An experimental study of the
secondary deformation
produced by oblique-slip
normal faulting
AUTHORS
Roy W. Schlische ⬃ Department of
Geological Sciences, Rutgers University, 610
Taylor Road, Piscataway, New Jersey, 08854–
8066; [email protected]
Roy W. Schlische, Martha Oliver Withjack,
and Gloria Eisenstadt
ABSTRACT
We have used scaled clay models to define the secondary deformation produced by oblique-slip normal faulting. In the models,
the master fault beneath the clay layer dips 45⬚ and strikes at a 45⬚
angle relative to the heave direction (i.e., the horizontal component
of the displacement direction). Thus, the master fault has both normal dip-slip and strike-slip components of displacement. The modeling results show the following. (1) The fault patterns produced
by oblique-slip normal faulting vary significantly with depth. Secondary faults that strike obliquely to the master-fault trend are
more abundant near the top of the clay layer, whereas secondary
faults that are subparallel to the master-fault trend are more abundant at depth. (2) A single episode of oblique-slip normal faulting
produces two populations of secondary faults that have different
trends and ages. Secondary faults that strike obliquely to the masterfault trend are more abundant during the early stages of the experiments, whereas secondary faults that strike subparallel to the
master-fault trend are more abundant during the later stages of the
experiments. (3) Relay ramps between overlapping secondary synthetic normal faults are wide and temporally persistent in obliqueslip models. The ramps are cut by numerous small-scale normal
faults that are subparallel to the ramp-bounding faults. Cross faults
are uncommon and begin to develop only during the final stages of
the experiments. (4) Both map and cross section data are necessary
to distinguish among the deformation patterns produced by strikeslip, oblique-slip, and dip-slip faulting. The map views of obliqueslip models closely resemble those of strike-slip models; in both
models, many of the secondary faults strike obliquely to the masterfault trend. These map views, however, differ considerably from
those of dip-slip models, in which most of the secondary faults
strike subparallel to the master-fault trend. Alternatively, the cross
sectional views of oblique-slip models are similar to those of dip-slip
Copyright 䉷2002. The American Association of Petroleum Geologists. All rights reserved.
Manuscript received May 19, 2000; revised manuscript received December 3, 2001; final acceptance
January 16, 2002.
AAPG Bulletin, v. 86, no. 5 (May 2002), pp. 885–906
885
Roy W. Schlische is an associate professor of
structural geology at Rutgers University. He
received his B.A. degree from Rutgers
University and his M.A. degree and Ph.D. from
Columbia University, where he studied the
structural and stratigraphic evolution of
Triassic–Jurassic rift basins in eastern North
America. His research interests include
extensional tectonics, fault-population studies,
experimental modeling of geologic structures,
and basin inversion.
Martha Oliver Withjack ⬃ Department
of Geological Sciences, Rutgers University, 610
Taylor Road, Piscataway, New Jersey, 08854–
8066
Martha Oliver Withjack received her Ph.D.
from Brown University in 1978, studying the
mechanics of continental rifting. She currently
is a professor of structural geology at Rutgers
University. Before joining Rutgers, she worked
as a research geoscientist at Cities Service,
ARCO, and Mobil. Her research interests
include extensional, inversion, and salt
tectonics; physical and analytical modeling of
structures; and structural interpretation of
seismic data. She received the Matson
Memorial Award in 1999 and is currently first
vice chair of the Geological Society of
America’s structural geology and tectonics
division.
Gloria Eisenstadt ⬃ Department of
Geology, University of Texas at Arlington,
Arlington, Texas, 76019–0049
Gloria Eisenstadt is a consultant and adjunct
assistant professor at the University of Texas
at Arlington. Eisenstadt worked at Mobil
Technology Company from 1989 to 2000 as a
researcher, international structural consultant,
and technical teacher. Eisenstadt received her
B.A. and M.A. degrees in geology from
Temple University and her Ph.D. from the
Johns Hopkins University, where she studied
the structural evolution of the Innuitian fold
belt in Ellesmere Island, arctic Canada. Her
current interests are physical modeling and
seismic interpretation of inversion structures.
ACKNOWLEDGEMENTS
We thank Mobil Technology Company for
providing Schlische access to its experimental
equipment. This research was partially supported by National Science Foundation grant
EAR 9706199 to Schlische. Our work greatly
benefited from discussions with Rolf
Ackermann, Amy Clifton, Karen Fredricks,
Geof Geary, Peter Hennings, Judith Sheridan,
and Sarah Tindall and technical support from
Charlie Wall and Frank Roof. We appreciate
the careful and helpful reviews of Rolf
Ackermann, Mark Baum, Albert W. Bally, Amy
Clifton, Karen Fredricks, Barry McBride, and
Bruno Vendeville.
886
Oblique-Slip Normal Faulting
models; in both models, a highly faulted extensional forced fold
develops. These cross sectional views are dissimilar to those of
strike-slip models, which show no appreciable folding and no
change of regional level.
INTRODUCTION
The Earth’s crust is riddled with zones of weakness that can be
reactivated during subsequent tectonic events. The movement on
these reactivated zones of weakness is rarely either pure dip slip or
pure strike slip. For example, many Paleozoic contractional structures in eastern North America were reactivated during an episode
of northwest-southeast extension associated with Triassic–Jurassic
rifting (e.g., Ratcliffe and Burton, 1985; Olsen and Schlische, 1990)
(Figure 1). During rifting, northeast-striking Paleozoic reverse faults
became normal faults, whereas north-northwest– and east-northeast–striking structures became oblique-slip normal faults, having
significant components of right-lateral strike-slip and left-lateral
strike-slip, respectively. In particular, the east-northeast–striking,
rift-basin boundary faults in the narrow-neck region between the
Newark and Gettysburg basins (NN in Figure 1) and in the eastern
Fundy basin (MFZ in Figure 1) had a significant left-lateral component of movement during early Mesozoic rifting (e.g., Olsen and
Schlische, 1990).
Although oblique slip along reactivated faults is likely to be the
rule rather than the exception, few experimental studies have simulated oblique-slip normal faulting. Numerous workers used clay
and sand models to study pure dip-slip normal faulting (e.g., Horsfield, 1977; Tsuneishi, 1978; Vendeville, 1987, 1988; Withjack et
al., 1990; Withjack and Callaway, 2000; Ackermann et al., 2001)
and pure strike-slip faulting (e.g., Cloos, 1928; Riedel, 1929; Wilcox et al., 1973; Naylor et al., 1986). Withjack and Jamison (1986),
Tron and Brun (1991), Smith and Durney (1992), McClay and
White (1996), and Clifton et al. (2000) used clay and sand models
to investigate oblique rifting. In these experiments, deformation
occurred in a wide zone above a flat base rather than in a narrow
zone above a dipping, oblique-slip normal fault. Richard (1991) and
Higgins and Harris (1997) experimentally simulated oblique-slip
normal faulting. In their models, either a single layer of sand or two
layers composed of sand and silicone putty covered moderately dipping master faults. During the experiments, the master faults were
reactivated with both normal dip-slip and strike-slip components
of displacement, causing the overlying layers to deform.
The goal of this article is to supplement this previous work by
using scaled, single-layer clay models. Our specific objectives are to
(1) better define the range of secondary fault and fold patterns produced by oblique-slip normal faulting, (2) determine how the magnitude of the displacement on the master fault and the thickness of
the sedimentary cover influence the secondary fault and fold patterns in map view, and (3) develop criteria to distinguish the de-
75
80
70
ure
North America
45
N
E. Jurassic
extension direction
G
NN
N
40
40
Atlantic Ocean
Figure 1. Simplified map of
eastern North America showing
major faults formed during
Appalachian orogenesis and
Triassic–Jurassic rift basins
whose border faults, in most
cases, represent reactivated Paleozoic faults. The Early Jurassic
regional extension direction was
oblique to many of the preexisting faults, particularly in
the narrow neck (NN) between
the Newark (N) and Gettysburg
(G) basins and the Minas fault
zone (MFZ) in Nova Scotia,
Canada. Modified from
Schlische (1993).
