This article was downloaded by: [Seoul National University]On: 21 April 2015, At: 22:16Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

International Geology ReviewPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tigr20

Deformation microstructures of olivine and chlorite inchlorite peridotites from Almklovdalen in the WesternGneiss Region, southwest Norway, and implications forseismic anisotropyDohyun Kima & Haemyeong Junga

a School of Earth and Environmental Sciences, Tectonophysics Laboratory, Seoul NationalUniversity, Seoul, Republic of KoreaPublished online: 01 Aug 2014.

To cite this article: Dohyun Kim & Haemyeong Jung (2015) Deformation microstructures of olivine and chlorite in chloriteperidotites from Almklovdalen in the Western Gneiss Region, southwest Norway, and implications for seismic anisotropy,International Geology Review, 57:5-8, 650-668, DOI: 10.1080/00206814.2014.936054

To link to this article: http://dx.doi.org/10.1080/00206814.2014.936054

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Deformation microstructures of olivine and chlorite in chlorite peridotites from Almklovdalen inthe Western Gneiss Region, southwest Norway, and implications for seismic anisotropy

Dohyun Kim and Haemyeong Jung*

School of Earth and Environmental Sciences, Tectonophysics Laboratory, Seoul National University, Seoul, Republic of Korea

(Received 20 February 2014; accepted 14 June 2014)

Chlorite peridotites from Almklovdalen in southwest Norway were studied to understand the deformation processes andseismic anisotropy in the upper mantle. The lattice preferred orientation (LPO) of olivine and chlorite was determined usingelectron backscattered diffraction (EBSD)/scanning electron microscopy. A sample with abundant garnet showed [100] axesof olivine aligned sub-parallel to lineation, and [010] axes aligned subnormal to foliation: A-type LPO. Samples rich inchlorite showed different olivine LPOs. Two samples showed [001] axes aligned sub-parallel to lineation, and [010] axesaligned subnormal to foliation: B-type LPO. Two other samples showed [100] axes aligned sub-parallel to lineation, and[001] axes aligned subnormal to foliation: E-type LPO. Chlorite showed a strong LPO characterized by [001] axes alignedsubnormal to foliation with a weak girdle subnormal to lineation. Fourier transform infrared (FTIR) spectroscopy of thespecimens revealed that the olivines with A-type LPO contain a small amount (170 ppm H/Si) of water. In contrast, theolivines with B-type LPOs contain a large amount (340 ppm H/Si) of water.The seismic anisotropy of the olivine and chlorite was calculated. Olivine showed Vp anisotropy of up to 3.8% and a

maximum Vs anisotropy of up to 2.7%. However, the chlorite showed a much stronger Vp anisotropy, up to 21.1%, and amaximum Vs anisotropy of up to 31.7%. A sample with a mixture of 25% of olivine and 75% of chlorite can produce a Vpanisotropy of 14.2% and a maximum Vs anisotropy of 22.9%. Because chlorite has a wide stability field at high pressureand high temperature in the subduction zone, the strong LPO of chlorite can be a source of the observed trench-normal ortrench-parallel seismic anisotropy in the mantle wedge as well as in subducting slabs depending on the dipping angle of slabin a subduction zone where chlorite is stable.

Keywords: chlorite; lattice preferred orientation; petrofabrics; seismic anisotropy

1. Introduction

Since olivine is the most dominant mineral in the uppermantle, the lattice preferred orientation (LPO) of olivine isuseful for studying mantle flow and the seismic anisotropyof the upper mantle (Nicolas and Christensen 1987; Karatoet al. 2008; Long and Silver 2008). Therefore, there havebeen numerous studies on the LPO of olivine. In the drycondition, A-type olivine LPO, which is characterized bythe [100] axes aligned sub-parallel to the shear directionand the [010] axes aligned subnormal to the shear plane,was found (Jung and Karato 2001a) and frequently foundin the upper mantle (Nicolas and Christensen 1987; BenIsmaïl and Mainprice 1998; Jung et al. 2009a).Deformation experiments at high pressures and high tem-peratures have revealed that A-, B-, C-, D-, and E-typeLPOs of olivine exist in other deformation conditions aswell (Bystricky et al. 2000; Jung and Karato 2001a;Holtzman et al. 2003; Couvy et al. 2004; Katayama et al.2004; Jung et al. 2006; Karato et al. 2008; Jung et al.2009b; Ohuchi et al. 2011, 2012). In the wet condition, B-,C-, or E-type olivine LPOs were found (Jung and Karato2001a; Katayama et al. 2004; Jung et al. 2006; Katayamaand Karato 2006). The correlation between olivine LPOs

and water content has been established (Jung et al. 2006).The B-type LPO is characterized by [001] axes alignedsub-parallel to the shear direction, and [010] axes alignedsubnormal to the shear plane. The C-type LPO is charac-terized by [001] axes aligned sub-parallel to the sheardirection, and [100] axes aligned subnormal to the shearplane. The E-type LPO is characterized by [100] axesaligned sub-parallel to the shear direction, and [001] axesaligned subnormal to the shear plane.

Water-related olivine fabrics have also been reported inseveral natural peridotites and mantle xenoliths. Mizukamiet al. (2004) found the B-type olivine LPO in garnetperidotites from the Higashi-akaishi peridotites body ofsouthwest Japan. Katayama et al. (2005) found theC-type olivine LPO in garnet peridotites from OtrøyIsland, in the Western Gneiss Region in Norway. Skemeret al. (2006) found the B-type olivine LPO in garnetperidotites from Cima di Gagnone, Switzerland.Michibayashi et al. (2007) and Park and Jung (2015)found B-type olivine LPOs in the southern Mariana trenchand in mantle xenoliths from Shanwang, eastern China,respectively. Webber et al. (2008) found the B-type olivineLPO in mantle peridotites from Red Hills, New Zealand.

*Corresponding author. Email: hjung@snu.ac.kr

International Geology Review, 2015Vol. 57, Nos. 5–8, 650–668, http://dx.doi.org/10.1080/00206814.2014.936054

© 2014 Taylor & Francis

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

Jung (2009) and Jung et al. (2014) found both B- and E-type olivine LPOs in peridotites from Val Malenco in Italyand from Bergen Arc in southwest Norway, respectively.Jung et al. (2013) also reported the C-type olivine LPO ingarnet peridotites in the North Qaidam ultrahigh-pressure(UHP) belt in Northwest China. Michibayashi and Oohara(2013) found both C- and E-type olivine LPOs in dunitesin the shear zone from the Fizh massif, Oman Ophiolite.Park et al. (2014) found, under hydrous conditions, theC-type olivine LPO in mantle xenoliths from AdamsDiggings in the Rio Grande Rift, USA.

Trench-parallel seismic anisotropy has been observedin the fore-arc mantle wedge in many subduction zones(Russo and Silver 1994; Savage 1999; Park and Levin2002; Long and Silver 2008; Long 2013). The polariza-tion direction of the fast S-wave is often trench-normal inthe back-arc area, but changes to trench-parallel in thefore-arc area (Smith et al. 2001; Huang et al. 2011a,2011b). This anisotropy can be attributed to the B-typeolivine LPO (Jung and Karato 2001a; Kneller et al.2007). Wang and Zhao (2013) also reported trench-par-allel seismic anisotropy above the mantle wedge, as wellas in the slab of Tohoku and Kyushu, Japan, which wasinterpreted to be caused by the fabric transition to B-typeolivine LPO in the presence of water. Recently,McCormack et al. (2013) found a thin layer (~6 km) ofstrong anisotropy (~10–14%) beneath Ryukyu with aslow axis of symmetry directly above the slab, whichcannot be explained by the olivine LPO. Wagner et al.(2013) also found trench-parallel seismic anisotropyusing Rayleigh waves beneath Cascadia, and suggestedthat hydrous minerals in the mantle wedge or in the slabmay also be responsible for trench-parallel seismic ani-sotropy. Typical anisotropic hydrous minerals are serpen-tine and chlorite, which are formed by dehydration of thesubducting slab. The seismic anisotropy of serpentine hasbeen previously reported, revealing that serpentine can beresponsible for trench-parallel seismic anisotropy if theserpentine is stable at depth (Katayama et al. 2009; Jung2011; Nishii et al. 2011; Watanabe et al. 2011). Thestability field of serpentine and chlorite at high pressureand high temperature is known (Ulmer and Trommsdorff1995; Hacker et al. 2003; Fumagalli and Poli 2004;Padrón-Navarta et al. 2010; Van Keken et al. 2011).Chlorite is known to be stable in a wide range of pres-sures and temperatures in subduction zones (Fumagalliand Poli 2004; Van Keken et al. 2011; Wada et al. 2012).However, the LPO type and seismic characteristics ofchlorite have not been well studied.

