLOQ)
11')1.O"<t"rC"l
r 'r r r
rC"l"<t"rMrl'.
COOMI1')I'.C"IO
rr'rrrN
00 N "<t" 0 N "<t" 1.0 0 N_"<t" 1.0 co 00
i=
I N
30 Estimation of Gas Reserves
Thus the cumulative water influx is given by
W, = C'L !J.pn = CL !J.p QD
After 4 years,
We = (30,800)(103.92) = 3.20 million res bbl
Calculations of We at 4year intervals throughout the production history of the
reservoir are shown below.
Time, t
!J.pn vt=t;;
We
years psi(yr) 112
million res bbl
4
103.92
3.20
8
715.97
22.05
12
1,613.53
49.70
16
2,622.51
80.77
20
3,590.40
110.58
The. estimated value of gas initially in place from the MBE straightline
method IS smaller than the volumetric estimate. The amount of water influx on
other hand, is than that calculated from the materialbalance equaiion
usmg the volumetnc estimate of gas initially in place.
40 0
38 0 z!
/'
V
0
/
/'
0
/G  304 MMMscf
./
C' == 30,800
V
320
0.5 1.0 1.5 2.0 2.5 3.0
P" ytt,,MM psi  (yrP'> scf/res bb!
Bg Bgi
Fig.1.14 MBE straightline plot for Example 1.6.
r
I
Ii
,
Reserves and Reservoir Performance Predictions 31
1.4.5 ReservQir Size
Sometimes it might be desirable to determine the areal extent of a gas reservoir.
This information is particularly advantageous when a well has been drilled into a
new gas reservoir and it is necessary to know the size of the reservoir in order to
decide whether additional wells can, or should, be proposed for drilling. The
technique uses the materialbalance equation in conjunction with the volumetric
equation, and it has the same limitation as any other application of the material
balance, that is, the accuracy increases as more production data become available.
Combining Eq. 1.1 (after proper adjustments for appropriate units) and Eq. 1.14,
the areal extent of the reservoir is given by
GpBgBgi
A=
7758h<!>(1  SWi)(Bg  Bg;) (1.33)
where
A = areal extent of the reservoir, acres
h = thickness of the reservoir, ft
All other terms have been defined in previous sections.
Example 1.7. One well has been drilled into a reservoir. Geologic data indicate
net pay thickness is 30 ft, porosity is 16%, and interstitial water saturation is 30%.
The initial reservoir pressure was 2100 psi a and formation temperature is 175F.
After producing 450 MMscf of 0.70gravity gas, pressure had declined to 1500
psia. Some indication of the size of the gas reservoir is necessary to decide whether
additional wells should be drilled.
Solution
zfactors are first calculated at the two pressure points using Figs. A.1 and A.2. At
2100 psia, z = 0.842; at1500psia, z = 0.869. Gas formation volume factors atthe
two pressUres are next calculated by using Eq. 1.23:
. = 0.005,04Tzi = (0.005,04)(175 + 460)(0.842) = 0.001283
B
g
, Pi 2100 '
 (0.005,04)(175 + 460)(0.869) = 0 00 53 bbll f
Bg1500  1,8 sc
1500
Sufficient data are now available to calculate A using Eq. 1.33:
_
____ __
A=
7758h<!>(1  SWi)(Bg  Bgi)
(450,000,000)(0.001,853)(0.001,283) = 72 acres
(7758)(30)(0.16)(1  0.30)(0.001,853  0.001,283)
32 Estimation of Gas Reseryes
Because of the small amount of acreage involved, another well would probably
not be justified. This problem may be solved using Eq. 1.7:
G = 7758Ah""(1  S .) (_1
P 'f' IVlB.B
g' g
Or
A = Gp / [7758h<l>(1  Sw;) 
Bg, Bg
450,000,000
,::,...,.,:;.,,..,.'''':..:..:..,. = 72 acres
(7758)(30)(0.16)(0.70)(779.4  539.7)
1.4.6 Rate Versus Time
Previous sections discussed how to calculate the total recovery from a gas reser
voir. Equally important is the necessity of relating future producing rates to time ..
Following is a method of determining rates of future gas production. The two
pieces of data needed are p/z vs. cumulative gas production and backpressure
test data to determine the productivity of the gas well.
