Download Thermal and metamorphic environment of subduction zone episodic

Click
Here
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 114, B00A07, doi:10.1029/2008JB005978, 2009
for
Full
Article
Thermal and metamorphic environment of subduction zone episodic
tremor and slip
Simon M. Peacock1
Received 1 August 2008; revised 10 December 2008; accepted 29 May 2009; published 5 August 2009.
[1] Episodic tremor and slip (ETS) have been detected in the Cascadia and southwest
Japan subduction zones, where the subducting crust is relatively warm because of the
young incoming lithosphere (<20 Ma) and modest plate convergence rates (40–60 mm/a).
In the southwest Japan subduction zone, low-frequency earthquakes occur on the plate
interface at depths of 30–35 km beneath Shikoku where finite element thermal
models predict temperatures of 425°C in the subducting oceanic crust and at depths
of 30–40 km beneath the Kii Peninsula where predicted temperatures are 325°C.
Warmer temperatures of 575°C are predicted at ETS depths beneath southern Vancouver
Island in the Cascadia subduction zone, but here tremor also occurs within the overlying
fore-arc crust where temperatures are lower. In the southwest Japan and Cascadia
subduction zones, subducting oceanic crust passes through the blueschist, greenschist, and
amphibolite metamorphic facies where mineral dehydration reactions are complex. The
different temperatures predicted for the two subduction zones suggest that ETS does not
coincide with a specific temperature or metamorphic reaction. Several lines of evidence
indicate that a free H2O-rich fluid is present, at least transiently, in subducting oceanic
crust and fluids released by prograde metamorphic dehydration reactions may help
trigger or enable ETS within the subducting plate. Less clear is the role H2O may play in
tremor observed in the Cascadia fore-arc crust where free H2O may exist locally in faults
and fractures, but retrograde hydration reactions are expected to consume H2O.
Citation: Peacock, S. M. (2009), Thermal and metamorphic environment of subduction zone episodic tremor and slip,
J. Geophys. Res., 114, B00A07, doi:10.1029/2008JB005978.
1. Introduction
[2] Many features of subduction zones are directly linked
to the subduction of H2O. Indeed, subduction on Earth
likely occurs only because H2O and hydrous minerals
weaken the lithosphere [e.g., Regenauer-Lieb et al., 2001;
O’Neill et al., 2007]. Over the past decade, continuous GPS
networks have revealed slow slip events near or within the
seismogenic zone in a number of circum-Pacific subduction
zones (see review by Schwartz and Rokosky [2007]). In the
Cascadia and southwest Japan subduction zones, slow slip
events correlate with seismic tremor, a phenomenon referred
to as episodic tremor and slip (ETS) [Rogers and Dragert,
2003; Obara et al., 2004]. Several authors have proposed
that H2O-rich fluids released by dehydration reactions in the
subducting plate play a role in triggering or enabling ETS
[e.g., Obara, 2002; Seno and Yamasaki, 2003; Katsumata
and Kamaya, 2003; Kodaira et al., 2004; Shelly et al.,
2006; Wang et al., 2006; Liu and Rice, 2007; Yoshioka et
al., 2008; Furukawa, 2009]. Alternatively, ETS may represent a mode of shear failure intermediate between seismic
stick slip and aseismic stable sliding, with or without the
1
Department of Earth and Ocean Sciences, University of British
Columbia, Vancouver, British Columbia, Canada.
Copyright 2009 by the American Geophysical Union.
0148-0227/09/2008JB005978$09.00
presence of fluids [e.g., Rogers and Dragert, 2003; Ide et
al., 2007]. In this contribution, I present the results of
thermal-petrologic modeling of the Cascadia and southwest
Japan subduction zones, which provide insight into the
thermal, metamorphic, and fluid conditions under which
ETS occurs.
[3] In subduction zones, the subducting plate and the
overriding plate experience different thermal and petrologic
histories. Mafic oceanic crust and ultramafic uppermost
mantle undergo retrograde metamorphism as the oceanic
lithosphere cools away from the ridge and variable
amounts of hydrous minerals form in the oceanic crust
and mantle as a result of hydrothermal circulation [e.g.,
Davis and Elderfield, 2003]. More H2O is added to the
oceanic lithosphere as sediments are deposited on top of the
plate and additional hydration reactions caused by hydrothermal circulation in the trench – outer rise region [Peacock,
2001; Ranero et al., 2003]. Global estimates of subduction
zone H2O fluxes reveal that oceanic sediments contain most
of the pore H2O entering subducting zones whereas the
variably hydrated oceanic crust contains most of the chemically bound H2O [e.g., Peacock, 2004].
[4] During subduction, the sediments, oceanic crust, and
uppermost mantle of the subducting plate undergo prograde
metamorphism, characterized by increasing temperature and
pressure, and by the release of H2O. Pore H2O is expelled
by porosity collapse caused by weakening of the solid
B00A07
1 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
Figure 1. Geotectonic map of the Cascadia subduction
zone showing location of thermal model constructed
through southern Vancouver Island (VI). Line labeled A-A00
represents location of complete thermal model; thicker line
labeled A-A0 represents portion of thermal model depicted
in Figure 3. Depth contours, in km, for subducting Juan de
Fuca plate and location of July 2004 seismic tremor beneath
southern Vancouver Island (small squares) from Kao et al.
[2005]. Solid triangles represent Holocene volcanoes.
matrix. Chemically bound H2O is liberated by metamorphic
dehydration reactions, most of which depend strongly on
temperature at depths less than 50 km. The thermal structure
of a subduction zone determines where dehydration reactions occur in subducting lithosphere. In cool subduction
zones, dehydration reactions are predicted to extend to
depths of 100 km or more, whereas in warm subduction
zones, like Cascadia and southwest Japan, dehydration
reactions are restricted to shallower depths and considerable
H2O is released beneath the fore arc [Kirby et al., 1996;
Peacock, 1996].
[5] Compared to the subducting plate, the overriding
(fore-arc) plate has typically undergone a complex thermal
and petrologic history related to formation and accretion.
