Regional petrophysical properties of Mesaverde low ... · Regional petrophysical properties of...

Regional petrophysical properties of Mesaverde low-permeability

sandstones

Regional petrophysical properties of Mesaverde low-permeability

sandstones

Alan P. Byrnes, KGS

John C. Webb, Discovery

Robert M. Cluff, Discovery

Alan P. Byrnes, KGS

John C. Webb, Discovery

Robert M. Cluff, Discovery

US DOE # DE-FC26-05NT42660US DOE # DE-FC26-05NT42660http://www.kgs.ku.edu/mesaverdehttp://www.kgs.ku.edu/mesaverde

US DOE Project Summary• Solicitation DE-PS26-04NT42072-0

– subtopic area 1: Understanding Tight Gas Resources– Award Date: October 1, 2005– Completion Date: June 30, 2008– Contract # DE-FC26-05NT42660

• Organization: University of Kansas, Kansas Geological Survey

• Principal Investigator: Alan P. Byrnes, KGS• KGS-Discovery Group, Inc. co-participants• DOE share $411,030 (80%)• Industry share $102,804 (20%)

Objectives• The project will provide petrophysical tools

that address: – 1) minimum gas flow, critical and residual gas

saturation, Sgc=f(lithofacies, Pc, architecture)– 2) capillary pressure, Pc=f(P), Pc=f(lithofacies, k,

φ, architecture)– 3) electrical properties, m* & n*– 4) facies and upscaling issues– 5) wireline log interpretation algorithms– 6) providing a web-accessible database of advanced

rock properties.

Sampling

• 4-5 wells/5 basins– 44 wells

• Describe core– 2500 ft

• 300 core samples– 1600

• 150 advanced properties samples– 150-300

Green River

Wind River

Washakie

Piceance

PowderRiver

Uinta

Wyoming

Colorado

Utah

N

Digital Core Description

• Sampling designed to sample across all lithofacies

• 5 digit system– basic type (Ss, Ls, coal)– grain size/sorting/texture– consolidation– sedimentary structure– cement mineralogy

• Property continuum not mnemonic

• Provides litholgy log traces and quantitative variables for multivariate analysis

Basic Properties Distributions

• Sampling QA• Distribution is sampling dependent but

interesting• Distribution = f(basin, Lith, M/NM, GRI, 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

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

mean median stdev ∆φ - 2.65All 2.654 2.652 0.033 0.2Green River 2.645 2.641 0.028 -0.2Piceance 2.663 2.662 0.035 0.7Powder River 2.679 2.674 0.026 1.5Sand Wash 2.633 2.639 0.020 -0.9Uintah 2.646 2.648 0.031 -0.2Washakie 2.662 2.661 0.034 0.6Wind River 2.672 2.673 0.029 1.1

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

1

10

100

1000

100 1000 10000Net Effective Stress (psi)

Pore

Vol

ume

Com

pres

sibi

lity

(10^

6/ps

i)

φ = 24%φ = 21%φ = 18%φ = 15%φ = 12%φ = 8%φ = 6%φ = 4%φ = 2%

Pore Volume Compressibility

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

psi

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 (%)

Pore

Vol

ume

Cha

nge

Inte

rcep

t (1

/psi

)

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)

• 113 Samples• Log-linear pore volume change

characteristic of fractures/sheet-pores• Slope and intercept change with porosity

β =10^[(0.000031φ2+0.00275φ-1.016)*logP+(0.000034φ3-0.00223φ2+0.056φ+4.238)]

• 113 Samples• Log-linear pore volume change

characteristic of fractures/sheet-pores• Slope and intercept change with porosity

β =10^[(0.000031φ2+0.00275φ-1.016)*logP+(0.000034φ3-0.00223φ2+0.056φ+4.238)]

In situ Klinkenberg Permeability

• b = 0.851 kik-0.34 (atm, Present Study)

• b = 0.867 kliq-0.33 (Jones & Owens)

• b = 0.777 kliq-0.39 (Heid)

0.1

1

10

100

1000

1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100 1000


Klin

kenb

erg

b fa

ctor

(atm

)

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)

kgas = kliq (1+4c /r) = kliq (1+b/P)

• Generalized = f(Ppore, Lith)• kik = 10^[0.0088 (logkair)3 - 0.072

(logkair)2+ 1.37 logkair +0.46]• kik = 10^[1.34 (logkair) - 0.6] (Byrnes, 1997)

