SPE 109662

Enhanced Mechanical Earth Modelling and Wellbore Stability Calculations Using

Advanced Sonic MeasurementsA Case Study of the HP/HT Kvitebjorn Field in the
Norwegian North Sea
Anke S. Wendt, SPE, Marit Kongslien, Bikash K. Sinha, SPE, Badarinadh Vissapragada, SPE, and Adrian Newton,
SPE, Schlumberger, and Eiliv Skomedal, Lasse Renlie, and Erik Sandtorv Pedersen, Statoil

Copyright 2007, Society of Petroleum Engineers

This paper was prepared for presentation at the 2007 SPE Annual Technical Conference and
Exhibition held in Anaheim, California, U.S.A., 1114 November 2007.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.

Abstract
Geomechanical
interpretations
of
inhomogeneous
anisotropic materials are enhanced by three dimensional (3D)
sonic measurements that provide compressional, fast shear,
slow shear, and Stoneley wave slownesses in (an)isotropic
drilled formations.
We investigated the Kvitebjorn high pressure high
temperature (HPHT) gas/condensate field for enhanced
Mechanical Earth Modelling, and wellbore stability analyses
using three-dimensional sonic measurements in overburden
and reservoir formations. Kvitebjorn field is situated on the
Norwegian continental shelf west of the Viking Graben
In the HPHT Kvitebjorn field, major development
challenges are the determination of the drawdown rate and the
maximum depletion for optimum production while
maintaining rock integrity. These challenges are directly
related to the effect of changes in reservoir pressure on the insitu formation stress state. Using advanced processing of sonic
data, we estimated the amount of stress anisotropy, stress
orientations, and stress magnitudes at particular reservoir
intervals. These stress data together with measured in-situ
minimum horizontal stress values were used to calibrate
horizontal stress logs perviously estimated in the MEM.
Furthermore, we constrained rock elastic properties
(deformation, strength) using sonic slownesses and elastic
shear moduli calculated from the sonic dataset. These data
were later used to estimate rock strength and the onset of
wellbore deformation. The two approaches allowed us to
calculate (1) isotropic rock properties from the sonic

slownesses, and (2) vertical and horizontal rock properties

from the calculated shear moduli.
Validity of the so-calibrated MEM was subsequently tested
estimating wellbore stabilities based on the isotropic rock
property data, and the vertical and horizontal rock property
data sets. In each case, the model was able to accurately
reproduce the events observed during drilling.
The work that we present here is intended to open a new
direction for geomechanical modelling: To demonstrate a new
measurement allowing practical consideration of rocks as an
anisotropic material, and to better constrain the effective stress
ratio.
Introduction
The Kvitebjorn field is an HPHTgas/condensate field,
operated by Statoil ASA in the Norwegian North Sea.
Reservoir formations comprise the Brent and Dunlin Groups.
They consist of clastic materials, subjected to reservoir
pressures of 77MPa and temperatures of 150oC. The field is
intensively faulted, with a predominant fault trend in NE-SW
direction. High resolution sonic data sets were obtained in
several wells, from which one field example is shown here.
Numerous theoretical and experimental works have been
undertaken in the recent past to understand the complexity of
high pressure high pressure fields. HPHT means oilfield
operations are technically challenging, and have high costs;
with the North Sea including some of the most highly
overpressured reservoirs in the world. Therefore it is
imperative to examine all the available data, and apply all the
current knowledge and new technology to assess stresses, and
rock properties.
In HPHT fields, it is often suggested that overburden,
horizontal stresses and pore pressures are nearly equal under
the high pressure and high temperature conditions.
Consequently the safe drilling window is narrow (e.g. Ray et
al., 1998). Therefore, the development of Enhanced
Mechanical Earth Models (MEM) based on precise horizontal
stress magnitudes and anisotropic rock elastic properties is
increasingly appreciated in the petroleum industry. These data
are required for advanced evaluation of wellbore stability and

