Frictional properties of low-angle normal fault gouges and

Earth and Planetary Science Letters 408 (2014) 57–65
Contents lists available at ScienceDirect
Earth and Planetary Science Letters
www.elsevier.com/locate/epsl
Frictional properties of low-angle normal fault gouges and
implications for low-angle normal fault slip
Samuel Haines a,b,∗ , Chris Marone a , Demian Saffer a
a
b
Department of Geosciences, Pennsylvania State University, 522 Deike Building, University Park, PA 16802, USA
Chevron ETC, 1500 Louisiana St., Houston, TX 77002, USA
a r t i c l e
i n f o
Article history:
Received 27 June 2014
Received in revised form 19 September
2014
Accepted 20 September 2014
Available online xxxx
Editor: P. Shearer
Keywords:
rock friction
clay gouge
authigenic gouge
low-angle normal faults
Cordillera
a b s t r a c t
The mechanics of slip on low-angle normal faults (LANFs) remain an enduring problem in structural
geology and fault mechanics. In most cases, new faults should form rather than having slip occur on
LANFs, assuming values of fault friction consistent with Byerlee’s Law. We present results of laboratory
measurements on the frictional properties of natural clay-rich gouges from low-angle normal faults
(LANF) in the American Cordillera, from the Whipple Mts. Detachment, the Panamint range-front
detachment, and the Waterman Hills detachment. These clay-rich gouges are dominated by neoformed
clay minerals and are an integral part of fault zones in many LANFs, yet their frictional properties
under in situ conditions remain relatively unknown. We conducted measurements under saturated
and controlled pore pressure conditions at effective normal stresses ranging from 20 to 60 MPa
(corresponding to depths of 0.9–2.9 km), on both powdered and intact wafers of fault rock. For the
Whipple Mountains detachment, friction coefficient (μ) varies depending on clast content, with values
ranging from 0.40 to 0.58 for clast-rich material, and 0.29–0.30 for clay-rich gouge. Samples from the
Panamint range-front detachment were clay-rich, and exhibit friction values of 0.28 to 0.38, significantly
lower than reported from previous studies on fault gouges tested under room humidity (nominally dry)
conditions, including samples from the same exposure. Samples from the Waterman Hills detachment
are slightly stronger, with μ ranging from 0.38 to 0.43. The neoformed gouge materials from all
three localities exhibits velocity-strengthening frictional behavior under almost all of the experimental
conditions we explored, with values of the friction rate parameter (a − b) ranging from −0.001 to
+0.025. Clast-rich samples exhibited frictional healing (strength increases with hold time), whereas
clay-rich samples do not. Our results indicate that where clay-rich neoformed gouges are present along
LANFs, they provide a mechanically viable explanation for slip on faults with dips <20◦ , requiring only
moderate (P f < σ3 ) overpressures and/or correcting for ∼5◦ of footwall tilting. Furthermore, the low
rates of frictional strength recovery and velocity-strengthening frictional behavior we observe provide an
explanation for the lack of observed seismicity on these structures. We suggest that LANFs in the upper
crust (depth <8 km) slip via a combination of a) reaction-weakening of initially high-angle fault zones
by the formation of neoformed clay-rich gouges, and b) regional tectonic accommodation of rotating fault
blocks.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Normal faults with low dips (<30◦ ) have been recognized in
the field for over a century (e.g. Ransom et al., 1910; Longwell,
1945), but the shallow observed dips stand in apparent contradiction with two fundamental predictions from rock mechanics. For typical values of rock internal friction (∼0.6–0.8), nor-
*
Corresponding author at: Department of Geosciences, Pennsylvania State University, 522 Deike Building, University Park, PA 16802, USA.
E-mail addresses: [email protected], [email protected] (S. Haines).
http://dx.doi.org/10.1016/j.epsl.2014.09.034
0012-821X/© 2014 Elsevier B.V. All rights reserved.
mal faults should not form at dips <∼50◦ ; for typical values
of sliding friction (μ = 0.6) those faults should not then slip at
dips much below 30◦ . The crust should fail instead by formation
of new higher-angle faults (e.g., Anderson, 1951; Byerlee, 1978;
Collettini and Sibson, 2001) (Fig. 1).
Several explanations for LANF slip have been offered. One set
of explanations argues that the low dips observed in the field
are caused by post-deformational rotation of the fault plane, either by the passage of a ‘rolling hinge’ (Wernicke and Axen,
1988; Buck, 1988), or by late-stage ‘domino’-style’ rotation of
normal fault blocks (e.g. Wong and Gans, 2008). A separate set
58
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
Fig. 1. A) Map of the southwestern US showing metamorphic core complexes with low-angle normal faults (LANFs). Core complexes with LANFs highlighted in red are those
sampled in this study: PM: Panamint range-front LANF, WM: Whipple Mountains Detachment, WH: Waterman Hills LANF. B) Panamint Detachment outcrop; this is the same
outcrop sampled by Numelin et al. (2007). Footwall is Neoproterozoic schist, hanging wall is weakly consolidated conglomerate. Clay gouge zone is 0.5–2.0 m thick. Inset
is clay-rich wafer used in our experiments. C) WM: Whipple Mountains detachment fault exposed at Bowman’s Wash. Hanging wall is quartz-cemented quartzite breccia
(interpreted megabreccia deposits – Forshee and Yin, 1995). Footwall is Cretaceous granite with characteristic chlorite/epidote ‘microbreccia’ texture and alteration. Clay
gouge zone is 0.5–2.0 m thick. D) Waterman Hills detachment fault. Hanging wall is Miocene rhyolite, footwall is Miocene granodiorite (Fletcher and Bartley, 1994). Clay
gouge zone is 1.0–2.0 m thick. Grey material is smectite-dominated, brown is illite-dominated. Illite-dominated material was sampled. E) Field photo of Whipple Moutain
Detachment at Bowman’s Wash, showing clay-rich gouge horizons we sampled (white arrows). F) Whipple Mountain Detachment at Bowman’s Wash, showing clast-rich
gouge (inset is a wafer used in our experiments). Scale in insets in 1B and 1F are the same, upper scale units are in 10’s of mm. Photo annotations: HW: Hanging wall. FW:
Footwall. G: Gouge.
of explanations argues for the presence of unusual conditions
within the fault core itself, through either: (a) rotation of principal
stresses in the immediate vicinity of the fault core (Chery, 2002;
Lecomte et al., 2012), (b) locally elevated pore-fluid pressures
in the fault zone (Rice, 1992; Axen, 1992), or (c) the presence
of low-friction materials in the fault core (Hayman et al., 2003;
Numelin et al., 2007; Collettini et al., 2009a, 2009b).
Many LANF in the American Cordillera contain well-developed
clay-rich gouges, ranging in thickness from cm’s to >10 m (Haines
and van der Pluijm, 2012, Fig. 1). These gouges are dominated
by authigenic clay minerals (illite, illite–smectite, smectite and
chlorite–smectite phases) formed at temperatures of 60–180 ◦ C.
