Chemical variations of abyssal peridotites in the central Oman ophiolite: Evidence of oceanic mantle heterogeneity

Gondwana Research 25 (2014) 1242–1262 Contents lists available at ScienceDirect Gondwana Research journal homepage: www.elsevier.com/locate/gr Chemical variations of abyssal peridotites in the central Oman ophiolite: Evidence of oceanic mantle heterogeneity Mohamed Zaki Khedr a,b,⁎, Shoji Arai a, Marie Python c, Akihiro Tamura a a b c Department of Earth Sciences, Kanazawa University, Kakuma, Kanazawa 920-1192, Japan Department of Geology, Faculty of Science, Kafrelsheikh University, 33516, Egypt Department of Natural History Science, Hokkaido University, Japan a r t i c l e i n f o Article history: Received 20 January 2013 Received in revised form 27 April 2013 Accepted 17 May 2013 Available online 23 May 2013 Handling Editor: M. Santosh Keywords: Basal lherzolites Slab metasomatism Mantle heterogeneity Wadi Sarami Central Oman ophiolite a b s t r a c t Basal peridotites above the metamorphic sole outcropped around Wadi Sarami in the central Oman ophiolite give us an excellent opportunity to understand the spatial extent of the mantle heterogeneity and to examine peridotites−slab interactions. We recognized two types of basal lherzolites (Types I and II) that change upward to harzburgites. Their pyroxene and spinel compositions display severely variations at small scales over b 0.5 km, and encompass the entire abyssal peridotite trend; clinopyroxenes (Cpxs) show wide ranges of Al2O3, Na2O, Cr2O3 and TiO2 contents. Primary spinels show a large variation of Cr# [= Cr/(Cr + Al)] from 0.04 to 0.53, indicating various degrees of partial melting. Trace-element compositions of peridotites and their pyroxenes also show a large chemical heterogeneity in the base of the Oman mantle section. This heterogeneity mainly resulted from variations of partial-melting degrees due to the change of a mantle thermal regime and a distance from the spreading ridge or the mantle diapir. It was overlapped with subsolidus modification during cooling and fluid metasomatism prior and/or during emplacement. The studied peridotites are enriched in Rb, Cs, Ba, Sr and LREE due to fluid influx during detachment and emplacement stages. Chondrite (CI)-normalized REE patterns for pyroxenes are convex upward with strong LREE depletion due to their residual origin, similar to abyssal peridotites from a normal ridge segment. The Cpxs are enriched in fluid mobile elements (e.g., B, Li, Cs, Pb, Rb) and depleted in HFSE (Ta, Nb, Th, Zr) + LREE, suggesting no effect of melt refertilization. Their HREE contents, combined with spinel compositions, suggest two melting series with 1–5% melting for type II lherzolites, 3– b 10% melting for type I lherzolites and ~15% for harzburgites. Hornblendes are enriched in fluid-mobile elements relative to HFSE + U inherited from their precursor Cpx. The clinopyroxenite lens crosscuts the basal lherzolites, forming small-scale (b5 cm) mineralogical and chemical heterogeneities. It was possibly formed from fractional crystallization of interstitial incremental melt that formed during decompression melting of a normal MORB mantle source. The studied peridotites possibly represent a chemical heterogeneity common to the mantle at an oceanic spreading center. © 2013 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved. 1. Introduction The Oman ophiolite is one of the largest and best-preserved sections of oceanic lithosphere in the world. Its mantle section consists mainly of harzburgites with minor dunites (e.g., Boudier and Nicolas, 1995). Few studies outlined the presence of basal lherzolites, i.e., lherzolites at the base of mantle section lying directly above the metamorphic sole in its northern part (Lippard et al., 1986; Takazawa et al., 2003: Khedr et al., 2013) or Cpx-bearing harzburgites at the base of the mantle section in the southern Oman ophiolite (Godard et al., 2000; Hanghøj et al., 2010). Lithological and chemical variations at the base of the Oman mantle section clearly confirm the existence of the upper mantle heterogeneity (e.g., Khedr et al., 2013). This heterogeneity likely resulted ⁎ Corresponding author at: Department of Earth Sciences, Kanazawa University, Kakuma, Kanazawa 920-1192, Japan. Tel.: +81 76 264 6513; fax: +81 76 264 6545. E-mail address: khedrzm@yahoo.com (M.Z. Khedr). from variations in the degree of melting and in the initial compositions, interaction with asthenospheric melts, modification by slab-derived melts or fluids at the mantle wedge, and mixing with the recycled oceanic crust/sediment in the deeper mantle (Bougault et al., 1988; Johnson et al., 1990; Bonatti et al., 1992; Hofmann, 1997; Hellebrand et al., 2002; Anderson, 2006; Stracke and Bourdon, 2009; Brandl et al., 2012). If the studied peridotites have suffered from subsolidus modifications at the mantle depth, this modification also leads to mantle heterogeneities. Because the mantle is the source of basaltic magmas, the mantle heterogeneity is one of the main factors controlling the diversity and chemical variations of erupted magmas (e.g., Brandl et al., 2012). Origin and dimension of the mantle heterogeneity, inferred from abyssal peridotites, are, however, still a matter of debate due to a scarcity of evidence (Dick et al., 1984; Michael and Bonatti, 1985; Bonatti et al., 1992; Hellebrand et al., 2002). In order to determine the origin and dimension of mantle heterogeneity, we conducted a detailed petrological and geochemical study on the basal mantle section in Wadi 1342-937X/$ – see front matter © 2013 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gr.2013.05.010 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Sarami, the central Oman ophiolite, where we can see the change of lherzolites to harzburgites and then dunites. This study depends mainly on compositions of unaltered peridotitic minerals that are reliable, compared to the whole-rock chemistry, to reflect mantle processes. The systematic change of trace-element compositions of peridotites and their pyroxenes reflects the upper mantle heterogeneity that is mainly related to variations in a partial-melting degree and an initial composition, whereas secondary processes (e.g., slab metasomatism) are possibly responsible for incompatible-element enrichment. This study will provide us with meter-scale lithological and chemical heterogeneities, if any, in the base of the mantle section, and will place constraints on peridotite− slab interactions during obduction and emplacement. 