• Statistically similar except for k > 1 mD

Gas

Liquid

Permeability vs Porosity• Generalized trend kik = 10[0.333φi-5] with 10X error• Different k-φ trends among basins due to lithologic variation• Beyond common k↑ with grain size↑, lithologic influence changes with porosity - nonlinear

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 24Calculated In situ Porosity (%)

In s

itu K

linke

nber

g Pe

rmea

bilit

y (m

D)

Green RiverPiceancePowder RiverUintaWind RiverWashakie

Permeability vs Porosity• logkik = 0.282φi + 0.18RC2- 5.13

(+4.5X MLRA)• logkik = 0.034φi

2-0.00109φi3 +

0.0032RC2 - 4.13 (+4.1X MNLRA)

• Neural Network +3.3X

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

)

hidden layer: 1Hidden layer nodes: 10

Mean> 8.239 4.280 6.294 hidden layer-Std Dev> 5.260 1.335 2.527 to-output

weightsNode Constant Phii RC2 RC4Constant -0.388

1 -0.760 2.946 -2.027 -6.438 -0.8852 -2.155 4.637 1.279 0.895 2.3233 -4.999 7.901 0.957 3.167 -2.5834 -1.484 -0.307 -1.695 6.175 -0.1545 -4.597 4.582 1.568 0.730 4.0226 -2.609 0.320 -2.201 -2.257 -2.4957 -1.765 -1.843 -1.122 0.145 -3.8598 2.839 -3.146 -9.237 0.264 0.7899 -1.566 1.029 -1.588 -3.390 2.400

10 2.951 0.778 3.316 0.179 -2.136

Input-to-hidden layer weights

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 24Calculated in situ Porosity (%)

in s

itu K

linke

nber

g Pe

rmea

bilit

y (m

D)

1XX9X1XX8X1XX7X1XX6X1XX5X1XX4X1XX3X1XX2X1XX1X1XX0X

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 24Calculated In situ Porosity (%)

in s

itu K

linke

nber

g Pe

rmea

bilit

y (m

D)

19XXX18XXX17XXX16XXX15XXX14XXX13XXX12XXX11XXX

Capillary Pressure under Pressure

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


Air-

Hg

Cap

illar

yPr

essu

re(p

sia)

LD43C4013.25 ftk = 0.190 mD

= 12.9%φ

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

PA4244606.5 ftk = 0.00107 mD

= 12.7%φ

1

10

100

1000

10000


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)R7802729.9 ftk = 7.96 mD

= 19.2%φ

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

B02913672.5 ftk = 0.000065 mD

= 2.6%φ

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

B02911460.6 ftk = 0.0255 mD

= 4.4%φ

1

10

100

1000

10000


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)

E9466530.3 ftk = 0.0416 mD

= 9.5%φ

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

E9466486.4 ftk = 0.637 mD

= 12.2%φ

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

tryPo

reDi

amet

er( µ

m)

y = 18.19x-0.50

R2 = 0.77

y = 18.94x-0.50

R2 = 0.93

1

10

100

1000

10000

1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100


Air-

Mer

cury

Thre

shol

dE

ntry

Pre

ssur

e(p

si)

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


Thre

shol

dEn

tryG

asCo

lum

nHe

ight

(ft)

A

B

C

113 mD 8 mD

0.6 mD 0.2 mD

0.04 mD 0.02 mD

0.001 mD 0.00007 mD

Capillary PressurePore Size Distribution (PSD)

• PSD expressed by Pcslope• Pcslope = f(k)• Pcslope ↑ with P ↑

0

1

2

3

4

5

6

7

8

9

0 10 20 30 40 50 60 70 80 90 100

Wetting Phase Saturation (%)

Leve

rett

J Fu

nctio

n

0.00025md0.00049md0.0012md0.0017md0.0018md0.0030md0.0040md0.0057md0.0085md0.012md0.013md0.032md0.046md0.085md0.25md0.41md0.56md0.84md2.24md

0

1

2

3

4

5

1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000


Bro

oks-

Cor

ey C

apill

ary

Pres

sure

Slo

pe

unconfinedin situ

Leverett J(Sw) = Pc (k/φ)0.5/τcosθ

Poor fit becausePcslope ≠ C = f(k)

Residual Gas Saturation

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

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


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


Air-

Hg

Cap

illar

yPr

essu

re(p

sia)


E393 7027.2 ft = 15.0% = 1.93 mD

φ

kik

1

10

100

1000

10000


Air

-Hg

Cap

illar

yP

ress

ure

(psi

a)


