IPA01-G-021

PROCEEDINGS, INDONESIAN PETROLEUM ASSOCIATION

Twenty-Eighth Annual Convention & Exhibition, October 2001
FAULT SEAL INTEGRITY IN THE TIMOR SEA AREA: PREDICTION OF TRAP
FAILURE USING WELL-CONSTRAINED STRESS TENSORS AND
FAULT SURFACES INTERPRETED FROM 3D SEISMIC
D. A. Castillo*
D. J. Bishop**
M. de Ruig**

ABSTRACT
Drilling in the Laminaria High and Nancar Trough
areas has shown that many hydrocarbon traps are
underfilled or completely breached. Previous studies
have shown that fault-trap integrity is strongly
influenced by the state of stress resolved on the
reservoir bounding faults, suggesting that careful
construction of a geomechanical model may reduce
the risk of encountering breached reservoirs in
exploration and appraisal wells. The ability of a fault
to behave as a seal and support a hydrocarbon column
is influenced in part by the principal stress directions
and magnitudes, and fault geometry (dip and dip
azimuth). If a fault is critically stressed with respect
to the present-day stress field, there is a high
likelihood that the fault will slip, thereby re-creating
fault zone permeability that enables hydrocarbons to
leak. Leakage could be intermittent depending on the
degree and rate of fracture healing, and on the
recurrence rate between reactivated slip events.
High-resolution wellbore images from over 15 wells
have been analyzed to construct a well-constrained
stress tensor. Constraints are based on geomechanical
parameters, along with drilling conditions that are
consistent with the style of drilling-induced
compressive and tensile wellbore wall failure seen in
each of these wells. This regional stress analysis of
permits AC/P8, AC/P16, Zone of Cooperation and
surrounding areas indicates a non-uniform strike-slip
stress regime (SHmax > Sv > Shmin) with the
orientation of the maximum principal horizontal
stress (SHmax) varying systematically from N-S
compression in the northern reaches to NE-SW
farther south.
___________________________________________________________
*
GeoMechanics International Inc.
** Woodside Energy Ltd.

Fault surfaces interpreted from 3D seismic data have

been subdivided into discrete segments for the
purpose of calculating the shear and normal stresses
in order to resolve the Coulomb Failure Function
(CFF) on each fault segment. Hydrocarbon
accumulation (column height) and leakage (residual
column) deduced from well results may be explained
in part by the CFF resolved on their respective
reservoir-bounding faults. By integrating the
geomechanical
model
with
fault
imaging
technologies, explorationists and reservoir engineers
will gain the ability to use these predictive tools to
help quantify the likelihood of encountering a
breached reservoir prior to drilling.
INTRODUCTION
Drilling in the AC/P8, AC/P16 and adjacent blocks
within the Timor Sea area has shown that many
reservoir fault traps have experienced hydrocarbon
leakage. Following the discoveries of the Laminaria
and Corallina Fields, exploration activity within the
greater AC/P8 and AC/P16 areas (including WA-260P and ZOCA 91-01) has produced a commercial
success ratio of 13% and an oil discovery rate of 35%.
A majority (67%) of the unsuccessful exploration
wells in the area show evidence of a residual or
palaeo hydrocarbon column, indicating trap breaching
as the main cause of exploration failure and subcommercial finds. In every case within the Timor Sea
area, the sealing integrity of these structural fault
traps depends largely upon a top seal of sufficient
quality and thickness to support the buoyancy forces
exerted by the hydrocarbon column, and the capacity
of the reservoir-bounding fault(s) to behave as an
impermeable seal or baffle to thwart lateral fluid
migration or vertical migration along the fault.
Vol. 1 - 261

Previous in-house studies within Woodside, Shell and

BHPP (De Ruig et al., 2001) have conducted mercury
injection capillary pressure tests on cap rock samples
within the Timor Sea area, and have found that
hydrocarbon column heights approaching several
hundred metres can be supported by the cap rock.
Although, this does not rule out the possibility that
critically-stressed natural fractures within the cap
rock have not breached the reservoir, detailed fracture
analysis using borehole image data from the Timor
Sea area indicates a relatively sparse population of
these natural fractures.
Alternatively, the faults with high trap integrity are
likely to be those associated with seals whose finegrained material, formed in grain-reduction process
during the fault slip history, remains intact or
unbroken.
Assuming hydrocarbon source and
migration pathways are present, successful charge and
fill of fault traps would likely be possible if the fault
zone seal material behaved as an impermeable barrier
that prevented further migration. If the fault becomes
critically stressed with respect to the ambient stress
field during the charge period or subsequent to being
charged, there is a high probability that the fault will
slip and rupture the lateral seal; thereby increasing
fault zone permeability and enabling hydrocarbons to
be discharged from the reservoir.
Previous reservoir stress studies in the Timor Sea
(Castillo et al., 1998) and North Sea (Wiprut and
Zoback, submitted) have shown that trap integrity
along the reservoir-bounding faults is partly
controlled by the interactions between fault geometry,
the principal stress directions and magnitudes, and
pore pressure. This paper reports the results of an
integrated study of the AC/P8 and AC/P16 areas by
Woodside Energy Limited, Shell Development
Australia and GeoMechanics International (GMI) to
formulate a geomechancial model for the region;
which has been interpreted in context with detailed
fault images based on 3D seismic data collected
within the Corallina and Laminaria Fields within the
greater AC/P8 area; and the Fannie Bay, Lameroo,
Ludmilla structures in the greater AC/P16 area
(Figures 1 and 2).

REGIONAL SETTING
The AC/P8 and AC/P16 blocks lie within the
Bonaparte Basin in the central Timor Sea, between

Timor Island and the coast of northern Australia.

These blocks comprise a significant portion of the
regional Londonderry High (Figures 1 and 2). The
Laminaria High, where the Laminaria and Corallina
Fields in the AC/P8 (and AC/L5) are situated within
the northern reaches of the Bonaparte Basin, is
characterized by a small, east-west oriented drowned
platform-remnant similar to the Sahul platform along
the eastern margins of the ZOCA area (Figures 1 and
2). Further south, the AC/P16 block contains the
northern part of the Londonderry High, most of the
Nancar Trough and the southern flank of the
Laminaria High.
The Bonaparte Basin was initiated as part of a multiphase rift event starting in Palaeozoic time, which
resulted in widespread subsidence and the formation
of a pronounced and complex structural fabric of NESW and N-S trending faults (Schuster et al., 1998;
Bishop and OBrien, 1998). The Triassic period was
a time of N-S wrench-type transpressional
reactivation of the major basin-bounding faults in the
greater Bonaparte Basin.
By the Middle -Late
Jurassic, the area had undergone major NW-SE
extension that produced a dominant NE-SW structural
fault fabric. This fabric effectively defined the
reservoir depositional centres represented by the
reservoirs present in the AC/P8 and AC/P16 blocks.
North-south extension in the Late Jurassic to Early
Cretaceous resulted in the development of E-W
trending normal faults that are pervasive throughout
the AC/P8, AC/P16, and ZOCA areas. This E-W fault
fabric that formed during the rifting phase comprises
the primary trap-bounding faults associated with
exploration targets in AC/P8 and AC/P16.
Beginning in Late Miocene (at about 3-5 Ma), and
continuing to the present-day, the area was
transformed to a collisional setting as a result of
oblique convergence between the Australian and
Eurasian plates. The recent GPS (Global Positioning
System) GEODYSSEA Survey between 1994 and
1996 (Wilson et al., 1998; Simmons et al., submitted)
in Southeast Asia, Indonesia, PNG, and Australia
indicates that there is differential motion occurring
between Timor Island and the Australian mainland.
The present-day convergence direction between
Australia and Indonesia is about 27N (Wilson et al.,
1998; Simmons et al. submitted). Considering the
Eurasian plate as a fixed reference frame, Timor
Island is moving in this direction at about 62 mm/yr,
while Australia is moving at about 75 mm/yr,
Vol. 1 - 262