1243 2. Geological setting and petrography The Wadi Sarami crosscuts two large ophiolitic blocks of the central Oman mountains (Sarami and Wuqbah blocks), which are separated by the north-westernmost and narrowest part of the Hawasina window (Fig. 1). Structural studies in this area showed that the sequence of the nappes, in which ophiolitic mantle lies above the metamorphic amphibolites and the Hawasina Formation, is similar in both sides of the Hawasina window. Thus, the Sarami and Wuqbah blocks in central Oman massifs were probably connected and actually formed one ophiolitic block that was separated into two blocks by deeper erosion or by diapiric intrusion of the Hawasina Formation into the ophiolite Fig. 1. Geological map of basal peridotites from Wadi Sarami, central Oman ophiolite. (a) A part of the main ridge-related structural map of the Oman ophiolite, central and north massifs (Nicolas et al., 2000). (b) Geological map of Wadi Sarami showing the locations of basal peridotite samples. Modified from map of Ministry of Petroleum and Minerals, 1992. 1244 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 (Nicolas et al., 1988, 2000). The mantle section in this area composed mainly of harzburgites with minor dunites and lherzolites is thrusted over the amphibolitic sole and the Hawasina Formation (Fig. 1). Our detailed study is based on more than 200 samples taken at intervals of few tens of centimeters to a meter in three oases of Al-Khabt, Al-Baks and Al-Qala to check lithological and chemical changes from the lherzolitic mantle upward to the harzburgitic mantle (Fig. 1b). We previously distinguished two types of basal lherzolites (Types I and II, see Khedr et al., 2013). The Type I lherzolites, being massive, crop out at various levels above the metamorphic sole. The Type II lherzolites, which are foliated, lie up to few meters (sometimes up to ~150 m) above the metamorphic sole, and are overlain and/or surrounded by Type I lherzolites (Khedr et al., 2013). Some of them show mylonitic structure at the direct contact with the metamorphic sole, and well-preserved mylonitic texture is easily recognized in the thin section (Khedr et al., 2013). We observed a lithological change from Type II lherzolites to harzburgites through Type I lherzolites or from Type I lherzolites to harzburgites within a few hundreds of meters (b 0.4 km from the sole contact) (Fig. 1a), suggesting small-scale mantle heterogeneities (Khedr et al., 2013). The studied peridotites contain variable amounts of Cpxs (0.5− 14.0 vol.%), which are unevenly distributed in each sample and sometimes show patchy concentrations in lherzolites (Fig. 2b). The modal abundance based on point counting (2000 counts for 2.5 × 4.5 cm) is listed in Supplementary Data 1. The Cpxs occur mainly as laminated porphyroclasts (Fig. 2a, b). Some coarse- and fine-grained Cpxs in Fig. 2. Textural characteristics of basal peridotites from Wadi Sarami. Photomicrographs taken by crossed-polarized light except for c (back scattered electron image: BSE) and e (slab taken by Scanner). (a) Coarse Cpx porphyroclasts containing Opx lamellae and surrounded by fine-grained Cpxs in massive Type I lherzolite (Tv.59). (b) Laminated coarse Cpxs as patchy or grain aggregates, different in morphology from thin film or vermicular interstitial Cpx formed from trapped melt, in foliated Type II lherzolite (N.277). (c) Cpx exsolution lamellae in Opx porphyroclasts in Type II lherzolites (G.53). (d) Equigranular olivine around coarse Opx and Cpx showing protogranular texture in harzburgite (Tv.136). (e) Scanned photo of hand-specimen slab showing a clinopyroxenite lens (G.43 M). (f) Cpx grain aggregates forming the lens (e) embedded in serpentines of hydrous peridotites (Olv, Serp, Hb) (G.43 M). Abbreviations; Opx, orthopyroxene: Cpx, clinopyroxene: Olv, olivine: Spl, primary spinel: Serp, serpentines: Hb, hornblende. M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Type I and II lherzolites are sieved and/or surrounded by hornblende patches or plates, showing metasomatic replacement. Nevertheless, Cpx grains are free of this metasomatic hornblende in harzburgites. Most orthopyroxene crystals in peridotites contain lamellae or blebs of Cpx (Fig. 2c). Type I lherzolites are composed of olivine, orthopyroxene (Opx), Cpx and spinel. They display mainly porphyroclastic textures (Khedr et al., 2013). Cpx shows various grain sizes. The coarse prismatic crystals (~2.2 mm across) of Cpx sometimes contain exsolution lamellae of Al-rich spinel. Type II lherzolites consist of olivine, Opx, Cpx and spinel. They show mainly mylonitic to porphyroclastic textures (Khedr et al., 2013). They are highly serpentinized relative to Type I (Supplementary Data 1). Tremolite occurs as fibrous crystals around Opx edges, like a corona shape. Harzburgites are massive and weakly serpentinized. They consist of olivine, Cpx, Opx and spinel, and are characterized by protogranular textures (Fig. 2d). Hydrous peridotites occur within the base of lherzolites, a few tens of centimeters, and consist mainly of serpentines (mainly lizardite and chrysotile) and hornblende with subordinate olivine and Cpx (Fig. 2e, f). Hornblende forms prismatic and fibrous crystals associated with fine-grained olivine (Fig. 2f). Hydrous peridotites contain a lens of clinopyroxenite, 2.2 cm in length × 1.2 cm in width (Fig. 2e), which is embedded in serpentines (Fig. 2f). This spindle shaped lens forms a sharp contact with its host rock (Fig. 2e). It is composed mainly of Cpx with a few spinel grains and secondary minerals (e.g., andradite and serpentine). The Cpx in the lens occurs mainly as anhedral to subhedral grains, showing a mosaic texture (Fig. 2f); another few subhedral Cpx grains are laminated, and display morphology of primary Cpx. This clinopyroxenite lens contains a few small spinel grains that are concentrated within and around the edge of Cpx grains. The interstitial andradite occurs as small flakes between Cpx grains and within serpentinite veins cutting the lens. 3. Sample preparation and analytical methods The basal peridotites selected for bulk-rock chemical analyses were homogeneous, and free of veins. The samples are crushed and grinded to 150 meshes (105 μm). Their powders were heated up to 1000 °C for 2 h to remove structural water before preparing fused beads for major- and trace-element analyses. The beads were prepared by fusion of the rock powders (0.2 g) with lithium metaborate/ tetraborate flux (1.2 g) in platinum crucibles. The