The inset on Fig. 1.10 illustrates in cross
section how the horizontal component of separation (sometimes called the heave) and the dip of
the normal fault are represented by the width of
the black stripes on maps. Greater stripe widths
(gray on the inset) imply greater horizontal separations of two points that originally were adjacent
to one another at the base of the Brent Formation.
Two faults with the same dip are displayed as
stripes with lesser and greater widths (horizontal
separations) corresponding to lesser and greater
dip separation. Two faults with the same dip separation are displayed as stripes with lesser and
greater widths corresponding to greater and
lesser dips. In regions where the dips of all the
normal faults are about the same, the widths of
the stripes on the structure map are used to infer
the relative magnitude of dip separations.
Similarly, if the dip does not vary along the length
of a fault, the changing width of the stripe is used
to infer changes in the distribution of dip separation.
From information gathered by drilling and
recovering core samples, measuring a variety of
indicators in the wells, and running flow and pressure tests in the sub-surface, it is believed that the
rocks in the fault zones shown in Fig. 1.10 are relatively impermeable to fluid flow compared to the
unfaulted portions of the sandstone reservoir.
Therefore these are referred to as sealing faults. A
number of mechanisms for changing the permeability of rock in a fault zone have been identified
(Jones et al., 1998). For example, the crushing of
sandstone grains during shearing in the fault
zone can fill the pores with fragments, thereby
decreasing the porosity and the permeability
(Antonellini and Aydin, 1994, 1995; Antonellini et
al., 1994). Also, it is possible for very fine-grained
clays to be dragged or injected into the fault zone
to form a seal (Aydin and Eyal, 2002). If the faults
are all of the sealing type, then they can divide the
reservoir into a number of isolated compartments
with little or no fluid communications between
compartments. The map shown in Fig. 1.10 suggests that the Brent Formation could be divided
into many compartments separated by the sealing
faults.
The presence of fault-sealed compartments
has important implications for locating wells and
producing the hydrocarbons from this reservoir
(Maerten et al., 1999, 2000). For example, if part of
the target reservoir is separated from a well by a
sealing fault those hydrocarbons beyond the fault
cannot be produced and another well must be
drilled. If the second well is drilled and it turns
out that the compartments do have a fluid connection because the faults are not continuous, or
do not extend as far as originally interpreted, the
unnecessary expense can be significant. Drilling
one additional well in these offshore settings can
be a multi-million dollar proposition. Clearly the
design of the production strategy is dependent
upon accurate interpretation of the fault geometry and sealing properties. Here we demonstrate
how models of fault slip distributions developed
by structural geologists can be helpful in making
such design decisions.
The two compartments of the Brent reservoir
in the Oseberg Syd Field are labeled in Fig. 1.10 as
Omega North and South. Both are bounded to the
west and east by major faults, presumed to be
sealing. Data from well No. 8, which intersects the
Brent Formation in Omega North, and data from
well No. 10, which intersects the formation in
Omega South, suggest that there is a fluid pressure difference between the northern and southern compartments. However, the interpreted fault
pattern does not define a complete compartment.
The two northwest–southeast striking faults just
to the southwest of Well No. 8 do not form a complete barrier between Omega South and Omega
North. Perhaps there are other explanations for
the pressure difference, but the hypothesis investigated here is that these two faults are actually
linked and thereby separate the reservoir into two
compartments.
This hypothesis was tested using a mechanical
model that relates fault geometry, rock properties,
and tectonic loading to fault slip (Crider and
Pollard, 1998; Maerten et al., 2000; Crider, 2001).
The modeling method subdivides faults into many
small triangular elements, as seen for a non-planar
fault with irregular tipline in Fig. 1.11, and these
elements approximate the three-dimensional
geometry. The model is based on elasticity theory
and the numerical method used to solve the governing equations is called the boundary element
method (Crouch and Starfield, 1983). Each element
14
MOTIVATIONS AND OPPORTUNITIES
section how the horizontal component of separation (sometimes called the heave) and the dip of
the normal fault are represented by the width of
the black stripes on maps. Greater stripe widths
(gray on the inset) imply greater horizontal separations of two points that originally were adjacent
to one another at the base of the Brent Formation.
