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

net/publication/249868372

Compaction trends for shale and clean sandstone in shallow sediments, Gulf of
Mexico

Article  in  The Leading Edge · May 2009

DOI: 10.1190/1.3124935

CITATIONS READS
31 3,493

3 authors:

Tanima Dutta Gary Mavko

12 PUBLICATIONS   96 CITATIONS   
Stanford University
300 PUBLICATIONS   10,523 CITATIONS   
SEE PROFILE
SEE PROFILE

Tapan Mukerji
Stanford University
183 PUBLICATIONS   6,715 CITATIONS   

SEE PROFILE

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

Writing a book on Pore Pressure View project

Improved granular medium model View project

All content following this page was uploaded by Tanima Dutta on 02 August 2017.

The user has requested enhancement of the downloaded file.

 SPECIAL
S e i s mSECTION:
i c m o d
S ee il si m
n ig c m o d e l i n g

Compaction trends for shale and clean sandstone in shallow

sediments, Gulf of Mexico
TANIMA DUTTA, GARY MAVKO, and TAPAN MUKERJI, Stanford Rock Physics Laboratory
TIM LANE, BP Exploration and Production Technology Group
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

C ompaction depth trends are important

in drilling, basin modeling, and seismic
exploration for several purposes: (1) to de-
tect overpressure and hydrocarbon zones
and distinguish them from seismic velocity
anomalies; (2) to calculate interval velocities
and depth conversion involving seismic data and
Earth models; (3) to predict seismic signatures
of sand-shale interfaces as a function of depth;
and (4) to recognize over-compacted zones
due to uplift. Several authors have studied the
effects of compaction on the porosity of sands
and shales (e.g., Magara, 1980; Ramm and
Bjorlykke, 1994). The effects of compaction
on velocity-depth trends have been provided
by different authors (e.g., Al-Chalabi, 1997;
Faust, 1951; Japsen, 2000). However, porosity
and velocity depth trends in the shallow section
are not well established. The main challenge in
computing such trends is the paucity of well-
log data in the shallow subsurface. Figure 1, a
typical well log from the Gulf of Mexico, lacks
measurements in the shallow section (< 3000 ft
or ~1000 m) due to riser-less drilling, and the
log response from the deeper section cannot be Figure 1. A typical well log from deepwater Gulf of Mexico. No velocity and density
data are available in the shallow section. The deeper interval is overpressured, and
used to compute the normal compaction trend the pore pressure as measured by repeat formation tester (RFT) is higher than the
due to overpressure. One way to overcome this hydrostatic pressure.
challenge is to integrate data from multiple
sources. In this paper, we compute porosity and velocity depth GOM)
trends by integrating data from multiple sources including 2) P-wave velocity measurements (Green Canyon, GOM)
well logs, geotechnical borehole data, and core measurements using the isonic (sonic-while-drilling) tool which records
from shallow sections of the Gulf of Mexico, and laboratory monopole acoustic waveforms
measurements at low effective pressure. 3) Density measurements using a soil-boring tool and a bore-
Our paper is organized in three sections. First we dis- hole gravimeter in geotechnical boreholes (Green Canyon,
cuss the data used to compute the compaction trends and GOM)
the petrophysical analysis of borehole data. Next we compute 4) Porosity and P-wave velocity measurements on cores
empirical trends of bulk property (porosity) and elastic prop- (GOM)
erties (VP, VS, and VP/VS ratio) for shales and sandstones. Fi- 5) Density, P-wave velocity, and S-wave velocity measure-
nally we show how these empirical trends in sandstones agree ments on unconsolidated, clean sandstones by Zimmer
with predictions of an effective-medium model (extended (2003) at low effective pressures (0–20 MPa)
Walton model) combined with appropriate coordination of
number-porosity-pressure relations. The sequence of loading, unloading, and reloading cor-
responds to sedimentation, erosion, and resedimentation in
Data integration and petrophysical analysis a basin. The behavior during unloading and reloading is very
To obtain the compaction depth trends in shallow sedi- different from that during the first loading. When the pre-
ments, we integrate measurements of density, porosity, and consolidation stress is exceeded, the subsequent stress-strain
compressional- and shear-wave velocities from the following curve follows the “virgin” or “normal compaction” curve.
sources: This curve is much steeper than the unloading and reloading
curves. In the present study, we consider experimental data
1) Conventional LWD measurements (Green Canyon, during the loading cycles.