Early Mesozoic
rift basin
Major Paleozoic fault
0
formation patterns produced by oblique-slip normal
faulting from those produced by dip-slip normal faulting and strike-slip faulting. The following sections describe the experimental procedure; review our modeling results and compare them with those from
previous studies; describe the implications of our modeling results for hydrocarbon exploration and production; and discuss how the models compare with actual
examples of oblique-slip normal faults from the eastern
Fundy basin of Nova Scotia, Canada, and the Carnarvon basin on the northwest shelf of Australia.
EXPERIMENTAL PROCEDURE
The modeling apparatus (Figure 2a) has three configurations: dip slip, strike slip, and oblique slip. In the
dip-slip configuration, the apparatus has a fixed footwall block and a mobile hanging-wall block (Figure
2b). The top surfaces of the blocks are initially level,
together forming a flat, 32 ⳯ 61 cm surface. The master fault between the blocks dips 45⬚, and its strike is
perpendicular to the heave direction, which we define
as the horizontal component of the displacement direction (following Bates and Jackson, 1987, p. 286).
km
65
35
400
Consequently, as the hanging-wall block slides down
the surface of the master fault during the experiment,
the master fault undergoes pure normal dip-slip displacement. In the strike-slip configuration, the apparatus consists of a fixed plate and a juxtaposed mobile
plate (Figure 2b). Together, the plates form a level surface (61 ⳯ 61 cm). The strike of the master fault between the plates is parallel to the displacement direction of the moving plate. Consequently, the master
fault undergoes pure left-lateral, strike-slip displacement during the experiment. In the oblique-slip configuration, the apparatus has a fixed footwall block and
a mobile hanging-wall block. The top surfaces of the
blocks are initially level, together forming a flat, 61 ⳯
61 cm surface. The master fault dips 45⬚ and strikes at
a 45⬚ angle relative to the heave direction. During the
experiments, the hanging-wall block slides down the
surface of the master fault and undergoes equal
amounts of normal dip-slip and left-lateral, strike-slip
displacement.
In our experiments, a layer of wet clay simulates
the sedimentary cover above the master fault. In several models, the clay layer is subdivided into colored
sublayers. These sublayers are mechanically identical
and are used as marker horizons to identify faults and
Schlische et al.
887
Figure 2. (a) Generalized
sketch of experimental modeling apparatus. The rigid
hanging-wall block of the master fault is attached to the
moveable wall, whereas the
rigid footwall block is attached
to the opposing fixed wall. Outward displacement of the
moveable wall causes the
hanging-wall block to slide
down the surface of the master
fault. (b) Configuration of rigid
blocks and master faults before
and after deformation for dipslip, oblique-slip, and strike-slip
faulting.
Fixed
Wall
Fixed
Wall
Clay with
colored
sublayers
Moveable
Wall
Fixed
Wall Fixed block
(a)
Master
Fault
Moveable
block
Stationary
Baseplate
Oblique Slip
Dip Slip
Strike Slip
(b)
90
45
HW
FW
HW
FW
90
45
FW
45
HW
HW
FW
folds in cross sectional view. The wet clay is composed
predominantly of kaolinite and water (⬃40% by
weight). Its density is about 1600 kg mⳮ3. The clay has
a low cohesion (⬃50 Pa) and a coefficient of internal
friction of about 0.5 (Fugro-McClelland, Inc., 1992,
personal communication; Sims, 1993; S. Dixon, 1996,
personal communication). Distributed cataclasis, as
defined by Rutter (1986), is the primary deformation
style in the wet clay when strains are small. With increasing strain, however, deformation becomes localized, discrete faults develop, and cataclastic faulting is
the primary deformation style (Withjack and Callaway, 2000). We preferred wet clay, rather than dry
sand, as the modeling medium because (1) both faults
and folds form in the wet clay; (2) more faults develop,
propagate, and link in wet clay than in dry sand; and
(3) fault zones are narrower and better defined in wet
clay than in dry sand (e.g., Withjack and Callaway,
2000).
Tables 1 and 2 provide specific information about
the experiments. We repeated most of the experi888
Oblique-Slip Normal Faulting
ments to verify the reproducibility of the modeling
results. In the experiments, the horizontal displacement rate of the moving wall and, thus, the rate of
heave on the master fault is 4 cm hrⳮ1. In most experiments, the clay layer is initially 4 cm thick. In
experiment III, however, the clay is initially 2 cm
thick. During the experiments, we photographed the
upper surface of the clay layer at regular increments
of displacement on the master fault (using multiple
lighting directions) to document the surface deformation through time. We used a low-angle light
source to emphasize the fault scarps (Figure 3a).
During experiments V and VII, we filled in the subsiding hanging wall with clay at regular increments
of the displacement on the master fault. These
growth layers initially had a horizontal upper surface.
After models IV, V, VI, and VII had dried, we cut
them into 1–2 cm–wide strips. These strips were
photographed, and some strips were sectioned vertically and horizontally to document the final deformation state in detail.
Table 1. Description of Models Used for Map Views
Experiment Number
I
II
IIA, IIB
III
IV
Experimental Configuration
Maximum Heave
(cm)
Maximum Throw
(cm)
Initial Overburden Thickness
(cm)
Dip slip
Oblique slip
Oblique slip
Oblique slip
Strike slip
2.8
2.8
3.8
2.8
2.8
2.8
1.9
2.7
1.9
0
⬃4
⬃4
⬃4
⬃2
⬃4
Table 2. Description of Models Used for Cross Section Views
Experiment Number
V
VI
VII
IV
Experimental Configuration
Maximum Heave
(cm)
Maximum Throw
(cm)
Initial Overburden Thickness
(cm)
Dip slip with growth beds
Oblique slip
Oblique slip with growth beds
Strike slip
2.0
2.8
2.9
2.8
2.0
1.9
2.0
0
⬃4
⬃4
⬃4
⬃4
For the experiments to be valid, the models must
be geometrically, kinematically, and dynamically similar to their geological counterparts (e.g., Hubbert,
1937). The scaling factor between the models and
geologic prototype is 104 ⳮ 105 (Appendix). Thus, a
4 cm–thick layer of wet clay in the models corresponds to a 0.4–4 km–thick sedimentary package in
nature. Also, 1 cm of displacement on the master fault
in the models corresponds to 0.1–1 km of fault displacement in nature. We emphasize that the experimental models are not exact scale models of natural
systems. Rock may deform differently than the clay in
the models. For example, rock that has preexisting
heterogeneities (e.g., fractures, bedding) may behave
very differently than the relatively homogeneous clay
in the models.
MAP-VIEW DEFORMATION
We used photographs of the clay surface to define the
map-view deformation. In the photographs, faults appear either very bright or very dark, depending on their
dip direction relative to the light source (Figure 3a).
The fault maps in Figures 4 and 5 represent the faults
as polygons bounded by the footwall and hanging-wall
cutoffs of the top of the clay layer. The width of the
polygon is approximately equal to the heave on the
faults. For ease of comparison, we oriented all of
the fault maps so that the master fault trends left to
right, defined as an azimuth of 090⬚. Also, we divided
the secondary faults in the dip-slip and oblique-slip experiments into those that dip in the same direction as
the master fault (synthetic faults) and those that dip in
the opposite direction from the master fault (antithetic
faults). Finally, we divided each of the secondary faults
into segments of uniform trend. The graphs in Figures
4 and 5 show the summed length of these fault segments vs. the trend of the fault segments (binned in 5⬚
intervals).
Experiment I: Dip-Slip Master Fault
During the early stages of experiment I, an extensional
forced fold develops in the clay above the master
fault. Soon after, synthetic and antithetic secondary
normal faults form in a wide zone above the master
normal fault (Figure 4a). The strike of most of the secondary normal faults is subparallel to the master-fault
trend (Figure 4b). Initially, the secondary normal faults
are isolated features with maximum displacements
near the center of their fault traces and diminishing
displacements toward their tips. They grow predominantly by tip propagation. As the experiment progresses, however, many of the propagating secondary
normal faults overlap each other, and narrow relay
ramps develop between them. Cross faults (e.g., CF in
Figure 4a) soon breach the narrow relay ramps,
Schlische et al.
