See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/290177612

Vs Prediction in Unconsolidated Sands –

Physical and Geological Controls on Shear Wave
Velocities

Article · June 2010

DOI: 10.3997/2214-4609.201401219

CITATIONS READS

0 72

4 authors, including:

Nazmul Haque Mondol Per Avseth

University of Oslo / Norwegian Geotechnical I… Norwegian University of Science and Techno…
110 PUBLICATIONS 574 CITATIONS 105 PUBLICATIONS 1,220 CITATIONS

SEE PROFILE SEE PROFILE

Manzar Fawad
University of Oslo
10 PUBLICATIONS 73 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

New advances in quantitative seismic interpretation View project

Reservoir Characterization of Traissic and Jurassic sandstones of Snorre field, the northern North Sea
View project

All content following this page was uploaded by Manzar Fawad on 02 May 2016.

The user has requested enhancement of the downloaded file.

P351
Vs Prediction in Unconsolidated Sands - Physical
and Geological Controls on Shear Wave Velocities
N.H. Mondol* (University of Oslo/NGI), P. Avseth (Odin Petroleum/NTNU),
M. Fawad (University of Oslo) & T. Smith (Conoco Phillips)

SUMMARY
Vp-Vs relations play a key role in lithology and pore fluid prediction from both sonic and seismic data.
Shear wave velocities are required as input to AVO modelling and analysis and for standard fluid
substitution using Gassmann equation. Shear waves are almost unaffected by pore fluid changes, as
opposed to compressional waves. Hence, Vp-Vs relations are useful fluid indicators. Furthermore, Vp-Vs
relations will be diagnostic of lithology and cement volume since different lithologies and amount of
cements will have different trends in the Vp versus Vs plane.

72nd EAGE Conference & Exhibition incorporating SPE EUROPEC 2010

Barcelona, Spain, 14 - 17 June 2010
Introduction

Together with compressional wave (Vp), shear wave (Vs) is an important parameter to discriminate
lithology and pore fluid. As a frame indicator, it also helps to estimate cement volume in silisiclastic
rocks. When Vs is unavailable, it is generally predicted from Vp using empirical relations. Castagna
et al. (1985) derived first the famous “Mudrock Line” which relates Vp to Vs for brine-saturated
silisiclastic rocks. Afterwards, several others have published various Vp-Vs relations for sandstones and
mudstones (Han, 1986; William, 1990; Krief et al., 1990; Greenberg and Cantagna, 1992; Mondol et al. 2009).
Castagna et al. (1993) extended the work of Castagna et al. (1985) and derived linear relationships for
various lithologies and mixed mineralogies. These relations are mostly useful for Vs prediction in
consolidated sandstones. Xu and White (1996) applied heuristic inclusion model to predict Vp and Vs
as functions of porosity and clay content. When mixing sands and clays, Xu and White assumed sands
with relatively stiff pores and aspect ratios ranging from 0.1 to 0.15, whereas the clay-related pores
were assumed to be more compliant with aspect ratios ranging from 0.02 to 0.05. A comparison of
typical Vp-Vs empirical relations commonly used for Vs predictor is shown in Figure 1 together with
published and unpublished data of brine-saturated sands and shales. It is clear from comparison that
all popular Vp-Vs linear regressions are no longer valid for shallow unconsolidated sands.

The variations between different Vs predictors 4000

can be explained by natural variability of

mineralogy and textural composition of sands 3500
and this could add significant error in any Vs
estimation. Although the empirical relations 3000
are found to be very useful in seismic
exploration, they provide little or no insight
Vp (m/s)