590 The Leading Edge May 2009

 S e i s m i c m o d e l i n g

volume of shale (vshale) is carefully interpreted

from gamma-ray, neutron, and resistivity logs
whenever they are available. For the purpose of
our study, it is crucial to identify any anoma-
lous zone that deviates from the normal com-
paction trend (based on petrophysical analysis).
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

For example, we observe an anomalous interval

at a depth ~1500 ft where resistivity and veloc-
ity are significantly lower than the background
trend (Figure 2). The similar depth interval in
the study area is characterized by higher poros-
ity than the background in a different borehole
(Figure 3). This interval corresponds to an over-
pressure zone, owing to the presence of shallow-
water flow sands. Identification of such anoma-
lous zones is valuable for our study so that we
can exclude these zones when establishing the
normal compaction trend.
Figure 2. Example of a well log showing shallow isonic measurements that were used We follow Terzaghi’s effective stress prin-
to compute velocity-depth-pressure trend. The interval of shallow water flow (SWF) ciple for computing effective pressure, which
sands is highlighted. Note that resistivity and velocity deviate from their normal trends is simply defined as the difference between the
in SWF sands.
overburden pressure and the pore fluid pressure.
The pore pressure is calibrated with repeat for-
mation tester (RFT) measurements whenever
available.

Results
We identified the normal compaction trends
for shale and sandstones in a shallow, suprasalt
section from the Gulf of Mexico. We present
the depth trends for porosity, velocities (V P and
VS), and V P /VS ratio. The depth trends of rock
properties can also be represented in terms of
effective pressure. We obtained the following
empirical relation that relates effective pressure
for normally compacted sediments to equiva-
lent depth in the study area:
(1)

where d = depth below mudline (ft) and P = ef-

fective pressure in psi/ft. Equation 1 was com-
Figure 3. Example of a borehole with density measurements that were used to puted using data from several wells in Green
compute porosity-depth-pressure trend. The probable interval of overpressure associated
with shallow water flow (SWF) sands is highlighted. Note that density and porosity Canyon, Gulf of Mexico.
deviate from their normal trends in SWF sands. Porosity-depth trends. Compaction disequi-
librium (undercompaction) is known as the
most dominant mechanism for overpressure development in
The downhole data are processed and interpreted for clastic basins and the deepwater environment (Dutta, 2002).
identification of different lithologies, especially clean sands, In order to predict pore pressure buildup in sediments due to
shales, and salts. To do this, we perform detailed petrophysi- compaction disequilibrium, it is necessary to establish a trend
cal analysis of wells containing shallow and intermediate log curve for porosity versus depth due to normal compaction.
data, and of geotechnical measurements. In the study area, One commonly used approach is based on Athy’s (1930) po-
some wells encounter salt at shallow depths. The compaction rosity versus depth relationship:
history of sediments beneath salt is different from that of the
(2)
suprasalt sediments. In the present study, we focus on the su-
prasalt section. Figures 2 and 3 show data from the shallow
interval in two boreholes with isonic compressional velocity where φ is the porosity at depth Z, φ0 is the depositional
and geotechnical density measurements, respectively. The porosity (i.e., critical porosity) at the mudline (Z=0), and c

May 2009 The Leading Edge 591

 S e i s m i c m o d e l i n g
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

Figure 4. (a) Porosity-depth trends in shale. The anomalously higher than background porosity at a depth of 1500 ft in well 2 corresponds to
shallow water flow (SWF). The log data from well 4 and well 6 are overpressured. The porosity-depth relation derived using the deeper section of
the log data underpredicts porosity in the shallow section. (b) Porosity-depth trends in clean brine sandstones.