889
Figure 3. (a) Overhead photograph of the top surface of
oblique-slip model (experiment
II) shown at 2.0 cm of heave
on the master fault. Synthetic
faults dip toward the light
source and are bright; antithetic
faults dip away from the light
source and are dark. HW is the
hanging-wall region of the master fault; FW is the footwall region. (b) Oblique view of deformed zone of oblique-slip
experiment IIA at 3.5 cm of
heave on the master fault
showing large en echelon synthetic normal faults separated
by relay ramps. With the exception of the incipient rampbreaching fault, most faults in
the relay ramps are subparallel
to the large synthetic faults. The
long axes of the strain ellipses
give the direction of maximum
horizontal extension. The scale
bar is in centimeters.
forming throughgoing fault zones during the early
stages of experiment I. These throughgoing fault zones
are subparallel to the master-fault trend and have normal dip-slip displacement. Fault linkage and the de890
Oblique-Slip Normal Faulting
velopment of throughgoing faults continue throughout
the duration of experiment I (Figure 5a, b). The
throughgoing fault zones in experiment I exhibit little
along-strike variation in displacement, although minor
Schlische et al.
891
CF
10 cm
Passive marker
on clay surface
L
1.5
cm
(g)
(e)
(c)
(a)
0
5
10
15
20
0
5
10
15
20
25
30
35
40
45
0
5
10
15
20
25
30
35
40
45
0
10
20
30
40
50
60
70
80
90
0
0
0
0
30
30
30
30
90
Trend
60
120 150
120 150
MF=D
90
D
120 150
D
120 150
MF
Trend
60
DN
90
MF
Trend
60
DN
90
Trend
60
MF=DN
(b)
180
(h)
180
(f)
180
(d)
180
D
Figure 4. Comparison of fault-trace maps for (a) dip-slip model, (c) thick oblique-slip model (4 cm–thick clay layer), (e) thin oblique-slip model (2 cm–thick clay layer), and
(g) strike-slip model shown at 1.5 cm of heave on the master fault. CF denotes cross faults, and L shows faults that will link in Figure 5a. Light gray areas in (a), (c), and (e) denote
the region between the footwall and hanging-wall cutoffs at the base of the clay layer; synthetic faults are black, antithetic faults are dark gray. Dashed lines in (g) are surficial
markers, initially straight, on the clay surface. Graphs (b, d, f, h) show the summed length of fault segments vs. the trend of fault segments binned in 5⬚ intervals. MF is the
master-fault trend, D is the trend of the displacement direction, and DN is the displacement-normal trend.
Experiment IV: Strike slip
Heave direction
Master fault
Experiment III: Oblique Slip (Thin)
Heave direction
Master fault
Experiment II: Oblique Slip (Thick)
Experiment I: Dip Slip
Heave direction
Master fault
Summed Length (cm)
Summed Length (cm)
Summed Length (cm)
Summed Length (cm)
CF
Passive marker
on clay surface
10 cm
2.75
cm
(g)
(e)
(c)
(a)
0
5
10
15
20
0
5
10
15
20
25
30
35
40
45
0
5
10
15
20
25
30
35
40
45
0
10
20
30
40
50
60
70
80
90
0
0
0
0
30
30
30
DN
30
60
60
DN
60
D
120 150
(b)
(d)
180
D
D
(f)
MF = D
120 150 180
(h)
120 150 180
Trend
90
90
Trend
MF
90 120 150 180
Trend
MF
90
Trend
60
MF=DN
Figure 5. Comparison of fault trace maps for (a) dip-slip model, (c) thick oblique-slip model, (e) thin oblique-slip model, and (g) strike-slip model shown at 2.75 cm of heave
on the master fault. CF denotes cross faults; L denotes complex fault that formed by linkage. Light gray areas in (a), (c), and (e) denote the region between the footwall and
hanging-wall cutoffs at the base of the clay layer; synthetic faults are black, antithetic faults are dark gray. Dashed lines in (g) are surficial markers, initially straight, on the clay
surface. Graphs (b, d, f, h) show the summed length of fault segments vs. the trend of fault segments binned in 5⬚ intervals. MF is the master-fault trend, D is the trend of the
displacement direction, and DN is the displacement-normal trend.
Experiment IV: Strike slip
Heave direction
Master fault
Experiment III: Oblique Slip (Thin)
Heave direction
Master fault
Experiment II: Oblique Slip (Thick)
Experiment I: Dip Slip
Heave direction
Master fault
Summed Length (cm)
Summed Length (cm)
Summed Length (cm)
Oblique-Slip Normal Faulting
Summed Length (cm)
892
displacement minima mark areas where fault segments
had linked.
Experiment II: Oblique-Slip Master Fault
Experiment II is similar to experiment I, except that
the master fault has oblique-slip displacement rather
than dip-slip displacement. In both experiments, an
extensional forced fold develops in the clay above the
master fault, and secondary normal faults form in a
wide zone above the master normal fault (Figures 4c,
5c). Unlike experiment I, however, the strike of most
of the secondary normal faults in experiment II is
oblique to the strike of the master fault (Figures 4d,
5d).
In both experiments I and II, the secondary normal
faults are initially isolated features with maximum displacements near the center of their fault traces and
diminishing displacements toward their tips. Fault
linkage and the development of throughgoing faults,
however, are delayed in experiment II compared to experiment I. During the later stages of experiment II,
some of the secondary normal faults begin to link, resulting in complex fault traces (L in Figure 5c). Generally, the synthetic secondary faults grow longer and
have more displacement than their antithetic counterparts. Well-developed relay ramps form between large,
overlapping (en echelon) synthetic faults (Figures 3b,
5c). The relay ramps are cut by minor, mostly antithetic faults subparallel to the synthetic faults that
bound the ramps. The relay ramps are never fully
breached during experiment II. Cross faults begin to
develop and cut the relay ramps only during the final
stages of the experiment (Figure 5c; also see Figure 3b
for an oblique-slip experiment with larger displacement on the master fault).
Experiment III: Oblique-Slip Master Fault (Thin Cover)
Experiment III (Figures 4e, 5e) is identical with experiment II, except that the clay layer is 2 cm thick
rather than 4 cm thick (Table 1). In both experiments II and III, an extensional forced fold develops
in the clay above the master fault, and secondary normal faults form in a wide zone located above the
master fault. In both experiments, the strike of most
of the secondary faults is oblique to the strike of the
master fault (Figures 4f, 5f). In experiment III, however, the secondary normal faults form and link together much sooner than those in experiment II. In
fact, the linkage of secondary faults forms a through-
going oblique-slip fault zone subparallel to the underlying master fault during the early stages of experiment III (Figure 4e). The surface trace of this
throughgoing fault zone is complex, consisting of
both oblique-fault segments and linking-fault segments oriented subparallel to the master-fault trend.
The hanging wall of the throughgoing fault zone contains numerous synthetic faults that are oblique to
the master-fault trend. Antithetic faults are much less
common in the thin model than in the thick model.
Also, the deformed zone is narrower in the thin
model than in the thick model.
Experiment IV: Strike-Slip Master Fault
As in the oblique-slip and dip-slip experiments, the
secondary faults in the strike-slip experiment develop
in a zone located above the master fault (Figures 4g,
5g). The deformed zone, however, is much narrower
in the strike-slip model than in the oblique-slip and
dip-slip models. The strike of most of the early formed
secondary faults in the strike-slip model is oblique to
the master-fault trend (Figure 4h), similar to the strike
of the early formed secondary faults in the oblique-slip
experiments (Figure 4d, f). Unlike the oblique-slip experiments, however, the early formed secondary faults
in the strike-slip experiment have predominantly leftlateral displacement rather than normal displacement.
The fault pattern observed in experiment IV is similar
to the R1 Riedel shears produced in previous experimental models of strike-slip fault systems (e.g., Cloos,
1928; Riedel, 1929; Tchalenko, 1970; Wilcox et al.,
1973; Naylor et al., 1986). During the final stages of
experiment IV, faults subparallel to the master-fault
trend develop, cutting the clay surface and linking the
early formed strike-slip faults (Figure 5g, h). An incipient throughgoing strike-slip fault zone develops during the final stages of experiment IV.
Summary of Map-View Deformation
Comparisons of the fault-trace maps (Figures 4, 5)
show the following. (1) A zone of folding and/or faulting forms in the clay above the master fault in all experiments. The deformation zone is wider in the dipslip and oblique-slip experiments than in the strike-slip
experiment. (2) Relay ramps develop between overlapping normal faults in the dip-slip and oblique-slip
experiments. In the dip-slip experiments, the relay
ramps are narrow and breached by cross faults.