If the p/z vs. cumulative gas production data are not available, then some
$;.:  . ._._.      ."' _.' " ... '", ...  ._ . , . .,
reasonable estimatEiinist De developea. This is relatively simple for a closed gas
reservoir (no water influx) because the plot is a straight line and two points on this
line can generally be ascertained. The first poi!!! that can usually be determined is
the initial pressure, at which point the cumulative gas production is zero.
second point that can also be determined is the total gas initially in place, which
would be equal to the total gas produced to zero pressure (see Fig. 1.3).
A typical backpressure graph is shown in Fig. 1.15. Theoretically, a plot of
PR
2
 Pw/ vs. q on loglog paper should plot as a straight line with a slope
ranging from 1.0 to 2.0 where
PR = reservoir pressure (shutin BHP), psia
Pwf = flowing sandface pressure, psia
q = flow rate, Mcfd
A stepbystep procedure for relating gas producing rate to time is outlined
below.
1. Draw a graph of p/z vs. cumulative gas production.
2. Plot a backpressure test data.
3. Arbitrarily select a value of p and, from the decline curve, read the cumula
tive gas produced (note the P Rand p are identical). In order to increase the
accuracy of the calculations, use small pressure steps.
4. At the selected value of p, determine the theoretical open flow potential. If
the contract producing rate does not exceed the allowable producing rate
(usually Y4 of the open flow potential), then use the contract rate for this time
interval. If the contract producing rate exceeds the allowable producing rate,
then the allowable producing rate must be used for the time interval.
r
i
,
Reserves and Reservoir Performance Predictions 33
5. Compute the time required to produce the gas during the first interval by
time = gas produced during intervallq average.
6. Repeat steps 2 to 5 for consecutively lower values of p until the abandonment
pressure is reached.
7. Prepare a ratetime plot (or tabulation) from the above calculation.
S. Use the ratetime plot (or the tabulation) as an analog for the proposed well.
1.4.7 Liquid Recovery
After the recoverable gas reserves have been calculated, the next step is a
determination of the amount of liquids that will be recovered at the surface, a fact
that must be considered in an economic evaluation for a proposed project.
Quite often, the value of the liquids recovered from the gas may exceed the
value of the gas itself. This is particularly true in the case of condensate reservoirs,
where the percentage of heavier hydrocarbons is relatively high. If there is a
considerable quantity of liquefiable hydrocarbons in the gas, it may be economic
to install liquid recovery units. If this is done, then the recoverable gas volume
x
20
10
8
6
4
2
"a
""
o.a
I l.
0
0.8
o. 6
o. 4
o. 2
o. 1
/
1/
p
w
[ 14.7 psia
V
/
//
/
1/
Calculated open flow
V
L potential ,I I
62,000 Mcf/day
/
2 4 6 8 10 20 30 40 50 60 80 100 200
q, Mcf/day x 10
3
Fig.I.IS Typical gas well backpressure test.
34 Estimation of Gas ReseITes
must be reduced by the amount of gas volume converted to liquids. This reduction
m volume IS often referred to as plant shrinkage.
Plant shrinkage will.vary from about 8 to 13%, with 10% being the average.
Before applymg the shnnkage factor, the recoverable gas is usually known as
recoverable wet gas; after the plant shrinkage factor has been applied, the terms
frequently used are reserves, recoverable dry reserves, or simply residue
gas. The term wet m thIS case means that the heavier hydrocarbons have not been
extracted; the term dry indicates that some of the heavier hydrocarbons have been
removed by some type of processing facility.
If the gas is not a condensate reservoir, the composition of the gas in
the reservOIr wIll never change and the liquid recovery will remain constant. If the
is a condensate liquid dropout will occur when the pressure
declmes to the dew pomt m the reservoir. Fig. 1.16 illustrates the variation of
liquidgas ratio (bbI/MMcf) for a gas reservoir and a condensate reservoir.
