RESEARCH

Geochemistry and geochronology of the Troodos ophiolite: An SSZ ophiolite generated by subduction initiation and an extended episode of ridge subduction?

Soichi Osozawa1,*, Ryuichi Shinjo2, Ching-Hua Lo3, Bor-ming Jahn3, Nguyen Hoang4, Minoru Sasaki5, Ken’ichi Ishikawa6, Harumasa Kano7, Hiroyuki Hoshi8, Costas Xenophontos9, and John Wakabayashi10 1DEPARTMENT OF EARTH SCIENCES, GRADUATE SCHOOL OF SCIENCE, TOHOKU UNIVERSITY, SENDAI 980-8578, JAPAN 2DEPARTMENT OF PHYSICS AND EARTH SCIENCES, UNIVERSITY OF THE RYUKYUS, NISHIHARA 903-0213, JAPAN 3DEPARTMENT OF GEOSCIENCES, NATIONAL TAIWAN UNIVERSITY, TAIPEI 10699, TAIWAN 4FORMER: GEOLOGICAL SURVEY OF JAPAN, ADVANCED INDUSTRIAL SCIENCE AND TECHNOLOGY (AIST), TSUKUBA 305-8567, JAPAN 5DEPARTMENT OF EARTH AND ENVIRONMENTAL SCIENCES, GRADUATE SCHOOL SCIENCE AND TECHNOLOGY, HIROSAKI UNIVERSITY, HIROSAKI 036-8561, JAPAN 6CENTER FOR THE ADVANCEMENT OF HIGHER EDUCATION, TOHOKU UNIVERSITY, SENDAI 980-8576, JAPAN 7TOHOKU UNIVERSITY MUSEUM, SENDAI 980-8578, JAPAN 8DEPARTMENT OF EARTH SCIENCES, GRADUATE SCHOOL OF EDUCATION, AICHI UNIVERSITY OF EDUCATION, KARITA 448-8542, JAPAN 9FORMER: CYPRUS GEOLOGICAL SURVEY, NICOSIA 1415, CYPRUS 10DEPARTMENT OF EARTH AND ENVIRONMENTAL SCIENCES, CALIFORNIA STATE UNIVERSITY, FRESNO, FRESNO, CALIFORNIA 93740, USA

ABSTRACT

New trace-element, radiogenic isotopic, and geochronologic data from the Troodos ophiolite, considered in concert with the large body of previously published data, give new insight into the tectonic history of this storied ophiolite, as well as demonstrating the variability of suprasubduction-zone ophiolites, and differences between them and commonly used modern analogs. Similar to earlier studies, we fi nd that island-arc tholeiite of the lower pillow lava sequence erupted fi rst, followed by boninite. We further divide boninitic rocks into boninite making up the upper pillow lava sequence, and depleted boninites that we consider late infi ll lavas. We obtained an Ar-Ar age from arc tholeiite of 90.6 ± 1.2 Ma, comparable to U-Pb ages from ophiolite plagiogranites. New biostratigraphic data indicate that most of the basal pelagic sedimentary rocks that conformably overlie the boninitic rocks are ca. 75 Ma. This suggests that voluminous eruption of boninitic rocks persisted until ca. 75 Ma. Limited eruption of boninitic lavas may have continued until 55.5 ± 0.9 Ma, based on the Ar-Ar age we obtained. The duration of arc magmatism at Troodos (at least 16 m.y., with some activity perhaps extending 35 m.y.) without the devel- opment of a mature arc edifi ce greatly exceeds that of other well-studied suprasubduction-zone ophiolites. We propose that Troodos was formed over a newly formed subduction zone, similar to many proposed models, but that the extended period of magmatism (boninitic) resulted from a prolonged period of ridge subduction.

LITHOSPHERE; v. 4; no. 6; p. 497–510 | Published online 14 November 2012 doi: 10.1130/L205.1

INTRODUCTION zone magmatism in the generation of many erated, given that LILE enrichment is a funda- ophiolites (e.g., Shervais, 2001; Pearce, 2003; mental characteristic of arc lavas. Analysis of Research on the Troodos ophiolite of Cyprus Nicolas and Boudier, 2003; Pearce and Robin- fresh volcanic glass of pillow and dike margins, has strongly infl uenced the evolution of the son, 2010). however, can largely surmount this diffi culty, as ophiolite concept and ideas on the signifi cance The interpretation of suprasubduction-zone evidenced by the extensive major-element data of on-land sheets of oceanic lithosphere in oro- generation of many ophiolites derives primarily set of Pearce and Robinson (2010), and trace- genic belts (e.g., Moores and Vine, 1971; Dilek, from geochemical affi nities of ophiolite lavas element and isotopic analyses (e.g., Rauten- 2003; Robinson et al., 2003). Accordingly, pet- and dikes. Since Miyashiro (1975) initially schlein et al., 1985), which demonstrate an arc rologic and geochemical studies of the Troodos proposed an island-arc setting for the Troodos origin for the lavas. ophiolite have also fi gured prominently in the ophiolite generation based on its major-element Although previous studies have fi rmly estab- wide acceptance of the role of suprasubduction- compositions, many subsequent studies, cited in lished the suprasubduction-zone origin of the this paper, have identifi ed island-arc characteris- Troodos ophiolite, further exploration of the tics, based on minor- and trace-element compo- geochemistry and geochronology of the lavas sitions and isotopic ratios. offers additional insight into details of mag- *E-mail: [email protected]. A potential problem in some of the early matic evolution of the Troodos ophiolite and Editor’s note: This article is part of a special issue ti- geochemical studies is the mobility of large other suprasubduction-zone ophiolites as well tled “Initiation and Termination of Subduction: Rock Re- ion lithophile elements (LILEs) during altera- as providing a means by which to better com- cord, Geodynamic Models, Modern Plate Boundaries,” edited by John Shervais and John Wakabayashi. The tion, which may result in enrichment of these pare the geochemistry of Troodos rocks to other full issue can be found at http://lithosphere.gsapubs elements in nonarc volcanic rocks, and an erro- ophiolites and modern seafl oor rocks (particu- .org/content/4/6.toc. neous interpretation of such rocks as arc gen- larly those in arc environments).

LITHOSPHEREFor permission to| Volumecopy, contact 4 | Number [email protected] 6 | www.gsapubs.org | © 2012 Geological Society of America 497

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

infill lava (depleted boninite) 106(th) 28(th) 15(bn) 103(bn) upper pillow lava sequence (boninite) dip and strike of lava 26(th) dip and strike of dike 104bn 21th 27th lower pillow lava sequence (tholeiite) Ar-Ar 3(bn) Vuoni sheeted dike complex Skouriotissa mine Akamas peninsula 5th gabbro

157(bn) CR Koronia Mitsero mine ultramafic rocks Akaki River sheeted dike complex KambiaAnaliondas Troulli Margi MR 18MR MC 35°N 44CRMR Troodos 99dbn MR 55(th) MR 111CR 88dbn MR Arakapas 125CR Ayios Nocolaos MR MR47MR Ayios Mamas MR N Kalavasos mine Pafos 144bn MR 139CR 137CR 152dbn 119dbn 132CR island of Cyprus Limassol Ar-Ar 138bn Troodos ophiolite 130bn Lefkara F. Nicosia 20 km R Late Campanian radiolarians 21th geochemical site C: chert, M: mudstone th: tholeiite, bn: boninite, dbn: depleted boninite (Perapedhi F.) 100 km 33°E (th) indicated by major elements only Figure 1. Geological map of the Troodos ophiolite, after Geological Survey Department Cyprus (1995) and Gass et al. (1994). There are 11 sample loca- tions, but 12 total samples (location 152 has f1 glass and f2 rock samples), with the sample numbers in the larger font size. Stars are sample localities of sediments (sample numbers in smaller font size), all of which contain late Campanian radiolarians.

To exploit the improvements in technology GEOLOGIC SETTING al., 1994). Ultramafi c rocks, gabbro, and sheeted and methodology and gain better insight into dikes also crop out in the Limassol Forest com- the tectonic settings of Troodos magmatism, The Troodos ophiolite extends 100 km in plex, but tectonism has severely disrupted the we analyzed trace elements of the 12 freshest E-W and 30 km in N-S dimensions (Geological original tectonomagmatic relationships. The samples from throughout the ophiolite (Fig. 1), Survey Department Cyprus, 1995). The Troodos Troulli inlier is isolated from the ophiolite, but it and also analyzed the Nd, Sr, and Pb isotope massif forms a gentle dome structure elongated is clearly an eastern extension of the main mas- ratios of fi ve of these samples. Whereas our E-W (Fig. 1). Because of the superposition of sif. Another isolated massif with the same lithol- samples span a greater geographical extent of this domal structure and erosion, the structurally ogy as the main massif crops out on the Akamas the Troodos ophiolite than the previous stud- lowest ultramafi c rocks constitute the center of Peninsula (Fig. 1). ies by Rautenschlein et al. (1985), Cameron the Troodos massif and the highest elevations at The Geological Survey Department Cyprus (1985), and others, our sample set is small Mount Olympus (1952 m altitude). The ultra- (1995) divided the pillow lavas into (1) a basal compared to the 137 samples of glass analyzed mafi c rocks are fl anked by gabbro associated group, (2) lower pillow lavas, (3) upper pillow by Pearce and Robinson (2010), although their with a small amount of plagiogranite, an exten- lavas, mostly in the axial sequence, and (4) lavas study presented major-element data only. sive sheeted dike complex, pillow lava, and and volcanic breccias in the Arakapas sequence In addition to geochemical data, we present sediments, in ascending structural-stratigraphic whose stratigraphic relationship with the other radiolarian biostratigraphic data, some paleo- order (Fig. 1). three groups is not clear. The basal group lavas magnetic data, and Ar-Ar age analyses of two The Troodos ophiolite has been subdivided display greenschist-facies hydrothermal meta- fresh volcanic glasses that directly date tholeiite into the northern main massif and its axial morphism (Geological Survey Department and boninite eruption in the Troodos ophiol- sequence, including Mt. Olympus, and the Cyprus, 1995) and are easily distinguished from ite. The combination of new geochemical and southern Limassol Forest complex and its Ara- the other lavas. However, the fi rst author could geochronologic data, when integrated with pub- kapas sequence, bounded by the E-W–trending not lithologically distinguish the upper and lower lished data, gives new insight into the sequence Arakapas transform fault zone (Fig. 1; Geologi- pillow lavas (cf. Murton, 1989), and sampled of the magmatism and its temporal extent. cal Survey Department Cyprus, 1995; Gass et lavas are classifi ed (Table 1) by their subunits as