Depth constraints on the formation of the neoformed clay gouges
we sampled from three well-studied detachments for this work
(the Whipple Mountains, Panamint range-front, and Waterman
Hills detachments) are harder to establish, given uncertainties
in heat flow and temperature gradients within exhuming LANF
footwalls, but range from <2 to >8 km (Haines and van der
Pluijm, 2012). The neoformation of clay-rich fault gouge, a lowtemperature metasomatic process confined to fault zones, thus has
a potential impact on the frictional strength of LANFs in the upper crust, where low-angle slip is mechanically most problematic.
Although reaction-softening has been discussed as a possible explanation for LANF slip (e.g. Manatschal, 1999), the actual frictional
properties of these neoformed clay-rich gouges and their implications for LANF slip remain poorly characterized.
Frictional properties of natural fault zones are commonly extrapolated or estimated from laboratory tests on synthetic gouges
of similar composition (e.g. crushed illite shale as a proxy for illitic
gouges; Brown et al., 2003; Ikari et al., 2009). Work comparing
frictional properties of chlorite separated from natural fault gouge
with crushed chlorite schist has highlighted dramatic differences
in both the frictional strength and stability between phyllosilicate
materials that have identical compositions, but very different parti-
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
59
Fig. 2. Hand specimen and SEM images of gouge materials sampled. A) Hand sample of gouge from the Panamint Detachment. B) Hand sample of clay-rich gouge from the
Whipple Detachment. Light irregular lines on surface are cracks in epoxy coating surface and not vein mineralization. C) Hand sample of grey illitic gouge from the Waterman
Hills Detachment. D) Wafers of clast-rich gouge from the Whipple Detachment. Compare with B. Wafers are slightly undersized for testing and were not used. Upper scale
is in millimeters. E) SEM image of gouge from the Panamint Detachment. Note scaly fabric in the ‘P’ orientation of Logan et al. (1979) and narrow through-going polished
Riedel surface. Sense of shear is top to the left. F) SEM image of gouge from the Waterman Hills Detachment. Note polished and striated Riedel surface. Sense of shear is
shown surface to bottom of image (missing other side of Riedel surface to top).
cle morphologies (Haines et al., 2013). While tests of natural fault
rocks are now common (e.g. Tembe et al., 2009; Collettini et al.,
2009a, 2009b; Carpenter et al., 2011), LANF gouges remain understudied relative to similarly mechanically problematic strike-slip
systems such as the San Andreas and Alpine Fault (e.g. Tembe et
al., 2009; Boulton et al., 2012) and seismogenic thrust-fault systems such as the Nankai Trough and Longman Shan (e.g. Verbane
et al., 2010; Ikari and Saffer, 2011). Fault rocks from the Zuccale
Fault, an exhumed LANF in central Italy, have been extensively
studied (Collettini et al., 2009a, 2009b; Smith and Faulkner, 2010;
Tesei et al., 2012), but the Zuccale LANF fault rocks are dominated by talc and chlorite-dominated phyllonites mineralogically
and structurally distinct from the illite- and smectite-dominated
clayey gouges common to the LANFs of the American Cordillera
(Haines and van der Pluijm, 2012).
Numelin et al. (2007) measured frictional strengths and stability
of gouges from the Panamint Detachment, but worked exclusively
with dry materials that were powdered and sieved. Their experimental conditions leave unresolved questions about the in situ
frictional properties of these fault gouges, given the importance of
fluid saturation, pore pressure and effective stress, and fault zone
fabric in governing the frictional behavior of clay-rich gouge (e.g.,
Ikari et al., 2009, 2011; Collettini et al., 2009a, 2009b). A need thus
exists to measure frictional properties of both intact and powdered
LANF natural gouges under saturated and controlled pore pressure conditions. Here, we report on shearing experiments designed
to measure the frictional properties of natural LANF neoformed
clay-rich fault gouges. Specifically, we (1) measure the frictional
strength of natural gouges, using intact gouge wafers where possible; (2) measure rate/state friction constitutive properties including
the rate of frictional healing and the velocity dependence of sliding friction to address the question of whether LANF seismicity is
expected; and (3) discuss these results in the context of LANF evolution, geometry, and seismicity.
2. Experimental samples and methods
We collected natural gouges from exposures of three LANF detachment faults in SE California: the Whipple Detachment, the
Panamint Detachment, and the Waterman Hills Detachment of the
Central Mojave Metamorphic Core Complex (CMMCC – Fig. 1).
The Panamint Detachment lacks a mylonitic footwall and was
apparently active entirely in the brittle regime (Cichanski, 2000)
during the Miocene to Pleistocene (Andrew and Walker, 2009).
We sampled the detachment in South Park Canyon (N36◦ 00 01.8 ),
W117◦ 12 02.0 , NAD 1927) where it locally dips 18◦ to the west
and juxtaposes likely Pliocene or latest Miocene weakly indurated
conglomerates against a footwall of Proterozoic schists. This exposure is the same exposure as ‘Location A’ of Numelin et al.
(2007) and was also characterized mineralogically by Haines and
van der Pluijm (2012) (their sample “South Park, central gouge”).
At this outcrop, clay gouge material is reddish brown or yellowish,
is relatively free of clasts larger than pebble size, and is only very
crudely foliated in hand specimen (Fig. 2A and inset in Fig. 1B).
Visual inspection of clasts in the gouge indicate that they consist of partially metasomatized schist and granitic material, indicating the neoformed clays in the gouge are apparently derived
from footwall lithologies (Haines and van der Pluijm, 2012). XRD
analysis indicates that both reddish and yellowish materials are
dominated by neoformed illite (specifically the disordered 1Md
polytype) derived from the breakdown of potassium feldspar, along
with catclastically-derived quartz and potassium feldspar, and minor gypsum (Haines and van der Pluijm, 2012). SEM images show
a scaly fabric with thin polished Riedel shear surfaces. The likely
temperature of gouge formation, based on the neoformed clay
mineral assemblage in the gouge, is 60–160◦ C (Haines and van der
Pluijm, 2012).
The Whipple Detachment was active in the mid-Miocene
(Davis, 1988). We sampled the detachment at Bowman’s Wash
60
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
(N34◦ 17 15.3 , W114◦ 17 30.3 ) where it juxtaposes cemented
monolithic quartzose breccia in the hanging wall and pervasively
brecciated and epidote–chlorite metasomatized Cretaceous granite
in the footwall (Forshee and Yin, 1995; Selverstone et al., 2012)
The Bowman’s Wash locality contains both laterally continuous
clay-rich gouge (Fig. 1E) and lenses of clast-rich material with
clasts to pebble and locally boulder size (Figs. 1C, 1F, 1F inset, and
Fig. 2). Gouge clasts are fresh sub-rounded to sub-angular granitic
material, indicating that the gouge was derived from footwall
lithologies, and not from the hanging wall quartzose monomict
breccia. Notably, gouge clasts are fresher than the epidote-altered
granitic material in the immediate footwall (Fig. 1C), indicating
formation from metasomatism of materials located some distance
away from the immediate outcrop sampled, presumably up-dip
on the footwall. XRD analysis of the gouge indicates that both
the clay-rich and clast-rich horizons (shown in Figs. 1E, 1F, and
Figs. 2B, 2D) are dominated by neoformed illite (specifically the
disordered 1Md polytype) derived from the breakdown of potassium feldspar, along with catclastically-derived quartz and potassium feldspar. The likely temperature of gouge formation, based on
the neoformed clay mineral assemblage in the gouge, is 60–160 ◦ C
(Haines and van der Pluijm, 2012). The epidote-rich metasomatism in the footwall (visible in Fig. 1C as a greenish layer in the
immediate footwall) formed at significantly higher temperatures
(380–420 ◦ C, Selverstone et al., 2012), at depths of ∼9 km and
likely earlier in the fault’s history. Both clay-rich and clast-rich materials were sampled for this study. The clay-rich material was too
friable to prepare wafers, so samples were disaggregated gently
with a mortar and pestle, and then sieved to <106 μm. Clastrich material was cut into wafers ∼8–10 mm × 60 mm × 60 mm
(Fig. 1F) for our tests.
