Downloaded from geology.gsapubs.

org on July 14, 2010

Geology
Mlange rheology and seismic style
ke Fagereng and Richard H. Sibson
Geology 2010;38;751-754
doi: 10.1130/G30868.1

Email alerting services

click www.gsapubs.org/cgi/alerts to receive free e-mail alerts when new

articles cite this article

Subscribe

click www.gsapubs.org/subscriptions/ to subscribe to Geology

Permission request

click http://www.geosociety.org/pubs/copyrt.htm#gsa to contact GSA

Copyright not claimed on content prepared wholly by U.S. government employees within scope of
their employment. Individual scientists are hereby granted permission, without fees or further
requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in subsequent
works and to make unlimited copies of items in GSA's journals for noncommercial use in classrooms
to further education and science. This file may not be posted to any Web site, but authors may post
the abstracts only of their articles on their own or their organization's Web site providing the posting
includes a reference to the article's full citation. GSA provides this and other forums for the
presentation of diverse opinions and positions by scientists worldwide, regardless of their race,
citizenship, gender, religion, or political viewpoint. Opinions presented in this publication do not reflect
official positions of the Society.

Notes

2010 Geological Society of America

Downloaded from geology.gsapubs.org on July 14, 2010

Mlange rheology and seismic style

ke Fagereng1* and Richard H. Sibson1
1

Department of Geology, University of Otago, PO Box 56, Dunedin 9054, New Zealand
elastic strain is stored in rigid wall rock out to a
distance ~100 km from the fault (proportional
to downdip width; e.g., Savage and Burford,
1973), and failure occurs episodically when
this strain reaches a critical value. This model
represents an end member where, if motion is
accommodated by seismic slip, accumulated
shear stress and elastic strain are released partly
(or wholly) from rocks surrounding the fault.
Some models consider subduction faults as
thick homogeneous layers of viscous material
(Fig. 1, model 2; e.g., Shreve and Cloos, 1986),
representing a second end member where most
of the displacement is accommodated by continuous, aseismic deformation to produce permanent finite strains within the shear zone.
The wall rocks undergo constant levels of shear
stress and elastic shear strain.
More realistic fault-rock assemblages, however, contain a mixture of materials with diverse
physical properties (Fig. 1, model 3), causing
spatial variations in shear strength, viscosity,
shear strain rate (), and faulting style. Displacement is accommodated by both finite and elastic strain within the shear zone, and the elastic

ABSTRACT
Shear displacements in crustal fault zones are accommodated by a range of seismic styles,
including standard earthquakes, non-volcanic tremor, and continuous and transitory aseismic slip. Subduction channel shear zones, containing highly sheared, fluid-saturated trenchfill sediments intermingled with fragments of oceanic crust, are commonly inferred to occur
along active subduction megathrusts. If this interpretation is correct, these plate boundary
faults are not discrete planes, but may resemble the mlange shear zones commonly found in
exhumed subduction-related rock assemblages. Mlange deformation appears to depend critically on the ratio of competent to incompetent material, with shear surfaces localized along
lithological contacts or within competent domains, while matrix flow accommodates shearing
by distributed strain. If the style of strain and/or displacement accommodation in a mlange
reflects the partitioning between aseismic and seismic slip, the proportion of competent material seems likely to be a significant factor affecting seismic style within subduction channel
shear zones, and along comparable mixed-lithology fault zones.

*Current address: Department of Geological Sciences, University of Cape Town, Private

Bag X3, Rondebosch 7701, South Africa; E-mail:
ake.fagereng@uct.ac.za.

ferent mineral flow mechanisms (Knipe, 1989).