(3) Throughgoing fault zones, subparallel to the
Schlische et al.
893
master-fault trend, develop in the dip-slip and strikeslip experiments but not in the oblique-slip experiment with the thick clay layer (experiment II). Apparently, the oblique orientation of the widely spaced
secondary faults in experiment II prevented the secondary faults from readily propagating along strike to
link with each other. Given enough displacement on
the master fault, however, the ramps between overlapping faults in the oblique-slip experiment would likely
become breached by cross faults and a throughgoing
oblique-slip fault system would develop. A throughgoing oblique-slip fault subparallel to the master-fault
trend did develop in the oblique-slip experiment with
the thin clay layer (experiment III). (4) The type and
orientation of secondary faults in the experimental
models depend on the type and orientation of the master fault relative to the heave direction. Most secondary
faults in the dip-slip experiment have normal displacements and are subparallel to the master-fault trend.
Most secondary faults in the thick oblique-slip experiment have normal displacements and are oblique to
the master-fault trend. In the strike-slip model, the secondary faults have strike-slip displacements and are
both parallel and oblique to the master-fault trend.
The orientation (but not the slip) of the faults that
form during the early stages of the oblique-slip experiment closely resembles those that form during the
early stages of the strike-slip experiment. Therefore,
the modeling results indicate that map-view fault patterns are insufficient to distinguish oblique-slip faulting
from strike-slip faulting. Map-view fault patterns,
however, are sufficient to distinguish dip-slip faulting
from oblique-slip and strike-slip faulting.
the master-fault trend or parallel to the heave
direction.
Experiment V: Dip Slip
The cross sections from the dip-slip experiment consist
of an extensional forced fold dissected by numerous
secondary faults (Figure 6a). Surface observations indicate that the extensional forced folding occurs before
and during the development of the secondary faults.
All of the secondary faults have normal separation.
Surface observations show that the secondary faults at
the model surface have normal displacement. Antithetic normal faults are more common in the upper
part of the cross sections. Many of the antithetic faults
are blind, dying out both upward and downward. The
surface of many of the antithetic faults in the footwall
region is curved, steepening with depth. These antithetic faults are probably related to bending associated
with the extensional forced folding. The upper, gently
dipping parts of these fault surfaces were initially more
steeply dipping but were rotated to shallower dips by
the subsequent forced folding. Synthetic normal faults
are more common in the lower part of the cross sections. Most of the synthetic faults splay off the master
fault and die out upward. The synthetic faults are probably related to aborted attempts of the master fault to
propagate upward through the folded and attenuated
clay layer. The secondary faults with the greatest displacement are the synthetic faults that emanate from
the footwall cutoff of the base of the clay on the master
fault. Synthetic faults preferentially propagate into the
growth layers.
Experiments VI and VII: Oblique Slip
DEFORMATION IN CROSS SECTIONAL
VIEW
We used vertical sections from the dried, layered models to define the deformation in cross sectional view
(Figure 6). All sections were decompacted to their approximate original thicknesses before drying. The dipslip and oblique-slip cross sections have approximately
the same throw (⬃2.0 cm; Table 2), whereas the
strike-slip and oblique-slip sections have approximately the same heave (⬃2.8 cm; Table 2). Dip-slip
cross sections are perpendicular to the master-fault
trend and parallel to the heave direction; strike-slip
cross sections are perpendicular to both the masterfault trend and the displacement direction; and
oblique-slip cross sections are either perpendicular to
894
Oblique-Slip Normal Faulting
The cross sections from the oblique-slip experiments
closely resemble those from the dip-slip experiment.
The cross sections from experiments VI and VII consist
of a highly faulted extensional forced fold in the clay
above the master fault (Figure 6b). All of the secondary
faults in the cross sections have normal separation. Surface observations show that the secondary faults at the
model surface have mostly normal displacement. Some
of the deeper secondary normal faults in the cross sections from experiments VI and VII, however, may have
oblique-slip displacement. Antithetic faults are more
common in the upper part of the cross sections. Many
of the antithetic faults are blind, dying out both upward and downward. As in the dip-slip cross sections,
the surfaces of many of the antithetic faults in the
Growth
Layers
PreGrowth
Layers
(Cover
Sequence)
Fixed wall
Rigid
Block
10 cm
12L
15L
Moving wall
(a)
Experiment V: Dip Slip
Experiment VI:
Oblique Slip
Experiment VII:
Oblique Slip
Fixed wall
G10R
10 cm
8R
G20L
18L
Moving wall
(b)
5 cm
Moving
plate
10 cm
8T
12B
13T
17B Fixed
plate
Experiment IV: Strike Slip
Movement
out of page
(c)
Movement
into page
Figure 6. Cross sectional geometry of (a) dip-slip models, (b) oblique-slip models, and (c) strike-slip models. Inset maps show
location of sections with respect to master-fault geometry. Growth layers are light gray, pregrowth layers are alternating dark gray
and white, and rigid fault blocks are dark gray. All sections are decompacted to their approximate original thicknesses before drying.
footwall region are curved, steepening with depth as a
result of the forced folding. Synthetic faults are more
common in the lower part of the cross sections. Most
of the synthetic faults splay off the master fault and die
out upward. The secondary faults that have the greatest displacement are the synthetic faults emanating
Schlische et al.
895
from the footwall cutoff of the base of the clay layer
on the master fault. More synthetic faults than antithetic faults propagate into the growth layers.
Experiment IV: Strike Slip
The cross sections from the strike-slip experiment (Figure 6c) show that a narrow, steeply dipping fault zone
is present in the clay layer directly above the master
fault. The secondary faults are high-angle splays off the
master fault and have very small reverse or normal separations (most likely a result of the lateral juxtaposition
of small irregularities in the clay sublayers). Surface
observations indicate that the secondary faults at the
model surface have predominantly left-lateral, strikeslip displacement. Cross sectional geometries change
along strike. For example, the attitude of the deformation zone in adjacent cross sections 13T and 12B
changes from right-dipping to left-dipping (Figure 6c).
No appreciable folding occurs in any of the cross
sections.
Summary of Deformation in Cross Sectional View
The oblique-slip and dip-slip models are similar in
cross sectional view. Cross sections in both models consist of a highly faulted extensional forced fold in which
the regional level steps down from the footwall block
to the hanging-wall block. Thus, the modeling results
suggest that cross sectional data alone are insufficient
to distinguish between dip-slip and oblique-slip faulting. The deformation patterns in the strike-slip cross
sections differ significantly from those in the obliqueslip and dip-slip cross sections. No appreciable folding,
no change in regional level, and no significant normal
separation occur on any of the secondary faults in the
strike-slip cross sections. Therefore, the modeling results suggest that cross sectional data alone are sufficient to distinguish strike-slip faulting from dip-slip
and oblique-slip faulting.
CHANGE IN FAULT GEOMETRY WITH
DEPTH
Selected pieces of experiment VI, an oblique-slip
model, were vertically and horizontally sectioned to
determine the geometries of the secondary structures
at depth. The spacing between the vertical sections is
approximately 1 mm. For each vertical slice, we defined the topography of three horizons (Figure 7a):
896
Oblique-Slip Normal Faulting
horizon 1 located in the upper half of the model, horizon 2 located in the lower half of the model, and the
base of the clay layer. Figure 7b–d shows the topography of the three horizons for all of the vertical slices
(aligned relative to the footwall cutoff of the base of
the clay layer on the underlying master fault). At the
shallow level of horizon 1, the secondary faults form a
series of en echelon horsts (shaded gray in Figure 7b)
and grabens. These secondary faults are located above
the footwall of the master fault, their strike is oblique
to that of the master fault, and they die out laterally
along strike. At the deep level of horizon 2, most deformation is localized on a series of gently dipping synthetic faults with significant displacement. These faults
are subparallel to the master fault, and they splay off
the master fault (Figure 7a). The strike of many of the
minor secondary faults, however, is oblique to the
master-fault trend. These minor faults appear to be cut
by the gently dipping synthetic faults. The fault patterns observed on horizontal slices through experiment
VI confirm this observation (Figure 8). Most of the secondary faults are subparallel to the master fault trend
in the lower parts of the clay model and are oblique to
the master-fault trend in the upper parts of the model.
