Professional Documents
Culture Documents
Rock and Fluid Characterization-DAS103358
Rock and Fluid Characterization-DAS103358
Rock and Fluid Characterization-DAS103358
talization, and determination of reservoir quality. Similarly, sampling techniques. The required fluid data also depend
fluid properties are used for determining reservoir continu- on the fluid type, ranging from simple compressibility-fac-
ity. Special-core-analysis laboratory (SCAL) data are used for tor estimation for dry gas to more-complex phase-behavior
rock-typing and flow-unit classification. Together with fluid changes for near-critical fluids. The rock-type-based relative
data, these analyses provide input for well-deliverability permeability and capillary pressure data are essential for
calculations, rock/fluid compatibility studies, stimulation all reservoir and fluid types. Sample size will depend
design, reservoir-engineering calculations, reservoir simu- on the nature and scale of heterogeneity (Honarpour
lation, production optimization, and design of improved- et al. 2003).
recovery processes.
Table 1 provides a comprehensive list of fluid pressure/ Data Collection
volume/temperature (PVT) and SCAL requirements for a A road map for rock and fluid characterization for reservoir
variety of reservoir fluids and rock types. Depending on management is shown in Fig. 1. The program involves a
whether the reservoir contains a dry gas; condensate; or vol- set of clear objectives, a comprehensive data-collection and
atile, black, or heavy oil, the sampling method varies from testing plan, experimental protocols, rigorous QC/quality-
simple wellhead sampling to more-complex bottomhole assurance (QA) procedures, and a sound data-management
Reservoir
Fluid Fluid Sampling and PVT Requirements Reservoir Rock SCAL Requirements Critical Considerations
Dry gas or Dry gas—wellhead samples, Z-Factor All rock types and Pc, kgswi, krg–krw, and Sgt Swi and distribution trapped-
wet gas Wet gas—separator samples, CGR and tight gas sand gas saturation, and krg–krw
Z-Factor
Gas DST and formation-tester samples All rock types kro–krg, krg–krw, Sgt, Reservoir-condition SCAL
condensate Isokinetic samples for rich condensate and Scc tests with reservoir fluids, Bo
and GOR, nonhydrocarbon
composition, and OBM
contamination
Near-critical Formation-tester samples All rock types Pc, kro–krw, remaining-oil Critical gas saturation,
fluid solution GOR, oil shrinkage below saturation PVT properties, and OBM
saturation pressure contamination
Volatile oil Formation-tester or DST samples, PVT All rock types Pc, kro–krw, remaining-oil Accurate GOR,
data for gas-injection processes saturation compositional PVT, IFT,
and OBM contamination
Black oil Surface samples, standard black-oil All rock types Pc, kro–krw, kro–krg, Bo, GOR, viscosity, IFT,
PVT hysteresis, Sgc, remaining- and OBM contamination
oil saturation
Viscous or Formation-tester sampling, viscosity, Mainly clastic rocks Pc, kro–krw, krw–krg, Viscosity, fluid contamina-
heavy oil density, and emulsion and unconsolidated hysteresis, Sgc, remaining- tion, sand production, Sgc,
sand oil saturation krw, kro, krg, and emulsion
All fluid types Sampling and PVT corresponding to Clay-rich sandstone SCAL on preserved cores Formation damage because
fluid types listed above or with properly cleaned of clay swelling/fines
cores and wettability migration, unrepresentative
restoration using SCAL if incompatible brine
compatible brine is used
All fluid types Sampling and PVT corresponding to Unconsolidated sand SCAL with minimum Grain reorientation, sample
fluid types listed above; avoid sand change to mechanical disintegration, clay swelling/
production properties fines migration and
unrepresentative SCAL
All fluid types Sampling and PVT corresponding to Heterogeneous SCAL on whole core Scale of measurement
fluid types listed above rocks, vugular and to capture the effect of
naturally fractured heterogeneity, mud solids
contamination
Fluids with Formation-tester sampling and depth- All rock types Compositional-dependent Depth-dependent fluid
compositional dependent PVT SCAL variation and the effect on
variation relative permeability
Quality-Control/Quality-
Assurance Program
Establish Detailed Sampling
and Data-Gathering Plan
Data Analysis/Management
Data Analysis/Management
• Data synthesis and modeling
• Data scaleup
• Data repository in database
- Raw data
- Analyzed data
system. The process starts with the collection of representa- hole samples. Special coring techniques are used to maximize
tive rock and fluid samples, sample preservation/restoration, core recovery while obtaining low-invasion core samples to
and testing procedures. The validated data are organized into avoid unnecessary cleaning and wettability restoration.
a data repository and managed digitally for retrieval and fur- Fig. 3 exhibits typical uncertainties observed in rock- and
ther analysis (e.g., synthesis and modeling). fluid-property measurements and the resulting effect on hydro-
carbon-in-place (HCIP) values, production/injection rates, frac-
Critical Rock and Fluid Data tional flow, and recovery factors. The magnitude of the effect is
The basic and critical rock and fluid data required for evalu- related to the rock and fluid types, their properties, wettability,
ation of various reservoir-depletion strategies are shown in and the balance of capillary, viscous, and gravitational forces.
