Click

Here

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115, B06314, doi:10.1029/2009JB007134, 2010

for

Full
Article

Lam parameters of common rocks in the Earths crust

and upper mantle
Shaocheng Ji,1 Shengsi Sun,1 Qian Wang,2 and Denis Marcotte1
Received 13 November 2009; revised 15 January 2010; accepted 28 January 2010; published 24 June 2010.

[1] Lam parameter (l) and shear modulus (m) are the most important, intrinsic, elastic
properties of rocks. The Lam parameter l, which relates stresses and strains in
perpendicular directions, is closely related to the incompressibility and contains a high
proportion of information about the resistance to a change in volume caused by a change in
pressure. Recent studies have emphasized the roles played by l in the discrimination of
gas sands from carbonates and shale in sedimentary basins and in the seismic reflection of
crustal fault zones. Here we analyze the equivalent isotropic elastic data of 475 natural
rocks in order to characterize l values for common types of crystalline rocks in the Earths
crust and upper mantle and their variations with pressure (P), temperature (T), and
mineralogical composition. When no partial melting, metamorphic reaction, dehydration,
or phase transformation occurs, l of a crystalline rock as a function of P and T can be
described by l = a + (dl/dP)P c exp( kP) (dl/dT)T, where a is the projected l value
at zero pressure if microcracks were fully closed; dl/dP is the pressure derivative in
the linear elastic regime; c is the initial l drop caused by the presence of microcracks at
zero pressure; k is a decay constant of the l drop in the nonlinear poroelastic regime; and
dl/dT is the temperature derivative. The parameter l increases nonlinearly and linearly
with increasing pressure at low (<300 MPa) and high (>300 MPa) pressures,
respectively. In the regime of high pressures, l decreases quasilinearly with increasing
temperature with dl/dT values in the range of 110 103 GPa/C. Approaching the ab
quartz transition temperature, quartzite displays negative l values. In the lr (density) and
ml plots, the main categories of lithology can be clearly distinguished. The ultramafic
rocks display systematic decreases in both m and l with increasing the degree of
serpentinization. Eclogites, mafic rocks (gabbro, diabase, mafic granulite, and mafic
gneiss), and felsic rocks (granite, diorite, felsic gneiss, intermediate gneiss, and
metasediments) are characterized by high, moderate, and low m and l values, respectively.
For pyroxene and olivine, both l and r increase, but m decreases with increasing the
Fe/Mg ratios. In the plagioclase series, both l and m increases with increasing the
anorthite content. Increases in the contents of garnets, sillimanite, rutile, zircon, ilmenite,
and spinel result systematically in an increase in rocks l and m values. The present
results provide improved constraints on the discrimination of composition for crustal and
upper mantle rocks in terms of l and m.
Citation: Ji, S., S. Sun, Q. Wang, and D. Marcotte (2010), Lam parameters of common rocks in the Earths crust and upper
mantle, J. Geophys. Res., 115, B06314, doi:10.1029/2009JB007134.

1. Introduction
[2] The elastic properties of an isotropic material or rock
can be described by two independent moduli termed lambda
(l) and mu (m), introduced and named after the 18th century
French mathematician and engineer G. Lam (17951870).
1
Dpartement des Gnies Civil, Gologique et des Mines, cole
Polytechnique de Montral, Montral, Qubec, Canada.
2
Key Laboratory of Marginal Sea Geology, Guangzhou Institute of
Geochemistry, Chinese Academy of Sciences, Guangzhou, China.

Copyright 2010 by the American Geophysical Union.

01480227/10/2009JB007134

In 1828, Lam formulated the modern version of Hookes

law relating stress (sij) to strain ("ij) in its general tensor
form, thereby creating the basis for the materials sciences
and rock mechanics:
ij ij D 2"ij

where the volume strain D = "11 + "22 + "33, dij is the

Kronecker delta, dij = 1 when i = j, and dij = 0 when i j;
l and m are the first and second Lam constants, respectively. The second Lam constant m is identical to the shear
modulus or rigidity, which is defined as the resistance to a
simple shear strain that produces a shape change without

B06314

1 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

changing total volume. Hereafter we will use Lam

parameter (l) in place of the first Lam constant as it varies
with pressure and temperature.
[3] For an isotropic medium, equation (1) can be written
in matrixvector form that is composed of 6 rows and
6 columns:
2

11

7 6
6
7 6
6
6 22 7 6
7 6
6
7 6
6
7 6
6
6 33 7 6
7 6
6
76
6
7 6
6
6 12 7 6
7 6
6
7 6
6
7 6
6
6 13 7 6
7 6
6
5 4
4
23

 2

 2

 2

2

2

32

"11

76
7
76
7
6 "22 7
0 7
76
7
76
7
76
7
6 "33 7
0 7
76
7
76
7
76
7
6 "12 7
0 7
76
7
76
7
76
7
6 "13 7
0 7
76
7
54
5
2
"23
2

The parameter l relates stresses and strains in perpendicular

directions [Jaeger, 1969]. The physical meaning of l can be
clearly illustrated in a special case of uniaxial strain where
"1 0, and "2 = "3 = 0 (i.e., no displacement occurs perpendicular to the x axis): l = s2/"1 = s3/"1. Interestingly and
most notably, only l and m appear in Hookes law, but not
Youngs modulus (E), the bulk modulus (K), or Poissons
ratio (u). This indicates that l and m are the most intrinsic
elastic coefficients to express stress in terms of strain.
[4] Goodway [2001] believes that l is closely related to
materials incompressibility (l = K 2m/3) and contains a
higher proportion of information about the resistance to a
change in volume caused by a change in pressure. The
parameter m corresponds to materials resistance to simple
shear. The parameter l is negative if K < 2m/3 as K > 0 and
m > 0.
[5] The most common geophysical parameters measurable are compressional (P) and shear (S) wave velocities
(Vp and Vs) and densities (r) of elastic media; l and m for
isotropic elasticity can be easily determined from the measured seismic data:


Vs2

from the inversion of P and S wave reflectivities [Estabrook

and Kind, 1996; Goodway et al., 1997; Goodway, 2001;
Gray, 2003]. ChvezPrez and Louie [1997, 1998] suggested that variations in l control seismic reflections of
crustal fault zones. Obviously, both l and m offer the most
fundamental and orthogonal parameterization of elastic
seismic waves to extract information about the composition
and structure of rocks in the Earths interior. In the literature,
however, only E, K, m, and u are usually reported although
l is an intrinsic and invariant property of elastic media
under given conditions. So far, little systematic research
work has been carried out on the characterization of l values
for crystalline rocks [Christensen, 1966; Kern, 1979]. In this
paper, we will investigate the pressure and temperature
dependences of l values for common categories of crystalline rocks from the Earths crust and upper mantle, and
explore the potential implications of l as a discriminant of
composition for common rock types in the Earths crust
and upper mantle. Our emphasis is placed on the properties
of crystalline igneous and metamorphic rocks. For the l
values of common sedimentary rocks (i.e., sandstone, shale,
limestone and dolomite) and their applications in the exploration of oil or gas reservoirs, the reader is referred to the
papers of Goodway et al. [1997], Goodway [2001], Gray
and Andersen [2000], Dufour et al. [2002], Gray [2003],
and Li et al. [2003].