Fore-arc terranes may include fragments of continental
crust, accreted island arcs, oceanic plateaus, and accretionary prisms. Low heat flow measurements and thermal
models demonstrate that fore-arc temperatures decrease
over time as heat conducts downward into the subducting
lithosphere [e.g., Hyndman and Wang, 1993]. Thus, during
subduction, the overriding fore-arc crust and mantle undergo retrograde metamorphism. Variable amounts of H2O may
be incorporated into the fore-arc crust and mantle as a result
of the infiltration of fluids derived from the subducting
plate.
2. Location of Subduction Zone ETS
[6] In the Cascadia subduction zone (Figure 1), continuous
GPS data reveal that slow slip events, lasting one to several
B00A07
weeks, occur periodically on the plate interface downdip of
the seismogenic zone [Dragert et al., 2001]. These events
partially release stresses at depth that accumulate from
ongoing plate convergence, but increase stresses on the
shallower locked portion of the plate boundary. Rogers
and Dragert [2003] demonstrated that slow slip events
correlate spatially and temporally with unusual seismic
tremor and refer to the two phenomena as episodic slip
and tremor, or ETS. It is difficult to precisely locate ETS.
Surface GPS data can be well modeled by 2 cm of
aseismic slip along the plate interface at depths of 30–
40 km [Dragert et al., 2001], but this solution is not unique.
Locating the seismic tremor is challenging because of the
lack of distinct P or S wave arrivals and the tendency of
tremor to occur in clusters. Kao et al. [2005, 2007]
overcame these difficulties by applying a source-scanning
algorithm to tremor observed in the northern Cascadia
subduction zone, which revealed that tremor occurs over a
considerable depth range extending from the shallow forearc crust down into the subducting oceanic crust. It remains
unknown whether the correlated slow slip occurs over a
similar depth range (10 –50 km) as the tremor, or if the slow
slip is confined to the plate interface [Kao et al., 2007].
[7] Obara [2002] observed seep seismic tremor along the
southwest Japan subduction zone (Figure 2). Using a crosscorrelation technique, Obara [2002] showed that tremor
originated from the plate interface at depths 30 km depth
Figure 2. Geotectonic map of southwest Japan subduction
zone showing location of the thermal models constructed
through Shikoku and Kii Peninsula. Lines labeled B-B00 and
C-C00 represent locations of complete thermal models;
thicker lines labeled B-B0 and C-C0 represent portions of
thermal models depicted in Figures 4 and 5, respectively.
Depth contours, in km, for subducting Philippine Sea plate
from Baba et al. [2002], Nakajima and Hasegawa [2007],
and Hirose et al. [2008] and location of low-frequency
earthquakes (small black pluses) from Hirose et al. [2008].
Solid triangles represent Holocene volcanoes.
2 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
beneath southwest Japan. In map view tremor occurs over
an area extending 600 km along strike, but a pronounced
gap exists beneath easternmost Shikoku and the Kii channel
[Obara, 2002] (Figure 2). Seismicity in southwest Japan
includes deep low-frequency earthquakes that are spatially
and temporally associated with deep tremor [Katsumata and
Kamaya, 2003; Obara and Hirose, 2006]. Hirose et al.
[2008] used double-difference tomography to determine
more accurately the geometry of the subducting Philippine
Sea plate. Deep low-frequency earthquakes occur along the
plate interface at a depth of 30 km, extending to slightly
greater depths beneath the Kii Peninsula and the Tokai
district to the northeast [Hirose et al., 2008] (Figure 2).
Slow slip events, correlating with periods of increased
tremor, have been detected in tilt data extracted from the
Hi-net high-sensitivity seismograph network [Obara et al.,
2004]. Tremor events in the southwest Japan subduction
zone appear to occur along or near the plate interface and, in
contrast to Cascadia, do not appear to extend upward into
the overlying fore-arc crust [Shelly et al., 2006; Wang et al.,
2006; Hirose et al., 2008]. Research continues to determine
if this contrast between the two subduction zones is real, or
reflects differences in the techniques use to locate the tremor
source.
3. Thermal Modeling
[8] Temperatures at depths of 30 km in subduction
zones are determined primarily by the thermal structure of
the incoming lithosphere and the plate convergence rate
[e.g., Peacock, 2003]. Both the Cascadia and southwest
Japan subduction zones are characterized by relatively
warm (young) incoming lithosphere and modest convergence rates. Here, two-dimensional finite element heat
transfer models are used to calculate temperatures at depth
in the two subduction zones. The two-dimensional models
are constructed perpendicular to the subduction zone trench
and volcanic arc, rather than parallel to the oblique convergence vector, in order to minimize heat transfer in the third
dimension. Because the structure of subduction zones varies
along strike and the geometry of subduction may vary over
time, two-dimensional thermal models are only approximations, with estimated uncertainties of 50 –100°C for forearc depths of 30 –40 km.
[9] Details of the numerical model are given by Peacock
and Wang [1999] and Peacock [2003]. Briefly, for each
subduction zone: (1) the geometry of the subducting slab is
defined by Wadati-Benioff seismicity and seismic reflection
and refractions surveys; (2) the orthogonal convergence rate
is defined using recent plate motion models; (3) the age of
the incoming lithosphere at the deformation front (trench) is
defined by marine magnetic anomalies on the Juan de Fuca
[Wilson, 1993] and Philippine Sea plates [Chamot-Rooke et
al., 1987] and an oceanic geotherm [Stein and Stein, 1992]
of corresponding age defines the trench-side boundary
condition; (4) a continental geotherm yielding a surface
heat flux of 65 mW/m2 is used for the arc-side boundary
condition; (5) the upper (0 – 15 km depth) and lower (15 –
30 km) crust contains radioactive heat-producing elements
generating 1.3 and 0.27 mW/m3, respectively; (6) the
thermal conductivity of the crust is 2.5 W/(m K) and the
B00A07
mantle is 3.1 W/(m K); and (7) flow in the mantle wedge is
simulated using Batchelor’s isoviscous corner flow solution,
truncated at the tip of the mantle wedge to satisfy surface
heat flow measurements. Note that the choice of mantle
rheology does not significantly affect the temperatures
calculated for the shallow fore arc, unless flow in the mantle
wedge corner were to extend updip to depths <50 km,
which would be inconsistent with fore-arc heat flow observations [e.g., Wiens et al., 2008].
