INTRODUCTION TO ROCK PHYSICS GARY MAVKO Rock Physics Laboratory Stanford University Hosted by Rock Solid Images
Feb 06, 2016
INTRODUCTION TO ROCK PHYSICS
GARY MAVKO
Rock Physics Laboratory Stanford University
Hosted by Rock Solid Images
Preface
Because of the increasing importance of oil recovery, the growing complexity of recently discovered oil fields, and the growing realization that reservoirs and recovery are more heterogeneous than assumed in the past, a major shift in the use of seismic methods has taken place during the past decade. One of the central aspects of this shift involves the need to better understand the relation between the seismic properties of reservoir rocks and their production properties (porosity, permeability) and state (mineralogy, saturation, pore pressure, etc.). Some obvious applications are the evaluation of stratigraphic traps, fracture detection, and the spatial distribution of porosity and permeability.
Reservoir complexity is typically related to significant spatial heterogeneity in porosity, permeability, clay content, fracture density, and other properties. A direct consequence of this heterogeneity is the complexity of reservoir recovery processes, ranging from migration of the gas cap in reservoirs with discontinuous shales, overpressured zones, and the tracking of injected water, steam, or temperature during recovery in reservoirs with large spatial variations of permeability. This spatial variability cannot be inferred at any level of detail from well testing data, logs, or cores. It can only be obtained from remote geophysical measurements, especially seismic measurements -- and only when these seismic measurements can be understood in terms of reservoir properties.
Over the last three decades enormous strides have been made to understand the relations between the physical properties of reservoir rocks and geophysical observables -- the science now known as Rock Physics. We have gradually discovered more and more order in relations that once appeared disappointingly scattered, for example, velocity vs. porosity, porosity vs. permeability, Vp/Vs vs. saturation and lithology. Some of the keys have been to explore effects of pore pressure, stress, temperature, clay content, compaction, fluid type, and saturation.
This course covers fundamentals of Rock Physics, ranging from basic laboratory and theoretical results to practical “recipes” that can be immediately applied in the field. We will present qualitative and quantitative tools for understanding and predicting the effects of lithology, pore fluid types and saturation, stress and pore pressure, fractures, and temperature on seismic velocity and attenuation. We will present case studies and strategies for seismic interpretation, upscaling seismic and rock properties from the lab to borehole to reservoir scales, suggestions for more effectively employing seismic-to-rock properties transforms in geostatistical methods, and especially emphasize subsurface fluid detection and recovery monitoring.
Stanford Rock Physics Laboratory - Gary Mavko
1
Units of Stress
1 bar = 106 dyne/cm2 = 14.50 psi
10 bar = 1 MPa = 106 N/m2
Mudweight to Pressure Gradient
1 psi/ft = 144 lb/ft3
= 19.24 lb/gal
= 22.5 kPa/m
1 lb/gal = 0.052 psi/ft
Stanford Rock Physics Laboratory - Gary Mavko
2
Summary:
whereρρρρ densityK bulk modulusµµµµ shear modulusλλλλ Lamé's coefficientE Young's modulusνννν Poisson's ratio
P wave velocity
S wave velocity
E wave velocity
In terms of Poisson's ratio we can also write:
Relating various velocities:
Moduli from velocities:
VS = µµµµρρρρ
VP = K + (4/3)µµµµ
ρρρρ = λλλλ + 2µµµµρρρρ
VE = Eρρρρ
VP2
VS2 = 2(1–νννν)
(1–2νννν)
VE2
VP2 = (1+νννν)(1–2νννν)
(1–νννν)
νννν = VP2 – 2VS
2
2 VP2 – VS
2 = VE2 – 2VS
2
2VS2
VP
2
VS2 =
4 – VE2
VS2
3 – VE2
VS2
VE2
VS2 =
3Vp2
VS2 – 4
VP2
VS2 – 1
µµµµ = ρρρρVS2
νννν = VP2 – 2VS
2
2 VP2 – VS
2
K = ρρρρ VP2 – 4
3 VS2
E = ρρρρVE2
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
3
Typical rock velocities, from Bourbié, Coussy, andZinszner, Acoustics of Porous Media, Gulf Publishing.
Type of formation P wavevelocity
(m/s)
S wavevelocity
(m/s)
Density(g/cm3)
Density ofconstituent
crystal(g/cm3)
Scree, vegetal soil 300-700 100-300 1.7-2.4 -Dry sands 400-1200 100-500 1.5-1.7 2.65 quartzWet sands 1500-2000 400-600 1.9-2.1 2.65 quartzSaturated shales and clays 1100-2500 200-800 2.0-2.4 -Marls 2000-3000 750-1500 2.1-2.6 -Saturated shale and sand sections 1500-2200 500-750 2.1-2.4 -Porous and saturated sandstones 2000-3500 800-1800 2.1-2.4 2.65 quartzLimestones 3500-6000 2000-3300 2.4-2.7 2.71 calciteChalk 2300-2600 1100-1300 1.8-3.1 2.71 calciteSalt 4500-5500 2500-3100 2.1-2.3 2.1 haliteAnhydrite 4000-5500 2200-3100 2.9-3.0 -Dolomite 3500-6500 1900-3600 2.5-2.9 (Ca, Mg)
CO32.8-2.9Granite 4500-6000 2500-3300 2.5-2.7 -Basalt 5000-6000 2800-3400 2.7-3.1 -Gneiss 4400-5200 2700-3200 2.5-2.7 -Coal 2200-2700 1000-1400 1.3-1.8 -Water 1450-1500 - 1.0 -Ice 3400-3800 1700-1900 0.9 -Oil 1200-1250 - 0.6-0.9 -
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
4
1
2
3
4
5
6
0 100 200 300
Bedford Limestone
Velo
city
(km
/s)
Effective Pressure (bars)
Sat.
Dry
Dry
Sat.
VP
V S
2
3
4
5
6
7
0 100 200 300
Westerly Granite
Velo
city
(km
/s)
Effective Pressure (bars)
Dry
Sat.
VP
VS
Dry
Sat.
2
3
4
5
6
0 100 200
Solenhofen limestone
Velo
city
(km
/s)
Effective Pressure (bars)
VS
VPSat. and Dry
Sat. and Dry
3
4
5
6
7
0 100 200 300
Webatuck dolomite
Velo
city
(km
/s)
Effective Pressure (bars)
Sat.
Sat.
Dry
Dry
VP
VS
The Saturation and Pressure Dependence of P- and S-wave Velocities.
F.1
Peffective = Pconfining – Ppore
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
5
Fundamental Observations of Rock Physics
• Velocities almost always increase with effectivepressure. For reservoir rocks they often tend toward a flat, high pressure asymptote.
• To first order, only the difference between confining pressure and pore pressure matters, not the absolute levels of each -- ”effective pressure law.”
• The pressure dependence results from the closing of cracks, flaws, and grain boundaries, which elastically stiffens the rock mineral frame.
• The only way to know the pressure dependence of velocities for a particular rock is to measure it.
• Make ultrasonic measurements on dry cores; fluid-related dispersion will mask pressure effects.
• The amount of velocity change with pressure is a measure of the number of cracks; the pressure range needed to reach the high pressure asymptote is a measure of crack shape (e.g. aspect ratio).
• Velocities tend to be sensitive to the pore fluid content. Usually the P-wave velocity is most sensitive and the S-wave velocity is less sensitive.
• Saturation dependence tends to be larger for soft (low velocity) rocks.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
6
Pressure-Dependence of Velocities
It is customary to determine the pressure dependence of velocities from core measurements. A convenient way to quantify the dependence is to normalize the velocities foreach sample by the high pressure value as shown here. This causes the curves to cluster at the high pressure point. Then we fit an average trend through the cloud, as shown. The velocity change between any two effective pressuresP1 and P2 can be conveniently written as:
Remember to recalibrate this equation to your own cores!
F29
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
0 5 10 15 20 25 30 35 40
P-Velocity Pressure Dependence
Vp /
Vp(4
0)
Effective Pressure (MPa)
Average:V
P /V
P(40) = 1.0-0.38*exp(-P
eff /12)
Remember:Dry Cores!
V(P2)V(P1) = 1.0 – 0.38 exp(–P2 /12)
1.0 – 0.38 exp(–P1 /12)
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
7
Effects of Pore Fluid on P-wave Velocity (Low Frequency)
10
10.5
11
11.5
12
12.5
13
13.5
0 10 20 30 40 50
P Im
peda
nce
(km
/s)-(
gm/c
m3 )
Effective Pressure (MPa)
dry
oil
water
4.4
4.6
4.8
5
5.2
5.4
0 10 20 30 40 50
Velo
city
(km
/s)
Effective Pressure (MPa)
dry
oil
water
16
18
20
22
24
26
28
30
0 10 20 30 40 50
Bulk
Mod
ulus
(GPa
)
Effective Pressure (MPa)
dry
oil
water
9
9.5
10
10.5
11
11.5
0 10 20 30 40 50 60
P Im
peda
nce
(km
/s)-(
gm/c
m3 )
Effective Pressure (MPa)
dry
oil
water
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
0 10 20 30 40 50 60
Velo
city
(km
/s)
Effective Pressure (MPa)
dry
oilwater
161718192021222324
0 10 20 30 40 50 60
Bulk
Mod
ulus
(GPa
)Effective Pressure (MPa)
dry
oil
water
Beaver Sandstone6% porosity
Fontainebleau Sandstone15% porosity
Calculations made from dry velocities, using Gassmann relation,Kmin = 36 GPa, Kwater = 2.2, Koil = 1.
F.2
Density Effect!
Density does not lead to ambiguitywhen Impedance is measured.
Imp = ρρρρV = ρρρρ modulus
V = modulusρρρρ
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
8
Effects of Pore Fluid on P-wave Velocity (Low Frequency)Ottawa Sand
1
2345
678
9
0 10 20 30 40 50 60
Bulk Modulus - Ottawa SandBu
lk M
odul
us (G
Pa)
Effective Pressure (MPa)
dry
oil
water
1
1.5
2
2.5
0 10 20 30 40 50 60
Vp - Ottawa Sand
Velo
city
(km
/s)
Effective Pressure (MPa)
dry
oil
water
0
1
2
3
4
5
6
0 10 20 30 40 50 60
P Impedance - Ottawa Sand
P Im
peda
nce
(km
/s)-(
gm/c
m3 )
Effective Pressure (MPa)
dry
oil
water
F.3
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
9
Beaver Sandstone6% porosity
Effects of Pore Fluid on P-wave Velocity (Low Frequency)Fontainebleau Sandstone
15% porosity
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0 10 20 30 40 50
Beaver Poisson's Ratio
Pois
son'
s R
atio
Effective Pressure (MPa)
dry
oil
water
1.42
1.44
1.46
1.48
1.5
1.52
1.54
0 10 20 30 40 50
Beaver Vp/Vs
Vp/V
s
P MPa
water
oil
dry
4.4
4.6
4.8
5
5.2
5.4
0 10 20 30 40 50
Beaver Vp
Velo
city
(km
/s)
Effective Pressure (MPa)
dry
oil
water
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
0 10 20 30 40 50 60
Fontainebleau Vp
Velo
city
(km
/s)
Effective Pressure (MPa)
dry
oil
water
0.1
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0 10 20 30 40 50 60
Fontainebleau Poisson's Ratio
Pois
son'
s R
atio
Effective Pressure (MPa)
dry
oil
water
1.5
1.51
1.52
1.53
1.54
1.55
1.56
1.57
1.58
0 10 20 30 40 50 60
Fontainebleau Vp/Vs
Vp/V
s
P MPa
water
oil
dry
F.4
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
10
Effects of Pore Fluid on P-wave Velocity (Low Frequency)Ottawa Sand
0
0.1
0.2
0.3
0.4
0.5
0 10 20 30 40 50 60
Ottawa Poisson's Ratio
Pois
son'
s R
atio
Effective Pressure (MPa)
dry
oil
water
1.5
2
2.5
3
0 10 20 30 40 50 60
Ottawa Vp/Vs
Vp/V
s
P MPa
water
oil
dry
1
1.5
2
2.5
0 10 20 30 40 50 60
Vp - Ottawa SandVe
loci
ty (k
m/s
)
Effective Pressure (MPa)
dry
oil
water
F.5
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
11
Velocities depend on fluid modulus and density
• When going from a dry to water saturated rock, sometimes the P-velocity increases; sometimes it decreases.
• The rock elastic bulk modulus almost always stiffens with a stiffer (less compressible) pore fluid.
• The stiffening effect of fluid on rock modulus is largest for a soft (low velocity) rock.
• The bulk density also increases when going from a dry to water-saturated rock.
• Because velocity depends on the ratio of elastic modulus to density, the modulus and density effects “fight” each other; sometimes the velocity goes up; sometimes down.
• Measures of modulus ( ), impedance ( ), and don’t have the density effect “ambiguity.”
• Be careful of ultrasonic data! At high frequencies, the elastic-stiffening effect is exaggerated for both bulk and shear moduli; so we don’t often see the density effect in the lab and the velocities will be contaminated by fluid-related dispersion.
Imp = ρρρρV = ρρρρ modulus VP / Vs
M = ρρρρV2
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
12
Beaver Sandstone6% porosity
Effect of Pore Pressure
Effect of pore pressure on velocity, calculated assuming effective pressure law is valid, and assuming a fixed confining pressure of 40MPa (low frequency calculations using Gassmannrelation).
4.4
4.6
4.8
5
5.2
5.4
0 5 10 15 20 25 30 35 40
Velo
city
(km
/s)
Pore Pressure (MPa)
dry
oil
water
F.6
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
13
Ways that Pore Pressure Impacts Velocities
• Increasing pore pressure softens the elastic mineral frame by opening cracks and flaws, tending to lower velocities.
• Increasing pore pressure tends to make the pore fluid or gas less compressible, tending to increase velocities.
• Changing pore pressure can change the saturation as gas goes in and out of solution. Velocities can be sensitive to saturation.
• High pore pressure persisting over long periods of time can inhibit diagenesis and preserve porosity, tending to keep velocities low.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
14
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
0 10 20 30 40 50 60
Fontainebleau
P-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
Sat.
Dry
15
15.2
15.4
15.6
15.8
0 10 20 30 40 50 60
Poro
sity
(%)
Effective Stress (MPa)
0
0.2
0.4
0.6
0.8
0 10 20 30 40 50 60
Soft
Poro
sity
(%)
Effective Stress (MPa)
2.6
2.7
2.8
2.9
3
3.1
3.2
3.3
0 10 20 30 40 50 60
Fontainebleau
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
DrySat.
F.7
Ultrasonic velocities and porosity in Fontainebleausandstone (Han, 1986). Note the large change in velocity with a very small fractional change in porosity. This is another indicator that pressure opens and closes very thin cracks and flaws.
Pressure Dependence of Velocities
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
15
Dry shaly sandstone data from Han (1986). Each vertical “streak” plotted with the same symbol is a single rock at different pressures. Note the large change in modulus with little change in porosity -- another illustration that cracks and flaws have a large change on velocity, even though they contribute very little to porosity.
Only the values at high pressure and with Han's empirical clay correction applied.
0
5
10
15
20
25
30
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Bul
k M
odul
us (G
Pa)
Porosity
Pressure
0
5
10
15
20
25
30
35
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4Sh
ear M
odul
us (G
Pa)
Porosity
Pressure
0
5
10
15
20
25
30
35
0 0.1 0.2 0.3 0.4
Shea
r Mod
ulus
(GPa
)
Equivalent Porosity
asymptotic stiffness
0
5
10
15
20
25
30
0 0.1 0.2 0.3 0.4
Bul
k M
odul
us (G
Pa)
Equivalent Porosity
asymptotic stiffness
F.8
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
16
The Information in a Rock'sVelocity-Pressure Curve
1. High pressure limiting velocity is a function ofporosity
2. The amount of velocity change with pressureindicates the amount of soft, crack-like pore space
3. The range of the greatest pressure sensitivityindicates the shape or aspect ratios of the crack-likepore space
Effective Pressure
Velo
city
Vmineral
indicates crack shape or aspect ratio
indicates "soft" or
crack porosity
indicates porosity
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
17
Soft, Crack-Like Porosity
1. Includes micro and macrofractures and compliant grain boundaries.
2. Soft Porosity:
• Decreases both P and S-wave velocities
• Increases velocity dispersion and attenuation
• Creates pressure dependence of V and Q
• Creates stress-induced anisotropy of V and Q
• Enhances sensitivity to fluid changes (sensitivity to hydrocarbon indicators)
3. High confining pressure (depth) and cementation, tend to decrease the soft porosity, and thereforedecreases these effects.
4. High pore pressure tends to increase the softporosity and therefore increases these effects.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
18
4
4.5
5
5.5
6
0 10 20 30 40
Vp (k
m/s
)
Effective Pressure (MPa)
Font D
BeaverFont CFont E
Font A
Font GPeter A
Peter CFont H
Increasing Pore Pressure
0 0.05 0.1 0.15 0.2 0.25 0.3
Porosity
A
BDiagenesis
Loading
Transient Unloading
Font D
BeaverFont C
Font E
Font A
Font GPeter A
Peter C
Font HIncreasing Pore Pressure
Seismic Velocity and Overpressure
F.25
Curves on the left show the typical increase of velocity with effective pressure. For each sample the velocity change is associated with the opening and closing of cracks and flaws. These are typical when rapid changes in effective pressure occur, such as during production.
Curves on the right show the same data projected on the velocity-porosity plane. Younger, high porosity sediments tend to fall on the lower right. Diagenesis and cementation tend to move samples to the upper left (lower porosity, higher velocity). One effect of over-pressure is to inhibit diagenesis, preserving porosity and slowing progress from lower right to upper left. This is called “loading” type overpressure. Rapid, late stage development of overpressure canopen cracks and grain boundaries, resembling the curves on the left. This is sometimes called “transient” or “unloading” overpressure. In both cases, high pressure leads to lower velocities, but along
different trends.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
19
7200
8000
8800
9600
10400
11200
2000 2500 3000 3500 4000 4500 5000
dept
h (ft
)
Vp (m/s)
Normal trend
Low velocity
Seismic Velocity and Overpressure
A typical approach to overpressure analysis is to look for low velocity deviations from normal depth trends. Caution: when overpressure is “late stage,” estimates of pressure can be too low.
F31
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
20
Experiments that illustrate the effective pressure law. In the first part of the experiment, effective pressure is increased by increasingconfining pressure from 0 to 80 MPa, while keeping pore pressure zero (solid dots). Then, effective pressure is decreased by keeping confining pressure fixed at 80 MPa, but pumping up the pore pressure from 0 to nearly 80 MPa (open circles). (Jones,1983.)
The curves trace approximately (but not exactly) the same trend.There is some hysteresis, probably associated with frictional adjustment of crack faces and grain boundaries. For most purposes, the hysteresis is small compared to more serious difficulties measuring velocities, so we assume that the effective pressure law can be applied. This is a tremendous convenience, since most
laboratory measurements are made with pore pressure equal 0.
2.5
3
3.5
4
4.5
5
0 20 40 60 80
St. Peter sandstoneVe
loci
ty (k
m/s
)
Effective Pressure (MPa)
pc=80 MPa
pc=80 MPa
pp=0
pp=0
VP
VS
2.6
2.8
3
3.2
3.4
3.6
0 20 40 60 80
Sierra White granite
Velo
city
(km
/s)
Effective Pressure (MPa)
pp=0
pc=80 MPa
VS
F.9
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
21
Effective pressure relationships, from Wyllie et al., 1958; see also Bourbié, Coussy, and Zinszner, 1987, Acoustics of Porous Media.
2.5
3
3.5
4
0 20 40 60 80
Berea sandstone
Velo
city
(km
/s)
Confining Pressure (MPa)
peff
=pc
peff
=28 MPap
eff=21 MPa
peff
=14 MPap
eff=7 MPa
peff
=0
F.10
Effective Pressure = Confining Pressure - Pore Pressure
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
22
0.5
1
1.5
2
2.5
3
3.5
4
0 50 100 150 200
Velo
city
(km
/s)
Confining Pressure (MPa)
pp=1MPa
pp=41 MPa
pp=0
pp=0
pp=1MPa
VP
V S
Sat.
Sat.
Dry
Dry
pp=41 MPa
Pierre shale (ultrasonic), from Tosaya, 1982.
F.11
For shales, we also often see an increase of velocity with effective pressure. The rapid increase of velocity at low pressures is somewhat elastic, analogous to the closing of cracks and grain boundaries that we expect in sandstones.
The high pressure asymptotic behavior shows a continued increase in velocity rather than a flat limit. This is probably due to permanent plastic deformation of the shale.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
23
Cotton Valley shale (ultrasonic), from Tosaya, 1982.
F27
2
2.5
3
3.5
4
4.5
5
0 200 400 600 800 1000 1200 1400
Cotton Valley Shale
Velo
city
(km
/s)
Pc (bars)
VpPp = 10
Pp = 400DRY
Vs Pp = 10
Pp = 400
DRY
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
24
Chalk (ultrasonic), from Gregory, 1976, replotted fromBourbié, Coussy, and Zinszner, 1987, Gulf Publishing Co.
1.5
2
2.5
3
3.5
0 10 20 30 40 50
Chalk
Velo
city
(km
/s)
Effective Pressure (MPa)
Sw
100
0
5102040
80
60
VP
VS
5
φ = 30.6%
F.12
For chalks, we also see an increase of velocity with effective pressure. The rapid increase of velocity at low pressures is somewhat elastic and reversible.
The high pressure asymptotic behavior shows a continued increase in velocity rather than a flat limit. This is probably due to permanent crushing of the fragile pore space.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
25
Velocity vs. pressure at various temperatures in water-saturated Westerly granite, from Nur (1980).
4.8
5
5.2
5.4
5.6
5.8
0 100 200 300 400 500
P-Ve
loci
ty (k
m/s
)
Confining Pressure (MPa)
200 oC
200 oC
300 oC
300 oC
400 oC
400 oC
Dry
WetP
p = P
c (lower effective pressure than the dry)
F.13
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
26
Velocity vs. temperature in Berea Sandstone. (a) dryultrasonic (Mobarek, 1971), (b,c) saturated resonant bar
(Jones, 1983), replotted from Bourbié, et al., 1987.
1.55
1.6
1.65
0 20 40 60 80 100 120
Berea sandstoneWater saturated
Exte
nsio
nal V
eloc
ity (k
m/s
)
Temperature (degrees C)
pp
=4.5 MPa
pc=10 MPa
V E
2.8
3
3.2
3.4
3.6
3.8
4
0 20 40 60 80 100
P-W
ave
Velo
city
(km
/s)
Confining Pressure (MPa)
20 oC100 oC200 oC
Berea Sandstone Dry
1
1.2
1.4
1.6
1.8
2
0 5 10 15 20
Berea sandstoneWater saturated
22 oC 0.95 cp 53 oC 0.52 cp 84 oC 0.34 cp110 oC 0.26 cp
S-W
ave
Velo
city
Confining Pressure (MPa)
Temp. η
pp=0.5 MPa
F.15
(a)
(b)
(c)
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
27
Influence of temperature on oil saturated samples, from Tosaya, et al. (1985).
2
2.5
3
3.5
0 50 100 150 200
Kern River sand
P Ve
loci
ty (k
m/s
)
Temperature oC
pe=10 MPa
100% oil
50% oil50% gas
100% gas
2
2.4
2.8
3.2
3.6
0 50 100 150 200
Venezuelan Oil Sand
P Ve
loci
ty (k
m/s
)
Temperature oC
100% oil
50% oil50% brine
100% brine
Pe = 10 MPa
F.16
We observe experimentally that velocities are most sensitive to temperature when the rocks contain liquid hydrocarbons (oil). We believe that this results from an increase of the oil compressibility and a decrease of the oil viscosity as the temperature goes up.
In field situations other factors can occur. For example, gas might come out of solution as the temperature goes up.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
28
0.5
0.6
0.7
0.8
0.9
1
1.1
-4 -2 0 2 4 6 8 10
Barre granite
µ/µµ/µµ/µµ/µοοοο
Log viscosity (poise)
Bedford limestone
Velocity vs. viscosity in glycerol saturated samples, from Nur (1980).
0.7
0.8
0.9
1
-4 -2 0 2 4 6 8 10
Bedford limestone
V/Vo
Log viscosity (poise)
VP
VS
F.17
Temperature
Temperature
In this experiment the pore fluid is glycerol, whose viscosity is extremely sensitive to temperature. The data show a classical viscoelastic behavior with lower velocity at low viscosity and higher velocity at higher viscosity. Viscosity is one of several pore fluid properties that are sensitive to temperature.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
29
Compressional velocities in the n-Alkanes vs. temperature. A drastic decrease of velocity with temperature! The numbers in the figure represent carbon numbers. From Wang, 1988, Ph.D. dissertation, Stanford University.
700
800
900
1000
1100
1200
1300
1400
-20 0 20 40 60 80 100 120 140
Alkanes
Com
pres
sion
al V
eloc
ity (m
/s)
Temperature (ÞC)
6
7
8
1011
1214 15161822
2836
F.18
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
30
800
900
1000
1100
1200
1300
1400
0 0.05 0.1 0.15 0.2
Alkanes
Com
pres
sion
al V
eloc
ity (m
/s)
1/Carbon Number
110 ÞC
75 ÞC
50 ÞC
22 ÞC
Compressional velocities in the n-Alkanes vs. inverse of the carbon numbers, at different temperatures. From Wang, 1988, Ph.D. dissertation, Stanford University.
F.19
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
31
Batzle-Wang formulas for fluid density and bulk modulus (Geophysics, Nov. 1992).