implying that the Australian continent is traveling

about 20% faster than Timor Island. The major
tectonic expressions of this convergence in the
Indonesia area are the partial subduction of the
Australian plate, pervasive reverse faulting, and the
development of an accretionary prism on Timor
Island. Several major NE-SW left-lateral strike-slip
faults that span the Banda Arc-Timor Sea provide
supportive evidence of oblique convergence (e g.,
Sumba Fracture Zone). Strike-slip faulting appears to
be contemporaneous based on the 1997 M6.3
Cockatoo Earthquake (insert in Figure 1).
Exploration wells within AC/P8 and AC/P16 have
targeted Middle to Upper Jurassic sandstone
reservoirs in structural traps on the margins of the
Nancar Trough and Laminaria High (Figure 2). With
the exception of the Laminaria and Corallina
discoveries, the drilling programme has been
disappointing.
The Barnacle -1, Vidalia-1, and
Claudea-1 wells (Figure 1) encountered little to no
direct evidence for hydrocarbon shows, implying a
lack of charge to these structures. Ludmilla-1, which
tested a Nancar Sandstone Member tilted fault-block
closure, encountered a 4 m live oil column overlying
a 50-70 m long residual oil column. The Mandorah-1
well (Figure 1) tested a Montara/Laminaria Formation
tilted fault-block closure but failed to encounter any
hydrocarbons. The structure is interpreted to have
had access to oil charge, but the lack of recoverable
hydrocarbons is attributed to either the inability of the
bounding faults to support a hydrocarbon column or a
leaky fault along the migration pathway. Fannie Bay1 and Lameroo-1 tested the Laminaria Formation in
two distinct tilted fault-block closures and were found
to be water-bearing with residual oil saturations only.
Both structures are interpreted as having retained
sizeable oil columns (in the case of Fannie Bay-1, in
excess of 80 m) which have since leaked. The Mindil1 well was drilled along the structural trend of the
Ludmilla-1 oil discovery.
The well failed to
encounter any moveable oil, although the possible
presence of residual hydrocarbons is suggested from
logs. A detailed description of the hydrocarbon
exploration history of the Nancar/Laminaria area of
the Timor Sea, can be found in De Ruig et.al., (this
volume).
STRESS ANALYSIS METHOD
Determining the relationship between the seismically
detected regional faults and the in-situ state of stress

involves
constructing
a
well-constrained
geomechanical model. This was accomplished by
reviewing available drilling information, pressure
data, and high-resolution microresistivity and
ultrasonic wellbore images from over 15 wells to
determine the magnitudes and orientations of the
principal tectonic stresses (SHmax, Shmin and Sv)
and pore pressure (Pp) distribution within the AC/P8
and AC/16 areas. The magnitude of the greatest
principal horizontal stress (SHmax) was calculated
using GMISFIB (Stress and Failure of Inclined
Boreholes) by forward modelling the style of stressinduced compressive and tensile wellbore wall failure
as observed in each of these wells. The orientation of
the greatest horizontal principal stress is parallel to
that of tensile failures (tensile wall fractures) and
perpendicular to that of compressive failures
(breakouts) in these near-vertical wells.
The
magnitude of the least principal horizontal stress
(Shmin) was determined from extended leak-off tests
(XLOT) and leak-off tests (LOT), while the
magnitude of the vertical stress (Sv) was simply
calculated based on density data collected in several
representative wells.
For a more complete description of the methodology
used in this study, the following original references
based on applying these techniques worldwide are
helpful (Haimson and Fairhurst, 1967; Bell and
Gough, 1979; Zoback and Healy, 1984; Plumb and
Hickman, 1985; Zoback et al., 1985; Moos and
Zoback, 1990; Zoback and Healy, 1992; Barton et al.,
1995; Castillo and Zoback, 1994; Peska and Zoback,
1995; Barton et al., 1998).

AC/P8 AND AC/P16 STRESS STATE

Over 5000 cumulative metres of high-resolution
wireline log image data and 4-arm caliper data have
been reviewed in order to identify stress-induced
compressive and tensile wellbore failures in about 15
wells within AC/P8, AC/16 and surrounding areas.
This information provided the basis upon which to
build the geomechancial model. To accurately assess
the uniformity of the stress field, we supplemented
this study with analysis of pressure data, well logs,
and other data from over 22 wells in the area. Results
indicate that the greater AC/P8 and AC/P16 area is
generally characterised by a strike-slip faulting
regime (SHmax > Sv > Shmin) in which the vertical
stress is the intermediate stress. Pressure data and
Vol. 1 - 263

drilling information indicates that the pore pressure

regime within the reservoir is approximately
hydrostatic throughout the area.
SHmax stress orientation
The quality of the image data ranged from good to
excellent.
Representative examples of wellbore
failure seen in the analyzed wells include the drillinginduced tensile wall fractures from Laminaria -2 (Fig.
3), and wellbore breakouts from Claudea-1 (Figure 4)
and Lameroo-1 (Figure 5). In addition to the
wellbore breakouts shown in Figures 4 and 5, tensile
wall fractures were also seen elsewhere in these wells
and were oriented close to 90 degrees to the direction
of the breakouts. Despite the regional variation in
SHmax stress directions, each particular well
indicated a relatively uniform stress orientation from
about 1,000 to 4,000 mTVD. Wellbore breakouts in
these study wells were relatively symmetric with
breakout widths ranging from 40 to 70. The
breakout width describes the angular coverage of the
borehole wall that failed in compression resulting in
the rock spalling from the wellbore.
The orientation of the maximum principal horizontal
stress (SHmax) varies systematically between the
AC/P8 and AC/P16 blocks (Figure 1). Details of
SHmax stress orientations from the individual wells
were analysed using azimuthal statistics (Mardia,
1979) and are listed in Table 1. The regional SHmax
stress direction in AC/P8 (Laminaria High) is 15N
6 (Figure 6a); while further south in AC/P16, the
SHmax stress direction changes to about 63N 6
(Fig. 6b). This marked transition occurs along the
northern reaches of the Nancar Trough (Figure 1).
The SHmax stress direction in the northern part of the
study are is remarkably sub-parallel to other stress
indicators seen in the northern section of ZOCA
(Castillo et al., 1998) as well as being sub-parallel to
the relative convergence direction between Indonesia
and Australia (Figures 1 and 2). This transition in
SHmax stress orientation to the south is nearly
identical to the stress rotation previously reported for
the western part of ZOCA (Castillo et al., 1998).
The systematic variation in SHmax stress directions
between the AC/P8 and AC/P16 areas is not clearly
understood. That the regional SHmax stress direction
in AC/P8 (Laminaria High) area is subparallel to the
convergence direction between Australia and
Indonesia (Figure 2), suggests that present-day