[10] A steady state thermal model of the Cascadia subduction zone was constructed through southern Vancouver
Island (Figure 1) using the subducting plate geometry of
Rogers [1998]. The age of the Juan de Fuca oceanic
lithosphere at the deformation front is 7.5 Ma and the
convergence rate is 40 mm/a based on DeMets et al.’s
[1994] NUVEL-1A plate motion model. At the deformation
front, a 3-km-thick sedimentary section overlies and insulates the oceanic crust [Hyndman and Wang, 1993].
[11] Two transient thermal models were constructed for
the southwest Japan subduction zone, one through Shikoku
and one through the Kii peninsula located 200 km to the
east (Figure 2). The geometry of the subducting Philippine
Sea plate was defined using the results of Baba et al.
[2002], Nakajima and Hasegawa [2007], and Hirose et al.
[2008] (Figure 2). At the Nankai Trough, a 1.2-km-thick
sedimentary section overlies and insulates the oceanic crust
[Hirahara, 1981]. The thermal structure beneath southwest
Japan has been affected by the subduction of the Shikoku
ridge, a back-arc spreading center that was active until 15 Ma
ago [Hibbard and Karig, 1990]. For both transects, a steady
state subduction thermal structure, calculated using a 100 Ma
incoming lithosphere and a 90 mm/a convergence rate, was
used as the initial thermal condition for the transient
solution. Fifteen million years of thermal evolution was
then simulated during which time the slab age, and therefore
the thermal structure, of the incoming oceanic lithosphere
was varied. For the Shikoku transect, a 0 Ma oceanic
geotherm was used to define the thermal structure of the
incoming lithosphere at the start of the transient simulation
and the age of the incoming oceanic lithosphere was
increased linearly from 0 Ma to 15 Ma during the 15 Ma
of simulation time. For the Kii peninsula transect, where the
Shikoku ridge was subducted earlier, a 5 Ma oceanic geotherm was used to define the thermal structure of the
incoming lithosphere at the start of the transient simulation
and the age of the incoming oceanic lithosphere was
increased linearly from 5 Ma to 20 Ma during the 15 Ma
of simulation time. Orthogonal convergence rates between
the Philippine Sea and Amuria plates were calculated using
the REVEL plate velocity model [Sella et al., 2002]. For the
Shikoku and Kii peninsula transects, orthogonal convergence rates of 58 and 52 mm/a, respectively, were used for
the transient solution.
4. Predicted Temperatures and Surface Heat Flux
[12] The calculated thermal structures for the Cascadia
(southern Vancouver Island transect) and southwest Japan
(Shikoku and Kii peninsula transects) subduction zones are
depicted in Figures 3, 4, and 5. The subduction of young
oceanic lithosphere at 40– 60 mm/a in the Cascadia and
3 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
B00A07
Figure 3. Calculated (a) surface heat flux and (b) thermal structure for southern Vancouver Island
(Cascadia subduction zone), with inferred location of ETS (shaded region) from Kao et al. [2005].
Contour interval is 100°C. Dashed line labeled M represents fore-arc Moho. See Figure 1 for location of
transect A-A0.
southwest Japan subduction zones results in relatively warm
calculated fore-arc temperatures as compared to most subduction zone fore arcs. For example, at 30 km depth,
calculated subduction interface temperatures for Cascadia
(510°C) and southwest Japan (Shikoku = 390°C; Kii =
280°C) are higher than temperatures of 160– 175°C calculated for the NE Japan and Izu-Bonin subduction zones
[Peacock, 2003], which are characterized by the rapid
subduction of old oceanic lithosphere. Beneath southwest
Japan, the numerical simulations predict the Kii Peninsula is
cooler than the Shikoku fore arc, consistent with the
findings of Hyndman et al. [1995]. These along-strike
thermal variations for southwest Japan reflect variations in
the slab geometry (steeper beneath the Kii Peninsula) and
the timing of the subduction of the fossil Shikoku spreading
center (subducting beneath Shikoku today).
[13] The predicted surface heat flux out of the Cascadia
(southern Vancouver Island transect) and southwest Japan
(Shikoku and Kii peninsula transects) fore arcs are depicted
in Figures 3a, 4a, and 5a. Surface heat flux decreases from
the trench toward the arc. In the fore arc, where the
subducting slab is at a depth of 30 km, the predicted surface
heat flux is 60 mW/m2 for Cascadia, 50 mW/m2 for
Shikoku, and 40 mW/m2 for the Kii Peninsula. The
predicted fore-arc heat fluxes are inversely proportional to
the convergence rate and age of the incoming plate, and are
Figure 4. Calculated (a) surface heat flux and (b) thermal structure for Shikoku transect (southwest
Japan subduction zone), with inferred location of ETS (shaded region) from Hirose et al. [2008]. Contour
interval is 100°C. Dashed line labeled M represents fore-arc Moho. See Figure 2 for location of transect
B-B0.
4 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
B00A07
of 30 – 50 km do not depend on the mantle viscosity
formulation. It is less clear why the two approaches predict
substantially different temperatures at 30 km depth, but
higher temperatures predicted by my models are probably
attributable to the inclusion (in my models) of (1) a 1.2-kmthick sedimentary section on top of the incoming Philippine
Sea plate, which acts like an insulating blanket to elevate
temperatures in the underlying crust; (2) a fore-arc crust that
has a lower thermal conductivity than the underlying
mantle; and (3) radioactive heat production in the fore-arc
crust. In addition, thermal models like Yoshioka et al.’s
[2008] that are constructed at an angle to the structure of a
subduction zone may introduce uncertainties owing to heat
flow in the third dimension (perpendicular to the thermal
model).