is a constant. val of anomalously higher porosity than background at ~1500

Although several empirical results for porosity versus ft. This interval corresponds to the overpressure regime due
depth trends are available, they often lack calibration data in to the presence of shallow water flow (SWF), and is therefore
the shallow section. The porosity-depth trends in sands and excluded from computation of the normal compaction trend.
shales during early burial are mostly affected by packing and For sandstones, porosity measured by Gregory (1977) and
ductile grain deformation (Surdam et al., 1989). Although Paxton (2002) agree with porosity computed using geotech-
cementation may occur during shallow burial (Dutton and nical measurements at the shallow section, as well as porosity
Diggs, 1990), it is more common at higher burial depths. measured by Zimmer at low effective pressure. Log-derived
Diagenesis of shales is restricted to mechanical compaction porosities at intermediate depth usually show scatter due to
during shallow burial (less than ~80°C). Chemical processes silt content, but they agree reasonably with the mean trend of
in shales begin at an intermediate diagenetic level (80–140° other data sources.
C), including transformation of smectite to illite and libera- We observe that the trend of porosity reduction in the
tion of organic acids from organic matter. Ramm and Bjor- shallow section is different for shale and sandstones. Here,
lykke (1994) suggested a clay dependent regression equation porosity decreases more rapidly in shale than in sandstones
(Equation 3) for Norwegian shelf sand and shale data which primarily due to dewatering. The existing porosity-depth re-
is valid for mechanical compaction at shallow depths: lation derived using deeper log data (e.g., Wendt model) is
not applicable in the shallow section, especially for shales. We
(3)
fit a general exponential equation of the following form to get
porosity-depth empirical trends:
where Cl = total clay relative to total stable framework grains.
For clean sandstones, Cl = 0. (4)
We show porosity-depth trends for the Gulf of Mexico
shale and clean sandstones in Figure 4a and 4b, respectively. where Z = DBML (depth below mudline) in feet. We use the
For shales, porosities measured by Gregory (1977) and Ham- same functional form as shown in Equation 4 to compute
ilton (1979) agree nicely with porosities derived from density depth trends for V P, VS, and V P/VS (Table 1). Although an ex-
measurements in geotechnical boreholes at the shallow sec- ponential equation with just one term has been reported ear-
tion (0–3000 ft). These data are integrated with log-derived lier (e.g., Equation 2), we require two terms to better fit the
porosities at an intermediate depth (3000–6000 ft) to com- rapid change of rock properties in the shallow subsurface.
pute normal porosity-depth trend. Figure 4a shows an inter- Velocity-depth trends. Estimates of overpressure and

592 The Leading Edge May 2009

 S e i s m i c m o d e l i n g
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

Figure 5. (a)VP-depth trends in shale. (b) VP-depth trends in clean brine sands. The velocity measurements on the Gulf of Mexico core provide
useful calibrations at shallow depths.

amount of exhumation based on

Coefficients of exponential fits using a general exponential equation: sonic data for a sedimentary for-
Rock property = aebZ + cedZ where Z = depth below mud-line (DBML) in ft. mation rely on identification of
a normal velocity-depth trend.
Coefficients of exponents Such trends describe how sonic
Rock
Lithology velocity increases with depth in
property a b c d R2
relatively homogeneous, brine-
Porosity Shale 0.2875 -0.00774 0.4384 -0.0001761 0.9 saturated sedimentary forma-
tions as porosity is reduced
Porosity Clean 0.01132 -0.004874 0.3923 -1.678e-005 0.9
during normal compaction.
brine
Storvoll et al. (2005) estimated
sand
linear velocity-depth trend lines
VP Shale 6917 4.633e-005 -1652 -0.0003646 0.8 for sandstones and shales on the
VP Clean 6350 2.532e-005 -500.3 -0.003647 0.9 Norwegian shelf using well log
brine data. However, the trends for
sand the shallow section (0–500 m)
VS Shale 3540 1.2e-005 -3536 -0.000421 0.9 were extrapolated.
Figures 5a and 5b show our
VS Clean 1951 7.962e-005 -1788 -0.001942 0.9 VP-depth trends for shale and
brine sands respectively for the Gulf
sand of Mexico. Figure 5a shows that
V P/VS Shale 14.86 -0.001911 2.714 -4.028e-006 0.9 VP measurements in shales by
Gregory and Hamilton provide
V P/VS Clean 19.67 -0.007416 4.094 -0.0001182 0.9 useful calibration in the shallow
brine section. They agree with the VP
sand obtained from isonic measure-
ments in the shallow subsurface,
Table 1. Empirical depth trends of rock properties for shallow sediments. The depths are in ft, porosities
in fraction, and the velocities in ft/s. and also match the well-log ve-
locities at greater depth. The