We review observations in compositionally
heterogeneous tabular shear zones over a range
of crustal levels. These observations lead us to
suggest critical rheological factors affecting the
mechanical behavior of natural faults.
FAULT MODELS
Simple fault models assume a single, planar,
discrete interface (Fig. 1, model 1). Interseismic
1. Discrete
fault plane

Fault
Model

2. Homogeneous
viscous shear zone

Elastic strain

Elastic strain

Interseismic
strain
accommodation

Elastic strain

< 100 km

Fault-perpendicular distance

w/2

Competent
blocks

Shear zone margin

Shear zone margin

Interseismic
shear strain
rate

< 100 km

3. Heterogeneous continuousdiscontinuous shear zone

Finite
strain

Fault

INTRODUCTION
Crustal fault zones accommodate displacement in a variety of styles, including standard
earthquakes (3 < MW < 9.5) corresponding to rupture lengths ranging from <1 m to
>1000 km, non-volcanic tremor, and steady and
transitory aseismic slip (e.g., Shelly et al., 2006;
Schwartz and Rokosky, 2007). Geological evidence suggests that mainshock ruptures are,
at least locally, restricted to thin principal slip
zones (Chester and Chester, 1998); however,
aftershocks, earthquake swarms, and non-volcanic tremor occur as seismic activity distributed
throughout substantial rock volumes.
A subduction channel mlange predominantly composed of sheared, fluid-saturated,
trench-fill sediments has been interpreted to
occur along subduction thrust interfaces (Shreve
and Cloos, 1986; Moore and Byrne, 1987; von
Huene and Scholl, 1991; Abers, 2005). These
channels exhibit a wide range of seismic styles
(e.g., Shelly et al., 2006; Schwartz and Rokosky,
2007). Major crustal fault zones are also generally tabular structures. They contain mixtures
of competent and incompetent material (where
competence is a qualitative term describing relative viscosity), and therefore have a volumetrically mixed rheology with mechanical properties distinctly different from a discrete plane.
For example, a clay-rich gouge may function as
a flowing matrix around more competent material in continental fault zones at high crustal
levels. In general, however, viscosity contrasts
between juxtaposed rock bodies are likely to
change with depth and time because of varying
temperature and grain-size sensitivities of dif-

w/2

w/2

Fault-perpendicular distance

w/2

Fault-perpendicular distance

Postseismic
appearance

Figure 1. Schematic fault models, where displacement is accommodated across: (1) a zone
of infinitesimal thickness; (2) a viscous shear zone of width, w, 2 km; and (3) a shear zone
comprising heterogeneous rock mixture undergoing mixed continuous-discontinuous deformation. In model 1, interseismic strain is taken up by elastic deformation of rigid wall
rock, and irrecoverable strain is accommodated by coseismic slip. In models 2 and 3, incremental elastic strains developed and released during single earthquake cycle are small compared to the finite strains accommodated by ductile flow and/or displacement on multiple
shear surfaces within the shear zone. Shear strain rate is approximately constant within a
model 2 fault, but highly variable within a model 3 shear zone.

2010 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.
GEOLOGY,
August
2010
Geology,
August
2010;
v. 38; no. 8; p. 751754; doi: 10.1130/G30868.1; 4 figures.

751

Downloaded from geology.gsapubs.org on July 14, 2010

OBSERVATIONS OF HETEROGENEOUS,
TABULAR FAULT ZONES
Rheological heterogeneity and mixed continuous-discontinuous deformation seem to be
common features of tabular shear zones from
the surface down to at least the lower crust
(Hudleston, 1999; Goodwin and Tikoff, 2002;
Waters et al., 2003). Just as observed deformation features vary within exhumed shear zones,
seismic style varies within large-scale active
fault zones. Along the San Andreas fault, California, microseismic activity is low along segments
known to rupture in large earthquakes, and high
in aseismically sliding regions (Wallace, 1970;
Hill et al., 1990). Burford and Harsh (1980) studied aseismic slip on the San Andreas fault from
alignment arrays, and demonstrated that some
slip is localized to zones only a few meters thick,
suggesting that some creeping segments may be
zones of continuous-discontinuous deformation
comprising numerous aseismic shear discontinuities. The Japan and New Zealand subduction
megathrusts are other examples of faults where
creeping segments exhibit high microseismic
activity, while locked portions experience almost
no microearthquakes through the interseismic
period (e.g., Mazzotti et al., 2000; Wallace et
al., 2009). These fault zones have finite thicknesses, and shearing is distributed across tabular zones containing heterogeneous material
with varying physical properties. From seismic
velocity anomalies, subduction channel shear
zones are commonly inferred to be as much as
a few kilometers thick (Eberhart-Phillips and
Reyners, 1999; Abers, 2005). On the Hayward
fault, California, for which high-resolution
(<100 m relative location uncertainty) seismic
data are available, the distribution of relocated