In this paper, we studied the LPO of chlorite andolivine in chlorite peridotites from Almklovdalen, south-west Norway, and found that chlorite has a strong LPOand can produce significant seismic anisotropy in themantle wedge and in the slab at high pressure and hightemperature.

2. Geological setting and sample characteristics

The Western Gneiss Region (WGR) in Norway experiencedcollision with Baltica and Laurentia and exhumation tocrustal levels at 425–380 Ma, an event called theScandian Orogeny (Lapen et al. 2009; Hacker et al. 2010;Wang et al. 2013). This collision triggered UHP meta-morphism at 800°C and 3.6 GPa (Hacker et al. 2010).The WGR is the lowest structure unit in the ScandinavianCaledonides (Lapen et al. 2009). The major domain of theWGR is biotite, hornblende, and garnet containing gneisseswith eclogites and peridotites (Hacker et al. 2010; Wanget al. 2013). Almklovdalen, located in the WGR of Norway(Figure 1), has some garnet peridotite bodies that camefrom Archaean mantle fragments embedded in theProterozoic Baltic continental crust (Lapen et al. 2009;Hacker et al. 2010; Wang et al. 2013). Almklovdalen hasalso some peridotites with high chlorite content, which maybe attributed to hydrated garnet.

We collected six peridotites from Almklovdalen. Themajor minerals in the samples consist of olivine (60–83%)and chlorite (2–35%), and the minor minerals are ortho-pyroxene (1–7%) and clinopyroxene (2–13%). The com-positional layering of the garnet/pyroxene, olivine, andchlorite is well developed (Figure 2(a) and (b)), and grainsare mostly elongated (Figure 3). Garnets are generallyround with a large grain size (~1 cm), and chlorites tendto be aligned to form a sheet structure (Figures 2(b) and3(b), 3(c)). One sample (434) is especially abundant ingarnet (19%) compared to the other chlorite-rich samples(Table 1). Figure 3(a) shows a large garnet grain withelongated olivines (sample 434). Three samples (435,436, and 438) contain more chlorite (31–35%) than theother samples (Table 1).

3. Methods

3.1. Measurement of LPO and calculation of seismicanisotropy

The foliation of the samples was determined by composi-tional layering of chlorite, garnet/pyroxene, and olivine.The lineation of the samples was determined by theshape-preferred orientation of minerals in the foliationplane, using the projection-function method (Panozzo1984). Samples were cut, and thin sections were made onthe x–z plane (x: lineation, z: normal to foliation). The LPOof samples was determined using electron backscattereddiffraction (EBSD) in a JEOL JSM-6380 scanning electronmicroscope (SEM) housed at the School of Earth andEnvironmental Sciences (SEES) at Seoul NationalUniversity (SNU). The accelerating voltage was 20 kV,and the working distance was 15 mm in the SEM. AnHKL EBSD system with channel 5 software was used forthe EBSD analysis. The EBSD patterns of every grain wereanalysed manually to ensure correct indexing, and the fabric

International Geology Review 651

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

strength of the LPO was estimated using the misorientationindex (M-index) (Skemer et al. 2005). The M-index rangesfrom 0 (random fabric) to 1 (single crystal).

The seismic anisotropy of these peridotites was calcu-lated from the LPO data of olivine and chlorite. Minorminerals such as orthopyroxene and clinopyroxene wereignored. The elastic constant of olivine came fromAbramson et al. (1997), and that of chlorite came fromAleksandrov and Ryzhova (1961). The newest elastic con-stant of chlorite at P = 1.8 GPa (Mookherjee andMainprice 2014) was also used to compare the resultantseismic anisotropy. The ANIS2k and VpG programs(Mainprice 1990) were used to calculate seismic wavevelocity and seismic anisotropy.

3.2. Measurement of water content in olivine

Fourier transform infrared (FTIR) spectroscopy was usedto measure water content in the olivine samples. A Nicolet6700 FTIR spectrometer with a continuum IR microscopehoused at the Tectonophysics laboratory at the SEES atSNU was used to collect unpolarized FTIR spectra usingthe transmitted light. Samples for FTIR analysis wereboth-side-polished to a thickness of 100–350 μm (Table1), depending on the grain size of the specimen.Additionally, they were heated to T = 120°C for24 hours to eliminate water from the surface. The samplechamber was purged with N2 gas to exclude atmosphericmoisture during the IR measurements. The aperture sizewas 100 × 100 μm, and only one sample (sample 436)used an aperture size of 50 × 50 μm. The IR beam was

penetrated through a clean area of olivine without anycracks, inclusions, or grain boundaries. The water contentof 10 different grains was measured and averaged. Thewater content of the olivine was calculated from infraredabsorptions in the wavenumbers 3400–3620 cm−1, therange of the O–H stretching vibrations. The calibrationof Paterson (1982) was used, following the former studiesof olivine LPOs in the wet condition (Jung and Karato2001a; Katayama et al. 2004; Jung et al. 2006; Katayamaand Karato 2006).

3.3. Dislocation microstructure of olivine

An oxygen decoration technique (Kohlstedt et al. 1976)was used to observe the dislocation microstructures of theolivine specimens. Samples were heated to T = 800°C forone hour to oxidize the specimens. After that, each samplewas polished with Syton (colloidal silica) to remove a thinoxide layer from the surface. Prepared samples were car-bon-coated to be used in the JEOL JSM-6380 SEM.Backscattered electron images (BEI) were taken with anacceleration voltage of 15 kV and a working distance of10 mm to observe the dislocation microstructures of theolivine specimens in the SEM (Karato 1987; Jung andKarato 2001b).

3.4. EPMA analysis and P–T estimation of specimen

A JEOL JXA-8900R electron probe micro analyser (EPMA)at the National Centre for Inter-university Research Facilities(NCIRF) at SNU was used to analyse the chemical

Figure 1. Location of study area, Almklovdalen, southwest Norway, with pressure and temperature contours (modified after Wang et al.(2013)).

652 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

composition of the minerals. The accelerating voltage was15 kV, the working distance was 11 mm, and the beam sizewas 5 μm. Samples with abundant garnet (434) and chlorite(438) were selected, and two garnet grains from sample 434and two chlorite grains from sample 438 were analysed.Three olivine grains from each sample were also analysed.The grains were observed at the core and rim areas, and theresults were averaged to represent each grain. The results ofthe EPMA analysis were recalculated using Minpet 2.02,written by Dr Yong-Joo Jwa, Gyeongsang NationalUniversity, Republic of Korea.

For the P–T calculations, the oxide mass percentagewas used. The oxides came from the olivine, clinopyrox-ene, and garnet grains of samples 434, 435, 436, and 439.The ptex13 program written by Dr Andrei Girnis, GoetheUniversity, Frankfurt, Germany, was used for the P–Tcalculation.

4. Results

4.1. Mineral chemistry and P–T estimation ofspecimens

The chemical compositions of major minerals from EPMAanalysis are shown in Table 2. The chemical compositionof each mineral did not differ much by grain and sample.

Figure 3. Optical photomicrographs in cross-polarized light. (a)Sample 434 showing relatively large garnet grain, with elongatedsmall olivines. (b) Sample 436 showing notable chlorite lenseselongated sub-parallel to the lineation. (c) Sample 438 withabundant chlorite grains elongated. (Ol, olivine; Cpx, clinopyr-oxene; Chl, chlorite; and Grt, garnet).

Figure 2. Outcrops of Almklovdalen peridotites. (a) Compositionallayering of olivine-rich and pyroxene/garnet-rich peridotite.Olivine is shown as a yellowish brown colour after oxidation onthe surface. Hammer is 40 cm long. (b) Chlorite-rich peridotitewith elongated lenses of chlorite. Coin has a diameter of22.86 mm. Ol, olivine; Chl, chlorite.

International Geology Review 653

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

The Mg#, 100*Mg/(Mg + Fe), of the olivines was 86.32–88.27, indicating that the olivines are close to forsterite(Fo86–Fo88). The garnets had high Al and Mg contents,and the averaged garnet composition wasAlm23Adr2Grs6Prp64Sps1Uv4 (Alm, almandine; Adr,andradite; Grs, grossular; Prp, pyrope; Sps, spessartine;Uv, uvarovite). The chlorites contained Mg and Fe2+,possibly close to clinochlore.

The P–T conditions were calculated from averagedcore and rim EPMA data of olivine, clinopyroxene, andgarnet. The pressure was calculated using the method ofclinopyroxene barometer for garnet peridotites (Nimis andTaylor 2000). The temperature was calculated using themethod of Fe–Mg exchange between clinopyroxene andgarnet (Berman et al. 1995; Nakamura and Hirajima2005), and Ca partitioning between olivine and clinopyr-oxene (Köhler and Brey 1990). The estimated pressurewas 2.2 GPa. The temperature for the A-type olivineLPO (sample 434) was 689°C using the Berman et al.(1995) method, and 769°C using the Nakamura andHirajima (2005) method. The temperature for the B-typeolivine LPO (samples 435 and 436) was 741°C using theKöhler and Brey (1990) method. The temperature for theC-type olivine LPO (sample 439) was 908°C using theKöhler and Brey (1990) method.