mud weight design, for sand prediction studies, for completion

optimisation, and are also a key element in the estimation of
reservoir compaction and surface subsidence.
As a result, there is a trend towards more use of
measurements that can characterize directly the minimum and
maximum horizontal formation stresses (magnitude and
directions), and the directional rock elastic properties. In the
present work, we developed enhanced Mechanical Earth
Models (MEM) that were based on horizontal stress ratios
estimations for the undisturbed rocks in the far-field, on stress
changes for the altered rocks in the near-wellbore area; and
which contained rock elastic properties that were characterized
isotropically, vertically and horizontally.
In the following, we present the methods used throughout
this work, the data sets, the difference between isotropic and
anisotropic rock elastic properties (deformation and strength),
and the stress estimations. We then proceed with an evaluation
of the results against conventional methods and wellbore
stability events.
Data acquisition and processing
Open hole sonic data were obtained in several wells of the
Kvitebjorn Field between 4200m and 4400m measured depth,
using Schlumberger's advanced sonic tool (Pistre et al., 2005.
In our field example well "A" the well is directed towards NE
with a deviation of 20o from vertical.
Compressional-, Shear-, Stoneley-, and flexural-wave
slownesses were obtained at varying radial formation depth
due to the short- and long- spaced transmitter-receiver
geometry with monopole and dipole sources. Flexural waves
were generated using orthogonally oriented dipole transmitters
for characterization of shear wave slowness. From these
measurements, we directly obtained slownesses in the planes
along the wellbore axis, and in the plane orthogonally oriented
to the wellbore axis. In addition, dispersion (slowness versus
frequency) data were derived for slightly dispersive Stoneley
waves, and highly dispersive flexural arrivals.
Thorough quality control and Best-Delta-Time processing
were performed on the sonic data for ensuring accuracy of the
slowness curves along the wellbore trajectory. Multishot
processing for higher data resolution and advanced processing
necessary for enhanced geomechanics applications were
carried out along the entire length of the measurements. The
advanced processing included (1) shear anisotropy processing
for fast and slow shear slownesses and directions based on a
four-component Alford rotation (Alford, 1986); (2) dispersion
analyses for characterizing the type of rock (an)isotropy; (3)
3D anisotropy processing of the flexural wave for estimating
the shear moduli C44 and C55 in the vertical planes, and of
the Stoneley wave for estimating shear modulus C66 in the
horizontal plane. Shear modulus C33 was derived from the
compressional wave; and additionally (4) dipole radial
profiling was carried out for near wellbore-alteration
characterization. Further reading on advanced sonic data
acquisitions, processing and interpretation can be found in e.g.
Haldorsen et al. (2006), and Mercado Ortiz et al. (2006).

SPE 109662

Slowness-time-coherences of good quality were obtained

for all wave modes (Figure-1) in the applied frequency filter
range. Generally, the high signal to noise ratio of the sonic
data, allows for dispersion analyses which are necessary for
the determination of the type of rock anisotropy (intrinsic or
stress induced). In such analyses, slownesses in the low
frequency area represent the far field, while slownesses in the
high frequency area represent the near wellbore environment.
This approach allowed us to distinguish homogeneous
isotropic and anisotropic materials; and inhomogeneous
isotropic and anisotropic materials: (1) in the homogeneous
isotropic case, no shear wave splitting occurs. This is a very
stable situation; (2) in the homogeneous anisotropic case, the
fast and the slow shear slownesses are different in magnitude.
This is an intrinsic material anisotropy due to layering and/or
fractures; (3) in the inhomogeneous isotropic case, no shear
wave splitting occurs, but slownesses are higher than for case
(1). This can be a result of elasto-plastic near-wellbore
alteration; and (4) in the inhomogeneous anisotropic case,
shear wave splitting exists (at least) in the near wellboreenvironment: The slow shear slowness becomes faster than the
fast shear slowness towards the near wellbore area. This is a
result of the differential interaction between hoop stresses and
far field stresses, and is typical for a stress-induced anisotropy.
In Figures-2a-d, we show typical results from dispersion
analyses, and stress induced anisotropy cross-over dispersion
patterns as observed in the reservoir sands of well "A" in the
Kvitebjorn field.
From the rotated waveforms, energy-, time-, and slownessbased anisotropy were obtained together with the fast and slow
shear formation slownesses, and the fast shear azimuth.
Figure-3 shows the result for several depth intervals. The
dispersion analyses clearly indicated stress induced
anisotropy. The fast shear azimuth direction derived
corresponded to NE65. Based on the relationship between
rock stiffness and wave propagation speed (e.g. Mavko et al.,
1998), we assume that NE65 represents the present day
maximum horizontal stress direction.
The results of the 3D anisotropy processing for derivation
of the shear moduli C33, C44, C55 and C66 are shown on Figure4. C44, C55, and C66 denote shear rigidity in the three
orthogonal planes of the well "A" rock formations, and change
as a function of lithology and/or mineralogy. C33 describes
shear rigidity along the borehole axis. The results of the shear
moduli analyses were subsequently used for estimating
anisotropic rock deformation and strength properties.

Estimation of deformation and strength properties

The sonic data processing described above provided the
necessary input for further calculations. Isotropic dynamic
deformation rock properties were calculated as a function of
compressional slowness (DTCO), fast shear slowness (DTS),
and density (). Static properties were obtained by correlating
them to empirical data from the North Sea, and to core test
results from this and offset wells. Isotropic rock strength
estimations were performed using static Young's Modulus
correlations, and equally results from laboratory core tests.

SPE 109662

Anisotropic dynamic deformation rock properties were

calculated based on the assumption of a transversely isotropic
(TI) medium (x and y directions in the horizontal plane display
the same properties) with the axis of symmetry oriented
parallel to the wellbore. For such a case, the elastic stiffness
matrix describing the rock properties takes the form as
presented below (Figure-5). For a complete property
characterization of a TI medium, five independent stiffness
tensors Cij are needed.