The Waterman Hills detachment fault was sampled in the
northern Waterman Hills (Walker et al., 1990; Anderson, 2007)
where it juxtaposes Tertiary rhyolites in the hanging wall and
mylonitic granodiorite in the footwall (Fig. 1D). The detachment
is a late brittle fault that is part of the ductile-to-brittle Central Mojave metamorphic core complex (Dokka, 1989; Walker et
al., 1990). The temporal and kinematic relationship of the brittle
fault to the mylonitic detachment is complex, in that kinematic
indicators associated with the brittle fault record north–south extension as opposed to the northeast–southwest mid-Miocene extension recorded by the mylonites (Walker et al., 1990; Anderson,
2007). At outcrop, the gouge consists of grey illitic layers and
brown smectite-dominated layers (Haines and van der Pluijm,
2012; Figs. 1D and 2C). We sampled grayish illitic material for
this study, as it is laterally continuous throughout the outcrop.
The illite in the Waterman Hills Detachment is also neoformed,
of the disordered 1Md polytype, and derived from the breakdown
of potassium feldspar, similar to the Panamint Detachment and
Whipple Detachment localities, and also likely formed at temperatures of 60–160 ◦ C (Haines and van der Pluijm, 2012). Sample
material was disaggregated gently with a mortar and pestle and
sieved to <106 μm.
Representative material from all 3 exposures was examined
by scanning electron microscopy (SEM) to image fault rock microfabric, and material from the Panamint and Waterman Hills
exposures was characterized by high-resolution X-ray texture goniometry (XTG) to quantify the degree of preferred orientation of
phyllosilicates. Details of the XTG method are found in van der
Pluijm et al. (1994). SEM images show a scaly fabric with welldeveloped polished and striated narrow (<10 μm) Riedel shear
surfaces, and with phyllosilicates weakly aligned in the aligned
in the P orientation (Figs. 2E and 2F). These fabrics are similar to
those described in experimental clay-rich fault rocks (Haines et al.,
2009, 2013), and the very thin nature of the Riedel surfaces is similar to those found in both natural and experimental fault gouges
Fig. 3. Data for a representative shearing experiment run at an effective normal
stress of 20 MPa. Friction (shear stress normalized by normal stress) is plotted
vs. shear strain determined by normalizing shear displacement by layer thickness.
A shear displacement of ∼16 mm corresponds to a shear strain of 10. Larger inset:
Frictional healing (μ) plotted versus hold time for example clast-rich and clayrich samples from the Whipple Detachment. Lines indicate best-fit healing rates
(μ/decade). Smaller inset: Example measurement of frictional healing parameter
μ for a single slide–hold-slide test.
and faulted mudrocks (Haines et al., 2013; Laurich et al., 2014).
XTG analysis indicate that the degree of preferred orientation of
phyllosilicates in the Panamint material is 2.3 and 3.1 multiples
of a random distribution (MRD), and that material in the Waterman Hills exposure has a preferred orientation of 2.5 and 3.4 MRD
units. These low values are very similar to those reported from clay
rich fault gouges elsewhere in the Cordillera and globally (Haines
et al., 2009).
We conducted our experiments on intact wafers of gouge from
the Panamint Detachment (clay-rich gouge) and the Whipple Detachment (clast-rich lenses), and on powdered gouges from all
three localities. Sample conditions did not allow tests of intact
wafers of clay-rich gouge from the Waterman Hills or Whipple
Detachments. Gouge material was either disaggregated gently using a mortar and pestle to preserve natural particle morphology
and sieved to <106 μm (for tests on powdered material), or cut
into wafers ∼8–10 mm × 60 mm × 60 mm using an oil-cooled
tile saw (for intact wafers). Wafers were cut parallel to the in situ
fault-zone shear direction and fault zone fabric (Fig. 1B). The plasticity and saturated nature of the sheared materials from all sites
prevented removal of intact samples post-test for fabric studies to
compare with pre-experiment fabric (shown in Fig. 2).
We sheared samples in a double-direct shear configuration inside a pressure vessel, under saturated and controlled pore pressure conditions at room temperature. Details of the experimental
configuration are given in Samuelson et al. (2009) and Ikari et al.
(2009). Samples were jacketed and saturated with distilled water at a pore fluid pressure of 5 MPa and a confining pressure
of 6 MPa, prior to application of effective normal stresses ranging
from 18 to 60 MPa. Samples were then sheared at 10 μm/s until reaching a steady-state friction value, followed by velocity-step
tests to measure frictional rate dependence (a − b), and slide–holdslide tests to measure the rate of frictional healing (β ) (Fig. 3).
We report friction values as the ratio of shear stress to normal stress (assuming no cohesion), taken at angular shear strains
of 2–3, past the point at which the stress–strain curve ‘rolls over’
and at which (in dry materials sheared in the same apparatus)
Riedel shears and a weak preferred orientation of phyllosilicates
in the P orientation (of Logan et al., 1979; Marone, 1998) is developed (Haines et al., 2013). The change in friction coefficient
with sliding velocity is described by the friction rate parameter
(a − b) = μss / ln V , where μss is the steady-state friction coefficient and V is the sliding velocity (Fig. 3). Positive values of
(a − b) reflect velocity-strengthening behavior (i.e. friction coefficient increases with increasing sliding velocity), whereas negative values of (a − b) reflect velocity-weakening behavior. The
rate of frictional healing (β ) is given by μ/ log t h , based on the
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
61
Fig. 4. A) Coefficient of sliding friction for clay gouges from (A) Panamint Detachment; (B) Waterman Hills Detachment; and (C) Whipple Mts. Detachment. Center key
applies to all panels. Panel (A) also shows data collected at room humidity from Numelin et al. (2007) for comparison, including results for powders comprised of <50%
phyllosilicates (orange shaded area), sample A3 (57% phyllosilicates; gray triangles), and sample B6 (63% phyllosilicates; gray circles). Numelin ‘A3’ was taken from the same
Panamint outcrop as we sampled. Note our wafers and powders are substantially weaker than reported by Numelin et al. (2007) for >50% clay gouges from same outcrop
sheared at room humidity. ‘B6’ is different clay mineralogy than we studied (predominantly smectite) and from a different Panamint outcrop. Whipple Detachment samples
are separated into clay-rich and clast-rich materials. Note Whipple clast-rich wafers and powders are 0.1 μ weaker than Numelin ‘<50% clay’ powders, although samples are
compositionally similar.