horizontal plate motion direction influences the

direction of the maximum horizontal principal stress
in this area.
The marked rotation in SHmax
beginning in the Nancar Trough and continuing into
the Londonderry High suggests that other
mechanisms are responsible for the systematic
variation in stress directions seen in the greater
AC/P16 area. Interestingly, the relatively uniform
SHmax stress direction seen within the Nancar
Trough and Mallee Terrace appear to be
systematically different from directions seen in the
Cartier Trough and Londonderry High area (Figure
2). The generalized SHmax stress direction in the
Nancar Trough and Mallee Terrace is approximately
65N, while SHmax in the Cartier Trough and
Londonderry High is about 30N. This regional
variation may be related to a major transition in the
regional fault trends at the Aptian Unconformity
level. Faults in the Nancar Trough and Mallee
Terrace area trend roughly E-W, while in the Cartier
Trough and Londonderry High area the prevailing
fault trend is NE-SW (Figure 2). This mutually
systematic rotation in both the SHmax stress direction
and regional fault trends in the greater AC/P16 area
would suggest that the regional tectonic patterns are
influencing the regional stress state. A more thorough
examination of the stress state in the AC/P4 area
(southwest of AC/P16) and ZOCA 96-16 (east of 9112) may help understand if these apparent stress
provinces distinct from the northern part of the study
area, are influenced by fault geometry .
Pore pressure, vertical stress, and Shmin
Direct measurements of formation pore pressure (Pp)
in AC/P8 and AC/P16 on wells analyzed in this study
are limited. Reservoir formation pore pressure was
generally considered to be normal and estimated to
range between 1.03 and 1.04 SG. The vertical
principal stress (Sv) was determined by integrating
density logs collected in AC/P8 and AC/P16 wells.
The magnitude of the regional minimum horizontal
principal stress (Shmin) in AC/P8, (including
AC/L5), AC/P16 and surrounding areas is inferred
from formation integrity tests (FIT) in 5 wells, LOTs
in 13 wells, and an XLOT in Claudea-1 (Figures 7
and 8, Table 1). A FIT generally provides only a
lower bound on the fracture gradient, which may not
be equivalent to the minimum principal stress. In the
case of AC/P8, the Shmin stress magnitude can be
assumed to be approximately equal to the leak off
Vol. 1 - 264

pressure because LOT results from these vertical

wells indicate a fracture gradient that is lower than the
vertical stress. The most robust measurement of
Shmin in the AC/P8 area was collected during an
XLOT in the Claudea-1 well at a depth of about 2750
mTVD. Stress profiles of pore pressure, vertical stress
and Shmin values for the study area are shown in
Figures 7 and 8, respectively. Without exception all
measurements (FIT, LOT, and XLOT) indicate an
Shmin that is considerably less than the vertical
stress. The thin dashed line in Figures 7 and 8
represents a best fitting approximation of Shmin
placing more weight on the LOT and XLOT rather
than the FIT, for reasons described above. These
results quantify four of the five unknown parameters
of the in-situ stress field (the magnitudes of Sv, Pp,
Shmin, and the orientation of SHmax). To determine
the remaining parameter (the magnitude of the
maximum horizontal stress, SHmax) requires a more
detailed analysis of wellbore failures from image
logs.
However, it is possible to use the existing data to
predict whether SHmax is greater or less than Sv,
based on the ratio between the effective vertical and
least horizontal stresses. For instance, frictional
constraints on the differential effective stress
magnitudes are limited by re-occurring slip on
optimally-oriented faults assuming a Mohr-Coulomb
failure criteria (Jaeger and Cook, 1979), namely
(S1-P p)/(S3-P p)

[(2+1)1/2 + ]2

(1)

where is the coefficient of friction and S1 > S3. In

seismically
active
regions,
in-situ
stress
measurements using hydraulic fracturing techniques
have confirmed that these laboratory constraints on
stress magnitudes are generally correct using values
between 0.6 and 1.0 (Byerlee, 1978; McGarr, 1980;
Zoback and Healy, 1984; Zoback and Healy, 1992).
In this paper we have used =0.7, well within the
0.6-1.0 range observed by Byerlee (1978).
Applying this to the top of the Laminaria (or Nancar)
Formation at an average depth of 3,200 m, the
frictional equilibrium relationship between Pp,
Shmin, and Sv is (Sv-Pp)/(Shmin-Pp) ~ 2.62 (Jaeger
and Cook, 1979). This is significantly less than a
value that would cause slip along optimally-oriented
faults in a normal faulting stress regime based on
laboratory-derived
coefficients
of
friction

(Byerlee, 1978). Therefore, while a least principal

stress that is less than the vertical stress could indicate
a normal faulting stress regime (i.e., Sv > SHmax >
Shmin), we will show next that the style of wellbore
failure seen in AC/P8, AC/P16 and surrounding
blocks can not be explained by a normal faulting
stress regime. In particular, we find that a strike-slip
stress regime (i.e., SHmax > Sv > Shmin) is not only
consistent with the style of wellbore failure, but is
also consistent with local tectonics associated with
present-day convergence between the Australian and
the Indonesian plates (Wilson et al., 1998; Shuster et
al., 1998; Simmons et al., submitted), and recent
earthquake activity (National Earthquake Information
Center web site, Fredrich et al., 1988).
Absolute magnitude of SHmax
To constrain SHmax stress magnitudes, we use the
GMISFIB module CSTR (Constrain Stress) at
several depths between about 1,000 and 4,100 mTVD
in AC/P8 and AC/P16 to forward model the
compressive (breakouts) and tensile (tensile wall
fractures) wellbore failure seen in the various wells
(Figures 3, 4 and 5). We also used GMISFIB
module BSFO (Borehole Stress and Failure
Orientation) to evaluate explicitly how borehole
geometry, rock strength, and stress conditions at the
borehole wall lead to the observed style of wellbore
failure. Additional input parameters required for this
analysis included the magnitudes of Sv and Shmin,
Pp, the mud weight, temperature, and wellbore
trajectory. Figures 9 and 10 show representative
examples of how GMISFIB was used in this study.
Figure 9 presents the results of the analysis for the
Claudea-1 well at about 2,900 mTVD where tensile
wall cracks were detected, but no breakouts were
observed. The figure plots the magnitude of Shmin on
the x-axis and the magnitude of SHmax on the y-axis.
The polygon constrains the horizontal stress
magnitudes assuming the crust is in frictional
equilibrium (Equation 1).
The blue contours
represent the magnitudes of the two horizontal
stresses required to induce tensile failure for a given
tensile strength (contours) and drilling conditions.
Because drilling-induced tensile fractures were
observed in this interval, the stresses must lie above
the blue contour corresponding to a finite effective
tensile strength of the rock. In the absence of strength
measurements, the contour corresponding to To=0
provides a lower bound on SHmax. The least
Vol. 1 - 265