Figure-5: Transversely isotropic stiffness matrix

From the five independent Cij, three can be directly

obtained from the advanced sonic measurements (C33,
C44(=C55), C66, Figure 4); the two remaining stiffness tensors
C11, and C12 can be calculated based on the ANNIE
approximation of Schoenberg and Sayers (1995) which
describes a three parameter family of TI media by a 6x6
density-normalized elastic stiffness matrix (see also Bikash et
al., 2004). Anisotropic rock strength properties were then
derived as a function of directional Young's modulus.
Static deformation and strength data were subsequently
obtained by relating the dynamic measurements through linear
regression to core measurements from offset wells and well
"A".
Estimation of formation stresses using sonic data
The underlying theory behind the estimation of formation
stresses using borehole sonic data is based on acoustoelastic
effects in rocks. Acoustoelasticity in rocks refers to changes in
elastic wave velocities caused by changes in pre-stress in the
propagating medium.
Elastic wave propagation in a prestressed material is
described by equations of motion for small dynamic fields
superposed on a statically deformed state of the material.
These equations can be solved by various methods including
the perturbation technique and finite-difference formulation.
Here, we used the changes in the borehole guided wave
dispersions in the presence of near-wellbore stress
concentrations (Sinha and Kostek, 1996; Liu and Sinha, 2003).
It has been demonstrated earlier that crossing dipole
dispersions in a vertical well are indicators of stress-induced
shear slowness anisotropy caused by differences in the
maximum and minimum horizontal stresses (Sinha and
Kostek, 1996; Winkler et al., 1998), and that measured crossdipole dispersions can be inverted to obtain radial variation of
the fast and slow shear slownesses (Sinha et al., 2003: Sinha,

2004; Sinha et al., 2006). The radial variations of the fast and
slow shear slownesses can be transformed into radial profiles
of the two shear moduli C44 and C55 that also exhibit a
crossover in the near-wellbore region (Liu and Sinha, 2003;
Bratton et al., 2004). Similarly, the lowest-order axisymmetric Stoneley dispersion can be inverted to obtain a
radial profile of the third shear modulus C66.
In nonlinear continuum mechanics, an acoustoelastic
model relates changes in the plane wave velocities
(compressional, Stoneley) in rocks to corresponding changes
in the static stresses in the propagating medium (Norris et al.,
1994). We generalized this assumption in order to relate
differences in the three principal stresses to corresponding
shear velocities (or equivalently, shear moduli), and their
changes in the presence of near-wellbore stresses (Sinha,
2002). Estimations of maximum and minimum horizontal
stress magnitudes were based on solving four nonlinear
equations. These equations defined an acousto-elastic
coefficient by relating far-field maximum and minimum
horizontal stress magnitudes to two nonlinear rock constants
(C144, and C155).
Two equations related differences in the far-field effective
horizontal stress magnitudes to the acoustoelastic coefficient,
using an effective overburden stress term (for a given depth).
(The overburden stress is easily obtained by integrating
formation bulk density from surface to depth (of interest) by
taking the gravity constant into account). The other two
equations related differences in the shear moduli at two radial
positions to corresponding differences in the radial, hoop, and
axial stresses at the same depth.
Subsequently, radial distributions of near-wellbore stresses
were determined from Kirschs equations as a function of the
far-field principal stresses. The four nonlinear algebraic
equations could then be solved for the four unkowns: Hmax,
hmin, C144, and C155.
Note, that radial positions close to the borehole interface
might exhibit plastic deformation of the rocks, if stress
concentrations exceed the rock yield strength. However, since
the plastically yielded area around the borehole is not well
quantified, a region of one wellbore radius around the
borehole is not taken into account for shear moduli derivation.
A more detailed description of stress estimations from
dispersion analyses can be found in Sinha et al. (2007).
Dispersion based stress estimations where performed for
three reservoir sand sections in the Brent Group. From the
calculations, based on the dispersion analyses shown in
Figures 2a-c, it became obvious that total stresses are in a
close-to-isotropic state, with effective stress ratio variations
between 4% and 10%. In Table-1 we show the results for the
three depth frames from one well in the Kvitebjorn Field.

SPE 109662

Depth
(m)

H/
V

H/
V

H
(MPa

h
(MPa

V
(MPa

4303.23

0.90

1.09

87.00

86.20

87.90

4319.50

0.90

1.06

87.00

86.00

88.30

4330.36

0.95

1.05

87.60

86.90

88.60

Table-1: Effective stress ratios and absolute stresses in the

reservoir

Enhanced Mechanical Earth Modelling (MEM)