2. Data for Analysis

[6] An isotropic rock can be characterized by only two
independent elastic constants: l and m. However, natural
rocks may be elastically anisotropic due to the presence of
compositional layering, lattice preferred orientation, shape
preferred orientation (i.e., foliation and lineation) of rock
forming minerals and alignment of microcracks. The average elastic property of such an anisotropic rock, which is
equivalent to the property of its isotropic counterpart, can be
computed from the mean Vp and mean Vs values according
to equations (3) and (4):


V p Vp X Vp Y Vp Z =3

 Vp2  2Vs2


B06314

2

= Vp =Vs 2

 Zp2  2Zs2

 Zs2

where Z is the acoustic impedance; P wave impedance Zpp =

rVp, and S wave impedance Zs = rVs; l = 0 if Vp =Vs 2,
p
p
and l < 0 if Vp =Vs < 2; l/m 1 if Vp =Vs  3 (e.g., serpentine, calcite, feldspar,
p hornblende,
p fayalite, dolomite),
and 0 l/m < 1 if 2  Vp =Vs < 3 (e.g., quartz, staurolite, bronzite, diallage, and enstatite). The information
about the impedance or lr and mr data can be extracted

V s Vs XY Vs XZ Vs YX
Vs YZ Vs ZX Vs ZY =6

In equation (8), Vp(X), Vp(Y) and Vp(Z) are the P wave

velocities along the propagation directions X, Y, and Z,
respectively, which lie parallel to the stretching lineation,
parallel to foliation and perpendicular to lineation, and
perpendicular to foliation, respectively. In equation (9),
Vs(XY), Vs(XZ), Vs(YX), Vs(YZ), Vs(ZX), and Vs(ZY) are the S
wave velocities with the first letter in the brackets to refer to
the propagation direction and the second letter to the
polarization direction. Previous theoretical analyses and
experimental studies [Christensen and Ramananantoandro,
1971; Ji et al., 2003; Wang and Ji, 2009] suggest that the
mean Vp and Vs values calculated from equations (8) and (9)
give values very close to true isotropic elastic properties
even in highly anisotropic rocks.
[7] The highquality data used for the analysis were taken
from a database of 475 samples on which P wave velocities

2 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

Figure 1. Lam parameter (l) as a function of pressure for selected lithologies from the Sulu ultrahigh
pressure metamorphic terrain (samples 1511416, 147211, SuluYK9, XG3, MB25, 125818, Sulu
YK22, TF3, and B2184R88P4s) and the Yunkai Mountains (samples YK52A, YK84, YK87, YK92,
YK102, and YK109), China. Dots indicate experimental data; line indicates the best fitting according
to equation (10) using a least squares method. Grt, garnet; Qtz, quartz.
have been well measured for the X, Y, and Z directions and
S wave velocities for six pairs of propagationvibration
directions at high hydrostatic pressures ( 500 MPa) using
the same laboratory equipment and the same methods (i.e.,
the pulse transmission technique [Birch, 1960, Christensen,
1974; Kern, 1982]). The uncertainty of measurements is
estimated to be < 0.5% for Vp, < 1% for Vs, and 0.005 g/cm3
for density. For each sample, l and m as a function of
pressure or temperature were calculated. In this database,
87 samples were measured by our group: 57 from the Dabie
Sulu ultrahigh pressure metamorphic terrane and 20 from
the Yunkai Mountains of China [Wang et al., 2005a, 2005b;
Ji et al., 2007; Wang and Ji, 2009], and 10 from the Tantato
high grade metamorphic domain in northern Saskatchewan,
Canada [Ji and Salisbury, 1993; Ji et al., 1993]. The literature
data were taken from Ji et al. [2002, and references therein].
Only the velocities measured during depressurization were
selected for the analysis because the values correspond to the

stable microstructural state [Burke and Fountain, 1990; Ji et

al., 1993; Wang et al., 2005a, 2005b; Ji et al., 2007].

3. Results
3.1. Pressure Dependence
[8] In using l to place constraints on lithology in the crust
and upper mantle, it is necessary to understand how l varies
with pressure and temperature. Figure 1 shows typical lP
data for common lithologies such as ultramafic rocks
(peridotite, serpentinite, and partially serpentinized peridotites, Figure 1a), mafic rocks (eclogite, mafic gneiss, mafic
granulite, and amphibolite, Figure 1b), felsic rocks (granite,
diorite, felsic gneiss and metasediments, Figure 1c), marble
and sedimentary rocks (limestone and sandstone, Figure 1d)
up to 1.0 GPa, a pressure equivalent to depths of approximately 35 km. Those rocks display systematically an initial
nonlinear increase in l at low pressures followed by a more

3 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

gradual linear increase at high pressure. This phenomenon

indicates that l is strongly affected by the presence of
microcracks within the rock. The rapid, nonlinear rise in l
below a critical pressure can be attributed to the progressive
closure of microcracks with varying aspect ratio spectra with
increasing pressure. The linear increase in l above the critical
pressure where all microcracks have been fully closed in the
rock marks an elastic volume contraction of crackfree
material under hydrostatic compression.
[9] The parameter l as a function of confining pressure
can be described by
 a d=dPP  c expkP

10

where a is the projected l value of a nonporous or crack

free compacted rock at zero pressure, which is determined
from extrapolation of the linear lP relationship obtained at
high pressures to zero pressure; dl/dP is the pressure
derivative in the linear elastic regime; c, which is the
ambient l drop caused by the presence of pores/microcracks
at zero pressure, determines the maximum magnitude of
the l increases due to the closure of pores and microcracks;
k, which is a decay constant of the l drop, controls the shape
of the nonlinear segment of the lP curve. The k value is a
parameter to characterize the facility of the successive closure of cracks of varying aspect ratio spectra with increasing
pressure. Flatter cracks yield a larger k value than more
spherical pores. The zeropressure l value of the rock
containing microcracks is the same as (ac). In equation (10),
a and dl/dP are two parameters which describe the l value
of the microcrack or porefree solid matrix, while c and k
are two parameters related to the porosity and geometrical
shape of pores (e.g., aspect ratio, spatial arrangement, orientation and size distribution), and in turn on the formation
and deformation processes of the rocks. The same type of
mathematical expression as equation (10) has been used by
previous workers for describing the pressure dependence of
seismic velocities [Stierman et al., 1979; Zimmerman et al.,
1986; EberhartPhillips et al., 1989; Shapiro, 2003; Wang
et al., 2005a, 2005b; Ji et al., 2009; Wang and Ji, 2009].
The equation can be derived based on an assumption that the
difference between the property (e.g., seismic velocities or
elastic moduli) of a nonporous material and its porous
counterpart at a given confining pressure has a maximum
value (c) at P = 0 and then decays progressively with
increasing P at a rate proportional to the value of the
property at the applied confining pressure [Ji et al., 2007].
The last term of equation (10) possesses the same form of
the expression that is commonly used to describe natural
phenomena such as radioactivity decay, cooling, and
vibration attenuation.
[10] We used equation (10) to curve fit lP data for 57
samples collected from the DabieSulu ultrahigh pressure
(UHP) metamorphic belt using a least squares method and
gave excellent results with wellconstrained a, dl/dP, c and
k parameters. Table 1 lists such parameters, together with
the data of density and modal composition, for only 27 representative UHP samples: 17 from the boreholes of the
Chinese Continental Scientific Drilling project [Ji and Xu,
2009] and 10 samples from the surface quarries. The information about the locality and recovery depth for these
samples were presented in Wang et al. [2005a, 2005b], Ji et

B06314

al. [2007], and Wang and Ji [2009]. Based on the values

reported in Table 1, extrapolation and interpolation can be
easily performed using equation (10).
3.2. Effect of Temperature
[11] Laboratory data on the effect of temperature on
seismic wave velocities and density of rocks are rather
limited and were mainly from H. Kerns laboratory. Figure 2
presents the l values as a function of temperature at a
confining pressure of 600 MPa for some common rocks
from the upper mantle and the continental crust. The l values
were calculated using the mean Vp and Vs data reported by
Barruol [1993], Kern et al. [1999], Kern et al. [2002], and
Kern and Tubia [1993]. When no partial melting [Spetzler
and Anderson, 1968], metamorphic reaction, dehydration or
phase transformation [Kern, 1982] occurs, most solid rocks
such as garnet peridotite, harzburgite, and pyroxenite
(Figure 2a), eclogites (Figure 2b), amphibolites (Figure 2c),
and gabbro, granite, and granitic gneiss (Figure 2d) display
that l decreases linearly with increasing temperature due to
thermal effects (e.g., thermal dilatation of mineral lattices,
microcracking and grain boundary widening induced by
differential thermal expansion). The gradual decrease in l
occurs over the range from room temperature to 600C,
which can be easily understood from equation (4) because
dVp/dT is significantly larger than dVs/dT for most rocks
[Kern and Tubia, 1993; Kern et al., 1996; Kern et al., 1999;
Ji et al., 2002; Kern et al., 2002]. For basalt, however, the
temperature dependence is negligible in the range from
room temperature up to 600C (Figure 2d and Table 2)
probably because the elastic velocities in basaltic glass at a
hydrostatic pressure of 600 MPa increase with temperature
due to reduction of porosity [Kern, 1982].
[12] Thus, the lT relationship for crystalline rocks can be
described by a linear equation:
 0  d=dT T