0
100
200
300
400
500
600
0 50 100 150 200
Gas Density
Gas gravity = 0.6Gas gravity = 1.2
Gas
Den
sity
(kg/
m3)
Temperature (degrees C)
P = 0.1 MPa
P = 25 MPa
P = 50 MPa
P = 25 MPa
P = 50 MPa
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 50 100 150 200
Gas Bulk Modulus
Gas gravity = 0.6Gas gravity = 1.2
Gas
Bul
k M
odul
us (G
Pa)
Temperature (degrees C)
0.1 MPa25 MPa
P = 50 MPa
25 MPa
P = 50 MPa
0.1 MPa
F30
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
32
600
650
700
750
800
850
900
950
0 50 100 150 200
Oil Density
API gravity = 10API gravity = 50
Oil
Den
sity
(kg/
m3)
Temperature (degrees C)
0.1 MPa25 MPa
P = 50 MPa
25 MPaP = 50 MPa
0.1 MPaGOR = 100
0
0.5
1
1.5
2
2.5
0 50 100 150 200
Oil Bulk Modulus
API gravity = 10API gravity = 50
Oil
Bulk
Mod
ulus
(GPa
)
Temperature (degrees C)
0.1 MPa25 MPa
P = 50 MPa
25 MPa
P = 50 MPa
0.1 MPa
F30
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
33
1
1.5
2
2.5
3
3.5
4
0 50 100 150 200
Brine Bulk Modulus
Salinity = 0Salinity = 240000 ppm
Brin
e Bu
lk M
odul
us (G
Pa)
Temperature (degrees C)
0.1 MPa
25 MPaP = 50 MPa
25 MPaP = 50 MPa
0.1 MPa
F30
850
900
950
1000
1050
1100
1150
1200
1250
0 50 100 150 200
Brine Density
Salinity = 0Salinity = 240000 ppm
Brin
e D
ensi
ty (k
g/m
3)
Temperature (degrees C)
0.1 MPa25 MPa
P = 50 MPa
25 MPaP = 50 MPa
0.1 MPa
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
34
Fluid Properties
• The density and bulk modulus of most reservoir fluids increase as pore pressure increases.
• The density and bulk modulus of most reservoir fluids decrease as temperature increases.
• The Batzle-Wang formulas describe the empirical dependence of gas, oil, and brine properties on temperature, pressure, and composition.
• The Batzle-Wang bulk moduli are the adiabaticmoduli, which we believe are appropriate for wave propagation.
• In contrast, standard PVT data are isothermal. Isothermal moduli can be ~20% too low for oil, and a factor of 2 too low for gas. For brine, the two don’t differ much.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
35
Stress-Induced Velocity Anisotropy -- the Historical Basis for Seismic Fracture Detection
Stress-induced velocity anisotropy in Barre Granite (Nur, 1969). In this classic experiment, Nur manipulated the crack alignment by applying uniaxial stress. Initially the rock is isotropic, indicating an isotropic distribution of cracks. Cracks normal to the stress (or nearly so) closed, creating crack alignment and the associated anisotropy.
3.6
3.8
4
4.2
4.4
4.6
4.8
5
0 20 40 60 80
Vp (k
m/s
ec)
angle from stress axis (degrees)
stress (bars)
0
100
200
300
2.6
2.7
2.8
2.9
3
3.1
3.2
0 20 40 60 80
Vs (S
H) (
km/s
ec)
angle from stress axis (degrees)
stress (bars)
0
100
200
300
400
Measurement angle related to theuniaxial stress direction
F.20
2.6
2.7
2.8
2.9
3
3.1
3.2
0 20 40 60 80
Vs (S
V) (k
m/s
ec)
angle from stress axis (degrees)
stress (bars)
0
100
200
300
400
P and S waves
θ
Axial Stress
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
36
Stress-Induced Velocity AnisotropyDue to Crack Opening Near Failure
Uniaxial stress-induced velocity anisotropy in WesterlyGranite (Lockner, et al. 1977).
P
PSource
Source
Source
S
S S
SII
II
Frac
ture
IIAS
0.6
0.7
0.8
0.9
1
1.1
1.2
0 20 40 60 80 100
PS⊥⊥⊥⊥SIISIIA
V/V
o
Percent Failure Strength
T
T
F.21
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
37
Seismic Anisotropy Due to Rock Fabric
Isotropic mixtureslight alignment
layered
102030405060708090
0 0.2 0.4 0.6 0.8 1
ρVp2
fabric anisotropy
vertical (c33 )
horizontal (c11 )
Virtually any rock that has a visual layering or fabric at a scale finerthan the seismic wavelength will be elastically and seismicallyanisotropic. Sources can include elongated and aligned grains andpores, cracks, and fine scale layering. Velocities are usually faster
for propagation along the layering.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
38
Velocity Anisotropy Due to Fabric
Anisotropic velocities vs. pressure. (a) and (b) Jones (1983), (c) Tosaya (1982).
3.5
4
4.5
5
5.5
6
0 20 40 60 80 100 120
GNEISS - Dry
ll foliation ⊥⊥⊥⊥ foliationll foliation
Com
pres
sion
al V
eloc
ity (k
m/s
)
Confining Pressure (MPa)
5.2
5.4
5.6
5.8
6
6.2
6.4
0 20 40 60 80 100 120
MYLONITE - Dry
⊥⊥⊥⊥ foliation, ll lineation⊥⊥⊥⊥ foliation⊥⊥⊥⊥ foliation, ⊥⊥⊥⊥ lineation
Com
pres
sion
al V
eloc
ity (k
m/s
)
Confining Pressure (MPa)
V33
2
1
V31 Vp 45Þ
V22 = V11
V21
V23 = V13 = V31
3
2
2.5
3
3.5
4
4.5
5
5.5
6
0 20 40 60 80 100 120 140
Pierre shaleSw = 100%
Velo
city
(km
/s)
Effective Pressure (MPa)
V11Vp 45Þ
V33
Vs V12
V13
V31
F.22
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
39
2
2.5
3
3.5
4
4.5
5
5.5
0 200 400 600 800 1000 1200
Cotton Valley Shale - Water Saturated
Velo
city
(km
/s)
Peff (bars)
V P
V S
V11
Vp 45 o
V33
V12
V13
V31
Velocity Anisotropy in Shale
Cotton Valley shale (ultrasonic), from Tosaya, 1982.
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
40
Velocity Anisotropy Resulting FromThinly Layered Kerogen
P-wave anisotropy in shales (from Vernik, 1990): (1) Bakken black shales, (2) Bakken dolomitic siltstone, (3) Bakken shaly dolomite, (4) Chicopee shale (Lo, et al, 1985).
Vernik found that kerogen-bearing shales can have verylarge anisotropy, easily 50%.
1
1.1
1.2
1.3
1.4
1.5
0 10 20 30 40 50 60 70 80
P-w
ave
Ani
sotr
opy
Confining Pressure (MPa)
1
1
1
432
C=13%
22.4%
30%
<0.5%
F.23
Stanford Rock Physics Laboratory - Gary Mavko
Parameters That Influence Seismic Velocity
41
Velocities in kerogen-rich Bakken shales (Vernik, 1990)and other low porosity argillaceous rocks (Lo et al., 1985; Tosaya, 1982; Vernik et al., 1987). Compiled by
Vernik, 1990.
1
2
3
4
0 20 40 60 80
S-w
ave
Velo
city
Vs(
0), k
m/s
Confining Pressure, MPa
Kola Phyllite, C=2-3%
Chicopee Shale, C<1%
Cotton Valley Shale, C<1%
Bakken Black Shales, C=13-30%
2
3
4
5
6
0 20 40 60 80
P-w
ave
Velo
city
Vp(0
), km
/s
Confining Pressure, MPa
Kola Phyllite, C=2-3%
Chicopee Shale, C<1%
Cotton Valley Shale, C<1%
Bakken Black Shales, C=13-30%
F.24
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
42
Bounding Methods for Estimating Effective Elastic Moduli
For many reasons we would like to be able to model or estimate the effective elastic moduli of rocks in terms of the properties of the various constituent minerals and pore fluids. To do it precisely one must incorporate
• the individual elastic moduli of the constituents
• the volume fractions of the constituents
• geometric details of how the various constituents
are arranged
The geometric details are the most difficult to know or measure. If we ignore (or don’t know) the details of geometry, then the best we can do is estimate upper and lower bounds on the moduli or velocities.
The bounds are powerful and robust tools. They give rigorous upper and lower limits on the moduli, given the composition. If you find that your measurements fall outside the bounds, then you have made a mistake - in velocity, volume fractions, or composition!
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
43
Voigt and Reuss Bounds
On a strictly empirical basis one can imagine defining a power law average of the constituents
where
Special cases are the Voigt average (an upper bound):
and the Reuss average (a lower bound):
Since these are upper and lower bounds, an estimate of the actual value is sometimes taken as the average of the two, known as the Voigt-Reuss-Hill average:
Mαααα = f1M1αααα + f2M2
αααα + f3M3αααα + ...
MV = f1M1 + f2M2
MR–1 = f1M1
–1 + f2M2–1
MVRH = MV + MR2
= the effective modulus of the composite
Mi = the modulus of the ith constituent
fi = the volume fraction of the ith constituent
αααα = a constant, generally between -1 and +1
M
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
44
The Voigt and Reuss averages are interpreted as the ratio of average stress and average strain within the composite.
The stress and strain are generally unknown in the composite and are expected to be nonuniform. The upper bound (Voigt) is found assuming that the strain is everywhere uniform. The lower bound (Reuss) is found assuming that the stress is everywhere uniform.
Geometric interpretations:
E = σσσσεεεε = σσσσ
ΣΣΣΣfiεεεεi
= σσσσfi σσσσ
EiΣΣΣΣ
E = σσσσ
εεεε = ΣΣΣΣfiσσσσ iεεεε
= fi εεεεEiΣΣΣΣεεεε
Voigt iso-strain model Reuss iso-stress model
E = ΣΣΣΣfiEi
1E
= ΣΣΣΣ fiEi
Since the Reuss average describes an isostress situation,it applies perfectly to suspensions and fluid mixtures.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
45
Backus Average for Thinly Layered MediaBackus (1962) showed that in the long wavelength limit a stratified medium made up of thin layers is effectively anisotropic. It becomes transversely isotropic, with symmetry axis normal to the strata. The elastic constants (see next page) are given by:
where
M = 12 A – B
A B F 0 0 0B A F 0 0 0F F C 0 0 00 0 0 D 0 00 0 0 0 D 00 0 0 0 0 M
,
A = 4µµµµ λλλλ + µµµµ
λλλλ + 2µµµµ + 1λλλλ + 2µµµµ
–1 λλλλλλλλ + 2µµµµ
2
B = 2µµµµλλλλλλλλ + 2µµµµ + 1
λλλλ + 2µµµµ–1 λλλλ
λλλλ + 2µµµµ2
C = 1λλλλ + 2µµµµ
–1
F = 1λλλλ + 2µµµµ
–1 λλλλλλλλ + 2µµµµ
D = 1µµµµ
–1
M = µµµµ
are the isotropic elastic constants of the individual layers. The brackets indicate averages of the enclosed properties, weighted by their volumetric proportions. This is often called the Backus average.
λλλλ, µµµµ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
46
Hooke’s law relating stress and strain in a linear elasticmedium can be written as
elastic stiffnesses (moduli) elastic compliancesA standard shorthand is to write the stress and strain as vectors:
T =
σ 1= σ 11σ 2= σ 22σ 3= σ 33σ 4= σ 23σ 5= σ 13σ 6= σ 12
E =
e1= ε11e2= ε22e3= ε33e4=2ε23e5=2ε13e6=2ε12
σ 1σ 2σ 3σ 4σ 5σ 6
=
A B F 0 0 0B A F 0 0 0F F C 0 0 00 0 0 D 0 00 0 0 0 D 00 0 0 0 0 M
e1e2e3e4e5e6
Note the factor of 2 in the definition of strains.
The elastic constants are similarly written in abreviated form, and the Backus average constants shown on the previous page now have the meaning:
σ ij = cijkl εklΣkl
ε ij = Sijkl σ klΣkl
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
47
Velocity-porosity relationship in clastic sediments compared with the Voigt and Reuss bounds. Virtually all of the points indeed fall between the bounds. Furthermore, the suspensions, which are isostress materials (points with porosity > 40%) fall very close to the Reuss bound.
Data from Hamilton (1956), Yin et al. (1988), Han et al. (1986). Compiled by Marion, D., 1990, Ph.D. dissertation, Stanford Univ.
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100suspensionssand-clay mixt.sandclay-free sandstoneclay-bearing sandstone
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt Avg.
Reuss Avg.(Wood's Relation)
G.1
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
48
Hashin-Shtrikman Bounds
Interpretation of bulk modulus:
where subscript 1 = shell, 2 = sphere. f1 and f2 are volume fractions.
These give upper bounds when stiff material is K1, µµµµ1(shell) and lower bounds when soft material is K1, µµµµ1.
KHS± = K1 + f2
K2 – K1–1 + f1 K1 + 4
3 µµµµ1–1
µµµµHS± = µµµµ1 + f2µµµµ2 – µµµµ1
–1 + 2f1 K1 + 2µµµµ1
5µµµµ1 K1 + 43µµµµ1
The narrowest possible bounds on moduli that we can estimate for an isotropic material, knowing only the volume fractions of the constituents, are the Hashin-Shtrikman bounds. (The Voigt-Reuss bounds are wider.) For a mixture of 2 materials:
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
49
Hashin-Shtrikman Bounds
A more general form that applies when more than two phases are being mixed (Berryman, 1993):
where
indicates volume average over the spatially varying K(r), µµµµ(r) of the constituents.
KHS+ = ΛΛΛΛ µµµµ max , KHS– = ΛΛΛΛ µµµµ min
µµµµHS+ = ΓΓΓΓ ζζζζ Kmax,µµµµmax , µµµµ HS– = ΓΓΓΓ ζζζζ Kmin,µµµµmin
ΛΛΛΛ z = 1K r + 4
3z–1
– 43z
ΓΓΓΓ z = 1µµµµ r + z
–1– z
ζζζζ K,µµµµ = µµµµ6
9K + 8µµµµK + 2µµµµ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
50
Bounds on the bulk and shear moduli of two-phase isotropic aggregates. Also shown is the Voigt-Reuss-Hill average (VRH). We observe that the upper and lower bounds are far apart when the materials being mixed are dissimilar, but close when materials are similar. Hence, we can often use the upper and lower bounds (or the VRH average) to model a mixture of minerals. But bounds are often not so useful when mixing minerals and pore-filling materials.From Watt, Davies, and O'Connell, 1976, Reviews of Geophysics and Space Physics, 14, 541-563.
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
K*/K
MgO
Volume Fraction AgCl
Voigt
HS+
HS-
VRH
Reuss
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
K*/K
stis
hovi
te
Volume Fraction MgO
Voigt
HS+, HS-
VRH
Reuss
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
µµ µµ */µµ µµ
MgO
Volume Fraction AgCl
Voigt
HS+
HS-
VRH
Reuss
0.996
0.997
0.998
0.999
1
0 0.2 0.4 0.6 0.8 1
µµ µµ */µµ µµ
stis
hovi
te
Volume Fraction MgO
Voigt, HS+, VRH, HS-, Reuss
K*/KMgO µµµµ*/µµµµMgO
K*/Kstishovite µµµµ*/µµµµstishovite
G.2
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
51
Distance between bounds depends on similarity/difference of end-member constituents. Here we see that a mixture of calcite and water gives widely spaced bounds, but a mixture of calcite and dolomite gives very narrow bounds.
697071727374757677
0 0.2 0.4 0.6 0.8 1
Calcite + Dolomite
Bul
k M
odul
us (G
Pa)
fraction of dolomite
HS-, HS+, Reuss, Voigt, VRH
01020304050607080
0 0.2 0.4 0.6 0.8 1
Calcite + Water
Bul
k M
odul
us (G
Pa)
porosity
HS+
HS-, Reuss
Voigt
VRH
G13
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
52
Wyllie Time Average
1
2
3
d1
d2
d3
D
Wyllie et al. (1956, 1958, 1962) found that travel time through water saturated consolidated rocks could be approximately described as the volume weighted average of the travel time through the constituents:
t = DV
1V
= f1V1
+ f2V2
+ f3V3
1V = d1 /D
V1+ d2 /D
V2+ d3 /D
V3
DV
= d1V1
+ d2V2
+ d3V3
t = t1 + t2 + t3
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
53
Limitations:
• rock is isotropic• rock must be fluid-saturated• rock should be at high effective pressure• works best with primary porosity• works best at intermediate porosity• must be careful of mixed mineralogy (clay)
The time-average equation is heuristic and cannot be justified theoretically. It is based on ray theory which requies that (1) the wavelength is smaller than the grain and pore size, and (2) the minerals and pores are arranged in flat layers.
Note the problem for shear waves where one of the phases in a fluid, Vs-fluid →→→→ 0!
Wyllie’s generally works best for
• water-saturated rocks• consolidated rocks• high effective pressures
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
54
Modification of Wyllie's proposed by Raymer
Still a strictly empirical relation.
This relation recognizes that at large porosities (φφφφ > 47%) the sediment behaves as a suspension, with the Reuss average of the P-wave modulus, M = ρρρρVp2.
V = (1 – φφφφ)2 Vmineral + φφφφ Vfluid φφφφ < 37%
1ρρρρV2 = φφφφ
ρρρρfluidVfluid2 + 1 – φφφφ
ρρρρmineralVmineral2 φφφφ > 47%
1V = 0.47 – φφφφ
0.101
V37+ φφφφ – 0.37
0.101
V4737% < φφφφ < 47%
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
55
Comparison of Wyllie's time average equationand the Raymer equations with Marion's compilation of shaly-sand velocities from Hamilton (1956), Yin et al. (1988), Han et al. 1986).
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100suspensionssand-clay mixt.sandclay-free sandstoneclay-bearing sandstone
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt Avg.
Wood's Relation(Reuss Avg.)
Wyllie
Raymer
G.3
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
56
Seismic Fluid Substitution
Pore fluids, pore stiffness,and their interaction
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
57
Typical Problem: Analyze how rock properties, logs, and seismic change, when pore fluids change.
Flood sandyintervals with brine
Example: We observe Vp, Vs, and density at a well and compute a synthetic seismic trace, as usual. Predict how the seismic will change if the fluid changes -- either over time at the same position, or if we move laterally away from the welland encounter different fluids in roughly the same rocks.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
58
Effective moduli for specific pore and grain geometries
Imagine a single linear elastic body. We do two separate experiments--apply stresses σσσσ1 and observe displacements u1, then apply stresses σσσσ2 and observe displacements u2.
The Betti-Rayleigh reciprocity theorem states that the work done by the first set of forces acting through the second set of displacements is equal to the work done by the second set of forces acting through the first set of displacements.
σσσσij(1), u(1)
∆σ∆σ∆σ∆σ
σσσσij(2), u(2)
∆σ∆σ∆σ∆σ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
59
Estimate of Dry Compressibility
Applying the reciprocity theorem we can write:
Assumptions• minerals behave elastically• friction and viscosity not important• assumes a single average mineral
∆σ∆σ∆σ∆σ∆σ∆σ∆σ∆σVbulkKdry
– ∆σ∆∆σ∆∆σ∆∆σ∆vpore = ∆σ∆σ∆σ∆σ∆σ∆σ∆σ∆σVbulkKmineral
limit as
∆σ∆σ∆σ∆σ→→→→0
1Kdry
= 1Kmineral
+ 1Vbulk
∂∂∂∂vpore∂∂∂∂σσσσ
∆σ∆σ∆σ∆σ
∆σ∆σ∆σ∆σ∆σ∆σ∆σ∆σ
∆σ∆σ∆σ∆σ
∆σ∆σ∆σ∆σ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
60
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Kdr
y/K
min
eral
Porosity
Modified Voigt
Voigt
Reuss
Kφφφφ/Kmineral =
.4
.3
.2
.1
φφφφc
.5
P
porous glassss - high pressss - variable press
Relation of Rock Moduli to Pore Space Compressibility -- Dry Rock
where
G.4
A fairly general and rigorous relation between dry rock bulk modulus and porosity is
1K dry
= 1K mineral
+ φK φ
1K φ
= 1vpore
ŽvporeŽσ
is the pore space stiffness. This is a new concept that quantifies the stiffness of a pore shape.
K φ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
61
What is a “Dry Rock”?
Many rock models incorporate the concept of a dry rock or the dry rock frame. This includes the work byBiot, Gassmann, Kuster and Toksoz, etc, etc.
Caution: “Dry rock” is not the same as gas-saturated rock. The dry frame modulus in these models refers to the incremental bulk deformation resulting from an increment of applied confining pressure, with pore pressure held constant. This corresponds to a “drained” experiment in which pore fluids can flow freely in or out of the sample to insure constant pore pressure. Alternatively, it can correspond to an undrained experiment in which the pore fluid has zero bulk modulus, so that pore compressions do not induce changes in pore pressure – this is approximately the case for an air-filled sample at standard temperature and pressure. However, at reservoir conditions (high pore pressure), gas takes on a non-negligible bulk modulus, and should be treated as a saturating fluid.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
62
Relation of Rock Moduli to Pore Space Compressibility -- Saturated Rock
where Pore spacecompressibilitymodified by fluids.
A similar general relation between saturated rock bulk modulus and porosity is
1K sat
= 1K mineral
+ φK φ
K φ = K φ + K mineralK fluid
K mineral – K fluid ≈ K φ + K fluid
So we see that changing the pore fluid has the effect ofchanging the pore space compressibility of the rock. Thefluid modulus term is always just added to K φ
When we have a stiff rock with high velocity, then its valueof is large, and changes in do not have much effect. But a soft rock with small velocity will have a small
and changes in will have a much larger effect.
K φ
K φ
K fluid
K fluid
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
63
Bulk Modulus For a Saturated Rock
cracks
pores
≈ 1 cracks≈ 0.2 pores
1Ksat
= 1Kmineral
+ φφφφKφφφφ
– φφφφKmineral
∂∂∂∂P∂∂∂∂σσσσ
∂∂∂∂P∂∂∂∂σσσσ = 1
1 + Kφφφφ1
Kfluid– 1
Kmineral
≈≈≈≈ 11 +
KφφφφKfluid
1Kdry
> 1Ksat
≈≈≈≈ 1Kmineral
1Kdry
> 1Ksat
> 1Kmineral
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
64
Gassmann's Relations
KsatKmineral – Ksat
=Kdry
Kmineral – Kdry+ Kfluid
φφφφ Kmineral – Kfluid
1µµµµsat
= 1µµµµdry
These are Transformations! Pore space geometry and stiffness are incorporated automatically by measurements of Vp, Vs. Gassmann (1951) derived this general relation between the dry rock moduli and the saturated rock moduli. It is quite general and valid for all pore geometries, but there are several important assumptions:
• the rock is isotropic• the mineral moduli are homogeneous• the frequency is low
“Dry rock” is not the same as gas saturated rock.
Be careful of high frequencies, high viscosity, and clay.
Useful for Fluid Substitution problem:gas
oilwater
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
65
Some Other Forms of Gassmann
Ksat =φ 1
Kmin– 1
K fluid+ 1
K min– 1
Kdryφ
Kdry
1Kmin
– 1K fluid
+ 1Kmin
1K min
– 1Kdry
K sat = K dry +1 –
K dryKmin
2
φK fluid
+ 1 – φK min
–K dry
Kmin2
1Ksat
= 1Kmin
+ φ
K φ +K minK fluid
K min – K fluid
K dry =Ksat
φK minK fluid
+ 1 – φ – Kmin
φKminK fluid
+ KsatK min
– 1 – φ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
66
1. Extract Moduli from Velocities measured with fluid 1:
2. Transform the bulk modulus using Gassmann
where K1, K2 are dynamic rock moduli with fluids 1, 2
Kfl 1, Kfl 2 bulk moduli of fluids 1, 2ρρρρ1, ρρρρ2 density of rock with fluids 1, 2Kmin, φφφφ mineral modulus and porosityρρρρfl 1, ρρρρfl 2 density of fluids 1, 2
3. µµµµ2 = µµµµ1 shear modulus stays the same
4. Transform density
5. Reassemble the velocities
K2Kmin – K2
– Kfl 2φφφφ Kmin – Kfl 2
= K1Kmin – K1
– Kfl 1φφφφ Kmin – Kfl 1
Fluid Substitution Recipe
VP =K2 + 4
3 µµµµ2ρρρρ2
VS = µµµµ2ρρρρ2
ρρρρ 2 = 1 – φφφφ ρρρρmin + φφφφρρρρ fl 2 = ρρρρ1 + φφφφ ρρρρ fl 2 – ρρρρ fl 1
K1 = ρρρρ1(VP2 – 4
3VS2), µµµµ1 = ρρρρ1VS
2
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
67
Why is the shear modulus unaffected by fluids in Gassmann’s relations?
Imagine first an isotropic sample of rock with a hypothetical spherical pore. Under “pure shear”loading there is no volume change of the rock sampleor the pore -- only shape changes. Since it is easy tochange the shape of a fluid, the rock stiffness is notaffected by the type of fluid in the pore.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
68
Why do the Gassmann relationsonly work at low frequencies?
Imagine an isotropic sample of rock with cracks at all orientations. Under “pure shear” loading there is no volume change of the rocksample or the pore space, because some cracks open while others close. If the frequency is too high, there is a tendency for local pore pressures to increase in some pores and decrease in others: hence the rock stiffness depends on the fluid compressibility.
However, if the frequency is low enough, the fluid has time to flow and adjust: there is no net pore volume change and therefore the rock stiffness is independent of the fluids.
This crack decreases involume. Its pore pressurelocally increases if the fluid cannot flow out of the crack.