These observations suggest that the secondary faults
associated with oblique-slip normal faulting have different orientations at different structural levels. If the
faults at depth are physically connected to the shallower faults, then these faults have a helicoidal or propeller blade–shaped geometry.
DISCUSSION
Implications of Modeling Results
Our modeling results have the following implications
for structural interpretation and for hydrocarbon exploration and production.
1. The unambiguous recognition of oblique-slip normal faulting requires observations from both map
and cross sectional views. In map view, the fault
patterns in the oblique-slip models resemble those
in the strike-slip models (Figures 4, 5). In both
cases, two fault trends develop in the cover sequence. One fault trend is oblique to the trend of
the underlying master fault, and one fault trend is
subparallel to the trend of the underlying master
fault. In cross sectional view, the fault patterns in
the oblique-slip models resemble those in the dip-
3 cm
Footwall cutoff of base of clay
(a)
Section in (a)
(b)
Horst blocks
Figure 7. Topography of
three horizons derived from
vertical serial sections through
part of experiment VI (obliqueslip normal faulting). (a) Photograph of a representative serial
section showing the location of
the three analyzed horizons.
Topography of horizon 1 (b),
horizon 2 (c), and base of clay
layer (d) for all of the serial
sections. Thin gray lines denote
bedding; thin black lines denote
faults; gray areas in (b) show
top of horst blocks. All sections
are aligned relative to the footwall cutoff at the base of the
clay layer. The vertical scale in
(b), (c), and (d) is the true
scale; the horizontal spacing between adjacent sections has
been increased by a factor of
approximately 1.3 to reduce
crowding.
Section in (a)
(c)
Section in (a)
(d)
slip models (Figure 6). In both cases, the regional
level of the cover sequence drops across the master
fault. A combination of extensional forced folding
and faulting accommodates the drop in regional
level.
2. The presence of two fault populations that have different trends and ages need not indicate two distinct
episodes of regional extension. A single episode of
oblique-slip normal faulting can produce such fault
populations. In the oblique-slip models, two populations of secondary faults develop in the cover sequence (Figures 4, 5). One population consists of
early formed faults whose strike is oblique to the
trend of the underlying master fault. The second
population consists of later formed throughgoing
faults whose strike is subparallel to the trend of the
master fault. The faults of the second population
generally cut the faults of the first population.
Schlische et al.
897
Level 1
1 cm
FW cutoff of base of clay
Level 2
(a)
Map view
of Level 1
1 cm
(b)
Map view
of Level 2
(c)
Figure 8. Horizontal sections through part of experiment VI (oblique-slip normal faulting). (a) Photograph of vertical side of sectioned
piece showing the locations of the two map levels (b and c). The inset diagrams show the average trend of small faults (thin black
line) relative to the trend of the master fault (thick gray line). The white arrow gives the heave direction.
3. The secondary structures associated with obliqueslip normal faulting are likely to vary with depth. In
the upper part of the oblique-slip models with the
thick clay layer (experiments II and VI), secondary
structures include en echelon horsts, grabens, and
tilted fault blocks (Figures 3b, 7b). The strike of
these secondary structures is oblique to the trend of
the master fault, and many of the horst blocks and
footwall highs have three-way closure. In the lower
part of these oblique-slip models, throughgoing
oblique-slip normal faults are the primary structures (Figure 7c). These faults link with the master
fault, and their strike is subparallel to that of the
master fault. Fewer structures with three-way clo898
Oblique-Slip Normal Faulting
sure exist in the lower part of the oblique-slip
models.
4. Long-lived, well-developed relay ramps are more
likely to form in the sedimentary cover above an
oblique-slip normal fault than a dip-slip normal
fault. In the models of oblique-slip normal faulting
with the thick clay layer, relay ramps are wide and
persist throughout the duration of the experiment.
In identical models of dip-slip normal faulting
(where the heave and fault dip are identical), relay
ramps are narrow and become breached by cross
faults during the early stages of the experiment (Figures 4, 5). Previous studies have shown that relay
ramps can have a profound impact on the petro-
leum system. They are preferred sites of sediment
transport from eroded footwall highs into hangingwall basins (e.g., Gawthorpe and Hurst, 1993; Peacock and Sanderson, 1994; Childs et al., 1995) (Figure 9), and they are potential pathways for
hydrocarbon migration from hanging-wall kitchens
into footwall structural traps.
5. The small-scale faulting affecting relay ramps can
vary considerably. In the models of oblique-slip normal faulting with the thick clay layer, relay ramps
are cut by numerous small-scale normal faults that
are subparallel to the ramp-bounding faults (Figure
5c). Cross faults are uncommon and begin to develop only during the final stages of the experiment
when the displacement on the master fault is large
(incipient ramp-breaching fault in Figure 3b). In
identical models of dip-slip normal faulting, relay
ramps are cut by some small-scale normal faults that
are oblique to the ramp-bounding faults. These
cross faults are abundant and developed throughout
the duration of the experiment (Figures 4a, 5a).
This latter fault pattern resembles the fault pattern
conventionally assumed for relay ramps (e.g., Peacock and Sanderson, 1994). The small-scale faults
that cut or breach relay ramps can retard or enhance
fluid migration.
Previous Modeling Studies and Deformational Styles
Associated with Oblique Deformation
Richard (1991) and Higgins and Harris (1997) also experimentally studied the secondary deformation associated with oblique-slip normal faulting (in which the
strike of the fault is at an approximately 45⬚ angle to
the heave direction) (Figure 10). In their models, the
cover sequence above the master fault was composed
of either a layer of dry sand or a layer of dry sand over
a layer of silicone putty. Dry sand deforms primarily
by cataclastic faulting, whereas silicone putty deforms
by viscous flow. Direct comparisons of these models
with our models are difficult because boundary conditions (e.g., displacement on the master fault, thickness of the cover sequence) and modeling materials
were substantially different. Nevertheless, comparisons suggest that three distinct deformation styles can
develop in the cover sequence above an oblique-slip
normal fault: partitioned, focused, and distributed
(Figure 11).
In the pure sand model of Richard (1991) (Figure
10a), deformation in the cover sequence is partitioned.
A steeply dipping, synthetic normal fault accommodates most of the dip-slip component of displacement
on the master fault, whereas a high-angle strike-slip
fault zone accommodates most of the strike-slip component of displacement. In map view, all faults are subparallel to the master fault. In cross sectional view, all
faults link directly with the master fault. During the
later stages of our thin clay model (experiment III; Figure 5e), and in the deeper parts of our other clay models (experiment VI; Figures 7c, 8c), most deformation
in the cover sequence is focused on a throughgoing
oblique-slip normal fault. In map view, these faults are
subparallel to the master fault. In cross sectional view,
they link directly with the master fault. Finally, in the
sand/putty models of Richard (1991) (Figure 10b),
during the early stages of our thin clay model (experiment III; Figure 4e), and in the shallow parts of our
other clay models (experiments II and VI; Figures 5c,
Experiment IIB
Figure 9. Inferred sedimenttransport pathways (arrows) associated with relay ramps between en echelon synthetic
faults in the model of obliqueslip normal faulting (experiment
IIB, shown at 3.75 cm of heave
on the master fault). Sediment
eroded from footwall highs
would move down the ramps
into hanging-wall lows.
Schlische et al.
899
A⬘
9 cm
A⬘
A
A
(a)
Sand
Geologic Examples
Rigid block
In this section, we show that our models reproduce
many of the key features observed in a known geologic
example (Minas subbasin, Canada) of oblique-slip
normal faulting. We then show that the modeling results can be used to infer oblique-slip normal faulting
for another geologic example (Dampier subbasin,
Australia).
B⬘
B
B⬘
Sand
B
9 cm
Putty
Rigid block
(b)
Figure 10. Results from Richard’s (1991) study of oblique-slip
normal faulting in which the master fault dips 45⬚ and strikes
45⬚ relative to the heave direction. (a) Fault-trace maps and
cross sections from experiment with sand over rigid blocks. The
thickness of the sand is 8 cm. (b) Fault-trace maps and cross
sections from experiment with sand over putty over rigid blocks.