4.2. LPO of minerals

4.2.1. LPO of olivine

The LPOs of olivine are plotted in pole figures (Figure 4).Six peridotite samples showed various olivine LPOs.Sample 434, with abundant garnet, showed [100] olivineaxes aligned sub-parallel to the lineation and [010] axesaligned subnormal to the foliation, close to A-type LPO.Two samples (435 and 436) showed [001] axes alignedsub-parallel to the lineation and [010] axes aligned sub-normal to the foliation, consistent with B-type LPO (Jungand Karato 2001a). Sample 437 showed [100] axesaligned sub-parallel to the lineation and [001] axes alignedsubnormal to the foliation, consistent with E-type LPO(Katayama et al. 2004). Sample 438 showed [100] axesaligned sub-parallel to the lineation and [010] and [001]axes aligned subnormal to the foliation. Sample 439showed [001] axes aligned sub-parallel to the lineationand [100] axes aligned subnormal to the foliation, closeto C-type LPO (Jung et al. 2006). The fabric strength wasrepresented as a misorientation index (M-index) (Skemeret al. 2005). The M-index of the olivine ranged from 0.035to 0.069, which is relatively weak (Table 1).

4.2.3. LPO of chlorite

The LPOs of chlorite are plotted in pole figures (Figure 5).Three samples (435, 436, and 438) had enough chloriteT

able

1.Sam

pledescriptionandresults.

Sam

ple

Mod

alcompo

sitio

nLPO

ofOl

M-ind

ex1

IRsample

thickn

ess(μm)

Water

contentof

Ol2

(ppm

H/Si)

Recrystallized

grainsize

ofOl3(μm)

Stress4

(MPa)

Ol(%

)Opx

(%)

Cpx

(%)

Spl

(%)

Chl

(%)

Grt(%

)Ol

Chl

434

60.3

6.5

12.6

0.5

1.6

18.5

A0.04

9–

130

170±30

496

17±15

435

58.3

1.5

4.9

0.5

34.4

0.4

B0.04

50.08

310

031

0±30

481

39±15

436

60.4

4.6

2.7

0.6

30.5

1.1

B0.03

50.09

923

021

0±30

389

47±15

437

81.9

2.7

5.4

3.0

5.1

1.9

E0.05

6–

350

180±30

314

50±15

438

61.8

0.6

1.7

0.4

35.3

0.2

~E0.06

90.08

216

017

0±30

328

50±15

439

83.0

1.7

4.8

5.7

2.1

2.7

C0.06

0–

100

–46

541

±15

Notes:1M-index

representsfabric

strength

(Skemer

etal.20

05).

2The

water

contentsof

10differentgrains

weremeasuredandaveraged.The

calib

ratio

nof

Paterson(198

2)was

used.The

water

contentof

sample439was

omitted

becauseof

itspoor

resolutio

nof

theIR

spectrum

.3Recrystallized

grainsize

was

measuredusingthelin

earinterceptmethod(G

ifkins

1970).

4Stresswas

estim

ated

from

therecrystallizedgrainsize

ofolivine(JungandKarato20

01b).

Ol,olivine;

Opx,orthopyroxene;

Cpx,clinopyroxene;

Spl,spinel;Chl,chlorite;andGrt,garnet.

654 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

content to plot. Two samples (435 and 436) showed [001]axes strongly aligned subnormal to the foliation, with aweak girdle subnormal to the lineation, and both [100]axes and poles of (010) and (110) are aligned sub-parallelto the lineation with a weak girdle along the foliation.Sample 438 showed [001] axes weakly distributed as agirdle subnormal to the lineation, and both [100] axes andpoles of (010) and (110) aligned sub-parallel to the linea-tion. The M-index of the chlorite ranged from 0.082 to0.099, which is much stronger than that of olivine(Table 1).

4.3. Seismic anisotropy

Seismic anisotropy was calculated for the individualminerals olivine and chlorite, and for the mixture of oli-vine and chlorite (Figures 6 and 7; Table 3). In the case ofolivine-only, the P-wave anisotropy (Vpmax–Vpmin)/Vpmean was weak in the range of Vp = 1.8–3.8% with anaverage of 2.9%, and the maximum shear-wave anisotropywas also weak, AVs = 1.66–2.68%. The maximum P-wavevelocity was 8.41–8.53 km s–1 with an average of

8.47 km s–1 (Figure 6). In the case of chlorite-only,using the elastic constant of chlorite (Aleksandrov andRyzhova 1961), the P-wave anisotropy was high, in therange of 14.9–21.1%, with an average of 18.2%, and themaximum shear-wave anisotropy was very high, up to14.46–31.74%. The maximum P-wave velocity was7.21–7.41 km s–1 with an average of 7.30 km s–1 (Figure7(a)). In the case of the mixture of olivine and chlorite,seismic anisotropy was based on the compositional ratiobetween olivine and chlorite for samples 435, 436, and438. In this case, the P-wave anisotropy was in the rangeof 6.2–8.3%, with an average of 7.5%, and the maximumshear-wave anisotropy was 6.93–13.10% (Figure 7(b)).These results indicate that seismic wave propagation inchlorite is much more anisotropic than that in olivine, andchlorite can provide a significant effect of overall seismicanisotropy. Using the elastic constant of chlorite atP = 1.8 GPa (Mookherjee and Mainprice 2014), seismicanisotropy was also calculated and compared with theresults, which were obtained using the elastic constant ofchlorite (Aleksandrov and Ryzhova 1961). In general, themaximum P-wave velocity of chlorite was increased, but

Table 2. Chemical composition of major minerals.

Sample 434 438

Mineral Grt 1 Grt 2 Ol 1 Ol 2 Ol 3 Chl 1 Chl 2 Ol 1 Ol 2 Ol 3

SiO2 41.50 41.73 40.54 39.46 40.67 31.25 30.98 40.33 40.33 39.90TiO2 0.04 0.08 0.03 0.02 0.01 0.03 0.04 0.01 0.01 0.04Al2O3 23.97 23.98 0.01 0.03 0.00 17.44 17.95 0.00 0.00 0.00Cr2O3 1.49 1.52 0.00 0.00 0.00 1.31 1.59 0.00 0.00 0.00FeO 12.42 12.01 11.68 11.60 11.57 4.12 3.87 13.19 13.19 13.37MnO 0.41 0.41 0.07 0.09 0.10 0.03 0.04 0.20 0.20 0.19MgO 17.76 18.03 48.67 48.32 48.76 31.87 31.63 47.57 47.57 47.34CaO 4.81 4.91 0.00 0.01 0.00 0.03 0.01 0.00 0.00 0.00Na2O 0.03 0.01 0.01 0.01 0.02 0.00 0.01 0.03 0.03 0.00K2O 0.00 0.00 0.01 0.04 0.00 0.01 0.00 0.00 0.00 0.00NiO 0.00 0.00 0.80 0.68 0.80 0.00 0.00 0.55 0.55 0.35

Total 102.41 102.66 101.81 100.24 101.94 86.10 86.13 101.89 101.89 101.18

Si 2.93 2.93 0.99 0.98 0.99 3.00 2.97 0.99 0.99 0.99Al 1.99 1.99 0.00 0.00 0.00 1.97 2.03 0.00 0.00 0.00Ti 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Fe2+ 0.68 0.66 0.24 0.24 0.24 0.33 0.31 0.27 0.27 0.28Fe3+ 0.06 0.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Cr 0.08 0.08 0.00 0.00 0.00 0.10 0.12 0.00 0.00 0.00Mn 0.02 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Mg 1.87 1.89 1.77 1.79 1.77 4.56 4.52 1.74 1.74 1.74Ca 0.36 0.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Na 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00K 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Ni 0.00 0.00 0.02 0.01 0.02 0.00 0.00 0.01 0.01 0.01Cations 8.00 8.00 3.01 3.02 3.01 9.96 9.95 3.01 3.01 3.02O 12.00 12.00 4.00 4.00 4.00 18.00 18.00 4.00 4.00 4.00Mg# 71.82 72.81 88.14 88.13 88.27 93.24 93.58 86.53 86.53 86.32

Note: Garnet-rich (434) and chlorite-rich (438) samples were selected, and each measurement is the average of the core and rim area of each grain.Ol, olivine; Chl, chlorite; and Grt, garnet.