May 2009 The Leading Edge 593

 S e i s m i c m o d e l i n g

suming shear-wave velocity approaches zero at

the mudline. This assumption is based on the
critical-porosity concept (Nur et al., 1998),
which suggests that sediments approach fluid
velocity at critical porosity. The porosity of
sediments immediately below the mudline cor-
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

responds to critical porosity (or depositional po-

rosity). For sandstones, no VS measurements are
available in shallow depths at the wells. Hence,
the VS measurements by Zimmer on uncon-
solidated sandstones conducted at low effective
pressures (0-20 MPa) contribute significantly in
computing the VS-depth trend (Figure 6b). We
use Equation 1 to convert effective pressures to
equivalent depths.
VP/VS-depth trends. The VP/VS ratio (which
is directly related to Poisson’s ratio) is a key pa-
rameter that relates to low effective pressure or
high pore pressure, especially under shallow-
Figure 6. (a)VS-depth trends in shale . (b) VS-depth trends in clean brine sands. The water flow conditions. Poor grain-to-grain con-
velocity measurements on unconsolidated sands by Zimmer (2003) provide useful tacts in overpressured zones with low effective
calibrations at shallow depths. stress affect the S-wave velocity more than the
P-wave velocity. In overpressured (softer) sedi-
ments, VP/VS is usually higher than the back-
ground trend. This has been observed in labo-
ratory experiments. The depth trend of VP/VS
is shown in Figure 7. For shales, we observe
an anomalous interval of low VP/VS at 1500-
2200 ft (Figure 7a). This anomaly corresponds
to partial gas saturation, as evident from core
observations at that well. VP/VS measurements
are not available in shale above 1000 ft due to
lack of VS measurements, and we extrapolate us-
ing critical-porosity concepts. For sandstones,
measurements by Zimmer provide useful cali-
brations for computing VP/VS-depth trends. We
observe that VS decreases much faster than VP at
low pressure/shallow depth, and VP/VS can be as
high as 10 in the sediments immediately below
mudline (Figure 7b).
An exponential equation similar to Equa-
Figure 7. (a) VP/VS-depth trends in shale. (b) VP/VS-depth trends in clean brine tion 4 describes depth trends of VP,VS, and
sands. VS decreases much faster than VP at low pressure/shallow depth. VP/VS can be as VP/VS. Table 1 lists the coefficients. Equation
high as 10 in the shallow sands immediately below mudline. 1 can be used along with these coefficients to
represent trends of rock-properties in terms of
effective pressure.
shallow subsurface is mostly dominated by shale; hence, log Figure 8a and Figure 8b explore the VP/VS relations in
data for sandstones tend to be rare. Therefore, velocity mea- shale and sand respectively. Vernik’s VP/VS equation for shale
surements on sandstones using core from the Gulf of Mexico (2002) and the mudrock relation by Greenberg and Castagna
and experimental data on unconsolidated sands provide valu- (1992) agree with the mean trend of log data from the Gulf
able calibration for computing VP-depth trends in the shallow of Mexico shales (Figure 8a). However, the VP/VS relation by
subsurface (Figure 5b). Greenberg and Castagna for sandstone fails to predict the
There are fewer data for describing VS-depth trends since trend for sandstones, especially at shallow depths (grey points
it is difficult to measure shear-wave velocities in the shallow in Figure 8b). Vernik’s VP/VS equations differ from Green-
section. The VS-depth trend for shale (Figure 6a) is obtained berg and Castagna at shallow depths since they honor the
from log data only, and no calibration data are available above critical-porosity concept. We find that Vernik’s VP/VS equa-
1000 ft. DBML. The VS above this depth is extrapolated, as- tion for sandstone is similar to Greenberg and Castagna at