This crack increases involume. Its pore pressurelocally decreases if the fluidcannot flow into the crack.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
69
Graphical Interpretation of Gassmann's Relations
1. Plot known effective modulus K, with initial fluid.
2. Compute change in fluid term:
3. Jump vertically up or down that number of contours.
Example: for quartz and water ~ 3 contours.
∆∆∆∆KfluidKmineral
≈≈≈≈ .06
∆∆∆∆KmineralKfluid
Kmineral – Kfluid≈≈≈≈ ∆∆∆∆Kfluid
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
0
0.2
0.4
0.6
0.8
1
K/K
min
eral
Porosity
K φφφφ/K mineral = .1
.4
.3
.2
.5
K flu
id/K
min
eral
( φφ φφ =
1 o
nly)
A
A'
B
B'
~
G.6
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
70
Graphical Interpretation of Gassmann's Relations
1. Plot the known modulus with initial fluid (point A).2. Identify Reuss averages for initial and final fluids.3. Draw straight line through through A to initial Reuss
curve.4. Move up or down to new Reuss Curve and draw
new straight line.5. Read modulus with new fluid (point A').
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 10
0.2
0.4
0.6
0.8
1
K/K
min
eral
Porosity
K φφφφ/K mineral = .1
.4
.3
.2
.5
K flu
id/K
min
eral
(φφ φφ
= 1
only
)A
A'
Reuss (water)
Reuss (air) water
φφφφ c
~
air
G.7
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
71
Approximate Gassmann Relationwhen Shear Velocity is Unknown
Normally, to apply Gassmann's relations, we needto know both Vp and Vs so that we can extract thebulk and shear moduli:
and then compute the change of bulk modulus withfluids using the usual expression:
The problem is that we usually don't know Vs.
One approach is to guess Vs, and then proceed.
We have also found that a reasonably good approximation to Gassmann is
where M is the P-wave modulus:
K = ρρρρ VP
2 – 43 VS
2 µµµµ = ρρρρ VS2
KsatKmineral - Ksat
=Kdry
Kmineral - Kdry+
Kfluidφφφφ(Kmineral - Kfluid)
MsatMmineral - Msat
≈≈≈≈Mdry
Mmineral - Mdry+
Mfluidφφφφ(Mmineral - Mfluid)
M = ρρρρ VP2
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
72
Approximate Gassmann RelationWhen Shear Velocity is Unknown
Vp sat From Exact Gassmann’s Equation
2.5
3
3.5
4
4.5
5
5.5
6
2.5 3 3.5 4 4.5 5 5.5 6
Vsat - 40 Full
Vsat - 30
Vsat - 20
Vsat - 10
Vsat - 5
x = y
Predictions of saturated rock Vp from dry rock Vp are virtually the same for the approximate and exact forms of Gassmann’s relations.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
73
Gassmann's is a Low Frequency Relation
Vp Sat Measured (high frequency)
2.5
3
3.5
4
4.5
5
5.5
6
2.5 3 3.5 4 4.5 5 5.5 6
5 MPa Wet M1
5 MPa Wet M2
5 MPa Wet Exact
40 MPa Wet M1
40 MPa Wet M2
40 MPa Wet Exact
x = y
It is important to remember that Gassmann’s relations assume low frequencies. Measured ultrasonic Vp in saturated rocks is almost always faster than saturated Vp predicted from dry rock Vp using Gassmann. Data here are for shaly sandstones (Han, 1986).
VelocityDispersion
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
74
Water Flood Example: Pore PressureIncrease and Change From Oil to Brine
Calculated using Gassmann from dry lab data from Troll (Blangy, 1992).Virgin condition taken as low frequency, oil saturated at Peff=30 MPa.Pressure drop to Peff=10 MPa, then fluid substitution to brine.Koil = 1., Kbrine = 2.2
G.12
2 2.5
1250
1300
1350
Brine Flood into Oil
Vp (km/s)
dept
h (m
)
Pressure
oil to water
original oil
oil atincreased Pp
brine atincreased Pp
1.61.8
22.22.42.62.8
0 5 10 15 20 25 30 35
Vp (k
m/s
)
Eff pressure (MPa)
brineoilgas
One typical depth point
• effect of pressure on frame• effect of pressure on fluids• frame+fluid: fluid substitution
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
75
Gas Flood Example: Pore PressureIncrease and Change From Oil to Gas
Calculated using Gassmann from dry lab data from Troll (Blangy, 1992).Virgin condition taken as low frequency, oil saturated at Peff=30 MPa.Pressure drop to Peff=10 MPa, then fluid substitution to gas.Koil = 1., Kbrine = 2.2
1.61.8
22.22.42.62.8
0 5 10 15 20 25 30 35
Vp (k
m/s
)
Eff pressure (MPa)
brineoilgas
G.12
1.8 2.4
1250
1300
1350
Gas Flood into Oil
Vp (km/s)
dept
h (m
)
Pressure
oil to gas
original oil
oil atincreased Pp
gas atincreased Pp
One typical depth point
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
76
Brine Flood Example: Pore PressureDecrease and Change From Oil to Brine
Calculated using Gassmann from dry lab data from Troll (Blangy, 1992).Virgin condition taken as low frequency, oil saturated at Peff=25 MPa.Pore pressure drop to Peff=30 MPa, then fluid substitution to brine.Koil = 1., Kbrine = 2.2
1.8 2.4 3
1250
1300
1350
Brine Flood with Pressure Decline
Vp (km/s)
dept
h (m
)
original oil
oil atdecreased Pp
brine atdecreased Pp
frame effectdecreased Peff
One typical depth point
1.61.8
22.22.42.62.8
0 5 10 15 20 25 30 35
Vp (k
m/s
)
Eff pressure (MPa)
brine
oil
gas
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
77
Calculated using Gassmann from dry lab data from Troll (Blangy, 1992).Koil = 1., Kbrine = 2.2
High Effective Pressure Stiffens the Rocks and Decreases Sensitivity to Pore Fluids
1.5 2 2.5 3
1250
1300
1350
Peff = 30 MPa
Vp (km/s)
dept
h (m
)
wateroil
gas
1.5 2 2.5 3
1250
1300
1350
Peff = 10 MPa
Vp (km/s)
dept
h (m
)
wateroil
gas
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
78
Velocity Dispersion Decreases ApparentSensitivity to Pore Fluids
From lab data on Troll sandstones (Blangy, 1992). Low frequency Vp calculated from lab, water-saturated, dispersive data using Gassmann. Illustrates that low frequency seismic might see larger differences than suggested by high frequency data.
1.5 2 2.5 3
1250
1300
1350
Vp
Vp (km/s)
dept
h (m
)
water
oil
lo f hi f
0 0.1 0.2 0.3
1250
1300
1350
Vp (brine) - Vp (oil)
delta Vp (km/s)
dept
h (m
)hi f lo f
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
79
Fluid Substitution in Anisotropic Rocks: Brown and Korringa’s Relations
where
effective elastic compliance tensor of dry rock
effective elastic compliance tensor of rock saturated with pore fluid
effective elastic compliance tensor of mineral
compressibility of pore fluid
compressibility of mineral material =
porosity
Sijkl(dry)
Sijkl(sat)
Sijkl0
ββββ fl
ββββ 0
φφφφ
Sααββααββααββααββ0
This is analogous to Gassmann’s relations. To apply it,one must measure enough velocities to extract the fulltensor of elastic constants. Then invert these for thecompliances, and apply the relation as shown.
Sijkl(dry) – Sijkl
(sat) =Sijαααααααα
(dry) – Sijαααααααα0 Sklαααααααα
(dry) – Sklαααααααα0
Sααααααααββ(dry) – Sααααααααββ
0 + β fl – β 0 φ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
80
3000
2500
2000
1500
1000 500
0.3
0.1
0.03 0.01
0.00
320
0040
0050
0030
00
Vs sat
Vp
sat
AV
O p
aram
eter
²νν νν/
(1- νν νν
)2
3000
2500
2000
1500
1000 500
2000
4000
5000
3000
Vs sat
Vp
sat
P-P
Ref
lect
ivity
0.4
0.25
0.15
0.1
0.06
3000
2500
2000
1500
1000 500
2000
4000
5000
3000
Vs sat
Vp
sat
0.8
0.6
0.4
0.2
0
AV
O p
aram
eter
(Vp
sat -
Vp
dry)
/Vp
ave
3000
2500
2000
1500
1000 500
2000
4000
5000
3000
Vs sat
Vp
sat
0.1
0.05
0 -0.0
5
-0.1
AV
O p
aram
eter
gra
dien
t ter
m
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
81
Marion (1990) discovered a simple, semi-empirical way to solve the fluid substitution problem. The Hashin-Shtrikman bounds define the range ofvelocities possible for a given volume mix of two phases, either liquid or solid. The vertical position within the bounds, d/D, is a measure of the relative geometry of the two phases. For a given rock, the bounds can becomputed for any two pore phases, 0 and 1. If we assume that d/D remains constant with a change of fluids, then a measured velocity with one fluid will determine d/D, which can be used to predict the velocity relative to the bounds for any other pore phase.
Bounding Average Method (BAM)
05
10152025303540
0 0.2 0.4 0.6 0.8 1
Fluid 2 (oil)
Bul
k M
odul
us (G
Pa)
Porosity
d'
D'
H-S Upper bound (M 2 + )
H-S Lower bound (M 2 - )
05
10152025303540
0 0.2 0.4 0.6 0.8 1
Fluid 1 (water)B
ulk
Mod
ulus
(GPa
)
Porosity
H-S Upper bound (M 1 + )
H-S Lower bound (M 1 - )d
D
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
82
Velocity in Massilon sandstone saturated with parowax. Data from Wang(1988). Wax saturated velocities were predicted using BAM, from Wang'smeasured velocities in the dry rock and in wax (from Marion, 1990)
2800
3000
3200
3400
3600
3800
4000
4200
0 20 40 60 80 100 120 140
Massillon Light Sandstone
P-Ve
loci
ty (m
/s)
Temperature ( °°°°C)
measured parowax
BAMcalculatedparowax
measured dry
G.8
An Example of the Bam Method. The wax saturated velocities are predicted from the dry
rock velocities.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
83
Velocity in dry and saturated Westerly granite. Data from Nur andSimmons (1969). Saturated velocities were predicted using BAM, from measured velocities in the dry rock (from Marion, 1990)
3000
3500
4000
4500
5000
5500
6000
6500
0 50 100 150 200 250 300
Westerley Granite
dry measuredwater sat. measuredBAM predicted
P-ve
loci
ty (m
/s)
Pressure (MPa)
HS+
HS-
G.9
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
84
Comparison of BAM and Gassmann predictions of saturated moduli in 10 different clay-free sandstones from measured dry velocities. BAM generally estimates the dispersion better than Gassmann (from Marion, 1990).
15000
20000
25000
30000
35000
40000
5 10 15 20 25
Bulk Modulus
water sat. measureddry measuredGassmann predictedBAM predicted
Bul
k M
odul
us (M
Pa)
Porosity (%)
-15
-10
-5
0
5
10
15
5 10 15 20 25
BAMGassmann
(Km
eas
- Kca
l)/K
mea
s (%
)
Porosity (%) G.10
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
85
bc
Ellipsoidal Models for Pore Deformation
Most deterministic models for effective moduli assume a specific idealized pore geometry in order to estimate the pore space compressibility:
The usual one is a 2-dimensional or 3-dimensional ellipsoidal inclusion or pore.
The quantity αααα = b/c is called the aspect ratio.
1
K φ=
1vpore
Žvpore
Žσ
Recall the general expression for the dry rock modulus:
1K dry
= 1K mineral
+ φK φ
Gassmann’s relation is a transformation, allowing us to predict how measured velocities areperturbed by changing the pore fluid. Now wediscuss a different approach in which we try to model the moduli “from scratch”.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
86
Estimating the Dry Rock Modulus
Mathematicians have worked out in great detail the 3-D deformation field U, of an oblate spheroid (penny-shaped crack) under applied stress. For example, the displacement of the crack face is:
We can easily integrate to get the pore volume changeand the dry modulus:
bc
σσσσ
σσσσ
An externally applied compression tends to narrow thecrack, with the faces displacing toward each other.
U(r) = σc
K mineral
4 1 – ν 2
3π 1 – 2ν 1 – rc
2
1K dry
= 1K mineral
+ 16 1 – ν 2
9 1 – 2ν1
K mineral
Nc 3
Vbulk
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
87
"Crack density parameter"
Dry Rock Bulk Modulus
Modulus depends directly on crack density. Crackgeometry or stiffness must be specified to get adependence on porosity.
1
Kdry= 1
Kmineral+ 16 1 – νννν2
9 1 – 2νννν1
Kmineral
Nc3
Vbulk
1
Kdry= 1
Kmineral1 + 16 1 – νννν2
9 1 – 2ννννNc3
Vbulk
1
Kdry= 1
Kmineral1 + 16 1 – νννν2
9 1 – 2νννν ∈∈∈∈
∈∈∈∈ = NVbulk
c3
≈≈≈≈ φφφφαααα
34ππππ
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
88
Crack Density Parameter
In these and other theories we often encounter the quantity:
This is called the Crack Density Parameter, and has the interpretation of the number of cracks per unit volume.
Example: 2 cracks per small cell. Each crack about 2/3 the length of a cell.
εεεε = Nc3
Vbulk
L2c
v = L3
εεεε = cL
3 ≈≈≈≈ 0.07
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
89
Distribution of Aspect Ratios
Modulus depends on the number of cracks andtheir average lengths
An idealized ellipsoidal crack will close when theamount of deformation equals the original crackwidth:
solving gives:
We generally model rocks as having a distributionof cracks with different aspect ratios. As thepressure is increased, more and more of themclose, causing the rock to become stiffer.
U = b
1
Kdry= 1
Kmineral+ 16 1 – νννν2
9Kmineral 1 – 2ννννNc3Vbulk
≈≈≈≈ ααααKmineral
σσσσclose ≈≈≈≈ ααααKmineral3ππππ4
1 – 2νννν1 – νννν2
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
90
Kuster and Toksöz (1974) formulation based on long-wavelength, first order scattering theory (non self-consistent)
KKT
* – KmKm + 4
3 µµµµm
KKT* + 4
3µµµµm= xi Ki – Km PmiΣΣΣΣi = 1
N
µµµµKT* – µµµµm
µµµµm + ζζζζmµµµµKT
* + ζζζζm= xi µµµµi – µµµµm QmiΣΣΣΣi = 1
N
ζζζζ = µµµµ6
9K + 8µµµµK + 2µµµµ
shape Pmi Qmi
Spheres
Needles
Disks
Penny cracks
Coefficients P and Q for some specific shapes. The subscripts m and i refer to the background and inclusion materials. From Berryman (1995).
15 1 + 8µµµµm
4µµµµi + παπαπαπα µµµµm+2ββββm+ 2
Ki + 23 µµµµi + µµµµm
Ki + 43 µµµµi + παπαπαπαββββ m
Km + 43 µµµµm
Ki + 43µµµµm
Km + µµµµm + 13 µµµµi
Ki + µµµµm + 13 µµµµi
Km + 43 µµµµi
Ki + 43µµµµi
Km + 43µµµµi
Ki + 43µµµµi + παπαπαπαββββm
µµµµm + ζζζζmµµµµi + ζζζζm
15
4µµµµmµµµµm + µµµµi
+ 2 µµµµm + γγγγmµµµµi + γγγγm
+Ki + 4
3µµµµm
Ki + µµµµm + 13 µµµµi
µµµµm + ζζζζiµµµµi + ζζζζi
ββββ = µµµµ (3K+µµµµ)(3K+4µµµµ)
γγγγ = µµµµ (3K+µµµµ)
(3K+7µµµµ)
ζζζζ = µµµµ6
(9K+8µµµµ)(K+2µµµµ)
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
91
Self-Consistent EmbeddingApproximation
Walsh's expression for the moduli in terms of the porecompressibility is fairly general. However attempts toestimate the actual pore compressibility are often basedon single, isolated pores.
The self-consistent approach uses a single porein a medium with the effective modulus.
Solving for Kdry gives:
1
Kdry= 1
Kmineral+ 16 1 – νννν2
9Kmineral 1 – 2ννννNc3
Vbulk
1
Kdry= 1
Kmineral+ 16 1 – νννν2
9Kdry 1 – 2ννννNc3
Vbulk
Kdry = Kmineral 1 – 16 1 – νννν2
9 1 – 2ννννNc3
Vbulk
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
92
Self-Consistent Approximations
O’Connell and Budiansky (1974) model for medium with randomly oriented thin dry cracks
K and µµµµ are the bulk and shear moduli of the uncracked medium, νννν is the Poisson’s ratio, and εεεε is the crack density parameter. The calculations are simplified by the approximation:
Assumes small aspect ratios (αααα →→→→ 0).
µµµµSC*
µµµµ = 1 – 3245
1 – ννννSC* 5 – ννννSC
*
2 – ννννSC* εεεε
KSC*
K = 1 – 169
1 – ννννSC*2
1 – 2ννννSC* εεεε
εεεε = 4516
νννν – ννννSC* 2 – ννννSC
*
1 – ννννSC*2 10νννν – 3ννννννννSC
* – ννννSC* εεεε
ννννSC* ≈≈≈≈ νννν 1 – 16
9 εεεε
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
93
shape P*i Q*i
Spheres
Needles
Disks
Penny cracks
Coefficients P and Q for some specific shapes. The * and i refer to the unknown effective and inclusion materials. From Berryman (1995).
ββββ = µµµµ (3K+µµµµ)(3K+4µµµµ)
γγγγ = µµµµ (3K+µµµµ)
(3K+7µµµµ)
ζζζζ = µµµµ6
(9K+8µµµµ)(K+2µµµµ)
Self-Consistent Approximations
Berryman’s (1980) model for N-phase composites
coupled equations solved by simultaneous iteration
xi K i – K * P*i = 0ΣΣΣΣi = 1
N
xi µµµµ i – µµµµ * Q*i = 0ΣΣΣΣi = 1
N
K* + 43µµµµ*
Ki + 43µµµµ*
µµµµ* + ζζζζ*µµµµ i + ζζζζ*
K* + µµµµ* + 13 µµµµi
Ki + µµµµ* + 13 µµµµi
15
4µµµµ*µµµµ* + µµµµi
+ 2µµµµ* + γγγγ*µµµµi + γγγγ* +
Ki + 43 µµµµ*
Ki + µµµµ* + 13 µµµµi
K* + 43µµµµi
Ki + 43µµµµ i
µµµµ* + ζζζζ iµµµµ i + ζζζζi
K* + 43µµµµ i
Ki + 43 µµµµi + παπαπαπαββββ*
15 1 + 8µµµµ*
4µµµµi + παπαπαπα µµµµ*+2ββββ* + 2Ki + 2
3 µµµµ i + µµµµ*
Ki + 43 µµµµi + παπαπαπαββββ*
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
94
Comparison of Han's (1986) sandstone data with modelsof idealized pore shapes. At high pressure (40-50 MPa),there seems to be some equivalent pore shape that is
more compliant than any of the convex circular or spherical models.
0
5
10
15
20
25
30
35
40
0 0.1 0.2 0.3 0.4
Shea
r Mod
ulus
(GPa
)
Equivalent Porosity
HS -
triangle
sphere
needle
HS +
0
5
10
15
20
25
30
35
0 0.1 0.2 0.3 0.4
Bul
k M
odul
us (G
Pa)
Equivalent Porosity
HS -
triangle
sphere
needle
HS +
G.11
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
95
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
interparticle porositymicro porositymoldic porosityintraframe porositydensely cemented
Velo
city
(m/s
)
porosity (percent)
spherical pores
tubular pores
crack-like pores(α = 0.1)
G14
Comparison of self-consistent elliptical crack modelswith carbonate data. The rocks with stiffer pore shapes are fit best by spherical pore models, whilethe rocks with thinner, more crack-like pores are fitbest by lower aspect ratio ellipsoids.
Data from Anselmetti and Eberli., 1997, in Carbonate Seismology, SEG.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
96
dry rock:cracks make the rock softer.
saturated “drained” rockwhen squeezed slowly, fluid can escape and rock behaves as though it is dry.
saturated “undrained” rockwhen jacketed or squeezed quickly, rock is stiffer.
Stanford Rock Physics Laboratory - Gary Mavko
Effective Medium Theories
97
soft stiff
Cracks deform the most when tractions are applied to their faces--in both normal and shear.
If the cracks are oriented randomly and equally in all directions, then the rock will probably behave isotropically. But if there is a preferred orientation, then the rock will be anisotropic: stiffer in some directions than other.
Crack Anisotropy
"Ex" "Ex"
"Ey" "Ey"Isotropic Anisotropic
"Ex" = "Ey" "Ex" > "Ey"c1111 c2222
Rock Physics Laboratory - Gary Mavko
Fluid Flow
98
Viscosity
σxz = 2η εxz
εxz = 12
ŽUxŽz
σxz = ηŽUxŽz
Shear stress in the fluid is proportional to the fluidvelocity gradient.
V
Stationary
z
x
Fluid Velocity Profile
where ηηηη is the viscosity. Or in terms of the strainrate:
Units:
1 Poise = 1 dyne–seccm2 = 0.1 newton–sec
m2
water at 20oC: η ≈ .01 Poise ≈ 1 centiPoise
Rock Physics Laboratory - Gary Mavko
Fluid Flow
99
Darcy’s Law:
where
volumetric flow ratepermeability of the mediumviscosity of the fluidcross sectional area
Differential form:
Q = – κκκκηηηηA∆∆∆∆P
∆∆∆∆I
Q =κκκκ =ηηηη =A =
U = – κκκκηηηη grad P
where is the filtration velocity U
PU
∆∆∆∆ l
P + ∆∆∆∆P
Rock Physics Laboratory - Gary Mavko
Fluid Flow
100
Units
Darcy’s law:
Permeability κκκκ has dimensions of area, or m2 in SI units. But the more convenient and traditional unit is the Darcy.
1 Darcy ≅≅≅≅ 10–12 m2
In a water saturated rock with permeability of 1Darcy, a pressure gradient of 1 bar/cm gives a flow velocity of 1 cm/sec.
Q = – κκκκ
ηηηηA∆∆∆∆P∆∆∆∆I
Rock Physics Laboratory - Gary Mavko
Fluid Flow
101
Kozeny-Carman RelationThe most common permeability model is to assume that rocks have nice round pipes for pore fluids to flow.
Compare this with general Darcy’s law:
Combining the two gives the permeability of a circular pipe:
We can rewrite this permeability in terms of familiar rock paramters, giving the Kozeny-Carman equation:
where: φφφφ is the porosityS is the specific pore surface areaττττ is the tortuosity
d is a typical grain diameterB is a geometric factor
κκκκ = Bφφφφ3
ττττ2S2
κκκκ = ππππR4
8A = ππππR2
AR2
8
κκκκ = Bφφφφ3d2ττττ
Q = – ππππR4
8ηηηη∆∆∆∆P∆∆∆∆I
Q = – κκκκηηηηA∆∆∆∆P
∆∆∆∆I
The classical solution for laminar flow through a circular pipe gives:
strong scale dependence!
2R
Rock Physics Laboratory - Gary Mavko
Fluid Flow
102
Schematic porosity/permeability relationship in rocks from Bourbié,Coussy, Zinszner, 1987, Acoustics of Porous Media, Gulf Publishing Co.
10 -9
10 -7
10 -5
10 -3
10 -1
10 1
1 10
Perm
eabi
lity
(Dar
cy)
Porosity (%)
Claysand
shales
Silts
Micriticsandstones
Shalysandstones
Granularlimestones
Crystallinerocks
Tightsediments
Clean coarse-grained sandstones
H.1
Rock Physics Laboratory - Gary Mavko
Fluid Flow
103
Demonstration of Kozeny-Carman relation in sintered glass,from Bourbié, Coussy, and Zinszner, 1987,
Acoustics of Porous Media, Gulf Publishing Co.
1
10
100
1000
0 10 20 30 40 50
280 µµµµm spheres50 µµµµm spheres
κκ κκ /d
2
(x10
e-6)
Porosity (%)
Sintered Glass
H.2
Here we compare the permeability for two synthetic porous materials having very different grain sizes. When normalized by grain-size squared, the data fall on top of each other -- confirming the scale dependence.
Rock Physics Laboratory - Gary Mavko
Fluid Flow
104
Porosity/permeability relationship in Fontainebleau sandstone,from Bourbié, Coussy, and Zinszner, 1987,
Acoustics of Porous Media, Gulf Publishing Co.
H.3
1
10
100
1000
10000
2 4 6 8 10
Per
mea
bilit
y (m
D)
Porosity (%)2 30
n = 8
n = 3
κκκκ = a φφφφ n
A particularly systematic variation of permeability withporosity for Fontainebleau sandstone. Note that the slope increases at small porosity, indicating an exponent on porosity larger than the power of 3 predicted by the Kozeny-Carman relation.
Rock Physics Laboratory - Gary Mavko
Fluid Flow
105
0.00001
0.0001
0.001
0.01
0.1
Perm
eabi
lity
(mD
)
.05Porosity
.20.10
φφφφc ≈≈≈≈ 0.045
Kozeny-Carman Relation with Percolation
Hot-pressed Calcite (Bernabe et al, 1982), showing a good fit to the data using the Kozeny-Carmanrelation modified by a percolation porosity.