Fig. 5. A) Friction rate parameter (a − b) as function of sliding velocity for our complete dataset. No significant variation of (a − b) between outcrops was observed, so
data from all localities are plotted together. B) Rate of frictional re-strengthening
(β) (μ after hold −μ before hold) measured during slide–hold-slide tests plotted
as a function of friction coefficient (μ). Key is the same as for Fig. 4A.
change in measured peak friction (μ) before and after a hold of
length t h (Fig. 3, inset). Positive healing rates indicate frictional restrengthening with hold time. Both velocity-weakening friction and
positive healing rates are pre-requisites for initiation of failure in
repeated stick-slip events (i.e. earthquakes) (e.g., Marone, 1998).
3. Frictional strength and properties of low-angle normal fault
gouges
The clay-rich gouges from all three detachment faults are frictionally weak, with steady-state friction values of μ = 0.28–0.43
(Fig. 3). In contrast, the clast-rich gouges are stronger, with friction coefficients of 0.43–0.56. In general, wafers and powders of
the same material exhibit similar frictional strengths. The friction
coefficient for wafers of clay-rich gouge from the Panamint Detachment was μ = 0.28–0.38, whereas μ for powders of the same
material ranged from 0.29 to 0.36. Wafers of clast-rich gouge from
the Whipple Detachment were slightly weaker (μ = 0.43–0.52)
than powders of the same material (μ = 0.47–0.56). Powdered
clay-rich gouge from the Waterman Hills Detachment was slightly
stronger (μ = 0.38–0.44) than clay-rich gouges from the other two
localities (Fig. 3). We observe no systematic trends in friction coefficient with effective normal stress for either wafers or powders
(Figs. 4A–4C).
Gouge materials from all three detachments exhibit velocitystrengthening behavior, with the friction rate parameter (a − b)
ranging from +0.0002 to +0.0250 (Fig. 5A). Examination of the
data reveals two distinct trends. First, (a − b) generally increases
with upstep sliding velocity for all of the gouges (Fig. 5A). Second, clay-rich materials have slightly higher values of (a − b) at
upstep velocities <100 μm/s than for clast-rich (clay-poor) materials (Fig. 5B). The parameter a − b does not vary significantly with
normal stress, or between wafers and powders.
Healing rates vary strongly with both clay content and frictional
strength for all gouges. Values of β range from −0.039μ/decade
(negative healing) for clay-rich powdered gouge from the Panamint
Detachment, to +0.069μ/decade for clast-rich gouge from the
Whipple Detachment. In general, the clay-rich gouges exhibit systematically lower healing rates than the clast-rich gouges; a clear
trend of decreasing healing rate with lower frictional strength of
the clay gouge is also evident in Fig. 4B. Overall, we do not observe a significant difference in healing between wafers and powders. Notably, all three detachments had at least 1 clay-rich gouge
sample that was weak (μ < 0.4), velocity-strengthening, and characterized by negative healing rates (Fig. 4B), including wafers of
gouge from the Panamint Detachment. Overall, we do not observe
a significant difference in healing between wafers and powders.
4. Discussion
The LANF gouges we tested are weak relative to Byerlee’s Law,
and are comparable in frictional strength to previously studied natural and synthetic clay-rich gouges from other major fault systems
(e.g. Carpenter et al., 2011; Ikari et al., 2011). Our observation
of similar frictional strength between intact wafers and powdered
material is consistent with previous studies of intact vs. powdered
clay gouges (Ikari et al., 2009, 2011; Carpenter et al., 2011, 2012)
and contrasts with large differences in frictional strength between
powders and intact wafers in visibly foliated and mineralogically
layered fault rocks (Collettini et al., 2009a, 2009b). Clay-rich fault
gouges lack strong fabrics at outcrop and fabric intensity studies using X-ray texture goniometry and synchrotron X-rays both
demonstrate that they are characterized by weak preferred orientation of phyllosilicates relative to visibly-foliated fault rocks (Haines
et al., 2009; Buatier et al., 2011).
Notably, our results suggest that in situ LANF gouges are significantly weaker than reported by Numelin et al. (2007) for similar,
or in some cases, the same materials. This is consistent with previous work demonstrating lower frictional strengths for saturated
versus nominally dry clay-rich gouges of the same material (e.g.,
Ikari et al., 2007; Behnsen and Faulkner, 2012). For example, at the
effective normal stress conditions we investigated (17–60 MPa),
Numelin et al., reported μ values ranging from μ = 0.48–0.64 for
samples with <50% phyllosilicates, and for sample A3 (57% phyllosilicates, and from the same outcrop we sample) that drop systematically with increasing normal stress from μ = 0.65 at 5 MPa
to μ = 0.42 at 70 MPa (Fig. 4A). Our results for saturated samples
62
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
from the same Panamint outcrop indicate μ = 0.28–0.38, with no
systematic decrease in frictional strength with increasing normal
stress.
The frictional properties of clay rich gouges, both natural and
synthetic, change markedly as a function of temperature. Previous works show that clay gouges strengthen markedly and
in some cases becoming frictionally unstable at temperatures
>200 ◦ C (Tembe et al., 2009; Den Hartog et al., 2012a, 2012b).
These results were obtained to understand the behavior of either:
(1) subduction-zone faults, where clay-rich material is entrained
in the subduction zone and transported to regions of higher temperatures and pressures; or (2) continental-scale strike-slip faults.
In contrast to those types of fault systems, exposure of LANF
gouges to higher temperatures and pressures after their formation is exceedingly unlikely, given the clear extensional nature of
LANFs. The temperature at the time of neoformed LANF gouge
formation is likely the highest temperature experienced by the
gouge, and is constrained by the clay mineral assemblage to be
60–160 ◦ C. Although our results were obtained at room temperature, they are likely valid for LANF fault-zone conditions, as the
gouges likely never experienced temperatures >200 ◦ C at which
clay-rich gouges with similar illite-dominated mineralogies are reported to strengthen and become frictionally unstable as reported
elsewhere (Den Hartog et al., 2012a, 2012b).
To explore the implications of our lab data for the paradox of
slip on shallowly-dipping (and thus severely misoriented) LANF, we
define a relationship between the dip of faults on which slip would
be permitted, and the effective horizontal stress (S, normalized to
the overburden) based on simple re-arrangement of the Coulomb–
Mohr failure criterion (after Abers, 2001):
S=
σ3
1 − μ cot θ
= (1 − λ)
ρ gz
1 + μ cot θ
(1)
where σ3 is the least principal effective stress, ρ is the mean
bulk density of the overburden, g is gravitational acceleration, z is
depth, λ is the ratio of pore fluid pressure to lithostatic stress, and
θ is the dip of the detachment (Fig. 6C). The lower limit for effective stress labeled ‘hydrofracture’ in Fig. 5C is slightly below 0,
reflecting an assumed rock tensile strength of 10 MPa and shown
for depths of 2, 5 and 8 km, assuming a mean bulk density of
2300 kg m−3 for the overburden.