5. Metamorphic Conditions Where ETS Occurs
Figure 5. Calculated (a) surface heat flux and (b) thermal
structure for Kii Peninsula transect (southwest Japan
subduction zone), with inferred location of ETS (shaded
region) from Hirose et al. [2008]. Contour interval is
100°C. Dashed line labeled M represents fore-arc Moho.
See Figure 2 for location of transect C-C0.
broadly consistent with the marine and land heat flow data
reported by Hyndman and Wang [1995] and Hyndman et al.
[1995].
[14] Recently, Yoshioka et al. [2008] presented twodimensional thermal models constructed across southwest
Japan in order to investigate along-strike regional variations
in earthquake occurrence frequency. Two of their models lie
close to the two southwest Japan profiles presented here.
Compared to the models presented here, Yoshioka et al.
[2008] predict cooler slab interface temperatures at 30 km
depth and dramatically hotter slab interface temperatures at
50 km depth. At 30 km depth beneath Shikoku and the Kii
peninsula, Yoshioka et al. [2008] predict slab interface
temperatures of 300°C and 240°C, respectively, as compared to 390°C and 280°C predicted by the models presented here (Figures 4 and 5). At 50 km depth beneath
Shikoku and the Kii peninsula, Yoshioka et al. [2008]
predict very warm slab interface temperatures of 870°C
and 810°C, compared to 490°C and 350°C predicted by the
models presented here (Figures 4 and 5).
[15] Yoshioka et al.’s [2008] thermal models differ from
those presented here in several respects. The very large
differences in calculated temperatures at 50 km depth,
primarily reflect different mantle flow models. Yoshioka et
al. [2008] employed a temperature- and depth-dependent
mantle rheology, where flow in the mantle wedge is driven
by full coupling between the slab and mantle wedge
beginning at 30 km depth. Thermal models like Yoshioka
et al.’s [2008], which employ a very shallow slab – mantle
wedge coupling depth, predict high fore-arc temperatures
and thermal gradients [Furukawa, 1993] that are inconsistent with observed low heat flow in fore arcs [Hyndman et
al., 1995; Peacock and Wang, 1999; Peacock, 2003; Wiens
et al., 2008]. In models where coupling between the slab
and mantle wedge begins at greater depth (e.g., 70 km in
the models presented here), fore-arc temperatures at depths
[16] To gain insight into the petrology of the subducting
oceanic crust and regions of dehydration, calculated
pressure-temperature (P-T) paths for the subducting oceanic
crust are overlain on a phase diagram constructed for
hydrous mid-ocean ridge basalt [Hacker et al., 2003a,
2003b] in Figure 6. The basaltic phase diagram presented
in Figure 6a depicts the P-T stability field of different
metamorphic facies characterized by specific mineral
assemblages. For example, a greenschist facies metabasalt
consists of epidote + albite (plagioclase feldspar) + actinolite
(calcic amphibole) + chlorite + quartz ± accessory minerals.
Metamorphic facies boundaries mark important metamorphic reactions that result in different mineral assemblages. It
is important to note that, with rare exceptions, metamorphic
facies boundaries do not correspond to a single, univariant
metamorphic reaction. Rather, metamorphic facies boundaries involve continuous, multivariant metamorphic reactions and are generally broader than depicted in Figure 6a.
Many metamorphic facies boundaries are marked by major
dehydration reactions, but dehydration reactions also occur
within metamorphic facies. The gray shading in Figure 6
reflects the maximum calculated H2O content of each
metamorphic facies; mineral assemblages stable in lowtemperature metamorphic facies are capable of containing
more bound H2O than higher temperature metamorphic
facies. See Hacker et al. [2003a] for a detailed description
of how the phase diagram was constructed and the H2O
calculations.
[17] P-T paths calculated for the three transects differ
considerably indicating that the subducting oceanic crust
experiences significantly different metamorphic conditions
in each region (Figure 6b). Beneath southern Vancouver
Island (Cascadia), subducting oceanic crust descends
along a warm P-T path and experiences greenschist !
amphibolite ! eclogite facies metamorphism. Beneath
Shikoku (southwest Japan), the predicted P-T paths are
slightly cooler, with the subducting oceanic crust predicted
to pass through the epidote blueschist facies. Beneath the
Kii Peninsula (southwest Japan), subducting oceanic crust
descends along even cooler P-T paths and experiences
blueschist facies metamorphism.
[18] The different P-T conditions in the Cascadia and
southwest Japan fore arcs suggest that ETS is not a result of
specific thermal or metamorphic conditions. At ETS depths,
5 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
B00A07
within subducting oceanic crust does not appear to be
confined to a specific temperature or metamorphic facies
or to a particular metamorphic transition.
6. Metamorphic Fluids and Permeability
Figure 6. Calculated metamorphic conditions encountered
by subducting oceanic crust in Cascadia and southwest
Japan subduction zones. (a) Phase diagram showing
metamorphic facies for mid-ocean ridge basalt and
calculated maximum H2O contents after Hacker et al.
[2003a]. Metamorphic facies: eA, epidote amphibolite;
eB, epidote blueschist; egA, epidote-garnet amphibolite;
jeB, jadeite-epidote blueschist; laE, lawsonite-amphibole
eclogite, PA; prehnite-actinolite; PP, prehnite-pumpellyite;
Z, zeolite facies. (b) Calculated P-T paths for top (solid
lines) and base (dashed lines) of subducting oceanic crust
for three transects, superimposed on metamorphic facies.
Stippled regions represent calculated P-T conditions within
subducting oceanic crust where ETS region occurs.
subducting oceanic crust is predicted to be 525– 650°C
and undergoing amphibolite facies metamorphism beneath
southern Vancouver Island, 375 – 475°C and undergoing
epidote-blueschist facies metamorphism beneath Shikoku,
and 275– 350°C and undergoing blueschist facies metamorphism beneath the Kii Peninsula (Figure 6b). Thus, ETS
[19] While ETS appears not to be linked to specific
metamorphic reaction, H2O-rich metamorphic fluids are
likely present where ETS occurs within the subducting
oceanic crust or along the subduction interface. In the
subducting oceanic crust and uppermost mantle, prograde
metamorphic dehydration reactions release H2O-rich fluids.