A laboratory of a gas sample is necessary to determine precisely
whether the reserVOIr IS a gas reserVOIr or a condensate reservoir. However, a rule
of thumb can be quite helpful to estimate the type of reservoir before laboratory
data are aVaIlable. If the liquidgas ratio exceeds 20 bbllMMscf and initial reser
voir pn.'ssure exceeds 5000 psia, then the reservoir is probably a condensate
reservOIr. In the absence of a laboratory or field data, Fig. 1.17 can be used to
estImate barrels of liquid per MMscf of gas.
1.4.8 Associated Gas Reserves
The initial gas reserves in contact with crude oil in the reservoir or the initial
quantity of gas in a gas cap can be readily determined by the volumetric method
(?q. 1.1), provided the gasoil contact position can be fixed. Subsequent estima
tIons of gas reserve in the cap are difficult to make after production of oil and
from the reservoir. Gas from the gas cap may have migrated into the oil zone,
dIssolved gas may have migrated into the cap, and some wells may have produced
"t
"
:0
D
0
m
I
.,
cr
:J
40
30
20
10
0
I
I
I Condensate reservoir
I
I
I
___ 1 __ _______ _

Pressure, psig
o
Fig.l.16 Variation of liquidgas ratio for a gas reservoir and a condensate reservoir.
Reseryes and Reservoir Performance Predictions 35
1.6
1.5
/
/
7
;:: II 1.4
11 .:
.!::
e 1.3
, m
/0
'0
7
1.1
1.0
o 20 40 60 80
Bbl condensate/MMscf gas
V
/
V
100 120
140
Fig.I.l7 Gravity ratio versus stocktank yield of condensate. (After Rsaza and Katz.)
a mixture of associated and dissolved gas with crude oil. The problem of estima
tion is simplified if no effort is made to distinguish between associated and dis
solved gas (Gray and Crichton). The total initial gas reserve may be computed and
the cumulative production of associated and dissolved gas may be subtracted
to obtain total gas reserves remaining at any time. also pressuredecline methods
are inaccurate for estimating gas remaining in a gas cap because volume of
reservoir occupied by the cap will remain constant only under exceptional con
ditions.
1.4.9 Dissolved Gas Reserves
The principal method of estimating dissolved gas reserves is the volumetric
method. The initial dissolved gas reserve in a reservoir is computed by estimating
initial reserve of stocktank oil and multiplying this oil volume by the initial
solution gasoil ratio. Recoverable dissolved gas reserve is determined by apply
ing a recovery factor that varies according to reservoir characteristics and mecha
nism by which the reservoir is produced.
In strong waterdrive reservoirs, where pressures remain above saturation
pressure or bubble point of the oil throughout producing life of the reservoir, the
recovery factor for dissolved gas will be the same as that for oil. If reservoir
pressure is reduced below the saturation pressure of reservoir oil, the recovery
factor for dissolved gas will be greater than that for oil. This results from increas
36 Estimation of Gas Reserves
ing relative permeability of porous media to gas and decreasing relative perme
ability to oil as gas saturation increases.
1.5 GENERAL MATERIALBALANCE EQUATION
The general materialbalance equation may be written as
NBti G
N(B,  B'i) + 1 _ Swi (CwSwi + Cj)(Pio  Po) + 5.615 (Bgc  Bg,)
GB,i
+ 5.615(1  Swi) (CwSwi + Cj)(Pig  pg)
+ ' g r W + WB = N B + P' "g
GB 1 [ (R  R)B ]
5.615 ' 'w 1" 5.615
GpcBgc
+ + W B
5.615 p w
(1.34)
where
N = STB originally in the oil zone
B, = Bo + (R'i  R,)/5.615 = the total volume factor
Swi = the original water saturation
C
w
= compressibility of water, psi
1
Cj = compressibility of formation, psi
l
Pio = original mean pressure in oil zone
Po = pressure in oil zone after producing Np bbl
G = original standard cubic feet in free gas zone
Bgc = gas volume factor for gas cap
Bgi = original gas volume factor
Pig = original mean pressure in gas zone
pg = mean pressure in gas zone after producing G
pc
scf
G
i
= cumulative standard cubic feet of gas injected
B/ = volume factor of injected gas
We = cumulative reservoir barrels of water influx into the volume orig
inally containing N Or G
Wi = the cumulative barrels of water injected
Bw = formation volume factor of water
Np = cumulative STB of oil produced
Rp, = Gp,/ Np = cumulative produced gasoil ratio, scf/STB
R'i = original standard cubic feet of gas in solution in 1 bbl of STO
R, = cubic feet of gas still in solution in 1 bbl of stock tank oil (STO)
J3g = formation volume factor of the evolved solution gas
G
pc
= the cumulative standard cubic feet of gas produced from the gas
cap
WI' = cumulative barrels of water produced
r
I
Ii
I
General MaterialBalance Equation 37
The restrictions of the materialbalance equation include:
