zone itself but locally vary in orientation (Fig.
9.39b). The traces of bands are nearly straight on
cross sections parallel to the direction of offset,
but form a wavy pattern on cross sections perpendicular to the offset. Lenses of sandstone may be
preserved between wavy bands but the bands
rarely appear to cross one another. The thickness
of a zone increases simply by the addition of deformation bands. The relative shearing displacement
across a zone is the sum of that across the
members and may be as great as 25 to 30 cm for a
zone containing 100 bands.
The third stage in the development of these
faults is the localization of discrete surfaces (Fig.
9.38c) with slickenlines and striations indicating
slip. These slip surfaces are found on the margins
of thick zones of deformation bands (Fig. 9.39c)
and they accommodate several decimeters to
several meters of offset. There is a reduction in
grain size and porosity within a few millimeters
of the slip surfaces and the surfaces themselves
may be highly polished relative to the sandstone.
Investigations at other localities suggest that slip
surfaces nucleate in small patches which link to
form an anastomosing network (Shipton and
Cowie, 2001).
This review of the physical mechanisms
involved in faulting of granitic rock in the Sierra
Nevada (Fig. 9.36) and of sandstone in the San
Rafael Desert (Fig. 9.38) serves to make the point
that faulting is a more complicated process than
the development of opening fractures such as the
veins or dikes described in the previous section.
Multiple physical mechanisms are involved and
these may differ depending upon the rock type
and tectonic setting (Bürgmann et al., 1994). The
fault zones in granite begin to form with relatively little internal deformation except for a
single set of oblique fractures. The fault zones in
sandstone only form by the accumulation of
bands accounting for considerable internal deformation. The thickness of zones in granite is determined by the original spacing of joints whereas
the thickness of zones in sandstone is determined
by how many deformation bands cluster together.
In neither case is the growth of these faults
readily conceptualized in terms of the propagation of a fracture tipline and the increase in area
of two surfaces that formerly were bonded
together. It is conceivable that the slip surfaces in
sandstone propagate in this way, but they require
the previous development of the zone of deformation bands in order to nucleate.
The development of faults in granitic rock is
not expected to conform to the example presented
here unless a pre-existing set of weak surfaces is
present. If such anisotropy is absent, or if more
than one set exists, we would anticipate a different
outcome. For porous sandstone the sequence from
deformation band to slip surface described above
is not the only possibility. For example, in the presence of a set of pre-existing joints the mechanisms
involved in the evolution of faults in sandstone is
quite different (Flodin and Aydin, 2004; Meyers
and Aydin, 2004). Furthermore, for a different
lithology, such as interbedded limestone and
shale, the mechanisms are different than those
described for granite or sandstone (Peacock, 1991;
Peacock and Sanderson, 1991, 1994; Willemse et al.,
1997; Cooke, 1997). Some phenomena described
here may be explained using quasi-static elastic
models; others may require solutions for dynamic
elastic problems (Rice, 1980; Poliakov et al., 2002).
9.6 Concluding remarks
The dominant behavior of rock in Earth’s upper
crust is elastic and brittle at scales that range from
that of mineral grains to the crust itself. Brittle
deformation is manifest in rock structures including microcracks, joints, veins, dikes, deformation
bands, compaction bands, and faults. In the laboratory, extension and shear fractures form as
loading conditions reach the strength of samples.
While these laboratory experiments do not necessarily reproduce the mechanisms responsible for
outcrop-scale structures, they do provide important data regarding strength, friction, and fracture toughness. The strength of rock samples
typically increases with increasing confining pressure and decreases with increasing pore pressure.
The fact that tensile strength is about one order of
magnitude less than compressive strength helps
to rationalize the abundance of opening fractures
in the crust despite the nominal compressive
stress regime due to the overburden weight.
Griffith’s concept of stress concentration at flaws
382
BRITTLE BEHAVIOR
9.39b). The traces of bands are nearly straight on
cross sections parallel to the direction of offset,
but form a wavy pattern on cross sections perpendicular to the offset. Lenses of sandstone may be
preserved between wavy bands but the bands
rarely appear to cross one another. The thickness
of a zone increases simply by the addition of deformation bands. The relative shearing displacement
across a zone is the sum of that across the
members and may be as great as 25 to 30 cm for a
zone containing 100 bands.
The third stage in the development of these
faults is the localization of discrete surfaces (Fig.
9.38c) with slickenlines and striations indicating
slip. These slip surfaces are found on the margins
of thick zones of deformation bands (Fig. 9.39c)
and they accommodate several decimeters to
several meters of offset. There is a reduction in
grain size and porosity within a few millimeters
of the slip surfaces and the surfaces themselves
may be highly polished relative to the sandstone.
Investigations at other localities suggest that slip
surfaces nucleate in small patches which link to
form an anastomosing network (Shipton and
Cowie, 2001).
This review of the physical mechanisms
involved in faulting of granitic rock in the Sierra
Nevada (Fig. 9.36) and of sandstone in the San
Rafael Desert (Fig. 9.38) serves to make the point
that faulting is a more complicated process than
the development of opening fractures such as the
veins or dikes described in the previous section.
Multiple physical mechanisms are involved and
these may differ depending upon the rock type
and tectonic setting (Bürgmann et al., 1994). The
fault zones in granite begin to form with relatively little internal deformation except for a
single set of oblique fractures. The fault zones in
sandstone only form by the accumulation of
bands accounting for considerable internal deformation. The thickness of zones in granite is determined by the original spacing of joints whereas
the thickness of zones in sandstone is determined
by how many deformation bands cluster together.
In neither case is the growth of these faults
readily conceptualized in terms of the propagation of a fracture tipline and the increase in area
of two surfaces that formerly were bonded
together. It is conceivable that the slip surfaces in
sandstone propagate in this way, but they require
the previous development of the zone of deformation bands in order to nucleate.
The development of faults in granitic rock is
not expected to conform to the example presented
here unless a pre-existing set of weak surfaces is
present. If such anisotropy is absent, or if more
than one set exists, we would anticipate a different
outcome. For porous sandstone the sequence from
deformation band to slip surface described above
is not the only possibility. For example, in the presence of a set of pre-existing joints the mechanisms
involved in the evolution of faults in sandstone is
quite different (Flodin and Aydin, 2004; Meyers
and Aydin, 2004). Furthermore, for a different
lithology, such as interbedded limestone and
shale, the mechanisms are different than those
described for granite or sandstone (Peacock, 1991;
Peacock and Sanderson, 1991, 1994; Willemse et al.,
1997; Cooke, 1997). Some phenomena described
here may be explained using quasi-static elastic
models; others may require solutions for dynamic
elastic problems (Rice, 1980; Poliakov et al., 2002).
9.6 Concluding remarks
The dominant behavior of rock in Earth’s upper
crust is elastic and brittle at scales that range from
that of mineral grains to the crust itself. Brittle
deformation is manifest in rock structures including microcracks, joints, veins, dikes, deformation
bands, compaction bands, and faults. In the laboratory, extension and shear fractures form as
loading conditions reach the strength of samples.
While these laboratory experiments do not necessarily reproduce the mechanisms responsible for
outcrop-scale structures, they do provide important data regarding strength, friction, and fracture toughness. The strength of rock samples
typically increases with increasing confining pressure and decreases with increasing pore pressure.
The fact that tensile strength is about one order of
magnitude less than compressive strength helps
to rationalize the abundance of opening fractures
in the crust despite the nominal compressive
stress regime due to the overburden weight.
Griffith’s concept of stress concentration at flaws
382
BRITTLE BEHAVIOR