498 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH rock standards 3.060.457 2.57 2.50 0.383 1.45 3.481.14 2.28 4.38 0.830 3.22 0.7610.370 0.255 0.153 1.230.637 0.471 0.79 1.063.11 0.869 0.472 2.60 0.396 0.491 0.254 2.15 1.14 0.1055.72 0.032 5.46 0.7644.91 0.605 4.00 1.79 13.4 5.12 2.31 11.2 6.28 11.1 26.2 11.6 8.84 11.7 6.6 85.1 47.5 31.6 25.1 13.5 6.57 12.0 40 30.3 57.8 304 215 266 178 108 598 below 104 104 close to

e v o 27 b s sequence lava, according to Geological Survey Department Cyprus a 5

e v o b a d e t a d

r A - r A

s a

e m 152f2 a (glass) (inside) s TABLE 1. MAJOR- AND TRACE-ELEMENT ANALYSES OF VOLCANIC ROCKS FROM THE TROODOS OPHIOLITE THE OF VOLCANIC ROCKS FROM ANALYSES TRACE-ELEMENT AND 1. MAJOR- TABLE d e t a d

r A - r A 1234567891011121314151617181920GSJ 1234567891011121314151617181920GSJ 1.30 1.24 1.29 0.59 0.42 0.40 0.44 0.31 0.29 0.21 0.29 0.38 0.63 1.30 0.47 2.91 1.00 0.44 1.17 0.32 0.09 0.09 0.11 0.03 0.05 0.03 0.03 0.02 0.01 0.01 0.02 0.02 0.12 0.23 0.16 0.30 0.06 0.14 0.24 0.11 14.28 13.95 14.03 14.28 15.27 14.03 14.09 12.69 12.30 11.83 14.00 15.65 15.78 13.80 15.08 10.10 14.38 15.88 14.22 14.75 TH—tholeiite; BON—boninite; d BON—depleted boninite; UPL—upper pillow lava; LPL—lower pillow lava; BS—basal group; ASL—Arakapa TH—tholeiite; BON—boninite; d BON—depleted boninite; UPL—upper pillow lava; LPL—lower BS—basal group; (wt%) 50.97 53.61 53.26 46.00 52.38 51.52 53.70 51.42 50.91 52.47 54.19 53.25 52.00 55.50 54.73 46.14 56.84 55.55 56.95 54.28 3 5 2 O 2.57 3.01 2.92 1.74 1.54 1.33 1.86 0.65 0.94 0.48 1.14 1.25 1.96 2.52 1.44 1.35 2.44 1.80 2.87 1.12 2 2 O O O 0.62 0.20 0.70 0.11 0.21 0.20 0.33 0.11 0.18 0.21 0.18 0.27 0.09 0.35 0.16 0.23 0.22 0.20 0.24 0.20 2 Note: 2 2 SiO Sm 2.48 2.68 2.63 1.17 1.03 0.914 0.825 0.415 0.385 0.308 0.53 0.61 ThU 0.227 0.248 0.107 0.298 0.134 0.128 0.146 0.152 0.06 0.223 0.077 0.131 0.08 0.113 0.091 0.096 0.061 0.2 0.049 0.088 0.137 0.107 0.084 0.103 Ho 1.06 1.08 1.15 0.507 0.446 0.387 0.473 0.323 0.301 0.263 0.339 0.443 CePrNd 6.65 1.14 7.3 6.57 1.27 7.2 6.89 1.16 3.12 6.94 0.529 2.86 3.03 0.472 0.523 3.56 2.66 0.357 2.17 2.63 0.142 0.913 1.99 0.121 0.824 0.156 0.799 1.24 0.189 0.708 1.2 0.224 0.736 1.11 1.43 1.28 YbLuHfTa 3.13Pb 0.492 3.28 1.72 0.495 0.062 3.62 0.536 2.03 1.22 0.068(1995). JA-1 and JB-2 data are average of three time measurements during analytical session. 0.233 1.54 1.9 0.069 1.38 0.2 1.33 0.041 1.38 0.811 0.056 0.188 1.23 0.727 1.08 0.075 0.204 1.41 0.748 0.033 0.698 0.178 0.77 1.14 0.076 0.633 0.17 0.315 0.07 0.98 1.09 0.158 0.292 0.065 1.2 0.183 1 0.269 0.024 0.236 0.499 0.835 0.028 1.18 0.565 2.3 1.48 1.45 1.7 Total 97.52 96.60 96.52 90.99 96.50 95.49 97.60 95.49 95.15 97.94 98.22 96.17 98.03 96.07 97.42 91.30 97.48 97.77 95.38 96.69 EuGdTbDy 0.906 3.6 0.99 0.693 0.932 4.63 3.9 0.754 0.467 0.746 5.12 3.84 0.412 0.33 5.15 1.72 0.368 0.297 2.25 0.327 1.5 0.262 0.174 1.97 0.272 1.35 0.162 0.178 1.79 0.122 1.27 0.17 1.98 0.215 0.806 0.138 0.26 1.33 0.765 0.203 1.25 0.602 0.242 0.93 1.07 1.11 1.5 1.83 ScVCr 36.7 420 36.8 8.4 444 33.8 16.2Zr 330 41.2Nb 7.33Cs 251 37.9 337Ba 61.6 40.9La 209 287 0.982 68.3 40.4 1.1 223 1.05 40.6 467 65.1 46.4 1.05 240 2.21 0.213 23.1 305 27.8 42.8 0.611 0.672 2.42 255 32.5 952 23.4 0.133 47 0.979 2.27 11.5 239 0.186 1.2 25 608 1.07 44.9 0.141 22.9 251 924 0.562 0.117 24.2 1.21 43 21 255 1.15 0.189 470 1.58 7.89 0.169 1.07 10.4 276 0.82 7.22 0.577 61 0.855 12.9 0.328 0.523 7.68 0.35 10.9 0.367 0.431 13.3 0.4 22.3 0.693 14.9 0.493 12.8 0.61 21.4 ErTm 3.12 0.501 3.27 0.503 0.549 3.53 0.234 1.52 0.205 1.34 0.186 0.224 1.2 0.165 1.49 0.159 0.142 1.01 0.174 0.978 0.225 0.852 1.1 1.43 P TiO K MgOCaONa 4.61 8.12 4.04 8.13 3.49 6.92 7.48 11.74 7.53 11.18 9.77 9.60 10.03 8.21 12.18 9.02 11.87 9.24 10.32 13.29 11.07 10.30 8.77 12.30 6.20 7.50 7.61 10.86 3.18 8.23 6.60 8.23 10.36 5.53 6.98 3.59 11.45 5.77 2.83 7.43 RbSr 9.28 170 4.27 93.1 8.28 122 2.79 75.5 4.09 95.3 5.17 88.2 6.44 82.5 4.03 43.2 4.18 150 6.99 28.6 5.07 47.4 6.67 197 Al FeO totalMnO 13.67 12.16 12.51 1.30 8.44 0.17 7.77 1.29 8.46 0.59 8.47 0.15 8.79 0.14 9.10 0.44 8.97 0.31 8.41 0.29 8.47 0.15 7.38 0.14 11.54 0.38 7.82 16.16 0.16 10.59 0.15 7.53 0.09 9.75 0.36 6.94 0.13 0.09 0.14 0.09 Mg#Li (ppm) 3.66 0.38 8.51 0.37 3.49 0.33 6.67 0.61 9.4 0.63 18.1 0.67 11.5 0.63 0.65 9.1 0.64 9.41 0.73 8.5 0.65 8.34 0.57 11.3 0.65 0.33 0.60 0.38 0.38 0.58 0.34 0.66 Be 0.431 0.423 0.696 0.182 0.175 0.207 0.169 0.085 0.078 0.152 0.131 0.158 Y 32.9 32.8 36 15.3 13.6 11.6 13.8 9.44 8.91 7.67 10.1 14.7 CoNi 38.7 12.3 40.9 24.9 35 8.26 43.7 107 35.3 66.3 45.5 211 36.7 72.4 47.6 208 50.4 288 54.4 321 41.7 110 38.5 55.5 Rock type:Horizon: TH LPL TH LPL TH LPL LPL, UPL BON LPL BONASL LPL, UPL BON UPL BON d BON UPL d BON d BON ASL d BON UPL d BON BON UPL TH UPL BON LPL TH UPL LPL TH UPL BON UPL TH d BON BG Andesite Basalt UPL Sample: CY21 CY27 CY5 CY144 CY138 CY130 CY104 CY152f1 CY152f2 CY119 CY88 CY99 CY3 CY26 CY15 CY28 CY55 CY103 CT106 CY157 JA-1 JB-2

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 499

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

defi ned by the Geological Survey Department fi eld relationships (Varga and Moores, 1985; may account for those element contents that Cyprus (1995) mapping, although the strati- Allerton and Vine, 1991; Moores et al., 1990). show greater than 10% difference. graphic position of several samples in the axial In Figure 1, three antiformal axes in lavas on Nd, Sr, and Pb isotope ratios (Table 2; total: sequence can be described (see Table 1). Volcanic the northern fl ank correspond to these grabens; fi ve samples) were measured in a multicollec- breccias are distributed along the Arakapas valley the antiformal dip confi guration results from tor VG Sector 54 thermal ionization mass spec- and its western and eastern extensions, and they tilting of lavas in the hanging wall of graben- trometer (TIMS) at the Geological Survey of characterize the transform fault zone. Associated bounding listric normal faults. The relationship Japan, AIST. Sample treatment, normalization lavas, especially those above the breccias, are between the lava stratigraphy and the proposed values for Sr and Nd isotope ratios, their within- expected to occupy the stratigraphically highest ridge jumps is not clear, and we will revisit and run precision, their standard sample ratios, mass level. Gass et al. (1994) called these infi ll lavas, discuss these relationships later in the context of fractionation correction for Pb isotope ratios, and we follow that designation here. Weathered the relation of suites A and B. and their internal precision are described in infi ll lavas tend to be gray colored. Gass et al. Hoang and Uto (2006). Measured isotope ratios (1994) classifi ed pillow lavas cropping out along GEOCHEMISTRY were corrected to initial ratios based on the 91.6 the southern margin of the Limassol Forest com- Ma U-Pb zircon age obtained from the ophiolite plex as upper pillow lavas instead of infi ll lavas. Analytical Methods (Mukasa and Ludden, 1987; we use 90 Ma for The lack of easily identifi ed marker hori- all samples) and corresponding element concen- zons with the volcanic rocks, lack of distinctive Major-element compositions for 12 samples trations obtained by ICP-MS (Table 2). fi eld characteristics of volcanic units, and the (Table 1) were measured on fused glass beads, enormous extent of the exposures and struc- using X-ray fl uorescence (XRF) spectrometry at Glass Samples tural complexity of the ophiolite necessitate Hirosaki University and Tohoku University. The the application of a chemostratigraphic clas- analytical procedure used has been described in Lavas comprise mostly pillow lavas with sifi cation of the lavas. For Figure 1, we used a detail by Fujimaki and Aoki (1987) and Yajima some massive sheet fl ows, and they are intruded somewhat different division of lavas than previ- et al. (2001). An additional eight samples by dikes and sills (Fig. 2 A1). Mutual crosscut- ously proposed that highlights key geochemical (Table 1) were analyzed for major elements by ting and overlapping relationships show the differences as well as stratigraphic setting, as electron microprobe at Tohoku University; these cogenetic nature of these extrusive and intru- will be explained in the sections on geochemis- samples were not analyzed for trace elements. sive rocks. In spite of the widespread weath- try. The problem with such an approach is that Trace elements (Table 1) were analyzed by ering and local hydrothermal alteration in the the classifi cation depends heavily on geochemi- inductively coupled plasma–mass spectrom- lavas, largely unaltered glasses of these lavas cal sampling density, so it is subject to some etry (ICP-MS) (Yokogawa Analytical Systems and intrusives can be found (Fig. 2 A2, B1, uncertainty, and differences in interpretation HP4500) at the University of the Ryukyus, B2). The glasses are black in outcrop, con- between different researchers are inevitable, with a 115In internal standard. A 50 mg pow- trasting with altered gray-, brown-, red-, and as well as some ambiguity regarding the affi n- dered sample was dissolved in a mixture of HF- green-colored rocks inside the glass margins of

ity of some outcrops. For the purposes of our HClO4, and the evaporated sample was re-dis- extrusives and intrusives. Microscopically, such

study, we believe that uncertainties in location solved in HNO3, as described in Shinjo (1999). glasses are not devitrifi ed (colorless in plane- of chemostratigraphic contacts, or the defi nition The uncertainty in the analyses is generally bet- polarized light and completely extinct in cross- of them, do not signifi cantly impact the main ter than 5% for most trace elements. Calibration polarized light), and contain only clinopyrox- interpretation and conclusions of this study. curves were constructed using rock standard ene microcrystals or microcrystallites (Fig. 2 We divided the lavas into the lower pillow lava JB-1 and a blank solution. In order to evaluate A2), although we found olivine pseudomorphs sequence (suite A; island-arc tholeiite), upper our analytical result, we also analyzed JA-1 and (iddingsite) and pores fi lled by smectite in some pillow lava sequence (suite B; boninite), and the JB-2, standard rock samples provided by the thin sections. Gass et al. (1994) reported simi- infi ll lava (suite C; depleted boninite). Geological Survey of Japan. JB-2 is a recently lar and characteristic microcrystallites in their The dikes of the sheeted complex strike erupted basalt from Oshima Island of the Izu- plate 2.12 (the same photo is in fi g. 11.1-a in N-S, parallel to the paleo–spreading axis, as Bonin arc, and its geochemical data favorably Rogers et al., 1989). Sample CY152 appeared concluded by many authors (e.g., Gass et al., compare with Troodos tholeiitic samples. Devi- exceptionally fresh (Fig. 2 B1, B2), so we ana- 1995). However, three fossil axial valleys on ation of our data from the recommended values lyzed both the glass rim (CY152-f1) and glassy the northern fl ank of the ophiolite may record (Imai, 1990) is mostly less than 5% (Table 1), basalt interior (CY152-f2; Fig. 2 B2), and we discrete eastward ridge jumps, based on their but error in the recommended values themselves also used this sample for Ar-Ar dating. Oxygen