resulting molten beads were rapidly digested in 5% nitric acid solution (120 g) before trace element and major oxide analysis. Major- and trace-element contents (Table 1) of basal peridotites were determined by Perkin Elmer Sciex ELAN 6100, 9000-inductively coupled plasma mass spectrometry (ICP−MS) at Activation Laboratories Ltd (Actlabs) Ancaster, Ontario in Canada (www.actlabs.com). International standard reference materials such as W-2A, BIR-1, DNC-1 (USGS Certified Reference Material), NIST 694, NIST 696 (National Institute of Standards and Technology) and other international certified reference material were analyzed with every batch of samples and reported as part of quality control. The internal standard for the ICP suite of analyses is cadmium (Cd) whereas the internal standard for the ICP−MS suite of elements is rhodium (Rh) and iridium (Ir). The NIST 694 is certified essentially for major elements instead of trace elements. The relative standard deviation from replicate analyses is b5% for major elements and b10% for minor/trace elements. The major-element contents of peridotitic minerals (Tables 2 and 3; Supplementary Data 2 and 3) were determined by JEOL wavelength dispersive electron probe X-ray micro-analyzer (JXA 8800 JEOL) at Kanazawa University, Japan. Accelerating voltage, beam current, and beam diameter for the analyses were 20 kV, 20 nA, and 3 μm, respectively. Chemical mapping of selected minerals was carried out by the microprobe at the same conditions, except for beam diameter of b 1 (nominally 0) μm and beam current of 80 nA. Mg# is Mg/(Mg + total Fe) atomic ratio for bulk rocks and silicates, and Mg/(Mg + Fe2+) atomic ratio in chromian spinel, calculated assuming spinel stoichiometry. Cr# is 1245 Cr/(Cr + Al) atomic ratio. Trace-element abundances (Tables 2 and 3; Supplementary Data 3) of silicates (Cpx, Opx, olivine and hornblende) were in-situ determined by laser-ablation (193 nm ArF excimer: MicroLas GeoLas Q-plus)–inductively coupled plasma mass spectrometry (Agilent 7500S) (LA–ICP–MS) at Kanazawa University. Analyses were performed by ablating 60-μm diameter spots for Cpx and hornblende, whereas spot diameter for olivine and Opx was 100 μm. All analyses were performed at 6 Hz with energy density of 8 J/cm2 per pulse. NIST 612 glass was used as an external standard, assuming the composition given by Pearce et al. (1997), and 29Si was used as an internal standard based on SiO2 concentration obtained by the electron microprobe. NIST 614 glass (secondary standard) was measured for quality control of each analysis (Table 2; Supplementary Data 3). Precision or reproducibility is better than 4% for most elements (Ti and B better than 7%), except Sc, Ni, Sr and Pb for which it is better than 19%. The accuracy and data quality based on the reference material (NIST 614) are high, and were described by Morishita et al. (2005). Details of the analytical procedures have been described by Morishita et al. (2005). 4. Geochemical characteristics 4.1. Whole-rock compositions The normative abundance of Cpx based on the scheme of Niu (1997) (Table 1) is nearly consistent with modal% of Cpx counted in the thin section (Supplementary Data 1), except for a few samples that include fine-grained Cpx (difficult to count) and high modal% of serpentines. The Sarami basal peridotites (=Sarami–Wuqbah peridotites) are mainly lherzolites with subordinate harzburgites (Table 1; Supplementary Data 1); their bulk-rock Mg# ranges from 0.90 to 0.92 and matches with that of their olivine (Mg#, 0.90–0.92), and Opx (Mg#, 0.89–0.92) (Tables 1 and 3; Supplementary Data 2). The Type I and II lherzolites are enriched in A12O3 (1.9–3.12 wt.%), CaO (1.6–3.0 wt.%), TiO2 (0.02–0.08 wt.%), and Na2O (0.02–0.1 wt.%) relative to harzburgite compositions (A12O3 = 0.8–1.3 wt.%, CaO = 0.96–1.5 wt.%, TiO2 = 0.01 wt.%, and Na2O = 0.01–0.02 wt.%, see Table 1). They are similar in major-element compositions to Fizh lherzolites and abyssal peridotites (Takazawa et al., 2003; Niu, 2004; Monnier et al., 2006; Fig. 3). The Sarami basal peridotites follow the residual abyssal peridotite trend from Pacific and Indian Ocean ridges (Niu, 1997, 2004) (Fig. 3). Their Al2O3, CaO, TiO2, Na2O, Sc, V, heavy rare earth elements (HREE), and Y contents decrease systematically as MgO increases, whereas Ni and Co show good positive correlations with MgO, coinciding with residual peridotite trends (Figs. 3 and 4). But light rare earth elements (LREE) (e.g., La, Ce, Pr and Nd), Sr and Ba (not shown here) show erratic or scattered plots, without any correlations with MgO (Table 1). The Sarami basal peridotites display spoon-shaped REE patterns (0.1–2.5 times CI) with inflection at La, Ce and Pr depending on their lithological types (Fig. 5a, c). Their primitive mantle (PM)-normalized multi-element patterns show slightly higher concentrations in large ion lithophile elements (LILE) with Cs and Sr positive anomalies than in high field strength elements (HFSE) (below detection limits) (Fig. 5b, d). The basal peridotites are similar in trace-element chemistry to residual abyssal peridotites (Niu, 2004) and in HREE to Fizh peridotites in north Oman (Takazawa et al., 2003), except that Type II lherzolites show enrichment in LREE in contrast to Fizh Type II lherzolites (Fig. 5a, c). 4.2. Mineral compositions 4.2.1. Major elements The major-element mineral compositions of lherzolites and harzburgites (Fig. 6) have been discussed in detail by Khedr et al. (2013), and some representative data are given as a Supplementary Data 2 in the online version. Pyroxenes and spinels show large intersample and intra-sample chemical heterogeneities (Figs. 6 and 7). The basal peridotites show compositional variations of Cpx (e.g., Al, Na, Cr 1246 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Table 1 Major (wt.