It has proven critical to present-day capability of creating
Mechanical Earth Models (MEM) that contain elastic moduli
information (with static equivalents), formation strength, and
stress predictions (e.g. Plumb et al. 2000).
The enhanced capability to create rock mechanical
properties and stresses from three-dimensional sonic
measurements and interpretations is a new concept, and their
integration with the full suit of conventional log-data, in-situ
and
laboratory-based
measurements
enables
the
characterization of inhomogeneous anisotropic rock
formations in space. This provides more accurate (1) safe mud
weight window design; (2) well integrity projections; and (3)
it even can distinguish between the altered near wellbore and
undisturbed far field rock characteristics, important for
completion and perforation optimisation.
Kvitebjorn field example well "A" is situated towards the
center East of the Kvitebjorn field. The well was drilled for
gas and condensate production from the Brent Group, starting
at a measured depth of 4051.6mTVD. Drilling was performed
using a high density Cs/K formate mud (in average 2.04g/cm3)
in the reservoir section. During reservoir drilling, seepage
losses were experienced, but no formation breakdown or
breakouts were observed. In shale above and below the
reservoir some washouts and drilling induced fractures were
identified.
Leak-off, formation integrity and minifrac tests were
performed throughout the field, and provided consistent
constrains on minimum horizontal stress magnitudes (Table
1). Pore pressure gradients were estimated for the overburden
depending on lithology, and they were constructed using MDT
measurements in the reservoir section for initial and slightly
depleted pressures in well "A". Oriented cores were taken in
offset wells and in well "A" in the Brent Group. Static
properties calculated from core deformation and strength
testing were subsequently used for correlating dynamic in-situ
properties to static rock properties.
We calculated a homogeneous isotropic and an
inhomogeneous anisotropic Mechanical Earth Model (vertical
and horizontal parameter input) for the well "A" reservoir
section. Necessary input for the isotropic model were gamma
ray (GR), DTCO, DTSfast, density, pore pressure, and
calibration points based on leak-off, formation integrity and
minifrac tests. Necessary input for the anisotropic model were
GR, and the shear moduli C11, C12, C33, C55, and C66, density,

pore pressure, and calibration points based on leak-off,

formation integrity and minifrac tests. Effective stress ratios
and absolute stresses were taken as from the constrained three
depth frames in sands of the Brent Group (described above).
We furthermore calculated the change in far field shear
modulus (G) and stresses, compared to near wellbore shear
moduli and stresses based on the DRP, and results from the
Enhanced Mechanical Earth Model.

Homogeneous isotropic and inhomogeneous anisotropic

rock properties. The conventional homogeneous isotropic
model provided one data point for each property along the
wellbore trajectory. Poisson ratio () (lateral versus axial
strain), Young's modulus (E) (uniaxial stress versus normal
strain), bulk and shear moduli (K, G) (K = hydrostatic pressure
versus volumetric strain; G = applied stress versus shear
strain) were calculated using the standard functions as shown
in equations 1 to 4 with Vc - compressional velocity, Vs shear velocity, - density. Note slowness is the inverse of
velocity. Equations after Mavko et al. (1998).

(Vc2-2Vs2)
2(Vc2-Vs2)
2
2
Vs2 (3Vc -4Vs )

(1)
(2)

(Vc2-Vs2)

2
2
= (Vc -4/3Vs )

(3)

G = Vs

(4)

In the inhomogeneous anisotropic approach, we assumed

that the rocks were transverseley isotropic with five
independent elastic stiffness coefficients Cij (Figure-5). Based
on Cij, one vertical (E33) and one horizontal (E11) Young's
modulus, together with three Poisson's ratios (31(vertical),
12, 13 (horizontal)) were calculated for each depth point
based on equations 5 to 9. Bulk modulus (K) and shear
modulus (G) can then be derived using conventional
expressions.
E33 = C 33

2C132
,
C11 + C12

(5)

E11 = C11 +

C132 (C11 + C12 ) + C12 (C 33C12 + C132 )

,
C11C 33 C132

(6)

31 = 32 =

C13
,
C11 + C12

(7)

13 =

C13 (C11 C12 )

,
C33C11 C132

(8)

12 =

C33C12 C132
,
C33C11 + C132

(9)

SPE 109662

The results for both property model types are shown in

Figures-6, -7. We display a comparison between the isotropic
and each component of the anisotropic calculations together
with GR along the wellbore trajectory in the reservoir section.
The differences between the property curves are highlighted in
color. From the figures, it is clear to see that (1) there are
small differences between the isotropic and the vertical
Poisson's ratios and the vertical Young's modulus; and (2)
there are large differences for the isotropic and the horizontal
Poisson's ratios and Young's moduli.
In details, the following observations were made. (1)
Poisson's ratio: Above the reservoir, and in the clay dominated
layers in the reservoir, the vertical Poisson's ratio (31) is
mostly smaller than the isotropic Poisson's ratio. In the sandy
reservoir layers, the vertical (31) and the isotropic Poisson's
ratios are nearly equal. However, in some areas along the
wellbore, the vertical Poisson's ratio is higher than the
isotropic Poisson's ratio. This seems to occur mostly, where
shaly or sandy layers contain other rock material (e.g. coal).
(2) Young's modulus: Above the reservoir, and in the clay
dominated layers in the reservoir, the vertical (33) Young's
modulus is mostly equal to the isotropic Young's modulus. In
the sandy reservoir layers, the vertical Young's modulus (33)
is slightly lower than the isotropic Young's modulus.
However, in some areas along the wellbore, the vertical
Young's modulus (33) is higher than the isotropic Young's
modulus. This seems to occur mostly, where shaly or sandy
layers contain other rock material (e.g. coal).
The differences in rock property values using a
homogeneous isotropic and an inhomogeneous anisotropic
approach demonstrate, that the isotropic approach reflects
mostly the properties of low clay sand materials. Lateral
variations in rock properties, and variations due to layering,
are not accounted for.
Continuous stress magnitudes. Continuous stress
calculations for the undisturbed far field formations parallel to
the wellbore path were performed using a Mohr-Coulomb
approach based on overburden stresses (V), pore pressure
profiles and friction angles. The Biot elastic constant was
taken equal to one in all calculations.
Overburden stress was conventionally calculated by
integrating density over depth taking into account the gravity
constant. The reservoir pore pressure profile was constructed
using in-situ pore pressure points from exploration drilling that
exhibited initial pore pressures, and pressure points directly
taken in the reservoir section of well "A". Depleted pore
pressures were in average 3MPa lower than the initial pore
pressures of 77MPa, thus representing a depletion of less than
4% in average. The friction angle was calculated as a linear
function of clay content of the reservoir rocks (Plumb, 1994),
and checked against core measurements and empirical North
Sea values.
Mohr Coulomb stress boundaries where calculated for both
pore pressure senarios (initial, depleted), and then used as the
vertical constraint for the stress development with increasing