International Geology Review 655

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

seismic anisotropy of P- and S-wave was decreasedslightly using the recent elastic constant of chlorite(Mookherjee and Mainprice 2014), but still preservedhigh anisotropy of chlorite (Table 3). The ratio change ofchlorite in a rock specimen can affect both Vp and Vs1

seismic anisotropy. Considering LPOs of olivine andchlorite (sample 436), Figure 8 shows that increasingchlorite ratio increases both Vp and Vs1 anisotropy.

The polarization direction of the fast S-wave (Vs1) isalso plotted in Figures 6 and 7. In the case of olivine-only

Figure 4. Pole figures of olivine presented in the lower hemisphere using an equal-area projection. The white line (S) indicates thefoliation and the red dot (L) indicates the lineation. N represents the number of grains, and a half-scatter width of 20° was used for thecontours. The red colour represents the high density of data points, and the numbers in the legend correspond to the multiples of uniformdistribution.

656 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

(Figure 6), the A-type olivine LPO (sample 434) showed apolarization direction of the fast S-wave sub-parallel to thelineation for vertically propagating S-waves, the B-typeolivine LPO (samples 435 and 436) showed a polarizationdirection of the fast S-wave subnormal to the lineation, theC-type olivine LPO (sample 439) showed a polarizationdirection subnormal to the lineation, and the E-type oli-vine LPO (samples 437 and 438) showed a polarizationdirection sub-parallel to the lineation. In the case of chlor-ite-only (Figure 7(a)), all three samples (435, 436, and438) showed a polarization direction of the fast S-wavesub-parallel to the lineation for vertically propagating S-waves. It is notable that chlorite (sample 436) producesVs1 polarization anisotropy of up to 31.74%. In the caseof the mixture of olivine and chlorite (Figure 7(b)), it isfound that the polarization direction of the fast S-waveclosely follows that of chlorite-only. The samples (435 and436) with a high Vs1 polarization anisotropy showed atrend of axial anisotropy, where high and low anisotropy isfound close to the horizontal and vertical planes, respec-tively (Figure 7(b)).

4.4. Water content of olivine

Typical FTIR spectra of olivine are shown in Figure 9.Olivine grains showed IR absorption peaks at and near thewavenumbers 3479, 3525, 3572, and 3613 cm−1

(Figure 9), which are related to the stretching vibrationsof O–H bonding (Kohlstedt et al. 1996; Mei and Kohlstedt2000). Two samples (435 and 436) that showed B-type

olivine LPOs had a high water content in the olivine: 210–310 ± 30 ppm H/Si (Table 1). If the calibration of Bellet al. (2003) is used, water content will be four timeshigher (Mosenfelder et al. 2006). These water contentsare relatively high, indicating that these peridotites weredeformed under wet condition. However, sample 434,which showed an A-type olivine LPO, had a low watercontent, approximately 170 ± 30 ppm H/Si (Table 1).

4.5. Dislocation microstructure

Dislocation microstructures of the olivines are shown in theBEIs from the SEM (Figure 10). Dislocations are shown aswhite dots and lines. The sample with A-type olivine LPO(sample 434) showed curved and straight dislocations(Figure 10(a) and (b)). Some dislocations were alignedand formed subgrain boundaries (Figure 10(b)). The samplewith ~E-type olivine LPO showed generally a straightdislocation (Figure 10(c)) and also showed aligned disloca-tions, forming a subgrain boundary (Figure 10(d)).Dislocations of the sample with B-type olivine LPOs (sam-ples 435 and 436) were distributed heterogeneously (Figure10(e) and (f)) and a subgrain boundary was rarely found,which is consistent with a previous experimental studyunder wet conditions (Jung et al. 2006).

4.6. Stress estimation of samples

A recrystallized grain size of olivine can be used to esti-mate the differential stress of a sample (Van der Wal et al.,

Figure 5. Pole figures of chlorite presented in the lower hemisphere using an equal-area projection. The white line (S) indicates thefoliation and the red dot (L) indicates the lineation. N represents the number of grains, and a half-scatter width of 20° was used for thecontours. The red colour represents the high density of data points, and the numbers in the legend correspond to the multiples of uniformdistribution.

International Geology Review 657

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

1993; Karato et al. 1980; Jung and Karato 2001b). Therecrystallized area of olivine was selected on the thinsection of the x–z plane. The recrystallized grain size ofolivine was measured using the linear intercept method

(Gifkins 1970; Russ and Dehoff 2000). The measured 2-Drecrystallized grain size was converted to 3-D recrystal-lized grain size by multiplying a factor of 1.5, and therecrystallized grain size of olivine in each sample was in

Figure 6. Seismic anisotropy and velocity calculated from the LPO of olivine-only, presented in the lower hemisphere using anequal-area projection. The centre of the pole figure (z) is the direction normal to the foliation, and the E–W direction (x) representsthe lineation. The P-wave velocity (Vp), amplitude of S-wave anisotropy (AVs), and the polarization direction of the fast S-wave(Vs1) are plotted.

658 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

the range of 314–496 μm (Table 1). To estimate the stressof samples, the recrystallized grain size piezometer (Jungand Karato 2001b) was used for both dry (A-type olivineLPO) and wet (B- and E-type olivine LPOs) conditions.

The sample with A-type olivine LPO (sample 434)showed a low stress 17 ± 15 MPa and other sampleswith other types of olivine LPOs showed a little higherstress 39–50 ± 15 MPa (Table 1).

Figure 7. Seismic anisotropy and velocity calculated from LPOs of (a) chlorite-only and (b) mixture of olivine and chlorite presented in thelower hemisphere using an equal-area projection. The centre of the pole figure (z) is the direction normal to the foliation, and the E–Wdirection(x) represents the lineation. The P-wave velocity (Vp), amplitude of S-wave anisotropy (AVs), and the polarization direction of the fast S-wave(Vs1) are plotted. The modal compositions of olivine and chlorite in Table 1 were used to calculate the seismic velocity and anisotropy in (b).

International Geology Review 659

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

5. Discussion

Various types of olivine LPOs, such as A-, B-, C-, and E-types, are found in chlorite peridotites from Almklovdalen,southwest Norway. Using FTIR spectroscopy, we foundthat samples with B- and E-type olivine LPOs containwater (170–310 ± 30 ppm H/Si). This water content ofolivine is thought to be the minimum water contentbecause water might have been lost due to the fast hydro-gen diffusion in olivine (Mackwell and Kohlstedt 1990).Our result that both B- and E-type olivine LPOs are foundin a wet olivine is consistent with the result of previousexperimental studies (Jung and Karato 2001a; Katayamaet al. 2004; Jung et al. 2006; Katayama and Karato 2006).Wang et al. (2013) also reported olivine LPOs such as A-,B-, and E-types in Almklovdalen peridotites, with smalleramounts of water in the olivine (10–180 ppm H/Si). Weplotted A- and E-type olivine LPOs in terms of watercontent and stress (Jung et al. 2006) (Figure 11(a)),where our data are located near the fabric boundarybetween A- and E-types. Because the former fabric dia-gram in terms of water content and stress (Jung et al.2006) was based on much higher temperature conditions(1470–1570 K) than the peridotites we studied (962–1181 K), we plotted our data of B- and C-type olivineLPOs in the diagram of B-/C-fabric transition (Katayamaand Karato 2006; Skemer et al. 2006) at lower temperatures(Figure 11(b)), where our data was fitted to the stress-temperature field with the data of previous studies.

The LPO of chlorite has not been studied much. Puelleset al. (2012) reported only one chlorite LPO so far from aspecimen in northwest Spain, which showed [001] axesaligned subnormal to the foliation, which is similar to thechlorite LPO that we found in this study. Two samples (435and 436) showed [001] axes are strongly aligned subnormalto the foliation, with a weak girdle subnormal to the linea-tion and [100] axes aligned sub-parallel to the lineation.Seismic anisotropy of single-crystal chlorite and olivine isshown in Figure 12. Both P- and S-waves propagate slowlyalong the [001] axis of chlorite, but propagate fast along

Table 3. Seismic velocity and anisotropy of olivine, chlorite, and whole rock.

Sample

Max. Vpof Ol

(km s–1)

Vpanisotropyof Ol (%)

Max. AVsanisotropyof Ol (%)

Max. Vpof Chl(km s–1)

Vpanisotropyof Chl (%)

Max. AVsanisotropy of

Chl (%)

Max. Vp ofwhole rock(km s–1)

Vp anisotropyof wholerock (%)

Max. AVsanisotropy ofwhole rock (%)

434 8.43 2.1 1.66 – – – – – –435 8.52 3.8 2.43 7.41 21.1 26.90 7.91 8.0 11.59

(8.27) (15.2) (20.25) (8.26) (5.7) (8.64)436 8.44 2.3 2.13 7.27 18.5 31.74 7.94 8.3 13.10

(8.17) (13.3) (22.82) (8.28) (6.0) (9.53)437 8.46 3.5 2.45 – – – – – –438 8.53 3.8 2.68 7.21 14.9 14.46 7.82 6.2 6.93

(8.10) (10.3) (10.61) (8.20) (4.3) (5.16)439 8.41 1.8 1.77 – – – – – –

Note: The values inside the bracket () are calculated using the elastic constants of chlorite from Mookherjee and Mainprice (2014). Ol, olivine; Chl, chlorite.