into the physics behind the changes in Vp/Vs 2500 Lista shale (Avseth et al., 2009)
Dvorkin, 2002
ratio. Neither do they serve well for Heimdal Sand (Avseth et al., 2009)
Sand database, UiO
interpretation of geological trends. The role of 2000 Prasad, 2002
Hamilton, 1979
this study is to systemize and improve the Castagna et al., 1993
understanding of the physical and geological 1500
Greenberg and Castagna, 1992
Krief et al., 1990
controls on development of Vp and Vs of William, 1990
Han, 1986
shallow unconsolidated sands. In particular, Castagna et al., 1985
1000
we study the effect of mineralogy, sorting, size
0 400 800 1200 1600 2000
and shape of grains that control compaction, Vs (m/s)
rotation, slip and breakage of grains and
influence Vp-Vs relations of shallow Figure 1 A comparison of Vp-Vs empirical relations
unconsolidated sandstones. Together with (Castagna et al., 1985; Han, 1986; William, 1990; Krief et
al., 1990; Greenberg and Cantagna, 1992 and Castagna et
literature data, a wide variety of well al., 1993) for sandstones are shown. Brine-saturated
characterized synthetic sands (known herein as published (Prasad, 2002; Dvorkin, 2002; Avseth et al.,
‘sand database, UiO) compacted mechanically 2009) and unpublished (sand database, UiO) data of
sandstones and shales are also shown in comparison.
in the laboratory were analyzed to investigate
physical and geological control on shear wave
velocities for unconsolidated sands. The Vs estimation from Vp using empirical regressions were then
evaluated by comparing the recent laboratoty measurements and literature data.

Materials and Methods

In addition to literature data (Prasad, 2002, Dvorkin, 2002, Avseth et al., 2009), a total of 29 dry
compaction experiments were performed using a wide variety of sands to observe the effect of size,
mineralogy, sorting and shape of grains on Vp and Vs as a function of stress. A high stress uniaxial
oedometer was used to perform compaction tests (up to a maximum 50 MPa effective stress) of 8
main sand samples classified by quartz arenites, subarkose, volcanic arenite, arkosic arenite and
greywacke (Table 1). These sands were then seperated to 21 samples which represent coarse, medium
and fine fractions of 8 end member sand. Detail mineralogical characterization and grain size analysis
of samples were peformed by XRD and sieve method.

72nd EAGE Conference & Exhibition incorporating SPE EUROPEC 2010

Barcelona, Spain, 14 - 17 June 2010
 Sand classification Minerals XRD (%) Sand classification Minerals XRD (%)
Quartz Arenite-1 Quartz 95.27 Arkosic Arenite Quartz 54.85
K-Feldspar 1.27 Kaolinite 2.51
Albite 3.46 Illite 8.47
K-Feldspar 14.39
Albite 19.79
Quartz Arenite-2 Quartz 97.88 Subarkose-1 Quartz 91.27
K-Feldspar 1.77 K-Feldspar 6.49
Ankerite 0.35 Albite 2.24
Quartz Arenite-3 Quartz 94.35 Subarkose-2 Quartz 77.19
Kaolinite 5.65 K-Feldspar 22.81
Volcanic Arenite Cristobalite 4.84 Greywacke Quartz 35.50
Aragonite 4.98 Amphibole 1.19
Calcite 7.81 Chlorite 4.16
Ankerite 3.80 Muscovite 18.82
Augite 26.04 K-Feldspar 24.21
Plagioclase 52.51 Albite 16.12
Table 1 Sand classification using XRD quantitative analysis.

Sorting was measured using the weight percentage of different grain sizes obtained from sieving. The
shape parameters like roundness and sphericity were computed using image analysis software ‘Image
J’. The dry results were theoretically substituted with brine using Gassman fluid substitution scheme.
Four brine-saturated experiments were carried out to check validity of Gassmann brine substitution
and found good agreement between Gassmann results and direct laboratory measurements. Rocks in
nature are generally fluid saturated, therefore only the Vp and Vs data derived from Gassmann brine
substitution were used in this study for comparison of published and lab data and empirical models.
No squirt flow dispersion was observed between laboratory and Gassmann predicted velocities, and
this can be explained by the high porosity and loose nature of the sands.