As porosity decreases from cementation and compaction, it is common to encounter a percolationthreshold where the remaining porosity is isolated ordisconnected. This porosity obviously does not contribute to permeability. Therefore, we suggest,purely heuristically, replacing giving
φφφφ →→→→ (φφφφ – φφφφc)
κκκκ = B (φφφφ – φφφφc)3 d2
H.4
κ = B (φ – .045)3 d2
Rock Physics Laboratory - Gary Mavko
Fluid Flow
106
Fused Glass Beads (Winkler, 1993)
0.0001
0.001
0.01
0.1
1
10
100
Perm
eabi
lity
/ D2 200 micron
Porosity
100 micron
50 micron
.05 .50.10
φφφφc ≈≈≈≈ 0.035
H.5
κ = B (φ – .035)3 d2
Rock Physics Laboratory - Gary Mavko
Fluid Flow
107
Fontainebleau Sandstone (Bourbié et al, 1987)
1
10
100
1000
10000
Perm
eabi
lity
(mD
)
Porosity.02 .30.05 .10
φφφφc ≈≈≈≈ 0.025
H.6
Here we show the same Fontainebleau sandstone dataas before with the Kozeny-Carman relation modified by a percolation porosity of 2.5%. This accounts for the increased slope at low porosities, while retaining theexponent of 3.
κ = B (φ – .025)3 d2
Rock Physics Laboratory - Gary Mavko
Fluid Flow
108
Diffusion
The stress-strain law for a fluid (Hooke’s law) is
which can be written as
combining with Darcy’s law:
gives the classical diffusion equation:
where D is the diffusivity
εεεεαααααααα = 1KP
∇∇∇∇2P = –ηηηηκκκκK
∂∂∂∂P∂∂∂∂t
∇∇∇∇2P = –1D
∂∂∂∂P∂∂∂∂t
∇∇∇∇••••U = 1K
∂∂∂∂P∂∂∂∂t
U = –κκκκηηηη ∇∇∇∇P
Rock Physics Laboratory - Gary Mavko
Fluid Flow
109
Examples of Diffusion Behavior
1-D diffusion from an initial pressure pulse
Standard result:
P = P0δδδδ x
P x,t = P0
4ππππDte
x2
–4Dt = P04ππππDt
eττττ–t
Characteristic time scale
ττττ = x2
4D
Rock Physics Laboratory - Gary Mavko
Fluid Flow
110
Examples of Diffusion Behavior
Sinusoidal pressure disturbance
Disturbance decays approximately as
ττττd = λλλλ2
4D
λλλλ
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
111
(1) Seismic velocities almost always increase with frequency, and
(2) Seismic waves are always attenuated as theytravel through rocks.
These two observations are usually intimately related. Both usually increase from dry to fluid saturated conditions, and both usually decrease with increasing effective pressure.
These effects complicate the comparison of laboratory and field data, but they also reveal details about the pore space and the pore fluids it contains.
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
112
In most rocks and sediments, the velocity tends toincrease with frequency. This is accompanied byattenuation. Attenuation tends to be highest in frequency range where velocity is increasing mostrapidly.
0
0.1
0.2
0.3
0.4
0.5
0.6
0
0.5
1
1.5
2
-5 0 5M
odul
us (M
/M0)
log( ωωωω/ωωωωr)
1/Q
M∆∆∆∆M
M0
M�
I.1
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
113
These plots (Winkler, 1985) show typical features of ultrasonic laboratoryvelocities in rocks. The low frequency Gassmann theory almost always
underestimates the saturated velocities relative to the dry velocities. The discrepancy is usually greatest at low effective pressure.
3
3.2
3.4
3.6
3.8
4
4.2
0 10 20 30 40 50
P-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
dry
saturatedBiot
(high f)
Gassmann(low f)
Brine
1.8
1.9
2
2.1
2.2
2.3
2.4
2.5
2.6
0 10 20 30 40 50
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
dry
saturated
Biot(high f)
Gassmann(low f)
Brine
I.2
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
114
The difference between dry and saturated velocities and the disagreementwith the low frequency Gassmann theory often increases with fluid
viscosity. Again the differences are greatest at low pressures.Data from Winkler (1985).
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
0 10 20 30 40 50
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
saturated
dry
Oil
Biot(high f)
Gassman(low f)
3
3.2
3.4
3.6
3.8
4
4.2
0 10 20 30 40 50
P-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
saturated
dry
Oil
Biot(high f) Gassmann
(low f)
I.3
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
115
Failure of Gassmann's theory to predict saturated ultrasonic velocitiesrelative to dry velocities. Navajo sandstone data from Coyner (1984).
4
4.2
4.4
4.6
4.8
5
5.2
0 20 40 60 80 100 120
P-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
SatDry
low frequency(Gassmann)
2.6
2.7
2.8
2.9
3
3.1
3.2
3.3
3.4
0 20 40 60 80 100 120
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
SatDry
low frequency(Gassmann)
I.4
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
116
E1
E2
η
ViscoelasticityWe have been talking about elastic materials where stress is proportional to strain.
volumetric
shear
general
Viscoelastic materials also depend on rate or history.
Maxwell model
Voigt model
Standard linear solid
σσσσ11 + σσσσ22 + σσσσ333 = K εεεε11 + εεεε22 + εεεε33
σσσσ ij = 2µµµµεεεε ij
εεεε ij =σσσσ ij2µµµµ +
σσσσ ij2ηηηη
σσσσ ij = 2ηηηηεεεε ij + 2µµµµεεεε ij
ηηηησσσσ ij + E1 + E2 σσσσ ij = E2 ηηηηεεεεij + E1εεεεij
σσσσ ij = λλλλδδδδ ijεεεεαααααααα + 2µµµµεεεεij
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
117
Wave Propagating in a Viscoelastic Solid
At any point, the stress and strain are out of phase:
The ratio of stress to strain is the complex modulus.
u x,t = u0exp –αααα ωωωω x exp i ωωωωt – kx
σσσσ = σσσσ0exp i ωωωωt – kx
εεεε = εεεε0exp i ωωωωt – kx – ϕϕϕϕ
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
118
Quality Factor “Q”
low Q: large dissipation high Q: small dissipation
Different views of Q:energy dissipated per wave cycle
peak strain energy of the wave
velocity
frequency
phase delay
amplitude loss per cycle
1Q = ∆∆∆∆W
2ππππW
1Q = ααααV
ππππf
1Q ≈≈≈≈ 1
ππππ ln u tu t + ττττ
1Q = tan–1 ϕϕϕϕ
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
119
frequency
spec
tra
(db) S(x1)
S(x2)
ln ratio
Spectral Ratio MethodWe can think of Q-1 as the fractional loss per wavelength or per oscillation. Therefore over a fixed distance there is a tendency for shorter wavelengths to attenuate more:
or
If we propagate the wave
Then we can compare the amplitudes at two different distances:
1Q ≈≈≈≈ ααααV
ππππf
αααα ≈≈≈≈ ππππfVQ
u = u0exp –ααααx
ln ux2
ux1= – αααα x2 – x1
ln ux2
ux1= – ππππf
QV x2 – x1
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
120
εεεε = εεεε0eiωωωωt
Standard Linear Solid
If we assume sinusoidal motion
Then we can write:
with the complex, frequency-dependent modulus
In the limits of low frequency and high frequency
σσσσ = σσσσ0eiωωωωt
M ωωωω = E2 E1 + iωηωηωηωη
E1 + E2 + iωηωηωηωη =M∞∞∞∞ M0 + i ωωωω
ωωωωrM0M∞∞∞∞
M∞∞∞∞ + i ωωωωωωωωr
M0M∞∞∞∞
M0 = E2E1
E1 + E2, ωωωω →→→→ 0
M∞∞∞∞ = E2 , ωωωω →→→→ ∞∞∞∞
Re M ωωωω =
M0M∞∞∞∞ 1 + ωωωωωωωωr
2
M∞∞∞∞ + ωωωωωωωωr
2M0
σσσσ 0 = M ωωωω εεεε0
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
121
Standard Linear Solid
1Q = Ml ωωωω
MR ωωωω=
ωωωωωωωωr
M0M∞∞∞∞ M∞∞∞∞ – M0
M0M∞∞∞∞ 1 + ωωωωωωωωr
2
Similarly, we can write Q as a function of frequency:
where
The maximum attenuation
occurs at
1Q = E2
E1 E1 + E2
ωωωωωωωωr
1 + ωωωωωωωωr
2
ωωωωr = E1 E1 + E2ηηηη
1Q max
= 12
E2E1 E1 + E2
1Q max
= 12
M∞∞∞∞ – M0M0M∞∞∞∞
≈≈≈≈ 12
∆∆∆∆MM
ωωωω = ωωωωr
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
122
Standard Linear Solid Model
Attenuation and velocity dispersion tend to be mostlocalized in frequency. Attenuation is largest wherevelocity is changing most rapidly with frequency.
0
0.1
0.2
0.3
0.4
0.5
0.6
0
0.5
1
1.5
2
-5 0 5M
odul
us (M
/M0)
log( ωωωω/ωωωωr)
1/Q
M∆∆∆∆M
M0
M�
I.5
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
123
Finally
V(ωωωω)V(ωωωω0)
= M0 + ∆∆∆∆MM0
= 1 + ∆∆∆∆MM0
Nearly Constant Q ModelLiu, et al. (1976) considered a model in which simple attenuation mechanisms are combined such that the attenuation is nearly a constant over a finite range of frequencies.
We can then write
which relates the velocity dispersion within the band of constant Q, to the value of Q and the frequency.
We can express as:
V(ωωωω)/V(ωωωω0)
Expanding for small and substituting in:
∆∆∆∆M /M
12
∆∆∆∆MM0
≈≈≈≈ 1ππππQ log ωωωω
ωωωω0
1Q ≈≈≈≈ ππππ
log ωωωωωωωω0
12
∆∆∆∆MM0
V ωωωωV ωωωω0
= 1 + 1ππππQ log ωωωω
ωωωω0
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
124
0
0.1
0.2
0.3
0.4
0.5
0.6
0
0.4
0.8
1.2
1.6
2
-6 -4 -2 0 2 4 6 8M
odul
uslog( )ω
1/Q
M
nearly constant
Nearly Constant Q Model
Attenuation is nearly constant over a finite range offrequencies. It is sometimes interpreted as a super-position of individual (Standard Linear Solid) attenuation peaks. The broadening of the attenuationpeak is accompanied by a broadening of the range offrequency where velocity increases.
I.6
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
125
As with the nearly constant Q model, we can simplify this for large Q, giving:
Constant Q Model
Kjartansson (1979) considered a model in which Q is strictly constant. In this case the complex modulus and Q are related by:
where
M ωωωω = M0iωωωωωωωω0
2γγγγ
1Q ≈≈≈≈ ππππ
log ωωωωωωωω0
12
∆∆∆∆MM0
γγγγ = 1
ππππarctan 1Q
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
126
1/Q
M
ω
log( )
Constant Q Model
Attenuation is constant for all frequencies, andvelocity always increases with frequency.
I.7
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
127
Fontainebleau Sandstone
Just as velocity increases with effective pressure, so does Q. The strong pressure dependence is a clue that cracks are important for the physical mechanism of attenuation. From Nathalie Lucet, 1989, Ph.D. dissertation, Univ. of Paris/IFP.
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 5 10 15 20 25 30 35 40
Velocity
BR-EBR-SUS-PUS-S
Velo
city
(m/s
)
Pressure (MPa)
0
50
100
150
200
0 5 10 15 20 25 30 35 40
Attenuation
BR-EBR-SUS-PUS-S
1000
/Q
Pressure (MPa)
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 50 100 150 200 250 300
Velocity Dispersion
ExtensionTorsionCQ ModelNCQ Model
Vbr/V
us
1000/Qbr I.8
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
128
Some values of Q in rocks and sediments, summarizedby Bourbié, Coussy, and Zinszner, 1987, Acoustics of
Porous Media, Gulf Publishing Co.
Location Type of rock Depth (m) Measurementfrequency (Hz)
Limon (Colorado) Pierre shale 0-225 50-450 32
Gulf Coast (30 kmsouth of Houston)
Loam/sand/claySands and shalesSandy clayClay/sand
0-33-30
30-150150-300
50-40050-40050-40050-400
218175136
Offshore-Lousiana (Pleistocene)Southeast TexasSoutheast TexasSoutheast Texas
Clay/sandSands and shalesSame but more sandySandbanks, silty shaleMostly shaleSand (23%) and claySand (20%) and clayLimestone and chalkSand (45%) and claySand (24%) and clay
1170-17701770-20702070-2850900-15601560-18001800-2100600-15601590-1755660-1320
>1020
Š125Š125Š125Š80Š80Š80Š80Š80
15-4040-70
67>273
2852
>2733041
>2732855
Beaufort Sea(Canada)
549-1193945-1311
125425
OffshoreBaltimore
Siliceous chalk
Siliceous chalk withporcellanite joints
278-442
442-582
5000-15000
5000-15000
68 onave.287on
ave.S an d b an k s, silty sh ale
McDonald et al.(1958)
Tullos and Reid(1969)
Hauge (1981)
Ganley, Kansewich(1980)
Golberg (1958)
from Carmichael (1984) and Goldberg (1985)
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
129
Dominant fluid mechanisms affecting velocity and Q
Gassmann Model• static, zero frequency limit• no viscous/inertial effects• uniform pore pressure
Biot "global flow" Model• viscous/inertial effects• average flow only• average relative motion of fluid
and solid lead to dispersion and attenuation
Squirt "local flow" Model• viscous/inertial effects• grain-scale fluid motion leads to
dispersion and attenuation• effectively stiffens the soft porosity• superimposed on Biot/Gassmann
Patchy Saturation Model• large scale patches of saturation• patch-scale diffusion leads to
dispersion and attenuation• microscale squirt can be
superimposed
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
130
Biot TheoryBiot developed a macroscopic theory to attempt to model the behavior of fluid-saturated poroelasticsystems.
His generalized form of Hooke’s law:
where λλλλ and µµµµ are the dry rock moduli, and the fluid pressure P is linearly related to the normal stresses (and not the shears) by a new constant ββββ. Similarly, the increment of fluid content ξξξξ in an elementary cell of solid is linearly related to the pore pressure and the solid volumetric strain εεεεαααααααα. These describe essentially the same mechanical problem as the Gassmann theory for coupling the fluid and solid.
The equations of motion are:
where ρρρρuw describes an inertial coupling between the solid and the fluid, and χχχχ is a dissipation term.
σσσσ ij = λλλλδδδδ ijεεεεαααααααα + 2µµµµεεεεij – ββββPδδδδ ij
ξξξξ = 1MP + ββββεεεεαααααααα
∂∂∂∂σσσσ ij∂∂∂∂xj
= ρρρρ ∂∂∂∂2ui∂∂∂∂t2 + ρρρρuw
∂∂∂∂2wi∂∂∂∂t2
∂∂∂∂P∂∂∂∂xi
= ρρρρuw∂∂∂∂2ui∂∂∂∂t2 + ρρρρw
∂∂∂∂2wi∂∂∂∂t2 + 1
χχχχ∂∂∂∂wi∂∂∂∂t
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
131
Biot Theory
The high frequency limiting velocities and , are given by [in Geertsma and Smit’s (1961) notation]:
VS∞∞∞∞
VP∞∞∞∞
VS∞∞∞∞ = µµµµfr
ρρρρ0 1 – φφφφ + φφφφρρρρ fl 1 – a–1
12
where
Kfr, µµµµfr bulk and shear moduli of dry rock frameK0 bulk modulus of mineralKfl effective bulk modulus of pore fluidφφφφ porosityρρρρ0 mineral densityρρρρfl fluid densityρρρρ low frequency density of saturated composite:
a-1 tortuosity
The low frequency limiting velocities are the same as predicted by Gassmann’s relations.
ρρρρ = 1 – φφφφ ρρρρ0 + φφφφρρρρ fl
VP∞∞∞∞ = 1ρρρρ0 1 – φφφφ + φφφφρρρρ fl 1 – a–1 Kfr + 4
3µµµµfr +φφφφ ρρρρ
ρρρρfla–1 + 1 – Kfr
K01 – Kfr
K0– 2φφφφa–1
1 – KfrK – φφφφ 1
K + φφφφK
12
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
132
Biot Theory
Biot and later Stoll (1977) considered adding frame attenuation on top of fluid effects. This plot by Stollshows attenuation vs. frequency for two extreme cases and for a typical sand showing how frame losses and fluid losses combine to control the overall response.
10 -8
10 -6
10 -4
10 -2
10 0
10 0 10 1 10 2 10 3 10 4 10 5
Effects of Two DifferentKinds of Energy Loss
no frame lossno fluid losscomplete sediment model
Atte
nuat
ion
(nep
ers/
met
er)
Frequency (Hz) I.9
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
133
Squirt Flow
When a rock is compressed by the stress of a passing wave, increments of pore pressure are induced in the pore fluid. At very low frequencies there is time for the pore pressure to equilibrate throughout the pore space, and the fluid effect is described by the Gassmanntheory.
However, at high frequencies we expect that unequal pore pressures are induced on the microscale of individual pores--larger increments in the soft, crack-like porosity and smaller increments in the stiffer, equi-dimensional pores. If these do not equilibrate, the rock will be stiffer, and the velocities will be faster, than at low frequencies when they do equilibrate.
This frequency-dependent distribution of pore pressure leads to velocity dispersion, and the tendency for the fluid to flow and adjust leads to attenuation.
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
134
Estimating the High Frequency Squirt Flow Modulus
We compute the high frequency bulk modulus in 2 steps:1. the unrelaxed bulk modulus of the “wetted” frame where liquid is trapped in the thinnest cracks and the remaining spaceis dry, is given by
pore pressure in the ith thin crack:
Combining gives:
1
Khigh f– 1
Kmineral= 1
Kfluid– 1
Kmineralφφφφi
∂∂∂∂Pi∂∂∂∂σσσσΣΣΣΣ
low P
high P
∂∂∂∂Pi∂∂∂∂σσσσ ≈≈≈≈ 1
1 + 1Kfluid
– 1Kmineral
φφφφi /∂∂∂∂φφφφi /∂∂∂∂σσσσ dry
1Khigh f
≈≈≈≈ 1Kdry high P
+ 1Kfluid
– 1Kmineral
φφφφsoft σσσσ + ...
P4
P1
P2
P3
So trapping water in the thinnest cracks is approximately the same as closing the cracks under high pressure.
2. Finally the remaining pore space is saturated usingGassmann with Khigh f used as the “dry” rock modulus.
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
135
Squirt Flow
The shear modulus:
Comparing with the bulk modulus:
1
µµµµhigh f≈≈≈≈ 1
µµµµdry– 4
15∂∂∂∂φφφφi /∂∂∂∂σσσσ dry
1 + 1Kfluid
– 1Kmineral
φφφφi /∂∂∂∂φφφφi /∂∂∂∂σσσσ dryΣΣΣΣ
low P
high P
1µµµµhigh f
– 1µµµµdry
≈≈≈≈ 415
1Khigh f
– 1Kdry
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
136
Constructing the Unrelaxed Moduli
Confining Pressure
Bul
k M
odul
us
dry data
Biot / Gassmann
Shea
r Mod
ulus
dry data
Biot / Gassmann
saturated data
unrelaxed frame
unrelaxed frame
saturated data
I.10
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
137
4
4.2
4.4
4.6
4.8
5
5.2
0 20 40 60 80 100 120
Navajo
P-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
SatDry
high frequency(Mavko and Jizba)
high frequency(Biot)
low frequency(Gassmann)
2.6
2.8
3
3.2
3.4
0 20 40 60 80 100 120
Navajo
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
Sat
Dryhigh frequency
(Mavko and Jizba)
high frequency(Biot)
low frequency(Gassmann)
I.11
In these plots, the dry data are taken as inputs. The ultrasonic water-saturated data are compared with predictions byGassmann, the high frequency Biot limit, and the high frequency squirt limit.
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
138
3
3.1
3.2
3.3
3.4
3.5
0 20 40 60 80 100 120
Westerly
S-Ve
loci
ty (k
m/s
)
Effective Stress (MPa)
Sat
Dry
high frequency(Mavko and Jizba)
Biot and Gassmann
4.8
5
5.2
5.4
5.6
5.8
6
6.2
0 20 40 60 80 100 120
WesterlyP-
Velo
city
(km
/s)
Effective Stress (MPa)
Sat
Dry
high frequency(Mavko and Jizba)
Biot and Gassmann
I.12
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
139
-0.01
0
0.01
0.02
0.03
0.04
0.05
0.06
0 0.05 0.1 0.15 0.2 0.25 0.3
Biot
local flow
1µµµµsat
– 1µµµµdry
1Ksat
– 1Kdry σσσσ
– 1Ksat
– 1Kdry highσσσσ
I.13
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
140
Most physical mechanisms of dispersion and attenuationhave a characteristic frequency where attenuation is largest and velocity is changing most rapidly with frequency. It also separates the low frequency “relaxed”behavior from the high frequency “unrelaxed” behavior.
It is very difficult to predict the characteristic frequencyvery accurately, because it depends on idealized modelassumptions, and details of the rock microstructure thatare not well known.
Nevertheless, here are some rough estimates:
Biot:
patchy saturation:
viscous shear in crack:
squirt:
fbiot = φηφηφηφη2ππππρρρρfk
fvisc = kKfl2ηηηη
fsquirt = K0αααα3
ηηηη
fvisc. crack = αµαµαµαµ2πηπηπηπη
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
141
Biot Theory
Compiled in Bourbié, Coussy, and Zinszner,1987, Acoustics of Porous Media, Gulf Publishing Co.
Parameter Porosity(%)
Permeability(mD)
Characteristic frequency
Sample Water(h = 1cP)
(4)
Normal oil(h = 10-50
cP) (4)
Heavy oil(h = 100-
500 cP) (4)
Fontainebleausandstone (1)
5 0.1 80 MHz 800-4000MHz
8-40 GHa
Fontainebleausandstone (1)
20 1000 30 kHz 300-1500kHz
3-15 MHz
Tight sand (2) 8 0.02 1 GHz 10-50 GHz 100-500GHz
CordovaCreamlimestone (2)
24.5 9 4.5 MHz 45-230MHz
450-300MHz
Sintered glass 28.3 1000 42 kHz 420-2100kHz
4.2-21 MHz
(1) Bourbié and Zinszner (1985) (2) Carmichael (1982) (3) Plona and Johnson (1980)
(4) Viscosity η is expressed in centipoises (1 cP = 1 mPa. s).
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
142
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0 0.05 0.1 0.15 0.2 0.25 0.3
Bio
t Dis
pers
ion
of V
p
Porosity
Permeability vs. Seismic Dispersion?
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0 0.1 0.2 0.3 0.4 0.5 0.6
Building SandstoneP-SandstoneGulf SandstoneClean Sandstone
Bio
t Dis
pers
ion
of V
p
Clay Content I.14
Stanford Rock Physics Laboratory - Gary Mavko
Velocity Dispersion and Q
143
0
0.02
0.04
0.06
0.08
0.1
0 0.05 0.1 0.15 0.2 0.25 0.3
Non
-Bio
t Dis
pers
ion
of V
p
Porosity
Permeability vs. Seismic Dispersion?
0
0.02
0.04
0.06
0.08
0.1
0 0.1 0.2 0.3 0.4 0.5 0.6
Building SandstoneP-SandstoneGulf SandstoneClean Sandstone
Non
-Bio
t Dis
pers
ion
of V
p
Clay Content I.15
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
144
Knight and Nolen-Hoeksema (GRL, 1990) found saturationhysteresis at ultrasonic frequencies.
We know now that velocities depend, not just on saturation, but also on the scales at which the phases are mixed. The curve labeled “imbibition” is typical when phases are mixed at a fine scale. The curve labeled “drainage” is typical when the phases are mixed at a coarse scale -- which we call “patchy.”
2
2.5
3
3.5
4
4.5
5
0 0.2 0.4 0.6 0.8 1
Vp (k
m/s
)
Sw (fraction)
Drainage
Imbibition
K.1
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
145
2.15
2.2
2.25
2.3
2.35
2.4
2.45
0 0.2 0.4 0.6 0.8 1
Vp
(km
/s)
Oil Saturation
sandstoneporosity = 30%
patchy
homogeneous
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
146
Murphy (GRL, 1984) found that the hysteresis anddisagreement with Gassmann is primarily a problem of
high frequencies -- partially saturated dispersion.
Velocity vs. Saturationat Different Frequencies
2
2.5
3
3.5
4
4.5
5
5.5
6
0 0.2 0.4 0.6 0.8 1
Velo
city
(km
/s)
Water Saturation
200 kHz
1-2 kHzP
S
CQ
BGD
BGD
CQ
1-2 kHz
200 kHz
K.2
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
147
Murphy (GRL, 1984) found that ultrasonic velocitiesincrease with saturation much faster than predicted by
the low frequency Gassmann theory.
Ultrasonic Velocity vs. Pressureat Different Saturations
1500
2000
2500
3000
3500
4000
4500
5000
5500
0 10 20 30 40 50
Gro
up V
eloc
ity (m
/s)
Effective Pressure (MPa)
100%
70%0%P
S
SR 6521.6Frequency 800kHz
K.3
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
148
Endres and Knight (The Log Analyst, 1989) modeled different microdistributions of pore fluids and gas in the stiff and soft portions of the pore space. They concluded that the scale and distribution of fluids influence velocities.
a.
b.
c.
d.
Increasing water saturation
1600
1700
1800
1900
2000
2100
2200
2300
0 0.2 0.4 0.6 0.8 1
P-W
ave
Velo
city
(m/s
)
a
b
c
d
K.4
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
149
Patchy Saturation
High frequency responseIsolated patches—somestiff, some softer
Overall a highereffective velocity
Very low frequency response
Gassmann behaviorwith a single "effectivefluid"
Overall the softest,lowest velocity
Critical frequency:
fvisc = kKfl2ηηηη
1Keff.fl
= SiKi
ΣΣΣΣi
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
150
Patchy Diffusion ScalesCharacteristic diffusion time for a pressure disturbancewith length scale L to relax:
Inverting this, we can find the characteristic diffusion length over which pressure differences can relax at seismic frequency f
1f = ττττ ≈≈≈≈ L2
4D
L ≈≈≈≈ 4κκκκKηηηηf
Hz
f L10 Hz 1000 mD
100 mD10 mD1 mD
1000 mD100 mD10 mD1 mD
1000 mD100 mD10 mD1 mD
.1 mD
1 m.3 m.1 m
.03 m.3 m.1 m
.03 m
.01 m
.01 m.003 m.001 m
.0003 m
.0001 m
κκκκ
105
100 Hz
D is the hydraulic diffusivity, K is the fluid bulk modulus, κκκκ is the permeability, and ηηηη is the viscosity.