11

where l0 is the projected l value when T = 0C, and

dl/dT is the temperature derivative. The values l0 and
dl/dT as well as the modal composition for each sample
illustrated in Figure 2 are given in Table 2. For these
samples, dl/dT varies generally from 3 103 to 10
103 GPa/C.
[13] Thus, the laboratory experimental results can be
extrapolated into the Earths interior using the following
equation if there is no partial melting, metamorphic reaction,
dehydration or phase `transformation:
P; T a d=dPP  c expkP  d=dT T

12

The results computed from equation (12) have the potential

of providing valuable constraints on the interpretation of the
lithological and mineralogical composition of the Earths
crust and upper mantle.
3.3. Effect of ab Quartz Transition
[14] A quartzite is taken as an example to illustrate the
effect of phase transition on l. Figure 3a presents the
mean Vp and Vs data as a function of temperature at a confining pressure of 600 MPa for an ultramylonitic quartzite
(89SB114a) consisting of 100% quartz. This sample was
collected from a ductile shear zone in the Saint Barthelemy

4 of 15

5 of 15

Amphibolite
Amphibolite
Bt dioritic gneiss
BtHblPlKfs paragneiss
BtMusPlKfs orthogneiss
Coarsegrained eclogite
Coarsegrained eclogite
Coarsegrained eclogite
Dunite
Felsic orthogneiss
Felsic orthogneiss
Finegrained eclogite
Granitic gneiss
Granitic Gneiss
GrtHarzburgite
GrtHarzburgite
HblBtPlKfs paragneiss
HblMag felsic orthogneiss
Lherzolite
Dolomitic marble
Metagabbro
Metagabbro
Phl dunite
Phn eclogite
Qtz eclogite
Serpentinite
Serpentinite

203515
B1651R37P41c
17866
B2184R88P4s
B1578R14P18t
MB22
MB25
MB26
147211
B2078R63P9r
B1694R49P7s
SB1
TF3
261017
1601211
1511416
B2242R100P16a
B1628R33P24
150320
YM1
SuluYK3B
SuluYK22
1664243
315111
125818
XG3
SuluYK9

3.07
3.00
2.74
2.66
2.65
3.50
3.59
3.56
3.16
2.63
2.63
3.37
2.66
2.64
3.17
3.21
2.66
2.65
3.12
2.86
3.00
2.96
3.25
3.43
3.38
2.66
2.60

Density
(g/cm3)
Pl 35.0, Amp 41.0, Bt 15.0, Qtz 8.0, Chl 1.0
Amp 48.0, Chl 13.0, Grt 3.0, Cpx 15, Pl 12.0, Qtz 5.0, Ep 2.0, Opq 2.0
Pl 49.0, Qtz 37.0, Bt 12.0, Opq 2.0
Pl. 40.0, Kfs 15.0, Qtz 40.0, Bt 4.0, Opq 1.0
Pl 40.0, Kfs 25.0, Qtz 25.0, Mus 4.0, Bt 3.0, Grt 1.0, Opq 2.0
Grt 63.0, Cpx 35.0, Rt 1.5, Qtz 0.5
Grt 65.0, Cpx 33.0, Rt 1.5, Qtz 0.5
Grt 70.0, Cpx 29.0, Rt 0.8, Qtz 0.2
Ol 88.0, Srp11.0, Opq 1.0
Pl 31.0, Kfs 35.0, Qtz 25.0, Bt 4.0, Amp 4.0, Opq 1.0
Pl 30.0, Kfs 42.0, Qtz 21.0, Cpx 4.0, Amp1.5, Opq 1.0, Rt 0.5
Grt 33.0, Cpx 60.0, Amp 2.0, Rt 2.0, Qtz 1.0, Symp 2.0
Qtz 20.0, Pl 30.0, Kfs 42.0, Opx 4.0, Opq 1.5, Rt 0.5
Pl 32.0, Kfs 31.0, Qtz 33.0, Bt 3.0, Opq 1.0
Ol 67.0, Opx 10.0, Cpx 13.0, Grt 4.0, Srp 4.0, Phl 2.0
Ol 68.0, Opx 14.0, Cpx 8.0, Grt 5.0, Srp 4.0, Phl 1.0
Pl 38, Kfs 18.0, Qtz 35.0, Amp 7.5, Opq 1.5
Pl 38.0, Kfs 15.0, Qtz 35, Amp 7.0, Grt 1.5, Opq 2.0
Ol 70.0, Opx 8.0, Cpx 11.0, Srp 10.0, Opq 1.0
Dol 87.0, Cal 7.0, Qtz 4.0, Grt 1.0, Cpx 1.0
Pl 72.0, Cpx 4.0, Grt 8.0, Qtz 4.0, Mus 9.0, Chl 3.0
Pl 45.0, Cpx 14.0, Grt 9.0, Qtz 10.0, Mus 5.0, Chl 11.0, Zo 5.0, Opq. 1.0
Ol 86.0, Opx 2.0, Phl 4.0, Srp 5.0, Opq 3.0
Grt 36.0, Cpx 40.0, Amp 5.0, Phn 17.0, Qtz 1.0, Rt 1.0
Grt 40.0, Cpx 35.0, Amp 4.0, Symp 12.0, Qtz 7.0, Phn 1.5, Rt 0.5
Srp 80.0, Ol 15.0, Opq 5.0
Srp 85.0, Tlc 10.0, Opq 5.0

Modal Composition (vol %)

32.22
49.44
34.27
29.03
29.71
74.79
86.67
90.46
71.89
31.38
39.31
81.05
30.87
31.82
67.09
83.41
39.52
36.89
64.92
41.15
50.13
51.07
63.20
80.26
76.63
40.03
41.99

a (GPa)

3.32
3.81
4.32
0.80
3.04
5.25
8.84
7.75
5.11
1.41
0.00
1.05
3.27
5.28
5.83
5.31
2.81
2.99
1.28
1.62
3.61
7.18
4.72
0.00
9.66
3.66
6.73

dl/dP

22.69
6.98
18.93
26.57
35.45
12.91
50.94
20.79
24.07
65.20
24.42
20.08
15.20
12.87
74.65
29.64
12.36
12.11
41.98
21.35
16.21
16.96
24.35
45.19
75.09
14.48
7.67

c (GPa)

Coefficients of Lam Parameterb

1

13.07
14.51
11.19
9.93
20.63
10.65
14.54
8.36
19.41
36.74
3.81
6.15
16.12
21.33
28.92
8.88
28.81
12.10
10.17
24.74
21.99
15.40
16.00
12.70
24.63
23.07
21.43

k (GPa

a
Abbreviations are Amp, amphibole; Bt, biotite; Cal, calcite; Chl, chlorite; Cpx, clinopyroxene; Dol, dolomite; Ep, epidote; Grt, garnet; Hbl, hornblende; Kfs, Kfeldspar; Mag, magnetite; Mus, muscovite;
Ol, olivine; Opq, opaque; Opx, orthopyroxene; Phl, phlogopite; Phn, phengite; Pl, plagioclase; Qtz, quartz; Rt, rutile; Srp, serpentine; Symp, symplectite; Tlc, talc; Zo, zoisite.
b
The goodness of fit coefficients R2 > 0.98 for all the samples.

