J Petrol Explor Prod Technol (2012) 2:117–123

DOI 10.1007/s13202-012-0027-9

ORIGINAL PAPER - PRODUCTION ENGINEERING

Modified analytical model for prediction of steam flood

performance
Ankit Dutt • Ajay Mandal

Received: 22 March 2012 / Accepted: 7 July 2012 / Published online: 8 August 2012
Ó The Author(s) 2012. This article is published with open access at Springerlink.com

Abstract Steam flooding as a tertiary recovery method loi Initial reservoir oil viscosity (cp)
for recovery of oil from heavy oil reservoir has been of qw Water density (lb/cu ft)
interest in recent years. Analytical models are very useful A Effective pattern area (acres)
to predict oil recovery by steam flooding for preliminary As Steam zone size (acres)
forecasting purposes and sensitivity studies. Though dif- Cw Specific heat of water (Btu/lbm-oF)
ferent models are available, the predicted values did not D The thermal diffusivity in the overburden and in the
satisfy the field value because of presumptions. In the reservoir beneath the steam zone (ft2/day)
present study, an attempt has been made to modify the Ehs Average thermal efficiency of steam zone,
existing Jeff Jones model and Chandra and Mamora model dimensionless
by considering the true profile of steam zone size in res- fsn Bottomhole injection steam quality, dimensonless
ervoir and vertical sweep efficiency for calculation of FhD Ratio of enthalpy of vaporization to liquid enthalpy,
capture efficiency. The reservoir characteristics and pro- dimensionless
duction data of three oil fields, viz., Schoonebeek in the Fos Cumulative ratio of oil displaced from steam zone
eastern part of Netherlands, San Ardo in Monterey County, to water injected as steam, dimensionless
California, USA and Hamaca in Venezuela’s Orinoco hfg Latent heat of steam (BTU/lbm)
heavy oil belt were analyzed for performance prediction of hn Net zone thickness (ft)
oil production. The modified model gave very satisfactory hst Steam zone thickness (ft)
results for production performance, compared to the ori- ht Gross formation thickness (ft)
ginal Jeff Jones and Chandra and Mamora model. is Steam injection rate, cold water equivalent (B/D)
kh Bulk thermal conductivity of cap rock and base
Keywords Enhanced oil recovery  Steam flooding  rock (BTU/ft-hr-oF)
Analytical model  Oil recovery  Production performance M1 Average heat capacity of steam zone
(BTU/cu ft- °F)
M2 Average heat capacity of cap rock and base rock
List of abbreviations
(BTU/cu ft-oF)
DSo Change in oil saturation before/after steam front
N Oil originally in place (STB)
passage (fraction)
Nc Ratio of the volume of moveable oil to that of
DTs Temperature difference = Tf - Ts
steam injection up to the critical time, tc
U Porosity, percent
ND Cumulative oil displacement (RB)
lo Average viscosity
Qinj Heat injection rate (BTU/h)
Qs Rate at which energy is delivered to the steam zone
by condensation before breakthrough
A. Dutt  A. Mandal (&)
qod Oil displacement rate/Myhill–Stegemeier’s oil
Department of Petroleum Engineering, Indian School of Mines,
Dhanbad 826004, India production rate (RB/D)
e-mail: mandal_ajay@hotmail.com Sg Gas saturation (fraction)