594 The Leading Edge May 2009

 S e i s m i c m o d e l i n g

where C is the coordination number, η is the

solid volume fraction, φ is the porosity, G is
the grain shear modulus, and P is the effec-
tive pressure. When α is zero, we obtain the
Walton smooth model, where all the grains
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

have zero friction at their contacts. And,

when α is equal to 1, we obtain the Wal-
ton rough model, which considers infinitely
high friction at the grain contacts of identi-
cal spheres. Since the effective shear moduli
(Geff ) of a granular aggregate is linearly pro-
portional to tangential stiffness, it is possible
to lower Geff by assuming tangential stiffness
is zero for some fraction of the grains. We use
Equations 5 and 6 to represent this scenario,
where α represents the fraction of complete-
ly adhered contacts, and (1-α) represents the
Figure 8. (a) VP/VS crossplot in shale. (b) VP/VS crossplot in clean brine sands. In the
fraction of completely slipping contacts. We
shallow sediments, Vernik’s VP-VS relation is more appropriate than the Greenberg and
refer to this as the extended Walton model.
Castagna relation.
Note that mixing of frictional and nonfric-
tional grains can also be achieved assuming
intermediate depths, but provides a better prediction at shal- some of the grains have zero tangential stiffness in the Hertz-
low depths. Mindlin effective medium model (Bachrach et al., 2000).
Effective medium modeling for velocity-depth trends in We use the extended Walton model (α = 0.6) to invert
sands. An understanding of rock physics models is important the coordination number (C) from measurements on dry, un-
for a reliable prediction of pore pressure in shallow sediments. consolidated sandstones by Zimmer. We choose this particu-
Japsen (2000) investigated the relations between rock phys- lar value of α since it provides realistic prediction of C for a
ics models and velocity-depth trends for normal compaction. random pack of grains. The relations between C-porosity and
He presented baselines for sandstones and shales based on C-pressure (Dutta et al.) are expressed as:
a modified Voigt trend, and on a constrained time-average
equation, respectively. Avseth et al. (2001) showed that the Cp(φ)= (–2.338e-014) φ 9.099 + 14.45 (7)
Hertz Mindlin theory (Mavko et al., 1998) can be applied to
predict normal velocity-depth trends in unconsolidated sedi- Cs(φ)= (–1.798e-014) φ 9.099 + 11.02 (8)
ments. However, the modeling approach by Avseth lacked
calibration data above ~6000 ft. In this paper, we use the Cp(P)= (777.1) P 0.001545 -770.7 (9)
extended-Walton model along with new coordination num-
ber porosity-pressure relations to compute velocity-depth Cs(φ)= (–4.457) P –0.2724 + 9.401 (10)
trends in clean, dry sandstones (Dutta, 2009). The coordina-
tion number represents average number of grain contacts in where Cp and Cs represent coordination numbers to predict
a granular pack. Next, Gassmann’s fluid substitution is per- V P and VS , respectively. φ is porosity in percentage, and P
formed to obtain depth trends in brine sandstones. is effective pressure in MPa. We find that prediction of VS
The Walton model (1987) provides effective elastic mod- requires a lower value of C than the prediction of V P.
uli for two special cases: infinitely rough spheres (friction We apply the above Cp-pressure and Cs-pressure relations
coefficient is very large) and ideally smooth spheres (friction (Equations 9 and 10) to the extended Walton model (α =
coefficient is zero). Jenkins et al. (2005) expressed the effec- 0.6) to predict the velocities as a function of pressure. Figure
tive elastic moduli of Walton model as follows: 9 shows that both VP and VS predictions by this model agree
nicely with the integrated data sets at shallow depths.
(5)
Conclusions
We provided normal compaction depth trends of porosity,
velocities, and VP/VS ratio for shallow, suprasalt sediments in
the Gulf of Mexico. For this, we combined data from mul-
(6) tiple sources, such as isonic, geotechnical, and core measure-
ments from shallow sections of the Gulf of Mexico, and labo-
ratory measurements on unconsolidated sands conducted at
low effective pressure. However, the lack of conventional log

May 2009 The Leading Edge 595

 S e i s m i c m o d e l i n g

cementation in the Lower Cretaceous Travis Peak

Formation, East Texas” by Dutton et al. (Journal
of Sedimentary Petrology, 1990). “Seismic velocity
as a function of depth and geologic time” by Faust
(Geophysics, 1951). “Shear-wave velocity estima-
tion in porous rocks: Theoretical formulation, pre-
Downloaded 08/02/17 to 12.109.29.106. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