Slip is permissible at dips as low as 16◦ (μ = 0.28) to 22◦
(μ = 0.43) (Fig. 6C) and at dips of >20◦ with overpressures similar to those found in sedimentary basins elsewhere (Figs. 6 and 7,
Gaarenstroom et al., 1993; Converse et al., 2000). Alternatively,
overpressure could be locally developed in the clay-rich and lowpermeability gouge layer itself and not require a regionally overpressured basin. Although slip is not permitted at hydrostatic pore
fluid pressure for μ = 0.28, slip is possible at the range of dips we
observe in the field for the Panamint Detachment with the introduction of only modest overpressure ( P f equivalent to a gradient
of 11.3 MPa/km, vs. a hydrostatic gradient of 10.1 MPa).
For the friction coefficients we measured, slip would be permitted on LANF with dips as low as ∼16◦ (μ = 0.28) to 22◦
(μ = 0.43) (Fig. 4) with higher overpressures (nearer lithostatic).
For μ = 0.43, moderate to high overpressures (but still P f < σ3 )
are required for slip to be permitted (Figs. 6–7). Estimates of
fluid pressures from field evidence in the vicinity of low-angle
normal faults vary with crustal depth. For LANFs active in the
upper few km of the crust, reliable estimates of pore pressure
are sparse. By contrast, LANF fault rocks exhumed from the midcrustal brittle–plastic transition at temperatures of 250 to 420 ◦ C,
several LANFs (the Woodlark Basin detachment, Zuccalle Fault, and
the Whipple Detachment) exhibit compelling field evidence for
overpressure, such as fault-parallel tensile veins (Roller et al., 2001;
Collettini et al., 2006; Selverstone et al., 2012)
Fig. 6. Mechanical analysis of measured friction results. A) Measurements of the
modern dip angle of the Panamint range-front LANF (after Andrew, 2002). B) Mohr
circle diagram for a stress state at failure consistent with our laboratory findings
and a mean fault dip of 21◦ . Circles are for our measured minimum and maximum
friction values. C) Minimum horizontal effective stress (σ3 − P f ) at failure normalized to the overburden (σ1 ; given by ρ gz), as a function of fault dip for frictional
failure, assuming hydrostatic conditions (λ = 0.4) and σ3 = 0.7 × σ1 (after Abers,
2001). Coefficients of friction from our experimental results are shown in bold. Dots
are Mohr circles shown in panel B. Shaded area below each curve of μ defines the
range of mechanically viable fault dips and stress states that would allow slip at that
value of μ. Pink area is mean fault dip shown in panel (A), showing ±1σ . Brown
areas define fields of hydrofracture for surrounding rock with tensile strength of
10 MPa at depths of 2, 5 and 8 km.
For comparison, the Panamint Detachment has dips ranging
from 6◦ to 44◦ (mean = 21◦ , 1σ variation = 8.8◦ ) (Andrew,
2002, Fig. 4A), and sub-horizontal horizontal lacustrine strata in
the hanging wall indicate that it slipped at or near its current
dips. Previous studies of LANF dips indicate that the Panamint Detachment mean dip of 21◦ is representative of many LANFs (e.g.,
Collettini, 2011). The amount of footwall tilting in the vicinity of
the Panammint Detachment is likely minor. We note that even for
well-exposed and undisturbed detachments such as the Panamint
Detachment, some footwall tilting (∼5◦ ) may have occurred that
is not resolvable. With 5◦ of footwall tilting, slip is possible at any
fluid pressure for μ = 0.28 and at moderate to high overpressures
for μ = 0.43 (Fig. 7).
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
Fig. 7. Minimum horizontal effective stress ( S ) at failure normalized to the overburden stress as a function of fault dip as in Fig. 5C, showing conditions required for
LANF slip at current Panamint Detachment dip. Stresses are calculated using Eq. (1)
and friction data we report. Mean fault dip for Panamint Detachment ± 1 standard
deviation shown in vertical transparent grey box. Effective stress ratios at failure
are calculated holding μ and σ3 constant (for values of μ we measure, μ = 0.28
and μ = 0.43, and σ3 = 0.7 × σ1 ) and varying λ. Note, by contrast, in Fig. 5C we
hold λ constant (λ = 0.44) and calculate states of stress for different values of μ.
A) μ = 0.28 and varying λ, dipping at present day dips. Hydrostatic fracture window is same as 5 km window in Fig. 5C. B) μ = 0.43, varying λ, and dipping at
present day dips. Note that with moderate or high (but sub-σ3 ) overpressures, slip
is possible at the observed dips.
The dip of the Waterman Hills detachment fault (present-day
10◦ ) is more difficult to establish at the time of slip, as field relations are significantly complicated by later high-angle normal
faulting and strike-slip faulting associated with the onset of the
San Andreas strike-slip system. An estimate based on apatite and
zircon fission-track dating and assumptions of a geothermal gradient elsewhere in the greater Central Mojave metamorphic core
complex put the dip at the time of slip to be 20–27◦ (Dokka,
1995), but is strongly sensitive to the choice of early Miocene
geothermal gradient. For a postulated 20–27◦ dip at the time of
slip, slip is similarly possible at μ = 0.38 to μ = 0.44 with moderate to high overpressures.
The dip of the Whipple detachment is also difficult to establish at the time of slip. At our sampling locality, the detachment surface dips 15◦ to the SW, and is clearly back-rotated at
outcrop, as the detachment has a clear overall top to the NE
sense of displacement (Davis, 1988). The detachment has undergone significant late doming due to isostatic rebound owing
to a probably ‘rolling-hinge’ evolution (Dorsey and Becker, 1995;
Axen, 2004) but was likely never above 45◦ . In the least-disrupted
northeastern portion of the detachment (10 km north of our sampling locality), the detachment dips at 12–28◦ (Yin and Dunn,
1992). For the clay-rich material at the Whipple detachment, slip
63
is possible with our measured friction values at the higher end of
the observed dip range for most values of overpressure, but slip is
impossible at the lowest dip with all but near-lithostatic values of
overpressure (Figs. 6–7). Two explanations have been suggested to
explain slip on the Whipple Detachment: high fluid pressures or
low-friction gouge material (Axen, 2004). Our data lend support to
the low-friction gouge material interpretation or a combination of
low friction and moderately elevated pore pressures. As paleo-dip
constraints for the Waterman Hill and Whipple Detachments are
significantly poorer than for the Panamint Detachment, we do not
treat them quantitatively in Fig. 7.
Clay-rich gouges are found at multiple exposures of many
Cordilleran LANFs (Haines and van der Pluijm, 2012). Our results
indicate that the widespread occurrence of these mechanically
weak fault zone materials, together with moderately-elevated pore
fluid pressures and/or correcting for only minor footwall tilting,
are a viable explanation for LANF slip at low dips. In particular,
field relations at the Whipple Detachment require slip at low dips
during the final period of activity on the detachment (Dorsey and
Becker, 1995; Axen, 2004).