Lithology

Sample

Table 1. Densities, Modal Compositions, and Four Coefficients to Describe the Lam Parameter l for 27 Typical Samples From DabieSulu UHP Metamorphic Belt, China

B06314
JI ET AL.: LAM PARAMETERS OF COMMON ROCKS
B06314

B06314

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

Figure 2. The effect of temperature on Lam parameter for (a) harzburgite, garnet (Grt) peridotite and
pyroxenite, (b) eclogites, (c) amphibolites, and (d) basalt, granite, granitic gneiss and gabbro at a hydrostatic pressure of 600 MPa. The original wave velocities and density data are from Barruol [1993], Kern
et al. [1999], Kern et al. [2002], and Kern and Tubia [1993].

Table 2. Densities, Modal Compositions, and Parameters l0 and dl/dT Describing the Effect of Temperature for 13 Samples Shown in
Figure 2
Sample

Lithology

Density
(g/cm3)

90VS12
90VS33
QS07
MB9802
MB9803
MB9819
91MA203
D9513
LS9809
WH9813
SL9812
Tb3573
90VS53b

Amphibolite
Amphibolite
Basalt
Eclogite
Eclogite
Eclogite
Gabbro
Granite
Granitic gneiss
Grt amphibolite
Grt peridotite
Harzburgite
Pyroxenite

3.14
3.07
2.74
3.57
3.45
3.61
2.90
2.65
2.64
3.24
3.39
3.30
3.24

Modal Compositiona

dl/dT
l0
(GPa) (GPa/C)

R2

Source of Data

Hbl 59.0, Pl 35.0, Qtz 4.0, Opq 1.0

57.92 0.0042 0.98 Barruol [1993]
Hbl 54.2, Pl 35.8, Pyx 8.0, Opq 2.0
52.32 0.0032 0.94 Barruol [1993]
Matrix 70.0, Pyx 19.0, Pl phenocryst 6.0, Idd 5.0
41.65 0.000005 0.81 Kern et al. [2002]
Grt 57.0, Omp 42.0, Rt 0.6, Mus 0.2
90.81 0.0075 0.99 Kern et al. [2002]
Grt 46.0, Omp 53.0, Mus 0.6
86.36 0.0062 0.99 Kern et al. [2002]
Grt 80.0, Omp 15.0, Phl 3.5, Qtz 1.0, Opq 0.5
94.05 0.0072 0.99 Kern et al. [2002]
Pl 75.0, Opx 22.0, Hbl 2.0, Bt 1.0
58.23 0.0042 0.98 Barruol [1993]
Pl 54.0, Kfs 23.0, Qtz 20.0, Bt 3.0
30.57 0.0071 0.99 Kern et al. [1999]
Kfs 53.0, Qtz 29.0, Mus 6.0, Pl 8.0, Ep 4.0, Sp 0.2, Bt 0.1 33.33 0.0112 1.00 Kern et al. [2002]
Hbl 65.0, Qtz 18.0, Grt 15.0, Mag 2.0, Pl 0.4
47.39 0.0054 0.99 Kern et al. [2002]
Ol 73.0, Mag 14.0, Grt 12.0, Pyx 1.0
73.5
0.0065 1.00 Kern et al. [2002]
Ol 74.0, Opx 21.5, Cpx 2.5, Spl 1.6, Srp 0.4
72.42 0.0098 1.00 Kern and Tubia [1993]
Pyx 68.5, Pl 25.4, Spl 4.5, Opq 1.6
64.97 0.0089 0.96 Barruol [1993]

a
Abbreviations are Bt, biotite; Cpx, clinopyroxene; Ep, epidote; Grt, garnet; Hbl, hornblende; Idd, iddingsite; Kfs, Kfeldspar; Mag, magnetite; Mus,
muscovite; Ol, olivine; Omp, omphacite; Opq, opaque; Opx, orthopyroxene; Phl, phlogopite; Pl, plagioclase; Pyx, proxene; Qtz, quartz; Rt, rutile; Srp,
serpentine; Sp, sphalerite; Spl, spinel.

6 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

the lower crusts, a significant decrease in l or u may mark

this boundary.

Figure 3. (a) Mean P and S wave velocities and (b) Lam

parameter as a function of temperature for a quartzite at a
confining pressure of 600 MPa. The wave velocity data
are from Barruol [1993]. The ab quartz transition occurs
at 650C.
Massif (northern Pyrenees, France), and its P and S wave
velocities and densities were measured in three propagation
directions (X, Y, Z) and six propagationpolarization directions (XY, XZ, YX, YZ, ZX and ZY), respectively
[Barruol, 1993; Barruol et al., 1992]. The P wave velocity
falls pronouncedly when approaching the ab quartz transition temperature (650C at 600 MPa), and jumps
abruptly after the transition while little change occurs in the
S wave velocity across the transition. A similar phenomenon
was observed in other quasiisotropic quartzites [Kern,
1979]. The l values (Figure 3b) computed from the mean
Vp, Vs, and density values for the quartzite become negative
in the temperature range from 450C to 650C. An abrupt
increase in l occurs above 650C at 600 MPa, reflecting a
pronounced increase in Vp/Vs ratio in the beta quartz. Previous experiments [e.g., Shen et al., 1993] showed that the
ab quartz transition temperature increases linearly with
increasing the confining pressure. In the quartzrich continental crust with an anomalously high geothermal gradient
(dT/dz > 25 C), one may expect to detect low, very low and
high l values for the upper crust (<450C), middle crust
(450650C) and lower crust (>650C), respectively. In
other words, as the ab quartz transition is expected to
occur in the quartzbearing rocks (i.e., granite, diorite, and
felsic gneiss) across the boundary between the middle and

3.4. The lr Correlation

[15] Figure 4 shows fields of different types of rocks
within the lr plots at a hydrostatic confining pressure
of 600 MPa under which the effect of microcracks is
eliminated. The field boundary contours were drawn by eye
according to the distribution of experimental data. For
peridotites, both l and r values decrease progressively with
increasing the content of serpentine (chrysotile, antigorite,
and lizardite [Watanabe et al., 2007]) (Figure 4a). This trend
is consistent with the preliminary results of Christensen
[1966] for ultramafic rocks. The field of peridotites is relatively large due to varying contents of garnet, spinel and
magnetite, which have high values in both l and r. For
mafic rocks, eclogite displays significantly higher values of
both l and r than gabbro, diabase, mafic gneiss, mafic
granulite and amphibolite. The large variations in both l and
r for eclogites are attributable to varying volume fractions
of garnet, rutile and magnetite. Acid and intermediate rocks
(i.e., granite, diorite, felsic gneiss, intermediate gneiss and
metasediments) have lower l and r values than the mafic
crystalline rocks (Figure 4b). The fields for limestone/marble,
anorthosite, and basalt are also identified in Figure 4c. As
shown in Figure 5a, l illustrates a clear trend of linear
increase with density. The best fitting of the data for
401 samples gives: l = 57.814 r 118.51 (R2 = 0.79), l is in
GPa, and r is in g/cm3. In the statistical analysis, porous
(basalt and sandstone) and hydrated (serpentinite) rocks
are not included. There is considerable scatter is the distribution of the lr data for peridotites and eclogites due to
mineralogical complexities (mainly the content of garnet).
This will be analyzed in more detail in the Discussion.
3.5. The lVp Correlation
[16] Figure 6 provides lVp plots for different categories
of rocks at a confining pressure of 600 MPa. The trend of l
variation with Vp for peridotites with various degrees of
serpentinization reflects essentially progressive decreases in
both l and Vp with increasing the content of serpentine
(Figure 6a). It is noted that the presence of garnet, spinel or
chromite may cause a significant increase in l for peridotites. Eclogite displays higher values of both l and Vp than
the other mafic rocks such as gabbro, diabase, mafic gneiss,
mafic granulite and amphibolite. Acid and intermediate
rocks such as granite, diorite, felsic gneiss, intermediate
gneiss and metasediments have lower l and Vp values
(Figure 6b). Limestones/marbles show intermediate l and Vp
values between anorthosite and sandstone (Figure 6c). The
best fitting of the data for 401 samples (except of basalt,
sandstone and serpentinite) gives: l = 25.775Vp 125.84
(R2 = 0.90), where l is in GPa, and Vp is in km/s (Figure 5b).
3.6. The lVs Correlation
[17] Variations of l values as a function of Vs are shown
in Figure 7 for different groups of rocks at a hydrostatic
confining pressure of 600 MPa. Fresh peridotites can be
clearly distinguished from serpentinite and partially serpentinized peridotites in the l Vs diagram (Figure 7a).
Eclogite can be easily separated from the other mafic rocks
such as gabbro, diabase, mafic gneiss and mafic granulite