Results and discussion

The comparison of experimental results (sand database, UiO), and literature data demonstrates clearly
the influence of size (Fig. 2), mineralogy (Fig. 3a), sorting (Fig. 3b) and shape (Fig. 3c) of sand grains
on Vs for unconsolidated sands. Shear waves are higher for smaller grains compared to larger grains
for common mineralogy (Fig. 2). Quartz arenite shows higher Vs compared to volcanic arenite,
greywacke and subarkose suggests dependency of Vs on mineralogy for unconsolidated sands (Fig.
3a). Spencer et al. (1994) found that the effective Poisson’s ratio (can also define by Vp/Vs ratio) in
unconsolidated sands does not depend on mineralogy contrary to our observations. A strong
dependency of Vs is observed for sorting and shape of sand grains where poorly sorted greywacke
shows lower Vs then well sorted quartz arenite (Fig. 3b) and angular quartz arenite shows higher Vs
then subangular and subrounded volcanic arenite and subarkose (Fig. 3c).

Figure 2 Shows grain size effects on Vp-Vs relations for unconsolidated sands and sandstones. Three quartz arenites
have more or less similar mineralogy (dominated by quatz ~95%) but vary in grain size. At same Vp value Vs is 80-
100 m/s higher for fine-grained quartz arenite-1 (about 60% silt-sized quartz grains) compared to medium-grained
quartz arenite-3 (about 80% medium-sized quartz grains).

72nd EAGE Conference & Exhibition incorporating SPE EUROPEC 2010

Barcelona, Spain, 14 - 17 June 2010
The difference beween Vs predictors
(e.g. Williams, 1990, Krief et al.,
1990, Castagna et al., 1993) and large
scattered of published data emphasize
the importance of sand textural and
mineralogical relations on the Vs
estimate from Vp in unconsolidated
sands. The popular linear regressions
break down for unconsolidated sands
at low stresses. However, the
quadratic relationship of Krief et al.
(1990), Hertz-Mindlin contact theory
and Walton model (Walton, 1987) are
found to work better for
unconsolidated sands (Figs. 1-3).
Walton model assumes that normal
and shear deformation of a two-grain
combination occur simultaneously.
Then, there is no partial slip in the
contact area. The slip will occur across
the whole contact area once applied
tangential forces exceed friction
resistance. Avseth and Bachrach
(2005) also found validation of
Walton contact theory for loose sands
where no friction was assumed at the
grain contacts.

Manificat and Guegen (1998) found

that rough grain contacts (i.e., grain
angularity) caused a dry Poisson’s
ratio in the range of 0.12-0.15,
considerably higher that what is
predicted from the ideal Hertz-Mindlin
theory (0-0.07). Accounting for the
roughness, they found that the
tangential stiffness was halved
because only half of the asperities are
active in a given shear direction. They
concluded that the shear asymmetry
has a determining role on the Vp/Vs
ratio of unconsolidated granular rocks. Figure 3 Cross plots of Vp and Vs showing effects of (a)
The same conclusions were made by mineralogy, (b) shape and (c) sorting on Vp-Vs relations for
Makse et al. (2004) via numerical unconsolidated sands. At same Vp value, Vs varies 80-100 m/s for
simulations where they found that variations of mineralogy, sorting and shape of sand grains.
contact theory fails because it does not
properly allow the grains to relax from the affine, macroscopic deformation (i.e. each grain translates
according to the direction of the macroscopic strain) or to rotate. The assumption of affinity given by
the Hertz-Mindlin theory is approximately valid for the bulk modulus, but seriously flawed for the
shear modulus. The limitations in utilizing inclusion based models (e.g. Xu and White, 1996) for Vs
prediction for unconsolidated sands is the input parameter ‘aspect ratio’. The aspect ratio is almost
impossible to determine from thin-sections or other observables. Hence, it represents a fudge
parameter and the inclusion models are therefore heuristic more than theoretical. Furthermore, the
inclusion models assume no pore connectivity, and will therefore not be realistic for high porosity
sands and sandstones. All the empirical formulas, except of the semi-analytical relationship of Krief,

72nd EAGE Conference & Exhibition incorporating SPE EUROPEC 2010

Barcelona, Spain, 14 - 17 June 2010
have been derived for consolidated rocks and do not take into account geological and physical
parameters. Hence, these models fail to predict Vs from Vp for shallow unconsolidated sands.