R829 5618.3 ft (B) = 9.2% = 0.287 mD

φ

kik

1

10

100

1000

10000


Air

-Hg

Cap

illar

yP

ress

ure

(psi

a)


B646 8294.4 ft (B) = 7.6% = 0.022 mD

φ

kik

1

10

100

1000

10000


Air

-Hg

Cap

illar

yP

ress

ure

(psi

a)


S685 6991.2 ft (B) = 8.6% = 0.0063 mD

φ

kik

1

10

100

1000

10000


Air-

Hg

Cap

illar

yPr

essu

re(p

sia)


E458 6404.8 ft (A) = 9.5% = 0.0019 mD

φ

kik

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)


KM360 8185.7 ft (B) = 5.9% = 0.00070 mD

φ

kik

C = 1/Snwr-1/SnwiSnwr = 1/[C + 1/Snwi]C = 0.55 (min ε)

C = 1/Snwr-1/SnwiSnwr = 1/[C + 1/Snwi]C = 0.55 (min ε)

Critical Gas Saturation

• Experimental work indicates Sgc < 10% often < 5% but krgcurves extrapolate to 35% < Sgc < 0%

• Issues– little krg data at Sw >

65%– Does p vary or Sgc vary

or both?

Sgc?

0.0001

0.001

0.01

0.1

1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation (fraction)

Rel

ativ

e Pe

rmea

bilit

y (f

ract

ion)

g-10 mdw -10 mdg-1 mdw -1 mdg-0.1 mdw -0.1 mdg-0.01 mdw -0.01 mdg-0.001 mdw -0.001 md

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 NonwettingPhase Saturation

• Electrical conductivity and Pc inflection indicate 0% < Sgc < 22%

• Higher Sgc in complex bedding lithofacies

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


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 Percolation

• Experimental results can be explained using four pore network architecture models

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

• Sgc (L) = A LD−E (Wilkinson and Willemsen, 1983)

– L is network dimension– A is a numerical constant (for simple cubic network A

= 0.65)– D is the mass fractal dimension of the percolation

cluster (D = 1.89 for 2-D, D = 2.52 for 3-D)– E is the Euclidean dimension (E = 2 for 2-D and E = 3

for 3-D)• As L → ∞ Sgc → 0

– Sgc = 21.5% for L = 10– Sgc = 2.4% for L = 1000– Sgc = 0.8% for L = 10000)

Archie Cementation Exponent

1.0

1.11.21.3

1.41.5

1.61.7

1.81.92.02.1

2.22.32.4

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24Porosity (fraction)

Arc

hie

Cem

enta

tion

Expo

nent

(m,A

=1)

• Empirical: m = 0.95 - 9.2φ + 6.35φ0.5

• Dual porosity: m = log[(φ-φ2)mm + φ2m2]/logφ

• φ = bulk porosity• φ2 = fracture or

touching vugporosity

• mm = matrix cementation exponent

• m2 = fracture or touching vugcementation exponent

High: mm = 2.1, φ2 = 0.0005Int: mm = 2.0, φ2 = 0.001Low: mm = 1.8, φ2 = 0.002

m2 = 1

High: mm = 2.1, φ2 = 0.0005Int: mm = 2.0, φ2 = 0.001Low: mm = 1.8, φ2 = 0.002

m2 = 1

Conclusions• Grain density for 2200 samples averages 2.654+0.033 g/cc (+1sd)

with grain density distributions differing slightly among basins. • Klinkenberg b(atm) = 0.851 kik-0.34• log kik = 0.282 φi + 0.18 RC2 - 5.13 (+4.5X,1 sd)• ANN analysis provides prediction within +3.3X• Capillary pressure (Pc) exhibits an log-log linear threshold entry

pressure (Pte) versus kik and kik/φi trend and variable Brooks-Corey slopes.

• Snwr ↑ with Snwi ↑consistent with the Land-type relation: 1/Snwr-1/Snwi = 0.55.

• Critical nonwetting-phase (e.g., gas) saturation is low (Sgc < 0.05) in massive and parallel bedded lithologies but may increase in rocks with more complex bedding Percolation theory provides a tool forpredicting limits..

• Archie cementation exponent (m) decreases with decreasing porosity (?) below approximately 6% and can be generally described by empirical or by a dual- porosity model

• These relationships are still being investigated. • Mesaverde Project website is http://www.kgs.ku.edu/mesaverde

Regional petrophysical properties of Mesaverde low ... · Regional petrophysical properties of...

Documents

Transcript of Regional petrophysical properties of Mesaverde low ... · Regional petrophysical properties of...