%) and trace element (ppm) abundances in whole rocks of Sarami basal peridotites. Rock type Type I lherzolites Sample no. TV.125 G.33 TV.59 TV.56 TV.41B G.52 TV.117 G.27 G.47 TV.123 TV.40 N.264 TV.60 TV.136 SiO2 TiO2 Al2O3 Fe2O3a MnO MgO CaO Na2O K2O P2O5 Total 45.87 0.08 3.12 8.02 0.13 39.46 3.09 0.10 bdl 0.03 99.9 45.72 0.02 1.91 7.26 0.15 41.57 1.89 0.02 bdl 0.03 98.6 44.63 0.04 2.21 8.02 0.12 39.79 2.60 0.03 bdl bdl 97.4 44.98 0.04 2.48 8.47 0.13 40.17 2.87 0.04 bdl bdl 99.2 45.85 0.08 3.00 7.02 0.15 40.71 2.49 0.08 bdl bdl 99.4 45.01 0.06 2.70 7.91 0.15 40.82 1.58 0.06 bdl bdl 98.3 44.99 0.05 2.46 8.32 0.10 42.00 1.63 0.05 bdl 0.01 99.6 44.23 0.06 2.67 8.12 0.11 41.28 1.59 0.04 bdl bdl 98.1 45.06 0.06 2.72 8.42 0.12 41.18 2.62 0.08 bdl 0.01 100.3 45.25 0.06 2.74 8.74 0.11 40.43 2.87 0.08 bdl bdl 100.3 45.97 0.07 2.88 7.09 0.18 41.24 2.29 0.06 bdl 0.02 99.8 45.24 0.01 1.30 8.28 0.10 43.94 1.47 0.02 bdl 0.02 100.4 43.44 0.01 0.81 8.69 0.12 45.07 1.04 bdl bdl 0.01 99.2 43.86 0.01 0.88 8.59 0.12 45.92 0.96 0.01 bdl 0.02 100.4 0.010 0.001 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 LOI Mg#b Mg#c-Olv 8.04 0.907 0.899 11.20 0.919 0.908 10.64 0.908 0.903 7.61 0.904 0.901 8.96 0.920 0.908 10.64 0.911 0.906 9.37 0.909 11.52 0.910 0.908 8.57 0.906 0.903 9.30 0.902 0.901 10.68 0.920 0.907 10.95 0.913 0.910 7.03 0.911 0.914 6.93 0.914 0.915 0.010 CIPW normd Olv Opx Cpx 53.88 31.28 14.62 58.75 32.49 8.64 58.93 29.00 11.99 59.96 26.91 13.06 54.92 32.70 12.23 58.01 33.47 8.41 62.76 28.88 8.25 62.10 29.46 8.32 61.99 25.42 12.45 60.20 26.19 13.50 55.53 33.73 10.57 69.32 24.10 6.52 79.12 16.31 4.53 79.67 16.02 4.25 (ppm) 2420 101 1770 75.0 14.0 30.0 50.0 2.00 1.20 36.00 2.10 bdl 463.2 2.40 13.00 0.06 0.11 0.03 0.22 0.13 0.042 0.23 0.04 0.37 0.08 0.24 0.038 0.29 0.052 18 1.932 2610 99.0 1980 62.0 12.0 bdl 170 2.00 1.20 8.00 0.80 bdl 150.9 3.40 4.00 bdl bdl bdl bdl bdl bdl bdl 0.01 0.12 0.03 0.11 0.017 0.12 0.027 8.8 0.434 2400 98.0 1740 69.0 13.0 30.0 50.0 2.00 1.00 22.00 1.30 bdl 262.4 2.70 8.00 bdl 0.09 0.01 bdl 0.02 0.014 0.10 0.02 0.17 0.04 0.16 0.026 0.19 0.033 8.3 0.873 2610 103 1780 76.0 15.0 30.0 40.0 2.00 1.10 10.00 1.50 bdl 261.8 0.70 bdl 0.05 0.08 bdl 0.09 0.04 0.006 0.11 0.02 0.20 0.05 0.17 0.028 0.19 0.031 15.1 1.065 2470 98.0 1740 73.0 14.0 10.0 140 3.00 1.00 9.00 2.10 bdl 483.7 1.40 bdl 0.06 0.10 0.02 0.16 0.10 0.035 0.17 0.05 0.34 0.07 0.24 0.045 0.32 0.05 15.1 1.76 2320 95.0 1780 60.0 12.0 10.0 180 2.00 0.80 5.00 1.60 bdl 382.9 0.90 bdl bdl 0.08 0.02 0.11 0.07 0.037 0.12 0.04 0.26 0.06 0.19 0.029 0.21 0.036 9 1.262 2510 100 1820 59.0 12.0 20.0 50.0 2.00 0.90 5.00 1.50 1.00 285.5 0.40 bdl 0.06 0.09 0.01 0.11 0.06 0.021 0.10 0.03 0.25 0.05 0.18 0.032 0.21 0.035 14.3 1.238 2580 100 1820 64.0 12.0 20.0 90.0 2.00 0.90 7.00 2.40 bdl 366.7 1.10 bdl 0.25 0.62 0.09 0.50 0.13 0.04 0.21 0.06 0.36 0.08 0.28 0.047 0.31 0.052 7 3.029 2360 100 1790 67.0 13.0 20.0 60.0 2.00 1.00 13.00 1.60 2.00 339.9 0.90 4.00 0.32 0.37 0.02 0.11 0.09 0.018 0.17 0.04 0.32 0.06 0.20 0.029 0.21 0.04 12.9 1.997 2380 102 1740 72.0 13.0 20.0 50.0 2.00 1.00 11.00 1.80 bdl 368.7 0.80 4.00 bdl 0.16 bdl bdl 0.08 0.025 0.15 0.04 0.30 0.07 0.21 0.035 0.24 0.04 16.8 1.35 2470 85.0 1720 69.0 13.0 bdl 150 2.00 1.10 13.00 1.80 bdl 423.5 1.50 4.00 bdl 0.09 0.01 0.08 0.11 0.043 0.15 0.04 0.29 0.06 0.23 0.039 0.27 0.045 10.2 1.457 2510 105 1930 49.0 10.0 bdl 40.0 1.00 1.10 21.00 bdl bdl 80.3 3.60 4.00 0.08 0.12 0.02 bdl 0.02 bdl bdl bdl 0.06 0.01 0.04 0.009 0.06 0.011 9.4 0.43 2340 115 2080 42.0 10.0 20.0 50.0 bdl 0.90 8.00 bdl bdl 52.0 0.50 4.00 0.11 0.17 0.01 bdl 0.03 bdl 0.02 bdl 0.01 bdl 0.03 0.007 0.06 0.013 28.4 0.46 2440 115 2070 39.0 9.0 20.0 40.0 bdl 1.10 bdl bdl bdl 64.1 0.50 bdl 0.12 0.23 0.02 bdl 0.02 bdl 0.02 bdl 0.05 bdl 0.04 0.006 0.05 0.011 32.8 0.567 Trace elements Cr Co Ni V Sc Cu Zn Ga Ge Sr Y Zr Ti Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu W ΣREE Type II lherzolites Harzburgite dl 20.00 1.00 20.00 5.00 1.00 10.00 30.00 1.00 0.50 2.00 0.50 1.00 6.00 0.10 3.00 0.050 0.050 0.010 0.050 0.010 0.005 0.010 0.010 0.010 0.010 0.010 0.005 0.010 0.002 bdl, below detection limits. Note Pb and Th (b0.05 ppm), U and Ta (b0.01 ppm), Hf (b0.1 ppm), Rb (b1.0 ppm), Nb (b0.2 ppm) analyses (not listed in the table) of all samples are below detection limits. a Fe, total Fe as Fe2O3. b Mg# = Mg/(Mg + Fe2+) of whole rocks and Fe2+ as total iron. c Mg#-Olv, average Mg# of olivines from EPMA analyses. d CIPW norm calculated following the scheme by Niu (1997). and Mg#), Opx (e.g., Al and Mg#) and spinel (Cr#) with a distance from the sole contact upward to harzburgites (Fig. 7). Cpxs show wide ranges of Mg# = 0.90–0.95, Al2O3 = 1.3–7.3 wt.%, Na2O = 0.004–1.2 wt.%, Cr2O3 = 0.2–1.4 wt.% and TiO2 = 0.01–0.4 wt.% (Table 2; Figs. 6a and 7). Their exsolution lamellae of spinel are Al-rich (Fig. 8). Cpxs in clinopyroxenite lens cutting hydrous peridotites (Fig. 2e) show highly magnesian and depleted characteristics, i.e., Mg# = 0.94–0.96, Al2O3 = 0.6–1.4 wt.%, Na2O = 0.00–0.11 wt.%, Cr2O3 = 0.05–0.33 wt.% and TiO2 = 0.01–0.07 wt.% (Table 2; Supplementary Data 2). Opxs in peridotites (=lherzolites and harzburgites) show high Mg#s, 0.89−0.92, and wide compositional ranges of Al2O3 from 0.7 to 6.7 wt.%, CaO from 0.3 to 2.6 wt.%, and Cr2O3 from 0.15 to 1.02 wt.% (Khedr et al., 2013; Table 3 and Fig. 6b). Olivines in peridotites show a range of forsterite content from 89.6 to 92.5 (90.7, on average) (see Supplementary Data 3). Their chemistry and morphology are similar to residual olivine in primary peridotites (Fig. 6c). Primary spinels (Mg#, 0.6–0.8) in peridotites exhibit a wide range of Cr# from 0.04 to 0.53 (Fig. 6d) with low YFe [=(Fe3+/(Cr + Al + Fe3+) atomic ratio, 0.001–0.05], and TiO2 (0.0–0.09 wt.%) (Khedr et al., 2013). They encompass the entire spaces of abyssal peridotites (Fig. 6d). Spinels (Mg#, 0.2–0.6) in clinopyroxenite lens display higher Cr# (0.4–0.6), TiO2 (0.04–0.17 wt.%) and YFe (0.04–0.12) than those of basal lherzolites (Cr# b 0.25, Fig. 6d). Pargasitic hornblendes in lherzolites show an intra-grain chemical heterogeneity and have wide ranges of 1247 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Table 2 Major (wt.%) and trace element (ppm) abundances of clinopyroxenes in Sarami basal peridotites. Rock type Type I lherzolites Sample no. Tv.59 SiO2 TiO2 Al2O3 Cr2O3 FeOa MnO MgO CaO Na2O K2O NiO Total 50.11 0.23 6.80 1.11 2.99 0.14 15.25 23.31 0.31 0.00 0.02 100.3 51.53 0.21 5.54 1.10 2.47 0.08 15.64 23.59 0.41 0.00 0.02 100.6 53.48 0.21 2.80 0.35 2.36 0.09 17.62 23.42 0.25 0.01 0.04 100.6 52.37 0.27 5.07 0.95 2.29 0.06 15.91 23.72 0.37 0.00 0.03 101.0 50.25 0.33 5.37 0.71 2.26 0.07 15.42 23.16 0.58 0.01 0.03 98.2 54.26 0.21 3.83 0.44 2.31 0.09 16.91 23.53 0.55 0.00 0.03 102.2 51.91 0.29 5.07 0.71 2.19 0.09 15.83 23.50 0.56 0.00 0.05 100.2 51.78 0.36 5.81 0.88 2.38 0.08 15.64 23.32 0.62 0.00 0.04 100.9 52.77 0.24 5.35 0.81 2.31 0.07 16.30 23.20 0.51 0.00 0.04 101.6 52.89 0.24 5.44 0.84 2.46 0.07 16.30 22.75 0.55 0.00 0.05 101.6 53.80 0.18 4.01 0.71 2.39 0.07 17.15 23.07 0.45 0.00 0.02 101.8 53.72 0.15 2.72 0.62 2.10 0.07 17.38 24.42 0.23 0.01 0.06 101.5 Cpx-Mg#b Spl-Cr#c 0.901 0.156 0.919 0.113 0.930 0.123 0.925 0.156 0.924 0.111 0.929 0.113 0.928 0.113 0.922 0.113 