depth. An estimated absolute value for the initial minimum

horizontal stress magnitudes (h) (before depletion) was
obtained by assuming that leak-off tests (LOTs) and formation
integrity tests (FITs) performed on shales in other wells, and
in well "A" represented h or the lowest limit of h (FIT) for
individual depth points. These pressure points were then
vertically correlated with respect to the Mohr Coulomb stress
boundaries, e.g. the trend of the averaged lower and upper
Mohr-Coulomb boundary was projected on the positions of the
LO tests and FIT tests in a depth-pressure (or stress) plot
(Figure 8a)
Subsequently, the same procedure was used for calculating
a h profile along the wellbore that would honor the pore
pressure depletion in the reservoir sands, and the depleted
minimum horizontal stress magnitude obtained from advanced
stress calculation (see above). We calibrated these calculations
with a minifrac test undertaken in the depleted reservoir sands.
A summary of the results is shown on Figures-8a,b.
The total stress ratios derived from the dipole dispersion
analyses revealed that reservoir total stresses where nearly
isotropic for the three depth points investigated at
4303.32mMD, 4319.5mMD, and 4330.36mMD. We used this
information to constrain the formerly continuous minimum
horizontal stress (H) profile along the wellbore path. Then,
the total stress ratio of h: H =0.99:1 was applied to the
minimum horizontal stress, respecting the relationship of
H: V = 0.99:1 to obtain the continuous maximum horizontal
stress curve. The calibration results are presented in Figure-8b.
The consistency of the minimum horizontal stress
calculations and the calibration points clearly demonstrate the
validity of the advanced sonic approach. Furthermore, the
sonic approach allows us to quantify maximum horizontal
stress profiles along the wellbore path.
Evaluation of maximum horizontal stress azimuth. The
four-component Alford rotation of the crossed-dipole shear
waveforms, revealed that the maximum horizontal stress
direction is oriented in NE-SW direction (NE65o) (Figure-3).
We tested this information against interpretations of induced
fracture sets (orientations and dips) derived from image logs,
and the overall recent stress field as published in the world
stress map. In Figure-9 we show the fracture analyses for 4
wells. The strike directions of the induced fractures align in
average in E-W direction (the green arrow points towards the
North), and are thus compatible with a NE-SW oriented H.
This result is equally compatible with other data from the
North Sea, that show Kvitebjorn positioned between two
maximum horizontal directions with a 15o variation around
E-W (Statoil, internal report; World Stress Map).

Near wellbore environment

Pressures exerted by the drilling fluid shall compensate the
main stresses acting upon the rock formation at the wellbore
interface. This pressure is variable as frictions factors and fluid
density changes during the drilling process (e.g. pump rate,

hole size, Surge, Swap). This results in a continuous change of

stress around the wellbore during the drilling process that
leads to near wellbore alteration (e.g. deformation, failure).
The phenomenon has been know for many years (e.g Gauer et
al., 1988), and is suspected to generate a zone of plastically
deformed rocks around the wellbore wall.
The dipole radial profiling is a sonic characterization of
the extent and the intensity of the alteration that occurs
azimuthally in the near wellbore environment induced by the
drilling process. The alteration decreases with radial distance
from the borehole towards the undisturbed rock formation. It
can be detected by a continuous change in slowness from the
far field (> 5 wellbore radii) towards the near field. On Figure
-10 we show an example from the Kvitebjorn field. It is clear
to see that slownesses increase in sand intervals, and decrease
in shale intervals in the near wellbore environment compared
to the far field. Figure-11 shows the results for wellbore stress
and shear moduli calculations based on 6 dipole radial profiles
for reservoir sands at 4283m, 4284, 4300m, 4301m, 4319.5m
and 4321.5m.
For each of the 6 depth frames, near wellbore stresses and
dynamic shear moduli were calculated using Kirsch's
equations and the conventional sonic based approach for shear
modulus calculation (see equation 4) (e.g. Bratton et al.,
2004). The calculations revealed that effective tangential
stresses increased to 14.48 MPa, and effective radial stresses
decreased to 5.1MPa close to the wellbore wall. They align
with the horizontal stresses in the far field. The deviation of
the effective horizontal stresses from undisturbed formation
values occured at about 2.5 wellbore radii depth into the
formation. Accordingly, shear moduli decreased, and probable
onset of yielding occured at the same radial formation depth.