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
151
Estaillades LimestoneE
wav
e Ve
loci
ty (m
/s)
Water Saturation
DryingDrying (Reuss average)Drying (Voigt average)Gassmann formula
DepressurizationDepressurization (Reuss average)Depressurization (Voigt average)
K.5
Thierry Cadoret studied velocity vs. saturation using the resonant bar and found the coarse-scale and fine-scale behavior.
Stanford Rock Physics Laboratory - Gary Mavko
Partial Saturation
152
Estaillades Limestone
2500
2550
2600
2650
2700
0.5 0.6 0.7 0.8 0.9 1
DryingDepressurizationE
wav
e Ve
loci
ty (m
/s)
Water Saturation
0
5
10
15
20
25
30
0.5 0.6 0.7 0.8 0.9 1
DryingNCQ Model
1000
/Qe
Water Saturation K.6
Cadoret’s velocity and attenuation vs. saturation. The fine scale distribution gives relaxed viscoelasticbehavior, and the coarse scale gives unrelaxed. Therefore, we expect the largest attenuation when the velocity dispersion is largest. Hence, we get the important result that P-wave attenuation in a partially saturated rock can be much larger than in the dry or fully saturated case.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
153
2.15
2.2
2.25
2.3
2.35
2.4
2.45
0 0.2 0.4 0.6 0.8 1
Vp
(km
/s)
Oil Saturation
sandstoneporosity = 30%
patchy
homogeneous
The problem that we address is the nonuniqueresponse of seismic velocity to fluid saturation.What are the physical conditions that cause patchybehavior? When do we use the patchy model andwhen do we use the homogeneous model?
Our approach is to use flow simulation to study theparameters that control fluid distributions at a fine scale.
study by Madhumita Sengupta and G. Mavko
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
154
Porosity and permeability models for flow simulation
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
155
We will consider two important cases: water flood into oil, and gas flood into oil.
The parameters that we consider are:• relative permeability• wettability• density contrast• permeability heterogeneity• capillary pressure
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
156
Relative Permeability Curves for Oil and Water
Water Injection in Oil
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
157
SaturationsSaturations obtained from flow simulations using the obtained from flow simulations using the dashed (top) and solid (bottom) relative permeability dashed (top) and solid (bottom) relative permeability curves. The irreduciblecurves. The irreducible saturationssaturations are critical controlsare critical controlson the saturation extremes.on the saturation extremes.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
158
The patchy and uniform saturation curves are upperand lower bounds. They describe the range of velocitysignatures that we can achieve by mixing the endmembers. Finite irreducible saturations drastically narrow the range of uncertainty.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
159
Water saturation map and histogramof saturation in an oil wet rock
Wettability
The saturation distributiondepends on the wettabilityof the rock.
Most sandstone reservoirsare water wet and most carbonate reservoirsare oil wet.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
160
Wettability
Wettability tends to determine whether the velocitiesfall high or low in the allowable range.
** ****
** ******
Oil WetOil Wet(Drainage)(Drainage)
Water WetWater Wet(Imbibition)(Imbibition)
Sw
Vp
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
161
Low Mobility Ratio
20 40 60 80 100
20
400.2
0.4
0.6
0.8
0 0.5 10
500
1000
1500
2000
2500
High Mobility Ratio
20 40 60 80 100
20
400.2
0.4
0.6
0.8
0 0.5 10
500
1000
1500
Mobility Ratios
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
162
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 13050
3100
3150
3200
3250
3300
Water Saturation
VP
High MR
Low MR
SSww
VVpp
Low MRLow MR
High MRHigh MR
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
163
Summary
• The uniform saturation model is good enough for waterflood oil-water cases.
• Exceptions: when the irreducible oil is very low in an oil wet rock.
• The main control is the finite irreduciblesaturations.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
164
Gas Injection Into Oil
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
165
Gas InjectionEffect of Mobility
(a) Low MobilityRatio
20 40 60 80 100
20
40
0
0.2
0.4
0.6
0 0.5 10
1000
2000
3000
(b) High MobilityRatio
20 40 60 80 100
20
40
0
0.2
0.4
0.6
0 0.2 0.4 0.6 0.80
500
1000
1500
2000
2500
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
166
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 12800
2850
2900
2950
3000
3050
3100
3150
Vp
So
High Mobility Ratio
Low Mobility Ratio
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
167
Heterogeneity of Perm
99
99.5
100
100.5
101
0 0.5 1
20406080100120
0
0.2
0.4
0.6
0 0.5 1
200
400
600
800
1000
0
0.2
0.4
0.6
0 0.5 1
0
0.2
0.4
0.6
Perm Models Saturation
SatnHistograms
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
168
0 0.2 0.4 0.6 0.8 12800
2850
2900
2950
3000
3050
3100
3150
Large ScaleHeterogeneities
Small ScaleHeterogeneities
Vp
So
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
169
Summary
• When there is gas in the reservoir it is quite easy to get patchy saturation.
• This patchy saturation is controlled by the irreducible saturations.
Stanford Rock Physics Laboratory - Gary Mavko
Flow Simulation and Saturation Scales
170
Conclusions
• Reservoirs with gas are very likely to show patchy behavior.
• The uniform saturation model may be good enough for reservoirs with only oil and water
• The main mechanism that causes patchy behavior at the field scale is gravity.
Stanford Rock Physics Laboratory - Gary Mavko
North Sea Turbidite
171
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
172
Velocity-porosity relationship in clastic sediments and rocks. Datafrom Hamilton (1956), Yin et al. (1988), Han et al. (1986). Compiled
by Marion, D., 1990, Ph.D. dissertation, Stanford Univ.
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100suspensionssand-clay mixt.sandclay-free sandstoneclay-bearing sandstone
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt Avg.
Wood's Relation
L.1
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
173
We observe that the clastic sand-clay system is divided into two distinct domains, separated by a critical porosityφφφφc. Above φφφφc, the sediments are suspensions. Below φφφφc , the sediments are load-bearing.
Critical Porosity
load-bearing suspension
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100suspensionssand-clay mixt.sandclay-free sandstoneclay-bearing sandstone
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt Avg.
Wood's Relation
L.1
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
174
Critical Porosity
Traditionally, bounding methods have been considered not very useful for quantitative predictions of velocity-porosity relationships, because the upper and lower bounds are so far apart when the end memebers are pure quartz and pure water.
However, the separation into two domains above and below the critical porosity helps us to recognize that the bounds are in fact useful for predictive purposes.
• φφφφ > φφφφc, fluid-bearing suspensions. In the suspension domain the velocities are described quite well by theReuss average (iso-stress condition).
• φφφφ < φφφφc, load-bearing frame. Here the situation appears to be more complicated. But again, there is a relatively simple pattern, and we will see that the Voigt average is useful.
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
175
The first thing to note is that the clean (clay free) materials fall along a remarkably narrow trend. These range from very low porosity, highly consolidatedsandstones, to high porosity loose sand.
(Data from Yin et al., 1988; Han et al., 1986. Compiled and plotted by Marion, D., 1990, Ph.D. dissertation, Stanford University.
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100suspensionssandclay-free sandstone
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt Avg.
Wood's Relation
L.2
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
176
Amos Nur discovered that this narrow trend can be described accurately with a modified Voigt bound. Recall that bounds give a way to use the properties of the “pure” end members to predict the properties in between. The trick here is to recognize that the critical porosity marks the limits of the domain of consolidated sediments, and redefine the right end member to be the suspension of solids and fluids at the critical porosity.
0
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100
P-Ve
loci
ty (m
/s)
Porosity (%)
Voigt
Reuss
Critical Porosity
Solid line: SaturatedDash line: Dry
L.3
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
177
The Modified Voigt Bound
Velocity in rocks
The usual Voigt estimate of modulus
Modified Voigt estimate of modulus
VP = Mρρρρ
φφφφ = φφφφφφφφc
0 ≤≤≤≤ φφφφ ≤≤≤≤ φφφφc 0 ≤≤≤≤ φφφφ ≤≤≤≤ 1.0
ρρρρ = 1 – φφφφ ρρρρmineral + φφφφρρρρfluid
M = 1 – φφφφ Mmineral + φφφφMfluid
M = 1 – φφφφ Mmineral + φφφφMcritical "mush"
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
178
0
0.2
0.4
0.6
0.8
1
0 0.1 0.2 0.3 0.4 0.5
SandstonesWater saturated
ρρ ρρVp2 (n
orm
aliz
ed)
porosity
φcReuss
average
approximatedata trend
suspensionload-bearing
L.4
Example of critical porosity behavior in sandstones.
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
179
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water saturated carbonates
interparticle porositymicro porositymoldic porosityintraframe porositydensely cemented
Velo
city
(m/s
)
porosity (percent)
average trend
Reuss average suspension line
L.29
Data from Anselmetti and Eberli, 1997, in Carbonate Seismology, SEG.
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
180
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
Velo
city
(m/s
)
porosity (percent)
Intraframe porosity
average trend
above averagepore stiffness
(f) L.30
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
Velo
city
(m/s
)
porosity (percent)
Moldic porosity
average trend
above averagepore stiffness
(e) L.31
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
Velo
city
(m/s
)
porosity (percent)
micro porosity
average trend
below averagepore stiffness
(d) L.32
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
Velo
city
(m/s
)
porosity (percent)
Interparticle, intercrystalline porosity
average trend
averagepore stiffness
(c) L.33
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water-saturated Carbonates
Velo
city
(m/s
)porosity (percent)
densely cemented,low porosity
average trend
above averagepore stiffness
(b) L.34
1000
2000
3000
4000
5000
6000
7000
0 20 40 60 80 100
Water saturated carbonates
interparticle porositymicro porositymoldic porosityintraframe porositydensely cemented
Velo
city
(m/s
)
porosity (percent)
average trend
Reuss average suspension line
(a)
L.29
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
181
L.5
0.0
0.2
0.4
0.6
0.8
1.0
0 0.2 0.4 0.6 0.8
Chalks
Nor
mal
ized
M M
odul
us
Porosity
Calcite
Brevik
Urmos and Wilkens
Reuss Bound
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
182
L.6
0.0
0.2
0.4
0.6
0.8
1.0
0 0.2 0.4 0.6 0.8 1
Glass Foam
Nor
mal
ized
M M
odul
us
Porosity
Glass
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
183
Effects of Clay
Han (1986, Ph.D. dissertation, Stanford University)studied the effects of porosity and clay on 80 sandstonesamples represented here.
0
0.1
0.2
0.3
0.4
0.5
0.6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Building SandstoneP-SandstoneGulf SandstoneClean SandstoneTight Gas Sandstone
Cla
y C
onte
nt
PorosityL.7
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
184
Han (1986) found the usual result: velocities tend to decrease with porosity, but with a lot of scatter about the regressions when clay
is present (water saturated).
L.8
3
3.5
4
4.5
5
5.5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Building SandstoneP-SandstoneGulf SandstoneClean SandstoneTight Gas Sandstone
Vp (k
m/s
)
Porosity
1.5
2
2.5
3
3.5
4
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Building SandstoneP-SandstoneGulf SandstoneClean SandstoneTight Gas Sandstone
Vs (k
m/s
)
Porosity
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
185
Han’s Relations (40 MPa)Clean sandstones (10 samples)
Clay-bearing sandstones (70 samples)
Ignoring the clay
Including a clay term
R = correlation coefficient; % = RMS
VP = 6.08 – 8.06φφφφVS = 4.06 – 6.28φφφφ
VP = 5.02 – 5.63φφφφVS = 3.03 – 3.78φφφφ
VP = 5.59 – 6.93φφφφ – 2.18CVS = 3.52 – 4.91φφφφ – 1.89C
VP = 5.41 – 6.35φφφφ – 2.87CVS = 3.57 – 4.57φφφφ – 1.83C
R = 0.99 2.1%R = 0.99 1.6%
R = 0.80 7.0%R = 0.70 10%
R = 0.98 2.1%R = 0.95 4.3%
R = 0.90R = 0.90
dry
wat
er s
atur
ated
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
186
22.5
33.5
44.5
55.5
6
0 0.1 0.2 0.3 0.4
Shaly sandstones
0-10%10-20%20-30%30-40%
Vp (k
m/s
)
porosity
c = .05.15.25.35
clay volume
Han’s water-saturated ultrasonic velocity data at 40 MPa compared with his empirical
relations evaluated at four different clay fractions.
Han’s empirical relations between ultrasonic Vp and Vs in km/s with porosity and clay volume fractions.
Clean Sandstones (determined from 10 samples) Water saturated40 MPa Vp = 6.08 - 8.06φφφφ Vs = 4.06 - 6.28φφφφ
Shaly Sandstones (determined from 70 samples)
Water saturated40 MPa Vp = 5.59 - 6.93φφφφ - 2.18C Vs = 3.52 - 4.91φφφφ - 1.89C30 MPa Vp = 5.55 - 6.96φφφφ - 2.18C Vs = 3.47 - 4.84φφφφ - 1.87C 20 MPa Vp = 5.49 - 6.94φφφφ - 2.17C Vs = 3.39 - 4.73φφφφ - 1.81C10 MPa Vp = 5.39 - 7.08φφφφ - 2.13C Vs = 3.29 - 4.73φφφφ - 1.74C5 MPa Vp = 5.26 - 7.08φφφφ - 2.02C Vs = 3.16 - 4.77φφφφ - 1.64C
Dry40 MPa Vp = 5.41 - 6.35φφφφ - 2.87C Vs = 3.57 - 4.57φφφφ - 1.83C
L.9
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
187
The critical porosity, modified Voigt bound incorporating Han's clay correction.
0
1
2
3
4
5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Shaly Sandstone Data Before Clay-Corrections
modeldata (clean-sandstones)data (shaly-sandstones)
Velo
city
(km
/s)
Porosity
P-Velocity
S-Velocity
0
1
2
3
4
5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Clay-Corrections With Empirical Relation
data (clean-sandstones)data (shaly-sandstones)critical por. models for clean sandstonecritical por. models for shaly sandstone
Velo
city
(km
/s)
Porosity
P-Velocity
S-Velocity
L.12
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
188
Porosity vs. clay weight fraction at various confining pressures. FromDominique Marion, 1990, Ph.D. dissertation, Stanford University. Data
are from Yin, et al., 1988.
Sand, shaley sand Shale, sandy shale
0
10
20
30
40
50
60
70
0 20 40 60 80 100
Poro
sity
(%)
Clay Content by Weight (%)
0 MPa
40 MPa
20 MPa
30 MPa
10 MPa
50 MPa
0
10
20
30
40
50
60
70
0 20 40 60 80 100
Poro
sity
(%)
Clay Content by Weight (%)
0 MPa
40 MPa
20 MPa
30 MPa
10 MPa
50 MPa
L.13
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
189
Velocity vs. clay weight fraction at various confining pressures. FromDominique Marion, 1990, Ph.D. dissertation, Stanford University. Data
are from Yin, et al., 1988.
Sand, shaley sand Shale, sandy shale
1500
2000
2500
3000
3500
0 20 40 60 80 100
P-Ve
loci
ty (m
/s)
Clay Content by Weight (%)
40 MPa
20 MPa30 MPa
10 MPa
50 MPa
1500
2000
2500
3000
3500
0 20 40 60 80 100
P-Ve
loci
ty (m
/s)
Clay Content by Weight (%)
40 MPa
20 MPa30 MPa
10 MPa
50 MPa
L.14
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
190
Influence of clay content on velocity-porosity relationship at a constant confining pressure (50 MPa). Distinct trends for shalysand and for shale are schematically superposed on experimental data on sand-clay mixture. From Dominique Marion, 1990, Ph.D. dissertation, Stanford University. Data are from Yin, et al., 1988, and Han, 1986.
2000
2500
3000
3500
0.1 0.15 0.2 0.25 0.3 0.35
P-Ve
loci
ty (m
/s)
Porosity
shaley sands
sandshale
sandy shales
50 MPa
c = øs
L.15
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
191
Gulf of Mexico Well, Herron et al, 1992, SPE 24726
40
20
0
60
3
2
1
Clay (vol%)
Poro
sity
(vol
%)
Vp (k
m/s
)3780-4800 ft 3780-4800 ft
Poro
sity
(vol
%)
Clay (vol%)
Vp (k
m/s
)
3
2
1
60
40
40 40
20
20 200
00
4800-5895 ft4800-5895 ft
L.17
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
192
Amoco's Well in the Hastings Field (On-Shore Gulf Coast)
Density vs. Neutron Porosity Poorly Consilidated Shaly Sands
Laminar Clay Model
2.30
Marion Model
Increasing Clay Content
nphi
rhob
(g/c
m )
2.00
2.10
2.20
2.40
2.50
2.60
2.700.00 0.10 0.20 0.30 0.40 0.50
3
L.18
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
193
Schlumberger, 1989
Density Porosity vs. Neutron Porosity in Shaly Sands
Sho
0.5
0.4
0.3
0.1
Q
QuartzPoint
0.1
0.2
0.3 0.4 0.5
T o D r y C l a y P o i n t
T o W a t e r P o
GasSand
Sd
C
ClSh
0.2
φφφφN
φφφφD
T o W a te r P o in t
C le an W a te r S a nd s
A
B
L.19
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
194
Yin’s laboratory measurements on sand-clay mixtures.
10 -2
10 -1
10 0
10 1
10 2
10 3
10 4
0 20 40 60 80 100
Permeability (Gas) vs. Clay Content
Perm
eabi
lity
(mD
)
Clay Content (% by weight)
0 MPa
10 MPa
20 MPa30 MPa
40 MPa50 MPa
shaly sand
sandy shale
L.20
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
195
Yin’s laboratory measurements on sand-clay mixtures.
10 -2
10 -1
10 0
10 1
10 2
10 3
10 4
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Permeability (Gas) vs. Porosity
Perm
eabi
lity
(mD
)
Porosity
0 MPa
30 MPa
10 MPa
50 MPa 40 MPa
20 MPa
0%
5%
10%
15%20%
25%
30%
40%
50%
65%
85%
100%
% clay content by weight
L.21
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
196
Permeability vs. porosity data in Gulf-Coast sandstones reflect the primary influence of clay content on both permeability and porosity.Kozeny-Carman relations for pure sand and pure shale are also shown (dashed lines) to illustrate the effect of porosity on permeability. FromDominique Marion, 1990, Ph.D. dissertation, Stanford University.
0.0001
0.01
1
100
10 4
10 6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Perm
eabi
lity
(md)
Porosity
K-C sand
K-C shale
L.22
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
197
Yin's laboratory measurements on sand-clay mixtures.
1800
2000
2200
2400
2600
2800
1 10 100 1000 10 4
P-Velocity vs. Permeability (Gas)
P-Ve
loci
ty (m
/s)
Permeability (md)
0% (pure sand)
5%
10%15%
20%
25%
30%40%
50%
65%
85%
100% (pure clay)% clay content by weight
L.23
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
198
L.36
0
1000
2000
3000
4000
5000
6000
0 0.1 0.2 0.3 0.4 0.5
Varied Velocity-Porosity Trends
Porosity
Gulf of Mexico (Han)
Vp Troll
Oseberg
Cementing Trend
Han’s large data set spans a large range of depths and clearly shows the steep cementing trend, which would be favorable for mapping velocity (or impedance) to porosity. Other data sets from the Troll and Oseberg indicate much shallower trends.
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
199
0
1000
2000
3000
4000
5000
6000
0 0.1 0.2 0.3 0.4 0.5
Cementing vs. Sorting Trends
Porosity
Troll
Gulf of Mexico (Han)
Oseberg
Vp
Reuss Bound(Deposition)
Cementing Trend
SortingTrend
The slope of the velocity-porosity trend is controlled by the geologic process that controls variations in porosity. If porosity is controlled by diagenesis and cementing, we expect a steep slope – described well by a modified upper bound. If it is controlled by sorting and clay content (depositional) then we expect a shallower trend – described well by a modified lower bound.
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
200
Generalized Sandstone Model
L.36
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4 0.5
Cementing vs. Sorting Trends
Vp
Porosity
clean cementing trend
Suspension Line(Reuss Bound)
sorting trend
New Deposition
Mineral point
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
201
0
1
2
3
4
5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
North Sea Clean sands
shallow oil sand deeper water sand
Vp
Total Porosity
increasing cement
Suspension Line
poor sorting
• all zones converted to brine• only clean sand, Vsh <.05
L.37
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
202
L.37
0
1
2
3
4
5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
North SeaClean vs. Shaly Sands
2508-2545 m, vsh<.052508-2545 m, Vsh>.32701-2750 m, vsh<.052701-2750 m, Vsh>.3
Vp
Total Porosity
increasing cement
Suspension Line
poor sorting
all zonesconverted to brine
more clay
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
203
2
2.5
3
3.5
4
4.5
5
5.5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
P-w
ave
Velo
city
(km
/s)
Porosity
?Jizba 9,000 ft (75 MPa)
Han 15,000 ft (40 MPa)R = 0.70
Han 12,000 ft (40 MPa)R = 0.80
Han 10,000 ft (40 MPa)R = 0.96
Blangy 5,000 ft, Troll (30 MPa)R = 0.76
L.27
Stanford Rock Physics Laboratory - Gary Mavko
Shaly Sands
204
Xu and White (1995) developed a theoretical model for velocities in shaly sandstones. the formulation uses the Kuster-Toksöz and Differential Effective Medium theories to estimate the dry rock P and S velocities, and the low frequency saturated velocities are obtained from Gassmann’s equation. The sand-clay mixture is modeled with ellipsoidal inclusions of two different aspect ratios. The sand fraction has stiffer pores with aspect ratio αααα
≈≈≈≈
0.1 - 0.15, while the clay-related pores are more compliant with αααα
≈≈≈≈
0.02-0.05. The velocity model simulates the “V” shaped velocity-porosity relation of Marion et al. (1992) for sand-clay mixtures. The total porosity φφφφ = φφφφsand + φφφφclay where φφφφsand and φφφφclay are the porosities associated with the sand and clay fractions respectively. These are approximated by
where Vsand and Vclay denote the volumetric sand and clay content respectively. Shale volume fromlogs may be used as an estimate of Vclay. through the log derived shale volume includes silts, and overestimates clay content, results obtained by Xuand White justify its use. The properties of the solid mineral mixture are estimated by a Wyllie time average of the quartz and clay mineral velocities, and arithmetic average of t heir densities:
where subscript 0 denotes the mineral properties. these mineral properties are then used in the Kuster-Toksöz formulation along witht he porosity and clay content, to calculate dry rock moduli and velocities. The limitation of small pore concentration of the Kuster-Toksöz model is handled by incrementally adding the pores in small steps such that the non-interaction criterion is satisfied in each step. Gassmann’s equations are used to obtain low frequency saturated velocities. High frequency saturated velocities are calculated by using fluid-filled ellipsoidal inclusions in the Kuster-Toksöz model.
The model can be used to predict shear wave velocities (Xu and White, 1994). Estimates of Vs may be obtained from known mineral matrix properties and measured porosity and clay content, or from measured Vp and either porosity or clay content. Su and White recommend using measurements of P-wave sonic log since it is more reliable than estimates of shale volume and porosity.
φsand = 1 – φ – Vclay
φ1 – φ = Vsand
φ1 – φ
φclay = Vclay
φ1 – φ
1VP0
=1 – φ – Vclay
1 – φ1
VPquartz
+Vclay1 – φ
1VPclay
1VS 0
=1 – φ – Vclay
1 – φ1
VSquartz
+Vclay1 – φ
1VSclay
ρ0 =
1 – φ – Vclay1 – φ ρquartz +
Vclay1 – φρclay
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
205
Grain
Contact cement
A B
R a a
Non-contact cement Scheme 1 Scheme 2
C
Dvorkin’s Cement ModelJack Dvorkin introduced a cement model that predicts
the bulk and shear moduli of dry sand when cement is deposited at grain contacts. The model assumes that the cement is elastic and its properties may differ from those of the grains.
It assumes that the starting framework of cemented sand is a dense random pack of identical spherical grains with porosity , and the average number of contacts per grain C = 9. Adding cement reduces porosity and increases the effective elastic moduli of the aggregate. The effective dry-rock bulk and shear moduli are (Dvorkin and Nur, 1996)
where
φ0 ≈ 0.36
G c = ρ cVSc2
Keff = 16 C 1 – φ0 M c Sn
M c = ρcVPc2
Geff = 35 K eff + 3
20 C 1 – φ0 G c S τ
is the cement's density; and and are its P-and S-wave velocities. Parameters and are proportional to the normal and shear stiffness, respectively, of a cemented two-grain combination. They depend on the amount of the contact cement and on the properties of the cement and the grains. (see next page)
ρc VPc VSc
Sn S τ
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
206
where G and are the shear modulus and the Poisson's ratio of the grains, respectively; and are the shear modulus and the Poisson's ratio of the cement; a is the radius of the contact cement layer; R is the grain radius.