123
118 J Petrol Explor Prod Technol (2012) 2:117–123

Soi Initial oil saturation (fraction) liquids (oil and water), each liquid phase exerts its own
Sor Residual oil saturation (fraction) vapor phase at the temperature of the system. Distillation
Swc Connate water saturation (fraction) begins when the sum of the vapor phase (oil and water)
tc Critical time (years) equals the total pressure on the system. As a result, oil
tcD Time of steam injection at onset of convective heat starts distilling at a temperature much lower than the nor-
transport through condensation front, dimensionless mal boiling point of its constituents. Thirdly, miscible
tD Time of steam injection, dimensionless displacement of residual oil takes place due to the solvent
Vs,inj Cumulative steam injection (RB) extraction process (Volek and Pryor 1972): Steam distil-
lation strips the more volatile component from residual oil.
Steam, enriched with these hydrocarbons, flows through
the steam zone and gets condensed at the condensation
Introduction front. The condensed hydrocarbon contains lighter con-
stituents than the oil resident in that region and can dis-
Steam flooding is a major EOR process applied to heavy oil place some of the oil miscibly. Finally, steam drive is
reservoirs (Alajmi 2011). As the world is moving toward a inherently more stable than hot water floods (Prats 1982).
depletion era, EOR is required in every field. The demand for A steam flood project typically proceeds through four
energy has been increasing day by day, while the conven- phases of development: (1) reservoir screening; (2) pilot
tional oil reserves are shrinking speedily. In this scenario, tests; (3) field-wide implementation; and (4) reservoir
heavy oil reserves have become a good source of energy and management. Performance prediction is essential to pro-
can fulfill oil demand for a long time in the future. All over vide information for proper execution of each of these
the world, billions of barrels of heavy oil and tar sand which development phases. Three different mathematical models
cannot be produced by conventional techniques have been (statistical, numerical and analytical models) are com-
reserved. To recover heavy oil, many EOR techniques monly used to predict steam flood performance. The ana-
(Thermal Method, Chemical flooding etc.) are used, of which lytical models (volumetric) generally require the entering
the thermal method is widely used presently. of few, but critical data. Since, it is much faster to obtain
Three processes have been evolved in the thermal results from analytical models than from simulation, ana-
method: cyclic steam stimulation, steam flooding and in lytical model are still useful tools for preliminary fore-
situ combustion. In cyclic steam stimulation, steam is casting purposes and sensitivity studies. In addition, the
injected into a production well for a period of 2–4 weeks. models provide a better insight than simulation into the
The well is shut and allowed to soak before re-starting physics of the thermal process. The economic feasibility of
production. The initial production is high because of any steam flooding project depends on the accuracy of the
reduced viscosity at the increased temperature. In steam predicted production data proposed by the model. One of
flooding, steam is injected from the injection well and the most widely used analytical models is the Jeff Jones
simultaneously production takes place from the production model. Later Chandra and Mamora (2005) tried to improve
well. In situ combustion is a displacement process in which this aspect and presented a new analytical model. The
an oxygen-containing gas is injected into a reservoir where objective of this study is focused on improving this aspect
it reacts with the crude oil to create a high-temperature of the Jeff Jones model and Chandra and Mamora model.
combustion front that is propagated through the reservoir. The results of the modified model are tested against results
In most cases, the injected gas is air. based on field performance for steam drive to verify its
The present study is concerned with steam flooding. In accuracy and validity. A more accurate steam flood model
steam flooding, as the steam zone grows, more oil moves will provide engineers with an improved and useful tool for
from the steam zone to the unheated zone ahead of the prediction of steam flood production performance. This
steam front. The oil gets accumulated to form an oil bank. new model will help to decide the economical feasibility of
The condensed hot water also moves across the steam the project and will help to estimate the total production
front, heating and displacing the accumulated oil. The throughout its economic life.
heated oil with reduced viscosity moves toward the pro-
ducing well and is produced usually by artificial lifting.
There are many factors that promote high displacement Theory
efficiency of a steam drive. Firstly, there is expansion of oil
and reduction in oil viscosity with temperature (Willman Jeff Jones model
et al. 1961). Secondly, steam distillation of oil left behind
the hot water flood increases the overall oil recovery. When Jones (1981) developed an analytical model to calculate the
a vapor phase (steam) is in the presence of two immiscible oil production rate during steam flooding, based on the

123
J Petrol Explor Prod Technol (2012) 2:117–123 119

works of Myhill and Stegemeier (1978) and Lookeren

(1977). Jeff Jones’ steam flood model is divided into two
different parts. The first part of the model calculates an
optimal steam injection rate (Eq. 1). The second part uses
the related data calculated in the first part in conjunction
with additional inputs to predict the oil production history
(Eq. 4). Jeff Jones modified Myhill–Stegemeier’s (Myhill
and Stegemeier 1978) oil displacement rate and converted
it into oil production rate based on correlation with 14
different steam flood projects.

First part

Ndn  Ndn1 Fig. 1 Chandra and Mamora: three stages of oil production profile
qod ¼ ð1Þ
Dt
where; Nd ¼ Fos Vs;inj ð2Þ tcD h2t M12
tc ¼ ð10Þ
35; 040Kh M2
q Cw hn
Fos ¼ w DSo /ð1 þ Fhd ÞEhs : ð3Þ Limit: AcDmax  VoD  0
M 1 ht
 
Second part Vs;inj  5:62 2
VpD ¼ ð11Þ
43; 560As hn /Sg
Capture efficiency ðsweep efficiency) ¼ AcD  VoD  VpD
ð4Þ Limit: 0 B VpD B 1.0 and VpD = 1.0 at Sg = 0.