TABLE 2. Sr, Nd, AND Pb ISOTOPIC COMPOSITIONS OF VOLCANIC ROCKS FROM THE TROODOS OPHIOLITE

87 87 ε 143 143 ε 206 207 208 206 207 208 Sample Rock Sr Sr Sr(90Ma) Nd Nd Nd(90Ma) Pb Pb Pb Pb Pb Pb 86 86 144 144 204 204 204 204 204 204 Sr(meas.) Sr(90Ma) Nd(meas.) Nd(90Ma) Pb(meas.) Pb(meas.) Pb(meas.) Pb(90Ma) Pb(90Ma) Pb(90Ma) CY27 Suite A0.703889 0.703719 –9.6 0.513009 0.512876 6.9 18.523 15.560 38.398 18.439 15.556 38.347 CY138 Suite B 0.704197 0.704038 –5.1 0.513035 0.512897 7.3 18.864 15.546 38.653 18.768 15.541 38.591 CY130 Suite B 0.704053 0.703836 –7.9 0.513022 0.512899 7.3 18.889 15.568 38.753 18.779 15.563 38.653 CY119 Suite C 0.707104 0.706199 25.6 0.512828 0.512679 3.1 18.647 15.622 38.692 18.595 15.620 38.667 CY88 Suite C 0.705197 0.704801 5.8 0.513029 0.512859 6.6 18.557 15.584 38.451 18.507 15.582 38.434 Note: Age-corrected Sr, Nd, and Pb isotope ratios and epsilon values were calculated using 90 Ma for Troodos ophiolite (Mukasa and Ludden, 1987). Although suites B and C are expected to be younger, the corrected ages do not change the Figure 6 plot pattern. U, Th, Pb, Sm, Nd, Rb, and Sr concentration values are the same as those in Table 1.

500 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH

Figure 2. Field photos of lava and dikes, and photomicrographs of fresh glass, (A) loc. CY5, and (B) loc. CY152. (A1) pillow lava and an intrusion, (A2) glass was obtained from another dike margin similar to the dike shown (A1), (B1) a fresh pillow, and (B2) glassy basalt, showing the fresh interior of a pillow.

isotope data from the Akaki River section sam- 1975). These samples are boninitic, as detailed Pearce and Robinson (2010) showed that ples show no evidence of alteration in nearly in the following. the calc-alkaline and tholeiitic rock series all of the samples (Rautenschlein et al., 1985). Our calc-alkaline samples have higher MgO classifi cation shown in Figure 3B (Miyashiro Although sample CY144 has a low (90.99) total and Mg# (Mg/[Mg + Fe]; 0.57–0.73) compared diagram) does not characterize the crystalliza- wt% (Table 1) that apparently refl ects altera- with tholeiitic samples and modern Izu-Bonin tion trends exhibited by the Troodos glasses. tion, the glass appeared fresh petrographically, arc volcanic front lavas (Fig. 3C). The calc- They concluded that a better rock series clas- so we used this sample. alkaline samples also have high Cr and Ni abun- sifi cation would class Troodos lavas as tholei- dances (Table 1). Sample CY119 has an excep- itic, depleted tholeiitic (suites A and B), and Rock Series Based on Major-Element tionally high MgO content (Fig. 3C) relative to boninitic (suite C). Composition its silica content. Such high-Mg rocks were also Combined with other geochemical char- analyzed by Robinson and Malpas (1990, see acteristics reviewed herein, we will employ a Major-element chemistry of the samples their fi g. 6). classifi cation that differs somewhat from that of allowed us to classify rock series (Fig. 3). Our The high-Mg calc-alkaline rocks are char- Pearce and Robinson (2010). This classifi cation

samples are subalkaline basalt to andesite in acterized by relatively low TiO2. In MgO-TiO2 consists of a tholeiitic suite A of the lower pil- composition (Fig. 3A). Although two tholeiitic space (Fig. 3D), data are separated into suite A low lava sequence, a boninite suite B or upper samples plot near the low-K/medium-K series (“lower suite” in Robinson et al., 1983; their sam- pillow lava, instead of depleted tholeiite as per boundary, all other samples fall into the low-K ples were more silicic than corresponding rocks Pearce and Robinson (2010), and a suite C of series fi eld. The Troodos glass can be divided of our sample set), suite B (“upper suite” in Rob- depleted boninite that occurs as stratigraphi- into at least two groups: one comprising tholei- inson et al., 1983), and suite C in Robinson and cally high units erupted into paleotopographic itic samples, and the other plotting in the fi eld of Malpas (1990) and Robinson et al. (2003). These lows along the Arakapas fault zone, which we the calc-alkaline rock series (Fig. 3B; Miyashiro, researchers interpreted suite C rocks as boninitic. refer to as infi ll lavas after Gass et al. (1995).

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 501

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

1.5 ABIzu-Bonin VF The classifi cation is based primarily on the FeO/MgO versus SiO2 and TiO2 versus MgO Medium-K 4 concentrations from our data, in which suites TH B and C are similar and highly distinct from 1 suite A. In addition, rare earth element (REE) CA patterns clearly separate the depleted suites B 2 and C from suite A, consistent with previously Low-K K2O (wt %) 0.5 BA FeO*//MgO published REE data on Troodos, particularly Cameron (1985), and, to a lesser extent Rauten- B Chichijima boninite schlein et al. (1985) (Fig. 4). Crawford et al. (1989) proposed a division 0 of boninite suites into two classes, high-Ca 45 50 55 60 50 55 60 65 boninites and low-Ca boninites. The Troodos SiO2 (wt %) SiO2 (wt %) boninites represent a reference suite of high-Ca 20 2 boninites, which have notably lower SiO2, Na2O, C Chichijima D and K2O, and higher CaO and FeO* contents boninite 15 CY119 suite A than low-Ca boninites. High-Ca boninites have also been found at the northern termination of the Tonga Trench and at the southern termination of 10 1 the Vanuatu Trench (Crawford et al., 1989). Some of the boninite from the Bonin Islands is high-Ca TiO2 (wt %) TiO2 MgO (wt %) suite B boninite (e. g., Maehara and Maeda, 2004). 5 Izu-Bonin VF CY119 suite C Chondrite-Normalized REE Patterns 0 0 The island-arc tholeiite (suite A), boninite 50 55 60 65 0 5 10 15 (suite B), and depleted boninite (suite C) show SiO2 (wt %) MgO (wt %) contrasting REE patterns (Fig. 4A). The tholei-

Figure 3. Major-element data. (A) K2O vs. SiO2. Open circle—island-arc tholeiite; solid tri- ites are characterized by depletion of light (L) angle—boninite; open triangle—depleted boninite. Boundaries for low-K and medium- REEs relative to heavy (H) REEs and REE abso- K series are taken from Tatsumi and Eggins (1995). B—basalt, BA—basaltic andesite. lute abundances more than ten times chondrite (B) FeO*/MgO vs. SiO . Boundary line between tholeiitic (TH) and calc-alkaline (CA) series 2 values. The REE patterns of the arc tholeiite sam- is from Miyashiro (1975). (C) MgO vs. SiO2. (D) TiO2 vs. MgO. Data sources: Taylor and Nes- bitt (1998) for Quaternary Izu-Bonin volcanic front (VF) lavas and Taylor et al. (1994) for the ples closely resemble normal mid-ocean-ridge Chichijima boninite. basalt (N-MORB), so the REE pattern cannot discriminate island-arc tholeiite from N-MORB. The Mariana forearc basalt and Oshima basalt 50 50 REE patterns (Fig. 4D) are also comparable to our tholeiitic REE patterns (Fig. 4A). A This study B Cameron (1985) Most of our boninite samples of suites B and suite A suite A Koronia C show a “spoon-shaped” REE pattern, with 10 10 lower overall REE abundance less than ten times suite B suite B chondrite values (Fig. 4A). Our boninite samples

Sample / Chondrite Figure 4. Chondrite-normalized rare earth ele- suite C suite C ment (REE) patterns for the Troodos ophiolite. 1 1 Open circle—island-arc tholeiite; solid triangle— La Ce PrNd SmEuGdTb DyHo ErTmYbLu La CePr Nd SmEuGd TbDy HoErTmYb Lu boninite; open triangle—depleted boninite. Nor- malizing values are from Sun and McDonough 50 50 (1989). (A) Data from this study. (B) Data of Rautenschlein et al. (1985) C D Izu-Bonin Arc Cameron (1985) from the northern fl ank of the Mariana fore arc basalt Troodos ophiolite and from the Arakapas fault suite A zone. (C) Data of Rautenschlein et al. (1985) from Arc tholeiite (Oshima, JB2) 10 10 the Akaki River section in the Troodos ophiol- ite. (D) Representative volcanic rocks from the Hahajima arc basalt modern Izu-Bonin arc. Data sources: Reagan suite B et al. (2010) for Mariana forearc lava, Taylor and Chichijima boninite Nesbitt (1998) for Quaternary volcanic front lava Sample / Chondrite (Oshima arc tholeiite, JB2), Taylor and Nesbitt (1995) for Hahajima arc basalt, and Cameron et 1 1 al. (1983) and Taylor et al. (1994) for Chichijima La Ce PrNd SmEuGdTb DyHo ErTmYbLu La CePr Nd SmEuGd TbDy HoErTmYb Lu boninites.