7 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

according to their lVs plots. Felsic rocks including granite,

diorite, felsic gneiss, intermediate gneiss and metasediments
are characterized by relatively lower l values (Figure 7b).
Important differences between Figure 6b and Figure 7b
originate from the high Vs value of quartz, which affects
the relative positions of quartzrich rocks (e.g., granite,
granodiorite, diorite, and felsic gneiss). Porous volcanic
basalts show significantly lower values in both l and Vs than
calciterich limestonemarble and plagioclaserich anorthosite (Figure 7c). The best fitting of the data from 401 samples
(excluding basalt, sandstone, and serpentinite) yields: l =
40.122Vs 104.34 (R2 = 0.62), where l is in GPa, and Vs in
km/s (Figure 5c). Obviously, the correlation of l with Vp
(Figure 5b) is remarkably better than that with Vs (Figure 5c).
3.7. The ml Correlation
[18] It is interesting to plot m as a function of l because
these two invariants form the basic elements within all the
elastic properties. If different types of rocks tend to separate
from each other in the ml diagram, these two parameters
may be used as a discriminant of lithological compositions.
As shown in Figure 8, fresh peridotites, serpentinites and
partially serpentinized peridotites can be distinguished each
other in the ml plot. The ultramafic rocks display systematic decreases in both m and l with increasing the degree
of serpentinization. The parameter l is always larger than
m with the mean m/l ratios equal to 0.810, 0.641 and
0.460 for the peridotite (38 samples), serpentinized peridotite
(15 samples) and serpentinite (12 samples), respectively.
The ratios correspond to the mean Poissons ratios of 0.275,
0.305 and 0.343 for the peridotite, serpentinized peridotite
and serpentinite, respectively. In Figure 8a, amphibolites
(31 samples) deviate clearly from the trend of the ultramafic rocks. According to the ml plot, eclogite can be
separated from the other mafic rocks (e.g., gabbro, diabase, amphibolite, mafic gneiss and mafic granulite) as
well as felsic rocks (e.g., granite, diorite, felsic gneiss,
intermediate gneiss, and metasediments). The mean m/l
ratios are 0.818, 0.915 and 1.057, corresponding to the
mean Poissons ratios of 0.275, 0.261 and 0.243, for the
mafic rocks (118 samples), eclogites (54 samples), and
felsic rocks (145 samples), respectively. Eclogite exhibits
remarkably higher values of both m and l than the other
common mafic rocks such as gabbro, diabase, mafic gneiss,
mafic granulite and amphibolite. Data of acid and intermediate rocks (e.g., granite, diorite, felsic gneiss, intermediate
gneiss and metasediments) cluster in a small area with
lower m and l values. Limestone/marble (29 samples), basalt
(21 samples), and anorthosite (8 samples) can also be separated from each other in the ml plots (Figure 8c).
[19] For reservoir sedimentary rocks that are generally
porous, Goodway et al. [1997] and Gray and Andersen
[2000] used a cross plot of mr versus lr in place of a
Figure 4. The lr plots for (a) 31 amphibolites, 38 peridotites, 12 serpentinites, and 15 partially serpentinized peridotites, (b) 54 eclogites, 118 mafic rocks (gabbro, diabase,
mafic gneiss and mafic granulite) and 145 felsic rocks (granite, diorite, felsic gneiss, intermediate gneiss, and metasediments), and (c) 8 anorthosites, 21 basalts, and 29 limestones/
marbles at a hydrostatic pressure of 600 MPa.
8 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

Figure 5. Variations in Lam parameter (l) with (a) density, (b) Vp, and (c) Vs for various types of rocks
at a hydrostatic pressure of 600 MPa for 401 samples. Serpentinite and porous rocks such as basalt and
limestone are not included.
ml diagram. According to these authors, different types of
lithology tend to separate along orthogonal boundaries in
such a mr lr cross plot: (1) coals have the lowest values
of both lr and mr. (2) Carbonates have high values of both
lr and mr. (3) Gasfilled sandstones have low lr values but
high mr values. A combination of lr and mr values has
been used for quick determination of reservoir lithology and
gas content [e.g., Li et al., 2003].

4. Discussion
[20] In this section, we analyze the relationship between l
and mineral composition of the crystalline rocks (Figures 48).
Figures 9a and 9b, plotting l versus r and m versus l,
respectively, provide some interesting perspectives on l
characteristics of pure monomineralic aggregates of common
rockforming minerals. The l value of a polymineralic
composite rock depends presumably on the volume fractions
and the l values of its constituent minerals. The contribution
of each constituent mineral to the bulk l value of a composite
rock can be analyzed for the moment only in a qualitative

manner because a relevant mixture rule for l is not available

[Lubarda, 1998; Nadeau, 1999; Berryman, 1995; Ji and Xia,
2002; Ji, 2004; Mavko et al., 2009]. The utilization of the lr
and ml plots for lithological discrimination is also discussed.
[21] Quartz is characterized by extremely low values in
both l (8.1 GPa) and r (2.65 g/cm3) but moderate value in
m (44.4 GPa) [Bass, 1995]. Hence, it lies obviously below
and above the general trends in the lr (Figure 9a) and ml
(Figure 9b) diagrams, respectively. Moreover, quartzrich
sandstones should have significantly lower l or lr values
than calciterich or dolomitedominant carbonates. This
feature can be used for quick determination of reservoir
lithology and in turn is economically important for oil or gas
field exploration in sedimentary basins [Goodway et al.,
1997; Goodway, 2001; Gray and Andersen, 2000; Gray,
2003; Dufour et al., 2002]. In the seismological investigation of carbonate reservoir in the Western Canadian sedimentary basin [Goodway et al., 1997; Li et al., 2003], for
example, lr and m/l were used as a good indicator for
distinguishing tight limestone (lr = 90.0 GPa g/cm3 and
m/l = 0.81), wet dolomite (lr =75.0 GPa g/cm3, m/l =

9 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

1.24) and gas dolomite (lr =37.5 GPa g/cm3, m/l =

2.08). The gas effect is apparent on reducing the lr
and m/l values because the presence of gas in porous,
grainsupported rocks causes a significant decrease in its
incompressibility (K) but does not affect its rigidity (m).
As a result, l is significantly reduced by the presence of gas.
In petroleum industry, therefore, layers of low l or lr
values derived from the inversion of amplitude versus
offset (AVO) stacks often suggest the presence of gas in a
sandstone or dolomite reservoir [Goodway and Tessman,
2000; Gray and Andersen, 2000]. In addition, coals have
low rigidity and thus low shear modulus, and have low
incompressibility and hence low l values.
[22] Feldspars, which are the most abundant minerals in
the Earths crust, form a ternary system of three end
members: KAlSi3O8 (orthoclase), NaAlSi3O8 (albite) and
CaAl2Si2O8 (anorthite). For the plagioclase series, l, m and
r all increase nearly linearly with increasing the amount of
anorthite component (Figure 9). There is a nearly linear
relationship between l and r or between m and l for plagioclase feldspars. The alkali feldspars have lower l and
m values than plagioclase feldspars. Nepheline (Ne) and
serpentine (Srp) lie at almost the same position as orthoclase
(Or) in the lr diagram, although these three minerals are
separated in the ml diagram with mNe > mOr > mSrp. In either
the lr or ml diagram (Figure 9), both alkali and plagioclase feldspars can be clearly separated from quartz. Thus,
the lr and ml plots may be used to discriminate quartz
from feldspardominant layers or domains in the continental
crust. Furthermore, Figure 9 illustrates the reason why acid
and intermediate rocks (e.g., granite, diorite, felsic gneiss,
intermediate gneiss and metasediments), which consist of
alkali feldspar, plagioclase, quartz, and mica, are low in both
l and r or m and l values.
[23] Accessory minerals such as zircon, rutile, ilmenite,
sillimanite and spinel show generally very high l and m
values with respect to the main rockforming silicate minerals
(Figure 9). Increasing the content of these minerals will result
in an increase of l and m for the bulk rock.
[24] In the ml diagram (Figure 9b), pyroxenes (ferrosilite,
augite, diallage, bronzite, enstatite, diopside, hedenbergite,
and aegirine) form a lozengeshaped domain while high
pressure pyroxenes, omphacite, and jadeite, lie at far outside
this domain. Compared with jadeite and omphacite, diallage,
enstatite, bronzite, and augite have relatively lower l values.
Ferrosilite has a much higher density than diopside, but both
minerals show similar l values. The role of FeMg substitution in orthopyroxenes can be seen by the comparison
between enstatite (MgSiO3) and ferrosilite (FeSiO3): r =
3.198 g/cm3, m = 75.7 GPa, l = 57.3 GPa, and m/l = 1.32
for enstatite whereas r = 4.002 g/cm3, m = 52.0 GPa, l =
66.3 GPa, and m/l = 0.78 for ferrosilite. Hornblende shows
similar l and r values to enstatite. In the ml diagram,
Figure 6. The lVp plots for (a) 38 peridotites, 12 serpentinites, and 15 partially serpentinized peridotites, (b) 54
eclogites, 118 mafic rocks (gabbro, diabase, mafic gneiss
and mafic granulite) and 145 felsic rocks (granite, diorite,
felsic gneiss, intermediate gneiss, and metasediments), and
(c) 31 amphibolites, 8 anorthosites, 21 basalts, and 29 limestones/marbles at a hydrostatic pressure of 600 MPa.
10 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