Figure 8. The relation between chlorite content and seismic ani-sotropy, considering LPOs of olivine and chlorite (sample 436).

Figure 9. Representative unpolarized FTIR spectra of the olivineshowing each type of LPO. Sample 439 was omitted because of itspoor resolution of the IR spectrum.

660 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

both the [100] and [010] axes. Both P-wave and S-waveanisotropies of single-crystal chlorite are very high, up toVp = 35.5% and AVs = 76.2%, respectively (Figure 12), aspreviously reported (Mainprice and Ildefonse 2009).Therefore, the formation of strong LPO of [001] axes inchlorite (Figure 5) can produce significant seismic aniso-tropy: P-wave anisotropy up to 21.1% and S-wave polar-ization anisotropy up to 31.7% (Figure 7(a)).

Olivine from Almklovdalen peridotites showed that A-and E-type olivine LPOs (samples 434, 437, and 438)have a Vs1 polarization direction sub-parallel to the linea-tion for vertically propagating S-waves (Figure 6). TheB-type olivine LPO (samples 435 and 436) showed aVs1 polarization direction subnormal to the lineation(Figure 6), which may be responsible for trench-parallelseismic anisotropy in the subduction zone (Jung andKarato 2001a). However, the strength of the olivine LPOin this study was low, with an M-index of M = 0.035–0.069 (Table 1), resulting in a low seismic anisotropy ofolivine only up to Vp = 3.8% and AVs = 2.7% (Figure 6).

In contrast, we found that the chlorite had developed astrong LPO, with a high M-index of M = 0.082–0.099(Table 1), which produced a very strong seismic aniso-tropy of up to Vp = 21.1% and AVs = 31.7% (Figure 7(a))for the sample with 34% chlorite. Previous study on theLPO of serpentinite showed that a sample with 40% anti-gorite produced a strong seismic anisotropy of up toVp = 23.6% and AVs = 23.2% (Jung 2011), indicatingthat chlorite produces a similar Vp anisotropy to serpen-tine, but much higher AVs anisotropy than serpentine(antigorite).

Considering application of current chlorite fabric inthe (meta)peridotites to the mantle wedge or the mantlepart of the subducting slab, important factors for devel-oping fabrics of minerals are deformation mechanism anddeformation conditions such as pressure, temperature,stress, water, etc. If a rock is deformed under the samedeformation mechanism and deformation conditions,then the rock fabric developed in one locality can beapplied to the one developed in the other locality.

Figure 10. BEIs of olivine showing dislocation microstructure. (a and b) Curved and straight dislocations in olivine with an A-type LPO(sample 434). (c and d) Straight and aligned dislocations forming a subgrain boundary with an E-type LPO (sample 438). (e and f)Heterogeneous distribution of dislocations in olivine with a B-type LPO (e, sample 436; f, sample 435).

International Geology Review 661

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

Olivine and chlorite fabrics in the present study weredeveloped in the meta-peridotite of HP-UHP meta-morphic locality, the WGR of Norway. As we mentionedin the text, chlorite has a wide stability field at high-pressure and high-temperature conditions (Fumagalliand Poli 2004; Till et al. 2012) so that chlorite peridotitecan be present in the mantle wedge and in the mantle partof the subducting slab (Hacker et al. 2003). Since defor-mation mechanism and deformation conditions of miner-als in the HP-UHP belt are considered to be similar tothose in the part of mantle wedge and subducting slab,we believe that current data of chlorite and olivine fabricscan be applicable to the mantle fabrics.

Considering a 2-D corner flow model in a subductionzone, Figure 13 shows the change of the polarizationdirection of the fast S-wave with a dip angle θ, using thechlorite LPO of sample 436. For the shallow dip angle(θ ≤ 45°), the polarization direction of the fast S-wave(Vs1) is sub-parallel to the flow direction, which can causetrench-normal seismic anisotropy in the subduction zone.On increasing θ, the blue band where Vs1 polarizationanisotropy is high rotates, and the polarization direction ofthe fast S-wave (Vs1) becomes subnormal to the lineation(flow direction). This is very important because this fea-ture can be used to explain a strong trench-parallel seismicanisotropy in subduction zones (Figure 14).

Figure 14 shows a schematic view of the stability fieldof chlorite in the subduction zone based on previousexperimental and modelling studies (Hacker et al. 2003;Fumagalli and Poli 2004; Till et al. 2012) and also illus-trates the signature of seismic anisotropy in a subductionzone that can be caused by the LPO of chlorite. For thelow dipping angle (θ ≤ 45º) of slab in the subduction zone(Figure 14(a)), LPO of chlorite will cause trench-normalseismic anisotropy for the vertically propagating fastS-waves. For example, the Juan de Fuca slab under theCascadia subduction zone is dipping at low angle and thetrench-normal seismic anisotropy observed under Oregon(Currie et al. 2004; Park et al. 2004; Wagner et al. 2013)may be attributed to the LPO of chlorite. In addition, thetrench-normal seismic anisotropy observed in northeastJapan with a low dipping angle of slab (θ < 45º) (Huanget al. 2011b) may also be caused by the LPO of chlorite inthe mantle wedge. Recently, a trench-parallel seismic ani-sotropy was observed in a limited area under Oregon usingthe Rayleigh waves (Wagner et al. 2013) where the sourcearea of the trench-parallel seismic anisotropy includes thenose of 2-D corner flow. In this area, chlorite can be stable(Hacker et al. 2003; Till et al. 2012) and the dip angle ofthe flow of chlorite peridotite in this area is high(Figure 14(a)) so that the trench-parallel seismic aniso-tropy observed in the limited area (Wagner et al. 2013)

Figure 11. (a) Olivine fabric diagram in terms of water contentand stress, modified after Jung et al. (2006). AD representsAlmklovdalen specimen in this study. AD in black rectanglerepresents the data of the A-type olivine LPO of Almklovdalen.AD in green rectangle represents the data of the E-type olivineLPO of Almklovdalen. (b) Olivine fabric diagram of B-/C-typeLPO transition, modified after Skemer et al. (2006). Solidsquares are the data from deformation experiments that formedB-type fabrics, and open squares that formed C-type fabrics(Katayama and Karato 2006). HA, Higashi Akaishi peridotiteswith B-type olivine LPO (Mizukami et al. 2004); CDG, Cima diGagnone (Skemer et al. 2006); and AA, Alpe Arami (Skemeret al. 2006). AD in a black rectangle represents the data of the B-type olivine LPO of Almklovdalen in this study. AD in a whiterectangle represents the data of the C-type olivine LPO ofAlmklovdalen in this study.

662 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

may be attributed to the LPO of chlorite. In contrast, forthe high dipping angle (θ > 50º) of slab in the subductionzone (Figure 14(b)), LPO of chlorite will cause trench-parallel seismic anisotropy for the vertically propagatingfast S-waves. For example, the slab under both Tonga andMariana is dipping at a high angle and a strong trench-parallel seismic anisotropy was observed under the fore-arc and arc areas (Smith et al. 2001; Pozgay et al. 2007),which may be attributed to the LPO of chlorite in additionto the B-type LPO of olivine in the mantle wedge. Figure14(c) illustrates the change of seismic anisotropy when adipping angle of slab in a subduction zone changes fromlow to high angle. In this case, LPO of chlorite is sup-posed to cause a change in seismic anisotropy from thetrench-normal anisotropy in the low-angle subductionzone to the trench-parallel anisotropy in the high-anglesubduction zone. It is very interesting to observe similarchange of the seismic anisotropy from trench-normal totrench-parallel in the Nicaragua subduction zone (Abtet al. 2009), where the dipping angle of the slab inNicaragua changes from low to high angle, moving fromfore-arc to back-arc areas.