AcD = dimensionless area (area sweep efficiency)

VoD = volume of displaced oil produced, fraction (vol- Modified volumetric model
umetric displacement efficiency)
VpD = initial pore void filled with steam as water Steam is lighter than oil, therefore when steam is injected
q0 = q0 9 capture efficeience. gravity segregation occurs and steam moves vertically
upward. This process takes place so rapidly that the res-
ervoir heating occurs mainly due to the vertical expansion
Chandra and Mamora model of steam zone. Since steam zone thickness varies with time,
a new vertical sweep efficiency (Eq. 13) term gets incor-
Chandra and Mamora presented an improved model of Jeff porated (ESV) into the capture efficiency. This vertical
Jones seam flood analytical model. AcD (Eqs. 5, 6) was sweep efficiency has a role when reservoir thickness is
modified to account for the decrease in oil viscosity during more than 200 ft. For reservoir thickness less than 200 ft, it
steam flood and its dependence on the steam injection rate. is found that steam zone thickness is equal to sand zone
VoD (Eqs. 7–10) was modified from its square-root format thickness (Green and Willhite 1998). The calculation of
to an exponential form. The third component, Vpd (Eq. 11), average steam thickness is based on energy and mass
was kept unchanged. The oil production profile was divi- balance over a thin slice of reservoir in the vertical cross
ded into three stages as shown in Fig. 1. section (Neuman 1985).
" #2
As Capture efficiency ¼ AcD  VoD  VpD  Esv ð12Þ
AcD ¼ ð5Þ
Afa lnðloi =100Þg1=2 where Esv is assumed to be the ratio of steam zone
thickness to that of net pay zone thickness:
a ¼ 0:00015is þ 0:05 ð6Þ

hst Cw DTs ptKh 0:5
Limit: 0 B AcD B 1.0 and AcD = 1.0 at l0 B 100 cp Esv ¼ ¼ ð13Þ
  N N S   ht hfg ht M2
d dmax oi
VoD ¼ AcDmax e b NDSo ð7Þ Vs
where; hst ¼ ð14Þ
As
b ¼ 17:93Nc þ 1:3401 ð8Þ   0:5
hfg fs is Kh t
7758Ah/ð1  Sor  Swc Þ As ¼ ð15Þ
Nc ¼ ð9Þ Kh DTs M2 p
365is tc

123
120 J Petrol Explor Prod Technol (2012) 2:117–123

Table 1 Reservoir characteristic and operating conditions of Scho- But in the reservoir, due to gravity segregation, steam
onebeek, San Ardo and Hamaca fields zone size differs from that calculated by the Marx and
Characteristics of reservoir and Schoonebeek San Hamaca Langenheim model. Neuman developed a new model to
operating conditions Ardo calculate steam zone size for the gravity override model
(Neuman 1975). According to Neuman model, steam zone
Effective pattern area, A (acres) 15 10 10
size (Eq. 18) is given by:
Porosity (fraction) 0.3 0.345 0.3
Permeability (md) 1,000–10,000 6,922 12,000    0:5
Qs hfg Dt
Initial oil saturation (fraction) 0.47 0.73 0.832 As ¼ ð18Þ
Kh DTs hfg þ Cw DTs p
Residual oil saturation (fraction) 0.25 0.15 0.15  
Initial gas saturation (fraction) 0.4 0 0 where Qs ¼ ms fs hfg þ Cw DTs : ð19Þ
Initial oil viscosity (cp) 180 at 100 °F 3,000 at 25,000 at    pffiffiffiffiffi
127 °F 125 °F hfg fs is Kh t 0:5 350  24
As ¼ : ð20Þ
Reservoir temperature (8F) 100 127 125 Kh DTs M2 p 43; 560  24
Oil formation volume factor 1.01 1.01 1.01
(STB/B)
Thermal conductivity of 1.2 1 1
formation [BTU/(ft-hr-8F)] Case study: field location and setting
Rock density (lb/cu ft) 174 174 174
Heat capacity of rock (BTU/ 0.21 0.21 0.21 Case 1: Schoonebeek oil field
(lbm-8F)
Water density (lb/cu ft) 62.428 62.428 62.428 The Schoonebeek oil field is situated in the eastern part of
Oil density (lb/cu ft) 56.444 61.550 62.864 the Netherlands at the German border. The oil is produced
Heat capacity of water (Btu/lbm- 0.96 0.96 0.96 from the Bentheim barrier sands of Neocom/Valanginian
8F) age. A completely sealing fault divides the accumulation
Heat capacity of oil (Btu/lbm- 0.5 0.5 0.5 into two separate reservoirs. The southwestern part pro-
8F)
duces by solution-gas drive and the remainder of the field
Total thickness (ft) 83 115 100
by edge-water drive. In the greater part of the field, the
Net thickness (ft) 83 115 100
reservoir consists of a single, unconsolidated sand body.
Volumetric heat capacity of rock 35 35.2 35.2
(BTU/(cu ft-8F))
Schoonebeek pilot project is an inverted five-spot pattern.
Injection rate, cold water 1,250 1,600 1,600
Due to steam drive, oil recovery from this field increased
equivalent (bbl/day) from 5 to 38 %.
Saturated steam pressure (psia) 600 600 600
Steam quality (fraction) 0.7 0.8 0.8 Case 2: San Ardo field
Injection temperature (8F) 350 582.3 582.3
Oil viscosity at steam 3.5 3.5 3.5 The San Ardo is a large oil field in Monterey County,
temperature (cp)
California, in the USA. It is in the upper Salinas Valley and
Volumetric heat capacity: over 42 60 60 has an anticline structure. The productive units are the
and under (BTU/(cu ft-8F)
Aurignac sands, which are a portion of the huge Monterey
Total day of calculations (days) 2,190 6,900 6,900
Formation. Above the Aurignac sands are the thinner but
also productive Lombardi sands. All of the productive units
Cw Qs t are of Miocene age with the underlying basement rocks
Vs ¼  : ð16Þ dated to the Jurassic period
M2 hfg þ Cw DTs