0.926 0.121 0.922 0.123 0.928 0.107 0.937 0.250 Trace elements (ppm) Li B Sc Ti V Cr Co Ni Rb Sr Y Zr Nb Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Pb ΣREE 18.68 3.32 43.86 1213 264.6 8029 28.01 441.1 1.079 3.849 8.023 0.402 0.058 0.721 1.553 bdl 0.007 0.011 0.230 0.317 0.155 0.794 0.168 1.372 0.306 0.925 0.133 0.932 0.131 0.08 0.121 5.48 22.62 5.86 41.53 1152 251.5 9111 28.26 479.4 0.932 3.913 8.467 0.397 0.065 0.921 1.719 bdl 0.008 0.011 0.226 0.318 0.160 0.826 0.172 1.425 0.328 0.983 0.142 1.032 0.142 0.09 0.074 5.77 3.25 3.60 52.97 1041 214.0 2755 29.07 414.3 0.058 0.818 7.605 0.337 0.018 0.093 0.214 bdl 0.007 0.010 0.181 0.288 0.129 0.770 0.154 1.274 0.298 0.867 0.129 0.855 0.124 0.09 0.053 5.09 8.37 1.82 53.03 1472 267.8 7114 22.48 359.4 0.374 1.885 10.701 0.583 0.049 0.322 0.650 bdl 0.010 0.015 0.283 0.399 0.194 0.997 0.228 1.822 0.434 1.275 0.194 1.287 0.188 0.12 0.042 7.33 4.25 4.45 36.03 1455 217.3 6188 41.30 533.0 0.180 4.355 8.071 2.492 0.043 0.281 1.567 0.001 0.062 0.045 0.582 0.506 0.206 0.955 0.174 1.442 0.308 0.928 0.138 0.946 0.125 0.22 0.040 6.42 5.69 1.30 51.79 2096 283.5 7086 22.10 359.5 0.073 3.536 13.317 2.691 0.046 0.267 0.775 bdl 0.085 0.063 0.819 0.750 0.349 1.440 0.295 2.318 0.514 1.529 0.216 1.550 0.208 0.22 0.034 10.14 7.81 2.65 51.93 1834 254.5 6191 21.69 345.9 0.409 2.620 11.977 2.707 0.049 0.613 1.217 bdl 0.074 0.055 0.766 0.658 0.309 1.303 0.268 2.154 0.455 1.412 0.217 1.416 0.199 0.23 0.044 9.28 3.28 17.40 46.50 1820 248.0 7385 24.41 469.4 0.085 7.595 12.192 2.372 0.055 0.178 2.857 bdl 0.070 0.056 0.738 0.662 0.293 1.290 0.265 2.059 0.459 1.387 0.208 1.398 0.190 0.21 0.056 9.08 8.04 49.38 47.20 1289 212.7 4630 39.70 664.5 bdl 1.516 9.802 2.340 0.057 0.037 1.180 bdl 0.058 0.048 0.636 0.566 0.243 1.096 0.222 1.731 0.374 1.116 0.164 1.114 0.151 0.23 0.242 7.52 9.40 9.99 53.22 1546 248.7 6043 23.37 358.9 bdl 1.546 11.584 2.455 0.064 bdl 1.004 0.005 0.067 0.049 0.666 0.608 0.280 1.207 0.250 2.052 0.463 1.340 0.195 1.322 0.171 0.24 0.110 8.67 6.33 6.91 52.54 1355 230.9 4687 22.61 353.5 bdl 2.296 10.368 2.212 0.052 bdl 0.882 0.004 0.061 0.047 0.687 0.598 0.270 1.163 0.249 1.926 0.422 1.239 0.181 1.240 0.172 0.21 0.124 8.26 4.71 3.66 63.82 1092 276.5 6907 20.05 347.9 bdl 0.188 8.283 0.097 0.039 0.019 0.064 bdl bdl bdl 0.056 0.160 0.087 0.573 0.142 1.386 0.328 1.047 0.161 1.080 0.149 0.05 bdl 5.17 Rock type Type II lherzolites Tv.125 G.39 G.52 G.33 G.27 Av.dl 0.292 0.681 0.054 0.263 0.044 2.034 0.027 0.131 0.047 0.007 0.005 0.019 0.005 0.028 0.034 0.003 0.003 0.002 0.015 0.021 0.005 0.023 0.004 0.012 0.005 0.009 0.004 0.014 0.003 0.017 0.028 Nist614 1.77 1.55 1.73 3.50 1.02 1.25 0.76 1.41 0.85 43.16 0.74 0.81 0.77 0.67 3.04 0.68 0.76 0.73 0.71 0.72 0.73 0.71 0.68 0.72 0.71 0.70 0.69 0.74 0.71 0.65 2.38 Sample no. Tv.123 SiO2 TiO2 Al2O3 Cr2O3 FeOa MnO MgO CaO Na2O K2O NiO Total 51.16 0.26 6.20 0.85 2.26 0.07 15.20 22.40 0.94 0.00 0.03 99.4 51.68 0.26 6.82 0.99 2.50 0.14 15.07 22.08 0.98 0.00 0.02 100.5 52.09 0.33 5.54 0.86 2.21 0.03 15.09 21.31 0.90 0.00 0.04 98.4 52.14 0.24 6.32 0.97 2.63 0.09 17.10 20.28 0.93 0.00 0.04 100.7 53.23 0.28 3.68 0.79 1.84 0.03 16.58 23.46 0.58 0.00 0.06 100.5 51.33 0.27 7.29 1.23 2.63 0.08 16.42 20.99 0.98 0.00 0.05 101.3 51.14 0.40 5.83 0.98 1.87 0.06 15.85 22.90 1.06 0.00 0.00 100.1 51.42 0.31 6.61 1.03 2.12 0.10 14.91 22.79 0.91 0.00 0.04 100.2 52.42 0.18 3.59 0.82 2.40 0.13 16.69 23.79 0.51 0.00 0.03 100.6 51.49 0.31 7.00 1.14 2.08 0.06 15.77 22.53 0.84 0.00 0.04 101.2 51.41 0.30 6.28 0.88 2.35 0.08 16.87 21.78 0.58 0.00 0.06 100.6 Tv.40 52.31 0.34 6.57 0.88 2.23 0.06 15.16 22.86 0.97 0.00 0.04 101.4 52.46 0.35 6.80 0.95 2.01 0.08 15.36 22.93 1.02 0.00 0.05 102.0 52.80 0.32 5.66 0.97 2.51 0.08 16.51 21.55 0.94 0.00 0.04 101.4 52.40 0.35 6.37 0.83 2.23 0.10 15.44 23.00 0.93 0.00 0.01 101.7 Cpx-Mg#b Spl-Cr#c 0.923 0.133 0.915 0.150 0.924 0.127 0.921 0.160 0.941 0.134 0.918 0.121 0.938 0.125 0.926 0.147 0.925 0.143 0.931 0.131 0.928 0.131 0.924 0.143 0.932 0.142 0.921 0.116 0.925 0.143 Trace elements (ppm) Li B Sc 3.46 15.95 47.58 7.52 38.48 40.30 3.10 2.34 43.06 14.54 4.78 51.66 13.01 20.71 58.63 5.31 3.74 52.31 2.52 2.19 47.19 14.28 24.99 41.49 13.76 10.66 56.01 13.74 2.58 52.33 10.91 8.97 47.58 5.67 1.42 42.14 5.60 3.48 43.74 4.88 2.13 53.56 8.39 7.53 52.50 (continued on next page) 1248 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Table 2 (continued) Rock type Type II lherzolites Sample no. Tv.123 Ti V Cr Trace elements (ppm) Co Ni Rb Sr Y Zr Nb Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Pb ΣREE 1956 269.3 7118 1557 247.6 7593 1695 259.9 7357 2148 257.8 7338 2186 254.0 6659 2036 247.2 7085 1843 243.9 7368 1613 246.9 7798 1996 292.7 7292 1976 257.0 7496 1862 252.2 7996 1769 233.4 6877 1915 245.7 7465 2169 259.0 6840 2277 260.5 6853 21.70 320.1 bdl 10.95 12.81 1.80 0.044 0.045 0.277 bdl 0.039 0.039 0.590 0.658 0.335 1.388 0.268 2.286 0.497 1.483 0.226 1.545 0.206 0.20 0.106 9.56 30.04 372.5 bdl 1.58 10.65 1.36 0.056 0.038 0.394 bdl 0.037 0.036 0.514 0.547 0.263 1.133 0.238 1.836 0.412 1.237 0.178 1.232 0.175 0.15 0.123 7.84 23.78 363.1 bdl 0.70 11.65 1.53 0.048 bdl 0.148 0.006 0.036 0.037 0.578 0.616 0.294 1.236 0.247 2.047 0.457 1.371 0.198 1.407 0.188 0.17 0.018 8.72 18.98 326.6 bdl 2.38 13.56 4.23 0.068 bdl 0.127 0.015 0.157 0.093 1.084 0.847 0.403 1.515 0.307 2.335 0.498 1.519 0.218 1.531 0.208 0.27 0.036 10.73 21.68 405.3 bdl 2.19 12.71 4.68 0.057 0.121 0.194 bdl 0.121 0.077 0.895 0.722 0.341 1.434 0.284 2.286 0.509 1.520 0.228 1.500 0.202 0.32 0.085 10.12 17.96 317.1 bdl 2.28 13.02 4.04 0.068 0.076 0.189 0.013 0.150 0.088 1.006 0.787 0.395 1.497 0.310 2.355 0.533 1.601 0.221 1.562 0.207 0.28 0.040 10.73 18.69 317.1 bdl 2.91 13.16 3.89 0.069 bdl 0.063 0.006 0.153 0.095 1.066 0.802 0.377 1.469 0.286 2.216 0.482 1.443 0.204 1.510 0.201 0.24 bdl 10.31 24.58 375.8 bdl 2.50 10.34 2.30 0.084 0.175 0.439 0.004 0.122 0.069 0.823 0.627 0.294 1.211 0.230 1.802 0.401 1.168 0.177 1.161 0.153 0.167 0.075 8.24 22.52 380.4 bdl 2.66 9.49 2.62 0.076 0.068 1.242 0.002 0.104 0.060 0.670 0.551 0.239 1.080 0.224 1.704 0.373 1.093 0.161 1.098 0.142 0.226 0.190 7.50 21.58 340.4 0.11 1.79 12.47 3.39 0.081 bdl 0.241 0.004 0.137 0.083 0.918 0.743 0.326 1.403 0.272 2.208 0.486 1.400 0.210 1.379 0.188 0.259 0.030 9.76 28.95 369.2 0.12 2.23 11.59 3.31 0.082 0.444 0.478 0.004 0.123 0.074 0.883 0.683 0.304 1.310 0.259 2.073 0.440 1.343 0.202 1.337 0.177 0.230 0.060 9.21 23.93 379.0 bdl 1.83 10.86 2.78 0.039 bdl 0.041 0.004 0.129 0.080 0.926 0.760 0.334 1.330 0.256 1.990 0.438 1.247 0.189 