Figure 7. The lVs plots for (a) 31 amphibolites, 38 peridotites, 12 serpentinites and 15 partially serpentinized peridotites, (b) 54 eclogites, 118 mafic rocks (gabbro, diabase,
mafic gneiss, and mafic granulite) and 145 felsic rocks
(granite, diorite, felsic gneiss, intermediate gneiss, and metasediments), and (c) 8 anorthosites, 21 basalts, and 29 limestones/marbles at a hydrostatic pressure of 600 MPa.
however, hornblende lies close to anorthite. Obviously,
mafic rocks (e.g., gabbro, diabase, mafic gneiss and mafic
granulite) in which pyroxene and labradoriterich plagio-

B06314

clase (50% to 70% anorthite) are principal constituents have

moderate m, l and r values between the endmembers.
[25] In the olivine group [(Fe, Mg)2SiO4], both l and r
values increases with increasing the content of Fe2SiO4
(fayalite) whereas the m value increases with increasing the
content of Mg2SiO4 (forsterite). In other words, l shows a
positive correlation with r, but a negative correlation with m
for olivine. The role of FeMg substitution in olivines is
particularly significant: r = 3.222 g/cm3, m = 81.7 GPa, l =
74.2 GPa, and m/l = 1.1 for forsterite whereas r = 4.377 g/cm3,
m = 50.8 GPa, l = 102.7 GPa, and m/l = 0.49 for fayalite.
Figure 9 also indicates the position of olivine with forsterite
contents Fo100 to Fo90, which represents the typical composition of common olivine in the upper mantle. In both the
lr and ml diagrams, the upper mantle olivine is located
close to omphacite. A peridotite which is a mixture of
olivine, orthopyroxene and clinopyroxene should lie somewhere among these three endmembers, and the exact
position depends on the volume fraction of each constituent and also on the amount of accessory minerals such as
spinel, garnet, magnetite and ilmenite. The m, l and r of
peridotites decrease with increasing the degree of serpentinization because serpentine as their hydrated product has
extremely low values in l (32.3 GPa), m (15.67 GPa) and
r (2.516 g/cm3). Thus, the l or lr data may allow the
prediction of the serpentine content in the oceanic lithosphere from the inversion of surface seismic data through
AVO methods.
[26] In the lr or ml diagram (Figure 9), silicate garnets
can be classified into two groups: pyralspite (pyrope
almandinespessartine) and ugrandite (uvarovitegrassular
andradite) groups. From pyrope to almandine or spessartine,
l remains almost unchanged, but r increases considerably.
The m and l values for pyrope, almandine and spessartine
are very similar (Figure 9b). Therefore differences in the
relative proportions of these components have little influence on the m and l values of pyralspite. Uvarovite displays
higher l values than andradite although they have similar
densities. From grossular to andradite, l increases slightly
but r increases and m decreases significantly (Figure 9).
[27] Eclogites, which consist of garnet and omphacite
with rutile and magnetite as accessory minerals, are certainly
very high in m, l and r values (Figures 4 and 8). The mafic
granuliteeclogite transformation occurs near the base of
thickened continental crust that is forced into the upper
mantle during a collisional orogeny. Such a huge mass of
eclogite may delaminate from the crust and sink into the
mantle. As indicated by Figures 4b and 8b, the transition
from mafic granulite or gabbro to eclogite should result in a
significant increase in l, m and r. Such striking contrasts
make distinction between eclogite and mafic granulite possible. Eclogite has seismic velocities similar to surrounding
mantle peridotite (Figures 5b and 5c). However, the distinction of eclogite from peridotite is likely in the lr diagram
(Figure 5a).
[28] Muscovite shows a relatively lower l value with
respect to its density (Figure 9). If the similar situation is
true for other phyllitic minerals (e.g., biotite, phengite,
phlogopite and chlorite) whose elastic constants are not
available due to experimental restrictions, one would expect
that the micarich rocks, which often occur in retrograde

11 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

shear zones within the continental crust, are characterized by

low l values.
[29] As illustrated in Figure 8, the variation range in either
m or l is much larger for mafic rocks than highsilica rocks.
These differences reflect the fact that there is considerable
variation in plagioclase, clinopyroxene, orthopyroxene,
hornblende, and garnet in mafic rocks of various origins
(magmatic crystallization for gabbro and diabase, and
metamorphism for mafic granulite and mafic gneiss). Both
the m and l values of these minerals are very disparate and
hence a change in their relative proportion has a significant
effect on these elastic parameters. In the felsic rocks, however, the dominant minerals are potassium feldspar and
sodic plagioclase, both having similar m and l values. Thus
variations in m and l correspond mainly to the difference in
the relative proportion of quartz in the felsic rocks. Similarly, the large range of variations in m and l values for
eclogites result principally from the difference in the relative
proportion of garnet.
[30] Metamorphism is also an important factor to influence the l value of a rock. For example, pelitic rocks at
greenschist facies grades contain chlorite, muscovite, quartz
and albite. At the upper amphibolite or granulite facies
grades, however, the rocks with the same chemical composition contain garnet, sillimanite, oligoclase or andesine,
quartz and hornblende or pyroxene [Burlini and Fountain,
1993]. Increases in the contents of garnet, sillimanite,
hornblende and pyroxene and in anorthite content of plagioclase during progressive metamorphism should systematically result in an increase in rocks l value.
[31] Finally, the presence of fluids in porous materials
such as sedimentary rocks affects significantly l but not m
[Goodway, 2001]. The parameters m and l are sensitive to
the rock matrix and pore fluids, respectively. For example, it
is impossible to distinguish shale (Vp = 2.898 km/s) from
gas sandstone (Vp = 2.857 km/s) according their P wave
velocities. However, it is easy to discriminate these rocks
according to their l values: 12.3 GPa for shale and 5.9 GPa
for
gas sandstone.
The reason is simple. As Vp =
p

 2=, the l decrease of sandstone due to the gas

porosity is nearly completely offset by an increase in 2m resulting from the predominance of quartz in sandstone over
shale.