6. Conclusion

Various olivine LPO types (A-, B-, C-, and E-) were foundin chlorite peridotites in Almklovdalen (Figure 4). FTIRspectroscopy of the olivine revealed that the A-type oli-vine LPO is found in dry conditions, whereas the B-typeolivine LPO is found in wet conditions. This result isconsistent with former experimental studies (Jung andKarato 2001a; Jung et al. 2006; Katayama and Karato

2006). The highest olivine water content in the sampleswas 340 ± 30 ppm H/Si, which belonged to a sample witha B-type olivine LPO (Figure 9). It is found that the LPOof olivine is relatively weak, whereas the LPO of chloriteis very strong. The LPO of chlorite showed [001] axesstrongly aligned subnormal to the foliation, with a weakgirdle subnormal to the lineation. Since chlorite is muchmore anisotropic elastically than olivine (Figure 12), seis-mic anisotropy of chlorite LPOs showed much higheranisotropy than that in olivine LPOs. Adding the seismicanisotropy of chlorite to that of olivine, the seismic aniso-tropy of a whole rock was increased up to Vp = 8.3% andAVs = 13.1% (Figure 7(b)). In addition, it is found thatincreasing the dip angle (θ) of a slab in a subduction zonehighly rotates the polarization direction of the fast S-wave(Figures 13 and 14), resulting in the polarization directionof the fast S-wave subnormal to the lineation (flow direc-tion) with increasing dip angle, which can explain trench-parallel seismic anisotropy in a high-angle subductionzone. Because chlorite has a wide stability field at highpressure and high temperature in a mantle wedge and in asubducting slab (Hacker et al. 2003; Fumagalli and Poli2004; Padrón-Navarta et al. 2010; Van Keken et al. 2011;Till et al. 2012; Wada et al. 2012), the strong LPO ofchlorite can be a source of the observed trench-normal ortrench-parallel seismic anisotropy in the mantle wedge(Smith et al. 2001; Currie et al. 2004; Park et al. 2004;Pozgay et al. 2007; Long and Silver 2008; Huang et al.2011b; Wagner et al. 2013) as well as in subducting slabs(Wagner et al. 2013; Wang and Zhao 2013) depending onthe dipping angle of slab in a subduction zone wherechlorite is stable.

Figure 12. Seismic velocity and anisotropy for a single crystal of olivine and chlorite, presented in the lower hemisphere using an equal-area projection. The P-wave velocity (Vp), amplitude of S-wave anisotropy (AVs), and the polarization direction of the fast S-wave (Vs1)are plotted. The centre of the pole figure represents the [001] axis, the E–W direction represents the [010] axis, and the N–S directionrepresents the [100] axis.

International Geology Review 663

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

Figure 13. Effect of dip angle (θ) on the seismic anisotropy of chlorite (sample 436). In the cartoon, green arrows represent verticallypropagating S-waves. The red arrow represents mantle flow direction. The P-wave velocity (Vp), amplitude of S-wave anisotropy (AVs),and the polarization direction of the fast S-wave (Vs1) are plotted. The plots are shown in equal-area and lower-hemisphere projections.The x and z directions represent the lineation and the direction normal to the foliation, respectively. θ shows the eastward dip angle of thesubducting slab. Horizontal flow occurs when θ = 0°. The polarization direction of the vertically propagating fast S-wave (Vs1) is shownat the centre of the pole figure.

664 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

AcknowledgementsH.J. thanks Professor Håkon Austrheim for his help in collectingsamples at Almklovdalen.

FundingThis work was supported by the Mid-career Research Programmethrough an NRF grant to H.J. funded by the MEST (3345-20130011) in Korea. The calculations of seismic anisotropy weremade using the programs of David Mainprice (http://www.gm.univ-montp2.fr/spip/spip.php?rubrique179).

ReferencesAbramson, E., Brown, J., Slutsky, L., and Zaug, J., 1997, The

elastic constants of San Carlos olivine to 17 GPa: Journal ofGeophysical Research: Solid Earth (1978–2012), v. 102, no.B6, p. 12253–12263. doi:10.1029/97JB00682

Abt, D.L., Fischer, K.M., Abers, G.A., Strauch, W., Protti, J.M.,and González, V., 2009, Shear wave anisotropy beneathNicaragua and Costa Rica: Implications for flow in themantle wedge: Geochemistry, Geophysics, Geosystems, v.10, no. 5, p. 1–26.

Aleksandrov, K., and Ryzhova, T., 1961, The elastic properties ofrock forming minerals, pyroxenes and amphiboles: Bulletin

of Academy of Sciences USSR, Geophysical Series, v. 9,p. 871–875.

Bell, D.R., Rossman, G.R., Maldener, J., Endisch, D., andRauch, F., 2003, Hydroxide in olivine: A quantitative deter-mination of the absolute amount and calibration of the IRspectrum: Journal of Geophysical Research: Solid Earth(1978–2012), v. 108, no. B2. doi:10.1029/2001JB000679

Ben Ismaïl, W., and Mainprice, D., 1998, An olivine fabricdatabase: An overview of upper mantle fabrics and seismicanisotropy: Tectonophysics, v. 296, no. 1–2, p. 145–157.doi:10.1016/S0040-1951(98)00141-3

Berman, R.G., Aranovich, L.Y., and Pattison, D.R.M., 1995,Reassessment of the garnet-clinopyroxene Fe−Mg exchangethermometer: II Thermodynamic Analysis: Contributions toMineralogy and Petrology, v. 119, no. 1, p. 30–42.

Bystricky, M., Kunze, K., Burlini, L., and Burg, J.-P., 2000, Highshear strain of olivine aggregates: Rheological and seismicconsequences: Science, v. 290, no. 5496, p. 1564–1567.doi:10.1126/science.290.5496.1564

Couvy, H., Frost, D.J., Heidelbach, F., Nyilas, K., Ungár, T.,Mackwell, S., and Cordier, P., 2004, Shear deformationexperiments of forsterite at 11 GPa – 1400°C in the multi-anvil apparatus: European Journal of Mineralogy, v. 16, no.6, p. 877–889. doi:10.1127/0935-1221/2004/0016-0877

Currie, C.A., Cassidy, J.F., Hyndman, R.D., and Bostock, M.G.,2004, Shear wave anisotropy beneath the Cascadia subduc-tion zone and western North American craton: Geophysical

θ ≤ 45°

θ ≤ 45°

θ > 50°

θ > 50°

θ

θ

θ

θ

Figure 14. Schematic diagram showing the effect of the dipping angle of the slab on the seismic anisotropy caused by the LPO ofchlorite, assuming 2-D corner flow. (a) Trench-normal seismic anisotropy in the low-angle subduction zone (θ ≤ 45º). (b) Trench-parallelseismic anisotropy in the high-angle subduction zone (θ > 50º). (c) Change in the seismic anisotropy corresponding to the change in thedipping angle of the slab. Blue and red bars represent trench-normal and trench-parallel seismic anisotropy of the fast S-wave,respectively. (a) Low-angle (warm) subduction zone. (b) High-angle (cold) subduction zone. (c) Angle-changing subduction zone.

International Geology Review 665

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

Journal International, v. 157, no. 1, p. 341–353. doi:10.1111/j.1365-246X.2004.02175.x

Fumagalli, P., and Poli, S., 2004, Experimentally determinedphase relations in hydrous peridotites to 6.5 GPa and theirConsequences on the dynamics of subduction zones: Journalof Petrology, v. 46, no. 3, p. 555–578. doi:10.1093/petrol-ogy/egh088

Gifkins, R.C., 1970, Optical microscopy of metals: New York,American Elsevier.

Hacker, B.R., Andersen, T.B., Johnston, S., Kylander-Clark, A.R.C., Peterman, E.M., Walsh, E.O., and Young, D., 2010,High-temperature deformation during continental-marginsubduction & exhumation: The ultrahigh-pressure westerngneiss region of Norway: Tectonophysics, v. 480, no. 1–4,p. 149–171. doi:10.1016/j.tecto.2009.08.012

Hacker, B.R., Peacock, S.M., Abers, G.A., and Holloway, S.D.,2003, Subduction factory – 2. Are intermediate-depth earth-quakes in subducting slabs linked to metamorphic dehydra-tion reactions?: Journal of Geophysical Research-SolidEarth, v. 108, no. B1, p. 1–16. doi:10.1029/2001JB001129

Holtzman, B., Kohlstedt, D., Zimmerman, M., Heidelbach, F.,Hiraga, T., and Hustoft, J., 2003, Melt segregation and strainpartitioning: Implications for seismic anisotropy and mantleflow: Science, v. 301, no. 5637, p. 1227–1230. doi:10.1126/science.1087132

Huang, Z., Zhao, D., and Wang, L., 2011a, Frequency-dependentshear-wave splitting and multilayer anisotropy in northeastJapan: Geophysical Research Letters, v. 38, no. 8.doi:10.1029/2011GL046804

Huang, Z., Zhao, D., and Wang, L., 2011b, Shear wave aniso-tropy in the crust, mantle wedge, and subducting Pacific slabunder northeast Japan: Geochemistry, Geophysics,Geosystems, v. 12, no. 1. doi:10.1029/2010GC003343

Jung, H., 2009, Deformation fabrics of olivine in Val Malencoperidotite found in Italy and implications for the seismicanisotropy in the upper mantle: Lithos, v. 109, no. 3–4, p.341–349. doi:10.1016/j.lithos.2008.06.007