The low frictional strength of LANF clay-rich gouge correlates
with both velocity-strengthening behavior and near-zero healing
rates, similar to both synthetic clay-rich gouges (Bos and Spiers,
2000) and those recovered from the SAFOD borehole (Carpenter
et al., 2011, 2012). The negative healing rates we observe indicate
that grain contacts in the clay-rich LANF gouges we studied do not
strengthen with hold time, and together with the rate strengthening behavior we observe (a − b > 0), provide direct experimental
evidence suggesting that LANFs should slip aseismically.
One of the most compelling arguments against low-angle normal fault slip has been the absence of an unequivocal focal mechanism for a low-angle slip event (Wernicke, 1995), although isolated
examples have been reported (Abers, 2001; Chairaluce et al., 2007).
The most compelling modern example is the Alto Tibernia Fault
in the Italian Appenines, which has both documented microseismicity (Chairaluce et al., 2007) and apparent aseismic creep apparent from GPS measurements (Hreinsdóttir and Bennett, 2009).
Our data suggest that an absence of LANF focal mechanisms for
large earthquakes should not be considered evidence of absence
of LANF slip. We note that the velocity-strengthening behavior of
the clay gouges we sampled indicates that rupture nucleation in
clay-rich gouges is likely impossible, but does not preclude the
possibility of permitting propagation of a rupture initiated elsewhere on the detachment surface. Our data thus indicate that if
LANF faults are capable of seismicity, slip events must nucleate at
depths greater than those at which clay-rich gouges form, and suggest that the depth of rooting of a low-angle normal fault may
control its propensity for seismicity and shallow coseismic slip.
Another key issue in the debate surrounding low-angle normal faults has been the mechanical problem of initiating a normal
fault at low dips (e.g. Axen, 2004). While our data do not shed
light directly on the problem of LANF initiation, neoformation of
frictionally-weak clay-rich gouges offers a potential solution to the
problem if the fault initiated at a higher dip. Neoformation of
clay-rich fault gouge can be thought of as a low-temperature metasomatic process confined to fault zones. Our data indicate that clay
gouge formation, a process that can be dated directly to occur during periods of fault activity (Haines and van der Pluijm, 2008, 2010,
Duvall et al., 2011), is a process that produces materials in LANF
fault cores with low frictional strength, velocity-strengthening behavior and little to no healing. These materials, once formed, can
act to weaken a pre-existing fault surface, and thus permit it to
slip at dips lower than it otherwise could.
Our data are consistent with field observations indicating
that some low-angle normal faults initially formed at high angles (∼60◦ ) and later rotated to their current low dips (e.g.
64
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
Wong and Gans, 2008). Faults that are metasomatically weakened
by the growth of neoformed clay could thus continue to slip to the
low dips observed, as long as larger-scale crustal extensional processes continue to create accommodation space for the ends of the
‘domino blocks’ to rotate. For faults that initiated at anomalously
low dips (at or <45◦ ), the formation of weak fault zone materials by clay metasomatism will still permit slip at dips significantly
lower than would be permitted by typically assumed frictional values of μ = ∼0.6.
Cordilleran LANF faults are observed to ‘cluster’ in space and
time, suggesting larger-scale crustal processes, such as gravitational collapse of thick lithosphere triggered by thermal weakening of subducted oceanic lithosphere and passage of the Mendocino fracture zone, play a role in their formation (Axen et al.,
1993). Examples include the Miocene–Pliocene ‘Turtleback’ LANFs
in the Death Valley area and the Panamint Range in CA (Badwater,
Mormon Point, and Copper Canyon Turtlebacks, Amargosa Detachment, and Panamint Detachment), and the Middle Miocene Colorado River Extensional Corridor LANFs, (Sacramento Detachment,
Chemehuevi Detachment, Whipple Detachment, Buckskin–Rawhide
Detachment, Harcuvar and Harquahala Detachments). Older examples include the Oligo-Miocene LANFs in eastern Nevada (RubyHumboldt Mts LANF, North and South Snake Range Detachments,
Schell Creek Detachment, Albion–Grouse Creek and Raft River detachments) and the Eocene-age belt of metamorphic core complexes of western Canada and the NW United States (Valhalla MCC,
Kettle Dome, Shushwap MCC, and Bitterroot Detachment). We suggest that LANFs may be the result of a combination of widespread
reaction-softening of the fault surface by formation of clay-rich
gouges during fault evolution together with larger-scale crustal
processes that create new accommodation space for fault block
rotation. LANF formation and slip may thus be the result of a confluence of distinct processes, one metasomatic, the other crustal in
scale.
Acknowledgements
This research was funded by NSF Award #0911589 to DS,
CM, and E. Kirby; and awards EAR054570, EAR0746192, and
OCE-0648331 to CM, and a grant from the GDL Foundation to
S. Haines. We are grateful to Gary Axen and Cristiano Collettini for
thorough and constructive reviews. We thank Steve Swavely for
technical assistance in the laboratory and B. Kaproth, K. Bhadra,
B. Carpenter, M. Ikari, and A. Rathbun for field and laboratory support.
Appendix A. Supplementary data
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.epsl.2014.09.034.
These data include the Google map of the most important areas
described in this article.
References
Abers, G., 2001. Evidence for seismogenic normal faults at shallow dips in continental rifts. In: Wilson, R., Whitmarsh, R., Taylor, B., Froitzheim, N. (Eds.),
Non-Volcanic Rifting of Continental Margins: A Comparison of Evidence from
Land and Sea. In: Geol. Soc. (Lond.) Spec. Publ., vol. 187, pp. 305–318.
Anderson, E.M., 1951. The Dynamics of Faulting. Olivier and Boyd, Edinburgh.
206 pp.
Anderson, R., 2007. A blister hypothesis for the Central Mojave metamorphic core
complex near Barstow, California. GSA Specialty Meetings Abstracts With Programs 39, 227.
Andrew, J., 2002. The mesozoic and tertiary tectonics of the Panamint Range and
Quail Mountains, California. PhD thesis. University of Kansas. 400 pp.
Andrew, J., Walker, J.D., 2009. Reconstructing late Cenozoic deformation in central
Panamint Valley, California: evolution of slip partitioning in the Walker Lane.
Geosphere 5, 172–198.
Axen, G., 1992. Pore pressure, stress increase and fault weakening in low-angle normal faulting. J. Geophys. Res. B, Solid Earth Planets 97, 8979–8991.
Axen, G., 2004. Mechanics of low-angle normal faults. In: Karner, G., Taylor, B.,
Driscoll, N., Kohlstedt, D. (Eds.), Rheology and Deformation of the Lithosphere
at Continental Margins. Columbia Univ. Press, New York, pp. 46–91.
Axen, G., Taylor, W., Bartley, J., 1993. Space–time patterns and tectonic controls of
Tertiary extension and magmatism in the Great Basin of the western United
State. Geol. Soc. Am. Bull. 105, 56–76.