Fig. 2. Rock, fluid, and rock/fluid data form the basis for the The largest effect on HCIP is caused by uncertainties in poros-
design of any recovery process, calibration of wireline logs, ity, particularly for low-quality rocks, and by uncertainties in
and evaluation of potential compaction and subsidence. The the formation volume factor (FVF) for volatile oils. The uncer-
specific rock-/fluid-data requirements are further classified tainties in permeability, relative permeability, viscosity, and FVF
under three major reservoir-fluid classes: viscous or heavy have similar effects on rates and fractional flow. Quantifying the
oils, light oils, or gas and gas condensates. The recovery pro- effect of rock/fluid uncertainties on the recovery factor is not
cesses involving light oils are grouped into three processes: straightforward because of heterogeneity, wettability, and the
immiscible, miscible, and tertiary gas injection, including combined effect of viscous, capillary, and gravitational forces.
water-alternating-gas (WAG) injection. Other recovery pro- However, the wettability, endpoint saturation, corresponding
cesses, such as thermal methods, will require customized relative permeability, and permeability contrast are critical
rock/fluid characterization as well. parameters that affect the recovery factor significantly. The
Acquiring representative rock and fluid samples is chal- recovery factor in heavy-oil reservoirs is highly influenced by
lenging and requires expertise. Major challenges in obtain- uncertainties in the live-oil viscosity.
ing fluid samples include avoiding liquid dropout in gas
condensates and gas evolution in heavy oils during sampling. QC Program
Potential fluid contamination by oil-based mud (OBM) High-quality rock and fluid data add significant value by
should be minimized for obtaining representative bottom- reducing uncertainty, especially when the initial investment
* Includes the effect of factors such as net-/gross-pay ratio and interwell connectivity.
is substantial, such as in deepwater operations and produc- the parts of both the service providers and the clients to
tion strategies for giant reservoirs. The evaluation of coring ensure quality data.
fluid, coring parameters, core handling, and wellsite-wetta-
bility preservations should be the first step in the QC process Data Management
(Bulau and Honarpour 1997). In case of contaminated core Millions of dollars are spent to gather, analyze, and interpret
samples, cleaning and wettability restoration should be cus- the rock and fluid data for use in various reservoir-engineer-
tom designed. Special tools and methods should be adopted ing and -simulation studies. However, if these data are not
to minimize fluid contamination by OBM. managed efficiently, the value of the data may not be fully
The laboratory QC program consists of prequalification realized. Often, the data gathered on rocks and fluids from
of the laboratories and subsequent consistency checks. The different sources must be archived for retrieval, then dis-
prequalification involves evaluation of laboratories on the played graphically for data comparison and integration. Key
basis of routine property measurements on a set of standard components of a data-management system should include
rock and fluid samples and an on-site inspection. This step a comprehensive data repository and a set of analysis tools
provides information on the laboratories’ technical capa- to guide in proper interpretation, modeling, and systematic
bilities including wellsite operation, rock and fluid handling, data integration. The data-repository system should include
screening, analysis, and safety practices. It also includes a database of both raw and interpreted data in a well-orga-
information about the company’s QC practices, communi- nized format. The raw data must be grouped on the basis of
cation, documentation, and reporting on previous projects. regional and geological information. Further grouping may
Strength and weaknesses of each laboratory in various cat- be organized on the basis of individual reservoir, well, and
egories should be documented and used to select qualified rock/fluid type. The data also are classified under different
laboratories. The consistency check of routine-core-analysis formations, petrophysical properties, and fluid types, then
data from prequalified laboratories is conducted by repeating ranked for quality by use of standard QC/QA procedures.