502 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH

appear to comprise two groups: suite B boninite 50 and suite C depleted boninite, with REE patterns A similar to those reported in Cameron (1985) (Fig. 4B). The depleted boninites (suite C) com- 10 prise fi ve samples, all of which are from the Arakapas transform fault zone, and have lower Ce-Nd contents than suite B boninite. However, suite A HREE concentrations of both types of boninite 1 overlap. The “spoon-shaped” REE pattern is suite B

well developed in the depleted boninites. Suite Sample / N-MORB B samples from the lower pillow lava sequence Figure 5. (A) Mid-ocean-ridge along the Akaki River section (Fig. 1) reported suite C by Rautenschlein et al. (1985) also show a 0.1 basalt (MORB)–normalized trace-element patterns. Nor- slightly “spoon-shaped” REE pattern (Fig. 4C). 0.05 mal-MORB normalizing values Rb Ba Th U Nb Ta K La Ce Pb Nd Sr SmHf Zr Ti Eu Gd Y Yb The “U-shaped” REE pattern (Fig. 4D) of the are from Sun and McDonough Chichijima boninites differs markedly from 50 (1989). (A) Our data from the those of Troodos boninites, as absolute HREE B Troodos ophiolite. (B) Patterns concentrations are lower and MREE/HREE val- for representative rocks from the Izu-Bonin arc. Data sources ues are higher in Chichijima boninites. 10 are the same as in Figure 4D. arc tholeiite (Oshima, JB2) MORB-Normalized Trace-Element Patterns Mariana fore arc basalt

The most distinctive feature in incompatible 1 elements is enrichment of large ion lithophile

elements (LILE; such as Rb, Ba, K, Sr), Th, U, Sample / N-MORB Hahajima and Pb relative to high fi eld strength elements basalts (HFSEs; Nb, Ta, Zr, Hf, Ti, Y), which are shown 0.1 in MORB-normalized multi-element patterns Chichijima boninite (Fig. 5A). In addition, high Pb/Ce and Ba/La 0.05 (LILE/LREE) are evident in the patterns. These Rb Ba Th U Nb Ta K La Ce Pb Nd Sr SmHf Zr Ti Eu Gd Y Yb characteristics are typical features of subduction zone–related magmas, and because these are 143 144 glass analyses, the enrichment is not an artifact “boninite” or suite C “depleted boninite” REE Nd/ Nd(t) between 0.512577 and 0.512816, 87 86 of alteration. patterns (Fig. 4A). Both types show enrichment and higher Sr/ Sr(t) between 0.707155 and Thorium (Th) is a key element, because it is of LILEs, Th, and Pb over HFSEs and LREEs. 0.707519. Thus, Nd and Sr isotopic ratios enriched in arc lavas, but it is immobile during Nb/La values of suite C depleted boninites are clearly distinguish suite C from suites A and B. metamorphism up to the melting temperature comparable to N-MORB and similar to Chichi- Izu-Bonin volcanic front lavas, Haha- (Pearce, 2003) and during seawater alteration, jima boninite (Fig. 5B), whereas Nb/La values jima basalts, and Mariana forearc basalts have contrasting with U, which is mobile in fl uids of suite B boninite are intermediate between 87Sr/86Sr similar to Troodos tholeiites, but they (Hawkesworth et al., 1997). Enriched Th relative suite C depleted boninite and tholeiite. The type have higher 143Nd/144Nd (Fig. 6A). Chichijima to Nb and Ta (Fig. 5A) is indicative of a supra- low-Ca boninites of Chichijima have character- boninite samples have lower Nd and higher Sr subduction-zone setting. istic Zr and Hf enrichment relative to middle (M) isotope ratios compared to relevant arc lavas; Three island-arc tholeiite patterns (suite REEs (Fig. 5B), unlike the Troodos boninites. such higher Sr isotopic ratios are observed in the A) are similar and very closely coincident in Troodos depleted boninites (suite C). HFSEs but show a bit more scatter for LILEs Sr, Nd, and Pb Isotope Ratios In Pb isotope space (Figs. 6C and 6D), our (Fig. 5A). However, every incompatible ele- Troodos tholeiites and boninites plot above the ment is enriched relative to the suite B and C Island-arc tholeiite sample CY27 (suite A) 0 Ma Northern Hemisphere Reference Line ε ε − boninites (Fig. 5A). Most HFSE and REE abun- has Nd(t) of 6.9 and Sr(t) of 9.6 (Fig. 6A; (NHRL). Among our samples, tholeiite sample dances of suite A are very close to N-MORB Table 2). The depleted boninites (suite C) have CY27 (suite A) has the lowest age-corrected ε 87 86 206 204 208 204 207 204 (i.e., normalized value of 1; Fig. 5A). Ratios lower Nd(t) (3.1–6.6) but much higher Sr/ Sr(t) Pb/ Pb and Pb/ Pb, but its Pb/ Pb of Nb (or Ta)/La are lower in Troodos tholeiite 0.704801–0.706199, whereas the boninites ratio is intermediate between boninites (suite than N-MORB values, which is also a diagnos- (suite B) have age-corrected Nd and Sr isotope B) and depleted boninites (suite C). Depleted tic characteristic of arc rocks. Collectively, the ratios comparable to the tholeiites. Our data boninites (suite C) have the highest age-cor- Troodos arc tholeiite patterns closely resemble from suite A tholeiite and suite B boninites are rected 207Pb/204Pb. Pb isotope data of Rauten- typical intra-oceanic-arc tholeiite of the Izu- similar to those reported in Rautenschlein et al. schlein et al. (1985) fall in the range outlined by Bonin arc (Oshima, JB2, and Hahajima basalt) (1985), along the Akaki River section. Rogers et our samples, excluding one sample (Fig. 6D). and that of the Mariana forearc (Fig. 5B). al. (1989) published Nd and Sr isotope analyses Pb isotope ratios for the Cypriot massive Troodos boninite patterns are fairly closely of six boninites including three suite C samples sulfi des are similar to those of fresh Troodos grouped for LILEs, but HFSEs patterns are from the Limassol Forest complex. Rogers et al. glasses reported by Rautenschlein et al. (1985), depleted (Fig. 5A), corresponding to suite B (1989) suite C data are characterized by lower and this suggests that the Pb incorporated in

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 503

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

Mariana fore arc basalt CY27 (suite A) has plateau age of 89.8 ± 0.5 B P-MORB 0.5131 A Izu-Bonin 0.5131 Ma and isochron age of 90.6 ± 1.2 Ma (Fig. 7, volcanic front lava (0 Ma) I-MORB left), and boninitic sample CY152 (suite C) has (t) (t) Hahajima tholeiite (0 Ma) plateau age of 56.7 ± 0.7 Ma and isochron age Nd Nd 0.5130 0.5130 of 55.5 ± 0.9 Ma (Fig. 7, right). Chichijima boninite 144 Samples CY27 and CY152 both exhibit 144 good plateaus in their Ar release spectra (Fig. 7, Nd/ Nd/ 0.5129 0.5129 top). The isochrons are tightly constrained for 143 143 suite C suite C both sample CY27 and CY152 (Fig. 7, bottom), CY119 CY119 with only the 400 and 600 °C steps, plotting sig- 0.5128 0.5128 nifi cantly for sample CY152 off of them (Fig. 7, 40 36 0.7025 0.7045 0.7065 17.5 18.5 19.5 bottom right). The Ar/ Ar intercepts for both samples are close to air, suggesting minimal 87 86 206 204 Sr/ Sr(t) Pb/ Pb(t) excess argon. However, the slightly higher 40 36 15.65 39.5 Ar/ Ar value for sample CY152, coupled C D I-MORB (0 Ma) with the fact that the isochron age (and plateau age) is younger than the total gas age, suggests Xu and Castillo some excess argon in this sample. Accordingly, (t)

(t) (2004) 15.55 38.5 P-MORB we prefer the isochron age for sample CY152. Pb Pb (0 Ma) For sample CY27, the isochron and plateau ages 204 204 are indistinguishable within their uncertainties, Pb/ Pb/ Rautenschlein but we will refer primarily to the isochron age 15.45 This study 37.5 et al. (1985) suite A, tholeiite 208 for this sample in the following discussion. The 207 suite B, boninite 90.6 ± 1.2 Ma Ar-Ar age of the tholeiite, suite suite C, depleted boninite NHRL (0 Ma) A, agrees well with the 90.3 ± 0.7 Ma and 92.4 NHRL (0 Ma) 15.35 36.5 ± 0.7 Ma U-Pb zircon ages of plagiogranites 17.5 18.5 19.5 17.5 18.5 19.5 of the Troodos ophiolite (Mukasa and Ludden, 206 204 206 204 1987). We will further address the signifi cance Pb/ Pb(t) Pb/ Pb(t) of the age data in the discussion section, includ- Figure 6. Isotopic composition of samples from the Troodos ophiolite. Troodos ophiolite data ing providing corroborating stratigraphic data to reported in Rautenschlein et al. (1985) and Xu and Castillo (2004) are also plotted for comparison. support the 55.5 Ma age included in suite C. Izu-Bonin arc data for Chichijima boninite, Hahajima basalts, and Quaternary volcanic front lavas (Taylor et al., 1994; Taylor and Nesbitt, 1995, 1998), Mariana forearc basalt data (Reagan et al., 2010), Sedimentary Cover and Radiolarian Ages and Pacifi c (P-) and Indian (I-) mid-ocean-ridge basalt (MORB) fi elds (Hickey-Vargas et al., 1995) are also shown for comparison. The isotopic compositions for MORB, Izu-Bonin samples, and those 87 86 143 144 Robertson and Hudson (1974) described the of Rautenschlein et al. (1985) are not corrected for age. (A) Plots of Sr/ Sr(t) vs. Nd/ Nd(t), (B) 206 204 143 144 206 204 207 204 206 204 208 204 Pb/ Pb(t) vs. Nd/ Nd(t), (C) Pb/ Pb(t) vs. Pb/ Pb(t), and (D) Pb/ Pb(t) vs. Pb/ Pb(t). sedimentary cover of the Troodos ophiolite, and NHRL (Northern Hemisphere Reference Line at 0 Ma) is from Hart (1984). Robertson (1975) reported umbers from the base of such cover. Umberiferous chert with radio- larians (main part of the Perapedhi Formation) the sulfi des is derived from the Troodos base- GEOCHRONOLOGY AND STRATIGRAPHY were observed at site 44CR on the northeastern ment (Booij et al., 2000), and the above men- fl ank of Troodos, at site 18MR at the Troulli tioned Pb ratios are primary. Our Pb isotope Ar-Ar Ages inlier, at sites 111CR and 125CR to the east of data are quite similar to those of the Izu-Bonin the Arakapas transform fault zone, and at sites boninites and Izu-Bonin volcanic front lavas Ar-Ar analyses were done at National Tai- 132CR, 137CR, and139CR around the south- (Figs. 6C and 6D). On the Pb-Nd isotopic dia- wan University. For full details on analytical ern margin of Limassol Forest complex (Fig. gram (Fig. 6B), all our samples and those of methodology, including irradiation, see Lee et 1). Although there is no exposure of overlying Rautenschlein et al. (1985) plot close to the al. (2009). sediments at site 44CR, the other umberiferous fi eld of Izu-Bonin boninites. We obtained Ar-Ar ages from tholeiitic chert localities are overlain by white mudstone Xu and Castillo (2004) reported Sr-Nd-Pb sample CY27, suite A, and depleted boninitic (included in the Perapedhi Formation), which isotope compositions of gabbro, its mineral sample CY152, suite C (Fig. 1). CY27 was from also contains radiolaria. Umberiferous cherts at separates (plagioclase and clinopyroxene), a fresh glass margin of a basaltic-andesitic sill, 111CR and 125CR overlie volcanic breccia, and and basaltic glass from the Troodos ophiolite. and CY152 was from the exceptionally fresh the others directly overlie pillow lavas. All con- Although their Sr and Nd isotope ratios are glassy basaltic-andesite interior of a pillow. Out- tacts appear conformable and unfaulted. similar to our tholeiites and boninites (Fig. 6A), crop and thin section photos of sample CY152 In addition to the white mudstone, red, green, age-corrected 206Pb/204Pb values of their samples are shown in Figure 2B. Our Ar data for each and white mudstones (also included in the Pera- are lower than our samples, plotting in the fi eld heating step are summarized in Table 3, and the pedhi Formation) conformably overlie pillow of Indian Ocean MORBs (Figs. 6C and 6D). Xu data and statistics associated with the isochrons lavas at the other MR sites (Fig. 1). Some white and Castillo (2004) thus suggested the Tethyan are presented in Table 4. mudstone is tuffaceous, and coarse-grained asthenosphere had the Indian Ocean MORB- Plateau and inverse isochron ages were silicic tuff (the Kannaviou Formation; Rob- type isotopic signature. obtained from both samples. Tholeiitic sample ertson, 1977; Geological Survey Department

504 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH

TABLE 3. DETAILED Ar/Ar DATING RESULTS (TROODOS) T(°C) Cumulation 39Ar(%) 36Ar/39Ar 37Ar/39Ar 38Ar/39Ar 40Ar/39Ar 40Ar/36Ar Date (Ma) 39Ar Atmosphere