Behnsen, J., Faulkner, D., 2012. The effect of mineralogy and effective normal stress
on frictional strength of phyllosilicates. J. Struct. Geol. 42, 49–61.
Bos, B., Spiers, C., 2000. Effect of phyllosilicates on fluid-assisted healing of gougebearing faults. Earth Planet. Sci. Lett. 184, 199–210.
Boulton, C., Carpenter, B.M., Toy, V., Marone, C., 2012. Physical properties of surface
outcrop cataclastic fault rocks, Alpine Fault, New Zealand. Geochem. Geophys.
Geosyst. 13, Q01018. http://dx.doi.org/10.1029/2011GC003872.
Brown, K., Kopf, A., Underwood, M., Weinberger, J., 2003. Compositional and fluid
pressure controls on the state of stress on the Nankai subduction thrust: a weak
plate boundary. Earth Planet. Sci. Lett. 214, 589–603.
Buatier, M.D., Chauvet, A., Kanitpanyacharoen, W., Wenk, H.R., Ritz, J.F., Jolivet, M.,
2011. Origin and behavior of clay minerals in the Bogd fault gouge, Mongolia. J.
Struct. Geol. 34, 77–90.
Buck, W., 1988. Flexural rotation of normal faults. Tectonics 7, 959–974.
Byerlee, J., 1978. Friction of rocks. Pure Appl. Geophys. 116, 615–626.
Carpenter, B., Marone, C., Saffer, D., 2011. Frictional strength of the San Andreas
fault from laboratory measurements of SAFOD drill samples. Nat. Geosci. 10.
http://dx.doi.org/10.1038/ngeo1089.
Carpenter, B.M., Saffer, D.M., Marone, C., 2012. Frictional properties and sliding stability of the San Andreas Fault from deep drill core. Geology 40, 759–762. http://
dx.doi.org/10.1130/G33007.1.
Chairaluce, L., Chiarabba, C., Collettini, C., Piccinini, D., Cocco, M., 2007. Architecture
and mechanics of an active low-angle normal fault: Alto Tibernia Fault, northern Apennines, Italy. J. Geophys. Res. B, Solid Earth Planets 112, B10310. http://
dx.doi.org/10.1029/2007JB005015.
Chery, J., 2002. Core complex mechanics: from the Gulf of Corinth to the Snake
Range. Geology 29, 439–442.
Cichanski, M., 2000. Low-angle range-flank faults in the Panamint, Inyo and Slate
ranges, California: implications for recent tectonics of the Death Valley region.
Geol. Soc. Am. Bull. 112, 871–883.
Collettini, C., 2011. The mechanical paradox of low-angle normal faults: current understanding and open questions. Tectonophysics 510, 253–268.
Collettini, C., Sibson, R., 2001. Normal faults, normal friction? Geology 29, 927–930.
Collettini, C., De Paola, N., Goulty, N., 2006. Swithces in the minimum compressive
stress direction induced by overpressure beneath a low-permeability fault zone.
Terra Nova 18, 224–231.
Collettini, C., Niemeijer, A., Viti, C., Marone, C., 2009a. Fault zone fabric and fault
weakness. Nature 462, 907–910. http://dx.doi.org/10.1038/nature08585.
Collettini, C., Viti, C., Smith, S., Holdsworth, R., 2009b. The development of interconnected talc networks and weakening of continental low-angle normal faults.
Geology 37, 567–570.
Converse, D., Nicholson, P., Pottorf, R., Miller, T., 2000. Controls on overpressure
in rapidly subsiding basins and implications for failure of top seal. In: Mello,
M., Katz, B. (Eds.), Petroleum Systems of South Atlantic Margins, AAPG Memoir,
vol. 73, pp. 133–150.
Davis, G., 1988. Rapid upward transport of mid-crustal mylonitic gneisses in the
footwall of a Miocene detachment fault, Whipple Mountains, southeastern California. Geol. Rundsch. 77, 191–209.
Den Hartog, S., Neimeijer, A., Spiers, C., 2012a. New constraints on megathrust stability under subduction zone P –T conditions. Earth Planet. Sci. Lett. 353–354,
240–252.
Den Hartog, S., Peach, C., Matthjis de Winter, D., Spiers, C., Shimamoto, T., 2012b.
Frictional properties of megathrust fault gouges at low sliding velocities: new
data on effects of normal stress and temperature. J. Struct. Geol. 38, 156–171.
Dokka, R., 1989. The Mojave extensional belt of southern California. Tectonics 8,
363–390.
Dokka, R., 1995. Original dip and subsequent modification of a Cordilleral detachment fault, Mojave extensional belt, California. Geology 21, 711–714.
Dorsey, R., Becker, U., 1995. Evolution of a large Miocene growth structure in the
upper plate of the Whipple detachment fault, northeastern Whipple Mountains,
California. Basin Res. 7, 151–163.
Duvall, A., Clark, M., van der Pluijm, B., Li, Chuanyou, 2011. Direct dating of Eocene
reverse faulting in northeastern Tibet using Ar-dating of fault clays and lowtemperature thermochronometry. Earth Planet. Sci. Lett. 304, 520–526.
Fletcher, J., Bartley, J., 1994. Constrictional strain in a non-coaxial shear zone: implications for fold and rock fabric development, central Mojave mertamorphic core
complex, California. J. Struct. Geol. 16, 555–570.
Forshee, E., Yin, A., 1995. Evolution of monolithic breccia deposits in supradetachment basins, Whipple Mountains, California. Basin Res. 7, 181–197.
Gaarenstroom, L., Tromp, A., DeJong, M., Brandenburg, A., 1993. Overpressure in the
Central North Sea: implications for trap integrity and drilling safety. In: Parker,
J.P. (Ed.), Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. Geological Society, London, pp. 1305–1313.
S. Haines et al. / Earth and Planetary Science Letters 408 (2014) 57–65
Haines, S., van der Pluijm, B., 2008. Clay quantification and Ar–Ar dating of synthetic
and natural gouge: application to the Miocene Sierra Mazatán detachment fault,
Sonora, Mexico. J. Struct. Geol. 30, 525–538.
Haines, S., van der Pluijm, B., 2010. Dating the detachment fault system of the Ruby
Mountains, Nevada: significance for the kinematics of low-angle normal faults.
Tectonics 29 (TC4028). http://dx.doi.org/10.1029/2009TC002552.
Haines, S., van der Pluijm, B., 2012. Patterns of mineral transformations in clay
gouge, with examples from low-angle normal fault rocks in the western USA.
J. Struct. Geol. 43, 2–32.
Haines, S., Ikari, M., van der Pluijm, B., Marone, C., Saffer, D., 2009. Clay fabric intensity in natural and artificial fault gouges: implications for brittle fault zone
processes and sedimentary basin clay fabric evolution. J. Geophys. Res. B, Solid
Earth Planets 114. http://dx.doi.org/10.1029/2008JB005866.