5. Conclusion
[32] Lam parameter (l), which is closely related to the
incompressibility and contains a high proportion of information about the resistance to a change in volume caused by
a change in pressure, is an important, intrinsic, elastic
property of rocks. Only l and shear modulus (m) appear in
Hookes law and not Youngs modulus (E), the bulk modulus (K) or Poissons ratio (u). Unfortunately, so far little
Figure 8. The ml plots for (a) 31 amphibolites, 38 peridotites, 12 serpentinites and 15 partially serpentinized peridotites, (b) 54 eclogites, 118 mafic rocks (gabbro, diabase, mafic
gneiss, and mafic granulite) and 145 felsic rocks (granite, diorite, felsic gneiss, intermediate gneiss, and metasediments),
and (c) 8 anorthosites, 21 basalts, and 29 limestones/marbles
(c) at a hydrostatic pressure of 600 MPa.
12 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

B06314

Figure 9. The (a) lr and (b) ml plots for main rockforming minerals. Ab, albite; Adr, andradite; Ab,
albite; Adr, andradite; Ae, aegirine; Alm, almandine; An, anorthite; Aug, augite; Br, bronzite; Cal, calcite;
Coe, coesite; Di, diopside; Dial, diallage; Dol, dolomite; En, enstatite; Ep, epidote; Fa, fayalite;
Fo, forsterite; Fs, ferrosilite; Grs, grossular; Hbl, hornblende; Hd, hedenbergite; Ilm, ilmenite; Jd, jadeite;
Law, lawsonite; Mag, magnetite; Ms, muscovite; Ne, nepheline; Omp, omphacite; Or, orthoclase; Prp,
pyrope; Qtz, quartz; Rt, rutile; Sil, sillimanite; Spl, spinel; Sps, spessartine; Srp, serpentine; Uvt, uvarovite;
Zrn, zircon. The values of l and m for each mineral were calculated from the elastic constants compiled
by Bass [1995].
has been known about the characteristic l value for each
common type of crystalline rocks that constitute the Earths
crust and upper mantle and its variation with pressure (P),
temperature (T) and mineralogical composition. Here we fill
such a gap by analyzing in details the l values of 475
natural rocks on which mean P and S wave velocities have
been measured at high hydrostatic pressures (400 MPa)
using the same laboratory equipment and the same method.

The l value of an equivalent isotropic crystalline rock as a

function of confining pressure (P) and temperature (T) can
be described by l = a + (dl/dP)P c exp( kP) (dl/dT)T,
where a is the projected l value at zero pressure if microcracks were fully closed; dl/dP is the pressure derivative in
the linear elastic regime; c is the initial l drop caused by the
presence of microcracks at zero pressure; k is a decay
constant of the l drop in the nonlinear poroelastic regime;

13 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

and dl/dT is the temperature derivative. The laboratory

experimental data may be extrapolated to the Earths interior
using the above equation if no partial melting, metamorphic
reaction, dehydration or phase transformation occurs. In the
lr and mr plots, the main categories of lithology can be
clearly discriminated. The ultramafic rocks display systematic decreases in both m and l with increasing the degree of
serpentinization. Eclogites, mafic rocks (gabbro, diabase,
mafic granulite, and mafic gneiss) and felsic rocks (granite,
diorite, felsic gneiss, intermediate gneiss and metasediments) are characterized by high, moderate and low m and l
values, respectively. For pyroxenes and olivines, both l and
r increase but m decreases with increasing the Fe/Mg ratios.
In the plagioclase series, both l and m increases with
increasing the anorthite content. Increases in the contents of
garnets, sillimanite, rutile, zircon, ilmenite and spinel result
systematically in an increase in rocks l and m values. The
results suggest that the lr and mr plots can be used as a
discriminant of composition for rocks in the Earths crust
and upper mantle. This is particularly important because the
connection between composition and seismic P or S wave
velocity is not unique due to similar velocities of many
common rock types.
[33] Acknowledgments. S. Ji thanks the Natural Sciences and Engineering Council of Canada for the discovery grants and the Chinese Academy of Geological Sciences for research grants. Qian Wang was supported
by the knowledge Innovation Program from Guangzhou Institute of Geochemistry, Chinese Academy of Sciences (GIGCX0902).

References
Barruol, G. (1993), Ptrophysique de la crote infrieure, Rle de lanisotropie sismique sur la rflectivit et le dphasage des ondes S, Ph.D. thesis, pp. 239258, Univ. de Montpellier II, Montpellier, France.
Barruol, G., D. Mainprice, H. Kern, M. Saint Blanquat, and P. Compte
(1992), 3D seismic study of a ductile shear zone from laboratory and petrofabric data (Saint Barthlmy Massif, northern Pyrnes, France),
Terra Nova, 4, 6376, doi:10.1111/j.1365-3121.1992.tb00451.x.
Bass, J. D. (1995), Elasticity of minerals, glasses, and melts, in Mineral Physics and Crystallography: A Handbook of Physical Constants, AGU Ref.
Shelf, vol. 2, edited by T. J. Ahrens, pp. 4563, AGU, Washington, D. C.
Berryman, J. G. (1995), Mixture theories for rock properties, in Rock Physics and Phase Relations: A Handbook of Physics Constants, AGU Ref.
Shelf, vol. 3, edited by T. J. Ahrens, pp. 205228, AGU, Washington,
D. C.
Birch, F. (1960), The velocity of compressional waves in rocks to 10 kilobar,
Part 1, J. Geophys. Res., 65, 10831102, doi:10.1029/JZ065i004p01083.
Burke, M. M., and D. F. Fountain (1990), Seismic properties of rocks from
an exposure of extended continental crustnew laboratory measurements
from the Ivrea Zone, Tectonophysics, 182, 119146, doi:10.1016/00401951(90)90346-A.
Burlini, L., and D. M. Fountain (1993), Seismic anisotropy of metapelites
from the IvreaVerbano zone and Serie dei Laghi (northern Italy), Phys.
Earth Planet. Inter., 78, 301317, doi:10.1016/0031-9201(93)90162-3.
ChvezPrez, S., and J. N. Louie (1997), Isotropic scattering and seismic
imaging of crust fault zones using earthquakes, paper presented at 67th
Annual International Meeting, Soc. of Explor. Geophys., Dallas, Tex.
ChvezPrez, S., and J. N. Louie (1998), Crustal imaging in southern
California using earthquake sequences, Tectonophysics, 286, 223236,
doi:10.1016/S0040-1951(97)00267-9.
Christensen, N. I. (1966), Elasticity of ultrabasic rocks, J. Geophys. Res.,
71, 59215931.
Christensen, N. I. (1974), Compressional wave velocities in possible mantle
rocks to pressures of 30 kilobars, J. Geophys. Res., 79, 407412,
doi:10.1029/JB079i002p00407.
Christensen, N. I., and R. Ramananantoandro (1971), Elastic moduli and
anisotropy of dunite to 10 kilobars, J. Geophys. Res., 76, 40034010,
doi:10.1029/JB076i017p04003.