horizontal stress, constrained by XLOT and LOTs is

predicted to be approximately 43.7 MPa. The
combination of the frictional faulting constraint and
the constraint imposed by the presence of tensile
failures limits the magnitude of SHmax at this depth
to be 79.0 +/- 7.8 MPa. The red contours delineate the
unconfined compressive strength required to prevent
breakouts from occurring for the given magnitudes of
the two horizontal stresses at this depth. The absence
of breakouts in this interval requires that the
unconfined compressive strength must be greater than
95 MPa (Figure 9 and Table 1).
Figure 10 shows a forward model of drilling-induced
compressive (wellbore breakouts) and tensile (tensile
wall fractures) failure seen in the Lameroo-1 well at
about 3,900 mTVD. The resultant stress distribution
around the borehole is shown in Figure 10a indicating
that breakouts would be about 55 degrees wide, whic h
is consistent with the breakouts seen in the STAR
image data. Breakouts will occur when the maximum
circumferential stress exceeds the uniaxial
compressive strength of the rock.
The missing
colours in Figure 10a correspond to sections in the
near-wellbore region that would be in tension, which
is also consistent with observations of the drillinginduced tensile wall fractures in Lameroo-1. Figure
10b is a modelled unwrapped view of the borehole
wall showing the location and width of the breakouts
as they are seen in image data. Figure 10c is again a
model of the borehole wall, and plots the position of
the inclined tensile wall fractures as they appear
within the Lameroo-1 well.
Modelling wellbore failure in the Claudea-1,
Lameroo-1 and other wells in the greater AC/P8 and
AC/P16 areas, reveals that the magnitude of SHmax
is consistently in excess of Sv (Figures 7, 8, 9, 10, and
Table 1). This implies that this section of the Timor
Sea area is subject to a strike-slip stress regime
(SHmax > Sv > Shmin). This strike-slip stress state is
consistent with results from a regional trap integrity
study (Castillo et al., 1998) where it was found that a
strike-slip stress regime exists within ZOCA 91-01
and much of ZOCA 91-12, which is also consistent
with recent strike-slip earthquake activity (e.g., 1997
M6.3 Cockatoo Earthquake). Inferences based on
structure, kinematic, and geodetic studies also suggest
that the Timor Sea Area is subject to oblique
convergence between Australia and Timor Island

resulting in strike-slip deformation (Shuster et al.,

1998; Wilson et al., 1998; Simmons et al., submitted).
DISCUSSION: IMPLICATIONS FOR FAULT
TRAP INTEGRITY IN THE AC/P8 AND AC/P16
AREAS
Many of the wells drilled in the AC/P8, AC/P16,
ZOCA, and surrounding regions encountered
significant residual oil columns, implying that the
associated fault trap structures were optimally suited
for hydrocarbon charge and retention early in the
charge history. This would further imply that the
faults bounding these reservoirs were not previously
critically-stressed, thus preserving the fault seal gouge
material (formed by cataclastic grain-reducing
processes associated with slip) within the fault zone.
Subsequent to hydrocarbon charge and oblique
collision between Australia and Indonesia (Shuster et
al., 1998), hydrocarbon leakage would have occurred
along specific faults that became critically-stressed
due to the change in the stress field.
The ability of a fault to behave as a seal is influenced
in part by the principal stress directions and
magnitudes, fault dip and dip azimuth. If a fault is
critically stressed with respect to the present-day
stress field, there is a high likelihood that the fault
will slip, thereby increasing the fault zone
permeability and enabling hydrocarbons to migrate
from the reservoir. To explore this fault-stress
relationship, results of the in-situ stress analysis
described above have been integrated with detailed
representations of the faults interpreted from 3D
seismic data from the AC/P8 and AC/P16 areas.
These interpretations of the reservoir fault surfaces
were depth converted and transformed into a series of
connected fault segments, fully described in terms of
a 3D coordinate system, dip and dip azimuth for each
individual segment. The shear and normal stresses
resolved on each of these segments were calculated
using GMIMohrFracs in order to identify which
segments are critically-stressed in the present-day
stress regime.
Fault segments which are critically-stressed are
associated with an applied shear stress that exceeds
the frictional strength of the fault plane. If the
difference between the shear stress and the frictional
strength of the fault is positive, the fault may rupture
due to Coulomb shear failure, which can be expresses

Vol. 1 - 266

as the following Coulomb Failure Function (Jaeger

and Cook, 1979):
- (Sn P p ) = CFF

(2)

where is the applied shear stress, Sn -P p is the

effective normal stress, is the coefficient of friction
(Byerlee, 1978). The second term in Equation 2 is the
frictional strength of the fault surface.
An overview of two representative fault surfaces in
stress space is shown in Figure 11 (Corallina Field)
and Figure 12 (Fannie Bay). Figure 11 shows model
results for two major faults in the Corallina Field area
(Figure 2).
There are critically-stressed fault
segments (red tadpoles, red poles in the Mohr
Diagram and white poles in the stereographic
projection), associated with the north-dipping (eastwest trending) fault in the Corallina Field. These
critically-stressed segments appear to be restricted to
those few that are steeply-dipping (70-75) to the
ENE or NW (Figure 11a). A similar situation exists
for the south-dipping fault in the Corallina Field
(Figure 11b). With the exception of the few fault
segments that dip to the southeast, there is a large
population of SSE dipping fault segments which are
not optimally oriented for shear failure. Because
SHmax is about 10N in this area (Figure 2 and Table
1) and the trend of the primary trap-bounding faults is
E-W, the trap integrity of the overall reservoir faults
appears to be high (CFF is negative), which is
consistent with the presence of hydrocarbon in the
Laminaria and Corallina Fields.
The pervasive presence of these high-angle faults
segments in the Fannie Bay and Lameroo areas
indicate that many of the reservoir-bounding faults
may be critically-stressed. The northward dipping
(approximately E-W trending) reservoir-bounding
fault adjacent to the Lameroo-1 well is shown in
Figure 12. This particular fault segment has a
morphologic shape with different fault dip segments
ranging from 50 to 70. The high angle faults (dip >
60) are better suited for slip failure in a strike-slip
stress regime, while fault segments which dip < 60
are not optimally oriented for shear failure.
Using 3D visualization technologies to display the
CFF plotted as fault attributes on the individual fault
surfaces helps to understand this fault-stress
relationship. This analysis provides an opportunity to
quantify the potential for shear failure on different