The thickness of the sand is approximately 6 cm, and the thickness of the putty is approximately 2 cm.
7b, 8b), deformation in the cover sequence is distributed. Secondary normal faulting and extensional
forced folding accommodate this distributed deformation. In map view, the axes of the extensional forced
folds are parallel to the master-fault trend. Most of the
secondary normal faults, however, are oblique to the
master-fault trend. In cross sectional view, some synthetic secondary normal faults link directly with the
master fault. Other secondary faults, however, either
die out at depth or detach within the viscous putty
layer.
These comparisons suggest that the deformation
style in the sedimentary cover above an oblique-slip
normal fault depends on several factors. Partitioned
and focused deformation are more likely to occur if the
cover sequence is thin, the structural level within the
cover sequence is relatively deep, the displacement on
the master fault is large, or the cover sequence deforms
900
primarily by cataclastic faulting. Alternatively, distributed deformation is more likely to occur if the cover
sequence is thick, the structural level within the cover
sequence is relatively shallow, the displacement on the
master fault is small, or the cover sequence deforms,
at least in part, by distributed cataclasis or flow (i.e.,
the cover sequence contains shales or salt).
Oblique-Slip Normal Faulting
Minas Subbasin of the Fundy Basin
Numerous rift basins developed in eastern North
America during the Late Triassic–Early Jurassic (Figure
1). One of the largest of these rift basins, the Fundy
basin of New Brunswick and Nova Scotia, Canada, has
three structural elements: the northeast-trending Chignecto and Fundy subbasins and the east-northeast–
trending Minas subbasin (Olsen and Schlische, 1990;
Withjack et al., 1995) (Figure 12a). The trends of
the subbasins reflect the trends of the Paleozoic
contractional fabric. Northwest-southeast extension
during the early Mesozoic (e.g., Schlische and Ackermann, 1995) reactivated northeast-trending Paleozoic
reverse faults along the northwest margins of the Chignecto and Fundy subbasins as normal faults. Simultaneously, east-northeast–trending Paleozoic structures
along the northern margin of the Minas subbasin (i.e.,
the Cobequid-Chedabucto fault system) became faults
with both normal dip-slip and left-lateral strike-slip
components of displacement. Thus, the northern
boundary of the Minas subbasin was an oblique-slip
normal fault zone during Late Triassic–Early Jurassic
rifting. Little subsidence and deposition occurred after
Paleozoic shortening and before Mesozoic extension.
Thus, it is likely that the sedimentary cover above the
reactivated Paleozoic structures was thin to absent.
Field mapping in the Minas subbasin shows that
rift-related faults have two trends (Figure 12b). Eastnortheast–striking faults form throughgoing fault zones
with combined normal dip-slip and left-lateral strikeslip displacement. They are subparallel to the trend of
the Minas subbasin and to the reactivated Paleozoic
fabric. Northeast-striking faults have normal displace-
Deformation & Attributes
Simplified Map Pattern
Simplified Cross Section
Partitioned (sand models)
• Normal and strike-slip faulting
parallel to master fault
Focused oblique-slip normal faulting
(sand & thin clay models)
• Oblique-slip normal fault parallel to
master fault
• Some secondary normal faults
oblique to master fault in clay models
• Moderate folding in clay models
Distributed (sand/putty &
thick clay models)
• Secondary normal faults oblique to
master fault
• Block rotation and reverse faulting in
sand/putty models
• Significant folding in clay models
Figure 11. Simplified map patterns and cross sectional geometries observed experimentally in the cover sequence above an obliqueslip normal fault. The three end-member types of deformation are partitioned, focused oblique-slip normal faulting, and distributed.
ment and are nearly perpendicular to the northwestsoutheast regional extension direction. They form predominantly in the hanging walls of the throughgoing
east-northeast–striking fault zones and locally link with
them. The observed fault pattern along the northern
margin of the Minas subbasin is very similar to the one
that developed in the thin oblique-slip model (experiment III, Figure 12c). Most deformation is localized
on east-northeast–striking, throughgoing, oblique-slip
normal faults; northeast-striking normal faults, however, accommodate some of the deformation. Using
the classification in Figure 11, the style of deformation
along the northern margin of the Minas subbasin is
focused.
Northwest Shelf of Australia
The northern Carnarvon basin is located on the northwest shelf of Australia (Figure 13a, b). Its main tectonic components include the Exmouth Plateau and
several elongate structural lows (e.g., the Exmouth,
Barrow, and Dampier subbasins) and highs (e.g., the
Brigadier and Rankin platforms) (Vincent and Tilbury, 1988; Stagg and Colwell, 1994; Romine et al.,
1997). The northeast-trending Exmouth, Barrow, and
Dampier subbasins developed during an episode of
Early Jurassic–Early Cretaceous rifting. Northeasttrending regional folds (synclines, monoclines) developed during rifting within and along the boundaries
of the subbasins (Figure 13a, b). For example, seismic
data show that the western boundary of the Dampier
subbasin is a northeast-trending monocline cut by numerous secondary normal faults (e.g., Veenstra, 1985;
Newman, 1994; Jablonski, 1997) (Figure 13c, d).
Units in the thick Paleozoic–Triassic cover sequence
consistently drop down from the northwest to the
southeast, suggesting the presence of a southeastdipping, northeast-striking master fault at depth. The
normal faults that were active during rifting have two
distinct trends: northeast, subparallel to the trend of
the Dampier subbasin, and north-northeast. In the
Dampier subbasin, seismic data suggest that the faults
with the north-northeast trend are slightly older than
those with the northeast trend (Newman, 1994).
Newman (1994) proposed that the presence of these
two fault sets that have different trends and ages reflected two distinct episodes of deformation in the
Dampier subbasin: west-northwest–east-southeast extension during the Early–Middle Jurassic followed
Schlische et al.
901
Chignecto subbasin
Figure 12b
45
Minas subbasin
Minas
fault zone
Nova Scotia
44
Atlantic
Ocean
N
75 km
68
66
64
45 25’
64 20’
Synrift
Prerift
N
5 km
64 10’
64 00’
63 50’
Figure 12. (a) Location map for the Fundy rift basin, southeastern Canada, and its three subbasins. (b) Simplified geologic map of
the northern margin of Minas subbasin. The thin black line is coastline, and thin gray lines are synrift sedimentary contacts. Faults
fall into two categories: east-northeast– to east-striking faults with a significant left-lateral strike-slip component of displacement during
rifting, and northeast-striking faults with mostly a normal dip-slip component of displacement during rifting. Arrow gives inferred Early
Jurassic regional extension direction (Schlische and Ackermann, 1995). Redrawn from Olsen and Schlische (1990). (c) Part of thin
oblique-slip model (experiment III) reoriented so that the trend of the throughgoing fault is similar to that in (b).
by northwest-southeast extension during the Late
Jurassic–Early Cretaceous.
We propose an alternate explanation for the observed deformation patterns in the Dampier subbasin:
distributed deformation (see Figure 11) associated
with oblique-slip normal faulting during a single episode of west-northwest–east-southeast regional extension. With this explanation, rifting during the Early
Jurassic–Early Cretaceous reactivated a southeastdipping, northeast-striking, preexisting zone of weakness in the basement beneath the northwestern edge
of the Dampier subbasin. In response to west-northwest–east-southeast regional extension, the zone of
weakness became an oblique-slip fault zone with components of both normal and left-lateral strike-slip displacement. Initially, an extensional forced fold developed in the thick Paleozoic–Triassic cover sequence
above the oblique-slip normal fault, forming the Dampier monocline. Soon after, north-northeast–striking
secondary normal faults developed in the Paleozoic–
Triassic strata above the footwall of the master fault.
These secondary normal faults, oblique to the masterfault trend, formed a series of relay ramps and en echelon horsts and grabens. As the displacement on the
902
Oblique-Slip Normal Faulting
master fault increased, the northeast-striking, obliqueslip normal fault eventually propagated upward, cutting the early formed north-northeast–striking secondary normal faults.
Oil fields in the Rankin trend (e.g., Newman,
1994) are primarily associated with en echelon horst
blocks that provide three-way closure (Figure 14a).
