lithofacies and petrophysical properties of lithofacies and … · 2008-11-07 · y p r es su r e...
TRANSCRIPT
Lithofacies and petrophysical properties of Mesaverde tight-gas sandstones in Western
U.S. basins
Lithofacies and petrophysical properties of Lithofacies and petrophysical properties of Mesaverde tightMesaverde tight--gas sandstones in Western gas sandstones in Western
U.S. basinsU.S. basins
Alan P. ByrnesKGS- now Chesapeake Energy
John C. WebbRobert M. CluffDaniel A. KrygowskiStefani D. Whittaker
The Discovery Group, Inc
Alan P. ByrnesKGS- now Chesapeake Energy
John C. WebbRobert M. CluffDaniel A. KrygowskiStefani D. Whittaker
The Discovery Group, Inc
2008 AAPG Annual Convention presentation, San Antonio
US DOE Project SummaryUS DOE Project Summary
DOE Contract # DOE Contract # DEDE--FC26FC26--05NT4266005NT42660completion date 30 June 2008completion date 30 June 2008
Organizations Organizations University of Kansas Center for Research, Inc.University of Kansas Center for Research, Inc.Kansas Geological Survey, Lawrence, KSKansas Geological Survey, Lawrence, KSThe Discovery Group Inc., Denver, COThe Discovery Group Inc., Denver, CO
Principal Investigator: Alan P. Byrnes, KGSPrincipal Investigator: Alan P. Byrnes, KGS
project website is project website is http://www.kgs.ku.edu/mesaverdehttp://www.kgs.ku.edu/mesaverde
Project objectivesProject objectives
investigate minimum gas flow, critical and residual gas investigate minimum gas flow, critical and residual gas saturation, Sgc=saturation, Sgc=f f (lithofacies, Pc, architecture)(lithofacies, Pc, architecture)investigate capillary pressure, Pc=investigate capillary pressure, Pc=f f (P), Pc=(P), Pc=f f (lithofacies, (lithofacies, k, k, φφ, architecture), architecture)investigate electrical properties, m = investigate electrical properties, m = f f ((φ, φ, salinity)salinity)characterize lithofacies and upscaling issuescharacterize lithofacies and upscaling issuesdevelop advanced wireline log interpretation algorithmsdevelop advanced wireline log interpretation algorithmsproviding a webproviding a web--accessible database of advanced rock accessible database of advanced rock properties properties this is the last of our overview presentations this is the last of our overview presentations –– from from here on they will be topic specific deep dives.........here on they will be topic specific deep dives.........
SamplingSampling
systematic systematic characterization of characterization of Kmv lithofacies over Kmv lithofacies over entire Rocky Mtn entire Rocky Mtn regionregion44 wells/6 basins44 wells/6 basinsDescribed 7000 ft Described 7000 ft core (digital)core (digital)2200 core samples2200 core samples120120--400 advanced 400 advanced properties samples
Green River
Wind River
Washakie
Piceance
PowderRiver
Uinta
Wyoming
Colorado
Utah
N
properties samples
Digital Core Digital Core DescriptionDescription
To provide lithologic input to To provide lithologic input to equations and predict lithology equations and predict lithology from logs used 5 digit systemfrom logs used 5 digit system
1 basic type (Ss, Ls, coal)1 basic type (Ss, Ls, coal)2 grain size/sorting/texture2 grain size/sorting/texture3 consolidation3 consolidation4 sedimentary structure4 sedimentary structure5 cement mineralogy5 cement mineralogy
Property continuum Property continuum -- not not mnemonic or substitution ciphermnemonic or substitution cipherSimilar to system used in 1994 Similar to system used in 1994 and subsequent studiesand subsequent studies
Core descriptionCore description
rock typing at 0.5 ft rock typing at 0.5 ft frequency to match log frequency to match log data resolutiondata resolutionlithology, color, grain lithology, color, grain size, sed structuressize, sed structuressample locationssample locationsimportant cementsimportant cementsEODEOD
Porosity consists of poorly to moderately connected moldic and secondary intergranular mesopores and traces of pore-lining ML/IS(?) clay containing microporosity. Quartz cement is prominent, ferroan calcite is sparse. Pore-lining clay cement causes elevated Swi and reduced relative permeability.