Pore pressures will be near lithostatic in rocks undergoing
active dehydration [e.g., Fyfe et al., 1978]. The extent to
which significant volumes of H2O exist at a given time
depends on fluid production rates and the permeability
structure of the subducting plate, the overriding plate, and
the plate interface. Subduction is a slow geologic process
and the amount of H2O produced is correspondingly small.
For warm subduction zones, like Cascadia and southwest
Japan, Hyndman and Peacock [2003] calculated that metamorphic dehydration reactions in subducting oceanic crust
and mantle beneath the fore-arc release on the order of
100 mL (a tea cup!) of H2O per square meter column of
rock per year.
[20] The permeability of metamorphic rocks undergoing
prograde metamorphism at depths of 30 to 40 km and
temperatures of 300 to 600°C is not well constrained.
Because metamorphic rocks exposed at the surface are
dry, permeabilities must be sufficient to allow metamorphic
fluids to escape on geologic (million year) time scales. Most
metamorphic dehydration reactions result in a reduction in
mineral volume, so porosity is expected to increase, at least
transiently while the dehydration reaction is occurring. If
the surrounding rocks are sufficiently permeable, the H2O
produced by dehydration reactions may drain away rapidly.
Alternatively, if the surrounding rocks are relatively impermeable, fluid draining may not occur until the fluid pressure
exceeds the confining pressure (s3) plus tensile strength of
the rock causing hydraulic fracture and fluid escape [Fyfe et
al., 1978]. Fluid release may be quasi-episodic with dehydration reactions increasing fluid pressure slowly over time
until hydraulic fracturing drains the fluid and the process
repeats.
[21] Subduction zone plate boundaries are Earth’s largest
thrust faults with cumulative slip commonly exceeding
thousands of kilometers. Field observations of large faults
reveal that rather than a single discrete surface, large faults
are complex zones made up of anastomosing slip surfaces,
gouge, cataclasite, and breccia (fault core) surrounded by
variably fractured and hydrothermally altered rocks (damage zone) [e.g., Sibson, 1977; Hickman et al., 1995; Caine
et al., 1996]. With increasing pressure and temperature,
brittle faults zones grade downward into ductile shear zones
characterized by mylonitic deformation [e.g., Scholz, 1990].
The very low permeability of fine-grained cataclasites and
mylonites, particularly when foliated, would suggest that
fault zones and mylonites zones act as barriers to fluid flow.
But extensive interaction between fluids and fault zone
rocks is evidenced by hydrothermal alteration, metamorphic
veins, and textures documenting repeated episodes of fracturing and brecciation followed by cementation and crack
6 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
healing [e.g., Moore and Vrolijk, 1992; Hickman et al.,
1995]. Permeability along the subduction plate interface
likely varies during the seismic cycle, with permeability
increasing during periods of slip due to fracturing and
brecciation, and decreasing between slip events due to crack
healing, silica and carbonate cementation, pressure solution,
and retrograde metamorphic reactions [e.g., Hickman et al.,
1995]. An important question is: Integrated over time, is the
subduction plate interface (i.e., the subduction thrust fault
zone at shallow depth and the ductile shear zone at greater
depth) relatively permeable or impermeable compared to the
underlying subducting plate and the overlying plate?
[22] Seismological observations suggest metamorphic
fluids are present where ETS is occurring in the southwest
Japan subduction zone. Beneath southwestern Shikoku,
low-frequency earthquakes associated with ETS occur in a
region of relatively high Vp/Vs, which are interpreted to
reflect relatively high pore pressures in this region [Shelly et
al., 2006]. Similarly, Wang et al. [2006] showed that lowfrequency earthquakes beneath the Kii peninsula occur in a
region of relatively high Poisson’s ratio indicative of the
presence of metamorphic fluids. Beneath southern Vancouver Island, receiver function analysis reveals anomalously
high Poisson’s ratios indicating the pervasive presence of
H2O at high pore pressures in the subducting oceanic crust
trapped beneath a low-permeability plate interface [Audet et
al., 2009].
[23] Are metamorphic fluids present above the plate
interface in the Cascadia fore-arc crust where episodic
tremor is observed? Depending on the permeability structure
of the subduction zone, fluids released from the subducting
plate may (1) remain trapped in the subducting plate,
(2) migrate updip along the plate interface, or (3) infiltrate
the overriding plate. Geologic observations, such as the
serpentine mud volcanoes in the Mariana fore arc [Fryer,
1996] and geophysical observations [Bostock et al., 2002;
Hyndman and Peacock, 2003] provide strong evidence that
fluids derived from the subducting slab infiltrate the overlying fore-arc mantle causing extensive hydration (serpentinization). Less clear is the extent to which the fore-arc crust
nearer the trench is hydrated by fluids released from the
subducting plate. While it seems likely that fluids escaping
from the subducting plate at shallower levels would infiltrate the overlying fore-arc crust, it is possible that the
subduction plate interface is less permeable at shallower
depths as suggested by Audet et al. [2009] and Song et al.
[2009]. Assuming H2O-rich fluids do infiltrate the overriding fore-arc plate, these fluids will tend to react with the
fore-arc crust to produce hydrous mineral assemblages [e.g.,
Yardley and Valley, 1997]. Different rocks have different
capacities to absorb H2O through retrograde hydration
reactions; thus, the extent to which H2O infiltrating the fore
arc will be absorbed by hydration reactions will depend, in
part, on the specific rock types encountered by the fluid.