Wellbore Stability Calculations

Wellbore stability calculations were performed for testing
the validity of the calculated stresses and rock properties
against observations in the wellbore. Losses were reported in
the daily drilling report at the top of the 8in section.
Formation breakouts or breakdowns were not reported.
However, the measured borehole caliper showed irregularities
(hole enlargments), and these were the best control for the
stability events.
We performed two types of wellbore stability calculations
for the sligthly depleted pore pressure szenario, with decreased
total stresses (see Figure-8b): In the first conventional
wellbore stability type (Figure-12a), we included the isotropic
rock properties and the stress magnitude profiles and
directions as described above. In the second advanced
wellbore stability type, we peformed two calculations which
were each based on a different directional rock property data
set (vertical and horizontal, (Figures-12b,c), and the stress
magnitude profiles and directions described above.
From the results presented in Figures-12a-c, it is easy to
see that the variations between horizontal and vertical stresses,
and rock strengths and Poisson's ratio between the vertical and
the horizontal directions where too small for producing visibly

SPE 109662

different effects on the wellbore stability. Therefore, we

performed two calculations with hypothetical parameters. In
the first calculation, the rock strenght between vertical and
horizontal direction varied of about 30% (Poisson's ratio was
kept constant). In this case, we were able to see differences in
the intensity of wellbore events occurring. However, vertical
positions of the incidents along the wellbore (hole enlargment)
remained the same (Figure-13a). In the second calculation, the
difference between horizontal stresses and the overburden
stress were increased from 1% to 4% (equals 3.5 MPa at top
reservoir level). This induced a similar change in wellbore
damage compare to the increased strength test (Figure-13b).
Therefore, we suggest that the enhanced MEM together
with the wellbore stability calculations based on isotropic and
directional data sets, demonstrate that the isotropic stress
distribution in the far field seems to be the most controlling
factor of stability in the Brent reservoir section of the
Kvitebjorn field (where rock layering is confined to relatively
small vertical sections). However, we expect that changes in
stress and strain also induce changes in Poisson's ratio and
Young's modulus.

Discussion and Conclusion

Advanced Mechanical Earth Modelling provided rock
property data sets, which quantify differences in rock elastic
properties between the vertical and the horizontal directions.
Furthermore, it became obvious from the sonic based stress
calculations and the in-situ measurements that the far field
stress situation is close to isotropic, with the maximum
horizontal stress being oriented in NE-SW direction. Wellbore
stability analyses demonstrated that under the given conditions
of stress isotropy and variations of 10% in vertical and
horizontal rock properties, wellbore events will be entirely
controlled by the stress distribution. Increasing differences in
rock strength (30%) between the horizontal and the vertical
directions or increasing differences between the vertical and
horizontal stresses (4%) would induce significant changes in
the intensity of wellbore incidents.
The situation analysed here, represented an early stage of
depletion, and can be regarded as a close-to-initial state of the
field. In HPHT normal faulted environments, as the Kvitebjorn
field, it is expected that small horizontal differential stresses
exist initially, and that large amounts of depletion can occur
rapidly through production. Uniform depletion of the field
would suggest uniform changes in the magnitudes of the
horizontal principal stresses (e.g. decrease in total stress,
increase in effective stress) without any change in stress
direction. However the Kvitebjorn field is characterized by
faults and sealing boundaries in the reservoir (e.g. Brent
reservoir shales with thickness greater than 1m; thick shales
between Cook formation and Rannoch formation). These
features can cause differential depletion in the reservoir layers,
and therefore cause the horizontal principal stresses to deplete
differentially, and to reorient in space (e.g. Wright and
Weijers, 2001; Day-Lewis and Zoback, 2007). Furthermore,
the depletion of deep HPHT reservoirs is typically
accompanied by an increase in effective vertical stress of

SPE 109662

several 100MPa. Most reservoir sandstone cannot withstand

such high stress increase without undergoing considerable
inelastic deformation of the load bearing grain framework
(Schutjens, 2005). Stress and strain will be redistributed to the
surrounding formations, resulting in changes in Poisson's
ratios, Young's moduli, in thickness, and fault activity (e.g.
Hawkins et al., 2006).
The workflow and the type of data set that we presented in
this work enable us to control the three-dimensional stress
state, and their vertical and horizontal rock properties
throughout the life of a field. The initial Kvitebjorn data
provide an excellent basis for forecasting potential time-lapse
stress and property changes during depletion, which is critical
information in any HPHT field development. Changes in
stress state and/or rock properties, and thus wellbore and field
stability, can be supported by repeat data logging, and 4Dfoward numerical modelling.