Similar en echelon horst blocks are also well developed
at shallow structural levels in our experimental models
of oblique-slip faulting (Figure 14b).
SUMMARY AND CONCLUSIONS
We have used clay models to define the secondary fault
and fold patterns produced by oblique-slip normal
faulting. Our modeling results, together with the results of previous modeling studies with sand and putty
as the modeling media, suggest the following.
1. Three deformation styles can develop in the sedimentary cover above an oblique-slip normal fault:
partitioned, focused, and distributed. With partitioned deformation, normal faulting accommodates
Gascoyne
abyssal plain
116
200 km
116
N
N
er
di
a
ig
Br
30 km
nd
tre
in
nk
a
R
nd
tre
Dampier
subbasin
Exmouth
plateau
20
Cuvier
abyssal
plain
Australia
Barrow
subbasin
Composite line
20
(a)
(b)
Exmouth subbasin
6
10 km
~1:1 at 4 km/s
Seconds
SE 0
NW
12
NW
SE
5
sec
Recent to Lower
Cretaceous
(e)
Lower Cretaceous
to Middle Jurassic
10 km
Middle to Lower
Jurassic
Lower J and older
strata & basement
5 cm
Figure 13. (a) Simplified geologic map of the northwest shelf of Australia showing the fault pattern on the Exmouth Plateau and
the distribution of subbasins and regional folds. Box shows area detailed in (b). Based on Vincent and Tilbury (1988), Stagg and
Colwell (1994), and Romine et al. (1997). (b) Structure map for Dampier subbasin. Faults active during rifting have two distinct trends:
northeast, subparallel to the trend of the Dampier subbasin, and north-northeast. Faults with the north-northeast trend are slightly
older than those with the northeast trend. The thick line gives the location of the composite seismic line in (c). Based on Vincent and
Tilbury (1988), Stagg and Colwell (1994), and Romine et al. (1997). (c) Interpretation of regional composite seismic profile across the
Dampier subbasin showing an inferred offset of top basement and secondary faults and folds in the Jurassic and Cretaceous sedimentary cover. Box shows area detailed in (d). Based on Jablonski (1997). (d) Part of composite seismic profile restored to top of
Upper Cretaceous. (e) Comparable cross section from oblique-slip model (experiment VI).
Schlische et al.
903
2.
3.
4.
Figure 14. Comparison of structural geometries from
(a) northwestern boundary of Dampier subbasin and (b) model
of oblique-slip normal faulting. Map from Dampier subbasin
modified from Newman (1994). Note that the north arrow
points toward the lower right. Horsts in the Rankin trend and
horsts in the model are en echelon and have three-way closure.
the dip-slip component of displacement on the master fault, whereas strike-slip faulting accommodates
the strike-slip component of displacement. With fo904
Oblique-Slip Normal Faulting
5.
cused deformation, throughgoing, oblique-slip normal faults accommodate the deformation. With distributed deformation, extensional forced folding
and secondary normal faulting, oblique to the
master-fault trend, accommodate the deformation.
Partitioned and focused deformation are more likely
to occur if the cover sequence is thin, the relative
depth within the cover sequence is large, the displacement on the master fault is large, and/or the
cover sequence deforms primarily by cataclastic
faulting. Alternatively, distributed deformation is
more likely to occur if the cover sequence is thick,
the relative depth within the cover sequence is
small, the displacement on the master fault is small,
and/or parts of the cover sequence deform by distributed cataclasis or viscous flow.
Focused deformation is the primary deformation
style during the later stages of the thin clay models
of oblique-slip normal faulting and at the deep levels of the models with the thick clay layer. Focused
deformation in the clay models is characterized by
a series of throughgoing oblique-slip normal faults.
These faults are subparallel to the master fault in
map view and link with the master fault at depth.
Some minor secondary normal faults also develop
and are oblique to the master-fault trend.
Distributed deformation is the primary deformation
style in the cover sequence during the early stages
of our clay models of oblique-slip normal faulting.
During the later stages of the experiments, distributed deformation occurs only at the shallow levels
of the models with the thick clay layer. Distributed
deformation in the clay models is characterized by
an extensional forced fold above the master fault.
Numerous secondary normal faults, both antithetic
and synthetic to the master fault, cut the cover sequence above the footwall of the master fault. In
map view, most of the secondary faults are oblique
to the master-fault trend. In cross sectional view,
some of the secondary normal faults die out at
depth, whereas others link with the master fault at
depth.
Both map and cross section data are needed to distinguish oblique-slip deformation from dip-slip and
strike-slip deformation. In map view, the fault patterns in the oblique-slip clay models resemble those
in the strike-slip models. In both cases, two fault
trends develop in the cover sequence. One fault
trend is oblique to the trend of the master fault, and
one fault trend is subparallel to the trend of the
master fault. In cross sectional view, the fault pat-
terns in the oblique-slip clay models resemble those
in the dip-slip models. In both cases, the cover sequence deforms by a combination of extensional
forced folding and faulting.
6. A single episode of oblique-slip normal faulting can
produce two fault populations that have different
trends and ages. In the oblique-slip clay models, two
populations of secondary faults develop in the cover
sequence. One population consists of early formed
faults whose strike is oblique to the trend of the
underlying master fault. The second population
consists of later formed throughgoing faults whose
strike is subparallel to the trend of the master fault.
The faults of the second population generally cut
the faults of the first population.
7. Long-lived, well-developed relay ramps are more
likely to form in the sedimentary cover above an
oblique-slip normal fault than a dip-slip normal
fault. In the thick clay models of oblique-slip normal faulting, relay ramps are wide and persist
throughout the duration of the experiment. The relay ramps are cut by numerous small-scale normal
faults that are subparallel to the ramp-bounding
faults. Cross faults are uncommon and begin to develop only during the final stages of the experiment.
In identical models of dip-slip normal faulting, relay
ramps are narrow. Cross faults are abundant and
develop throughout the duration of the experiment.
8. The deformation patterns in our experiments
closely resemble those observed in the Minas subbasin of the Fundy rift and the Dampier subbasin
on the northwest shelf of Australia. These similarities suggest that the modeling results can apply to
many oblique-slip normal faults and can provide
guidelines for interpreting field, well, or seismic
data in extensional provinces.
APPENDIX
The strength of most upper crustal rocks increases with depth, obeying a Mohr-Coulomb criterion of failure (e.g., Byerlee, 1978). According to this criterion,
s ⳱ c Ⳮ lrn
(1)
where s and rn are, respectively, the shear and normal stresses on a
potential fault surface, c is the cohesion, and l is the coefficient of
internal friction. This empirical criterion of failure describes the initiation of new faults rather than the frictional reactivation of existing
faults. For most sedimentary rocks, l ranges from about 0.55 to 0.85
(e.g., Handin, 1966; Byerlee, 1978). For intact sedimentary rocks, c
is about 10–20 MPa (Handin, 1966), whereas for highly fractured
sedimentary rocks, c is significantly less (e.g., Byerlee, 1978; Brace
and Kohlstedt, 1980). To insure dynamic similarity between the
models and natural prototypes, two conditions must be satisfied.
First, the modeling materials and the rocks in nature must have similar coefficients of internal friction (e.g., Nalpas and Brun, 1993; Weijermars et al., 1993). This condition is satisfied using wet clay as the
modeling material. Second,
c* ⳱ q*g*l*
(2)
where c*, q*, g*, and l* are model to natural prototype ratios for
cohesion, density, gravity, and length, respectively (e.g., Hubbert,
1937; Weijermars et al., 1993; Vendeville et al., 1995). In our models, the values of q* and g* are about 0.7 and 1, respectively, and l*
is 10ⳮ4–10ⳮ5 (i.e., 1 cm in the models equaled 100–1000 m in nature). Thus, to insure dynamic similarity between the models and
natural prototypes, the cohesion of rock must be about 104–105
greater than that of the modeling material. This condition is satisfied
using wet clay as the modeling material.
REFERENCES CITED
Ackermann, R. V., R. W. Schlische, and M. O. Withjack, 2001, The
geometric and statistical evolution of normal fault systems: an
experimental study of the effects of mechanical layer thickness
on scaling laws: Journal of Structural Geology, v. 23, p. 1803–
1819.