measurements on a subset of samples at another qualified
laboratory. Special-core- and fluid-analysis data need to be Examples: Effect of Rock and Fluid Properties
compared against a database on similar core samples or flu- Three examples listed in Table 2 show the effect of rock
ids analyzed under similar conditions. In summary, reliable and fluid data on reservoir-performance prediction. The
core- and fluid-analysis data require a qualified laboratory, examples include volatile oil, black oil, and gas/condensate
supervision by experienced staff, and coordinated effort on fluid systems from both sandstone and carbonate reser-
TABLE 2—KEY ROCK AND FLUID PROPERTIES AND THEIR EFFECT ON PERFORMANCE PREDICTIONS
The initial compositional-simulation model was set up ducing high-permeability streaks at selected locations. The
with permeability maps that were generated with stochastic result was a forced history match that honored neither the
modeling because of a lack of basic rock data at appropriate rock nor the fluid data. The reservoir-performance prediction
measurement scale. Analog relative permeability data were based on this simulation model was unreliable, as indicated
used to simulate gas/oil fractional flow. Simulation studies by the actual field performance encountered in the later life
were performed to history match the production data. A of the field, with very low sweep efficiency, earlier gas break-
comparison of the produced GOR with simulation predic- through, and a significantly lower liquid recovery.
tions in Fig. 5 shows a poor match. To overcome this diffi- A new simulation model was set up honoring all the rock
culty, the geological model was altered significantly by intro- and fluid data. The geologic model was based on a determin-
15
40
Field data
30 10
Prediction with correct Prediction with incorrect
rock and fluid data
rock and fluid models
20 Prediction with correct
Prediction with 5 rock and fluid models
10 incorrect rock
and fluid models Field data
0 0
1980 ‘81 ‘82 ‘83 ‘84 ‘85 ‘86 ‘87 ‘88 ‘89 ‘90 ‘91 ‘92 ‘93 1994 1996 1998 2000 2002 2004 2006 2008 2010
Time, year Time, year
Fig. 5—Separator GOR profile: prediction vs. field Fig. 6—Condensate rate as a function of time.
data.
istic model for porosity and permeability. A fluid gradient
model based on gravity/chemical equilibrium (GCE) was expansion and solution-gas drive with little aquifer support.
used to match the observed compositional gradient, cor- Gravity drainage was recognized as the main recovery mech-
rectly predicting a thin volatile-oil leg at the base and a large anism because of adequate gas-/oil-density contrast, excel-
gas/condensate fluid above it. The GCE-model predictions lent vertical and horizontal permeability (1,000 and 3,000
for methane and C7+ also are shown in Fig. 4. An improved md, respectively), a large oil column, and fieldwide pressure
history match was obtained, as shown in Fig. 5 for the communication. In the gravity-drainage process, oil-bypass-
produced high-pressure-separator GOR. Even though the ing and -trapping events are minimized because of the stable
nitrogen-breakthrough time was not matched, the late-life gas front displacing the oil. The oil maintains hydraulic con-
performance was accurately captured. The condensate rate as tinuity and drains to very low residual-oil saturation because
a function of time from the two simulation runs along with of positive spreading coefficient.
field data are compared in Fig. 6. The simulation model that The gravity-drainage process was characterized in the
honored rock and fluid data with the correct compositional laboratory with several types of drainage gas/oil relative
gradient matched the field performance more closely. permeability and capillary pressure tests at reservoir condi-
tions using live oil (Edwards et al. 1998). The relevant tests
Effect of Gas/Oil Relative Permeability on Reserves and included gravity-drainage experiments with X-ray for in-situ
Recovery—Ubit Field. The Ubit field is one of the largest saturation monitoring on 1- to 5-ft-long vertically oriented
producing light-oil reservoirs in Nigeria, with more than whole cores to determine oil relative permeability, kro, resid-
2 billion STB oil in place. The reservoir has an areal extent ual-oil saturation, and endpoint gas permeability; steady-
of 15,000 acres, with an oil column of 160 ft and a dip of state-drainage gas/oil relative permeability tests to obtain
2.5°. The reservoir energy was provided by natural gas-cap gas and oil relative permeabilities; primary-drainage gas/oil
1
k ro 70
k rg Transition zone
60
Relative Permeability, fraction
0.1
50
Oil Saturation, %
0.01
Siw = 9%
40
Primary Secondary
0.001 30 gas cap gas cap
20
0.0001 Current
10 gas/oil
contact
0.00001
0 10 20 30 40 50 60 70 80 90 100 0
Gas Saturation + Initial Water Saturation, % 5,720 5,730 5,740 5,750 5,760 5,770 5,780 5,790 5,800
Core Depth, ft
Fig. 7—Typical gas/oil relative permeabilities for Fig. 8—Residual-oil saturations in gas cap, secondary
high-quality geologic facies. gas cap, and oil zone.