Two faults with the same dip are displayed as
stripes with lesser and greater widths (horizontal
separations) corresponding to lesser and greater
dip separation. Two faults with the same dip separation are displayed as stripes with lesser and
greater widths corresponding to greater and
lesser dips. In regions where the dips of all the
normal faults are about the same, the widths of
the stripes on the structure map are used to infer
the relative magnitude of dip separations.
Similarly, if the dip does not vary along the length
of a fault, the changing width of the stripe is used
to infer changes in the distribution of dip separation.
From information gathered by drilling and
recovering core samples, measuring a variety of
indicators in the wells, and running flow and pressure tests in the sub-surface, it is believed that the
rocks in the fault zones shown in Fig. 1.10 are relatively impermeable to fluid flow compared to the
unfaulted portions of the sandstone reservoir.
Therefore these are referred to as sealing faults. A
number of mechanisms for changing the permeability of rock in a fault zone have been identified
(Jones et al., 1998). For example, the crushing of
sandstone grains during shearing in the fault
zone can fill the pores with fragments, thereby
decreasing the porosity and the permeability
(Antonellini and Aydin, 1994, 1995; Antonellini et
al., 1994). Also, it is possible for very fine-grained
clays to be dragged or injected into the fault zone
to form a seal (Aydin and Eyal, 2002). If the faults
are all of the sealing type, then they can divide the
reservoir into a number of isolated compartments
with little or no fluid communications between
compartments. The map shown in Fig. 1.10 suggests that the Brent Formation could be divided
into many compartments separated by the sealing
faults.
The presence of fault-sealed compartments
has important implications for locating wells and
producing the hydrocarbons from this reservoir
(Maerten et al., 1999, 2000). For example, if part of
the target reservoir is separated from a well by a
sealing fault those hydrocarbons beyond the fault
cannot be produced and another well must be
drilled. If the second well is drilled and it turns
out that the compartments do have a fluid connection because the faults are not continuous, or
do not extend as far as originally interpreted, the
unnecessary expense can be significant. Drilling
one additional well in these offshore settings can
be a multi-million dollar proposition. Clearly the
design of the production strategy is dependent
upon accurate interpretation of the fault geometry and sealing properties. Here we demonstrate
how models of fault slip distributions developed
by structural geologists can be helpful in making
such design decisions.
The two compartments of the Brent reservoir
in the Oseberg Syd Field are labeled in Fig. 1.10 as
Omega North and South. Both are bounded to the
west and east by major faults, presumed to be
sealing. Data from well No. 8, which intersects the
Brent Formation in Omega North, and data from
well No. 10, which intersects the formation in
Omega South, suggest that there is a fluid pressure difference between the northern and southern compartments. However, the interpreted fault
pattern does not define a complete compartment.
The two northwest–southeast striking faults just
to the southwest of Well No. 8 do not form a complete barrier between Omega South and Omega
North. Perhaps there are other explanations for
the pressure difference, but the hypothesis investigated here is that these two faults are actually
linked and thereby separate the reservoir into two
compartments.
This hypothesis was tested using a mechanical
model that relates fault geometry, rock properties,
and tectonic loading to fault slip (Crider and
Pollard, 1998; Maerten et al., 2000; Crider, 2001).
The modeling method subdivides faults into many
small triangular elements, as seen for a non-planar
fault with irregular tipline in Fig. 1.11, and these
elements approximate the three-dimensional
geometry. The model is based on elasticity theory
and the numerical method used to solve the governing equations is called the boundary element
method (Crouch and Starfield, 1983). Each element
14
MOTIVATIONS AND OPPORTUNITIES