Because serpentine minerals contain 13 wt % H2O, ultramafic rocks in the fore-arc mantle and crust have the
capacity to absorb large amounts of H2O through serpentinization reactions. Mafic rocks, which compose a significant part of the fore-arc crust, can absorb significant
amounts of H2O through greenschist facies hydration reactions that produce chlorite, amphibole, and other hydrous
minerals. In contrast, felsic rocks have limited capacity to
B00A07
absorb H2O because quartz and feldspar are relatively stable
under fore-arc conditions. Integrated over tens of millions of
years of subduction, enough H2O is liberated from subducted materials to fully hydrate the overlying fore-arc
crust. More commonly, exposures of fore-arc crust reveal
spatially heterogeneous retrograde hydration that is localized along fault zones, fractures, and metamorphic veins.
Free H2O may exist in faults and fractures lined by hydrous
minerals and may be linked to fore-arc ETS, but free H2O is
unlikely to be uniformly present through the fore arc.
[24] Must a free H2O-rich fluid phase be present in order
for ETS to occur? If so, then the gap in ETS beneath eastern
Shikoku and the Kii channel (Figure 2) might reflect a
lack of fluids in this area. The thermal and metamorphic
environment in the subducting Philippine Sea plate in this
area should be intermediate between the Shikoku and Kii
Peninsula models presented in Figures 4 and 5, suggesting
that prograde metamorphic dehydration reactions should be
occurring in this region. Seno and Yamasaki [2003] propose
that the crust subducting in this region is composed of
tonalite (a felsic plutonic rock) and therefore may not be as
hydrated prior to subduction as mafic oceanic crust composed of basalt and gabbro. Alternatively, perhaps deformation of the slab, as evidenced by the contorted slab
contours, elevates permeabilities in this region, such that
metamorphic fluids drain out of the subducting crust rapidly
and fluid pressures do not reach sufficiently high levels. In
the northeast Japan subduction zone, predicted slab P-T
paths are much cooler [Peacock and Wang, 1999] suggesting that the absence of ETS in this region may reflect the
lack of significant metamorphic dehydration reactions in the
subducting slab at shallow depth.
[25] Nadeau and Dolenc [2005] reported ETS along the
San Andreas Fault, which is a very different thermal and
metamorphic setting than a subduction zone. Prograde
metamorphism occurs primarily at convergent plate boundaries where rocks are tectonically buried; metamorphic
fluids are released by dehydration reactions that occur as
hydrous minerals become unstable with increasing pressure
and temperature. In contrast, at a transform plate boundary,
the two plates slip past each other horizontally with no
tectonic burial, so metamorphic dehydration reactions do
not generally occur along transform faults. However, the
San Andreas fault cuts across former subduction zone forearc rocks, which were likely variably hydrated by the earlier
subduction process. Postsubduction heating (thermal relaxation) of the previously hydrated fore-arc rocks could generate
significant volumes of metamorphic fluids, through serpentine and other dehydration reactions [Hyndman and Peacock,
2003]. Alternatively, fluids may be present in the San
Andreas fault zone as a result of downward circulation of
meteoric fluids or upward migration of fluids released from
deeper magmas or the mantle [Hickman et al., 1995].
7. Conclusions
[26] Thermal and petrologic models of the Cascadia and
southwest Japan subduction zones suggest that prograde
metamorphic reactions are likely taking place in the subducting crust in both regions, but the range in calculated
temperatures suggest episodic tremor and slip does not
coincide with a specific temperature or dehydration reac-
7 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
tion. Several lines of evidence indicate a free H2O-rich fluid
phase is present where ETS occurs in subducting oceanic
crust, where prograde metamorphic dehydration reactions
occur. It is less likely that large amounts of fluids are
pervasively present in the overlying fore-arc crust where
tremor has been observed beneath Vancouver Island. However, free H2O may exist locally in faults and fractures lined
by hydrous minerals, and fore-arc ETS may be related to
such structures.
[27] Acknowledgments. I thank the organizers of the Cascadia ETS
workshop held in Sidney, British Columbia in February 2008 for stimulating this research. I thank Susan Ellis and two anonymous reviewers for
constructive comments.
References
Audet, P., M. G. Bostock, N. I. Christensen, and S. M. Peacock (2009),
Seismic evidence for overpressured subducted oceanic crust and megathrust fault sealing, Nature, 457, 76 – 78, doi:10.1038/nature07650.
Baba, T., Y. Tanioka, P. R. Cummins, and K. Uhira (2002), The slip distribution of the 1946 Nankai earthquake estimated from tsunami inversion using a new plate model, Phys. Earth Planet. Inter., 132, 59 – 73,
doi:10.1016/S0031-9201(02)00044-4.
Bostock, M. G., R. D. Hyndman, S. Rondenay, and S. M. Peacock (2002),
An inverted continental Moho and the serpentinization of the forearc
mantle, Nature, 417, 536 – 538, doi:10.1038/417536a.
Caine, J. S., J. P. Evans, and C. B. Forster (1996), Fault zone architecture
and permeability structure, Geology, 24, 1025 – 1028, doi:10.1130/00917613(1996)024<1025:FZAAPS>2.3.CO;2.
Chamot-Rooke, N., V. Renard, and X. Le Pichon (1987), Magnetic anomalies in the Shikoku Basin: A new interpretation, Earth Planet. Sci. Lett.,
83, 214 – 228, doi:10.1016/0012-821X(87)90067-7.
Davis, E., and H. Elderfield (Eds.) (2003), Hydrogeology of the Oceanic
Lithosphere, 706 pp., Cambridge Univ. Press, Cambridge, U. K.
DeMets, C., R. G. Gordon, D. F. Argus, and S. Stein (1994), Effect of
recent revisions to the geomagnetic reversal timescale on estimates of
current plate motions, Geophys. Res. Lett., 21, 2191 – 2194, doi:10.1029/
94GL02118.
Dragert, H., K. Wang, and T. S. James (2001), A silent slip event on the
deeper Cascadia subduction interface, Science, 292, 1525 – 1528,
doi:10.1126/science.1060152.
Fryer, P. (1996), Evolution of the Mariana convergent plate margin system,
Rev. Geophys., 34, 89 – 125, doi:10.1029/95RG03476.
Furukawa, Y. (1993), Depth of the decoupling plate interface and thermal
structure under arcs, J. Geophys. Res., 98, 20,005 – 20,013, doi:10.1029/
93JB02020.