1.247 0.168 0.23 bdl 9.10 22.29 370.2 bdl 1.73 12.23 3.06 0.046 0.050 0.076 0.003 0.148 0.090 1.028 0.793 0.388 1.439 0.284 2.210 0.492 1.393 0.202 1.408 0.196 0.25 bdl 10.07 22.88 373.7 bdl 1.67 12.33 3.49 0.038 bdl 0.029 bdl 0.127 0.084 1.005 0.792 0.356 1.429 0.300 2.264 0.502 1.418 0.206 1.352 0.190 0.30 bdl 10.02 20.94 350.6 0.09 2.33 13.22 3.70 0.042 0.137 0.486 0.003 0.133 0.088 1.012 0.804 0.371 1.552 0.298 2.382 0.512 1.516 0.224 1.513 0.206 0.30 bdl 10.62 G.52 G.27 Tv.40 Rock type Harzburgites Sample no. Tv.60 Clinopyroxenite in hydrous peridotites SiO2 TiO2 Al2O3 Cr2O3 FeOa MnO MgO CaO Na2O K2O NiO Total 54.97 0.06 2.56 0.75 2.06 0.08 17.82 24.00 0.01 0.01 0.03 102.3 53.40 0.03 2.97 0.97 2.19 0.07 17.35 23.76 0.02 0.00 0.05 100.8 54.80 0.04 2.42 0.66 2.13 0.07 18.16 23.96 0.01 0.01 0.06 102.3 54.80 0.04 2.42 0.66 2.13 0.07 18.16 23.96 0.01 0.01 0.06 102.3 54.08 0.06 2.84 0.85 1.85 0.07 17.49 24.55 0.02 0.01 0.06 101.9 52.58 0.08 2.98 0.90 1.94 0.08 17.21 24.20 0.02 0.00 0.04 100.0 52.54 0.06 3.59 1.04 2.23 0.09 17.17 24.06 0.01 0.00 0.07 100.9 55.90 0.05 0.73 0.10 1.51 0.06 17.85 25.29 0.04 0.00 0.01 101.5 55.18 0.04 0.76 0.16 1.51 0.05 17.52 25.15 0.04 0.00 0.01 100.4 55.97 0.04 0.60 0.12 1.48 0.05 17.82 25.91 0.01 0.00 0.02 102.0 55.40 0.04 1.15 0.24 1.99 0.08 17.94 23.92 0.11 0.01 0.03 100.9 Cpx-Mg#b Spl-Cr#c 0.939 0.456 0.934 0.358 0.938 0.391 0.938 0.407 0.944 0.294 0.941 0.398 0.932 0.400 0.955 0.115 0.954 0.116 0.956 0.142 0.941 0.146 Trace elements (ppm) Li B Sc Ti V Cr Co Ni Rb Sr Y Zr Nb Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy 6.21 4.24 51.92 355 220.6 6004 24.56 393.7 0.064 0.483 2.231 0.044 0.042 0.090 0.203 bdl bdl bdl bdl bdl 0.007 0.073 0.027 0.296 6.69 2.11 53.01 353 229.8 7806 23.95 405.5 0.154 0.049 2.310 0.049 0.042 0.153 bdl bdl bdl bdl bdl 0.014 0.006 0.081 0.027 0.313 5.79 3.65 54.25 353 218.6 5945 23.65 389.9 bdl 0.796 2.141 0.043 0.037 0.034 bdl bdl bdl bdl bdl bdl 0.006 0.077 0.024 0.311 7.19 2.68 52.36 341 213.2 5458 23.21 390.7 0.149 0.625 2.009 0.035 0.034 0.244 0.277 bdl bdl bdl bdl bdl 0.008 0.071 0.024 0.269 4.04 0.97 51.17 502 229.0 7732 21.22 374.6 bdl 0.076 3.510 0.074 0.059 bdl bdl bdl 0.003 bdl bdl 0.029 0.019 0.167 0.048 0.525 4.84 1.91 53.09 530 220.5 6945 20.29 361.4 0.173 0.077 3.653 0.051 0.057 0.035 bdl bdl bdl bdl bdl 0.027 0.020 0.167 0.049 0.555 4.37 1.18 52.20 512 223.6 7250 20.79 361.5 0.101 0.076 3.461 0.054 0.061 0.084 0.037 bdl 0.004 bdl bdl 0.032 0.016 0.142 0.048 0.496 11.62 14.31 26.28 282 41.5 1171 15.91 221.5 bdl 6.930 0.569 0.949 0.008 0.061 0.252 0.043 0.089 0.012 0.081 0.041 0.018 0.071 0.014 0.111 11.04 16.90 30.23 324 49.4 1117 14.91 212.7 bdl 7.683 0.886 1.356 bdl 0.032 0.491 0.071 0.134 0.019 0.117 0.062 0.025 0.118 0.022 0.173 9.05 14.46 30.12 304 45.3 929 14.53 200.6 bdl 8.512 0.782 1.186 0.007 bdl 0.524 0.055 0.113 0.018 0.106 0.055 0.024 0.085 0.023 0.161 10.87 13.20 30.24 311 48.5 1241 15.39 217.8 bdl 7.613 0.923 1.364 0.006 bdl 0.072 0.063 0.131 0.019 0.120 0.059 0.027 0.110 0.024 0.181 Tv.136 G.43 M 1249 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Table 2 (continued) Rock type Harzburgites Sample no. Tv.60 Clinopyroxenite in hydrous peridotites Ho Er Tm Yb Trace elements (ppm) Lu Hf Pb Th U ΣREE 0.089 0.318 0.053 0.391 0.094 0.318 0.054 0.423 0.085 0.308 0.049 0.382 0.079 0.289 0.048 0.368 0.138 0.479 0.074 0.593 0.143 0.497 0.079 0.570 0.132 0.473 0.070 0.554 0.025 0.069 0.011 0.091 0.038 0.107 0.017 0.124 0.033 0.095 0.016 0.105 0.039 0.112 0.018 0.125 0.056 bdl bdl bdl bdl 1.310 0.062 bdl bdl bdl bdl 1.391 0.054 bdl 0.082 bdl bdl 1.295 0.053 bdl 0.079 bdl bdl 1.210 0.079 bdl 0.069 bdl bdl 2.153 0.079 bdl 0.091 bdl bdl 2.187 0.078 bdl 0.080 bdl bdl 2.048 0.012 0.039 0.077 bdl bdl 0.688 0.019 0.047 0.086 bdl 0.004 1.045 0.015 0.038 0.082 bdl bdl 0.904 0.017 0.048 0.077 bdl bdl 1.046 Tv.136 G.43 M Note that Ta (b0.006 ppm), Th (b0.007 ppm), and U (b0.005 ppm) are not listed in the table (below detection limits). a Total iron given as FeO. b Cpx-Mg#, [Mg# = (Mg/(Mg + Fe2+)] of clinopyroxene. c Spl-Cr#, [Cr# = (Cr/(Cr + Al)] of spinel. Table 3 Major (wt.%) and trace element (ppm) abundances of orthopyroxenes in Sarami basal peridotites. Rock type Type I lherzolites Type II lherzolites Harzburgites Sample no. Tv.59 Tv.59 G.39 G.39 Tv.125 Tv.125 Tv.123 Tv.123 G.52 G.52 G.27 G.27 Tv.136 Tv.136 Tv.60 Tv.60 SiO2 TiO2 Al2O3 Cr2O3 FeOa MnO MgO CaO Na2O K2O NiO Total 53.17 0.07 5.03 0.68 6.21 0.16 31.20 2.10 0.05 0.05 0.07 98.8 52.99 0.08 5.46 0.80 6.33 0.13 31.80 1.47 0.02 0.00 0.09 99.2 55.16 0.09 5.88 0.66 6.33 0.16 32.90 0.74 0.00 0.00 0.08 102.0 54.49 0.08 5.50 0.60 6.41 0.13 33.06 0.87 0.01 0.00 0.08 101.2 54.89 0.08 6.05 0.56 6.43 0.14 32.35 1.51 0.03 0.00 0.08 102.1 55.02 0.09 5.50 0.51 6.44 0.13 32.70 0.62 0.03 0.00 0.07 101.1 55.80 0.08 5.33 0.59 6.38 0.16 33.09 0.76 0.02 0.00 0.10 102.3 55.01 0.07 6.01 0.67 6.42 0.09 33.03 0.56 0.02 0.00 0.07 102.0 55.51 0.08 5.08 0.61 6.14 0.14 33.52 0.55 0.05 0.00 0.07 101.7 53.38 0.13 5.74 0.64 5.72 0.15 31.28 1.81 0.07 0.00 0.09 99.0 53.64 0.09 5.06 0.58 6.36 0.13 33.34 1.08 0.02 0.00 0.09 100.4 53.66 0.07 6.47 0.80 6.10 0.11 32.53 1.11 0.07 0.00 0.09 101.0 57.10 0.04 2.88 0.70 5.46 0.12 34.25 1.34 0.00 0.00 0.08 102.0 57.12 0.03 2.87 0.68 5.53 0.15 34.46 0.92 0.00 0.00 0.08 101.9 57.43 0.03 2.35 0.63 5.79 0.10 34.61 0.85 0.00 0.01 0.08 101.9 57.72 0.01 2.53 0.62 5.86 0.15 34.85 0.68 0.00 0.00 0.10 102.5 Mg#b 0.900 0.900 0.903 0.902 0.900 0.901 0.902 0.902 0.907 0.907 0.903 0.905 0.918 0.917 0.914 0.914 Trace elements Li B Sc Ti V Cr Co Ni Rb Sr Y Zr Nb Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Pb ΣREE (ppm) 3.99 2.22 23.22 435.2 140.9 4066 51.57 755.1 0.600 1.191 0.795 0.069 0.029 0.421 0.385 bdl 0.001 bdl bdl 0.006 0.004 0.027 0.007 0.086 0.026 0.117 0.023 0.224 0.041 0.013 0.037 0.56 4.00 0.81 24.88 520.2 149.3 5788 56.25 720.7 0.109 0.622 1.223 0.100 0.042 0.042 0.251 bdl bdl bdl 0.010 0.017 0.010 0.055 0.016 0.156 0.042 0.169 0.028 0.265 0.043 0.019 0.041 0.81 3.89 0.64 22.42 741.5 141.9 4940 58.86 770.5 bdl 0.087 1.675 0.491 0.058 bdl 0.247 bdl 0.005 0.003 0.039 0.037 0.019 0.101 0.025 0.230 0.062 0.223 0.042 0.334 0.056 0.044 0.033 1.18 4.16 1.93 22.94 779.4 144.0 5090 56.76 751.4 bdl 0.275 1.899 0.509 0.059 0.018 0.182 0.001 0.007 0.004 0.055 0.056 0.026 0.141 0.032 0.281 0.072 0.245 0.043 0.354 0.057 0.045 0.032 1.37 1.75 0.75 21.59 752.1 143.4 4484 58.96 723.1 bdl 0.083 1.422 0.478 0.035 0.040 0.026 bdl 0.004 0.003 0.037 0.037 0.017 0.085 0.020 0.202 0.052 0.195 0.032 0.295 0.051 0.041 