752

Surface
2 trace

6
8

Surface
trace

25 cm

8
10

12

12
0

10

-2

Distance (km)

-4

-2

Shear zone margin

Competent phacoid

MLANGE DEFORMATION STYLES

The Chrystalls Beach Complex accretionary mlange of New Zealand (Nelson, 1982)
is a well-exposed example of a heterogeneous
shear zone deformed at seismogenic depths
(Figs. 2C and 3). Relatively competent phacoids
(chert > sandstone > metabasalt) generally

Shear zone margin

Depth (km)

next consider a specific example of an exhumed

mlange as a preliminary model for bulk rheology of continuous-discontinuous shear zones.

Competent phacoid

earthquakes (Waldhauser and Ellsworth, 2002)

suggests that this fault zone varies in thickness
from <1 km (Fig. 2A) to a 5-km-thick fracture
zone (Fig. 2B). Many fault zones, both ancient
and active, are therefore inferred to be tabular
structures over a range of depths and scales, and
to contain a mixture of materials with varying
competence deforming by different deformation modes. In the rock record, deformation of
mixed lithologies is well represented in tectonic
mlanges (e.g., Meneghini et al., 2009). We

Depth (km)

strain in the surrounding rock (Horsman and

Tikoff, 2005). Elastic strain release occurs not
only at the shear zone boundaries, but also on
distributed slip surfaces within the shear zone.
Variations in competence lead to fluctuations
in (Lister and Williams, 1983; Goodwin and
Tikoff, 2002). Consider a 1-km-thick, viscous
subduction channel shear zone accommodating displacement of 30 mm/yr by homogeneous
simple shear. Across such a shear zone, = v/w
= (30 mm/yr)/(1 km) 1012 s1. If, however,
viscous deformation is restricted to a 100 m
cumulative thickness of incompetent material,
is amplified to 1011 s1. If only 1% of the 1-kmthick shear zone is composed of incompetent
rock, will locally be ~1010 s1. The value of
may therefore vary significantly across a model
3 fault zone, affecting local rheological response
and perhaps causing a transition to more brittle
behavior in higher strain-rate regions (White,
2004). In the following section, we consider
the prevalence of heterogeneous, tabular shear
zones, and their bulk mechanical behavior.

Distance (km)

Distance

Figure 2. Examples of heterogeneous deformation and shear strain-rate distribution in continuous-discontinuous tabular shear zones: A and B illustrate the width of an active crustal
fault zone in fault-normal vertical cross sections of relocated microearthquakes on the Hayward fault. A: Union City, where fault zone is <1 km wide. B: San Leandro, where fault is an
~5-km-thick fracture zone (after Waldhauser and Ellsworth, 2002). C: Meter-scale shear zone
in Chrystalls Beach Complex. D: Schematic drawing of a heterogeneous shear zone (white
incompetent, blackcompetent) at any scale and depth. E: Curve shows shear strain-rate
variation along transect (A-A) through D. Note sharp strain-rate gradients at boundaries of
competent lenses, and dependence of strain rate on shear zone width.
Increasing ratio of incompetent/competent material
Discontinuities dominant

Mixed continuous-discontinuous

15 cm

10 cm

Seismic slip at kilometerscale possible in interacting

clusters of competent bodies

< meter scale seismic slip

possible

Continuous deformation dominant

5 cm

Microseismically active, flowing

zone, large ruptures do not nucleate
but may propagate through

High interaction through stress

bridges

Moderate interaction
between competent bodies

Low interaction between

competent bodies

Localized peaks in shear

strain rate

Fluctuating shear strain rates

Fairly uniform shear strain rates

Figure 3. Photographs and illustrations of three end members of mlange deformation based
on observations in Chrystalls Beach Complex, and inferred relations between these rheological end members and shear zone seismic styles. Note that these end members are likely time
dependent, and may represent a progression from an immature to a mature fault zone as finegrained material is produced through cataclasis (e.g., Knipe, 1989). Hammer is 35 cm long.