Bates, R. L., and J. A. Jackson, eds., 1987, Glossary of Geology, 3rd
ed.: Alexandria, Virginia, American Geological Institute, 788 p.
Brace, W. F., and D. L. Kohlstedt, 1980, Limits on lithospheric stress
imposed by laboratory experiments: Journal of Geophysical Research, v. 85, p. 6248–6252.
Byerlee, J., 1978, Friction of rocks: Pure and Applied Geophysics,
v. 116, p. 615–626.
Childs, C., J. Watterson, and J. J. Walsh, 1995, Fault overlap zones
within developing normal fault systems: Journal of the Geological Society, v. 152, p. 535–549.
Clifton, A. E., R. W. Schlische, M. O. Withjack, and R. V. Ackermann, 2000, Influence of rift obliquity on fault-population systematics: results of analogue modeling experiments: Journal of
Structural Geology, v. 22, p. 1491–1509.
Cloos, H., 1928, Experimente zur inneren tektonik: Centralblatt für
Mineralogie, part B, p. 609–621.
Gawthorpe, R. L., and J. M. Hurst, 1993, Transfer zones in extensional basins: their structural style and influence on drainage
development and stratigraphy: Journal of the Geological Society, v. 150, p. 1137–1152.
Handin, J., 1966, Strength and ductility, in S. P. Clark Jr., ed., Handbook of physical constants: Geological Society of America
Memoir 97, p. 223–289.
Higgins, R. I., and L. B. Harris, 1997, The effects of cover composition on extensional faulting above re-activated basement
faults: results from analogue modelling: Journal of Structural
Geology, v. 19, p. 89–98.
Horsfield, W. T., 1977, An experimental approach to basementcontrolled faulting: Geologie en Mijnbouw, v. 56, p. 363–
370.
Hubbert, M. K., 1937, Theory of scale models as applied to the study
of geologic structures: Geological Society of America Bulletin,
v. 48, p. 1459–1520.
Jablonski, D., 1997, Recent advances in the sequence stratigraphy of
the Triassic to Lower Cretaceous succession in the northern
Schlische et al.
905
Carnarvon basin, Australia: Australian Petroleum Production
and Exploration Association Journal, v. 37, p. 429–454.
McClay, K. R., and M. J. White, 1996, Analogue modelling of orthogonal and oblique rifting: Marine and Petroleum Geology,
v. 12, p. 137–151.
Nalpas, T., and J.-P. Brun, 1993, Salt flow and diapirism related to
extension at crustal scale: Tectonophysics, v. 228, p. 349–
362.
Naylor, M. A., G. Mandl, and C. H. K. Sijpestein, 1986, Fault geometries in basement induced wrench faulting under different
initial stress states: Journal of Structural Geology, v. 8, p. 737–
752.
Newman, S., 1994, Clues to the structural history of the Rankin
Trend from 3D seismic data, in P. G. Purcell and R. R. Purcell,
eds., The sedimentary basins of Western Australia: Proceedings
of the Petroleum Exploration Society of Australia Symposium,
p. 497–507.
Olsen, P. E., and R. W. Schlische, 1990, Transtensional arm of the
early Mesozoic Fundy rift basin: penecontemporaneous faulting
and sedimentation: Geology, v. 18, p. 695–698.
Peacock, D. C. P., and D. J. Sanderson, 1994, Geometry and development of relay ramps in normal fault systems: AAPG Bulletin,
v. 78, p. 147–165.
Ratcliffe, N. M., and W. C. Burton, 1985, Fault reactivation models
for the origin of the Newark basin and studies related to U.S.
eastern seismicity: U.S. Geological Survey Circular 946, p. 36–
45.
Richard, P. D., 1991, Experiments on faulting in a two-layer cover
sequence overlying a reactivated basement fault with oblique
(normal-wrench or reverse-wrench) slip: Journal of Structural
Geology, v. 13, p. 459–469.
Riedel, W., 1929, Zur Mechanik geologischer Brucherscheinungen;
Ein Betrag zum Problem der “Fiederspalten”: Centralblatt für
Mineralogie, Geologie, und Paleontologie, part B, p. 354–368.
Romine, K. K., J. M. Durrant, D. L. Cathro, and G. Bernardel, 1997,
Petroleum play element prediction for the Cretaceous–Tertiary
basin phase, northern Carnarvon basin: Australian Petroleum
Production and Exploration Association Journal, v. 37, p. 315–
339.
Rutter, E. H., 1986, On the nomenclature of mode of failure transition in rocks: Tectonophysics, v. 122, p. 381–387.
Schlische, R. W., 1993, Anatomy and evolution of the Triassic–
Jurassic continental rift system, eastern North America: Tectonics, v. 12, p. 1026–1042.
Schlische, R. W., and R. V. Ackermann, 1995, Kinematic significance of sediment-filled fissures in the North Mountain Basalt,
Fundy rift basin, Nova Scotia, Canada: Journal of Structural
Geology, v. 17, p. 987–996.
Sims, D., 1993, The rheology of clay: a modeling material for geologic structures: EOS, Transactions of the American Geophysical Union, v. 74, p. 569.
906
Oblique-Slip Normal Faulting
Smith, J. V., and D. W. Durney, 1992, Experimental formation of
brittle structural assemblages in oblique divergence: Tectonophysics, v. 216, p. 235–253.
Stagg, H. M. J., and J. B. Colwell, 1994, The structural foundations
of the northern Carnarvon basin, in P. G. Purcell and R. R.
Purcell, eds., The sedimentary basins of Western Australia: Proceedings of the Petroleum Exploration Society of Australia
Symposium, p. 349–364.
Tchalenko, J. S., 1970, Similarities between shear zones of different
magnitudes: Geological Society of America Bulletin, v. 81,
p. 1625–1640.
Tron, V., and J.-P. Brun, 1991, Experiments on oblique rifting in
brittle-ductile systems: Tectonophysics, v. 188, p. 71–84.
Tsuneishi, Y., 1978, Geological and experimental studies on mechanism of block faulting: Bulletin of the Earthquake Research
Institute, v. 53, p. 173–242.
Veenstra, E., 1985, Rift and drift in the Dampier sub-basin, a seismic
and structural interpretation: Australian Petroleum Exploration
Association Journal, v. 25, p. 177–189.
Vendeville, B., 1987, Champs de failles et tectonique en extension:
modélisation expérimentale: Ph.D. thesis, Université de
Rennes, France, 395 p.
Vendeville, B., 1988, Modèles expérimentaux de fracturation de lat
couverture controlée par des failles normales dans le socle:
Comptes rendus de l’Académie des sciences, v. 307, série II,
p. 1013–1019.
Vendeville, B. C., G. Hongxing, and M. P. A. Jackson, 1995, Scale
models of salt tectonics during basement-involved extension:
Petroleum Geoscience, v. 1, p. 179–183.
Vincent, P., and L. Tilbury, 1988, Gas and oil fields of the Rankin
trend and Northern Barrow-Dampier sub-basin, in P. G. Purcell
and R. R. Purcell, eds., The North West Shelf Australia: Proceedings of the Petroleum Exploration Society of Australia
Symposium, p. 341–369.
Weijermars, R., M. P. A. Jackson, and B. Vendeville, 1993, Rheological and tectonic modeling of salt provinces: Tectonophysics,
v. 217, p. 143–174.
Wilcox, R. E., T. E. Harding, and D. R. Seely, 1973, Basic wrench
tectonics: AAPG Bulletin, v. 57, p. 74–96.
Withjack, M. O., and S. Callaway, 2000, Active normal faulting beneath a salt layer: an experimental study of deformation in the
cover sequence: AAPG Bulletin, v. 84, p. 627–651.
Withjack, M. O., and W. R. Jamison, 1986, Deformation produced
by oblique rifting: Tectonophysics, v. 126, p. 99–124.
Withjack, M. O., J. Olson, and E. Peterson, 1990, Experimental
models of extensional forced folds: AAPG Bulletin, v. 74,
p. 1038–1054.
Withjack, M. O., P. E. Olsen, and R. W. Schlische, 1995, Tectonic
evolution of the Fundy basin, Canada: evidence of extension
and shortening during passive-margin development: Tectonics,
v. 14, p. 390–405.