40X
100X
Williams PA 424, 6148.8’ 152769.9% 2.66 g/cc Ka=0.0237 mD
40X
100X
Porosity consists almost entirely of sparse, poorly connected, clay-filled intergranular microporosity. Quartz cement is prominent, ferroan calcite is sparse. Pore-filling clay cement causes elevated Swi and reduced relative permeability.
Williams PA 424, 4686.4’ 152867.9% 2.65 g/cc Ka=0.211 mD
Sample QA & Sample QA & distributionsdistributionsPetrophysical property distributions are Petrophysical property distributions are generally normal or loggenerally normal or log--normalnormalSubSub--distributions = distributions = f f (basin, lithofacies, (basin, lithofacies, marine/nonmarine/non--marine, etc.)
0
5
10
15
20
25
30
35
40
45
50
1E-7
- 1E
-6
1E-6
- 1E
-5
1E-5
- 1E
-4
0.00
01-0
.001
0.00
1-0.
01
0.01
-0.1
0.1-
1
1-10
10-1
00
100-
1,00
0
In situ Klinkenberg Permeability (mD)
Perc
ent o
f Pop
ulat
ion
(%)
AllGreen RiverPiceancePowder RiverSand WashUintahWind RiverWashakie
marine, etc.)
0
10
20
30
40
50
60
2.58-2.60
2.60-2.62
2.62-2.64
2.64-2.66
2.66-2.68
2.68-2.70
2.70-2.72
2.72-2.74
Grain Density (g/cc)
Perc
ent o
f Bas
in P
opul
atio
n
Green RiverPiceancePowder RiverUintahWind RiverWashakieSand Wash
0
5
10
15
20
25
30
35
40
45
0-2
2-4
4-6
6-8
8-10
10-1
2
12-1
4
14-1
6
16-1
8
18-2
0
20-2
2
22-2
4
In situ Porosity (%)
Perc
ent o
f Pop
ulat
ion
(%)
AllGreen RiverPiceancePowder RiverSand WashUintahWind RiverWashakie
0.70
0.75
0.80
0.85
0.90
0.95
1.00
100 1000 10000
Confining Pressure (psi)
Frac
tion
ofPo
rosi
tyat
200
psiPore Volume Pore Volume
CompressibilityCompressibility
y = 0.0060x + 0.03R2 = 0.59
0.00
0.05
0.10
0.15
0.20
0.25
0 2 4 6 8 10 12 14 16 18 20 22 24
Routine Helium Porosity (%)
Por
eVo
lum
eC
hang
eSl
ope
(-1/p
si)
y = 0.013x + 1.08R2 = 0.51
1.0
1.1
1.2
1.3
1.4
1.5
0 2 4 6 8 10 12 14 16 18 20 22 24
Routine Helium Porosity (%)Po
re V
olum
e C
hang
e In
terc
ept
(1/p
si)
Previously documented in literatureno large datasets in public domain113 SamplesLog-linear pore volume change seen in EVERY sample, avg. R2 = 0.99 characteristic of cracks/sheet-poresSlope and intercept increase with increasing porosity
Previously documented in literaturePreviously documented in literatureno large datasets in public domainno large datasets in public domain113 Samples113 SamplesLogLog--linear pore volume change seen in linear pore volume change seen in EVERY sample, avg. REVERY sample, avg. R22 = 0.99 = 0.99 characteristic of cracks/sheetcharacteristic of cracks/sheet--poresporesSlope and intercept increase with Slope and intercept increase with increasing porosity increasing porosity
Stress dependence of Stress dependence of permeabilitypermeability
Known for many years that lowKnown for many years that low--K K sandstones are stress sensitivesandstones are stress sensitiveGeneralized = Generalized = f f (P(Pporepore, Lith), Lith)1997 Byrnes equation:1997 Byrnes equation:kkikik = 10^[1.34 (logk= 10^[1.34 (logkairair) ) -- 0.6] 0.6] This study:This study:kkikik = 10^[0.0088 (logk= 10^[0.0088 (logkairair))33 -- 0.072 0.072 (logk(logkairair))22+ 1.37 logk+ 1.37 logkairair +0.46]+0.46]Statistically similar except for k > Statistically similar except for k > 1 mD1 mDno meaningful stress dependence no meaningful stress dependence over 10 mDover 10 mD
y = -0.0088x3 - 0.0716x2 + 1.3661x - 0.4574R2 = 0.9262
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
-7 -6 -5 -4 -3 -2 -1 0 1 2 3log Routine Air Permeability Ppore = 100 psi (mD)
log
In s
itu K
linke
nber
g Pe
rmea
bilit
y (m
D)
Permeability vs PorosityPermeability vs PorosityOverall trend allows prediction of Kik from porosity with 10X erOverall trend allows prediction of Kik from porosity with 10X errorrorBreaking into two subtrends at Breaking into two subtrends at φφ~12% improves to 5X error~12% improves to 5X errorDifferent kDifferent k--φφ trends among basins trends among basins Beyond common kBeyond common k↑↑ with grain sizewith grain size↑↑, lithologic influence changes are complex , lithologic influence changes are complex and nonlinearand nonlinear