CY27 tholeiite 400 0.001 16.025 0.5940E–01 0.1382E+01 0.3621E–03 0.1089E+03 0.1834E+04 600.9 ± 93.9 600 0.007 77.291 0.2014E+0 0.7903E+01 0.3405E–04 0.7627E+02 0.3786E+03 130.8 ± 11.8 700 0.053 56.611 0.6178E–01 0.1603E+02 0.6319E–01 0.3014E+02 0.4879E+03 100.1 ± 2.8 800 0.194 28.236 0.2020E–01 0.1809E+02 0.6628E–01 0.1634E+02 0.8090E+03 90.0 ± 1.2 850 0.379 20.940 0.1509E–01 0.1835E+02 0.6782E–01 0.1472E+02 0.9756E+03 89.4 ± 1.2 900 0.522 20.444 0.1477E–01 0.1837E+02 0.6585E–01 0.1461E+02 0.9890E+03 89.2 ± 1.3 950 0.648 21.752 0.1577E–01 0.1850E+02 0.6505E–01 0.1504E+02 0.9540E+03 90.4 ± 1.2 1000 0.745 30.203 0.2174E–01 0.1869E+02 0.6408E–01 0.1663E+02 0.7651E+03 89.2 ± 1.3 1050 0.854 33.263 0.2463E–01 0.1986E+02 0.6407E–01 0.1741E+02 0.7069E+03 89.3 ± 1.3 1100 0.915 47.000 0.4046E–01 0.2062E+02 0.6318E–01 0.2216E+02 0.5477E+03 90.3 ± 2.3 1150 0.950 65.130 0.8159E–01 0.2098E+02 0.6098E–01 0.3462E+02 0.4243E+03 92.8 ± 2.8 1200 0.967 84.248 0.2439E+00 0.2090E+02 0.7088E–01 0.8369E+02 0.3432E+03 101.2 ± 5.2 1250 0.979 96.862 0.1234E+01 0.2089E+02 0.2244E+00 0.3750E+03 0.3038E+03 90.6 ± 7.7 1300 0.990 98.080 0.2059E+01 0.2122E+02 0.3624E+00 0.6186E+03 0.3005E+03 91.5 ± 9.1 1450 1.000 98.370 0.3021E+01 0.2061E+02 0.5302E+00 0.9060E+03 0.2999E+03 113.0 ± 10.3 CY152 boninite 400 0.001 55.109 0.2773E+00 0.7578E+00 0.2271E–03 0.1486E+03 0.5359E+03 456.9 ± 41.3 600 0.046 76.938 0.1745E+00 0.7749E+01 0.3813E–01 0.6630E+02 0.3799E+03 115.9 ± 2.2 700 0.142 83.641 0.1463E+00 0.2677E+02 0.3794E–01 0.4929E+02 0.3370E+03 62.8 ± 1.9 800 0.288 75.392 0.7513E–01 0.1621E+02 0.2783E–01 0.2786E+02 0.3708E+03 53.1 ± 1.6 850 0.438 70.318 0.6118E–01 0.1456E+02 0.2699E–01 0.2418E+02 0.3952E+03 55.5 ± 1.1 900 0.560 70.815 0.6424E–01 0.1422E+02 0.2860E–01 0.2532E+02 0.3942E+03 57.1 ± 1.1 950 0.647 76.152 0.8402E–01 0.1452E+02 0.2947E–01 0.3120E+02 0.3713E+03 57.5 ± 2.6 1000 0.717 78.399 0.9431E–01 0.1398E+02 0.3251E–01 0.3423E+02 0.3630E+03 57.2 ± 1.4 1050 0.770 79.832 0.1032E+00 0.1850E+02 0.3168E–01 0.3649E+02 0.3535E+03 57.1 ± 1.9 1100 0.836 74.415 0.8232E–01 0.4098E+02 0.2635E–01 0.2857E+02 0.3470E+03 57.5 ± 4.3 1150 0.884 76.948 0.1111E+00 0.7067E+02 0.2822E–01 0.3579E+02 0.3220E+03 66.1 ± 4.7 1200 0.924 87.437 0.2094E+00 0.9071E+02 0.4322E–01 0.6298E+02 0.3008E+03 64.3 ± 8.3 1250 0.957 95.756 0.6123E+00 0.1059E+03 0.1083E+00 0.1807E+03 0.2950E+03 63.0 ± 8.2 1300 0.982 97.103 0.1223E+01 0.1164E+03 0.2149E+00 0.3633E+03 0.2970E+03 86.6 ± 8.4 1450 1.000 98.504 0.3011E+01 0.1032E+03 0.5310E+00 0.8953E+03 0.2974E+03 108.6 ± 8.8 Note: For sample CY27: sample mass = 400.0 mg; J-value = 0.004319107 ± 0.000005036; integrated date = 91.3 ± 0.5 Ma; 39Ar volume = 0.2575E–10 ccSTP/g; 40Ar* volume = 0.3094E–09 ccSTP/g. For sample CY152: sample mass = 400.0 mg; J-value = 0.004319107 ± 0.000005036; integrated date = 62.7 ± 0.8 Ma; 39Ar volume = 0.2003E–10 ccSTP/g; 40Ar* volume = 0.1640E–09 ccSTP/g. ccSTS/g—cubic centimeters of gas at standard temperature (0 degrees C) and pressure (1 atm) per gram of water.

Cyprus, 1995) is distributed along the southwest- stone, contained Amphipyndax tylotus, which is A tholeiitic rocks and suite B boninitic rocks ern margin of the Troodos main massif. These an index fossil of the late Campanian to Maas- crop out primarily along the northern fl anks of mudstones, also intercalated with thick silicic trichtian (ca. 75–66 Ma) (Sanfi lippo and Riedel, the Troodos massif, whereas suite C depleted tuff, contain well-preserved radiolarians (e.g., 1985; Osozawa and Okamura, 1993). Other boninites appear restricted to an infi ll geometry Osozawa and Okamura, 1993). radiolarians are A. pseudoconulus, which is an along the Arakapas fault zone. We note that the The umberiferous chert at the Kalavasos index fossil of the early Campanian but coexists complex chemostratigraphic relationships at mine yielded a Turonian (ca. 89–94 Ma) radio- with A. tylotus in samples from 111CR, 125CR, this large ophiolite can lead to differing inter- larian age (near the sample point of 137CR; and 44CR, and Dictyomitra koslovae, Myllocer- pretations, but there is agreement that tholeiite Blome and Irwin, 1985; Pseudodictyomitra cion acineton, and others. Considering the coex- eruption preceded boninite eruption at Troodos, pseudomacrocephala is the key index fossil; istence of A. tylotus and A. pseudoconulus, all and this relationship and associated details will Sanfi lippo and Riedel, 1985), but according to cover sediments we analyzed can be considered form a key part of our analysis of the tectono- Urquhart and Robertson (2000), their sample late Campanian (ca. 75 Ma; Yamasaki, 1987). magmatic evolution of this ophiolite. was not from the ophiolite but from the Moni mélange. The radiolarian assemblage collected DISCUSSION Chronology of Troodos Ophiolite Formation from umberiferous chert at Perapedhi east of 47MR (Bragina and Bragin, 1996) also con- Spatial Distribution of Chemostratigraphic Here, we discuss our new geochronologic tains old species, Theocorys antique, which was Types and Order of Eruption data in relation to our geochemical data and in extinct at the Santonian and Campanian bound- the context of the generation, evolution, and ary (ca. 83.5 Ma) (Sanfi lippo and Riedel, 1985), Our assessment of the chemostratigraphy, emplacement of the ophiolite. Whereas the 90.6 and Amphipyndax pseudoconulus. However, all described in the geochemistry sections, com- ± 1.2 Ma Ar-Ar date from tholeiite glass (suite the radiolarian samples we analyzed, including bines our data with previously published data A) is comparable to U-Pb zircon ages of pla- the umberiferous chert underlying white mud- sets to arrive at the conclusion that our suite giogranites by Mukasa and Ludden (1987), the