Conclusions

We demonstrate physical and geological controls on shear wave velocities for unconsolidated sands
by well characterized laboratory measurements and literature data. Though the dependency of Vs on
size, mineralogy, sorting and shape of sand grains are found important but at this stage it is difficult to
isolated individual factor as several factors may contribute simultaneously and influence the Vp-Vs
relations. The variations of mineralogy, shape, size and sorting of sand grains are controlled largely
by provenance and depositional environments. To predict more reliable Vs from Vp for unconsolidated
sands, it is necessary to understand the geological process. The comparison of published data and lab
measurements confirms that the Vp-Vs relations are not linear for shallow unconsolidated sands, and
using popular linear regressions to estimate Vs from Vp will definately yield inaccurate results. The
linear relations do not take into account different depositional environments or digenetic history.
Neither do they provide a physical links to geologic parameters. However, Hertz-Mindlin contact
theory, Krief et al. and Walton models are useful to predict Vs for shallow unconsolidated sands.

References
Avseth, P. and Bachrach, R. [2005] Seismic properties of unconsolidated sands: Tangential stiffness, Vp/Vs ratios and
diagenesis. SEG Extended Abstract, 1473-1476.
Avseth, P., Bachrach, R., Barsås, T. and Haaland, A. N. [2009] Fluid and stress sensitivity in cemented sandstones. SEG
Extended Abstract, 2015-1019.
Castagna, J. P., Batzle, M. L. and Eastwood, R. L. [1985] Relationships between compressional-wave and shear-wave
velocities in clastic silicate rocks. Geophysics 50(4), 571-581.
Castagna, J. P., Batzle, M. L. and Kan, T. K. [1993] Rock physics-The link between rock properties and AVO response.
Offset-dependent reflectivity-Theory and practice of AVO analysis, J. P. Castagna and M. M. Backus (eds.), 135-171.
Society of Exploration Geophysicists, Tulsa, OK, United States.
Dvorkin, J. [2002] Rock Physics hand notes.
Greenberg, M. L. and Castagna, J. P. [1992] Shear-wave velocity estimation in porous rocks; theoretical formulation,
preliminary verification and applications. Geophysical Prospecting 40(2), 195-209.
Hamilton, E. L. [1979] Vp/Vs and Poisson's ratios in marine sediments and rocks. Journal of the Acoustic Society of
America 66(4), 1093-1101.
Han, D., Nur, A. and Morgan, D. [1986] Effects of porosity and clay content on wave velocities in sandstones. Geophysics
51(11), 2093-2107.
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 31(6), 355-369.
Makse, H. A., Gland, N., Johnson, D. L. and L. Schwartz, L. [2004] Granular packings: Nonlinear elasticity, sound
propagation and collective relaxation dynamics: Physical Review E, 70, 061302 1-19.
Manificat, G. and Guéguen, Y. [1998] What does control Vp/Vs in granular rocks? Geophysical Research Letters, 25(3),
381–384.
Mindlin, R. D. [1949] Compliance of elastic bodies in contact. Journal of Applied Mechanics 16, 259-268.
Mondol, N. H., Fawad, M., Grande, L., Jahren, J., and Bjørlykke, K. [2009] Vs estimation in mudrocks-Implications about
mineralogy and grain size. EAGE Extended Abstract, Z012.
Prasad, M. [2002] Acoustic measurements in unconsolidated sands at low effective pressure and overpressure detection.
Geophysics 67(2), 405-412.
Spencer, J. W. J., Cates, M. E. and Thompson, D. D. [1994] Frame moduli of unconsolidated sands and sandstones.
Geophysics 59(9), 1352-1361.
Walton, K. [1987] The effective elastic moduli of a random packing of spheres. Journal of the Mechanics and Physics of
Solids 35, 213-226.
Williams, D. M. [1990] The acoustic log hydrocarbon indicator. Transactions of the SPWLA thirty-first annual Logging
symposium, W1-W22. Society of Professional Well Log Analysts, Houston, TX, United States.
Xu, S. and White, R. E. [1996] A physical model for shear-wave velocity prediction. Geophysical Prospecting 44(4), 687-
717.

72nd EAGE Conference & Exhibition incorporating SPE EUROPEC 2010

Barcelona, Spain, 14 - 17 June 2010
View publication stats