Dvorkin’s Cement Model
Constants in the cement model: S n = A n α2 + B n α + C n
A n = – 0.024153 Λ n–1.3646
Bn = 0.20405 Λ n–0.89008
Cn = 0.00024649 Λ n–1.9864
S τ = A τ α2 + B τ α + C τ
A τ = –10–2 2.26ν 2 + 2.07ν + 2.3 Λ τ0.079ν2 + 0.1754ν – 1.342
B τ = 0.0573ν 2 + 0.0937ν + 0.202 Λ τ0.0274ν2 + 0.0529ν – 0.8765
C τ = –10–4 9.654ν 2 + 4.945ν + 3.1 Λ τ0.01867ν2 + 0.4011ν – 1.8186
Λ n = 2G c
πG1 – ν 1 – ν c
1 – 2ν cΛ τ = G c
πG α = a
R
ννcGc
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
207
The amount of the contact cement can be expressed through the ratio of the radius of the cement layer a to the grain radius R:
α
α = a/RThe radius of the contact cement layer a is not necessarily directly related to the total amount of cement: part of the cement may be deposited away from the intergranular contacts. However by assuming that porosity reduction in sands is due to cementation only, and by adopting certain schemes of cement deposition we can relate parameter to the current porosity of cemented sand . For example, we can use Scheme 1 (see figure above) where all cement is deposited at grain contacts:
αφ
α = 2 φ0 – φ
3C 1 – φ0
0.25= 2 Sφ0
3C 1 – φ0
0.25
or we can use Scheme 2 where cement is evenly deposited on the grain surface:
α =
2 φ0 – φ3 1 – φ 0
0.5
= 2Sφ03 1 – φ0
0.5
In these formulas S is the cement saturation of the pore space - the fraction of the pore space occupied by cement.
Dvorkin’scement model
Grain
Contact cement
A B
R a a
Non-contact cement Scheme 1 Scheme 2
C
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
208
If the cement's properties are identical to those of the grains, the cementation theory gives results which are very close to those of the Digby model. The cementation theory allows one to diagnose a rock by determining what type of cement prevails. For example, it helps distinguish between quartz and clay cement. Generally, Vp predictions are much better than Vs predictions.
2
2.5
0.2 0.3
Vs (k
m/s
)
porosity
quartz-cemented
clay-cementedTHEORETICAL
CURVES
3
4
0.2 0.3
Vp (k
m/s
)
porosity
quartz-cemented
clay-cemented
THEORETICALCURVES
Predictions of Vp and Vs using the Scheme 2 model for quartz and clay cement, compared with data from quartz and clay cemented rocks from the North Sea.
Dvorkin’s Cement Model
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
209
LOAD-BEARING(CONTACT)
CEMENT
POROSITY
MO
DUL
US
NO CEMENT
NON-LOAD-BEARING
(NON-CONTACT) CEMENT
2
3
4
0.1 0.2 0.3 0.4
Vp
(km
/s)
Porosity
UncementedTrajectory
Cement-ClayTrajectory
Cement-QuartzTrajectory
a
2
3
4
0.1 0.2 0.3 0.4
TrollOseberg QuartzOseberg Clay
Vp
(km
/s)
Porosity
b
Sand models can be used to “Diagnose” sands
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
210
Dvorkin’s Uncemented Sand Model
This model predicts the bulk and shear moduli of dry sand when cement is deposited away from grain contacts. The model assumes that the starting framework of uncemented sand is a dense random pack of identical spherical grains with porosity , and the average number of contacts per grain C = 9. The contact Hertz-Mindlin theory gives the following expressions for the effective bulk ( ) and shear( ) moduli of a dry dense random pack of identical spherical grains subject to a hydrostatic pressure P:
φ0 = 0.36
KHM GHM
KHM = C2 1 – φ0
2 G2
18 π2 1 – ν 2 P1/3
GHM = 5 – 4ν
5 2 – ν3C2 1 – φ 0
2 G 2
2π2 1 – ν 2 P1/3
where is the grain Poisson's ratio and G is the grain shear modulus.
ν
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
211
Dvorkin’s Uncemented Sand ModelIn order to find the effective moduli at a different porosity, a heuristic modified Hashin-Strikman lower bound is used:
Keff = φ / φ0
K HM + 43 G HM
+ 1 – φ / φ0
K + 43 G HM
–1– 4
3 GHM
G eff = [ φ / φ0
G HM +G HM
69KHM + 8G HMK HM + 2GHM
+ 1 – φ / φ0
G +GHM
69K HM + 8GHMKHM + 2GHM
]–1
–GHM
69KHM + 8G HMKHM + 2GHM
0
20
40
60
80
100
0 0.1 0.2 0.3 0.4
M-m
odul
us (G
Pa)
porosity
solid
HERTZ-MINDLIN
increasingpressure
M = ρρρρVP
2
Illustration of the modified lower Hashin-Shtrikman bound for various effective pressures. The pressure dependence follows from the Hertz-
Mindlin theory incorporated into the right end member.
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
212
Dvorkin’s Uncemented Sand ModelThis model connects two end members: one has zero porosity and the modulus of the solid phase and the other has high porosity and a pressure-dependent modulus as given by the Hertz-Mindlin theory. This contact theory allows one to describe the noticeable pressure dependence normally observed in sands.The high-porosity end member does not necessarily have to be calculated from the Hertz-Mindlin theory. It can be measured experimentally on high-porosity sands from a given reservoir. Then, to estimate themoduli of sands of different porosities, the modifiedHashin-Strikman lower bound formulas can be used where KHM and GHM are set at the measured values. This method provides accurate estimates for velocities in uncemented sands. In the figures below the curves are from the theory.
0
1
2
3
0.2 0.3 0.4
velo
city
(km
/s)
porosity
Vp
Vs
saturatedP
eff = 5 MPa
0
1
2
3
0.2 0.3 0.4porosity
Vp
Vs
saturatedP
eff = 15 MPa
0
1
2
3
0.2 0.3 0.4porosity
Vp
Vs
saturatedP
eff = 30 MPa
Prediction of Vp and Vs using the lower Hashin-Shtrikmanbound, compared with measured velocities from
unconsolidated North Sea samples.
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
213
0.8
1.2
1.6
2
2.4
2.8
0.3 0.4 0.5
velo
city
(km
/s)
porosity
Vp
Vs
North Seasand
Ottawasand
theory
This method can also be used for estimating velocities in sands of porosities exceeding 0.36.
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
214
2 3
2.1
2.2
2.3
Vp (km/s)
Well #1
A40 80 120
GRB2 3 4
1.7
1.8
1.9
Vp (km/s)
Well #2
C
Marl
Limestone
40 80 120GRD
2.5
3
3.5
0.25 0.3 0.35 0.4
p(
)
Porosity
Contact CementLine
UnconsolidatedLine
ConstantCement Fraction (2%) Line
Well #1
Well #2
North Sea Example
Study by Per Avseth, along with J. Dvorkin, G. Mavko, and J. Rykkje
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
215
0 50 100
1730
1740
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
Dep
th(m
)
gamma ray (API)0 2 4
1730
1740
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
Vp, Vs and Vp/Vs
VpVs
Vp/Vs
1.8 2 2.2 2.4 2.6
1730
1740
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
density
Well-Log Measurements of Cemented High-Porosity Sandstone Interval
2.5 3 3.5
1780
1785
1790
1795
1800
1805
1810
1815
1820
1825
1830
Vp
Dep
th(m
)
0.3 0.35 0.4
1780
1785
1790
1795
1800
1805
1810
1815
1820
1825
1830
Porosity
Dep
th(m
)
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
216
Sorting Analysis of Thin-Sections
100 300 500 700 9000
50
100
150
Grain diameter − @1785.1m
Fre
quen
cy
100 300 500 700 9000
50
100
150
Grain diameter − @1790.1mF
requ
ency
100 300 500 700 9000
50
100
150
Grain diameter − @1815.1m
Fre
quen
cy
100 300 500 700 9000
50
100
150
Grain diameter − @1820.1m
Fre
quen
cy
0.4m m0.4m m
0.4m m0.4m m
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
217
Sorting and Rock Physics Properties
0.25 0.3 0.35 0.4
2.6
2.8
3
3.2
3.4
3.6
density porosity
Vp
0.6 0.8 1 1.22.6
2.8
3
3.2
3.4
3.6
sorting (std/mean)
Vp
0.6 0.8 1 1.2 1.40.28
0.3
0.32
0.34
0.36
0.38
sorting (std/mean)
dens
ity p
oros
ity
0.6 0.8 1 1.2 1.4
0.3
0.32
0.34
0.36
0.38
0.4
sorting (std/mean)
heliu
m p
oros
ity
0.6 0.8 1 1.21.3
1.4
1.5
1.6
1.7
1.8
sorting (std/mean)
Vs
(km
/s)
0.6 0.8 1 1.21.8
1.9
2
2.1
2.2
sorting (std/mean)
Vp/
Vs
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
218
Sorting Versus Cementation
0.25 0.3 0.35 0.42
2.5
3
3.5
4
porosity
Vp
(km
/s)
contact cement
line4%
2%
0.25 0.3 0.35 0.42
2.5
3
3.5
4
porosity
Vp
(km
/s)
2%
0.25 0.3 0.35 0.42
2.5
3
3.5
4
porosity
Vp
(km
/s)
4%
0.25 0.3 0.35 0.42
2.5
3
3.5
4
porosity
Vp
(km
/s)
2%
4%
Stanford Rock Physics Laboratory - Gary Mavko
Sand Models
219
Thin-Section and SEM Analyses
Well #2 Cemented
0.25 mm
Well #1 Uncemented
0.25 mm
SEM cathode-luminescent image: Well #2
0.1 mm0.1 mm
SEM back-scatter image: Well #2
Unconsolidated(Facies IIb)
Cemented(Facies IIa)
Back-scatter light Cathode lum. light
Qz-cement rim Qz-grain
Stanford Rock Physics Laboratory - Gary Mavko
220
A North Sea Example
PROBLEM
Can we predict and characterize reservoirs in North Sea deep-water clastic systems
using seismic data?
1 km1 km
SEISMIC AMPLITUDE MAP OF A NORTH SEA SUB-MARINE FAN
Stanford Rock Physics Laboratory - Gary Mavko
221
A North Sea Example
SEISMIC LITHOFACIES IN DEEP- WATER CLASTIC SYSTEMS
A seismic lithofacies is a seismically resolvable sedimentary unit characterized by its:
• lithology (clay content)
• bedding configuration (massive, interbedded, chaotic)
• petrography (grain size, clay location and cementation)
• seismic properties (seismic velocities and density)
Stanford Rock Physics Laboratory - Gary Mavko
222
A North Sea Example
I
II
III
IV
V
VI
FACIES I: Gravels and conglomerates
FACIES II:Thick-bedded sandstone
FACIES III:Interbedded sandstone-shale
FACIES IV:Silty shale and silt-laminated shale
FACIES V:Pure shale
FACIES VI: Chaotic deposits
Stanford Rock Physics Laboratory - Gary Mavko
223
A North Sea Example
1.5 2 2.5 3 3.5 4
2100
2150
2200
Vp (km/s)
Dep
th (m
)V
IV II b
II c
III
Identification of Seismic Lithofacies From Well Logs (Field A, Well A-1)
V
IV
40 60 80 100 120 140
2100
2150
2200
Gamma Ray (API)
Dep
th (m
)
V
IV II b
II c
III
Stanford Rock Physics Laboratory - Gary Mavko
224
A North Sea Example
IIa
III
IIcIIb
IVV
Overburden
Tuff
Chalk1.0 km
200 m
Depth = 2.0 km
2-D Geological Model
Stanford Rock Physics Laboratory - Gary Mavko
225
A North Sea Example
40 60 80 100 120 140
2.2
2.6
3
3.4
Gamma Ray (API)
P-w
ave
velo
city
(km
/s)
IIb
IIcIII
IVV
Rock Physics Analysis
40 60 80 100 120 140
0.2
0.3
0.4
Gamma Ray (API)
Poro
sity
IIbIIc
III
IV V
Stanford Rock Physics Laboratory - Gary Mavko
226
A North Sea Example
40 60 80 100 120 140
4.5
5.5
6.5
7.5
Gamma Ray (API)
Aco
ustic
Impe
danc
e
IIb
IIc III
IV
V
40 60 80 100 120 1401.5
2
2.5
3
Gamma Ray (API)
Vp/V
s
IIb IIc
IIIIV
V
Seismic Properties
Stanford Rock Physics Laboratory - Gary Mavko
227
A North Sea Example
Seismic Reflectivity Modeling(AVO)
R(θθθθ) = R0 + G * sin (θθθθ)
Ref
lect
ivity
0
- 0.1
0.1
II bV IIc
IIaIII
Angle of incidence ( θθθθ )
10 20 504030
Cap-rock = IV
0 60
Stanford Rock Physics Laboratory - Gary Mavko
228
A North Sea Example
Facies-Guided Reflectivity Modeling
IIb
VIV
III
IIc
Schematic facies map of sub-marine fan
Stanford Rock Physics Laboratory - Gary Mavko
229
A North Sea Example
Facies-Guided Reflectivity Modeling
Zero offset reflectivity
0.02
0.04
-0.02
0
-0.04
-0.12
-0.16
-0.08
Zero offset reflectivity+ AVO gradient
0
Stanford Rock Physics Laboratory - Gary Mavko
230
A North Sea Example
Seismic Response of Unconsolidated and Cemented High-Porosity Sands
Well #2
A B C D
Well #1
E F
Well #1 Well #2
Unconsolidated (oil) Cemented (oil)
Stanford Rock Physics Laboratory - Gary Mavko
231
A North Sea Example
Zero-offset Reflectivity, R(0)
Relatively high R(0) (blue)
Relatively low R(0) (yellow)
Stanford Rock Physics Laboratory - Gary Mavko
232
A North Sea Example
AVO Gradient, G
Relatively large negativegradient G (yellow)
Relatively smallgradient (blue)
Stanford Rock Physics Laboratory - Gary Mavko
233
A North Sea Example
Bivariate pdfs: R(0) vs. G
Stanford Rock Physics Laboratory - Gary Mavko
234
A North Sea Example
Facies classification usinglinear discriminant analysis
Sample to be classifiedFacies 1
Facies 2 Facies 3
R(0)
G
M2=(x-µi)TΣ-1(x- µi )
= Facies 2
Mahalanobis distance
Stanford Rock Physics Laboratory - Gary Mavko
235
A North Sea Example
Oil sands
Shale
Brine sands
Interbeddedsand-shales
3D seismic lithofacies prediction
Stanford Rock Physics Laboratory - Gary Mavko
236
A North Sea Example
Variability of AVO Response
Monti-Carlo, drawn from pdf, and assuming siltyshale cap rock.
Stanford Rock Physics Laboratory - Gary Mavko
237
A North Sea Example
Seismic Lithofacies PredictionLine 1 Through Well #1
Well #1 (CDP 1464)
1380 1400 1420 1440 1460 1480 1500 1520 1540 1560 1580−0.1
0
0.1
0.2
Seismic lithofacies prediction − Line #1
R(0
)
1380 1400 1420 1440 1460 1480 1500 1520 1540 1560 1580
−0.5
0
0.5
G
1380 1400 1420 1440 1460 1480 1500 1520 1540 1560 1580
2
4
6
8
Fac
ies
1380 1400 1420 1440 1460 1480 1500 1520 1540 1560 1580
2
4
6
8
Top Heimdal
Oil ?
Stanford Rock Physics Laboratory - Gary Mavko
238
A North Sea Example
Seismic Lithofacies PredictionLine 3 Through Well #3
Well #3 (CDP1844)
1700 1750 1800 1850 1900−0.1
0
0.1
R(0
)
Seismic lithofacies prediction − Line #3
1700 1750 1800 1850 1900−0.4
−0.2
0
0.2
G
1700 1750 1800 1850 1900
2
4
6
8
Fac
ies
1700 1750 1800 1850 1900
2
4
6
8
Top Heimdal
Stanford Rock Physics Laboratory - Gary Mavko
239
A North Sea Example
Seismic Lithofacies PredictionLine 2 Through Well #2
Well # 2Well #2 (CDP 2232)
2000 2050 2100 2150 2200 2250 2300−0.1
0
0.1
R(0
)
Seismic lithofacies prediction − Line #2
2000 2050 2100 2150 2200 2250 2300
−0.5
0
0.5
G
2000 2050 2100 2150 2200 2250 2300
2
4
6
8
Fac
ies
2000 2050 2100 2150 2200 2250 2300
2
4
6
8
Top Heimdal
Stanford Rock Physics Laboratory - Gary Mavko
240
A North Sea Example
Conclusions
• Seismic lithofacies link sedimentology to rockphysics.
• Paleocene deep-water clastic systems in the NorthSea show:
- Sand-shale ambiguity in acousticimpedance.
- Sand-shale discrimination in Vp/Vsratio.
• AVO analysis should be applied to predict seismiclithofacies from seismic data in our case.
• Seismic lithofacies can improve our understanding ofseismic signatures in any depositional system.
Stanford Rock Physics Laboratory - Gary Mavko
AVO
241
2
2.5
3
3.5
4
4.5
5
5.5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Vp (k
m/s
)
Porosity
2
2.5
3
3.5
4
4.5
5
5.5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Vp (k
m/s
)
Porosity
2
2.5
3
3.5
4
4.5
5
5.5
6
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
water-saturatedgas-saturated
Vp (k
m/s
)
Porosity
Water-saturated40 MPa
Water-saturated10-40 MPa
Gas andWater-saturated10-40 MPa
L8
Stanford Rock Physics Laboratory - Gary Mavko
AVO
242
N.1
More than 400 sandstone data points, with porositiesranging over 4-39%, clay content 0-55%, effectivepressure 5-40 MPa - all water saturated.
When Vp is plotted vs. Vs, they follow a remarkablynarrow trend. Variations in porosity, clay, and pressuresimply move the points up and down the trend.
0 1 2 3 4 5 6 7 80
0.51
1.52
2.53
3.54
Shaly SandsWater Saturated
φφφφ = .22-.36φφφφ = .04-.30φφφφ = .32-.39
Vp (km/s)
Vs (k
m/s
)
(Data from Han, Blangy, Yin)
increasing φφφφ , clay, and pore pressure
Stanford Rock Physics Laboratory - Gary Mavko
AVO
243
N.2
Variations in porosity, pore pressure, and shalinessmove data along trends. Changing the pore fluid causes the trend to change.
00.5
11.5
22.5
33.5
4
0 1 2 3 4 5 6 7 8
Vs (k
m/s
)
Vp (km/s)
gas
water
saturation
increasing φφφφ , clay, and pore pressure
Stanford Rock Physics Laboratory - Gary Mavko
AVO
244
00.5
11.5
22.5
33.5
4
0 1 2 3 4 5 6 7 8
Vs (k
m/s
)
Vp (km/s)
saturation
Poor fluid discrimination.Shear not much value.
Good fluid discrimination.Shear could be valuable.
gas
waterpore
pressure
0
2
4
6
8
10
0 5 10 15 20
S-Im
peda
nce
P-Impedance
saturation
Good fluid discrimination.Shear could be valuable.
Less fluid discrimination gas
waterpore
pressure
1
1.5
2
2.5
3
0 1 2 3 4 5
Vp/V
s
Vs
gas
watersaturation
Good fluid discrimination.
Poor fluid discrimination pore pressure
Different shear-related attributes.N.2
Stanford Rock Physics Laboratory - Gary Mavko
AVO
245
Seismic Amplitudes
Many factors influence seismic amplitude:
• Source energy and spectrum• Source coupling• Source radiation pattern• Receiver response, coupling, and pattern• Scattering• Spherical divergence• Focusing• Anisotropy• Intrinsic attenuation• Reflection coefficient
S R
Stanford Rock Physics Laboratory - Gary Mavko
AVO
246
Because there are so many factors affecting amplitude, it is desireable, in almost any analysis of amplitudes, to take advantage of relative measurements, to look for redundancy or changes in time and/or space:
• Surface-consistent analysis of data• Changes in reflection character along a horizon • Differences in time during production or an EOR operation• Variations with azimuth
Stanford Rock Physics Laboratory - Gary Mavko
AVO
247
Very large amplitude reflections on a stacked section are sometimes called “Bright Spots”.
The reflection coefficient of a normally-incident P-wave on a boundary is given by:
where ρρρρV is the acoustic impedence. So that anything that causes a large contrast in impedance can cause a large reflection. Candidates include:
• changes in lithology• changes in saturation
Bright Spots
R = ρρρρ2V2 – ρρρρ1V1
ρρρρ2V2 + ρρρρ1V1
Stanford Rock Physics Laboratory - Gary Mavko
AVO
248
Example of a bright spot.
N.3
Stanford Rock Physics Laboratory - Gary Mavko
AVO
249
In an isotropic medium, a wave that is incident on a boundary will generally create two reflected waves (oneP and one S) and two transmitted waves. The total sheartraction acting on the boundary in medium 1 (due to thesummed effects of the incident an reflected waves) mustbe equal to the total shear traction acting on the boundary in medium 2 (due to the summed effects of the transmitted waves). Also the displacement of a point inmedium 1 at the boundary must be equal to the displace-ment of a point in medium 2 at the boundary.
VP1, VS1, ρ1
VP2, VS2, ρ2
θ1
φ1
θ2φ2
Reflected P-wave
Incident P-wave
Reflected S-wave
Transmitted P-wave
Transmitted S-wave
sin θθθθ1VP1
= sin θθθθ2VP2
= sin φφφφ1VS1
= sin φφφφ2VS2
N.4
Stanford Rock Physics Laboratory - Gary Mavko
AVO
250
sin θ1 cos φ1 –sin θ2 cos φ2
–cos θ1 sin φ1 –cos θ2 –sin φ2
sin 2θ1V1S1
cos 2φ1ρ2S2
2V1ρ1S1
2V2sin 2θ2 –ρ2S2V1
ρ1S12 cos 2φ2
cos 2φ1 –S1V1
sin 2φ1 –ρ2V2ρ1V1
cos 2φ2 – ρ2S2ρ1V1
sin 2φ2
ARPARSATPATS
=
–sin θ1–cos θ1sin 2θ1
–cos 2φ1
By matching the traction and displacement boundary conditions, Zoeppritz (1919) derived the equations relating the amplitudes of the P and S waves:
Stanford Rock Physics Laboratory - Gary Mavko
AVO
251
AVO - Shuey's Approximation
P-wave reflectivity versus angle:
R(θθθθ) ≈≈≈≈ R0 + ER0 + ∆ν∆ν∆ν∆ν
1–νννν 2 sin2θθθθ + 12
∆∆∆∆VPVP
tan2θθθθ – sin2θθθθ
R0 ≈≈≈≈ 12
∆∆∆∆VPVP
+ ∆ρ∆ρ∆ρ∆ρρρρρ
E = F – 2(1 + F)1 – 2νννν
1 – νννν
F = ∆∆∆∆VP /VP∆∆∆∆VP /VP + ∆ρ∆ρ∆ρ∆ρ /ρρρρ
∆∆∆∆VP = VP2 – VP1
∆ρ∆ρ∆ρ∆ρ = ρρρρ2 – ρρρρ1
VP = VP2 + VP1 /2
ρρρρ = ρρρρ2 + ρρρρ1 /2
∆∆∆∆VS = VS2 – VS1 VS = VS2 + VS1 /2
Stanford Rock Physics Laboratory - Gary Mavko
AVO
252
AVO - Aki-Richard's approximation:
P-wave reflectivity versus angle:
∆∆∆∆VP = VP2 – VP1
∆ρ∆ρ∆ρ∆ρ = ρρρρ2 – ρρρρ1
VP = VP2 + VP1 /2
ρρρρ = ρρρρ2 + ρρρρ1 /2
R(θθθθ) ≈≈≈≈ R0 + 1
2∆∆∆∆VPVP
– 2VS2
VP2
∆ρ∆ρ∆ρ∆ρρρρρ + 2∆∆∆∆VS
VSsin2θθθθ
+ 12
∆∆∆∆VPVP
tan2θθθθ – sin2θθθθ
R0 ≈≈≈≈ 1
2∆∆∆∆VPVP
+ ∆ρ∆ρ∆ρ∆ρρρρρ
∆∆∆∆VS = VS2 – VS1 VS = VS2 + VS1 /2
Stanford Rock Physics Laboratory - Gary Mavko
AVO
253
AVO Response
P-Velocity Poisson ratio AVO response contrast contrast
negative negative increasenegative positive decreasepositive negative decreasepositive positive increase
Stanford Rock Physics Laboratory - Gary Mavko
AVO
254
Vp-Vs Relations
There is a wide, and sometimes confusing, variety of published Vp-Vs relations and Vs prediction techniques, which at first appear to be quite distinct. However, most reduce to the same two simple steps:
1. Establish empirical relations among Vp, Vs, and porosity for one reference pore fluid--most often water saturated or dry.
2. Use Gassmann’s (1951) relations to map these empirical relations to other pore fluid states.
Although some of the effective medium models predict both P and S velocities assuming idealized poregeometries, the fact remains that the most reliable and most often used Vp-Vs relations are empirical fits to laboratory and/or log data. The most useful role of theoretical methods is extending these empirical relations to different pore fluids or measurement frequencies. Hence, the two steps listed above.