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 505

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

TABLE 4. INTERCEPT AND AGE DATA (TROODOS) because of late Campanian (ca. 75 Ma) radio- this, and voluminous boninite eruption may larian ages obtained from the base of this unit. have continued until ca. 75 Ma, after which CY27 tholeiite, suite A This contradicts the conformable nature of the minor boninitic magmatism continued until at Points used to fi t the line= contact of the cover over both suites B and C. A least 55 Ma. 3 4 5 6 7 8 9 10 11 12 13 14 15 Slope = –0.40274E–01 ± 0.25455E–03 reasonable interpretation of these relationships Modifi ed error = 0.48340E–03 is that fairly boninitic volcanism persisted until Metamorphic Sole of the Troodos Y intercept = 0.33764E–02 ± 0.38306E–05 ca. 75 Ma, although additional Ar-Ar dating of Ophiolite? Does it Constrain Subduction Modifi ed error = 0.72745E–05 suite B and additional suite C boninite samples Initiation There? Reciprocal of Y intercept = 0.29617E+03 is desired for confi rmation. The occurrence of ± 0.33601E+00 Modifi ed error = 0.63809E+00 siliceous tuff intercalated with the umbers sug- Metamorphic soles, i.e., thin sheets of high- X intercept = 0.83836E–01 ± 0.47349E–03 gests some volcanism persisting to ca. 75 Ma. grade, mostly mafi c, metamorphic rocks, crop Modifi ed error = 0.10937E–02 This, however, does not require that boninitic out beneath many ophiolites and are interpreted Reciprocal of X intercept = 0.11928E+02 eruption continued until then, for signifi cant as the product of subduction initiation in young ± 0.67367E–01 gaps in age between siliceous volcanism and (hot) oceanic lithosphere (Williams and Smyth, Modifi ed error = 0.15561E+00 MSWD (mean square weighted deviation) = 1.9 earlier main-stage ophiolite formation are doc- 1973; Spray, 1984; Jamieson, 1986; Hacker, Xbar = 0.44085E–02 umented in other ophiolites, such as the Coast 1990). As such, analysis of the high-grade Ybar = 0.31989E–02 Range ophiolite of California, where siliceous metamorphic rocks associated with the Troo- Age = 90.6 ± 1.2 Ma volcanism may have occurred >20 m.y. after dos ophiolite has some relevance in evaluating CY152 depleted boninite, suite C main-stage ophiolite development (e.g., Hopson possible connections between the formation Points used to fi t the line= et al., 2008). Whereas these relationships sug- and/or emplacement of the Troodos ophiolite 3 4 5 6 7 8 9 10 11 12 13 14 15 gest the possibility of Troodos volcanism lasing and subduction initiation. The Troodos ophio- Slope = –0.24257E–01 ± 0.28341E–03 until ca. 75 Ma, they do not explain the 55 Ma lite lacks a coherent metamorphic sole, but Modifi ed error = 0.35979E–03 boninite age. We address this issue next. scattered outcrops of high-grade metamorphic Y intercept = 0.33561E–02 ± 0.39628E–05 Whereas the Perapedhi Formation conform- rocks have been interpreted as a dismembered Modifi ed error = 0.50308E–05 Reciprocal of Y intercept = 0.29796E+03 ably overlies much of the suite B and C volca- sole (Malpas et al., 1992; Chan et al., 2007). ± 0.35183E+00 nic rocks, as noted already, the younger Lefkara Controversy exists over whether these rocks Modifi ed error = 0.44665E+00 Formation overlies volcanic rocks in the vicin- represent a dismembered metamorphic sole or X intercept = 0.13836E+00 ± 0.15450E–02 ity of the 55 Ma boninite sample. The Lefkara whether they formed in a different tectonic envi- Modifi ed error = 0.23005E–02 Formation (Robertson and Hudson, 1974) is a ronment, such as oceanic transform fault (Spray Reciprocal of X intercept = 0.72276E+01 ± 0.80706E–01 thickly bedded chalk unit that nearly encircles and Roddick, 1981; Malpas et al., 1992; Chan et Modifi ed error = 0.12017E+00 the Troodos ophiolite in map distribution (Fig. al., 2007). The occurrence of metasedimentary MSWD (mean square weighted deviation) = 1.3 1). If the base of the Lefkara Formation is Maas- rocks, as well as the more common metabasites, Xbar = 0.70343E–02 trichtian, as interpreted by Gass et al. (1994), this may suggest a metamorphic sole origin, rather Ybar = 0.31855E–02 would indicate that our 55 Ma boninite age is than a deep fracture zone or ocean core com- Age = 55.5 ± 0.9 Ma erroneous. Recent biostratigraphic work on the plex, because the latter settings would not result Lefkara Formation shows that the Maastrichtian in suffi cient burial of metasediments to produce strata are either lacking or patchy in distribution. amphibolites-grade metamorphism in them. 55.5 ± 0.7 Ma age for boninitic sample CY152 The age for most of the Lefkara Formation is The 40Ar/39Ar ages of hornblendes from (included in suite C) requires additional expla- middle Paleogene to early Miocene (Kähler and these amphibolites, interpreted as Troodos nation, as does the difference between the ca. Stow, 1998). Strata as old as late Paleocene are metamorphic sole remnants, span 14 m.y. (ca. 91 Ma ages given previously and the ca. 75 Ma found only in the Ayios Nocolaos area (Fig. 1), 90 and 80 Ma for northeastern coast of the age of the basal cover strata overlying much of along the southwestern margin of the Troodos Akamas Peninsula, and 76 Ma for two samples the ophiolite. main massif (far from sample 152), whereas the near Pafos; Fig. 1), which is an unusually long The tholeiitic lower pillow lava sequence (suite bases of other sections are early Eocene (Kähler age range for metamorphic soles (Chan et al., A) has a genetic link with a brittlely deformed and Stow, 1998). Thus, the revised Lefkara For- 2007). This spread of ages suggests multiple early suite of plutonic rocks and sheeted dikes mation biostratigraphy indicates that its base is alternatives, including the following. (1) The (therefore spreading related), and the boninitic mostly younger than the 55 Ma boninite. This high-grade rocks did not form in a metamor- upper pillow lava sequence (suite B) has a genetic fact, coupled with the data indicating a robust phic sole environment and were juxtaposed link with an undeformed later suite of cumulate age (see geochronology sections), suggests that with the ophiolite by subsequent tectonic and (e.g., Dilek et al., 1990). Because plagiogranite the age is valid. possibly sedimentary processes (if olistos- belongs to the early suite of plutonic complex, Although the Ar-data and stratigraphic data trome, blocks in mélange for example). (2) The sheeted dikes, and lower pillow lavas (Dilek and suggest the 55 Ma boninite age is valid, the amphibolite samples include rocks formed in a Thy, 2009), the agreement between the tholeiite widespread distribution of Campanian ages of metamorphic sole environment as well as rocks (suite A) age and the plagiogranite ages indicates basal cover strata (Fig. 1) suggests that post- formed in other environments, such as deep a reasonable volcanic date. Campanian mafi c volcanism is rare, even for the along oceanic fracture zones. (3) The amphibo- In contrast, a hypothetical ca. 91 Ma age for infi ll unit suite C. The collective geochemical, lite formed in a metamorphic sole environment boninite (suite B) and depleted boninite (suite C) stratigraphic, and geochronologic data suggest but in an environment associated with slow would require a major unconformity or tectonic that Troodos ophiolite formation began with arc subduction initiation. The latter has been pro- contact between lavas of suites B and C and their tholeiite magmatism at ca. 91 Ma. The onset of posed to explain the ~15 m.y. range of high- cover sediments (the Perapedhi Formation), boninitic magmatism occurred sometime after temperature–high-pressure metamorphic ages

506 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH

Sample CY27 tholeiite (suite A) Sample CY152 depleted boninite (suite C) 200 200

Step 4 (800°C)-11 (1150°C) 150 150 Step 4 (800°C)-11 (1150°C) ± Plateau date = 89.8 0.5 Ma Plateau date = 56.7 ± 0.7 Ma

100 100

50 50 Apparent Age (Ma) Apparent Age (Ma)

0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 39 Fraction of 39 Ar released Fraction of Ar released 0.004 0.004 1450 400 - 1450°C 1450 1300 ± 1300 1250 Intercept date = 90.6 1.2 Ma 40 40 36 40 1200 1200 ( Ar/ Ar) = 296.2 1250 700 1050 Ar/ Ar Ar/ Mean square Ar Ar/ 950 800 36 36 1150 900 600 1000 600 weighted deviation = 1.9 1100 1150 400 850 0.002 0.002 700 1100

1000 400 - 1450 °C 1050 0.001 0.001 Intercept date = 55.5 ± 0.9 Ma 800 850 40 36 950 900 ( Ar/ Ar) = 298.0 400 Mean square weighted deviation = 1.3

0 0 39 40 39 0 0.04Ar/ Ar 0.08 0.12 0 0.02Ar/ 40 Ar 0.04 0.06

Figure 7. Age spectra of Ar-Ar data (top) and isochron plots (bottom) of sample CY27 tholeiite (left) and sample CY152 boninite (right).

associated with high-grade metamorphic rocks (ca. 91–75 Ma). This similarity between igne- Shervais, 2001; Dilek and Thy, 2009). Whereas of the Franciscan Complex, which may repre- ous formation ages of the ophiolite and meta- the eruptive sequence of tholeiite before boninite sent a dismembered metamorphic sole formed morphic ages of the amphibolite mirrors the characterizes the Troodos ophiolite as well, at during subduction initiation (Anczkiewicz et near-synchronous ophiolite formation and met- least one well-preserved suprasubduction-zone al., 2004). That conclusion is consistent with amorphic sole metamorphism ages that appear ophiolite, the Betts Cove ophiolite of Newfound- the results of thermal modeling (Cooper et al., to characterize many suprasubduction-zone land, has an eruptive sequence of boninite before 2011). (4) Amphibolite formed in a subduction ophiolites (Wakabayashi and Dilek, 2000, 2003; tholeiite (Bédard et al., 1998), as does the Izu- zone beneath the ophiolite over an extended Wakabayashi et al., 2010), although Troodos is Bonin arc itself (e.g., Ishizuka et al., 2006). period of time owing to the close approach unusual, or perhaps unique, for the long duration Based on our new geochronologic and bio- of another spreading ridge on the downgoing of both amphibolites-grade metamorphism and stratigraphic data, the Troodos ophiolite may plate (Aoya et al., 2002). The position of ridge the arc magmatism, possibly as a consequence represent a much longer record of suprasubduc- crest segments on the downgoing plate coupled of slow initial subduction rate and/or the stag- tion-zone magmatism than the <10 m.y. (usu- with relative plate motions (nearly parallel to gered subduction of multiple ridge segments. ally <5 m.y.) recorded in most suprasubduction- fracture zones) may have led to a series of zone ophiolites (e.g., Shervais, 2001; Tremblay ridge crest–subduction zone interactions that Similarities and Differences among et al., 2011). Voluminous magmatism appears may have spanned a signifi cant amount of time Troodos, Other Suprasubduction-Zone to have persisted at Troodos from ca. 91 Ma to (Wakabayashi, 2004). The last scenario may Ophiolites, and Proposed Modern Analogs 75 Ma, with some boninitic magmatism taking best explain the generation of boninites over an place as late as 55 Ma. Although the relation- extended period of time, given that such rocks Many models of suprasubduction-zone ophi- ship, and even existence of a metamorphic sole require high temperatures of formation. olite generation propose generation of the ophio- for the Troodos ophiolite is somewhat contro- The ages derived from the amphibolites spa- lite shortly after subduction initiation, with the versial, metamorphic ages and metamorphic tially associated with the Troodos ophiolite are ophiolite forming during slab rollback and erup- relationships are permissive of high-tempera- identical to the age range we have interpreted tion of refractory boninites following initial arc ture metamorphism that spanned approximately for most of the volcanic rocks of the ophiolite tholeiite eruption (e.g., Stern and Bloomer, 1992; the same period of time as the magmatism.

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 507

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

This is also unusual for metamorphic soles, MAP VIEW SCHEMATIC DIAGRAMS which commonly have a limited age range (e.g., Hacker, 1991; Wakabayashi and Dilek, 2000). A. Ca. 91 Ma: Early stages B. <91 Ma: After early stage of These relationships illustrate a signifi cant of Troodos ophiolite generation SSZ crust formation; approach degree of variability in basic characteristics of pre-Troodos oceanic crust of spreading ridge. suprasubduction-zone ophiolites and suggest Early formation of SSZ crust that the formational history of such ophiolites above slab that rolls back oldest part of Troodos ophiolite as well as their subsequent emplacement may vary more than commonly surmised. Bédard pre-Troodos oceanic crust

plate convergence et al. (1998) raised this point for ophiolites in a direction approach of spreading ridge single orogenic belt by pointing out differences on downgoing plate in the eruptive order of boninite before tholei- ite in the Betts Cove ophiolite versus tholeiite before boninite in the Thetford Mines and Bay of Islands ophiolites of the Northern Appala- chians. Pearce (2003) also presented a much broader range of modern seafl oor possible ana- C. <91 Ma to 75 Ma (or 55 Ma): logs of suprasubduction-zone ophiolites than is Long period of ridge commonly considered by ophiolite geologists. subduction The unusually long duration of arc magma- Figure 8. Map view schematic dia- tism without development of an arc edifi ce in boninites depleted boninites grams showing a proposed model for the Troodos ophiolite may require a somewhat form early in ridge subduction form later in extended duration formation of the Troodos ophiolite. event ridge subduction, lavas fill (A) Subduction initiates and, as the different tectonic model than that proposed for topographic lows on sea floor slab rolls back, extension above the other suprasubduction-zone ophiolites. Next, slab results in upwelling of hot man- we present a brief, speculative model that may tle with partial melting of this mantle explain the unusual features of the Troodos fl uxed by fl uids from the downgo- ophiolite (Fig. 8). ing slab. This magmatism forms the Adopting other models of suprasubduction- older part of the Troodos ophiolite in the upper plate. (B) A spreading ridge zone ophiolite generation, we propose as oth- alternative (gray lines): spreading ridge axis approaches the trench on the down- staggered series of ridge ers have, that the Troodos ophiolite formed subparallel to going plate. (C) A prolonged period of segments with overlap following subduction initiation with subduc- convergence subparallel to convergence ridge subduction generates boninitic direction tion rollback and creation of fi rst arc tholeiite direction magmatism in the Troodos ophiolite (double black lines) and then boninite during subduction rollback, as a result of higher temperatures extension above the subducting plate, and an imposed on previously depleted increasing degree of depletion of the mantle mantle above. Such a prolonged ridge subduction event may result wedge in the zone of partial melting above the from subduction of a ridge with an sinking slab (e.g., Stern and Bloomer, 1992; axis subparallel to the convergence Shervais, 2001; Dilek and Flower, 2003; Trem- direction or subduction of a series of blay et al., 2011). Whether subduction initia- ridge segments with the overlap sub- tion was spontaneous or induced (Stern, 2004) parallel to the convergence direction. cannot be determined with existing data, and it SSZ—suprasubduction zone. is also unclear whether subduction initiated in young or old oceanic lithosphere. The extended history of boninite eruption exhibiting progressively greater depletion, as well as the somewhat fragmentary evidence subparallel to the plate convergence vector or P. Robinson, and J. Bédard for helpful com- of long-lived, high-grade metamorphism, sug- (2) the subduction of a series transform-offset ments. We thank H. Fujimaki, Tohoku Univer- gests the arrival of an additional heat source, ridge segments wherein the convergence vector sity, and K. Uto, AIST, for their assistance at probably a spreading ridge, on the downgoing leads to subduction of the ridge segments along analytical facilities. Y. Kato, University of the plate. Such a scenario resembles that proposed approximately the same reach of the trench (Fig. Ryukyus, is also appreciated for his guidance. by Shervais (2001), except that we propose that 8C; e.g., Wakabayashi, 2004). Measurement of paleomagnetism was done by suprasubduction-zone oceanic crust produc- T. Koitabashi. Financial support was provided tion did not end with ridge subduction. Instead, ACKNOWLEDGMENTS by the Overseas Research Fund offered by the ridge subduction promoted a further partial Ministry of Education, Culture, Sports, Science melting of already depleted mantle, producing Scientifi c support was provided by the Cyprus and Technology, Japan, in 1997. boninite, and then depleted boninite. The ridge Geological Survey, and accommodation was subduction event was a prolonged one, resulting provided by the Cyprus American Archaeologi- REFERENCES CITED in an extended period of boninitic magmatism. cal Research Institute, Nicosia, and Axiothea Allerton, S., and Vine, F.J., 1991, Spreading structure of the Troo- Such a prolonged ridge subduction event can Hotel, Paphos, Cyprus, all in 1997 (for the dos ophiolite, Cyprus: Geology, v. 19, p. 637–640, doi: result from (1) the subducting ridge axis being fi rst author). We thank R. Russo, J. Shervais, 10.1130/0091-7613(1991)019<0637:SEOTTO>2.3.CO;2.