liminary verification and applications” by Green-

berg and Castagna (Geophysical Prospecting, 1992).
“Aspects of rock physics from laboratory and log
data that are important to seismic interpretation”
by Gregory (in Seismic Stratigraphy Application
to Hydrocarbon Exploration, AAPG Memoir 26,
1977). “Sound speed and related physical proper-
ties of sediments from experimental Mohole (Gua-
dalupe site)” by Hamilton (Geophysics, 1965).
“VP/VS and Poisson’s ratio in marine sediments and
rocks” by Hamilton (Journal of Acoustic Society of
America, 1979). “Construction of an intergranu-
lar volume compaction curve for evaluating and
predicting compaction and porosity loss in rigid-
grain sandstone reservoirs” by Paxton (AAPG Bul-
letin, 2002). “Investigation of multiphase erosion
using reconstructed shale trends based on sonic
data, Sole Pit axis, North Sea” by Japsen (Global
Figure 9. (a) VP-depth trends in clean brine-sands (left); (b) VS-depth trends in clean and Planetary Change, 2000). “Fluctuations and
brine sands (right). The solid line represents velocity predictions using the extended the effective moduli of an isotropic, random ag-
Walton model with appropriate coordination number-pressure relations. gregate of identical, frictionless spheres” by Jenkins
et al. (Journal of the Mechanics and Physics of Sol-
data in the shallow section remains a challenge for calibra- ids, 2005). “Comparison of porosity-depth relationships of shale
tion of these trends. We find that the trends of rock prop- and sandstone” by Magara (Journal of Petroleum Geology, 1980).
erties in the shallow subsurface can be drastically different Th e Rock Physics Handbook by Mavko et al. (Cambridge, 1998).
from that of the deeper section. Therefore, simple extrapola- “Critical porosity; a key to relating physical properties to porosi-
tion of depth trends upwards, computed using data from the ty in rocks” by Nur et al. (TLE, 1998). “Porosity/depth trends in
deeper section, can lead to erroneous results in the shallow reservoir sandstones: assessing the quantitative effects of varying
section. Furthermore, we offer an effective medium model- pore-pressure, temperature history and mineralogy, Norwegian
ing approach for clean sands using appropriate coordination Shelf data” by Ramm and Bjørlykke (Clay Minerals, 1994). “Ve-
number-pressure-porosity relations in the extended Walton locity-depth trends in Mesozoic and Cenozoic sediments from
model. These results can be useful in establishing normal the Norwegian Shelf ” by Storvoll et al. (AAPG Bulletin, 2005).
compaction trends in the shallow subsurface, where log data “Conceptual models for the prediction of porosity evolution
are usually not available and are often poor quality. Estab- with an example from the Frontier sandstone, Bighorn Basin,
lishing such trend curves is important in seismic exploration Wyoming” by Surdam et al. (Rocky Mountain Association of Ge-
for several reasons, especially for detecting drilling hazards ologists, 1989). “Estimation of net-to-gross from P and S imped-
and distinguishing shallow resource-potential prior to drill- ance in deepwater turbidites” by Vernik et al. (TLE, 2002). “The
ing in deep water. effective elastic moduli of a random pack of spheres” by Walton
(Journal of the Mechanics and Physics of Solids, 1987). “Seismic
Suggested reading. “Instantaneous slowness versus depth func- velocities in unconsolidated sands: Measurements of pressure,
tions” by Al-Chalabi (Geophysics, 1997). “Compaction and sorting, and compaction effects” by Zimmer (PhD dissertation,
oil migration” by Athy (AAPG Bulletin, 1930). “Rock phys- Stanford University, 2003).
ics and seismic properties of sands and shales as a function of Acknowledgments: The authors thank BP for permission to use the
burial depth” by Avseth et al. (SEG 2001 Expanded Abstracts). borehole data from Gulf of Mexico. We thank Juan Florez, Richard
“Rock lithology and porosity determination from shear and Weiland, and J.P. Blangy from BP for valuable discussions. Special
compressional wave velocity” by Domenico (Geophysics, thanks to James Packwood (formerly at BP) for his guidance with
1984). “Deepwater geohazard prediction using prestack inver- detailed petrophysical analysis. This work was supported by the
sion of large offset P-wave data and rock model” by Dutta (TLE, Stanford Rock Physics and Borehole Geophysics project and DOE
2002). “Integrating sequence stratigraphy and rock physics to contract DE-FC26-04NT15506.
interpret seismic amplitudes and predict reservoir quality” by
Dutta (PhD dissertation, Stanford, 2009). “History of quartz Corresponding author: [email protected]

596 The Leading Edge May 2009

View publication stats