Stanford Rock Physics Laboratory - Gary Mavko
AVO
255
0
1
2
3
4
5
6
7
0 0.5 1 1.5 2 2.5 3 3.5 4
SandstonesWater Saturated
Vp (k
m/s
)
Vs (km/s)
mudrockVs = .8621V p-1.1724
Castagna et al. (1993)Vs = .8042V p-.8559
Han (1986)Vs = .7936V p-.7868
(after Castagna et al., 1993)
0
1
2
3
4
5
6
7
0 0.5 1 1.5 2 2.5 3 3.5 4
ShalesWater Saturated
Vp (k
m/s
)
Vs (km/s)
mudrockVs = .8621 V p- 1.1724
Castagna et al. (1993)V s = .8042V p-.8559
Han (1986)Vs = .7936V p-.7868
(after Castagna et al., 1993)
N.5
Stanford Rock Physics Laboratory - Gary Mavko
AVO
256
0
1
2
34
5
6
78
0 0.5 1 1.5 2 2.5 3 3.5 4
LimestonesWater Saturated
Vp (k
m/s
)
Vs (km/s)
V s = V p/1.9Pickett (1963)
Castagna et al. (1993)Vs = -.05508 V P2 + 1.0168 V p - 1.0305
water (after Castagna et al.,1993)
01
2
3
45
6
7
8
0 0.5 1 1.5 2 2.5 3 3.5 4
DolomiteWater Saturated
Vp (k
m/s
)
Vs (km/s)
Castagna et al. (1993)V s = .5832V p -.07776
Pickett (1963)Vs = V p/1.8
(after Castagna et al., 1993)
N.6
Stanford Rock Physics Laboratory - Gary Mavko
AVO
257
0
1
2
3
4
5
6
0 0.5 1 1.5 2 2.5 3 3.5 4
Shaly SandstonesWater Saturated
Vp-sat c>.25Vp-sat c<.25
Vp (k
m/s
)
Vs (km/s)
clay > 25 % Vs=.8423Vp-1.099
clay < 25 %Vs=.7535Vp-.6566
mudrock Vs=.8621Vp-1.1724
(Data from Han, 1986)
0
1
2
3
4
5
6
0 0.5 1 1.5 2 2.5 3 3.5 4
Shaly SandstonesWater Saturated (hf)
Vp-sat Phi>.15Vp-sat Phi<.15
Vp (k
m/s
)
Vs (km/s)
porosity > 15 % Vs = .7563Vp-.6620
porosity < 15 %Vs = .8533Vp-1.1374
mudrock Vs = .8621Vp-1.1724
(Data from Han, 1986)
N.7
Stanford Rock Physics Laboratory - Gary Mavko
AVO
258
Dry Poisson’s Ratio Assumption
02468
10121416
0 5 10 15 20 25 30 35
clay < 10%clay > 10%
Vs2
dry
Vp2 dry
ν = 0.01
ν = 0.1
ν = 0.2
ν = 0.3
ν = 0.4
Shaly Sandstones - Dry
The modified Voigt Average Predicts linear moduli-porosity relations. This is a convenient relation for use with the critical porosity model.
These are equivalent to the dry rock Vs/Vp relation and the dry rock Poisson’s ratio equal to their values for pure mineral.
The plot below illustrates the approximately constant dry rock Poisson’s ratio observed for a large set of ultrasonic sandstonevelocities (from Han, 1986) over a large rance of effective pressures (5 < Peff < 40 MPa) and clay contents (0 < C < 55% by volume).
Kdry = K0 1 – φφφφ
φφφφc
µµµµdry = µµµµ0 1 – φφφφφφφφc
VSVP dry rock
≈≈≈≈ VSVP mineral
ννννdry rock ≈≈≈≈ ννννmineral
N.8
Stanford Rock Physics Laboratory - Gary Mavko
AVO
259
Krief’s Relation (1990)
The model combines the same two elements:
1. An empirical Vp-Vs-φφφφ relation for water-saturated rocks, which is approximately the same as the critical porosity model.2. Gassmann’s relation to extend the empirical relation to other pore fluids.
Dry rock Vp-Vs-φφφφ relation:
where ββββ is Biot’s coefficient. This is equivalent to:
where
ββββ and Kφφφφ are two equivalent descriptions of the pore space stiffness. Determining ββββ vs. φφφφ or Kφφφφ vs φφφφdetermines the rock bulk modulus Kdry vs φφφφ.
Krief et al. (1990) used the data of Raymer et al. (1980) to empirically find a relation for ββββ vs φφφφ:
Kdry = Kmineral 1 – ββββ
1Kdry
= 1K0
+ φφφφKφφφφ
1Kφφφφ
= 1vp
dvpdσσσσ PP = constant
; ββββ =dvpdV PP = constant
=φφφφ Kdry
Kφφφφ
1 – ββββ = 1 – φφφφ m(φφφφ) where m(φφφφ) = 3/ 1 – φφφφ
Stanford Rock Physics Laboratory - Gary Mavko
AVO
260
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Mod
ulus
(Nor
mal
ized
)
Porosity
Critical Porosity Model
Krief Model
φφφφc
Assuming dry rock Poisson’s ratio is equal to the mineral Poisson’s ratio gives
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Bio
t Coe
ffici
ent
Porosity
Critical Porosity Model
Krief Model
φφφφc
Kdry = K0 1 – φφφφ m(φφφφ) where m(φφφφ) = 3/ 1 – φφφφ
µµµµdry = µµµµ0 1 – φφφφ m(φφφφ)
N.9
Stanford Rock Physics Laboratory - Gary Mavko
AVO
261
VP–sat
2 – Vfl2
VS–sat2 = VP0
2 – Vfl2
VS02
Expressions for any other pore fluids are obtained fromGassmann’s equations. While these are nonlinear, they suggest a simple approximation:
where VP-sat, VP0, and Vfl are the P-wave velocities of the saturated rock, the mineral, and the pore fluid; and VS-sat and VS0 are the S-wave velocities in the saturated rock and mineral. Rewriting slightly gives
where VR is the velocity of a suspension of minerals in a fluid, given by the Reuss average at the critical porosity.
This modified form of Krief’s expression is exactly equivalent to the linear (modified Voigt) K vs φφφφ and µµµµ vs φφφφrelations in the critical porosity model, with the fluid effectsgiven by Gassmann.
VP–sat
2 = Vfl2 + VS–sat
2 VP02 – Vfl
2
VS02
which is a straight line (in velocity-squared) connecting the mineral point ( ) and the fluid point ( ). A more accurate (and nearly identical) model is to recognize that velocities tend toward those of a suspension at high porosity, rather than toward a fluid, which yields the modified form
VP02 , VS0
2 Vfl2 , 0
VP–sat
2 – VR2
VS–sat2 = VP0
2 – VR2
VS02
Stanford Rock Physics Laboratory - Gary Mavko
AVO
262
0
5
10
15
20
25
30
35
0 5 10 15 20
Vp-Vs Relation in Dry andSaturated Rocks
V p2 (k
m/s
)2
Vs2 (km/s)2
saturated
dry
Sandstones mineral point
fluid point
N.10
Stanford Rock Physics Laboratory - Gary Mavko
AVO
263
N.11
0
10
20
30
40
50
0 4 8 12 16
Vp-Vs Relation in Sandstoneand Dolomite
V p2 (k
m/s
)2
Vs2 (km/s)2
Sandstone
Dolomite
mineralpoints
fluid points
Stanford Rock Physics Laboratory - Gary Mavko
AVO
264
VS = 1
2 XiΣΣΣΣi = 1
LaijVP
jΣΣΣΣj = 0
Ni
+ XiΣΣΣΣi = 1
LaijVP
jΣΣΣΣj = 0
Ni –1 –1
Greenberg and Castagna (1992) have given empirical relations for estimating Vs from Vp in multimineralic, brine-saturated rocks based on empirical, polynomial Vp-Vs relations in pure monomineralic lithologies (Castagna et al., 1992). The shear wave velocity in brine-saturated compositelithologies is approximated by a simple average of the arithmetic and harmonic means of the constituent purelithology shear velocities:
Castagna et al. (1992) gave representative polynomial regression coefficients for pure monomineralic lithologies:
Regression coefficients for pure lithologies with Vp and Vs in km/s:
Xi = 1ΣΣΣΣi = 1
L
VS = ai2VP2 + ai1VP + ai0 (Castagna et al. 1992)
whereL number of pure monomineralic lithologic constituentsXi volume fractions of lithological constituents aij
empirical regression coefficientsNi order of polynomial for constituent iVp, Vs P and S wave velocities (km/s) in composite brine-saturated,
multimineralic rock
Lithology a i2 a i1 a i0
S andstone 0 0.80416 -0.85588Limestone -0.05508 1.01677 -1.03049Dolomite 0 0.58321 -0.07775S hale 0 0.76969 -0.86735
Stanford Rock Physics Laboratory - Gary Mavko
AVO
265
1
2
3
4
5
6
7
0 1 2 3 4 5
SandstoneLimestoneDolomiteShale
Vp (k
m/s
)
Vs (km/s)
Note that the above relation is for 100% brine-saturated rocks. To estimate Vs from measured Vp for other fluid saturations,Gassmann’s equation has to be used in an iterative manner. In the following, the subscript b denotes velocities at 100% brine saturation and the subscript f denotes velocities at any other fluid saturation (e.g. this could be oil or a mixture of oil, brine, and gas). The method consists of iteratively finding a (Vp, Vs) point on the brine relation that transforms, withGassmann’s relation, to the measured Vp and the unknown Vs for the new fluid saturation. the steps are as follows:
1. Start with an initial guess for VPb.2. Calculate VSb corresponding to VPb from the empirical regression.3. Do fluid substitution using VPb and VSb in the Gassmannequation to get VSf.4. With the calculated VSf and the measured VPf, use theGassmann relation to get a new estimate of VPb. Check with previous value of VPb for convergence. If convergence criterion is met, stop; if not go back to step 2 and continue.
N.12
Stanford Rock Physics Laboratory - Gary Mavko
AVO
266
0
5
10
15
0 5 10 15 20 25 30 35
Vs2
Vp
2
dry
saturatedlow frequency
saturatedultrasonic
1
1.5
2
2.5
3
3.5
4
2 2.5 3 3.5 4 4.5 5 5.5 6
Vs
Vp
dry
saturatedlow frequency
saturatedultrasonic
Ultrasonic data from Han (1986)
N.14
Stanford Rock Physics Laboratory - Gary Mavko
AVO
267
Coe fficie n t s for t h e e q u a t ion ρb = aVp2 + bVp + c
Li t h o l o g y a b c V p Ra n g e ( Km / s )
Sh a le - .0 2 6 1 .3 7 3 1 .4 5 8 1 .5 - 5 .0Sa n d s t on e - .0 1 1 5 .2 6 1 1 .5 1 5 1 .5 - 6 .0Lim e s t on e - .0 2 9 6 .4 6 1 0 .9 6 3 3 .5 - 6 .4Dolom it e - .0 2 3 5 .3 9 0 1 .2 4 2 4 .5 - 7 .1An h y d r it e - .0 2 0 3 .3 2 1 1 .7 3 2 4 .6 - 7 .4
Coe fficie n t s for t h e e q u a t ion ρb = dVpf
Li t h o l o g y d f V p Ra n g e( Km / s )
Sh a le 1 .7 5 .2 6 5 1 .5 - 5 .0Sa n d s t on e 1 .6 6 .2 6 1 1 .5 - 6 .0Lim e s t on e 1 .5 0 .2 2 5 3 .5 - 6 .4Dolom it e 1 .7 4 .2 5 2 4 .5 - 7 .1An h y d r it e 2 .1 9 .1 6 0 4 .6 - 7 .4
1
2
3
4
5
6
1.8 2 2.2 2.4 2.6 2.8
Shales
Vp (k
m/s
)
density (g/cm 3)
ρρρρ = -.0261Vp 2 +.373Vp+1.458
ρρρρ = 1.75Vp .265
Both forms of Gardner’s relations applied to log and lab shale data, as presented by Castagna et al. (1993)
Polynomial and powerlaw forms of the Gardner et al. (1974) velocity-density relationships presented by Castagna et al. (1993). Units are km/s and g/cm3.
N.15
Stanford Rock Physics Laboratory - Gary Mavko
AVO
268
3
3.5
4
4.5
5
5.5
6
6.5
7
2 2.2 2.4 2.6 2.8 3
Dolomite
Vp (k
m/s
)
density (g/cm 3)
ρρρρ = -.0235Vp 2 +.390Vp +1.242
ρρρρ = 1.74Vp .252
Both forms of Gardner’s relations applied to laboratory dolomite data.
N.16
Stanford Rock Physics Laboratory - Gary Mavko
AVO
269
1
2
3
4
5
6
1.8 2 2.2 2.4 2.6 2.8
SandstonesVp
(km
/s)
density (g/cm 3)
ρρρρ = -.0115Vp 2 +.261Vp+1.515
ρρρρ = 1.66Vp .261
1
2
3
4
5
6
1.8 2 2.2 2.4 2.6 2.8
Limestones
Vp (k
m/s
)
density (g/cm 3)
ρρρρ = -.0296Vp 2 +.461Vp +0.963
ρρρρ = 1.5Vp .225ρρρρ = 1.359Vp .386
Both forms of Gardner’s relations applied to laboratory limestone data. Note that the published powerlaw form does not fit as well as the polynomial. we also show a powerlaw form fit to these data, which agrees very well with the polynomial.
Both forms of Gardner’s relations applied to log and lab sandstone data, as presented by Castagna et al. (1993).
N.17
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
270
Reservoir modeling
Flow simulation
Rock physics modeling of
elastic properties
Synthetic seismic imaging
Comparison with field
4D seismic
Update model Reservoir management decisions
Steps involved in monitoring
reservoir performance
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
271
Rock Physics 4D Modeling Steps
Model-Building• Build geologic reservoir model:
lithology, porosity, perm at each pixel• Specify reservoir fluid properties:
gravity, GOR, salinity, etc.• Assign Vp, Vs, and density to each pixel
of geologic model for reference fluid
Upscale and Flow Simulate
Rock Physics Mapping to Seismic – ateach time step and each pixel:• Simulator predicts P, T, So, Sw, Sg• Model K, ρρρρ of water, oil, gas • Compute “average” fluid properties• downscale/upscale?• Use Gassmann to update Vp, Vs, density• Adjust Vp, Vs for changes in P
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
272
� Assessing feasibility � Interpreting data after acquisition
Reservoir Monitoring Analysis
� Realistic synthetic data� Based on North Sea:– Braided river channels– 5 km by 5 km, 2 faults– 24 production wells and 17 water injectors
�Geophysics:Biondo Biondi, David Lumley, Gary Mavko, Tapan Mukerji, James Rickett�Petroleum Engineering:Clayton Deutsch, Runar Gundesø, Marco Thiele�SRB, SEP, SCRF, SUPRI-B�Chevron, Norsk-Hydro
Study was performed by various groups at Stanford:
Example feasibility study:
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
273
Reservoir Model
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
274
� Hierarchical object-based approach– Define stratigraphic layers– Model channel sand/shale facies– Assign porosity/permeability within facies
� 3 layers with net:gross = 0.8, 0.4, 0.6
� Within channel sands φφφφmean = 23%
� 200 million geological modelling cells
Geologic/Geostatistical Modeling
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
275
Hierarchical Modeling
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
276
� Lithologies and Fluids1. Porosity from geologic modeling2. Vp-Vs-φφφφ-permeability relations 3. Saturations and pressures from flow simulation4. Gassmann for fluid substitution
� Scaling and Gridding
Rock Physics
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
277
Rock Physics Relations
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
278
� PVT-properties and relative permeabilities from a North-Sea dataset� Faults as non-nearest neighbor connections (sealing and non-sealing)� Simulation run for 3 years: 6 months primary production 2.5 years with water injection� Computationally intensive: importance of scale-up
Flow Simulation
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
279
Flow Simulation Grid
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
280
Gas and Water Saturations
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
281
• Scattering using first-order Born approximation- Plane-layer methods cannot deal with
heterogeneities- Finite difference methods too slow
• Amplitudes from Rayleigh scattering
Seismic Modeling
• Survey design- Area: 2.4 km x 3 km - Wavelet: 45 Hz - Offsets: 0 km and 3 km - Completed at t = 0 and t = 3 years
• Constant offset migration preserving true relative amplitudes
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
282
4-D Seismic
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
283
4-D Seismic
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
284
4-D Seismic
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
285
4-D Seismic AVO Effects
Stanford Rock Physics Laboratory - Gary Mavko
Example 4D Feasibility Study
286
� Rock/fluid properties� Scales of saturation� Frequency effects� Upscaling
Rock Physics Issues
� Survey Repeatability� Resolution� AVO information� Acquisition geometry � Differential seismic attributes
Seismic Imaging Issues
General Issues
• Different data formats andcoordinates• Integration of expertise, not
just data
Chalks
Water saturated Measurement type: sonic logs Effective pressure: Reference: Urmos J., and Wilkens R. H., 1993, In situ velocities in pelagic carbonates:
new insights from ocean drilling program leg 130, Ontong Java, Journ. Geophys. Res. 98, No. B5, pp 7903-7920.
min max mean std. dev
Vp 1.5250 4.2960 2.1571 0.3053 Vs 1.5930 2.5120 2.0304 0.3030
Vp/Vs 1.6174 1.7872 1.6700 0.0592 porosity 0.0995 0.7488 0.5026 0.0757 density 1.4296 2.5590 1.8498 0.1285
impedance 2.2973 10.9935 4.0249 0.8859
1.4
1.6
1.8
2
2.2
2.4
2.6
1.5 2 2.5 3 3.5 4 4.5
dens
ity (g
/cm
3)
Vp (km/s)
ρρρρ = 1.045 + 0.37308VpR = 0.88676
0 50 100 150 200 2501.4
1.6
1.8
2
2.2
2.4
2.6
count
dens
ity (g
/cm
3) mean: 1.8498
std. dev: 0.12845no. of points: 552
1.4
1.6
1.8
2
2.2
2.4
2.6
1.5 2 2.5 3 3.5 4 4.5
Vs (k
m/s
)
Vp (km/s)
Vs = 0.34543 + 0.49511VpR = 0.98899
1.6
1.65
1.7
1.75
1.8
2.5 3 3.5 4 4.5
Vp/V
s
Vp (km/s)
Vp/Vs = 1.3886 + 0.082698VpR = 0.84521
0
50
100
150
200
250
300
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
coun
t
porosity
mean: 0.5026std. dev: 0.075654no. of points: 552
1.5
2
2.5
3
3.5
4
4.5
5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Vp (k
m/s
)
porosity
Vp = 5.0598 φφφφ2 - 8.5052 φφφφ + 5.1284R = 0.92231
0 50 100 1501.5
2
2.5
3
3.5
4
4.5
count
Vp (k
m/s
) mean: 2.1571std. dev: 0.30532no. of points: 552
1.4
1.6
1.8
2
2.2
2.4
2.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Vs (k
m/s
)
porosity
Vs = 2.7665 + -2.9334 φφφφR = 0.9927
0 0.5 1 1.5 2 2.51.4
1.6
1.8
2
2.2
2.4
2.6
count
Vs (k
m/s
)
mean: 2.0304std. dev: 0.30296no. of points: 12
2
4
6
8
10
12
14
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
impe
danc
e 1
06
(kg/
m3
)(m/s
)
porosity
I = 1.678 φφφφ2 + -2.712 φφφφ + 1.333R = 0.96574
0 50 100 150 2002
4
6
8
10
12
count
impe
danc
e 1
06
(kg/
m3
)(m/s
)
mean: 4.0249std. dev: 0.88592no. of points: 552
Dolomite
Water saturated calculated from dry data using Gassmann equations Measurement type: ultrasonic Effective pressure Reference: Geertsma, J., 1961, Velocity-log interpretation: the effect of rock bulk
compressibility, Society of Petroleum Engineers Journal, December. Yale, D. P., and Jamieson Jr., W. H., 1994, Static and dynamic rock mechanical
properties in the Hugoton and Panoma fields, Kansas, SPE paper 27939, presented at the SPE Mid-Continent Gas symposium, Amarillo, Texas, May, 1994.
min max mean std. dev
Vp 3.4068 7.0214 5.3901 0.6935 Vs 2.0116 3.6443 2.9697 0.36614
Vp/Vs 1.5916 2.0855 1.8152 0.0745 porosity 0.0030 0.3210 0.1297 0.0649 density 0 3.6443 2.5673 0.2538
impedance 0.7836 1.9302 1.4010 0.2321
2
2.2
2.4
2.6
2.8
3
3 4 5 6 7 8
dens
ity (g
/cm
3)
Vp (km/s)
ρρρρ = 1.8439 + 0.13786VpR = 0.82496
0 5 10 15 202.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
count
dens
ity (g
/cm
3)
mean: 2.5673std. dev: 0.2538no. of points: 131
2
2.5
3
3.5
4
3 4 5 6 7 8
Vs (k
m/s
)
Vp (km/s)
Vs = 0.25524 + 0.5036VpR = 0.95378
1.5
1.6
1.7
1.8
1.9
2
2.1
3 4 5 6 7 8
Vp/V
s
Vp (km/s)
Vp/Vs = 1.6558 + 0.029589VpR = 0.27552
0
5
10
15
20
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
coun
t
porosity
mean: 0.12969std. dev: 0.064875no. of points: 130
3
4
5
6
7
8
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Vp (k
m/s
)
porosity
Vp = 6.6067 + -9.3808 φφφφR = 0.87762
0 5 10 15 20 253
4
5
6
7
8
count
Vp (k
m/s
)
mean: 5.3901std. dev: 0.69345no. of points: 130
2
2.5
3
3.5
4
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Vs (k
m/s
)
porosity
Vs = 3.5817 + -4.7194 φφφφR = 0.83621
0 2 4 6 8 10 12 14 162
2.5
3
3.5
4
count
Vs (k
m/s
)
mean: 2.9697std. dev: 0.36614no. of points: 130
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
impe
danc
e 1
07
(kg/
m3
)(m/s
)
porosity
I = 1.833 + -3.3081 φφφφR = 0.92461
0 5 10 15 200.6
0.8
1
1.2
1.4
1.6
1.8
2
count
impe
danc
e 1
07
(kg/
m3
)(m/s
)
mean: 1.401std. dev: 0.23211no. of points: 130
Sandstones
Water saturated Measurement type: ultrasonic Effective pressure: 30, 40 MPa Reference: Han, D.-H., 1986, Effects of Porosity and Clay Content on Acoustic
Properties of Sandstones and Unconsolidated Sediments, Ph.D. dissertation, Stanford University.
min max mean std. dev
Vp 3.1300 5.5200 4.0904 0.5051 Vs 1.7300 3.6000 2.4094 0.3966
Vp/Vs 1.5333 1.8866 1.7091 0.0833 porosity 0.0412 0.2993 0.1642 0.0707 density 2.0900 2.6400 2.3678 0.1348
impedance 6.6044 13.9656 9.7344 1.6378
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
3 3.5 4 4.5 5 5.5 6
dens
ity (g
/cm
3)
Vp (km/s)
Vp = 1.5694 + 0.19519 ρρρρR = 0.7312
0 5 10 152
2.1
2.2
2.3
2.4
2.5
2.6
2.7
count
dens
ity (g
/cm
3)
mean: 2.3678std. dev: 0.13483no. of points: 108
1.5
2
2.5
3
3.5
4
3 3.5 4 4.5 5 5.5 6
Vs (k
m/s
)
Vp (km/s)
Vs = -0.73327 + 0.7683VpR = 0.97843
1.2
1.3
1.41.5
1.6
1.7
1.8
1.9
2
3 3.5 4 4.5 5 5.5 6
Vp/V
s
Vp (km/s)
Vp/Vs = 2.1909 + -0.1178VpR = 0.71418
0
5
10
15
20
0 0.1 0.3 0.4
coun
t
porosity
mean: 0.16417std. dev: 0.070719no. of points: 108
3
3.5
4
4.5
5
5.5
6
0 0.1 0.3 0.4
Vp (k
m/s
)
porosity
Vp = 4.9442 + -5.2011 φφφφR = 0.72826
0 5 10 153
3.5
4
4.5
5
5.5
6
count
Vp (k
m/s
)
mean: 4.0904std. dev: 0.50507no. of points: 108
1.5
2
2.5
3
3.5
4
0 0.1 0.3 0.4
Vs (k
m/s
)
porosity
Vs = 2.9814 + -3.4843 φφφφR = 0.6213
0 2 4 6 8 10 12 14 161.5
2
2.5
3
3.5
4
count
Vs (k
m/s
)
mean: 2.4094std. dev: 0.3966no. of points: 108
6
8
10
12
14
0 0.05 0.1 0.15 0.2 0.25 0.3
impe
danc
e 1
06 (k
g/m
3 )(m/s
)
porosity
I = 12.954 + -19.612 φφφφR = 0.84679
0 2 4 6 8 10 12 146
8
10
12
14
count
impe
danc
e 1
06
(kg/
m3
)(m/s
)
mean: 9.7344std. dev: 1.6378no. of points: 108
Tight Gas Sandstones
Dry Measurement type: ultrasonic Effective pressure: 40 MPa Reference: Jizba, D.L., 1991, Mechanical and Acoustical Properties of Sandstones and
Shales, Ph.D. dissertation, Stanford University.