Acknowledgement
We would like to thank Statoil for the permission to
publish this work. It has been carried out with continuous
support from John Fuller, Ken Russel, Rob Marsden, and
Vivian Pistre, all part of Schlumberger Data and Consulting
Services, and Schlumberger Wireline. The authors would like
to thank them all. We wish also to thank Tom Bratton for
helping with near-wellbore calculations, Katrine Andersen,
who made the fracture plots, Alisdair McDowell, who helped
with the petrophysics plots, and Neil M. Stevenson for
ongoing patience, all part of Schlumberger Data and
Consulting Services.

References
Alford, R.M. 1986. Shear Data in the Presence of Azimuthal
Anisotropy: Dilley, Texas presented at the 56th SEG Annual
International Meeting, Houston, 2-6 November. Expanded
Abstracts: 476-479.
Bratton, T., Bricout, V., Lam, R., Plona, T., Sinha, B.K.,
Tagbor, K., Venkitraraman, A. and Borbas, T. 2004. Rock
Strength Parameters from Annular Pressure While Drilling
and Dipole Sonic Dispersion Analysis presented at SPWLA,
Annual Logging Symposium, 6-9 June.
Day-Lewis, A. and Zoback, M. 2007. Depletion-Induced
Stress Orientation Changes in Bounded Reservoirs presented
at the AAPG Annual Convention and Exhibition, Long Beach,
1-4 April.
Gauer, P., Janot, P., Baudemont, D. and Ruhland, M. 1988.
Borehole Breakout. Bibliographic Synthesis. Revue de
l'institut francais du ptrole (43), 3, 389-403.
Haldorsen, J.B.U., Johnson, D.L., Plona, T., Sinha, B., Valero,
H_P., and Winkler, K. 2006. Borehole Acoustic Waves.
Oilfield Review 18 (1): 34-43.
Hawkins, K., Harris, P. and Conroy, G. 2006: Estimating
Production-Induced Stress Changes from 4D Finite Offset
Timeshifts presented at 68th EAGE Conference & Exhibition
incorporating SPE Europec.

Liu, Q.H., and B.K. Sinha, 2003, A 3D cylindrical

PML/FDTD Method for Elastic Waves in Fluid-Filled
Pressurized Boreholes in Triaxially Stressed Formations:
Geophysics (68): 1731-1743.
Mavko, G., Mukerji, T., and Dvorkin, J. 1998. The Rock
Physics Handbook, first edition, 51-53. Cambridge University
Press, Cambridge.
Mercado Ortiz, J.L.A.F.M.A., De, G.S., Renlie, L., and
Williams, S. 2006. Sonic Investigations In and Around the
Borehole. Oilfield Review 18 (1): 14-33.
Norris, A.N., B.K. Sinha, and S. Kostek, 1994.
Acoustoelasticity of Solid/Fluid Composite Systems: Geophys.
J. Internat. (118): 439-446.
Pistre, V., Kinoshita, T., Endo, T., Schilling, K., Pabon, J.,
Sinha, B., Plona, T., Ikegami, T. and johnson, D. 2005. A
Modular Wireline Sonic Tool for Measurements of 3D
(Azimuthal, Radial, and Axial) Formation Acoustic Properties.
Transactions of the SPWLA, 46th Annual Logging Synposium,
New Orleans, June26-29.
Plona, T., Kane, M., Alford, J., Endo, T., Walsh, J. and
Murray, D. 2005. Slowness-Frequency Projection Logs: A
New QC Method for Accurated Sonic Slowness Evaluation
presented at SPWLA, 46th Annual Logging Symposium: 2629.
Plumb, R.A. 1994. Influence of Composition and Texture on
the Failure Properties of Clastic rocks. SPE 28022: 13- 20.
Plumb, R.A., Edwards, S., Pidcock, G. and Lee, D. 2000. The
Mechanical Earth Model Concept and Its Application to HighRisk Well Construction Projects. SPE 59128.
Ray, P.L., Goodman, H.E. and Perrin, V.P. 1998. Application
of rock mechanics to HPH and horizontal well drilling and
completion: A case study presented at 6th international oil
and gas conference and exhibition China, Beijing, 2-6
November.
Schutjens, P. 2005. Production-Induced Deformation of
Sandstone from a deep HP/HT-Reservoir: Laboratory
Experiments and Sand Production Modeling presented at Sand
Control and Management US, Houston.
Sinha, B.K., and S. Kostek, 1996. Stress-Induced Azimuthal
Anisotropy in Borehole Flexural Waves: Geophysics (61)
1899-1907.
Sinha, B.K., 2002, Determining stress parameters of
formations using multi-mode velocity data: U.S. Patent 6 351
991.
Sinha, B.K., Burridge, R. and Kane, M.R. 2003. Sonic Well
Logging for Radial Profiling: U.S. Patent 6 661 761 B2.
Sinha, B.K. 2004. Near-Wellbore Characterization Using
Radial Profiles of Shear Slownesses presented at 74th SEG
Annual International Meeting, Expanded Abstracts: 326-329.
Sinha, B.K. and Endo, T. 2004. Determination of Anisotropic
Moduli of Earth Formations. U.S. Patent 6 774 480.
Sinha, B.K., Vissapragada, B., Renlie, L. and Tysse, S. 2006.
Radial Profiling of the Three Formation Shear Moduli and its
Application to Well Completions: Geophysics (71): 65-77.
Sinha, B.K., Vissapragada, B., Wendt, A.S., Kongslien, M.,
Eser, H., Skomedal, E., Renlie, L. and Pedersen, E.S. 2007.
Estimation of Formation Stresses Using Radial Variation of
the Three Shear Moduli in a Well - A Case Study from a
High-Pressure and High-Temperature Field in the Norwegian
North Sea. SPE 109842.