0.0000001
0.000001
0.00001
0.0001
0.001
0.01
0.1
1
10
100
1000
0 2 4 6 8 10 12 14 16 18 20 22 24In situ calc Porosity (%)
Klin
kenb
erg
Perm
eabi
lity
(4,0
00 p
si, m
D)
Green RiverPiceancePowder RiverUintahWashakieWind RiverlogK=0.3Phi-3.7logK=0.3Phi-5.7
Permeability vs PorosityPermeability vs Porositylogklogkikik = 0.234= 0.234φφii + 0.12RC+ 0.12RC22-- 4.71 4.71 ((++4X; 4X; φφ<12%)<12%)logklogkikik = 0.265= 0.265φφii + 0.11RC+ 0.11RC2 2 --4.80 4.80 ((++5X; 0<5X; 0<φφ<24%)<24%)Artificial Neural Network Artificial Neural Network ++3.3X3.3X
0.00001
0.0001
0.001
0.01
0.1
1
10
100
1000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000
Measured in situ Klinkenberg Permeability (mD)
Pred
icte
d in
situ
Klin
kenb
erg
Perm
eabi
lity
(mD
)
0.0000001
0.000001
0.00001
0.0001
0.001
0.01
0.1
1
10
100
1000
0 2 4 6 8 10 12 14 16 18 20 22 24In situ calc Porosity (%)
Klin
kenb
erg
Perm
eabi
lity
(4,0
00 p
si, m
D)
X9XXXX8XXXX7XXXX6XXXX5XXXX4XXXX3XXXX2XXXX1XXX
0.0000001
0.000001
0.00001
0.0001
0.001
0.01
0.1
1
10
100
1000
0 2 4 6 8 10 12 14 16 18 20 22 24In situ calc Porosity (%)
Klin
kenb
erg
Perm
eabi
lity
(4,0
00 p
si, m
D)
XXX9XXXX8XXXX7XXXX6XXXX5XXXX4XXXX3XXXX2XX1XXXXXX0X
Capillary pressureCapillary pressure
investigating Pc as investigating Pc as ff (lithology, (lithology, φφ, K), K)120 high120 high--low pairslow pairssampled across basins, permeability range, & lithologysampled across basins, permeability range, & lithology
stress sensitivity of Pcstress sensitivity of Pcmost MICP curves are run under laboratory conditions, but most MICP curves are run under laboratory conditions, but given stress dependence of permeability we expect Pc to also given stress dependence of permeability we expect Pc to also be stress sensitivebe stress sensitive
relationship between initial and residual nonrelationship between initial and residual non--wetting wetting phase saturations (phase saturations (““scanning curvesscanning curves””))
only published data are for conventional reservoir rocksonly published data are for conventional reservoir rocks
Stress effect on PcStress effect on Pc
threshold entry pressure threshold entry pressure is entirely predictable is entirely predictable from from √√K/K/φφ ratio at any Pratio at any P
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yPr
essu
re(p
sia)
R091255.9 ftk = 113 mD
= 24.5%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yPr
essu
re(p
sia)
LD43C4013.25 ftk = 0.190 mD
= 12.9%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Capi
llary
Pres
sure
(psi
a)
PA4244606.5 ftk = 0.00107 mD
= 12.7%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yP
ress
ure
(psi
a)R7802729.9 ftk = 7.96 mD
= 19.2%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Capi
llary
Pres
sure
(psi
a)
B02913672.5 ftk = 0.000065 mD
= 2.6%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yPr
essu
re(p
sia)
B02911460.6 ftk = 0.0255 mD
= 4.4%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yP
ress
ure
(psi
a)
E9466530.3 ftk = 0.0416 mD
= 9.5%φ
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yPr
essu
re(p
sia)
E9466486.4 ftk = 0.637 mD
= 12.2%φ
113 mD 8 mD
0.6 mD 0.2 mD
0.04 mD 0.02 mD
0.001 mD 0.00007 mD
y = 11.77x0.50
R2 = 0.77
y = 11.28x0.50
R2 = 0.93
0.01
0.1
1
10
100
1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100
Klinkenberg Permeability/Porosity (mD/%)
Thre
shol
dEn
try
Pore
Dia
met
er( µ
m)
A
y = 6.48x-0.50
R2 = 0.77
y = 6.75x-0.50
R2 = 0.93
1
10
100
1000
10000
1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100