Jung, H., 2011, Seismic anisotropy produced by serpentine inmantle wedge: Earth and Planetary Science Letters, v. 307,no. 3–4, p. 535–543. doi:10.1016/j.epsl.2011.05.041

Jung, H., and Karato, S.-I., 2001a, Water-induced fabric transi-tions in olivine: Science, v. 293, no. 5534, p. 1460–1463.doi:10.1126/science.1062235

Jung, H., and Karato, S.-I., 2001b, Effects of water on dynami-cally recrystallized grain-size of olivine: Journal of StructuralGeology, v. 23, no. 9, p. 1337–1344. doi:10.1016/S0191-8141(01)00005-0

Jung, H., Katayama, I., Jiang, Z., Hiraga, T., and Karato, S.,2006, Effect of water and stress on the lattice-preferredorientation of olivine: Tectonophysics, v. 421, no. 1–2, p.1–22. doi:10.1016/j.tecto.2006.02.011

Jung, H., Lee, J., Ko, B., Jung, S., Park, M., Cao, Y., and Song,S.G., 2013, Natural type-C olivine fabrics in garnet perido-tites in North Qaidam UHP collision belt, NW China:Tectonophysics, v. 594, p. 91–102. doi:10.1016/j.tecto.2013.03.025

Jung, H., Mo, W., and Choi, S., 2009a, Deformation microstruc-tures of olivine in peridotite from Spitsbergen, Svalbard andimplications for seismic anisotropy: Journal of MetamorphicGeology, v. 27, no. 9, p. 707–720.

Jung, H., Mo, W., and Green, H.W., 2009b, Upper mantle seis-mic anisotropy resulting from pressure-induced slip transi-tion in olivine: Nature Geoscience, v. 2, no. 1, p. 73–77.doi:10.1038/ngeo389

Jung, S., Jung, H., and Austrheim, H., 2014, Characterization ofolivine fabrics and mylonite in the presence of fluid andimplications for seismic anisotropy and shear localization:Earth, Planets and Space, v. 66, p. 46. doi:10.1186/1880-5981-66-46

Karato, S., 1987, Scanning electron microscope observation ofdislocations in olivine: Physics and Chemistry of Minerals, v.14, no. 3, p. 245–248. doi:10.1007/BF00307989

Karato, S.-I., Jung, H., Katayama, I., and Skemer, P., 2008,Geodynamic significance of seismic anisotropy of the uppermantle: New insights from laboratory studies: AnnualReview of Earth Planetary Science, v. 36, p. 59–95.doi:10.1146/annurev.earth.36.031207.124120

Karato, S.I., Toriumi, M., and Fujii, T., 1980, Dynamic recrys-tallization of olivine single crystals during high-temperaturecreep: Geophysical Research Letters, v. 7, no. 9, p. 649–652.doi:10.1029/GL007i009p00649

Katayama, I., Hirauchi, K.-I., Michibayashi, K., and Ando, J.-I.,2009, Trench-parallel anisotropy produced by serpentinedeformation in the hydrated mantle wedge: Nature, v. 461,no. 7267, p. 1114–1117. doi:10.1038/nature08513

Katayama, I., Jung, H., and Karato, S.-I., 2004, New type ofolivine fabric from deformation experiments at modest watercontent and low stress: Geology, v. 32, no. 12, p. 1045.doi:10.1130/G20805.1

Katayama, I., and Karato, S., 2006, Effect of temperature on theB- to C-type olivine fabric transition and implication for flowpattern in subduction zones: Physics of the Earth andPlanetary Interiors, v. 157, no. 1–2, p. 33–45. doi:10.1016/j.pepi.2006.03.005

Katayama, I., Karato, S.-I., and Brandon, M., 2005, Evidence ofhigh water content in the deep upper mantle inferred fromdeformation microstructures: Geology, v. 33, no. 7, p. 613.doi:10.1130/G21332.1

Kneller, E.A., Van Keken, P.E., Katayama, I., and Karato, S.,2007, Stress, strain, and B-type olivine fabric in the fore-arcmantle: Sensitivity tests using high-resolution steady-statesubduction zone models: Journal of Geophysical Research-Solid Earth, v. 112, no. B4. doi:10.1029/2006JB004544

Köhler, T., and Brey, G., 1990, Calcium exchange betweenolivine and clinopyroxene calibrated as a geothermobarom-eter for natural peridotites from 2 to 60 kb with applications:Geochimica et Cosmochimica Acta, v. 54, no. 9, p. 2375–2388. doi:10.1016/0016-7037(90)90226-B

Kohlstedt, D., Goetze, C., Durham, W., and Sande, J., 1976, Newtechnique for decorating dislocations in olivine: Science, v.191, no. 4231, p. 1045–1046. doi:10.1126/science.191.4231.1045

Kohlstedt, D.L., Keppler, H., and Rubie, D.C., 1996, Solubilityof water in the α, β and γ phases of (Mg,Fe)2 SiO4:Contributions to Mineralogy and Petrology, v. 123, no. 4,p. 345–357. doi:10.1007/s004100050161

Lapen, T.J., Medaris, L.G., Beard, B.L., and Johnson, C.M.,2009, The Sandvik peridotite, Gurskøy, Norway: Three bil-lion years of mantle evolution in the Baltica lithosphere:Lithos, v. 109, no. 3–4, p. 145–154.

Long, M.D., 2013, Constraints on subduction geodynamics fromseismic anisotropy: Reviews of Geophysics, v. 51, no. 1, p.76–112. doi:10.1002/rog.20008

Long, M.D., and Silver, P.G., 2008, The subduction zone flowfield from seismic anisotropy: A global view: Science, v.319, no. 5861, p. 315–318. doi:10.1126/science.1150809

Mackwell, S.J., and Kohlstedt, D.L., 1990, Diffusion of hydro-gen in olivine: Implications for water in the mantle: Journal

666 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

of Geophysical Research: Solid Earth (1978–2012), v. 95,no. B4, p. 5079–5088. doi:10.1029/JB095iB04p05079

Mainprice, D., 1990, A Fortran program to calculate seismicanisotropy from the lattice preferred orientation of minerals:Computers & Geosciences, v. 16, no. 3, p. 385–393.doi:10.1016/0098-3004(90)90072-2

Mainprice, D., and Ildefonse, B., 2009, Seismic anisotropy ofsubduction zone minerals–contribution of hydrous phases,Subduction zone geodynamics: Berlin, Springer, p. 63–84.

McCormack, K., Wirth, E.A., and Long, M.D., 2013, B-typeolivine fabric and mantle wedge serpentinization beneaththe Ryukyu arc: Geophysical Research Letters, v. 40, no. 9,p. 1697–1702. doi:10.1002/grl.50369

Mei, S., and Kohlstedt, D.L., 2000, Influence of water on plasticdeformation of olivine aggregates: 1. Diffusion creep regime:Journal of Geophysical Research-Solid Earth, v. 105, no. B9,p. 21457–21469. doi:10.1029/2000JB900179

Michibayashi, K., and Oohara, T., 2013, Olivine fabric evolutionin a hydrated ductile shear zone at the Moho TransitionZone, Oman Ophiolite: Earth and Planetary ScienceLetters, v. 377–378, p. 299–310. doi:10.1016/j.epsl.2013.07.009

Michibayashi, K., Tasaka, M., Ohara, Y., Ishii, T., Okamoto, A.,and Fryer, P., 2007, Variable microstructure of peridotitesamples from the southern Mariana Trench: Evidence of acomplex tectonic evolution: Tectonophysics, v. 444, no. 1–4,p. 111–118. doi:10.1016/j.tecto.2007.08.010

Mizukami, T., Wallis, S.R., and Yamamoto, J., 2004, Naturalexamples of olivine lattice preferred orientation patternswith a flow-normal a-axis maximum: Nature, v. 427, no.6973, p. 432–436. doi:10.1038/nature02179

Mookherjee, M., and Mainprice, D., 2014, Unusually large shearwave anisotropy for chlorite in subduction zone settings:Geophysical Research Letters, v. 41, no. 5, p. 1506–1513.doi:10.1002/2014GL059334

Mosenfelder, J.L., Deligne, N.I., Asimow, P.D., and Rossman, G.R., 2006, Hydrogen incorporation in olivine from 2–12 GPa:American Mineralogist, v. 91, no. 2–3, p. 285–294.doi:10.2138/am.2006.1943

Nakamura, D., and Hirajima, T., 2005, Experimental evaluationof garnet–clinopyroxene geothermometry as applied to eclo-gites: Contributions to Mineralogy and Petrology, v. 150, no.6, p. 581–588. doi:10.1007/s00410-005-0023-x