0.018 1.03 2.09 1.93 21.28 630.1 141.3 4356 59.50 720.5 0.093 0.337 1.122 0.325 0.034 0.096 0.200 bdl 0.003 0.002 0.020 0.023 0.012 0.056 0.013 0.147 0.041 0.160 0.031 0.265 0.046 0.031 0.074 0.82 2.26 6.98 13.46 295.4 99.1 3742 73.32 1032.3 0.171 1.462 0.526 0.108 0.026 0.074 0.294 bdl 0.001 bdl 0.008 0.007 0.004 0.018 0.005 0.063 0.017 0.077 0.017 0.150 0.030 0.010 0.031 0.40 1.59 1.58 22.47 640.1 154.5 5541 63.00 774.2 bdl 0.050 1.421 0.271 0.039 bdl bdl bdl 0.001 0.001 0.017 0.024 0.013 0.074 0.019 0.196 0.049 0.194 0.034 0.288 0.048 0.036 0.061 0.96 2.43 1.25 19.33 570.4 117.7 4494 54.21 694.6 bdl 0.082 1.038 0.386 0.044 0.131 bdl bdl 0.004 0.002 0.027 0.023 0.010 0.057 0.013 0.137 0.036 0.146 0.027 0.240 0.043 0.029 0.018 0.77 2.93 0.49 22.49 745.9 129.5 4949 55.68 737.1 bdl 0.120 1.799 0.684 0.048 bdl bdl bdl 0.007 0.004 0.044 0.040 0.024 0.105 0.028 0.242 0.066 0.238 0.040 0.336 0.057 0.045 0.038 1.23 3.861 1.883 19.84 601.0 134.3 4438 51.89 654.4 0.049 0.207 1.065 0.319 0.050 0.060 0.069 bdl 0.004 0.002 0.024 0.023 0.011 0.055 0.015 0.147 0.039 0.152 0.028 0.242 0.041 0.029 0.037 0.78 4.097 1.076 21.98 747.3 145.4 6086 61.27 813.6 bdl 0.369 1.437 0.709 0.066 bdl 0.078 0.002 0.013 0.005 0.044 0.035 0.016 0.085 0.020 0.201 0.054 0.191 0.033 0.279 0.047 0.058 0.035 1.03 1.93 bdl 25.51 202.2 120.7 5530 58.60 755.8 bdl bdl 0.405 0.039 0.042 bdl bdl bdl bdl bdl bdl bdl bdl 0.007 0.003 0.042 0.015 0.067 0.013 0.132 0.025 bdl 0.037 0.30 1.78 0.28 27.16 213.1 126.5 5487 56.76 712.8 0.167 0.004 0.416 0.040 0.040 0.057 bdl bdl bdl bdl bdl bdl bdl 0.007 0.003 0.041 0.014 0.070 0.016 0.147 0.026 bdl 0.054 0.32 2.22 1.38 28.86 164.8 127.1 5050 58.14 716.9 0.234 0.007 0.275 0.027 0.028 0.227 0.009 bdl bdl bdl bdl bdl bdl bdl 0.002 0.023 0.010 0.048 0.011 0.120 0.024 bdl 0.036 0.24 3.05 1.17 26.38 148.5 126.5 4489 57.59 695.0 0.365 0.008 0.197 0.025 0.026 0.188 bdl bdl bdl bdl bdl bdl bdl bdl 0.001 0.015 0.007 0.038 0.009 0.101 0.020 bdl 0.049 0.19 Note that Ta (b0.002 ppm), Th (b0.002 ppm), and U (b0.001 ppm) are not listed in the table (below detection limits). a Total iron given as FeO. b Mg#, [Mg# = Mg/(Mg + Fe2+)]. 1250 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 3. Major oxides versus MgO variation diagrams for the bulk Sarami basal peridotites, compared with abyssal peridotites (Niu, 2004) and Fizh peridotites (Takazawa et al., 2003). Note that correlations of oxides with MgO reflect partial-melting degree of lherzolites (b15%) and harzburgites (b25%) based on model of Niu (1997). Primitive mantle (PM) compositions are after McDonough and Frey (1989), McDonough and Sun (1995) and Niu (1997). Abyssal lherzolites (Lrz) and harzburgites (Hrz) from Pacific and Indian Ocean ridge-transform systems are after Niu (2004). Fizh Type I lherzolites (Lrz) and Type 11 lherzolites in northern Oman ophiolite are obtained from Takazawa et al. (2003). Oman harzburgite (Hrz) data from central to north Oman ophiolites are after Monnier et al. (2006). The major-element concentrations are recalculated to 100% on the LOI-free basis. Mg# (0.88–0.93), SiO2 (42.7–52.2 wt.%), Al2O3 (6.0–15.4 wt.%), Cr2O3 (0.5–1.4 wt.%), Na2O (0.9–3.3 wt.%) and TiO2 (0.2–1.0 wt.%) (Supplementary Data 3; Fig. 9). Tremolites in peridotites with high Mg# (0.90–0.95) show wide ranges of SiO2 (46.8–57.4 wt.%), Al2O3 (1.83–5.9 wt.%), Na2O (0.03–1.0 wt.%) and Cr2O3 (0.24–1.6 wt.%) (Supplementary Data 3). 4.2.2. Trace elements Cpx REE contents (ΣREE = 5.17−10.14 ppm) and MREE/HREE ratio (Sm/Yb)N = 0.16–0.56) in Type I lherzolites are slightly lower than those in Type II [(ΣREE = 7.5 to 10.73 ppm; (Sm/Yb)N =0.46–0.67)] (Table 2), but the two lherzolites have nearly the same ratio of Cpx LREE/HREE [e.g., (Ce/Yb)N = 0.007 to 0.028]. Cpx in harzburgites is depleted in REE (ΣREE = 1.2 to 2.2 ppm), and shows lower MREE/HREE ratio [(Sm/Yb)N = 0.0–0.06] than Cpxs in lherzolites (Table 2). The CI-normalized REE patterns of Cpxs in lherzolites and harzburgites, convex upward (0.005–10 times CI), display high depletions in LREE (0.005–1 times CI) without inflection at La and Ce, in contrast to Cpx in Fizh lherzolites (Type I), which were interpreted to be modified by melt refertilization (Takazawa et al., 2003) (Fig. 10a). They also differ from flat to spoon-shaped REE patterns of the Cpx lens (Fig. 2e and f) in hydrous lherzolites (Fig. 10c). REE concentrations and patterns of Cpx in Type II lherzolites are similar to those in Fizh Type II lherzolites and abyssal peridotites from the normal ridge segment (Johnson et al., 1990) (Table 2; Fig. 10a). The Cpx REE of Type I lherzolites lie in the gap between the Fizh Type I and II lherzolites of Takazawa et al. (2003) (Fig. 10a). Cpx in harzburgites exhibits REE patterns parallel to those of Cpxs in both lherzolite types with lower concentrations (Fig. 10a). Our Cpxs are enriched in fluid-mobile elements (e.g., B, Li, Cs and Pb; 1–200 times PM) and depleted in HFSE (e.g., Ta, Nb, Th and Zr; b0.6 times PM) + U (Fig. 10b, d) in the same way of their corresponding host rock (Fig. 5b, d). Opx REE patterns (Fig. 11a, c) show the same general characteristics with those of Cpxs; they exhibit depletion in LREE (0.001–0.1 times CI), in contrast to Opx LREE in Fizh Type I lherzolites (Fig. 11a). Opxs in harzburgites are low in MREE to LREE (below detection limits) (Table 3; Fig. 11a). PM-normalized multi-element patterns of Opxs show high concentrations of fluid-mobile elements (e.g., Pb, Ba, Cs and Li) and Ti relative to adjacent elements, but display low values of Th, Y and Ta (below detection limits) (Table 3; Fig. 11b, d). Pargasitic hornblendes are also highly depleted in LREE (0.01–2 times CI) (Fig. 12a), and show similar REE characteristics to their associated Cpx but with slightly high REE (ΣREE = 5 to 16.5 ppm) concentrations (Supplementary Data 3; Fig. 12a). They are enriched in some fluid-mobile elements (e.g., B, Li, Cs and Pb; 0.7–300 times PM) and depleted in HFSE (e.g., Ta, Nb, Th and Zr; b 0.6 times PM) + U (Fig. 12b), similar to those in their host rocks and associated Cpxs (Figs. 5 and 10b, d). Olivines are highly depleted in trace elements (below detection limits) but are the main host for Ni and Co. They are also enriched in fluid-mobile elements (e.g., B, Pb, Cs and Ba) relative to HFSE (see Supplementary Data 3). We selected five peridotite samples (Tv.59, Tv.56, Tv.123, Tv.136 and G.27), which contain the same modal amount of Cpxs (Supplementary Data 1), and calculated Cpx from bulk rocks (scheme of Niu, 1997; Table 1) for mass balance