508 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 Geochemistry and geochronology of the Troodos ophiolite | RESEARCH

Anczkiewicz, R., Platt, J.P., Thirlwall, M.F., and Wakabayashi, Hacker, B.R., 1990, Simulation of the metamorphic and Mukasa, S.B., and Ludden, J.N., 1987, Uranium-lead isoto- J., 2004, Franciscan subduction off to a slow start: Evi- deformational history of the metamorphic sole of the pic ages of plagiogranites from the Troodos ophiolite, dence from high-precision Lu-Hf garnet ages on high Oman ophiolite: Journal of Geophysical Research, v. 95, Cyprus, and their tectonic signifi cance: Geology, v. 15, grade-blocks: Earth and Planetary Science Letters, p. 4895–4907, doi:10.1029/JB095iB04p04895. p. 825–828, doi:10.1130/0091-7613(1987)15<825:UIAOPF v. 225, p. 147–161, doi:10.1016/j.epsl.2004.06.003. Hacker, B.R., 1991, The role of deformation in the formation >2.0.CO;2. Aoya, M., Uehara, S., and Wallis, S., 2002, Thermal conse- of metamorphic fi eld gradients: Ridge subduction Murton, B.J., 1989, Tectonic controls on boninite genesis, in quences of a subduction boundary jump: A numeri- beneath the Oman ophiolite: Tectonics, v. 10, p. 455– Saunder, A.D., and Norry, M.J., eds., Magmatism in the cal model for generating subduction-related clockwise 473, doi:10.1029/90TC02779. Ocean Basins: Geological Society of London Special pressure-temperature paths: Tectonics, v. 21, p. 4-1–4-17. Hart, S.R., 1984, A large-scale isotope anomaly in the South- Publication 42, p. 347–377. Bédard, J.H., Lauziére, K., Tremblay, A., and Sangster, A., ern Hemisphere mantle: Nature, v. 309, p. 753–757, Nicolas, A., and Boudier, F., 2003, Where ophiolites come 1998, Evidence for forearc seafl oor-spreading from the doi:10.1038/309753a0. from and what they tell us, in Dilek, Y., and Newcomb, Betts Cove ophiolite, Newfoundland: Oceanic crust of Hawkesworth, C., Turner, S., Peate, D., McDermott, F., and S., eds., Ophiolite Concept and the Evolution of Geo- boninitic affi nity: Tectonophysics, v. 284, p. 233–245, van Calsteren, P., 1997, Elemental U and Th variations logic Thought: Geological Society of America Special doi:10.1016/S0040-1951(97)00182-0. in island arc rocks: Implications for U-series isotopes: Paper 373, p. 137–152. Blome, C.D., and Irwin, W.P., 1985, Equivalent radiolarian ages Chemical Geology, v. 139, p. 207–221, doi:10.1016/S0009 Osozawa, S., and Okamura, M., 1993, New radiolarian ages from ophiolitic terranes of Cyprus and Oman: Geol- -2541(97)00036-3. from theTroodos ophiolite and their tectonic implications: ogy, v. 13, p. 401–404, doi:10.1130/0091-7613(1985)13 Hickey-Vargas, R., Hergt, J.M., and Spadea, P., 1995, The The Island Arc, v. 2, p. 1–6, doi:10.1111/j.1440-1738.1993 <401:ERAFOT>2.0.CO;2. Indian Ocean–type isotopic signature in western Pacifi c .tb00070.x. Booij, E., Bettison-Varga, L., Farthing, D., and Staudigel, H., marginal basins: Origin and signifi cance, in Taylor, B., Pearce, J.A., 2003, Supra-subduction zone ophiolites: The 2000, Pb-isotope systematics of a fossil hydrothermal and Natland, J., eds., Active Margins and Marginal search for modern analogues, in Dilek, Y., and New- system from the Troodos ophiolite, Cyprus: Evidence Basins of the Western Pacifi c: American Geophysical comb, S., eds., Ophiolite Concept and the Evolution of for a polyphased alteration history: Geochimica et Union Geophysical Monograph 88, p. 175–197. Geologic Thought: Geological Society of America Spe- Cosmochimica Acta, v. 64, p. 3559–3569, doi:10.1016 Hoang, N., and Uto, K., 2006, Upper mantle isotopic com- cial Paper 373, p. 269–293. /S0016-7037(00)00449-X. ponents beneath the Ryukyu arc system: Evidence for Pearce, J.A., and Robinson, P.T., 2010, The Troodos ophiolitic Bragina, L.G., and Bragin, N.Y., 1996, Stratigraphy and radio- ‘back-arc’ entrapment of Pacifi c MORB mantle: Earth complex probably formed in a subduction initiation, larians from the type section of Pelapedhi Formation and Planetary Science Letters, v. 249, p. 229–240, slab edge setting: Gondwana Research, v. 18, p. 60–81, (Upper Cretaceous of Cyprus): Stratigraphy and Geo- doi:10.1016/j.epsl.2006.07.021. doi:10.1016/j.gr.2009.12.003. logical Correlation, v. 4, p. 246–253. Hopson, C.A., Mattinson, J.M., Pessagno, E.A., Jr., and Luy- Rautenschlein, M., Jenner, G.A., Hertogen, J., Hofmann, Cameron, W.E., 1985, Petrology and origin of primitive endyk, B.P., 2008, California Coast Range ophiolite: A.W., Kerrich, R., Schmincke, H.-U., and White, W.M., lavas from the Troodos ophiolite, Cyprus: Contribu- Composite Middle and Late Jurassic oceanic litho- 1985, Isotope and trace element composition of volca- tions to Mineralogy and Petrology, v. 89, p. 239–255, sphere, in Wright, J.E., and Shervais, J.W., eds., Ophiol- nic glasses from the Akaki Canyon, Cyprus: Implications doi:10.1007/BF00379457. ites, Arcs, and Batholiths: A Tribute to Cliff Hopson: Geo- for the origin of the Troodos ophiolite: Earth and Plane- Cameron, W.E., McCulloch, M.T., and Walker, D.A., 1983, logical Society of America Special Paper 438, p. 1–101, tary Science Letters, v. 75, p. 369–383, doi:10.1016/0012- Boninite petrogenesis: Chemical and Nd-Sr isotopic doi:10.1130/2008.2438(01). 821X(85)90180-3. constraint: Earth and Planetary Science Letters, v. 65, Imai, N., 1990, Multielement analysis of rocks with the use Reagan, M.K., Ishizuka, O., Stern, R.J., Kelley, K.A., Ohara, Y., p. 75–89. of geological certifi ed reference material by inductively Blichert-Toft, J., Bloomer, S.H., Cash, J., Fryer, P., Hanan, Chan, G.H.-N., Malpas, J., Xenophontos, C., and Lo, C.-H., coupled plasma mass spectrometry: Analytical Sci- B.B., Hickey-Vargas, R., Ishii, T., Kimura, J., Peate, D.W., 2007, Timing of subduction zone metamorphism during ences, v. 6, p. 389–395, doi:10.2116/analsci.6.389. Rowe, M.C., and Woods, M., 2010, Fore- arc basalts and the formation and emplacement of Troodos and Baër- Ishizuka, O., Kimura, J., Li, Y.B., Stern, R.J., Reagan, M.K., Tay- subduction initiation in the Izu–Bonin–Mariana system: Bassit ophiolites: Insights from 40Ar-39Ar geochronology: lor, R.N., Ohara, Y., Bloomer, S.H., Ishii, T., Hargrove, U.S., Geochemistry Geophysics Geosystems, v. 11, Q03X12, Geological Magazine, v. 144, p. 797–810, doi:10.1017 III, and Haraguchi, S., 2006, Early stages in the evolution doi:10.1029/2009GC002871. /S0016756807003792. of Izu-Bonin arc volcanism: New age, chemical, and iso- Robertson, A.H.F., 1975, Cyprus umbers: Basalt-sediment Cooper, F.J., Platt, J.P., and Anezkiewicz, R., 2011, Constraint topic constraints: Earth and Planetary Science Letters, relationships on a Mesozoic ocean ridge: Journal of on early Franciscan subduction rates from 2-D thermal v. 250, p. 385–401, doi:10.1016/j.epsl.2006.08.007. the Geological Society of London, v. 131, p. 511–531, modeling: Earth and Planetary Science Letters, v. 312, Jamieson, R.A., 1986, P-T paths from high temperature doi:10.1144/gsjgs.131.5.0511. p. 69–79, doi:10.1016/j.epsl.2011.09.051. shear zones beneath ophiolites: Journal of Metamor- Robertson, A.H.F., 1977, The Kannaviou Formation, Cyprus: Crawford, A.J., Falloon, T.J., and Green, D.H., 1989, Classifi ca- phic Geology, v. 4, p. 3–22, doi:10.1111/j.1525-1314.1986 Volcaniclastic sedimentation of a probable Late Creta- tion, petrogenesis and tectonic setting of boninites, in .tb00335.x. ceous volcanic arc: Journal of the Geological Society of Crawford, A.J., ed., Boninites: London, Unwin Hyman Kähler, G., and Stow, D.A.V., 1998, Turbidites and contourites London, v. 134, p. 269–292, doi:10.1144/gsjgs.134.3.0269. Ltd., p. 1–49. of the Palaeogene Lefkara Formation, southern Cyprus: Robertson, A.H.F., and Hudson, J.D., 1974, Pelagic sediments Dilek, Y., 2003, Ophiolite concept and its evolution, in Dilek, Sedimentary Geology, v. 115, p. 215–231, doi:10.1016/ in the Cretaceous and Tertiary history of the Troodos Y., and Newcomb, S., eds., Ophiolite Concept and the S0037-0738(97)00094-8. massif, Cyprus, in Hsü, K.J., and Jenkyns, H.C., eds., Evolution of Geologic Thought: Geological Society of Lee, H.Y., Chung, S.-L., Lo, C.-H., Ji, J., Lee, T.-Y., Qian, Q., Pelagic Sediments: On Land and under the Sea: Inter- America Special Paper 373, p. 1–16. and Zhang, Q., 2009, Eocene Neotethyan slab breakoff national Association Sedimentologists Special Publica- Dilek, Y., and Flower, M.F.J., 2003, Arc-trench rollback and in southern Tibet inferred from the Linzizong volcanic tion 1, p. 403–436. forearc accretion: 2. A model template for ophiolites in record: Tectonophysics, v. 477, p. 20–35. Robinson, P.T., and Malpas, J., 1990, The Troodos ophiolite of Albania, Cyprus, and Oman, in Dilek, Y., and Robinson, Maehara, K., and Maeda, J., 2004, Evidence for high-Ca Cyprus: New perspectives on its origin and emplace- P.T., eds., Ophiolites in Earth History: Geological Society boninite magmatism from Paleogene primitive low-K ment, in Marpas, J., Moores, E.M., Panayiotou, A., and of London Special Publication 218, p. 43–68. tholeiite, Mukoojima, Hahajima Island group, southern Xenophontos, C., eds., Ophiolites, Oceanic Crustal Ana- Dilek, Y., and Thy, P., 2009, Island arc tholeiite to boninitic melt Bonin (Ogasawara) forearc, Japan: The Island Arc, v. 13, logues: Proceedings of the Symposium “Troodos 1987”: evolution of the Cretaceous Kizildag (Turkey) ophiolite: p. 452–465, doi:10.1111/j.1440-1738.2004.00437.x. Nicosia, Cyprus, The Geological Survey Department, Model for multi-stage early arc–forearc magmatism in Malpas, J., Xenophontos, C., and Williams, D., 1992, The Ayia Ministry of Agriculture and Natural Resources, p. 13–26. Tethyan subduction factories: Lithos, v. 113, p. 68–87, Varvara Formation of S.W. Cyprus: A product of com- Robinson, P.T., Melson, W.G., O’Hearn, T., and Schmincke, doi:10.1016/j.lithos.2009.05.044. plex collisional tectonics: Tectonophysics, v. 212, p. 193– H.-U., 1983, Volcanic glass compositions of the Troodos Dilek, Y., Thy, P., Moores, E.M., and Ramsden, T.W., 1990, 211, doi:10.1016/0040-1951(92)90291-D. ophiolite, Cyprus: Geology, v. 11, p. 400–404. Tectonic evolution of the Troodos ophiolite within Miyashiro, A., 1975, The Troodos complex was probably Robinson, P.T., Malpas, J., and Xenophontos, C., 2003, The the Tethyan framework: Tectonics, v. 9, p. 811–823, formed in an island arc: Earth and Planetary Science Let- Troodos massif of Cyprus: Its role in the evolution of doi:10.1029/TC009i004p00811. ters, v. 25, p. 217–222, doi:10.1016/0012-821X(75)90199-5. the ophiolite concept, in Dilek, Y., and Newcomb, S., Fujimaki, H., and Aoki, K., 1987, Determination of trace ele- Moores, E.M., and Vine, F.J., 1971, The Troodos massif, eds., Ophiolite Concept and the Evolution of Geologic ments in rock samples by X-ray fl uorescence: Journal Cyprus, and other ophiolites as oceanic crust; evalu- Thought: Geological Society of America Special Paper of Mineralogy, Petrology and Economic Geology, v. 82, ation and implications, in Bullard, E., Cann, J.R., and 373, p. 295–308. p. 411–414. Matthews, D.H., eds., A Discussion on the Petrology of Rogers, N.W., MacLeaod, C.L., and Murton, B.J., 1989, Petro- Gass, I.G., MacLeod, C.J., Murton, B.J., Panayiotou, A., Simo- Igneous and Metamorphic Rocks from the Ocean Floor: genesis of boninitic lavas from the Limassol Forest com- nian, K.O., and Xenophontos, C., 1994, The Geology of Royal Society of London Philosophical Transactions, plex, Cyprus, in Crawford, A.J., ed., Boninites: Dordrecht, the Southern Troodos Transform Fault Zone: Nicosia, ser. A, v. 268, p. 443–466. Netherlands, Kluwer Academic Publishers, p. 288–313. Cyprus, Geological Survey Department Cyprus, Minis- Moores, E.M., Varga, K.L., Verosub, K.L., and Ramsden, Sanfi lippo, A., and Riedel, W.R., 1985, Cretaceous radiolaria, in try of Agriculture, Natural Resources and Environment, T., 1990, Regional structure of the Troodos dike com- Bolli, H.M., Saunders, J.B., and Perch-Nielsen, K., eds., Memoir 9, 218 p. plex, in Marpas, J., Moores, E.M., Panayiotou, A., and Plankton Stratigraphy: Cambridge, UK, Cambridge Uni- Geological Survey Department Cyprus, 1995, Geological Xenophontos, C., eds., Ophiolites, Oceanic Crustal Ana- versity Press, p. 573–630. Map of Cyprus: Nicosia, Cyprus, Ministry of Agricul- logues: Proceedings of the Symposium “Troodos 1987”: Shervais, J.W., 2001, Birth, death, and resurrection:The life cycle ture, Natural Resources and Environment, Government Nicosia, Cyprus, The Geological Survey Department, of suprasubduction zone ophiolites: Geochemistry Geo- of Cyprus, scale 1:250,000. Ministry of Agriculture and Natural Resources, p. 27–35. physics Geosystems, v. 2, doi:10.1029/2000GC000080.