Jones and Chandra and Mamora both calculated

Case 3: Hamaca field
dimensionless steam zone size in their analysis. In the
calculation of AcD, they both took Marx and Langenheim
The Hamaca field is located in Venezuela’s Orinoco heavy
(frontal advance model) (Marx and Langenheim 1959) as a
oil belt and is a huge stratigraphic trap on the southern
reference to calculate steam zone size, As.
 flank of the Oriente basin. The Hamaca concession area,
Qinj hn M1 pffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi which covers 160,000 acres, contains oil trapped in shallow
As ¼ etD erfc tD þ tD =p  1 :
4Kh ðts  tf ÞM2  43; 560 fluvial–deltaic reservoirs of the Oficina Formation (Mio-
ð17Þ cene age). Sandstone reservoirs of the Oficina Formation at

123
J Petrol Explor Prod Technol (2012) 2:117–123 121

Fig. 2 Oil production profile

(barrels of oil produced per day
vs. days of production) of the
Schoonebeek field. a Jeff Jones
Model, b Modified volumetric
model

Fig. 3 Oil production profile

(barrels of oil produced per day
vs. days of production) of the
San Ardo field. a Jeff Jones
Model, b Modified volumetric
model

Fig. 4 Oil production profile

(barrels of oil produced per day
vs. days of production) of the
Hamaca field. a Jeff Jones
Model, b Modified volumetric
model

123
122 J Petrol Explor Prod Technol (2012) 2:117–123

Table 2 Comparison of maximum oil production rate as predicted Results and discussion
by different models
Calculated oil Schoonebeek San Ardo field Hamaca The results obtained using the modified volumetric model
peak rate field (bbl/day) (bbl/day) field agrees with the actual field data in comparison to those
(bbl/day) obtained from Jeff Jones model and Chandra and Mamora
Jeff Jones model 181 620 280 model. Three fields were analyzed for this using macro
Suandy Chandra 1,250 1,600 1,600 programming in Excel. The barrels of oil produced per day
model were plotted against the days of production using the Jeff
Modified volumetric 1,250 1,880 1,480 Jones model as shown in Figs. 2a, 3a, 4a. In contrast to
model this, the same was plotted using the modified volumetric
Actual field value 1,317 2,010 1,500 model as shown in Figs. 2b, 3b, 4b. The graphs clearly
show the consistency and accuracy of the new model.
Moreover, to further prove the validity of this model, the
results of the prediction of the model were compared with
Table 3 Comparison of Cumulative oil production as predicted by that of actual field data. The oil peak rate calculated by the
different models modified volumetric model is close to the actual data, as
Cumulative oil Schoonebeek San Ardo Hamaca compared to that calculated using Jeff Jones model and
production field (bbl) field (bbl) field (bbl) Chandra and Mamora model as shown in Table 2. Chandra
and Mamora model’s prediction of cumulative oil pro-
Jeff Jones model 243,828 508,826 312,555
duction data is very large as compared to the actual field
Suandy Chandra model 633,758 1,022,254 784,155
data as shown in Table 3. In this table, it can be clearly
Modified volumetric 384,069 540,639 491,989
model
seen that the new model gives the best prediction. Simi-
larly, maximum recovery prediction for the three fields
Actual field value 444,562 659,820 589,200
shows that modified the volumetric prediction is better off
than the other widely accepted models as shown in Table 4.