segments of the fault, and therefore, the likelihood of

fault trap failure. Figures 13 to 16 illustrate how this
stress-fault visualization analysis was used to evaluate
the trap integrity along the reservoir-bounding fault
traps in the Corallina Field, Fannie Bay, Lameroo and
Ludmilla structures in AC/P8, AC/P16 and
surrounding areas.
Corallina and Laminaria structures in AC/P8
Figure 13 is a oblique perspective view of the
Corallina Field facing northeast, showing the
relationship between the Top Laminaria Formation
and the major faults defining the structural trap. The
colour contours on the 3D seimic faults are the CFF
fault attributes, based on the stress analysis described
above (Figures 2 and 7, Table 1). Because the
SHmax
stress
direction
is
approximately
perpendicular to the east-west trending faults, these
structures are not critically-stressed since the
frictional fault strength of the fault exceeds the
applied shear stress resulting in a CFF that is less than
zero (see Equation 2).
A change in fault geometry, particularly the fault
bounding the Corallina horst to the north, results in a
corresponding change in the CFF resolved on the
fault segment (Figure 13). The consequence of this
change in fault trend to a southeast dipping structure
is a positive CFF (i.e., critically-stressed segment).
Interestingly, this fault attribute transition from noncritically-stressed to critically-stressed occurs
approximately where the palaeo-oil-water contact in
Corallina, defined on the basis of Grains Oil Inclusion
(GOI) analysis (Figure 13), implying that the current
stress state may be controlling the maximum potential
hydrocarbon column height in the Corallina Field
over geologic time. Because the Mohr-Coulomb
failure criteria evaluate the static stress state along a
particular fault segment, there may be crack growth
effects related to dynamic fault-slip propagation
linking a stable fault segment that is at the threshold
of shear failure. The current oil-water contact within
the Corallina reservoir is up-dip from the paleo-oilwater contact, suggesting that there may have been
some dynamic slip propagation along the fault.
However, we cannot rule out the possiblility that
temporal variations in stress and varying episodes of
charge have not played a role in the current
hydrocarbon accumulation in the Corallina Field.
The results for the Laminaria Field also indicate that
Vol. 1 - 267

the main E-W bounding faults (Figure 2) are not

critically-stressed and would therefore, be capable of
supporting the hydrocarbon column that exists in the
field.
There is some limited evidence that a
secondary population of NE-SW trending faults
within the reservoir may be critically-stressed.
However, the resolution of the seismic data does not
place strong constraints on the geometry and extent of
these apparently minor fault segments (Smith et al.,
1996).
Fannie Bay and Lameroo structures in AC/P16
The implications of this stress-fault analysis differ
regionally. Despite the consistency in fault structure
throughout the region (Figure 2), the rotation in
SHmax from nearly N-S in AC/P8 to NE-SW in
AC/P16 produces an increase in applied shear stress
on the ENE-WSW trending faults (compare Figures
11 and 12). The net result of the poor fault trap
integrity of these charged reservoirs appears to be a
nearly complete discharge of the hydrocarbons.
Residual oil columns were detected in the Fannie
Bay-1 and Lameroo-1 wells, implying that at
sometime in the past the major faults were not
critically-stressed and were therefore, capable of
trapping hydrocarbons. A detailed examination of the
major reservoir-bounding fault within the Fannie Bay
structure indicates that the entire fault is criticallystressed, both above and below the top Laminaria
horizon (Figure 14). The fault segment adjacent to the
Lameroo structure is critically-stressed above the
horizon, adjacent to the cap rock (Figure 14). The
fault that once supported a hydrocarbon column in the
Lameroo reservoir, but is now critically-stressed, may
have initiated shear failure within the upper sections
of the fault (adjacent to the cap rock) and
subsequently propagated downward towards fault
segments near the threshold of shear failure (Figure
14).

Ludmilla and Mindil structures in AC/P16

Figures 15 and 16 illustrate this methodology applied
to the Ludmilla and Mindil area in AC/P16. These
figures show the top Nancar Formation along with the
major near-vertical faults and well locations.
Coloured contours on the faults are the specific CFF
attributes based on the stress tensor determined in this

study. Figure 15a shows a regional overview of the

area with the locations of the Nancar-1, Ludmilla-1
and Mindil-1 wells. The view direction is to the
southwest, sub-parallel to the local SHmax stress
direction of 53N (Figure 2).
A close-up view of the top Nancar Formation in the
Ludmilla structure is shown in Figure 15b along with
the two major trap-bounding faults. The CFF is
clearly different on each fault. The southwest dipping
fault is not optimally oriented for shear failure and is
therefore capable of supporting high trap integrity
(i.e. CFF << 0). In contrast, the northwest dipping
fault is critically-stressed (CFF > 0), particularly
along the northeast segment of the fault where the
CFF contours are shown in different shades of purple.
The greenish-brown contours on this northwest fault,
closer to the crest of the reservoir, represent sections
that are near the shear failure threshold. The
transition to a positive CFF attribute (purple shades)
along this section of the northwest faults marks the
approximate location of the palaeo-oil-water contact
based on GOI analysis. Drilling results of the
Ludmilla-1 well indicated a live oil column overlying
a residual oil column with the current oil-water
contact situated several tens of metres above the
palaeo-oil-water contact (Figure 15b).
The Mindil-1 well drilled a similar structure located
further to the west (Figure 15a). A close-up view is
shown in Figure 16. Again, the fault bounding the
reservoir to the north is critically-stressed (CFF > 0)
suggesting that the probability of encountering a
breached reservoir adjacent to this fault is high. The
Mindil-1 well was drilled into the structural high
situated adjacent to the southern trap-bounding fault
(left fault in Figure 16). The CFF associated with this
fault indicates that it is near the threshold of being
critically-stressed.

Pore pressure -induced shear failure

It is possible to quantify what the critical pore
pressure (Pp Critical) is required for an otherwise
stable fault segment to fail in shear (Wiprut and
Zoback, submitted). The Pp Critical (Pp/Sv scale bar
in Figures 11 and 12) reflects the incremental increase
in pore pressure required to decrease the effective
normal stress enough to induce shear failure on
segments that are currently not critically-stressed.
This is equivalent to replacing pore pressure (Pp) with
Vol. 1 - 268

the critical pore pressure (Pp Critical) in Equation 2.

The current pore pressure regime normalized by the
vertical stress (Sv) is about Pp/Sv ~ 0.4 (Figures 11
and 12). Increasing the pore pressure within or
adjacent to faults that are at the threshold of shear
failure (CFF slightly negative), would effectively
increase the likelihood of shear failure occurring
along these segments. Potentially, this could also
extend the zone of slip failure from the segment that
is already critically-stressed and prone for failure.
If elevated pore pressure is to be considered a viable
mechanism for enhancing the likelihood of shear
failure on segments that are currently at their
threshold of failure, it is necessary to identify the
source and timing of the overpressure. However, the
source of the elevated fluid pressure does not
necessarily have to be within the reservoir. These
elevated pore pressures can be unique to the fault
zone and unrelated to the adjacent rock formations.
Brines have been found to be associated with residual
oil columns and faults (Lisk et al., 1999) which could
have migrated from the same salt deposits laid down
in the Early Palaeozoic (Shuster et al., 1998). These
apparently high temperature, high pressure brines
migrating upward from a deeper source along
critically-stressed faults could provide a pulse of high
pressure fluid leading to dynamic shear failure along
segments at the threshold of shear failure (CFF
slightly negative).
Alternatively, the reservoir pore pressure in the
Londonderry High, Nancar Trough, and Laminaria
High areas at the time of hydrocarbon charge could
have been elevated with respect to the present-day
pore pressure. For instance, if the reservoir pore
pressure exceeded Pp Critical fault slip would result
in a discharge of hydrocarbons, which would
subsequently lead to a decrease in reservoir pore
pressure. If the pore pressure decreased to levels
below the critical pore pressure, then fault slip would
be inhibited and a fault seal would begin to redevelop
(i.e., secondary mineralization would heal the fault
zone). Although this palaeo pore pressure mechanism
is difficult to reconcile given the expanse of the
reservoir sands in the region, it may be possible to
determine the paleo-reservoir pore pressure by
conducting fluid inclusion analysis similar to George
et al. (1997) and Lisk et al. (1996).