LITHOSPHERE | Volume 4 | Number 6 | www.gsapubs.org 509

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021 OSOZAWA ET AL.

Shinjo, R., 1999, Geochemistry of high Mg andesites and the Bonin Arc, Japan: Earth and Planetary Science Letters, metamorphic soles, in Dilek, Y., and Robinson, P.T., eds., tectonic evolution of the Okinawa Trough–Ryukyu arc v. 164, p. 79–98, doi:10.1016/S0012-821X(98)00182-4. Ophiolites in Earth History: Geological Society of Lon- system: Chemical Geology, v. 157, p. 69–88, doi:10.1016 Taylor, R.N., Robert, W., Nesbitt, R.W., Harmon, R.S., Auvray, don Special Publication 218, p. 427–447, doi:10.1144/GSL /S0009-2541(98)00199-5. B., and Croudace, I.W., 1994, Mineralogy, chemistry, .SP.2003.218.01.22 2003. Spray, J.G., 1984, Possible causes and consequences of upper and genesis of the boninite series volcanics, Chichi- Wakabayashi, J., Ghatak, A., and Basu, A.R., 2010, Tectonic mantle decoupling and ophiolite displacement, in jima, Bonin Islands, Japan: Journal of Petrology, v. 35, setting of suprasubduction zone ophiolite generation Gass, I.G., Lippard, S.J., and Shelton, A.W., eds., Ophi- p. 577–617, doi:10.1093/petrology/35.3.577. and subduction initiation as revealed through geo- olites and Oceanic Lithosphere: Geological Society of Tremblay, A., Ruffet, G., and Bédard, J.H., 2011, Obduction chemistry and regional fi eld relationships: Geologi- London Special Publication 13, p. 255–268, doi:10.1144 of Tethyan-type ophiolites—A case-study from the cal Society of America Bulletin, v. 122, p. 1548–1568, /GSL.SP.1984.013.01.21. Thetford Mines ophiolitic complex, Quebec Appala- doi:10.1130/B30017.1. Spray, J.G., and Roddick, J.C., 1981, Evidence for Upper chians, Canada: Lithos, v. 125, p. 10–26, doi:10.1016/j Williams, H., and Smyth, W.R., 1973, Metamorphic aureoles Cretaceous transform fault metamorphism in West .lithos.2011.01.003. beneath ophiolite suites and alpine peridotites: Tectonic Cyprus: Earth and Planetary Science Letters, v. 55, Urquhart, E., and Robertson, A.H.F., 2000, Radiolarian evi- implications with west Newfoundland example: Ameri- p. 273–291, doi:10.1016/0012-821X(81)90107-2. dence of Late Cretaceous exotic blocks at Mangaleni, can Journal of Science, v. 273, p. 594–621, doi:10.2475 Stern, R.J., 2004, Subduction initiation: Spontaneous and Cyprus, and implications for the origin and emplace- /ajs.273.7.594. induced: Earth and Planetary Science Letters, v. 226, ment of the related Moni mélange, in Panayides, I., Xu, J.-F., and Castillo, P.R., 2004, Geochemical and Nd-Pb p. 275–292. Xenophontos, C., and Malpas, J., eds., Proceedings of isotopic characteristics of the Tethyan asthenosphere: Stern, R.J., and Bloomer, S.H., 1992, Subduction zone infancy: the Third International Conference on the Geology of Implications for the origin of the Indian Ocean mantle Examples from the Eocene Izu-Bonin-Mariana and Juras- the Eastern Mediterranean: Nicosia, Cyprus, Ministry of domain: Tectonophysics, v. 393, p. 9–27, doi:10.1016/j sic California arcs: Geological Society of America Bulle- Agriculture and Natural Resources, Geological Survey .tecto.2004.07.028. tin, v. 104, p. 1621–1636, doi:10.1130/0016-7606(1992)104 Department, Republic of Cyprus, p. 299–307. Yajima, K., Ono, M., and Fujimaki, H., 2001, Analytical accu- <1621:SZIEFT>2.3.CO;2. Varga, R.J., and Moores, E.M., 1985, Spreading structure of the racy and precision of major and trace elements for bulk Sun, S., and McDonough, W.F., 1989, Chemical and isoto- Troodos ophiolite, Cyprus: Geology, v. 13, p. 846–850, rocks using a 1:5 dilution glass bead by XRF: Japanese pic systematics of oceanic basalts: Implications for doi:10.1130/0091-7613(1985)13<846:SSOTTO>2.0.CO;2. Magazine of Mineralogical and Petrological Sciences, mantle composition and processes, in Saunders, A.D., Wakabayashi, J., 2004, Tectonic mechanisms associated with v. 30, p. 28–32, doi:10.2465/gkk.30.28. and Norry, M.L., eds., Magmatism in the Ocean Basins: P-T paths of regional metamorphism: Alternatives to Yamasaki, T., 1987, Radiolarian assemblages of the Izumi Geological Society of London Special Publication 42, single-cycle thrusting and heating: Tectonophysics, Group in Shikoku and western Awaji Island, southwest p. 313–345. v. 392, p. 193–218, doi:10.1016/j.tecto.2004.04.012. Japan: Journal of Geological Society of Japan, v. 93, Tatsumi, Y., and Eggins, S., 1995, Subduction Zone Magma- Wakabayashi, J., and Dilek, Y., 2000, Spatial and temporal p. 403–418, doi:10.5575/geosoc.93.403. tism: Cambridge, Blackwell Science, Inc., 211 p. relations between ophiolites and their subophiolitic Taylor, R.N., and Nesbitt, R.W., 1995, Arc volcanism in an soles: A test of models of forearc ophiolite genesis, in extensional regime at the initiation of subduction: A Dilek, Y., Moores, E.M., Elthon, D., and Nicolas, A., eds., geochemical study of Hahajima, Bonin Islands, Japan, Ophiolites and Oceanic Crust: New Insights from Field in Smellie, J.L., ed., Volcanism Associated with Exten- Studies and Ocean Drilling: Geological Society of Amer- MANUSCRIPT RECEIVED 6 FEBRUARY 2012 sion at Consuming Plate Margins: Geological Society of ica Special Paper 349, p. 53–64. REVISED MANUSCRIPT RECEIVED 13 SEPTEMBER 2012 London Special Publication 81, p. 115–134. Wakabayashi, J., and Dilek, Y., 2003, What constitutes MANUSCRIPT ACCEPTED 27 SEPTEMBER 2012 Taylor, R.N., and Nesbitt, R.W., 1998, Isotope characteristics ‘emplacement’ of an ophiolite?: Mechanisms and of subduction fl uids in an intra-oceanic setting, Izu- relationship to subduction initiation and formation of Printed in the USA

510 www.gsapubs.org | Volume 4 | Number 6 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/4/6/497/3039274/497.pdf by guest on 25 September 2021