Klinkenberg Permeability/Porosity (mD/%)
Thre
shol
dEn
try
Gas
Col
umn
Hei
ght(
ft)
C
Residual Gas Residual Gas Saturation
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100
Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yP
ress
ure
(psi
a)Primary DrainageFirst ImbibitionSecondary DrainageSecond ImbibitionTertiary DrainageThird Imbibition
E393 7001.1ft = 17.4% = 28.9 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yP
ress
ure
(psi
a)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
B049 9072.1 ft (A) = 12.3% = 6.74 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yPr
essu
re(p
sia)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
E393 7027.2 ft = 15.0% = 1.93 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yP
ress
ure
(psi
a)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
R829 5618.3 ft (B) = 9.2% = 0.287 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yP
ress
ure
(psi
a)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
B646 8294.4 ft (B) = 7.6% = 0.022 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yP
ress
ure
(psi
a)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
S685 6991.2 ft (B) = 8.6% = 0.0063 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air-
Hg
Cap
illar
yPr
essu
re(p
sia)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
E458 6404.8 ft (A) = 9.5% = 0.0019 mD
φ
kik
1
10
100
1000
10000
0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)
Air
-Hg
Cap
illar
yPr
essu
re(p
sia)
Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition
KM360 8185.7 ft (B) = 5.9% = 0.00070 mD
φ
kik
Saturation• Snwi and Snwr are ~ = for Sw > 80%• e.g., for Swi of 30%, Swr is ~50%
• Snwi and Snwr are ~ = for Sw > 80%• e.g., for Swi of 30%, Swr is ~50%
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Initial Nonwetting Phase Saturation (Snwi)
Res
idua
l Non
wet
ting
Phas
e Sa
tura
tion
(Snw
r) unconfinedconfinedLand C=0.66, Swi=0
Land C =0.54, Swi=0
Critical Gas SaturationCritical Gas Saturation
Experimental work indicates Experimental work indicates Sgc < 10% often < 5%Sgc < 10% often < 5%butbut krg curves extrapolate to krg curves extrapolate to 35% < Sgc < 0%35% < Sgc < 0%IssuesIssues
little krg data at Sw > 65%little krg data at Sw > 65%two different ways to model two different ways to model the data, which is better?the data, which is better?
0.00001
0.0001
0.001
0.01
0.1
1
0 10 20 30 40 50 60 70 80 90 100Water Saturation
Gas
Rel
ativ
e Pe
rmea
bilit
y
Critical Nonwetting Critical Nonwetting Phase SaturationPhase Saturation
Electrical conductivity and Pc inflection indicate 0% Electrical conductivity and Pc inflection indicate 0% < Sgc < 22%< Sgc < 22%Higher Sgc as bedding complexity increasesHigher Sgc as bedding complexity increases
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000
In situ Klinkenberg Permeability (mD)
Crit
ical
Non
-wet
ting
Phas
e Sa
tura
tion
MICP-inflectionElectrical Resistance High Pressure Vessel
Oil Hgpositive
displacementpump
Hg
Highpressure
oilpump
Core
316 SSend caps
Rubbersleeve
VoltmeterV∆
Vacuum
Sgc and Sgc and percolation theorypercolation theory
experimental results can be experimental results can be explained using four explained using four -- pore pore network architecture modelsnetwork architecture models
critical gas saturation strongly critical gas saturation strongly controlled by sedimentary controlled by sedimentary structures/rock fabricstructures/rock fabricanyany bedding parallel bedding parallel laminations result in low Sgclaminations result in low Sgc
1) Percolation Network N ( ) - Macroscopically homogeneous, random distribution of bond sizes, e.g., Simple Cubic Network (z=6)
p
2) Parallel Network N
N
( ) preferential orientation of pore sizes or beds of different
networks parallel to the invasion direction.