CONCLUSION
Increasing our understanding of fault seal integrity
requires a well-constrained geomechanical model.
This includes detailed knowledge of the principal
stress magnitudes (SHmax, Shmin and Sv) and stress
directions, pore pressure, coupled with a detailed
representation of fault surfaces from depth-converted
3D seismic data. If a fault segment is critically
stressed with respect to the present-day stress field,
there is a high likelihood that the fault will slip due to
shear failure, resulting in hydrocarbon leakage along
the slip-induced fault zone permeability structure.
Quantifying the stress state in the AC/P8 and AC/P16
areas indicates a non-uniform strike-slip stress regime
(SHmax > Sv > Shmin) with the orientation of the
maximum principal horizontal stress (SHmax)
varying systematically across the study area, similar
to that previously reported for the western part of
ZOCA. Within the AC/P8 region, SHmax is about
15N, while further south in AC/P16, the regional
SHmax stress direction systematically changes to
about 63N.
Assessing some of the risks associated with trap
integrity in structures drilled in AC/P8 and AC/P16
has been accomplished by combining details on the
stress field with 3D representations of the important
faults defining the traps. Evaluating the stress state
resolved on these faults to determine which fault
segments are critically-stressed, followed by 3D
visualization to interpret the results, provides an
important exploration tool for assessing risks
associated with fault seal failure. Applying this
integrated stress-fault approach to the Laminaria and
Corallina Fields, and the Fannie Bay, Lameroo and
Ludmilla structures has greatly increased our
understanding of the mechanism(s) responsible for
the successful and not-so successful drilling
enterprises in the Timor Sea area.

ACKNOWLEDGEMENTS
Author DAC is specially grateful for the opportunity
to have worked with co-authors DJB and MdR at
Woodside Australian Energy on this Timor Sea fault
seal project. Extended thanks also to Marian Magee
and David Moffat for their constructive and
supportive comments.

Vol. 1 - 269

REFERENCES
Barton, C.A., Zoback, M.D., and Moos, D., 1995.
Fluid flow along potentially active faults in crystalline
rock. Geology, v. 23 (8), p. 683-686.
Barton, C.A., Castillo, D.A., Moos, D., Peska, P., and
Zoback, M.D., 1998. Characterizing the full stress
tensor based on observations of drilling-induced
wellbore failures in vertical and inclined boreholes
leading to improved wellbore stability and
permeability prediction, APPEA Journal, v. 38, Part
1, p. 466-487.
Bell, J.S., and Gough, D.I., 1979. Northeastsouthwest compressive stress in Alberta: Evidence
from oil wells. Earth and Planetary Science Letters, v.
45, p. 475-482.
Bishop, D.J., and OBrien, G. W., 1998. A MultiDisciplinary
Approach
to
Definition
and
Characterisation of Carbonate Shoals, Shallow Gas
Accumulations and Related Complex Near-Surface
Sedimentary Structures in the Timor Sea, APPEA
Journal, v. 38, Part 1, p. 98-114.
Byerlee, J., 1978. Friction of rocks, Pure and Applied
Geophysics, v. 116, p. 615-626.
Castillo, D.A., Hillis, R.R., Asquith, K., and Fischer,
M., 1998. State of Stress in the Timor Sea Area,
Based on Deep Wellbore Observations and Frictional
Failure Criteria: Application to Fault Trap Integrity;
in the Sedimentary Basins of Western Australia 2,
eds. P. G. and R. R. Purcell, Proceedings West
Australian Basins Symposium, Perth, Western
Australia, p. 325-341.
Castillo, D.A., and Zoback, M.D., 1994. Systematic
variations in stress state in the southern San Joaquin
valley: Inferences based on wellbore data and
contemporary seismicity, American Association of
Petroleum Geologists Bulletin, 78 (8) 1257-1275.

Preserved at Jabiru Field, Vulcan Basin, Exploration

Geophysics, v. 28, p. 66-71.
Fredrich, J, McCaffery, R., and Denham, D., 1988.
Source parameters of seven large Australian
earthquakes determined by body waveform inversion,
Geophysical Journal, v. 95, p. 1-13.
Haimson, B.C., and Fairhurst, C., 1967. Initiation and
Extension of Hydraulic Fracture in Rocks, Soc.
Petrol. Eng., v. 7, p. 310-318.
Jaeger, J.C. and Cook, N.G.W., 1979. Fundamentals
of rock mechanics, 3rd Edition, Chapman and Hill,
New York, p. 593.
Lisk, M., George, S.C., Summons, R.E., Quezada,
R.A., and OBrien, G. W., 1996. Mapping
Hydrocarbon
Charge
Histories:
Detailed
Characterization of the South Pepper Oil Field,
Carnarvon Basin, APPEA Journal, v. 36, Part 1, p.
445-464.
Lisk, M., Brincat, M.P., OBrien, G. W., Eadington,
P.J., and Faiz, M, 1999. Palaeo-hydrology of the
Vulcan Sub-Basin: Implications for Trap Integrity,
APPEA Journal, v. 39, Part 1, p. 208-226.
Mardia, K.V., 1979. Statistics of directional data,
Academic Press, London, p. 234.
Mcgarr, A., 1980. Some constraints on levels of shear
stress in the crust from observations and theory,
Journal of Geophysical Research, v. 85, p. 62316238.
Moos, D., and Zoback, M.D., 1990. Utilization of
Observations of Well Bore Failure to Constrain the
Orientation and Magnitude of Crustal Stresses:
Application to Continental, Deep Sea Drilling Project,
and Ocean Drilling Program Boreholes, Journal of
Geophysical Research, v. 95, p. 9305-9325.

De Ruig, M.J., Trupp, M., Bishop, D.J., Kuek, D., and

Castillo, D. A., 2001. Fault Architecture in the Nancar
Trough/Laminaria Area of the Timor Sea, Northern
Australia, (This Volume).

Peska, P., and Zoback, M.D., 1995. Compressive and

tensile failure of inclined wellbores and determination
of in situ stress and rock strength, Journal of
Geophysical Research, v. 100 (7), p. 12791-12811.