GEOLOGY, August 2010

Downloaded from geology.gsapubs.org on July 14, 2010

MLANGE BULK RHEOLOGY

Considering viscous deformation at the
grain scale, Handy (1990) recognized three
end members for rocks with two dominant
mineral phases: (1) a load-bearing framework
where a strong phase controls the bulk rheology; (2) a relatively strong phase forming elongate boudins within a weaker matrix with both
phases accommodating significant strain; and
(3) bulk rheology controlled by a flowing weak
matrix containing isolated undeformed lenses of

GEOLOGY, August 2010

the stronger phase. As these three end members

resemble the end members of mlange deformation (Fig. 3), we extrapolate Handys (1990)
model to a mlange shear zone incorporating
lithological components of varying competence
under seismogenic conditions.
Following Handy (1990), end members are
plotted in a diagram of normalized rock strength
[strength of rock (r)/strength of strongest
lithology (l)] versus volume fraction of the
weak lithology, w (Fig. 4). The area of dominantly discontinuous deformation is approximated by the strength curve for rocks with elliptical pores (Handy, 1990). For end members
where competent lenses float in a fine-grained
matrix, we adopt the general mixture rule (Ji,
2004). This model approximates experimental
data for crystal-melt suspensions, an analogue
to rigid blocks in a flowing matrix if the strength
contrast is large and the aggregate is matrix supported (Ji, 2004). The boundary between zones
2 and 3 is diffuse, and depends on geometry,
strength contrast, and w (Handy, 1990). Figure 4 highlights the point that the bulk strength
of a heterogeneous assemblage is not a simple
average of the respective materials properties.
The abrupt boundary of the discontinuities dominant field shows that if the strength contrast is
large, and the weak material is well connected,
only a small amount of weak material (perhaps
10%20%) may be needed for distributed, continuous deformation to predominate.
Increasing heterogeneity and diversity of strain rate
1.0

Mixed continuousdiscontinuous behavior

0.8
2
0.6
0.4
0.2
0

Discontinuities
dominant

rock / strong

have observable dimensions ranging from centimeters to tens of meters (though larger may
occur) and are within a relatively incompetent
cleaved mudstone matrix. Meter-scale shear
zones within the mlange (Fig. 2C) display
mixed continuous-discontinuous deformation
with rigid blocks in a flowing matrix, developed
at pumpellyite-actinolite facies metamorphic
conditions (220 C < T < 350 C; Frey et al.,
1991). These conditions correspond to the lower
reaches of the seismogenic zone in continental
crust (Scholz, 1988). The original mlange fabric is believed to have developed during early
soft-sediment deformation at shallow depth
(Nelson, 1982). These early fabrics are crosscut by anastomosing discontinuities, defined by
shear surfaces coated with stacked quartz slickenfibers (shear veins) that are typically 110 mm
thick. Matrix flow, occurring coincident with
shear vein formation, was accommodated by
dissolution-precipitation creep. The structural
fabrics observed in the Chrystalls Beach Complex are not unique; similar features have been
described in numerous exhumed mlanges, e.g.,
in paleosubduction complexes in California,
Alaska, and Italy (Meneghini et al., 2009, and
references therein), and in the Shimanto Complex in Japan (e.g., Ujiie, 2002).
The style of mlange deformation varies
depending on the relative proportions of competent and incompetent material, as shown by three
end-member behaviors (Fig. 3): (1) where high
competence material is dominant, shearing was
localized along discontinuities; (2) where mudstone is the dominant lithology, deformation
was distributed; and (3) a mixture of competent
and incompetent material exhibits both discrete
and distributed deformation. In this intermediate case, displacement occurred on numerous
discontinuities localized along contacts between
competent lenses and surrounding matrix.
Shear localization corresponds to steep gradients at competence boundaries (Figs. 2E and
3), emphasizing the role of compositional heterogeneity in localizing deformation onto discrete planes. We suggest that this intermediate
behavior may occur in fault zones over a wide
range of depths and tectonic settings (Fig. 2D)
(see reviews by Hudleston, 1999; Goodwin and
Tikoff, 2002, and references therein).