Haines, S., Kaproth, B., Marone, C., Saffer, D., van der Pluijm, B., 2013. Shear zones
in clay-rich fault gouge: a laboratory study of fabric and evolution. J. Struct.
Geol. 51, 206–225.
Hayman, N., Knott, J., Cowan, D.S., Nemser, E., Sarna-Wojcicki, A., 2003. Quaternary lowangle slip on detachment faults in Death Valley, California. Geology 31,
343–346.
Hreinsdóttir, S., Bennett, R., 2009. Active aseismic creep on the Alto Tibernia lowangle normal fault, Italy. Geology 37, 683–686.
Ikari, M., Saffer, D., 2011. Comparison of frictional strength and velocity dependence
between fault zones in the Nankai accretionary complex. Geochem. Geophys.
Geosyst. 12. http://dx.doi.org/10.1029/2010GC003442.
Ikari, M., Saffer, D., Marone, C., 2007. Effect of hydration state on the frictional
properties of montmorillonite-based fault gouge. J. Geophys. Res. B, Solid Earth
Planets 112. http://dx.doi.org/10.1029/2006JB004748.
Ikari, M., Saffer, D., Marone, C., 2009. Frictional and hydrologic properties of a major splay fault system, Nankai subduction zone. Geophys. Res. Lett. 36, L20313.
http://dx.doi.org/10.1029/2009GL040009.
Ikari, M., Niemeijer, A., Marone, C., 2011. The role of fault zone fabric and lithification state on frictional strength, constitutive behavior, and deformation
microstructure. J. Geophys. Res. B, Solid Earth Planets 116. http://dx.doi.org/
10.1029/2011JB008264.
Laurich, B., Urai, J., Desbois, G., Vollmer, C., Nussbaum, C., 2014. Microstructural
evolution of an incipient fault zone in Opalinus Clay: insights from an optical and electron microscopic study of ion-beam polished samples from the
Main Fault in the Mt-Terri Underground Research Laboratory. J. Struct. Geol. 67,
107–128.
Lecomte, E., Le Pourhiet, L., Lacomb, O., 2012. Mechanical basis for slip along lowangle normal faults. Geophys. Res. Lett. 39, L03307. http://dx.doi.org/10.1029/
2011GL050756.
Logan, J., Freidman, M., Higgs, M., Dengo, C., Shimamoto, T., 1979. Experimental
studies of simulated fault gouge and their application to studies of natural fault
zones. In: Proc. Conf. VIII, Analysis of Actual Fault Zones in Bedrock. U.S. Geological Survey, Menlo Park, CA, pp. 305–343.
Longwell, C., 1945. Low-angle normal faults in the Basin and Range province. Eos 26,
1087–1118.
Manatschal, G., 1999. Fluid- and reaction-assisted low-angle normal faulting: evidence from rift-related brittle fault rocks in the Alps (Err Nappe, eastern
Switzerland). J. Struct. Geol. 21, 777–793.
65
Marone, C., 1998. Laboratory-derived friction laws and their application to seismic
faulting. Annu. Rev. Earth Planet. Sci. 26, 643–696.
Numelin, T., Marone, C., Kirby, E., 2007. Frictional properties of natural gouge from
a low-angle normal fault, Panamint Vallet, California. Tectonics 26, TC2004.
http://dx.doi.org/10.1029/2005TC001916.
Ransom, F., Emmons, W., Garrey, G., 1910. Geology and ore deposits of the Bullfrog
District, Nevada. U.S. Geol. Surv. Bull. 407, 1–130.
Rice, J., 1992. Fault stress states, pore pressure distributions, and the weakness of
the San Andreas Fault. In: Evans, B., Wong, T.-F. (Eds.), Fault Mechanics and
Transport Properties of Rocks; a Festschrift in Honor of W.F. Brace. Academic
Press, San Diego, CA, pp. 475–503.
Roller, S., Behrmann, J., Kpof, A., 2001. Deformational fabrics of faulted rocks, and
some syntectonic stress estimates from the active Woodlark Basin detachment
zone. In: Geol. Soc. (Lond.) Spec. Publ., vol. 187, pp. 319–334.
Samuelson, J., Elsworth, D., Marone, C., 2009. Shear-induced dilatancy of fluidsaturated faults: experiment and theory. J. Geophys. Res. B, Solid Earth Planets 114, B12404. http://dx.doi.org/10.1029/2008JB006273.
Selverstone, J., Axen, G., Luther, A., 2012. Fault localization controlled by fluid infiltration into mylonites: formation and strength of low-angle normal faults in
the midcrustal brittle-plastic transition. J. Geophys. Res. 117, B06210. http://
dx.doi.org/10.1029/2012JB009171.
Smith, S., Faulkner, D., 2010. Laboratory measurements of the frictional properties
of the Zuccale low-angle normal fault, Elba Island, Italy. J. Geophys. Res. B, Solid
Earth Planets 115, B02407. http://dx.doi.org/10.1029/2008JB006274.
Tembe, S., Lockner, D., Wong, Teng-Fong, 2009. Constraints on the stress state of the
San Andreas Fault with analysis based on core and cuttings from San Andreas
Fault Observatory at Depth (SAFOD) drilling phases 1 and 2. J. Geophys. Res. 114,
B11401. http://dx.doi.org/10.1029/2008JB005883.
Tesei, T., Collettini, C., Carpenter, B., Viti, C., Marone, C., 2012. Frictional strength
and healing behavior of phyllosilicate-rich faults. J. Geophys. Res. B, Solid Earth
Planets 117, B09402. http://dx.doi.org/10.1029/2012JB09402.
van der Pluijm, B., Ho, N., Peacor, D., 1994. High-resolution X-ray texture goniometry. J. Struct. Geol. 16, 1029–1032.
Verbane, B.A., He, C., Spiers, C.J., 2010. Frictional properties of sedimentary rocks
and natural fault gouge from the Longmen Shan Fault Zone, Sichuan, China. Bull.
Seismol. Soc. Am. 100, 2767–2790.
Walker, J., Bartley, B., Glazner, A., 1990. Large-magnitude Miocene extension in the
central Mojave desert: implications for Paleozoic to Tertiary paleogeography and
tectonics. J. Geophys. Res. B, Solid Earth Planets 95, 557–569.
Wernicke, B., 1995. Low-angle normal faults and seismicity: a review. J. Geophys.
Res. B, Solid Earth Planets 100, 20159–20174.
Wernicke, B., Axen, G., 1988. On the role of isostasy in the evolution of low-angle
normal fault systems. Geology 16, 848–851.
Wong, M., Gans, P., 2008. Geologic, structural, and thermochronologic constraints
on the tectonic evolution of the Sierra Mazatán core complex, Sonora, Mexico:
new insights into metamorphic core complex formation. Tectonics 27, TC4013.
http://dx.doi.org/10.1029/2007TC002173.
Yin, A., Dunn, J., 1992. Structural and stratigraphic development of the Whipple–
Chemehuevi detachment system, southeastern California: implications for the
geometrical evolution of domal and basinal low-angle normal faults. Geol. Soc.
Am. Bull. 104, 659–674.