1. Point solutionindependent of path.
2. Requires average (volumetric) pressure for B" Bg, Po> Pg, Bb" R" and Pwoc
for We'
3. Requires separation of produced gas into solution gas and gas cap gas.
4. Requires all free gas produced from the oil zone to be gas at the surface.
S. Requires that no free liquid be produced from the gas cap.
6. Requires good recordkeeping.
Some of the advantages are:
1. Permits the determination of unseen volumes with data obtainable at the
surface.
2. Permits solving for two unknowns simultaneously.
3. Permits a sequential analysis with simpler equations.
1.5.1 Gas Reservoirs
The general materialbalance equation reduces to the conventional gas material
balance equation if Nand Np are set equal to zero:
G{Bg  B,,[l  (CwSwi + Cj)(Pi  p)/(l = Sw,)]} _
+ We X 5.615 = GpBg + WpBw X 5.615 (1.35)
where
Qp = gas equivalent of the gas and liquid hydrocarbons on the surface
WI' = free water from the reservoir
If A I:;,(Tb/T,Pb) and all terms in Eq. 1.35 are multiplied by A, it reduces to
{
z Zi }
G    [1  (CwSwi + Cf)(Pi  p)/(l  Swill
P Pi
 Z 
+ AWe X 5.615 = G
p
 A WpBw X 5.615
P
If We = 0 and Wp = 0, then Eq. 1.36 reduces t6
Or
P
Z
Z Zi Z G
p
 =  [1  (CwSwi + Cf)(Pi  p)/(l  Swill +  G
P 0 P
(1.36)
(1.37)
(1.38)
38 Estimation of Gas ReseITes
The latter form is probably the more familiar, especially when the denominator of
the righthand term is reduced to the constant 1, yielding
Pi Pi G
p
p/Z =   
Zj Zj G
(1.39)
This equation has the following limitations: constant reservoir temperature,
no phase change in the reservoir, no water influx, no rock compaction, and no
connate water expansion or evaporization. The equation results in a natural
linearization of the gas material balance and hence is usually plotted as shown in
Fig. 1.18.
If, as is the real case, the formation compressible is not zero, then a plot of
p/z vs. Gp should not be a straight line. One can solve both gas equations for G
and determine the error introduced by omitting the connate water and rock com
paction terms:
(1.40)
When the formation and connate water volume changes are omitted, the error in
the results is given by
Or
_
G...:a",p,,:p::_G...:a:;:,c( x
Percent error =
G
act
100
(CwSwi + cf)(p  p)/(l  S .)
Percent error = ' w, x 100
(PiZ/PZ;)  1
PilZ;
Pi
m=
ziG
G
p
Fig.1.IB p/z versus G
p
.
(1.42)
(1.43)
General MaterialBalance Equation 39
It is very evident that the error increases as the magnitude of C
w
and C f
increase; however, it is not so evident that the error is at a maximum at the higher
values of pressure and decreases as the pressure in the reservoir is decreased. The
following example illustrates the variation of the error with pressure.
Example 1.8
Then
Pi = 10,000 psia
C
w
= 3 X 10
6
Swi = 0.25
cf = 4 X 10
6
Z = Zj
6.33 x 10
4
(10,000  p)
Percent error = 10,000
1
P
Pressure
9900
9500
9000
8000
7000
6000
5000
Percent Error
Z=Z;
6.26
6.01
5.70
5.06
4.43
3.80
3.17
The introduction of Z into the relationship increases the error, as Z/Zi will be
less than 1 and hence decrease the value of the denominator. If the reservoir
temperature is 250F and the gas has a gravity of 0.65 the error would be as shown
below. The p/z vs. G
p
plots are shown in Fig. 1.19.