5
10

0.2

100
1000
0.4

Continuous
deformation
dominant
0.6

0.8

1.0

Figure 4. Graph, based on analyses of Handy

(1990) and general mixture rule of Ji (2004),
of normalized rock strength () against volume fraction of the weak phase (w) in a
two-phase aggregate. Contours represent
variable strength ratio (strength of strong
lithology/strength of weak lithology). Three
zones are defined based on rock deformation in Chrystalls Beach Complex (Fig. 3).

IMPLICATIONS FOR SEISMIC STYLE

In a mlange, the distribution of competent
material imposes length scales of deformation
(Fig. 3). Where w is low, shear localization
within a thin zone increases and may lead to
embrittlement and discontinuous deformation.

If w is high, strain is distributed, and steady

matrix flow is likely to persist (Figs. 3 and 4).
Zones or isolated volumes of low w therefore
act as asperities, here defined as regions where
brittle failure tends to nucleate. Cloos (1992)
correlated asperity dimensions with earthquake
magnitude; we therefore suggest that the size of
low w zones may be critical to potential rupture
size. Mega-asperities may arise where atypical
features such as seamounts or clusters of interacting competent phacoids create log-jams
in the subduction channel shear zone (Fig. 3)
(Cloos and Shreve, 1996).
If the range in mlange structures reflects
seismic styles exhibited during a period of
active deformation, discontinuous deformation
records localized seismic and/or aseismic slip,
whereas volumetrically continuous fabrics predominantly record distributed aseismic deformation (Fig. 3). Matrix-dominated regions are
predicted to deform aseismically, with microseismic activity along localized shear discontinuities. While positive discrimination between
the rock products of seismic and aseismic deformation is problematic, we infer that ruptures are
more likely to nucleate where w is small and
is locally elevated, thereby restricting nucleation
sites to regions with relatively high proportions
of competent rock. Once a rupture has nucleated, high strain rate at the rupture tips may
allow propagation into normally aseismic, relatively incompetent regions. One could argue,
therefore, for a critical distinction in mechanical behavior between subduction channel shear
zones containing phacoids (or phacoid clusters)
with dimensions comparable to channel thickness, and those with phacoid dimensions significantly less than the channel thickness. Seismic
style in the former would likely be dominated
by large ruptures along channel boundaries,
while in the latter shearing would be accommodated volumetrically by a mixture of distributed
aseismic slip and microseismicity.
In active volumetric shear zones undergoing
mixed continuous-discontinuous deformation
of a heterogeneous rock assemblage, the composition and proportions of different shear zone
constituents exert a strong control on strain-rate
distribution and the dominant seismic style.
Though not considered further here, heterogeneous distribution of fluid overpressure within
the shear zone is another critical factor controlling rheology (Sibson, 1996), and possibly
seismic style (Fagereng and Ellis, 2009). The
subduction channel assemblage is thus considered to exert a first-order control on the rheology
of the subduction thrust interface. Comparable
effects of heterogeneous strain localization are
seen in other fault zones over a range of metamorphic conditions. There is therefore a need to
develop volumetric shear zone models that take
into account mixed competency content, which