George, S.C., Lisk, M., Eadington, P.J., Krieger,

F.W., Quezada, R.A., Greenwood, P.F. and Wilson,
M.A., 1997. Fluid Inclusion Record of Early Oil

Plumb, R.A., and Hickman, S.H., 1985. Stressinduced borehole elongation a comparison between
the four-arm dipmeter and the borehole televiewer in
Vol. 1 - 270

the Auburn geothermal well, Journal of Geophysical

Research, v. 90, no. B7, p. 5513-5521.

EOS, Transactions, American Geophysical Union, v.

79, (45), November 10.

Shuster, M.W., Eaton, S., Wakefield, L.L., and

Kloosterman, H. J., 1998. Neogene Tectonics, Greater
Timor Gap Offshore Australia: Implications for Trap
Risk, APPEA Journal, v. 38, Part 1, p. 351-379.

Wiprut, D., and Zoback, M.D., submitted. Fault

Reactivation and Hydrocarbon Migration in the
Northern North Sea, Geology.

Simmons, W., Ambrosius, B, Noomen, R.,

Angermann, D., Wilson, P., Becker, M., Reinhart, E.,
Walpersdorf, A., and Vigny, C., submitted. Observing
Plate motions in S.E. Asia : Geodetic results of the
GEODYSSEA Project, Geophysical Research Letters.
Smith, G.C., Tilbury, L.A., Chatfield, A., Senycia, P.,
and Thompson, N., 1996. Laminaria A New Timor
Sea Discovery, APPEA Journal, v. 36, (1), p. 12-29.
Wilson et al. (14 others), 1998. Study Provides Data
On Active Plate Tectonics in Southeast Asia Region,

Zoback, M.D. and Healy, J.H., 1984. Friction,

Faulting and In Situ Stress, Annales Geophysicae, v.
2, p. 689-698.
Zoback, M.D., and Healy, J.H., 1992. In Situ Stress
Measurements to 3.5 km Depth in the Cajon Pass
Scientific Research Borehole: Implications for the
Mechanics of Crustal Faulting, Journal of
Geophysical Research, v. 97, p. 50395057.
Zoback, M.D., Moos, D., and Mastin, L., 1985.
Wellbore breakouts and in situ stress, Journal of
Geophysical Research v. 90, p. 5523-5530.

Vol. 1 - 271

TABLE 1
Well
Name

Image
Data

Alaria-1
Banka Banka-1
Cleia-1
Claudea-1

STAR
FMI
*
STAR
*
FMS

SHmax
Stress
Direction
15N 1
27N 6
*
7N 1
*
11N 4

STAR
*
*
*
*
HDT
*
*
STAR
*
*
*
*
*
FMS
*
FMS
*
*
FMS
*
*
STAR

62N 2
*
*
*
*
67N 7
*
*
12N 5
*
*
*
*
*
no failure
*
36N 4
*
*
77N 4
*
*
72N 4

*
*
FMI
*
*
*
*
*
*
*

*
*
24N 2
*
*
*
*
*
*
*

Ludmilla-1

STAR
*

53N 4
*

Mallee East-1
Mandorah-1
Nancar-1
Vidalia-1

*
*
*
STAR
*
*
*
*
*
*

*
*
*
19N 6
*
*
*
*
*
*

Corallina-1

Fannie Bay-1

Fulica-1
Halimeda-1

Jarrah-1
Keppler-1
Kittiwake-1
Lameroo-1

Laminaria-1
Laminaria-2

Laminaria-3
Laminaria East -1
Lorikeet -1

Depth
mTVD
2400
*
2115
2750
2900
883
2612
3150
3250
3300
*
3007
3340
3430
4100
*
1634
2100
*
2900
3000
3200
3300
3400
*
1805
*
1153
1620
*
2395
2416
*
915
2959
3885
3900
3926
889
2653
*
2406
3300
3350
3400
2645
2815
*
1614
*
2160
3200
3500
1772
3016
2040
*
797
2699
1150
2000
2400
2950

Shmin
(observed)
MPa
*
*
31.71
41.48
*
10.99
36.40
*
*
*
*
45.38
*
*
*
*
20.14
*
*
*
*
*
*
*
*
30.42
*
16.64
*
*
34.37
*
*
10.31
44.82
*
*
*
10.80
36.18
*
31.65
*
*
*
37.58
38.62
*
25.15
*
30.23
*
*
26.05
46.94
30.79
*
10.39
38.26
*
*
*
*

SHmax
(modelled)
MPa
55.50 5.0
*
*
*
79.01 7.84
*
*
72.90 7.40
76.44 7.63
77.72 7.72
*
*
78.73 5.42
81.13 5.43
97.71 4.34
*
*
48.00 8.71
*
76.60 6.97
70.08 7.08
76.60 6.97
77.45 7.41
79.96 7.48
*
*
*
*
37.35 3.65
*
*
58.55 5.57
*
*
*
91.00 5.71
91.97 5.07
92.90 5.30
*
*
*
*
76.20 6.16
77.48 6.28
78.84 6.37
*
*
*
*
*
*
81.42 7.02
89.27 6.61
*
*
*
*
*
*
25.28 2.30
45.98 5.98
58.11 5.32
73.70 5.47

Co
modelled
MPa
85 15
*
*
*
120 25
*
*
80 10
100 25
102 22
*
*
55 25
32 17
130 20
*
*
55 20
*
40 20
37 15
60 20
62 22
55 20
*
*
*
*
50 10
*
*
80 20
*
*
*
83 13
105 20
70 10
*
*
*
*
82 22
90 20
95 20
*
*
*
*
*
*
95 35
72 28
*
*
*
*
*
*
34 8
56 16
75 15
95 25

Vol. 1 - 272

FIGURE 1 Regional stress map of the Timor Sea Area based on data within AC/P8 and AC/P16 and ZOCA. Stress indicators are represented by the
inward-facing arrows showing the direction of the maximum principal horizontal stress (SHmax) in the ZOCA 91-01 and ZOCA 91-12
areas (black stress indicators) after Castillo et al. 1998, and for the AC/P8 and AC/16 area (red stress indicators) based on this study.
The black/white earthquake focal mechanism solution for the August 10, 1997 M6.3 Cockatoo Earthquake indicates nearly pure strikeslip fault movement on either NNE-SSW or NW-SE nodal planes. The large blue arrow represents the convergence direction between
Australia and Indonesia (Wilson et al., 1998). Well names abbreviations: Undan-1 (e.g., U-1) and Bayu-2 (e.g., B-2).
Vol. 1 - 273

FIGURE 2 Depth map at the top of the Aptian Unconformity (Early Cretaceous Darwin Formation)
in the AC/P8 and AC/P16 study area showing SHmax stress orientation. The inwardfacing arrows correspond to the direction of the maximum principal horizontal stress
(SHmax) inferred from wellbore failure in the respective wells.
Vol. 1 - 274

FIGURE 3 Example of FMI image data from Laminaria -2 showing tensile wall fractures, as
indicated by their highly conductive nature (dark colour).

Vol. 1 - 275

FIGURE 4 - Example of STAR image data from Ludmilla -1 showing wellbore breakouts as indicated by the vertical dark regions of the
borehole. Sinusoidal feature on the amplitude and travel-time images are natural fractures.