Pressure zlz; Percent r r o ~
9900 0.994 15.67
9500 0.967 17.69
9000 0.940 14.24
8000 0.890 11.25
7000 0.829 10.30
6000 0.771 8.88
5000 0.714 7.39
40 Estimation of Gas Resel1'es
G
p
Fig. 1.19 Gas plot showing error.
1.5.2 Abnormally Pressured Gas Reservoirs (after Hammer/indl)
In abnormally highpressure, depletion type gas reservoirs two distinct slopes are
evident when a p / z vs. cumulative plot is used to predict reserves because of
formation and fluid compressibility effects (see Figs. 1.19 and 1.20). The final
slope of the p/z plot is steeper than the initial slope; consequently, reserve
estimates based on the early life portion of the curve are erroneously high. The
initial slope is due to gas expansion and significant pressure maintenance brought
about by formation compaction, crystal expansion, and water expansion. At
approximately normal pressure gradient, the formation compaction is essentially
complete and the reservoir assumes the characteristics of a normal gas expansion
reservoir. This accounts for the second slope. Most early decisions are made based
on the early life extrapolation of the p/z plot; therefore, the effects of hydrocar
bon pore volume change on reserve estimates, productivity, and abandonment
pressure must be understood.
All gas reservoir performance is related to effective compressibility, not gas
compressibility. \Vhen the pressure is abnormal and high, effective compressibil
[
"
l:!
!J
i
"
I
e.
I
I
b
!!
"
62
60
,
58
56
54
52
50
48
46
o
General MaterialBalance Equation 41
... Extrapolates
to 220 MMMcf
....
"\.\'
\\
......
......
......
Corrected estimate /'
,\,
from early life performance
indicates 114 MMMcf .
I
ExtrapOllates to 118
10 20 30 40
Gas production, (MMMcfl
Fig. 1.20 p/z versus cumulative production. North Ossum Field, Lafayette Parish,
Louisiana NS2B Reservoir. (After HammerlindL)
ity may equal two or more times that of gas compressibility. If effective compres
sibility is equal to twice the gas compressibility, then the first cubic foot of gas
produced is due to 50% gas expansion and 50% formation compressibility and
water expansion. As the pressure is lowered in the reservoir, the contribution due
to gas expansion becomes greater because gas compressibility is approaching
effective compressibility. Using formation compressibility, gas production, and
shutin bottomhole pressures, two methods are presented for correcting the
reserve estimates from the early life data (assuming no water influx).
p/z Corrections
To obtain the actual gas in place, the following method is used:
1. Using Fig. 1.21, determine cf from the appropriate curve (all of these points
assume no water influx).
2. Calculate CRi and C
R
2 (P2 should be ;;, 0.5 psi/It of depth) (Craft & Hawkins):
where
av 1 1 t:.z
=
vap pi Zi t:.p
t:.z
 = slope of the z curve at Zi
t:.p
(1.44)
42 Estimation of Gas Reserves
3. Calculate effective compressibility at p, and P2 (Craft & Hawkins):
4.
CgiSgi + CwiSwi + cf
Ceft; =
Sgi
F
d h (Ceff) h' h . .
III t e  w IC IS approxImately equal to
Cg ave
Cefl; + Ce ff
2
Ggi Cg2
2
(1.45)
5. Solve the following equation:
A IG
ctua =
(Cefrl cg)ave
Apparent G is obtained from conventional pi z plot extrapolation of the early
life performance.
6. To predict the producible reserves, (a) determine the point of normal gradi
ent on the extrapolated pi z line (0.45 to 0.5 psi/ft of depth in plz value), (b)
plot the actual G from step 5 on the abscissa at zero pi z, ( c) draw a straight
lme between the normal gradient point and the actual G, and (d) determine
the reserves using the desired plz abandonment pressure.
16
, ,
Gradient derived
\ frim these 3
1
Reservoirs
0.8 Grr,nt (E'r.l'\017 (E.t.)
0.725 Gradient
0.811 Gradient sand
osrm Field'I
La
.
1 35 M.1. Deep
(S.W. Tex.)
14
""
o 12
8
.=
0.714 Gradient sand r
K
V
34 W.M.1. (S.W. Tex.)
'\
"
8
6
35 30 25 20 15 10 5