753

Downloaded from geology.gsapubs.org on July 14, 2010

seems likely to be an important factor affecting
rupture nucleation and seismic style in a variety
of settings.
ACKNOWLEDGMENTS
We thank Laurel Goodwin and Francesca
Meneghini for critical reviews that significantly improved the paper. Fagereng was supported by a GNS
Science Hazards scholarship, and a University of
Otago postgraduate publishing bursary. Funding for
the project was provided by GNS Science.
REFERENCES CITED
Abers, G.A., 2005, Seismic low-velocity layer at the
top of subducting slabs: Observations, predictions and systematics: Physics of the Earth
and Planetary Interiors, v. 149, p. 729, doi:
10.1016/j.pepi.2004.10.002.
Burford, R.O., and Harsh, P.W., 1980, Slip on the
San Andreas fault in central California from
alinement array surveys: Seismological Society of America Bulletin, v. 70, p. 12331261.
Chester, F.M., and Chester, J.S., 1998, Ultracataclasite structure and friction processes on the
Punchbowl fault, San Andreas system, California: Tectonophysics, v. 295, p. 199221, doi:
10.1016/S0040-1951(98)00121-8.
Cloos, M., 1992, Thrust-type subduction zone earthquakes and seamount asperities: A physical
model for earthquake rupture: Geology, v. 20,
p. 601604, doi: 10.1130/0091-7613(1992)020<
0601:TTSZEA>2.3.CO;2.
Cloos, M., and Shreve, R.L., 1996, Shear-zone thickness and the seismicity of Chilean- and Marianastype subduction zones: Geology, v. 24, p. 107
110, doi: 10.1130/0091-7613(1996)024<0107:
SZTATS>2.3.CO;2.
Eberhart-Phillips, D., and Reyners, M., 1999, Plate interface properties in the northeast Hikurangi subduction zone, New Zealand, from converted seismic waves: Geophysical Research Letters, v. 26,
p. 25652568, doi: 10.1029/1999GL900567.
Fagereng, A., and Ellis, S., 2009, On factors controlling the depth of interseismic coupling on the
Hikurangi subduction interface, New Zealand:
Earth and Planetary Science Letters, v. 278,
p. 120130, doi: 10.1016/j.epsl.2008.11.033.
Frey, M., de Capitani, C., and Liou, J.G., 1991, A new
petrogenetic grid for low-grade metabasites:
Journal of Metamorphic Geology, v. 9, p. 497
509, doi: 10.1111/j.1525-1314.1991.tb00542.x.
Goodwin, L.B., and Tikoff, B., 2002, Competency
contrast, kinematics and the development of
foliations and lineations in the crust: Journal of
Structural Geology, v. 24, p. 10651085, doi:
10.1016/S0191-8141(01)00092-X.
Handy, M.R., 1990, The solid-state flow of polymineralic rocks: Journal of Geophysical Research, v. 95, p. 86478661, doi: 10.1029/
JB095iB06p08647.
Hill, D.P., Eaton, J.P., and Jones, L.M., 1990, Seismicity, 198086, in Wallace, R.E., ed., The San An-

754

dreas fault system, California: U.S. Geological

Survey Professional Paper 1515, p. 115151.
Horsman, E., and Tikoff, B., 2005, Quantifying
simultaneous discrete and distributed deformation: Journal of Structural Geology, v. 27,
p. 11681189, doi: 10.1016/j.jsg.2004.08.001.
Hudleston, P., 1999, Strain compatibility and shear
zones: is there a problem?: Journal of Structural Geology, v. 21, p. 923932, doi: 10.1016/
S0191-8141(99)00060-7.
Ji, S., 2004, A generalized mixture rule for estimating the viscosity of solid-liquid suspensions
and mechanical properties of polyphase rocks
and composite materials: Journal of Geophysical Research, v. 109, B10207, doi: 10.1029/
2004JB003124.
Knipe, R.J., 1989, Deformation mechanismsRecognition from natural tectonites: Journal of Structural Geology, v. 11, p. 127146, doi: 10.1016/
0191-8141(89)90039-4.
Lister, G.S., and Williams, P.F., 1983, The partitioning of deformation in flowing rock masses: Tectonophysics, v. 92, p. 133, doi: 10.1016/0040
-1951(83)90083-5.
Mazzotti, S., LePichon, X., Henry, P., and Miyazaki,
S., 2000, Full interseismic locking of the Nankai and Japan-west Kurile subduction zones: An
analysis of uniform elastic strain accumulation
in Japan constrained by permanent GPS: Journal of Geophysical Research, v. 105, p. 13,159
13,177, doi: 10.1029/2000JB900060.
Meneghini, F., Marroni, M., Moore, J.C., Pandolfi,
L., and Rowe, C.D., 2009, The processes of underthrusting and underplating in the geologic
record: Structural diversity between the Franciscan Complex (California), the Kodiak Complex (Alaska) and the Internal Ligurian Units
(Italy): Geological Journal, v. 44, p. 126152,
doi: 10.1002/gj.1144.
Moore, J.C., and Byrne, T., 1987, Thickening of fault
zones: A mechanism of mlange formation in
accreting sediments: Geology, v. 15, p. 1040
1043, doi: 10.1130/0091-7613(1987)15<1040:
TOFZAM>2.0.CO;2.
Nelson, K.D., 1982, A suggestion for the origin of
mesoscopic fabric in accretionary melange, based
on features observed in the Chrystalls Beach
Complex, South Island, New Zealand: Geological Society of America Bulletin, v. 93, p. 625
634, doi: 10.1130/0016-7606(1982)93<625:
ASFTOO>2.0.CO;2.
Savage, J.C., and Burford, R.O., 1973, Geodetic determination of relative plate motion in central California: Journal of Geophysical Research, v. 78,
p. 832845, doi: 10.1029/JB078i005p00832.
Scholz, C.H., 1988, The brittle-plastic transition
and the depth of seismic faulting: Geologische
Rundschau, v. 77, p. 319328, doi: 10.1007/
BF01848693.
Schwartz, S.Y., and Rokosky, J.M., 2007, Slow slip
events and seismic tremor at Circum-Pacific
subduction zones: Reviews of Geophysics,
v. 45, RG3004, doi: 10.1029/2006RG000208.