Table 4 Comparison of maximum recovery as predicted by different

models Conclusion

Oil recovery Schoonebeek San Ardo Hamaca field

1. Analytical models (volumetric) are used to predict
(fraction) field (%) field (%) (%)
steam flood performance (preliminary forecasting
Jeff Jones model 18 22.89 16 purposes and sensitivity studies) because these require
Suandy Chandra 46 43 40 few data, give quick results and provide better insight,
model than simulation, into the physics of the thermal
Modified 28 24.3 25 process.
volumetric model
2. The most widely used analytical model given by Jeff
Actual field value 33 27 30
Jones and then modified by Chandra and Mamora has
the limitation of the frontal advance model, and hence
give unsatisfactory prediction of oil production.
Hamaca were generally deposited in a bedload-dominated, 3. The new modified volumetric model, as presented in
fluvial–deltaic environment. Hamaca crude is considered this paper, removes the above limitation by consider-
‘‘foamy’’ and is generally saturated with gas at reservoir ing gravity segregation in the development of the
conditions. model.
4. Results based on the modified model agree with field
results for three different sets of reservoir and fluid
Reservoir characteristic and operating condition properties: Schoonebeek field, San Ardo field and
Hamaca field.
At maximum production rate, the vertical sweep efficiency 5. Engineers will find the modified model an improved
is unity because the steam zone thickness is equal to the net and useful tool for prediction of steam flood produc-
pay zone thickness. The various reservoir parameters and tion performance. In this study, it has been proved that
characteristics data which have been used to predict the the new modified model can predict more accurately
performance of steam flooding are shown in Table 1. than the earlier existing models.

123
J Petrol Explor Prod Technol (2012) 2:117–123 123

6. Engineers can predict the economic feasibility and Green DW, Willhite GP (1998) Enhanced oil recovery. Society of
optimum injection rate with the new volumetric model. Petroleum Engineers, Richardson, pp 330–370
Jones J (1981) Steam drive model for hand-held programmable
Performance prediction is essential to provide infor- calculators. J Petroleum Technol 33(9):1583–1598
mation for proper execution of each of these develop- Lookeren JV (1977) Calculation methods for linear and radial steam
ment phases. Therefore, this model is very useful for flow in oil reservoirs. In: SPE 6788, SPE 52nd annual technical
engineers in the decision-making process. conference and exhibition, Denver, 9–12 Oct
Marx JW, Langenheim RH (1959) Reservoir heating by hot fluid
injection. Trans AIME 216:312–314
Open Access This article is distributed under the terms of the Myhill NA, Stegemeier GL (1978) Steam-drive correlation and
Creative Commons Attribution License which permits any use, dis- prediction. J Petroleum Technol 30:173–182
tribution, and reproduction in any medium, provided the original Neuman CH (1975) A mathematical model of the steam drive
author(s) and the source are credited. process—applications. In: SPE 4757, SPE 50th annual technical
conference and exhibition, Dallas, 28 Sept–1 Oct
Neuman CH (1985) A Gravity override model of steam drive.
J Petroleum Technol 37(1):163–169
References Prats M (1982) Thermal recovery, SPE monograph, vol 7. Society of
Petroleum Engineers of AIME, Dallas
Alajmi AF (2011) Modeling of oil bank formation during steam flood. Volek CW, Pryor JA (1972) Steam distillation drive—Brea Field,
(SPE 142618), SPE Middle East Oil and Gas Show and California. J Petroleum Technol 24(8):899–906
Conference held in Manama, Bahrain, 25–28 Sept 2011 Willman BT et al (1961) Laboratory studies of oil recovery by steam
Chandra S, Mamora DD (2005) Improved steam flood analytical injection. JPT (July) 681–690, Trans-AIME, 222
model. In: Paper (SPE 97870), SPE-PE/CIM-CHAO Interna-
tional Thermal operations and heavy oil symposium, Calgary,
1–3 Nov

123