min max mean std. dev Vp 3.8090 5.5730 4.6688 0.3778 Vs 2.5940 3.5030 3.0603 0.2317
Vp/Vs 1.4169 1.6847 1.5258 0.0539 porosity 0.0080 0.1440 0.0521 0.0370 density 2.5780 2.6730 2.5119 0.1129
impedance 0.8902 1.4897 1.1744 12.5620
2.2
2.3
2.4
2.5
2.6
2.7
3.5 4 4.5 5 5.5 6
dens
ity (g
/cm
3)
Vp (km/s)
ρρρρ = 1.9617 = 0.11787VpR = 0.39424
0 1 2 3 4 5 6 7 82.2
2.3
2.4
2.5
2.6
2.7
count
dens
ity (g
m/c
m3
)
mean: 2.5119std. dev: 0.11294no. of points: 42
2.4
2.6
2.8
3
3.2
3.4
3.6
3.5 4 4.5 5 5.5 6
Vs (k
m/s
)
Vp (km/s)
Vs = 0.4855 + 0.55149VsR = 0.8997
1.3
1.35
1.4
1.45
1.5
1.55
1.6
1.65
1.7
3.5 4 4.5 5 5.5 6
Vp/V
s
Vp (km/s)
Vp/Vs = 1.2686 + 0.055098VpR = 0.38609
0
2
4
6
8
10
0 0.1 0.2
coun
t
porosity
mean: 0.052143std. dev: 0.036989no. of points: 42
3.5
4
4.5
5
5.5
6
0 0.1 0.2
Vp (k
m/s
)
porosity
Vp = 4.8689 + -3.8366 φφφφR = 0.37566
0 1 2 3 4 5 6 73.5
4
4.5
5
5.5
6
count
Vp (k
m/s
)mean: 4.6688std. dev: 0.3778no. of points: 42
2.4
2.6
2.8
3
3.2
3.4
3.6
0 0.1 0.2
Vs (k
m/s
)
porosity
Vs = 301491 + -1.7039 φφφφR = 0.27218
0 1 2 3 4 5 6 72.4
2.6
2.8
3
3.2
3.4
3.6
count
Vs (k
m/s
)
mean: 3.0603std. dev: 0.23156no. of points: 42
0.8
0.9
1.1
1.2
1.3
1.5
1.6
0 0.03 0.06 0.09 0.11 0.14 0.17
impe
danc
e 1
07
(kg/
m3
)(m/s
)
porosity
I = 1.2957 + -2.325 φφφφR = 0.68459
0 2 4 6 8 10
0.8
1
1.2
1.3
1.5
1.7
count
impe
danc
e 1
07
(kg/
m3
)(m/s
)
mean: 1.1744std. dev: 12.562no. of points: 42
Limestone
Water saturated and calculated from dry data using Gassmann equations Measurement type: ultrasonic, resonant bar Effective pressure: 10, 30, 40, 50 MPa Reference: Cadoret, T., 1993, Effet de la Saturation Eau/Gaz sur les Propriétés
Acoustiques des Roches, Ph.D. dissertation, University of Paris, VII. Lucet, N., 1989, Vitesse et attenuation des ondes elastiques soniques et ultrasoniques
dans les roches sous pression de confinement, Ph.D. dissertation, University of Paris. Yale, D. P., and Jamieson Jr., W. H., 1994, Static and dynamic rock mechanical
properties in the Hugoton and Panoma fields, Kansas, SPE paper 27939, presented at the SPE Mid-Continent Gas symposium, Amarillo, Texas, May, 1994.
min max mean std. dev Vp 3.3853 5.7930 4.6297 0.6590 Vs 1.6660 3.0350 2.4359 0.3688
Vp/Vs 1.7215 2.0390 1.87713 0.0805 porosity 0.0340 0.4130 0.1496 0.0878 density 2.0038 2.6512 2.4264 0.1579
impedance 0.6871 1.5080 1.4320 0.21978
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
3 3.5 4 4.5 5 5.5 6
dens
ity (g
/cm
3)
Vp (km/s)
ρρρρ = 1.5138 + 0.20238VpR = 0.89241
0 2 4 6 8 102
2.1
2.2
2.3
2.4
2.5
2.6
2.7
count
dens
ity (g
/cm
3)
mean: 2.4264std. dev: 0.15792no. of points: 55
1.6
1.8
2
2.2
2.4
2.62.8
3
3.2
3 3.5 4 4.5 5 5.5 6
Vs (k
m/s
)
Vp (km/s)
Vs = -0.039208 + 0.5443VpR = 0.9586
1.6
1.7
1.8
1.9
2
2.1
3 3.5 4 4.5 5 5.5 6
Vp/V
s
Vp (km/s)
Vp/Vs = 1.9511 + -0.017304VpR = 0.14147
0
2
4
6
8
10
0 0.1 0.2 0.3 0.4 0.5
coun
t
porosity
mean: 0.14964std. dev: 0.087839no. of points: 49
3
3.5
4
4.5
5
5.5
6
0 0.1 0.2 0.3 0.4 0.5
Vp (k
m/s
)
porosity
Vp = 5.6248 + -6.65 φφφφR = 0.88643
0 2 4 6 8 10 123
3.5
4
4.5
5
5.5
6
count
Vp (k
m/s
)
mean: 4.6297std. dev: 0.65897no. of points: 49
1.6
1.8
2
2.2
2.4
2.62.8
3
3.2
0 0.1 0.2 0.3 0.4 0.5
Vs (k
m/s
)
porosity
Vp = 3.053 + -3.8664 φφφφR = 0.9179
0 2 4 6 8 10 12 141.6
1.8
2
2.2
2.4
2.6
2.8
3
3.2
count
Vs (k
m/s
)
mean: 2.4359std. dev: 0.36878no. of points: 54
0.6
0.8
1
1.2
1.4
1.6
0 0.08 0.15 0.22 0.3 0.37 0.45
impe
danc
e 1
07
(kg/
m3
)(m/s
)
porosity
I = 1.4922 + -2.3319 φφφφR = 0.93198
0 2 4 6 8 100.6
0.8
1
1.2
1.4
1.6
1.8
count
impe
danc
e 1
07
(kg/
m3
)(m/s
)
mean: 1.432std. dev: 0.21978no. of points: 49
High Porosity Sandstones
Water saturated Measurement type: ultrasonic Effective pressure: 35, 40 MPa Reference: Strandenes, Sverre, 1991, Rock Physics Analysis of the Brent Group
Reservoir in the Oseberg Field, SRB special volume.
min max mean std. dev Vp 3.4547 4.7921 3.7989 0.2356 Vs 1.9533 2.6612 2.1573 0.1529
Vp/Vs 1.6836 1.8813 1.7548 0.1251 porosity 0.0170 0.3206 0.1815 0.0808 density 2.6947 2.1198 2.3282 0.1251
impedance 7.5747 9.9798 8.5693 0.6717
2.1
2.2
2.3
2.4
2.5
2.6
2.7
3.4 3.6 3.8 4 4.2 4.4
dens
ity (g
/cm
3)
Vp (km/s)
ρρρρ = 1.4507 + 0.2192VpR = 0.47861
0 2 4 6 8 10 12 142.1
2.2
2.3
2.4
2.5
2.6
2.7
count
dens
ity (g
/cm
3)
mean: 2.3282std. dev: 0.12514no. of points: 111
1.9
2
2.1
2.2
2.3
2.42.5
2.6
2.7
3.4 3.6 3.8 4 4.2 4.4 4.6 4.8
Vs (k
m/s
)
Vp (km/s)
Vs = 0.0068626 + 0.56839VpR = 0.9438
1.6
1.65
1.7
1.75
1.8
1.85
1.9
3.4 3.6 3.8 4 4.2 4.4 4.6 4.8
Vp/V
s
Vp (km/s)
Vp/Vs = 1.7691 + -0.0037876VpR = 0.023229
0
5
10
15
20
0 0.1 0.3 0.4
coun
t
porosity
mean: 0.18149std. dev: 0.080763no. of points: 159
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
0 0.1 0.3 0.4
Vp (k
m/s
)
porosity
Vp = 4.3038 + -2.2279 φφφφR = 0.51345
0 5 10 15 203.4
3.6
3.8
4
4.2
4.4
4.6
4.8
count
Vp (k
m/s
)
mean: 3.7989std. dev: 0.23562no. of points: 96
1.9
2
2.1
2.2
2.3
2.42.5
2.6
2.7
0 0.1 0.3 0.4
Vs (k
m/s
)
porosity
Vs = 2.4868 + -1.6268 φφφφR = 0.54027
0 2 4 6 8 101.9
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
count
Vs (k
m/s
)
mean: 2.1573std. dev: 0.15287no. of points: 58
6.7
7.5
8.3
9
9.8
10.5
0.13 0.17 0.21 0.25 0.29 0.33
impe
danc
e 1
06
(kg/
m3
)(m/s
)
porosity
I = 10.63 + -8.912 φφφφR = 0.71932
0 2 4 6 8 10 127.5
8
8.5
9
9.5
10
10.5
count
impe
danc
e 1
06
(kg/
m3
)(m/s
)
mean: 8.5693std. dev: 0.67166no. of points: 78
Poorly Consolidated Sandstones
Water saturated Measurement type: ultrasonic Effective pressure: 30 MPa Reference: Blangy, J.P., 1992, Integrated Seismic Lithologic Interpretation: The
Petrophysical Basis, Ph.D. dissertation, Stanford University.
min max mean std. dev Vp 2.4340 3.1400 2.7303 0.1823 Vs 1.2120 1.6610 1.3659 0.1243
Vp/Vs 1.8791 2.2433 2.0193 0.0856 porosity 0.2220 0.3640 0.3058 0.0405 density 2.0087 2.2343 2.1101 0.0537
impedance 4.8892 7.0157 5.7684 0.5118
2
2.1
2.2
2.3
2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2
dens
ity (g
/cm
3)
Vp (km/s)
Vs = 2.1003 + -2.4245 φφφφR = 0.81711
0 2 4 6 8 102
2.1
2.2
2.3
count
dens
ity (g
/cm
3) mean: 2.1101
std. dev: 0.053682no. of points: 38
1.2
1.3
1.4
1.5
1.6
1.7
2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2
Vs (k
m/s
)
Vp (km/s)
Vs = -0.31459 + 0.61088VpR = 0.90174
1.7
1.8
1.9
2
2.1
2.2
2.3
2.4 2.5 2.6 2.7 2.8 2.9 3 3.1 3.2
Vp/V
s
Vp (km/s)
Vp/Vs = 2.4319 + -0.14998VpR = 0.32147
0
1
2
3
4
5
6
7
0.2 0.3 0.4
coun
t
porosity
mean: 0.30576std. dev: 0.040505no. of points: 38
2.4
2.5
2.6
2.7
2.8
2.93
3.1
3.2
0.2 0.3 0.4
Vp (k
m/s
)
porosity
Vp = 3.77420 + -3.4144 φφφφR = 0.75854
0 1 2 3 4 5 62.4
2.5
2.6
2.7
2.8
2.9
3
3.1
3.2
count
Vp (k
m/s
) mean: 2.7303std. dev: 0.18232no. of points: 38
1.2
1.3
1.4
1.5
1.6
1.7
0.2 0.3 0.4
Vs (k
m/s
)
porosity
Vs = 2.1003 + -2.4245 φφφφR = 0.81711
0 1 2 3 4 5 61.2
1.3
1.4
1.5
1.6
1.7
count
Vs (k
m/s
)
mean: 1.3659std. dev: 0.12433no. of points: 33
4
5
6
7
8
0.2 0.23 0.27 0.3 0.33 0.37 0.4
impe
danc
e 1
06
(kg/
m3
)(m/s
)
porosity
I = 8.9745 + -10.485 φφφφR = 0.82938
0 2 4 6 8 10
5
6
7
8
count
impe
danc
e 1
06
(kg/
m3
)(m/s
)
mean: 5.7684std. dev: 0.51179no. of points: 38
REFERENCES Backus, G. E., 1962, Long-wave elastic anisotropy produced by horizontal layering: J. Geophys. Res., 67 , 4427-
4440. Batzle, M. and Wang, Z., 1992, Seismic properties of pore fluids, Geophysics, 57, 11. 1396-1408. Berryman, J.G., 1980, "Confirmation of Biot's theory": Appl. Phys. Lett., 37, 382-384. Berryman, J.G., 1992, "Mixture theories for rock properties", in Handbook of Physical Constants, American
Geophysical Union. Blangy, J.P., 1992, Integrated seismic lithologic interpretation: The petrophysical basis: Ph.D. dissertation, Stanford
University. Bourbié, T., Coussy, O., and Zinszner, B., 1987, Acoustics of Porous Media : Gulf Publishing Co.. Bourbié, T., Coussy, O., and Zinszner, B., 1986, Acoustique Des Milieux Poreux, Editions Technip, Paris. Boyse, W.E., 1986, Wave propagation and inversion in slightly inhomogeneous media: Ph.D. dissertation, Stanford
University. Brown, R., and Korringa, J., 1975, On the dependence of the elastic properties of a porous rock on the compressibility
of the pore fluid, Geophysics, 40, 608-616. Castagna, J.P. and Backus, M.M., 1993, Offset-Dependent Reflectivity-Theory and Practice of AVO Analysis,
Investigations in Geophysics Series, Vol. 8, SEG.
Clark, V.A., 1992, The effect of oil under in-situ conditions on the seismic properties of rocks, Geophyics, 57, 894-
901. Coyner, K.B., 1984, Effects of stress, pore pressure, and pore fluids on bulk strain, velocity, and permeability in rock:
Ph.D. thesis, Massachusetts Institute of Technology. De, G.S., Winterstein, D., and Meadows, M., 1994, Comparison of P- and S-wave velocities and Qs from VSP and
sonic log data, Geophysics, 59, 1512-1529. De Buyl, M., Optimum field development with seismic reflection data, Geophysics: The Leading Edge of Exploration,
(4), 1420, 1989. Dilay, A., and Eastwood, J., 1995, Spectral analysis applied to seismic monitoring of thermal recovery, The Leading
Edge, November 1995. Dutta, N.C., and Levin, F.K, 1987, Geopressure, Society of Exploration Geophysicists.
Dvorkin, J., Mavko, G., and Nur, A., 1990, The oscillation of viscous compressible fluid in an arbitrarily shaped pore:
Mech. of Mat., 9, 165-179.
Dvorkin, J., Nolen-Hoeksema, R., and Nur, A., 1993, The squirt-flow mechanism: Macroscopic description:
Geophysics, 59, 428-438.
Dvorkin, J., Mavko, G., and Nur, A., 1993, Squirt flow in fully saturated rocks: 63th Ann. Internat. Mtg., Soc. Expl.
Geophys., Expanded Abstracts, 805-808. Dvorkin, J., and Nur, A., 1993, Dynamic poroelasticity: A unifies model with squirt and the Biot mechanisms:
Geophysics, 58, 524-533.
Endres, A.L., and Knight, R., 1991, "The effects of pore scale fluid distribution on the physical properties of partially saturated tight sandstones": J. Appl. Physics, 69, 1091-1098.
Fertl, W.H., 1976, Abnormal formation pressures: Elsevier, N.Y. Fertl, W.H., Chapman, R.E., and Hotz, R.F., 1994, Studies in abnormal pressures, Elsevier. Frazer, L. N., 1994, A pulse in a binary sediment, Geophys. J. Int., 118, 75-93. Gardner, G.H.F., Gardner, L.W., and Gregory, A.R., 1974, Formation velocity and density – the diagnostic basis for
stratigraphic traps, Geophysics, 39, 770-780. Geertsma, J., and Smit, D.C., 1961, "Some aspects of elastic wave propagation in fluid-saturated porous solids":
Geophys., 26, 169-181. Greenberg, M.L. and Castagna, J.P., 1992, Shear-wave velocity estimation in porous rocks: theoretical formulation,
preliminary verification and applications, Geophysical Prospecting, 40, 195-209. Gordon, R.B., and Davis, L.A., 1968, "Velocity and attenuation of seismic waves in imperfectly elastic rock": J.
Geophys. Res., 73, 3917-3935. Gordon, R.B., and Radar, D., 1971, "Imperfect elasticity of rock: Its influence on the velocity of stress waves", in The
Structure and Physical Properties of the Earth's Crust, Geophys. Monogr. Ser., vol. 14, edited by J.G. Heacock, p. 235, AGU, Washington, D. C..
Greenberg, M.L. and Castagna, J.P., 1992, Shear-wave velocity estimation in porous rocks: theoretical formulation, preliminary verification and applications, Geophysical Prospecting, 40, 195-209.
Gregory, A.R., 1976, "Fluid saturation effects on dynamic elastic properties of sedimentary rocks": Geophys., 41, 895-921.
Gregory, A.R., 1977, "Aspects of rock physics from laboratory and log data that are important to seismic interpretation": Seismic stratigraphy AAPG memoir, 26, 15-46.
Guéguen, Yves, and Palciauskas, Victor, 1994, Introduction to The Physics of Rocks, Princeton University Press, Princeton, New Jersey.
Guéguen, Yves, and Palciauskas, Victor, 1992, Introduction a la physique des roches, Hermann Editeurs Des Sciences Et Des Arts, Paris.
Hamilton, E.L., 1956, "Low sound velocities in high porosity sediments": J. Acoust. Soc. Am. , 28, 16-19. Han, D., 1986, Effects of porosity and clay content on acoustic properties of sandstones and unconsolidated sediments:
Ph.D. dissertation, Stanford University. Han, D., Nur, A., and Morgan, D., 1986, Effect of porosity and clay content on wave velocity in sandstones":
Geophysics, 51, 2093-2107. Herron, et. al., 1992, SPE 24726. Hudson, J.A., 1980, Overall properties of a cracked solid: Math. Proc. Cambr. Phil. Soc., 88, 371-384. Hudson, J.A. , 1981, Wave speeds and attenuation of elastic waves in material containing cracks: Geophys. J. Roy.
Astr. Soc., 64, 133-150. Jizba, D., 1991, Mechanical and acoustical properties of sandstones and shales: Ph.D. dissertation, Stanford
University. Jones, T., 1983, "Wave propagation in porous rocks and models for crystal structure": Ph.D.. Thesis, Stanford
University, California.
Kan, T.K, and Sicking, C.J., 1994, Pre-drill geophysical methods for geopressure detection and evaluation: in Studies in abnormal pressures, ed. Fertl, W.H., Chapman, R.E., and Hotz, R.F., Elsevier.
Kennett, B. L. N., 1983, Seismic Wave Propagation in Stratified Media, Cambridge University Press, 342 pp. Kjartansson, E., 1979, ""Constant Q-wave propagation and attenuation": J. Geophys. Res., 84, 4737-4748. Knight, R., and Nolen-Hoeksema, R., 1990, "A laboratory study of the dependence of elastic wave velocities on pore
scale fluid distribution": Geophys. Res. Lett., 17, 1529-1532. Krief, M., Garat, J., Stellingwerff, J., and Ventre, J., 1990, A petrophysical interpretation using the velocities of P and S
waves (full-waveform sonic), The Log Analyst, November. Kuster, G.T., and Toksöz, N.M., 1974, "Velocity and attenuation of seismic waves in two-phase media, I. Theoretical
formulations": Geophys., 39, 607-618. Lo, T.W., Coyner, K.B., and Toksöz, M.N., 1986, "Experimental determination of elastic anistropy of Berea
sandstone, Chicopee shale, and Chelmsford granite": Geophysics, 51, 164-171. Lockner, D.A., Walsh, J.B., and Byerrlee, J.D., 1977, "Changes in seismic velocity and attenuation during deformation
of granite": J. Geophys. Res., 82, 5374-5378. Lucet, Nathalie, 1989, Ph.D. dissertation, Univ. of Paris/IFP. Marion, D., 1990, Ph.D. dissertation, Stanford Univ., California. Marion, D., Mukerji, T., and Mavko, G., 1994, Scale effects on velocity dispersion: from ray to effective medium
theories in stratified media, Geophysics, 59, 10, 1613-1619. Mavko, G., and Jizba, D., 1991, Estimating grain-scale fluid effects on velocity dispersion in rocks: Geophysics, 56,
1940-1949. Mavko, G., and Mukerji, T., 1995, Pore space compressibility and Gassmann’s relation, Geophysics, 60, 1743-1749. Mavko, G., Mukerji, T., and Dvorkin, J., 1998, The Rock Physics Handbook, Cambridge University Press, 329 pp. Mavko, G., Chan, C., and Mukerji, T., 1995, Fluid substitution: estimating changes in Vp without knowing Vs,
Geophysics, 60, 1750-1755. Milholland, P., Manghnani, M.H., Schlanger, S.O., and Sutton, G.H., 1980, Geoacoustic modeling of deep-sea
carbonate sediments, J. Acoust. Soc. Am., 68, 1351-1360. Mobarek, S.A.M., 1971, "The effect of temperature on wave velocities in porous rocks": Master's Thesis, Univ. of
California, Berkeley. Müller, G., Roth, M., and Korn, M., 1992, "Seismic wave travel-times in random media": Geophys. J. Int., 110, 29-
41. Murphy, W.F., 1982, Effects of microstructure and pore fluids on the acoustic properties of granular sedimentary
materials: Ph.D. dissertation, Stanford University. Nobes, D.C., Villinger, H., Davis, E.E., and Law, L.K., 1986, "Estimation of marine sediments bulk physical properties
at depth from sea-floor geophysical measurements": J. Geophys. Res., 51, 2193-2199. Nur, A., 1980, "Seismic velocities in low porosity rocks": in: Source mechanism and earthquake prediction, in:
ˆHommage au Prof. J. Coulomb, Ed. CNRS, Paris. Nur, A., and Murphy, W., 1981, "Wave velocities and attenuation in porous media with fluids": Proc. of 4th Int.
Conf. on Continuum models of discrete systems, Stockholm, 311-327. Nur, A., and Simmons, G., 1969, "The effect of viscosity of a fluid phase on velocity in low porosity rocks": Earth and
Planet Sc. Lett., 7, 99-108.
Nur, A.N., and Wang, Z., eds., 1988, Seismic and acoustic velociteis in reservoir rocks, Vol. 1, Experimental Studies, SEG Geophysics Reprint Series No. 10.
O'Connell, R.J., and Budiansky, B., 1974, "Seismic velocities in dry and saturated cracked solids": J. Geophys. Res., 79, 5412-5426.
Palaz, I., and Marfurt, K.J., eds., 1997, Carbonate Seismology, SEG. Pickett, G.R., 1963, Acoustic character logs and their applications in formation evaluation, J. Petrol. Technol., 15,
650-667. Raymer, L.L., Hunt, E.R., and Gardner, J.S., 1980, An improved sonic transit time-to-porosity transform, SPWLA,
21st Ann. Logg. Symp., July 1980. Rio, P., Mukerji, T., Mavko, G., and Marion, D., 1995, Velocity dispersion and upscaling in a laboratory-simulated
VSP", Geophysics, 61, 584-593. Schlumberger, 1989, Log interpretation manual. Shuey, R.T., 1985, "A simplification of the Zoeppritz equations": Geophysics, 50, 609-614. Smith, G.C., and Gidlow, P.M., 1987, "Weighted stacking for rock property estimation and detection of gas":
Geophys. Prosp., 35, 993-1014. Stoll, R.D., 1977, "Acoustic waves in ocean sediments": Geophysics, 42, 715-725. Stoll, R.D., 1979, "Experimental studies of attenuation in sediments": J. Acoust. Soc. Am., 66, 1152-1160. Stoll, 1980. Timur, A., 1977, "Temperature dependence of compressional and shear wave velocities in rock": Geophys., 42, 950-
956. Tosaya, C., 1982, "Acoustical properties of clay-bearing rocks": Ph.D.. Thesis, Stanford University, California. Tosaya, C., Nur, A., Vo-Thanh, D., and Da Prat, G., 1985, "Laboratory seismic methods for remote monitoring of
thermal EOR": 3éme Coll. Europ. sur l'amélioration de la récupération du pétrole, Rome. Vernick, L., and Nur, A., 1990, Ultrasonic velocity and anisotropy of some petroleum source rocks: The Bakken
formation, Williston basin, N. Dakota: Stanford Rock and Borehole Project. Vernick, L.I. et al., 1987, "Acoustic properties of rocks": in The Superdeep Well of the Kola Peninsula, Ye. A.
Kozlovsky, Ed., Springer Verlag, Berlin, 339-350. Wang, 1988, Ph.D. dissertation, Stanford University.
Wang, Z. and Nur, A., 1988, Effect of temperature on wave velocities in sands and sandstones with heavy
hydrocarbons: SPE Reservoir Engineering, 3, 158-164.
Wang, Z. and Nur, A., 1988, Velocity dispersion and the “local flow” mechanism in rocks: 58th Ann. Internat. Mtg.,
Soc. Expl. Geophys., Expanded Abstracts, 928-930. Wang, Z. and Nur, A., 1990, Dispersion analysis of acoustic velocities in rocks: J. Acoust. Soc. Am., 87, 2384-2395. Wang, Z., and Nur., A.M., eds., 1992, Seismic and Acoustic Velocities in Reservoir Rocks, Vol. 2, Theoretical and
Model Studies, SEG Geophysics Reprint Series, No. 10. Watt, Davies, and O'Connell, 1976, Reviews of Geophysics and Space Physics, 14, 541-563. Williams, D.M., 1990, The acoustic log hydrocarbon indicator, Soc. of Prof. Well Log Analysts, 31st Ann. Logging
Symposium, Paper W.
Winkler, K., Nur, A., and Gladwin, M., 1979, "Friction and seismic attenuation in rocks": Nature, 277, 528-531. Winkler, K. W., 1985, Dispersion analysis of velocity and attenuation in Berea sandstone: J. Geophys. Res., 90, 6793-
6800. Winkler, 1993, personal communication. Wyllie, M.R.J., Gregory, A.R., and Gardner, G.H.F., 1958, "An experimental investigation of factors affecting elastic
wave velocities in porous media": Geophys. ,23, 459-493. Xu, S., and White, R. E., 1994, A physical model for shear-wave velocity prediction, Expanded Abstracts, 56th
EAEG Mtg. and Tech. Exhib., Vienna, 117. Xu, S., and White, R. E., 1995, A new velocity model for clay-sand mixtures, Geophys. Prosp., 43, 91-118. Yamamoto, T., Nye, T., and Kuru, M., 1994, Porosity, permeability, shear strength: crosswell tomography below an
iron foundry, Geophysics, 59, 1530-1541. Yamamoto, T., Nye, T., and Kuru, M., 1995, Imaging the permeability structure of a limestone aquifer by crosswell
acoustic tomography, Geophysics, 60, 1634-1645. Yin, H., Han, D.H., and Nur, A., 1988, Study of velocity and compaction on sand-clay mixtures, Stanford Rock and
Borehole Project, Vol. 33.