Winkler, K.W., Sinha, B.K. and Plona, T.J. 1998. Effects of

Borehole Stress Concentrations on Dipole Anisotropy
Measurements: Geophysics, (63): 11-17.
Wright, C.A. and Weijers, L. 2001. Hydraulic fracture
reorientation: Does it occur? Does it matter? The Leading
Edge (20), 10, 1185- 1189.

SPE 109662

SPE 109662

Figure-1: Slowness-time-coherency curves (Slowness-time-projection receiver (STPR), slowness-frequency analysis receiver

(SFaR)). Further explanation under Plona et al., (2005).

Failure

Vs

Shear

Inhomogeneous Isotropic
Damaged,
Vs(r)
Near
Failure,
Or
Alteration
Frequency

Frequency

Vs()

Frequency

fast shear
slow shear

Inhomogeneous Anisotropic
S lo w n e s s

S lo w n e s s

Homogeneous Anisotropic
Intrinsi
c:
Shales,
Fractur
es

Stoneley

S lo w n e s s

Far
from

S lo w n e s s

Homogeneous Isotropic

Vs(r,)

Stress
Induced

cross over

Frequency

Stoneley

Stoneley

fast shear
slow shear

fast shear
slow shear

C
C

cross over

cross over

Figure-2a-d: Dispersion plots for three depth frames in the reservoir section. The dispersion analyses show the typical crossover of the fast and the slow shear waves in the near wellbore environment, when subjected to stress (stress-induced
anisotropy).

10

SPE 109662

Figure-3: Derivation of fast shear azimuth based on rotated waveforms, shear wave splitting and dispersion analysis.

Figure-4: 3D anisotropy analyses providing the shear moduli C33, C44, C55, and C66.

SPE 109662

11

Figure-5: Transversely isotropic stiffness matrix based on five independent stiffness tensors.

Figure-6: Comparison between elastic staticPoisson's ration (PR) and Young's modulus (E) for isotropic and
directional calculations. All dynamic isotropic curves (PR_dyn, E_dyn) are shown in red. The tracks show from left
to right the following comparisons: PR_dyn: PR31,PR_dyn:PR12, PR_dyn:PR13, E_dyn:E33, E_dyn:E11. Poisson's
ratios are plotted between 0 and 0.5; Young's moduli are plotted between 0GPa and 60GPa.

12

SPE 109662

Young's moduli

GPa

Young's moduli

30 0
E_isotropic
E_33

GPa

Poisson's ratio

30 0.1
E_isotropic
E_11

0.4
PR_isotropic
PR_13
PR_31
PR_12

Figure-7: Comparison between elastic static properties for isotropic and directional calculations.

Figure-8a: Initial and depleted pore pressure situations from surface to TD in well "A". Further explanations see text.
PPRS - pore pressure, MC_LL - Lower Mohr Coulomb Limit, Sig - stress, h-minimumhorizontal, H - maximum
horizontal, V - vertical.

SPE 109662

Figure-8b: Initial and depleted pore pressure situationsfor the reservoir section in well A. In-situ calibration points,and
calibration points from advance d stress calculations are given.

Figure-9: Azimuth of induced fractures in 4 wells. The green arrow points towards North. Right: In situ image of induced fractures.

13

14

SPE 109662

Figure-10: Dipole radial profile demonstrating that wellbore alteration occurs of up to 2 borehole diameter. Green
represents the far field slowness, any other color represents a deviation from the far field slowness. Colors which
increase towards red on the color bar show an increase in slowness; colors which decrease towards blue on the color
bar show a decrease in slowness.

Figure-11: Calculation of pseudo tangential (pt) and pseudo radial (pr) effective wellbore stresses and shear
modulus G based on the dipole radial profile slownesses for the depth interval shown in Figure-10.

SPE 109662

Figure-12a-c: Wellbore stability calculations base on static elastic properties. (a) isotropic parameters; (b) vertical
parameters; (c) horizontal parameters. Explanation see text. CALI - caliper.

Figure-13a-b: Wellbore stability calculations. (a) stress test: the difference between horizontal and vertical stress was
increased to 4% (from initially 1%); (b) strength test: the difference between vertical and horizontal rock strength was
increased to an average of 30%. Explanation see text.

15

16

SPE 109662

Depth
(m)

H/
V

H/
V

H
(MPa

h
(MPa

V
(MPa

4303.23

0.90

1.09

87.00

86.20

87.90

4319.50

0.90

1.06

87.00

86.00

88.30

4330.36

0.95

1.05

87.60

86.90

88.60

Table-1: Effective stress ratios and absolute stresses in the reservoir