B06314

Dufour, J., J. Squires, W. N. Goodway, A. Edmunds, and I. Shook (2002),

Integrated geological and geophysical interpretation case study, and
Lam rock parameter extractions using AVO analysis on the Blackfoot
3C3D seismic data, southern Alberta, Canada, Geophysics, 67, 2737,
doi:10.1190/1.1451319.
EberhartPhillips, D., D. Han, and M. D. Zoback (1989), Empirical relationships among seismic velocity, effective pressure, porosity, and clay
content in sandstone, Geophysics, 54, 8289, doi:10.1190/1.1442580.
Estabrook, C. H., and R. Kind (1996), The nature of the 660kilometer
uppermantle seismic discontinuity from precursors to the PP phase,
Science, 274, 11791182, doi:10.1126/science.274.5290.1179.
Goodway, B. (2001), AVO and Lam constants for rock parameterization
and fluid detection, CSEG Rec., 26, 3960.
Goodway, B., and J. Tessman (2000), Comparative reservoir imaging using
new seismicacquisition technology, World Oil, 221, 4954.
Goodway, B., T. Chen, and J. Downton (1997), Improved AVO fluid
detection and lithology discrimination using Lam petrophysical parameters, lr, mr, and l/m fluid stack, from P and S inversions, paper presented at 67th Annual International Meeting, Soc. of Explor. Geophys.,
Dallas, Tex.
Gray, D. (2003), A better way to extract fundamental rock properties with
much less noise, World Oil, 224, 4953.
Gray, D., and E. Andersen (2000), The application of AVO and inversion
to formation properties, World Oil, 221, 8590.
Jaeger, J. C. (1969), Elasticity, Fracture and Flow, With Engineering and
Geological Applications, 268 pp., Methuen, London.
Ji, S. C. (2004), A generalized mixture rule for estimating the viscosity of
solidliquid suspensions and mechanical properties of polyphase rocks
and composite materials, J. Geophys. Res., 109, B10207, doi:10.1029/
2004JB003124.
Ji, S. C., and M. H. Salisbury (1993), Shearwave velocities, anisotropy
and splitting in the high grade mylonites, Tectonophysics, 221, 453
473, doi:10.1016/0040-1951(93)90173-H.
Ji, S. C., and B. Xia (2002), Rheology of Polyphase Earth Materials,
259 pp., Polytech. Int. Press, Montreal, Que., Canada.
Ji, S. C., and Z. Q. Xu (Eds.) (2009), The Chinese Continental Scientific
Drilling (CCSD), Tectonophysics, 475(2), 201203, doi:10.1016/j.tecto.2009.04.007.
Ji, S. C., M. H. Salisbury, and S. Hanmer (1993), Petrofabric, P wave
anisotropy and seismic reflectivity of highgrade tectonites, Tectonophysics, 222, 195226, doi:10.1016/0040-1951(93)90049-P.
Ji, S. C., Q. Wang, and B. Xia (2002), Handbook of Seismic Properties of
Minerals, Rocks and Ores, 630 pp., Polytech. Int. Press, Montreal, Que.,
Canada.
Ji, S. C., K. Saruwatari, D. Mainprice, R. Wirth, Z. Q. Xu, and B. Xia
(2003), Microstructures, petrofabrics and seismic properties of ultrahighpressure eclogites from Sulu region, China: Implications for rheology of subducted continental crust and origin of mantle reflections,
Tectonophysics, 370, 4976, doi:10.1016/S0040-1951(03)00177-X.
Ji, S. C., Q. Wang, D. Marcotte, M. H. Salisbury, and Z. Q. Xu
(2007), P wave velocities, anisotropy and hysteresis in ultrahighpressure
metamorphic rocks as a function of confining pressure, J. Geophys. Res.,
112, B09204, doi:10.1029/2006JB004867.
Ji, S. C., Q. Wang, and M. H. Salisbury (2009), Composition and tectonic
evolution of the Chinese continental crust constrained by Poissons ratio,
Tectonophysics, 463, 1530, doi:10.1016/j.tecto.2008.09.007.
Kern, H. (1979), Effect of highlow quartz transition on compressional and
shear wave velocities in rocks under high pressure, Phys. Chem. Miner.,
4, 161171, doi:10.1007/BF00307560.
Kern, H. (1982), P and Swave velocities in crustal and mantle rocks under
the simultaneous action of high confining pressure and high temperature
and the effect of the rock microstructure, in HighPressure Research in
Geosciences, edited by W. Schreyer, pp. 1545, E. Schweizerbartsche,
Stuttgart, Germany.
Kern, H., and J. M. Tubia (1993), Pressure and temperature dependence of
P and Swave velocities, seismic anisotropy and density of sheared
rocks from Sierra Alpujata massif (Ronda peridotites, southern Spain),
Earth Planet. Sci. Lett., 119, 191205, doi:10.1016/0012-821X(93)
90016-3.
Kern, H., S. Gao, and Q.S. Liu (1996), Seismic properties and densities of
middle and lower crustal rocks exposed along the North China Geoscience Transect, Earth Planet. Sci. Lett., 139, 439455, doi:10.1016/
0012-821X(95)00240-D.
Kern, H., S. Gao, Z. M. Jin, T. Popp, and S. Y. Jin (1999), Petrophysical
studies on rocks from the Dabie ultrahighpressure (UHP) metamorphic
belt, central China: Implications for the composition and delamination of
the lower crust, Tectonophysics, 301, 191215, doi:10.1016/S0040-1951
(98)00268-6.

14 of 15

B06314

JI ET AL.: LAM PARAMETERS OF COMMON ROCKS

Kern, H., Z. M. Jin, S. Gao, T. Popp, and Z. Q. Xu (2002), Physical properties of ultrahighpressure metamorphic rocks from the Sulu terrain,
eastern central China: Implications for the seismic structure at the Donghai
(CCSD) drilling site, Tectonophysics, 354, 315330, doi:10.1016/S00401951(02)00339-6.
Li, Y. Y., J. Downton, and B. Goodway (2003), Recent applications of
AVO to carbonate reservoirs in the western Canadian sedimentary basin,
Leading Edge, 22, 670674, doi:10.1190/1.1599694.
Lubarda, V. A. (1998), A note on the effective Lam constants of polycrystalline aggregates of cubic crystals, J. Appl. Mech., 65, 769770,
doi:10.1115/1.2789122.
Mavko, G., T. Mukerji, and J. Dvorkin (2009), The Rock Physics Handbook:
Tools for Seismic Analysis in Porous Media, 511 pp., doi:10.1017/
CBO9780511626753, Cambridge Univ. Press, Cambridge, U. K.
Nadeau, J. C. (1999), A note on the effective Lam constants of polycrystalline aggregates of cubic crystals, J. Appl. Mech., 66, 577.
Shapiro, S. A. (2003), Elastic piezosensitivity of porous and fractured
rocks, Geophysics, 68, 482486, doi:10.1190/1.1567216.
Shen, A. H., W. A. Bassett, and I.M. Chou (1993), The ab quartz transition at high temperatures and pressures in a diamondanvil cell by laser
interferometry, Am. Mineral., 78, 694698.
Spetzler, H. A., and D. L. Anderson (1968), The effect of temperature and
partial melting on velocity and attenuation in a simple binary mixture,
J. Geophys. Res., 73, 60516060, doi:10.1029/JB073i018p06051.
Stierman, D. J., J. H. Healy, and R. L. Kovach (1979), Pressureinduced
velocity gradient: An alternative to a Pg refractor in the Gabilan range,
central California, Bull. Seismol. Soc. Am., 69, 397415.

B06314

Wang, Q., and S. C. Ji (2009), Poissons ratios of crystalline rocks as a

function of hydrostatic confining pressure, J. Geophys. Res., 114,
B09202, doi:10.1029/2008JB006167.
Wang, Q., S. C. Ji, M. H. Salisbury, B. Xia, M. B. Pan, and Z. Q. Xu
(2005a), Shear wave properties and Poissons ratios of ultrahighpressure
metamorphic rocks from the DabieSulu orogenic belt: Implications for
the crustal composition, J. Geophys. Res., 110, B08208, doi:10.1029/
2004JB003435.
Wang, Q., S. C. Ji, M. H. Salisbury, M. B. Pan, B. Xia, and Z. Q. Xu
(2005b), Pressure dependence and anisotropy of P wave velocities in
ultrahighpressure metamorphic rocks from the DabieSulu orogenic belt
(China): Implications for seismic properties of subducted slabs and origin
of mantle reflections, Tectonophysics, 398, 6799, doi:10.1016/j.tecto.
2004.12.001.
Watanabe, T., H. Kasami, and S. Ohshima (2007), Compressional and
shear wave velocities of serpentinized peridotites up to 200 MPa, Earth
Planets Space, 59, 233244.
Zimmerman, R. W., W. H. Somerton, and M. S. King (1986), Compressibility of porous rocks, J. Geophys. Res., 91, 12,76512,277,
doi:10.1029/JB091iB12p12765.
S. Ji, D. Marcotte, and S. Sun, Dpartement des Gnies Civil,
Gologique et des Mines, cole Polytechnique de Montral, Montral,
QC H3C 3A7, Canada. (sji@polymtl.ca)
Q. Wang, Key Laboratory of Marginal Sea Geology, Guangzhou Institute
of Geochemistry, Chinese Academy of Sciences, Wushan, Guangzhou
510640, China. (wangqian@gig.ac.cn)

15 of 15