Furukawa, Y. (2009), Convergence of aqueous fluid at the corner of the
mantle wedge: Implications for a generation mechanism of deep lowfrequency earthquakes, Tectonophysics, 469, 85 – 92, doi:10.1016/j.tecto.
2009.01.027.
Fyfe, W. S., N. J. Price, and A. B. Thompson (1978), Fluids in the Earth’s
Crust, 383 pp., Elsevier, Amsterdam.
Hacker, B. R., G. A. Abers, and S. M. Peacock (2003a), Subduction factory: 1. Theoretical mineralogy, densities, seismic wave speeds, and H2O
contents, J. Geophys. Res., 108(B1), 2029, doi:10.1029/2001JB001127.
Hacker, B. R., S. M. Peacock, G. A. Abers, and S. D. Holloway (2003b),
Subduction factory: 2. Are intermediate-depth earthquakes in subducting
slabs linked to metamorphic dehydration reactions?, J. Geophys. Res.,
108(B1), 2030, doi:10.1029/2001JB001129.
Hibbard, J. P., and D. E. Karig (1990), Structural and magmatic responses
to spreading ridge subduction—An example from southwest Japan,
Tectonics, 9, 207 – 230, doi:10.1029/TC009i002p00207.
Hickman, S., R. Sibson, and R. Bruhn (1995), Introduction to special section: Mechanical involvement of fluids in faulting, J. Geophys. Res., 100,
12,831 – 12,840, doi:10.1029/95JB01121.
Hirahara, K. (1981), Three-dimensional seismic structure beneath southwest Japan: The subducting Philippine Sea plate, Tectonophysics, 79,
1 – 44, doi:10.1016/0040-1951(81)90231-6.
Hirose, F., J. Nakajima, and A. Hasegawa (2008), Three-dimensional seismic velocity structure and configuration of the Philippine Sea slab in
southwestern Japan estimated by double-difference tomography, J. Geophys. Res., 113, B09315, doi:10.1029/2007JB005274.
Hyndman, R. D., and S. M. Peacock (2003), Serpentinization of the forearc
mantle, Earth Planet. Sci. Lett., 212, 417 – 432, doi:10.1016/S0012821X(03)00263-2.
B00A07
Hyndman, R. D., and K. Wang (1993), Thermal constraints on the zone of
major thrust earthquake failure: The Cascadia subduction zone, J. Geophys. Res., 98, 2039 – 2060, doi:10.1029/92JB02279.
Hyndman, R. D., and K. Wang (1995), The rupture zone of Cascadia great
earthquakes from current deformation and the thermal regime, J. Geophys. Res., 100, 22,133 – 22,154, doi:10.1029/95JB01970.
Hyndman, R. D., K. Wang, and M. Yamano (1995), Thermal constraints on
the seismogenic portion of the southwestern Japan subduction thrust,
J. Geophys. Res., 100, 15,373 – 15,392, doi:10.1029/95JB00153.
Ide, S., D. R. Shelly, and G. C. Beroza (2007), Mechanism of deep low
frequency earthquakes: Further evidence that deep non-volcanic tremor is
generated by shear slip on the plate interface, Geophys. Res. Lett., 34,
L03308, doi:10.1029/2006GL028890.
Kao, H., S.-J. Shan, H. Dragert, G. Rogers, J. F. Cassidy, and
K. Ramachandran (2005), A wide depth distribution of seismic tremors
along the northern Cascadia margin, Nature, 436, 841 – 844, doi:10.1038/
nature03903.
Kao, H., S.-J. Shan, G. Rogers, and H. Dragert (2007), Migration characteristics of seismic tremors in the northern Cascadia margin, Geophys.
Res. Lett., 34, L03304, doi:10.1029/2006GL028430.
Katsumata, A., and N. Kamaya (2003), Low-frequency continuous tremor
around the Moho discontinuity away from volcanoes in the southwest
Japan, Geophys. Res. Lett., 30(1), 1020, doi:10.1029/2002GL015981.
Kirby, S. H., E. R. Engdahl, and R. Denlinger (1996), Intraslab earthquakes
and arc volcanism: Dual physical expressions of crustal and uppermost
mantle metamorphism in subducting slabs, in Subduction: Top to Bottom,
Geophys. Monogr. Ser, vol. 96, edited by G. Bebout et al., pp. 195 – 214,
AGU, Washington, D. C.
Kodaira, S., T. Iidaka, A. Kato, J.-O. Park, T. Iwasaki, and Y. Kaneda
(2004), High pore fluid pressure may cause silent slip in the Nankai
Trough, Science, 304, 1295 – 1298, doi:10.1126/science.1096535.
Liu, Y., and J. R. Rice (2007), Spontaneous and triggered aseismic deformation transients in a subduction fault model, J. Geophys. Res., 112,
B09404, doi:10.1029/2007JB004930.
Moore, J. C., and P. Vrolijk (1992), Fluids in accretionary prisms, Rev.
Geophys., 30, 113 – 135, doi:10.1029/92RG00201.
Nadeau, R. M., and D. Dolenc (2005), Nonvolcanic tremors deep beneath
the San Andreas Fault, Science, 307, 389, doi:10.1126/science.1107142.
Nakajima, J., and A. Hasegawa (2007), Subduction of the Philippine Sea
plate beneath southwestern Japan: Slab geometry and its relation to arc
magmatism, J. Geophys. Res., 112, B08306, doi:10.1029/2006JB004770.
Obara, K. (2002), Nonvolcanic deep tremor associated with subduction in
southwest Japan, Science, 296, 1679 – 1681, doi:10.1126/science.
1070378.
Obara, K., and H. Hirose (2006), Non-volcanic deep low-frequency tremors
accompanying slow slip in the southwest Japan subduction zone, Tectonophysics, 417, 33 – 51, doi:10.1016/j.tecto.2005.04.013.