Vol. 1 - 276

FIGURE 5 Example of STAR image data from Lameroo-1 showing wellbore breakouts as indicated
by the dark vertical regions of the borehole.

Vol. 1 - 277

FIGURE 6 a) Rose diagram and b) associated histogram showing the regional SHmax stress direction based on all stress indicators seen in the
wells analysed in the AC/P8 area. Over 500 m of cumulative wellbore failure suggests that the regional SHmax stress directions is
150 60 N. c) Rose diagram and d) associated histogram showing the regional SHmax stress direction based on all stress indicators
seen in the wells analysed in the AC/P16 area. Over 450 m of cumulative wellbore failure suggests that the regional SHmax stress
directions is 630 60 N.
Vol. 1 - 278

FIGURE 7 Profile of stress magnitudes inferred from data collected in the AC/P8 (and AC/L5 Laminaria
Area). The pore pressure gradient is 1.03 SG. Minimum horizontal stress (Shmin)
measurements are considerably less than the vertical stress (Sv). The vertical stress is based on
integration of the density log. The Shmin gradient, represented by the dashed line is primarily
constrained by the most robust data types (e.g., LOT and XLOT). Results indicate that the
AC/P8 area is associated with a strike-slip stress regime (SHmax>Sv>Shmin). Well
abbreviations: Alaria -1 (A-1), Vidalia-1 (V-1), Halimeda-1 (H-1), Corallina-1 (CR-1), Claudea1 (CL-1), Laminaria -2 (L-2) and Laminaria -5 (L-5).

Vol. 1 - 279

FIGURE 8 Profile of stress magnitudes inferred from data collected in AC/P16. The pore pressure gradient is
1.03 SG above 3000 m and 1.04 SG bellow 3000 m, based on well completion reports. FIT and
LOT indicate that the minimum horizontal stress (Shmin) measurements are considerably less
than the vertical stress (Sv). The vertical stress is based on integration of the density log. The
Shmin gradient, represented by dashed line in primarily constrained by the most robust data
types (e.g., LOT) which actually propagated an hydraulic fracture away from the borehole. The
ranges of the maximum horizontal stress (SHmax) were calculated based on the style of
wellbore failure seen in image logs in each respective well. Results indicate that the AC/P16
area is associated with a strike-slip stress regime (SHmax > Sv > Shmin). Well abbreviations:
Banka Banka-1 (BB-1), Fannie Bay-1 (FB-1), Lameroo-1 (L-1), Kittiwake-1 (KW-1), Jarrah-1
(J-1), Kellper-1 (K-1), and Fulica-1 (F-1).

Vol. 1 - 280

FIGURE 9 Stress state for the Claudea-1 well at a depth of 2,900 mTVD. No wellbore breakouts were
observed but tensile wall fractures were detected. The presence of tensile failures requires
that the stress state lines above the blue lines. Shmin values, determined from the XLOT
performed in the Claudea-1 well at 2750 MD and LOT performed in nearby wells (Figure
4), restrict the allowable range of SHmax magnitudes to between 71.2 and 86.8 Mpa. The
lack of breakouts requires that UCS > 95 Mpa.

Vol. 1 - 281

FIGURE 10 Predicted wellbore breakout (BO) to constrain SHmax that is consistent with the observed
drilling-induced wellbore breakouts at 3926 mTVD in the Lameroo-1 well. The borehole crosssection a) shows the effective borehole circumferential stress distribution. The results of
forward modelling an unwrapped view of a borehole image for b) compressive failure and c)
tensile wall failure is consistent with the observed failure seen in Lameroo-1.
Vol. 1 - 282

FIGURE 11 Calculated stress state resolved on a) the southwest trap-bounding fault and b) the northwest trapbounding fault for the Corallina Field. These structural fault surfaces, based on 3D seismic data,
were defined as individual elements or polygons. Far left plot shows the principal stress and pore
pressure model. The natural fractures are indicated in three different styles: tadpoles (second from
left; lower hemisphere stereographic projection; and 3D Mohr diagram including the Coulomb
frictional failure line corresponding to coefficient of friction (=0.6). The red and white dots
represent the critically-stressed elements of the fault plane.
Vol. 1 - 283

FIGURE 12 Calculated stress state resolved on structural faults based on 3D seismic data in the Lameroo area. Figure description
same as Figure 11.

Vol. 1 - 284

FIGURE 13 A 3D perspective view of the Corallina Field in AC/P8 (AC/L5) area, looking to the
northeast, showing the top of the Laminaria Formation and the major structural faults
mapped from 3D seismic data, defined as a series of connected faults segments. The CFF
associated with each of the faults is contoured as a fault attribute based on the orientation
of the individual fault element with respect to the stress tensor derived from this study
(stress direction and absolute stress magnitudes). The dark red contours correspond to
positive CFF (unstable), while orange-green to green corresponds to negative CFF
(stable). Also shown is the approximate location of the current and palaeo-oil-water
contact inferred from GOI analysis. This corresponds to where segments of the fault
become critically-stressed and therefore, less likely to behave as an adequate fault seal
for trap integrity.

Vol. 1 - 285

FIGURE 14 A 3D perspective view of the Fannie Bay and Lameroo structures looking to the
southwest, showing the top of the Laminaria Formation and the major structural
faults inferred from 3D seismic data, defined as a series of connected faults
segments. Figure description same as Figure 13. The primary fault that defines
the Fannie Bay structure is critically-stressed for shear failure and therefore, less
likely to behave as an adequate fault seal for trap integrity. If only the upper
sections of the primary fault surface adjacent to the Lameroo structure (within
the cap rock) is critically-stressed, slip along this segment may propagate
downward into the reservoir.

Vol. 1 - 286

FIGURE 15 A 3D perspective view of a) the Mallee Terrace in central AC/P16, looking to the southwest,
showing the top of the Nancar Sands and the major structural faults mapped from 3D seismic data,
defined as a series of connected fault segments. The Nancar-1, Ludmilla-1 and Mindil-1 wells are
indicated as near vertical red lines. The grey to purple contours corresponds to critically-stressed
segments (CFF 0), while contour gradations from yellow, green to blue correspond to stable
segments (CFF < 0). B) Close-up view of the Ludmilla-1 area showing the approximate location of
the palaeo-oil-water-contact inferred from GOI analysis in Ludmilla-1. Note that this is also the
location at which sections of the fault become critically-stressed for shear failure and therefore,
less likely to behave as an adequate fault seal for trap integrity.
Vol. 1 - 287

FIGURE 16 A Close-up 3D perspective view of the Mindil-1 area, looking to the west, showing the top of
the Nancar Sands and the major structural faults inferred from 3D seismic data, defined as a
series of connected faults elements. The Mindil-1 well is indicated as a near vertical red line.
The CFF contouring as for Figure 15. The ENE-WSW trending high-angle fault bounding the
reservoir to the north (right side of plot) is critically-stressed and has a high probability of fault
leakage due to shear failure. The ESE-WNW trending high-angle fault bounding the reservoir
to the south is at the threshold of failure due to fault slip.

Vol. 1 - 288