200 2000
Modified relative
Field
* *** *
50 500
0
1970 1986 2002 2018 2034 2050
Time, year
Fig. 9—Comparison of oil rate and cumulative oil production for cases with conventional relative permeability
and gravity-drainage relative permeability.
relative permeability with the centrifuge technique under production and a corresponding increase in cumulative
confining stress to determine kro at very low oil saturation; recovery. Actual performance was not as good as the predic-
and primary-drainage gas/oil capillary pressure on core plugs tion because of changes to the development plan, mainly a
by use of the centrifuge technique under confining stress to reduction in the number of new wells because of capital-
determine minimum residual-oil saturation. A typical gas/oil spending constraints.
relative permeability curve for high-quality facies is shown
in Fig. 7. Residual-oil saturation for major reservoir facies Effect of Relative Permeability, Water Vaporization,
was low, in the range of 6.5% ±5%. Laboratory residual-oil- and Measurement Scale on Performance—Arun Field.
saturation measurements were validated with a low-invasion Arun field is a gas/condensate field off the northern coast of
coring program using water-based mud in a zone in which Aceh Province, North Sumatra, Indonesia. The initial reser-
the gas cap expanded approximately 25 ft over a 25-year pro- voir pressure and temperature were 7,100 psia and 352°F,
duction period. The residual-oil saturation measured on the respectively, at a datum of 10,050 ft subsea. The gas-bearing
cored interval, in the secondary gas cap, was approximately formation had an areal extent of 23,000 acres and formation
6%, as shown in Fig. 8. thickness of 1,000 ft (Pathak et al. 2004). The extensive
The simulation model used black-oil PVT properties and core analysis identified vuggy (reef facies), moldic (lagoonal
comprised 18 layers, mostly in the oil zone to capture oil facies), and intercrystalline porosities. The reservoir fluid
displacement by advancing gas cap. The history-match was a retrograde-gas condensate containing 15 mol% CO2
parameters were primarily the pressure, GOR, and water/oil with a liquid yield of 65 separator bbl/MMscf at 1,250 psia
ratio (WOR). The residual-oil saturation and the gas/oil and 68°F. The dewpoint pressure was 4,400 psi at 352°F.
relative permeability for the gravity-drainage process played The connate-water saturation in the gas zone was in the
a significant role in the history match. Initially, only centri- range of 5 to 20% for reef facies and higher for lagoonal
fuge relative permeability measurements with stock-tank oil facies. Pressure maintenance by peripheral gas injection
without correction for capillary end effect were used in the (mainly methane and CO2) was implemented to delay
simulation. Even when the correct residual-oil saturation of condensate dropout. Although the gas injection helped to
6% was used, the simulation model calculated remaining displace rich gas toward the central clusters of producers,
oil saturations as high as 20% in the secondary gas cap after it also resulted in CO2-rich gas production in later stages of
25 years of production. This error was caused by adverse oil the reservoir life. The focus of the reservoir-rock and -fluid
relative permeability near residual-oil saturation. After using characterization and modeling was an accurate account of
the correct gas/oil relative permeability curves, residual- the reduction in gas deliverability caused by condensate
oil saturation in the range of 4.0 to 6.5% was predicted in dropout, water vaporization, and increased CO2 content of
the reservoir-simulation model in the actual production produced gas.
time span. The difference in the shape of the oil relative Core Characterization—Measurement Scale. The effect
permeability curve for the gravity-drainage process had a of measurement scale on both routine and special core anal-
major effect on the reserves and the plateau production, ysis was captured though measurements on whole cores and
adding 150 million STB of incremental oil recovery. Fig. 9 core plugs. Permeability was measured on whole cores and
compares the base case with a case that used correct gas/oil core plugs taken at high- and low-permeability locations
relative permeability, resulting in a 3-year increase in plateau identified by minipermeametry. Fig. 10 shows significant
differences in absolute permeabilities between the whole 50% for the lagoonal facies. The drop in productivity index
core and core plugs as a result of the presence of small-scale caused by liquid accumulation in the near-wellbore region,
heterogeneities (Honarpour et al. 2003). Whole-core data predicted by a radial single-well compositional model using
provided the most representative values because they cor- measured relative permeability data, showed an excellent
rectly averaged the high- and low-permeability zones into match with field data.
the overall volume, whereas the plug measurements were
significantly lower because they did not capture the 3D con- Conclusions
nectivity of the pore system. Ignoring this scale effect can Reliable rock and fluid characterization require a set of
lead to unnecessary adjustment of relative permeability or well-defined objectives, a clear road map, and a multidis-
other data. Capillary pressure and electrical-property mea- ciplinary approach. A systematic and integrated proce-
surements on whole cores showed the characteristics of a dure, involving well-designed coring and sampling pro-
dual vugular/matrix-pore system for reef facies (Honarpour cedures, best-practice laboratory tests, rigorous QC, data
et al. 2003). The whole-core pore-volume compressibility integration, and validation, should be followed to reduce
measured from reef facies showed 20 to 100 microsips, com- measurement uncertainty and increase data reliability for
pared with 5 to 16 microsips measured on core plugs. accurate reservoir-performance prediction. Rock and fluid
Water Vaporization. Water vaporization was recognized data are important company assets; therefore, they require
as the main source of water production in the field.