Nicolas, A., and Christensen, N., 1987, Formation of anisotropyin upper mantle peridotites: A review, in Fuchs, K., andFroidevaux, C., eds., Composition, structure and dynamicsof the lithosphere-asthenosphere system: GeodynamicsSeries: Washington, D.C., American Geophysical Union,v. 16, p. 111–123. doi:10.1029/GD016p0111

Nimis, P., and Taylor, W.R., 2000, Single clinopyroxenethermobarometry for garnet peridotites. Part I. Calibrationand testing of a Cr-in-Cpx barometer and an enstatite-in-Cpx thermometer: Contributions to Mineralogyand Petrology, v. 139, no. 5, p. 541–554. doi:10.1007/s004100000156

Nishii, A., Wallis, S., Mizukami, T., and Michibayashi, K., 2011,Subduction related antigorite CPO patterns from forearcmantle in the Sanbagawa belt, southwest Japan: Journal ofStructural Geology, v. 33, no. 10, p. 1436–1445.doi:10.1016/j.jsg.2011.08.006

Ohuchi, T., Kawazoe, T., Nishihara, Y., and Irifune, T., 2012,Change of olivine a-axis alignment induced by water: Originof seismic anisotropy in subduction zones: Earth andPlanetary Science Letters, v. 317–318, p. 111–119.doi:10.1016/j.epsl.2011.11.022

Ohuchi, T., Kawazoe, T., Nishihara, Y., Nishiyama, N., andIrifune, T., 2011, High pressure and temperature fabric tran-sitions in olivine and variations in upper mantle seismicanisotropy: Earth and Planetary Science Letters, v. 304, no.1–2, p. 55–63. doi:10.1016/j.epsl.2011.01.015

Padrón-Navarta, J.A., Tommasi, A., Garrido, C.J., Sánchez-Vizcaíno, V.L., Gómez-Pugnaire, M.T., Jabaloy, A., andVauchez, A., 2010, Fluid transfer into the wedge controlledby high-pressure hydrofracturing in the cold top-slab mantle:Earth and Planetary Science Letters, v. 297, no. 1–2, p. 271–286. doi:10.1016/j.epsl.2010.06.029

Panozzo, R., 1984, Two-dimensional strain from the orientationof lines in a plane: Journal of Structural Geology, v. 6, no. 1–2, p. 215–221. doi:10.1016/0191-8141(84)90098-1

Park, J., and Levin, V., 2002, Seismic anisotropy: Tracing platedynamics in the mantle: Science, v. 296, no. 5567, p. 485–489. doi:10.1126/science.1067319

Park, J., Yuan, H.Y., and Levin, V., 2004, Subduction zoneanisotropy beneath Corvallis, Oregon: A serpentinite skidmark of trench-parallel terrane migration?: Journal ofGeophysical Research-Solid Earth, v. 109, no. B10.doi:10.1029/2003JB002718

Park, M., Jung, H., and Kil, Y., 2014, Petrofabrics of olivine in arift axis and rift shoulder and their implications for seismicanisotropy beneath the Rio Grande rift: Island Arc (inrevision).

Park, Y., and Jung, H., 2015, Deformation microstructures of oli-vine and pyroxene in mantle xenoliths in Shanwang, easternChina, near the convergent plate margin and implications forseismic anisotropy: International Geology Review, v. 57,p. 629–649. doi:10.1080/00206814.2014.928240

Paterson, M., 1982, The determination of hydroxyl by infraredabsorption in quartz, silicate glasses and similar materials:Bull Mineral, v. 105, no. 1, p. 20–29.

Pozgay, S., Wiens, D., Conder, J., Shiobara, H., and Sugioka, H.,2007, Complex mantle flow in the Mariana subduction sys-tem: Evidence from shear wave splitting: GeophysicalJournal International, v. 170, no. 1, p. 371–386.doi:10.1111/j.1365-246X.2007.03433.x

Puelles, P., Gil Ibarguchi, J.I., Beranoaguirre, A., and Ábalos, B.,2012, Mantle wedge deformation recorded by high-tempera-ture peridotite fabric superposition and hydrous retrogression(Limo massif, Cabo Ortegal, NW Spain): InternationalJournal of Earth Sciences, v. 101, no. 7, p. 1835–1853.doi:10.1007/s00531-012-0761-8

Russ, J., and Dehoff, R., 2000, Practical stereology: New York,Kluwer Academic Press.

Russo, R.M., and Silver, P.G., 1994, Trench-parallel flow beneaththe nazca plate from seismic anisotropy: Science, v. 263, no.5150, p. 1105–1111. doi:10.1126/science.263.5150.1105

Savage, M., 1999, Seismic anisotropy and mantle deformation:What have we learned from shear wave splitting?: Reviewsof Geophysics, v. 37, no. 1, p. 65–106. doi:10.1029/98RG02075

Skemer, P., Katayama, I., Jiang, Z., and Karato, S.-I., 2005, Themisorientation index: Development of a new method forcalculating the strength of lattice-preferred orientation:Tectonophysics, v. 411, no. 1–4, p. 157–167. doi:10.1016/j.tecto.2005.08.023

Skemer, P., Katayama, I., and Karato, S.-I., 2006, Deformationfabrics of the Cima di Gagnone peridotite massif, CentralAlps, Switzerland: Evidence of deformation at low tempera-tures in the presence of water: Contributions to Mineralogyand Petrology, v. 152, no. 1, p. 43–51. doi:10.1007/s00410-006-0093-4

International Geology Review 667

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15

Smith, G.P., Wiens, D.A., Fischer, K.M., Dorman, L.M., Webb,S.C., and Hildebrand, J.A., 2001, A complex pattern ofmantle flow in the Lau backarc: Science, v. 292, no. 5517,p. 713–716. doi:10.1126/science.1058763

Till, C.B., Grove, T.L., and Withers, A.C., 2012, The beginningsof hydrous mantle wedge melting: Contributions toMineralogy and Petrology, v. 163, no. 4, p. 669–688.doi:10.1007/s00410-011-0692-6

Ulmer, P., and Trommsdorff, V., 1995, Serpentine stability tomantle depths and subduction-related magmatism: Science,v. 268, no. 5212, p. 858–861. doi:10.1126/science.268.5212.858

Van Der Wal, D., Chopra, P., Drury, M., and Gerald, J.F., 1993,Relationships between dynamically recrystallized grain sizeand deformation conditions in experimentally deformed oli-vine rocks: Geophysical Research Letters, v. 20, no. 14, p.1479–1482. doi:10.1029/93GL01382

Van Keken, P.E., Hacker, B.R., Syracuse, E.M., and Abers, G.A.,2011, Subduction factory: 4. Depth-dependent flux of H2Ofrom subducting slabs worldwide: Journal of GeophysicalResearch: Solid Earth (1978–2012), v. 116, no. B1.doi:10.1029/2010JB007922

Wada, I., Behn, M.D., and Shaw, A.M., 2012, Effects of hetero-geneous hydration in the incoming plate, slab rehydration,and mantle wedge hydration on slab-derived H2O flux in

subduction zones: Earth and Planetary Science Letters, v.353–354, p. 60–71. doi:10.1016/j.epsl.2012.07.025

Wagner, L.S., Fouch, M.J., James, D.E., and Long, M.D., 2013,The role of hydrous phases in the formation of trench parallelanisotropy: Evidence from Rayleigh waves in Cascadia:Geophysical Research Letters, v. 40, no. 11, p. 2642–2646.doi:10.1002/grl.50525

Wang, J., and Zhao, D., 2013, P-wave tomography for 3-D radialand azimuthal anisotropy of Tohoku and Kyushu subductionzones: Geophysical Journal International, v. 193, no. 3, p.1166–1181. doi:10.1093/gji/ggt086

Wang, Q., Xia, Q.K., O’Reilly, S.Y., Griffin, W.L., Beyer, E.E.,and Brueckner, H.K., 2013, Pressure- and stress-inducedfabric transition in olivine from peridotites in the WesternGneiss region (Norway): Implications for mantle seismicanisotropy: Journal of Metamorphic Geology, v. 31, no. 1,p. 93–111. doi:10.1111/jmg.12011

Watanabe, T., Shirasugi, Y., Yano, H., and Michibayashi, K.,2011, Seismic velocity in antigorite-bearing serpentinitemylonites: Geological Society, London, SpecialPublications, v. 360, no. 1, p. 97–112. doi:10.1144/SP360.6

Webber, C.E., Little, T., Newman, J., and Tikoff, B., 2008, Fabricsuperposition in upper mantle peridotite, Red Hills, NewZealand: Journal of Structural Geology, v. 30, no. 11, p.1412–1428. doi:10.1016/j.jsg.2008.08.002

668 D. Kim and H. Jung

Dow

nloa

ded

by [

Seou

l Nat

iona

l Uni

vers

ity]

at 2

2:16

21

Apr

il 20

15