Shelly, D.R., Beroza, G.C., Ide, S., and Nakamula,

S., 2006, Low-frequency earthquakes in Shikoku, Japan, and their relationship to episodic
tremor and slip: Nature, v. 442, p. 188191,
doi: 10.1038/nature04931.
Shreve, R.L., and Cloos, M., 1986, Dynamics of
sediment subduction, melange formation, and
prism accretion: Journal of Geophysical Research, v. 91, p. 10,22910,245, doi: 10.1029/
JB091iB10p10229.
Sibson, R.H., 1996, Structural permeability of
fluid-driven fault-fracture meshes: Journal of
Structural Geology, v. 18, p. 10311042, doi:
10.1016/0191-8141(96)00032-6.
Ujiie, K., 2002, Evolution and kinematics of an ancient dcollement zone, mlange in the Shimanto accretionary complex of Okinawa Island,
Ryukyu Arc: Journal of Structural Geology,
v. 24, p. 937952, doi: 10.1016/S0191-8141
(01)00103-1.
von Huene, R., and Scholl, D.W., 1991, Observations
at convergent margins concerning sediment
subduction, subduction erosion, and the growth
of continental crust: Reviews of Geophysics,
v. 29, p. 279316, doi: 10.1029/91RG00969.
Wallace, L., and 21 others, 2009, Characterizing the
seismogenic zone of a major plate boundary
subduction thrust: Hikurangi Margin, New Zealand: Geochemistry, Geophysics, Geosystems,
v. 10, Q10006, doi: 10.1029/2009GC002610.
Wallace, R.E., 1970, Earthquake recurrence intervals
on the San Andreas fault: Geological Society
of America Bulletin, v. 81, p. 28752890, doi:
10.1130/0016-7606(1970)81[2875:ERIOTS]2
.0.CO;2.
Waldhauser, F., and Ellsworth, W.L., 2002, Fault structure and mechanics of the Hayward Fault, California, from double-difference earthquake locations: Journal of Geophysical Research, v. 107,
no. B3, 2054, doi: 10.1029/2000JB000084.
Waters, C., Tikoff, B., Goodwin, L.B., and Little,
T.A., 2003, Ductile instabilities and structural
heterogeneity in the lower continental crust, in
Goff, J.A., and Holliger, K., eds., Heterogeneity in the crust and upper mantle: Nature, scaling and seismic properties: New York, Kluwer
Academic, p. 136.
White, J.C., 2004, Instability and localization of deformation in lower crustal granulites, Minas
fault zone, Nova Scotia, Canada, in Alsop, G.I.,
et al., eds., Flow processes in faults and shear
zones: Geological Society of London Special
Publication 224, p. 2537, doi: 10.1144/GSL
.SP.2004.224.01.03.

Manuscript received 17 November 2009

Revised manuscript received 19 March 2010
Manuscript accepted 25 March 2010
Printed in USA

GEOLOGY, August 2010