II
p
Invasion direction
3) Series network N
N
( ) - preferential sample-spanning orientation of pore sizes or beds of different networks perpendicular to the invasion direction.
p
4) Discontinuous series network N
Np
N N
( ) - preferential non-sample-spanning orientation of pore sizes or beds of different networks perpendicular to the invasion direction. Represents continuum between and
d
p.
0
100
200
300
400
500
600
700
800
900
1000
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Water Saturation
Gas
-Wat
erC
apill
ary
Pres
sure
(kPa
)
0.001 md0.1 md
AB
Archie porosity (cementation) exponentArchie porosity (cementation) exponentNearly all cores exhibit some salinity dependenceNearly all cores exhibit some salinity dependencetested plugs with 20K, 40K, 80K, and 200K ppm brinestested plugs with 20K, 40K, 80K, and 200K ppm brines
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
0.01 0.1 1
Brine Resistivity (ohm-m)
In s
itu A
rchi
e C
emen
tatio
n Ex
pone
nt,
(m, A
=1)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 2 4 6 8 10 12 14 16 18 20 22
Brine Conductivity (mho/m)
Cor
e C
ondu
ctiv
ity (m
ho/m
)
n=335
Porosity dependence of Porosity dependence of ““mm””Empirical: m = 0.234 ln Empirical: m = 0.234 ln φφ + 1.33+ 1.33Dual porosity: m = log[(Dual porosity: m = log[(φφ--φφ22))m1m1 + + φφ22
m2m2]/log ]/log φφφφ22 = 0.35% m= 0.35% m11=2, m=2, m22=1; SE both = 0.11=1; SE both = 0.11rock behaves like a mixture of matrix porosity and rock behaves like a mixture of matrix porosity and cracks or fracturescracks or fractures
both models fit databoth models fit data
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
0 2 4 6 8 10 12 14 16 18 20 22
In situ Porosity (%)
In s
itu A
rchi
e C
emen
taito
n Ex
pone
nt
(m, a
=1, X
brin
e=40
Kpp
mN
aCl)
φφ = bulk porosity= bulk porosityφφ22 = fracture porosity= fracture porosity
mm11 = matrix = matrix cementation cementation exponentexponent
mm22 = fracture = fracture cementation cementation exponentexponent
Salinity dependence of Salinity dependence of ““mm””m = a ln m = a ln φ φ + b+ ba, b = a, b = ff (salinity)(salinity)low porosity rocks hold low porosity rocks hold more gas than we more gas than we thoughtthought
20K ppm
y = 0.2267Ln(x) + 2.2979
R2 = 0.6619
0.00
0.50
1.00
1.50
2.00
2.50
0.000 0.050 0.100 0.150 0.200 0.250
insitu porosity (%)
Axis Title
Series1
Log. (Series1) 40K ppm
y = 0.2328Ln(x) + 2.409
R2 = 0.6547
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0.000 0.050 0.100 0.150 0.200 0.250
insitu porosity (%)
Axis Title
Series1
Log. (Series1) 80K ppm
y = 0.2149Ln(x) + 2.4354
R2 = 0.5132
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0.000 0.050 0.100 0.150 0.200 0.250
insitu porosity (%)
Axis Title
Series1
Log. (Series1)
200K ppm
y = 0.1621Ln(x) + 2.3222
R2 = 0.3633
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0.000 0.050 0.100 0.150 0.200 0.250
insitu porosity (%)
Axis Title
Series1
Log. (Series1)
Log modelingLog modeling
ran a ran a ““basicbasic”” log analysis on every study welllog analysis on every study welltotal and effective porositytotal and effective porosityoptimize grain density, matrix valuesoptimize grain density, matrix valuesArchie and Dual Water Sw using Archie and Dual Water Sw using ““standardstandard”” tight gas tight gas electrical parameters from earlier workelectrical parameters from earlier workTimur equation absolute permeability estimateTimur equation absolute permeability estimate
imported digital rock numbers, routine core data, KGS imported digital rock numbers, routine core data, KGS SCAL data, and converted all routine to SCAL data, and converted all routine to in situin situ using using standard equationsstandard equationsCore data depth shifted using rock number curveCore data depth shifted using rock number curve
core rock numbervs. GR
grain density
alternativeporosity models
Kabs
Sw model
Advanced log modelingAdvanced log modeling