Increasing water vaporization during depletion was char-
acterized by a nine-component equation-of-state (EOS) 60
model and validated with laboratory measurements ranging
from 4 mol% at the initial reservoir pressure to 16 mol% at 50 Actual
1,000 psia. Fig. 11 shows good agreement between field Simulation
Water, thousand B/D
a well-designed data-management system for future access βω =water formation volume factor
and application. φ =porosity
Rock and fluid characterization strongly affect in-place ρ =density
volumes, recovery factors, injectivity/productivity, and deliv- µ =viscosity
erability. Therefore, accurate rock and fluid characterization µo =oil viscosity
are key for minimizing technical risks and maximizing the µw =water viscosity
asset value. ∆P =pressure drop
Acronyms Acknowledgments
CGR =condensate/gas ratio We gratefully acknowledge the support and encourage-
EOS =equation-of-state ment of ExxonMobil Upstream Research Co., ExxonMobil
FVF =formation volume factor Exploration Co., ExxonMobil Oil Indonesia, Mobil Producing
IFT =interfacial tension Nigeria, and ExxonMobil Production Co.
OBM =oil-based mud
OHCIP =original hydrocarbon in place References
PU =porosity unit Afidick, D., Kaczorowski, J., and Bette, S. 1994. Production
PVT =pressure/volume/temperature Performance of Retrograde Gas Reservoir: A Case Study of the
QA =quality assurance Arun Field. Paper SPE 28749 prepared for presentation at SPE
QC =quality control Asia Pacific Oil & Gas Conference, Melbourne, Australia, 7–10
RF =recovery factor November.
SU =saturation unit Al-Hussainy, R. and Humphreys, N. 1996. Reservoir Management
WAG =water-alternating-gas Principles and Practices. JPT 48(12): 1129–1135. SPE 30144.
Bulau, J.R. and Honarpour, M.M. 1997. A Quality Control Program
Nomenclature in Core Analysis. Soc. of Core Analysts, Paper 9713.
A =cross-sectional area Edwards, J.T., Honarpour, M.M., Hazlett, R.D., Cohen, M., Pebdani,
Bo =oil formation volume factor F., Clayton, C., and Al-Hussainy, R. 1998. Validation of Gravity-
EA =areal-sweep efficiency Dominated Relative Permeability and Residual Oil Saturation in
EV =vertical-sweep efficiency a Giant Oil Reservoir. Paper SPE 49316 prepared for presentation
Fw =fractional flow at the SPE Annual Technical Conference and Exhibition, New
h =formation thickness Orleans, 27–30 September.
k =permeability Honarpour, M.M., Djabbarah, N.F., and Sampath, K. 2003. Whole
kgswi =gas permeability at initial water saturation Core Analysis—Experience and Challenges. Paper SPE 81575
krg =gas relative permeability prepared for presentation at the SPE Middle East Oil Show,
kro =oil relative permeability Bahrain, 5–8 April.
krw =water relative permeability Metcalfe, R.S., Vogel, J.L., and Moris, R.W. 1988. Compositional
kr(S) =relative permeability as a function of saturation Gradients in the Anschutz Ranch East Field. SPERE 3(8): 1025–
Pc =capillary pressure 1032. SPE 14412.
Q =flow rate Nagarajan, N., Honarpour, M.M., Sampath, K., and McMichael,
re =drainage radius D. 2004. Comparison of Gas/Condensate Relative Permeability
rw =well radius Using Live Fluid vs. Model Fluids. Paper 2004-41A presented
Scc =critical condensate saturation at the Intl. Symposium of the Soc. of Core Analysts, Abu Dhabi,
Sgc =critical gas saturation UAE, 5–9 October.
Sgt =trapped gas saturation Pathak, P., Fidra, Y., Kahar, Z., Agnew, M., and Hidayat, D. 2004.
Sorw =residual oil saturation in displaced zone The Arun Gas Field in Indonesia: Resource Management of a
Swi =initial water saturation Mature Field. Paper SPE 87042 prepared for presentation at
Z =gas deviation factor the SPE Asia Pacific Conference on Integrated Modeling for
βhc =hydrocarbon formation volume factor Asset Management, Kuala Lumpur, 29–30 March.
βo =oil formation volume factor
JPT