Obara, K., H. Hirose, F. Yamamizu, and K. Kasahara (2004), Episodic slow
slip events accompanied by non-volcanic tremors in southwest Japan
subduction zone, Geophys. Res. Lett., 31, L23602, doi:10.1029/
2004GL020848.
O’Neill, C., M. A. Jellinek, and A. Lenardic (2007), Conditions for the
onset of plate tectonics on terrestrial planets and moons, Earth Planet.
Sci. Lett., 261, 20 – 32, doi:10.1016/j.epsl.2007.05.038.
Peacock, S. M. (1996), Thermal and petrologic structure of subduction
zones, in Subduction: Top to Bottom, Geophys. Monogr. Ser, vol. 96,
edited by G. Bebout et al., pp. 119 – 133, AGU, Washington, D. C.
Peacock, S. M. (2001), Are the lower planes of double seismic zones
caused by serpentine dehydration in subducting oceanic mantle?, Geology,
29, 299 – 302, doi:10.1130/0091-7613(2001)029<0299:ATLPOD>2.0.
CO;2.
Peacock, S. M. (2003), Thermal structure and metamorphic evolution of
subducting slabs, in Inside the Subduction Factory, Geophys. Monogr.
Ser, vol. 138, edited by J. M. Eiler, pp. 7 – 22, AGU, Washington, D. C.
Peacock, S. M. (2004), Insight into the hydrogeology and alteration of
oceanic lithosphere based on subduction zones and arc volcanism, in
Hydrogeology of the Oceanic Lithosphere, edited by E. E. Davis and
H. Elderfield, pp. 659 – 676, Cambridge Univ. Press, Cambridge, U. K.
Peacock, S. M., and K. Wang (1999), Seismic consequences of warm
versus cool subduction zone metamorphism: Examples from northeast
and southwest Japan, Science, 286, 937 – 939, doi:10.1126/science.
286.5441.937.
Ranero, C. R., J. Phipps Morgan, K. McIntosh, and C. Reichert (2003),
Bending, faulting, and mantle serpentinization at the Middle America
trench, Nature, 425, 367 – 373, doi:10.1038/nature01961.
Regenauer-Lieb, K., D. A. Yuen, and J. Branlund (2001), The initiation of
subduction: Criticality by addition of water?, Science, 294, 578 – 580,
doi:10.1126/science.1063891.
Rogers, G. C. (1998), Earthquakes and earthquake hazards in the Vancouver area, Bull. Geol. Surv. Can., 525, 17 – 25.
8 of 9
B00A07
PEACOCK: METAMORPHISM AND SUBDUCTION ZONE EPISODIC TREMOR AND SLIP
Rogers, G., and H. Dragert (2003), Episodic tremor and slip on the Cascadia subduction zone: The chatter of silent slip, Science, 300, 1942 – 1943,
doi:10.1126/science.1084783.
Scholz, C. H. (1990), The Mechanics of Earthquakes and Faulting, 439 pp.,
Cambridge Univ. Press, Cambridge, U. K.
Schwartz, S. Y., and J. M. Rokosky (2007), Slow slip events and seismic
tremor at circum-Pacific subduction zones, Rev. Geophys., 45, RG3004,
doi:10.1029/2006RG000208.
Sella, G. F., T. H. Dixon, and A. Mao (2002), REVEL: A model for Recent
plate velocities from space geodesy, J. Geophys. Res., 107(B4), 2081,
doi:10.1029/2000JB000033.
Seno, T., and T. Yamasaki (2003), Low-frequency tremors, intraslab and
interplate earthquakes in southwest Japan—From a viewpoint of slab
dehydration, Geophys. Res. Lett., 30(22), 2171, doi:10.1029/
2003GL018349.
Shelly, D. R., G. C. Beroza, S. Ide, and S. Nakamula (2006), Lowfrequency earthquakes in Shikoku, Japan, and their relationship to episodic tremor and slip, Nature, 442, 188 – 191, doi:10.1038/nature04931.
Sibson, R. H. (1977), Fault rocks and fault mechanisms, J. Geol. Soc.
London, 133, 191 – 231, doi:10.1144/gsjgs.133.3.0191.
Song, T. A., D. V. Helmberger, M. R. Brudzinski, R. W. Clayton, P. Davis,
X. Pérez-Campos, and S. K. Singh (2009), Subducting slab ultra-slow
velocity layer coincident with silent earthquakes in southern Mexico,
Science, 324, 502 – 506.
B00A07
Stein, C. A., and S. Stein (1992), A model for the global variation in
oceanic depth and heat flow with lithospheric age, Nature, 359, 123 –
129, doi:10.1038/359123a0.
Wang, Z., D. Zhao, O. P. Mishra, and A. Yamada (2006), Structural heterogeneity and its implications for the low frequency tremors in southwest
Japan, Earth Planet. Sci. Lett., 251, 66 – 78, doi:10.1016/j.epsl.
2006.08.025.
Wiens, D. A., J. A. Conder, and U. H. Faul (2008), The seismic structure
and dynamics of the mantle wedge, Annu. Rev. Earth Planet. Sci., 36,
421 – 455, doi:10.1146/annurev.earth.33.092203.122633.
Wilson, D. S. (1993), Confidence intervals for motion and deformation of
the Juan de Fuca plate, J. Geophys. Res., 98, 16,053 – 16,071,
doi:10.1029/93JB01227.
Yardley, B. W. D., and J. W. Valley (1997), The petrologic case for a dry
lower crust, J. Geophys. Res., 102, 12,173 – 12,185, doi:10.1029/
97JB00508.
Yoshioka, S., M. Toda, and J. Nakajima (2008), Regionality of deep lowfrequency earthquakes associated with subduction of the Philippine Sea
plate along the Nankai Trough, southwest Japan, Earth Planet. Sci. Lett.,
272, 189 – 198, doi:10.1016/j.epsl.2008.04.039.
S. M. Peacock, Department of Earth and Ocean Sciences, University of
British Columbia, Vancouver, BC V6T 1Z4, Canada. (simon.peacock@
science.ubc.ca)
9 of 9