variable grain density model as variable grain density model as ff (lith)(lith)improved permeability modelimproved permeability modelimproved Sw model, including variable m, nimproved Sw model, including variable m, nrock typingrock typing
really, really tough. May not be possible with really, really tough. May not be possible with conventional conventional ““triple combotriple combo”” log suiteslog suitesLooking at NMR, other options to improve Looking at NMR, other options to improve predictabilitypredictability
Work is still in progressWork is still in progress
ConclusionsConclusions
Average grain density for 2200 samples is 2.654+0.033 Average grain density for 2200 samples is 2.654+0.033 g/cc (+1sd) g/cc (+1sd)
but grain density distributions differ slightly among but grain density distributions differ slightly among basins & lithofacies. basins & lithofacies.
Pore volume compressibility shows a logPore volume compressibility shows a log--linear linear relationship characteristic of sheet like pores and cracksrelationship characteristic of sheet like pores and cracksStress dependence of permeability is consistent with Stress dependence of permeability is consistent with prior work (Byrnes, 1997)prior work (Byrnes, 1997)PorosityPorosity--permeability data exhibit two subtrends with permeability data exhibit two subtrends with permeability prediction approaching 5X within eachpermeability prediction approaching 5X within each
Adding rock types or using an ANN model improves Adding rock types or using an ANN model improves perm prediction to 3.3X perm prediction to 3.3X –– 4X4X
Capillary pressure (Pc) is stress sensitive as expectedCapillary pressure (Pc) is stress sensitive as expectedthreshold entry pressure is entirely predictable from threshold entry pressure is entirely predictable from √√K/K/ φφ at any confining pressureat any confining pressureMinimal impact at low SwMinimal impact at low Sw’’s (high capillary pressures)s (high capillary pressures)
Residual gas saturation increases with increasing initial Residual gas saturation increases with increasing initial gas saturationgas saturation
LandLand--type relation: (1/Snwr)type relation: (1/Snwr)--(1/Snwi) = 0.55 (1/Snwi) = 0.55 Critical gas saturation is low (Sgc < 0.05) in laminated Critical gas saturation is low (Sgc < 0.05) in laminated sandstones but tends to increase in rocks with more sandstones but tends to increase in rocks with more complex bedding complex bedding
Percolation theory provides a tool for predicting Percolation theory provides a tool for predicting limits.limits.
Archie porosity/cementation exponent (m) decreases Archie porosity/cementation exponent (m) decreases with decreasing porosity below 10% with decreasing porosity below 10%
Can predict using empirical or a dualCan predict using empirical or a dual-- porosity porosity modelmodelLittle impact over 10Little impact over 10--12% porosity (constant m)12% porosity (constant m)
Project completion within 70 daysProject completion within 70 daysWatch our website for new data Watch our website for new data –– most project results most project results will be posted this summerwill be posted this summerDetailed topicDetailed topic--specific presentations begin in July.specific presentations begin in July.
Forthcoming presentationsForthcoming presentations
AAPGAAPG--Rocky Mtn Section meeting, Denver, Rocky Mtn Section meeting, Denver, July 9July 9--1111thth
Cluff & ByrnesCluff & Byrnes: Evidence for a variable Archie : Evidence for a variable Archie porosity exponent porosity exponent ““mm”” and impact on saturation and impact on saturation calculationscalculationsWebb et alWebb et al: Lithofacies and petrophysical properties : Lithofacies and petrophysical properties of Mesaverde tightof Mesaverde tight--gas sandstonesgas sandstonesByrnes et alByrnes et al: Capillary pressure properties of : Capillary pressure properties of Mesaverde Group lowMesaverde Group low--permeability sandstonespermeability sandstones
Questions?Questions?