calculation based on in-situ analyses of olivine and pyroxenes (Tables 2 and 3; Supplementary Data 3). This calculation suggests that Cpxs contribute greatly for most trace elements to the whole-rock characteristics, while Opxs are the main source for Cr, V, M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 1251 Fig. 4. Conservative trace elements (ppm) versus MgO (wt%) variation diagrams for bulk Sarami basal peridotites. HREE, V, Sc, Co and Ni show systematic correlations with MgO, reflecting partial-melting trend. The better negative correlations of HREE and Sc with MgO reflect partial melting in the spinel field. Fields of Fizh Type I lherzolites (Lrz) and Type II lherzolites are the same as in Fig. 3. Fig. 5. The chondrite (CI)-normalized REE patterns and primitive (PM)-normalized multi-element patterns for the bulk Sarami basal peridotites in central Oman ophiolite. REE patterns (a, c) and spider diagram of trace elements (b, d). The Sarami peridotites show spoon-shaped REE patterns and enrichment in fluid-mobile elements with spikes at Cs and Sr coupled with depletion of HFSE (Ta, Hf, Zr, Nb, and Th) + U. Note that Pb and Th (b0.05 ppm), U and Ta (b0.01 ppm), Hf (b0.1 ppm), Rb (b1.0 ppm), Nb (b0.2 ppm) analyses (not displayed in figure) are below detection limits for all samples. Abyssal peridotites from the Pacific and Indian Ocean are after Niu (2004), and peridotites (mainly harzburgites) from central and northern Oman ophiolite are after Monnier et al. (2006). Normalized primitive mantle (PM) values are from McDonough and Sun (1995), and chondrite (CI)-normalized values are from Anders and Grevesse (1989). 1252 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 6. Chemical characteristics of pyroxenes, olivines and spinels in the Sarami peridotites. (a) Mg# versus Na2O of Cpx. (b) Mg# versus Al2O3 of Opx. Note fields of Fizh Type I and II lherzolites (Lrz) and harzburgites (Hrz) in northern Oman ophiolite (Takazawa et al., 2003), central and northern Oman harzburgites (Om Hrz; Monnier et al., 2006), Tayin harzburgites from south Oman (Hanghøj et al., 2010) and abyssal peridotites (Johnson et al., 1990; Bonatti et al., 1992; Johnson and Dick, 1992; Batanova et al., 1998; Hellebrand et al., 2002) are shown for comparison. (c) Forsterite (Fo) vs. NiO content for olivines. (d) Forsterite (Fo) vs. Cr# of spinels. OSMA (olivine–spinel mantle array) is a spinel peridotite restite trend, and melting trend (annotated by % melting) of Arai (1994). The studied peridotites occupy the whole field of abyssal peridotites. Fields of chromian spinels in forearc peridotites (Ishii et al., 1992; Parkinson and Pearce, 1998), and abyssal peridotites (Arai, 1994) are shown for comparison. Sc, Nb and Rb. In addition, olivines contribute for B, Ni, Co and Cs to their whole-rock budget. The calculated bulk-rock compositions for Cr, Sr, Cs and Ba are very low compared to the measured concentrations (Table 1). This discrepancy may be explained by minerals neglected for the calculation: primary spinels are the principal host for Cr and serpentines could contain a significant amount of LILE that are possibly concentrated along mineral-grain boundaries. Finally, most of bulkrock trace elements reside in Cpxs that form up to 14.0 modal vol.% of our peridotite samples (see Supplementary Data 1). 5. Discussion 5.1. Tectonic setting for Sarami basal peridotites The tectonic setting and the polygenetic origin of the Oman ophiolite have been widely discussed for many years: the main portion of Oman mantle sections is of a mid-ocean ridge origin (e.g., Boudier et al., 1988; Nicolas, 1989; Kelemen et al., 1995; Takazawa et al., 2003; Monnier et al., 2006), however some Oman peridotites are of an arc origin (e.g., Arai et al., 2006; Tamura and Arai, 2006). Khedr et al. (2013) reported that the Sarami basal lherzolites are similar in mineral compositions to abyssal peridotites formed in a mid-ocean ridge after Arai (1994) (Fig. 6d). The Sarami basal peridotites plot in the entire chemical field of abyssal peridotites (Fig. 6) and show that large chemical variations may occur in a small scale (b0. 4 km, Fig. 7). Our samples are localized in three places and cannot give a general view of the tectonic setting of the entire Oman ophiolite, but only give us a local context for the formation and evolution of the Wadi Sarami peridotites. HREE, HFSE, Sc and Cr are relatively immobile during alteration processes and can be used to know the tectonic setting for the peridotite genesis (Tatsumi et al., 1986; Pearce and Parkinson, 1993; Kogiso et al., 1997; Bizimis et al., 2000). In our samples, the CI-normalized REE patterns of Cpxs, which show depletion in LREE (Figs. 10a, c and 13a), are typically similar to those of residual Cpx in abyssal peridotites (Johnson et al., 1990; Johnson and Dick, 1992; see Fig. 13a). Yb, Dy, Ti and Cr concentrations in Sarami Cpxs (see Fig. 14) suggest their residual character, which is similar to that of abyssal peridotites (Johnson et al., 1990; Johnson and Dick, 1992; Dick and Natland, 1996; Ross and Elthon, 1997; Hellebrand et al., 2001, 2002) and lies far away from supra-subduction zone (SSZ) peridotites (Fig. 14a). This is consistent with the similarity in the whole-rock major and trace elements (e.g., Yb, Dy, Ti and Cr) between the Sarami peridotites and abyssal peridotites from Pacific and Indian Ocean ridges M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 7. Vertical mineral chemical profiles of Sarami basal peridotites in terms of the distance from the metamorphic sole contact upward to the harzburgitic mantle. Note the heterogeneous compositions of the basal mantle section in Cpx Al2O3, Cpx Na2O, Cpx Mg#, spinel Cr# and Opx Mg#, almost systematically changed upward to harzburgites. The dash line indicates the spread of the element (e.g., Al, Na) within one sample. Cpxs appear to decrease in a modal amount and in Na2O content from basal lherzolites to harzburgites. 1253 1254 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 8. Fine textural characteristics of coarse subhedral Cpx in Type I lherzolites. Distribution of Ca (a), Al (b), Mg (c) and Cr (d) within the Cpx crystal in a sample (G. 35), showing exsolution of Opx (a, c) and Al-rich spinel (b, d). Warm colors indicate higher concentrations than cooler colors. (e) Back-scattered image of the same Cpx grain containing exsolution lamellae of Opx and blebs of Al-rich spinels. (f) Back-scattered image of coarse exsolution lamellae of Al-rich spinels associated with Opx lamellae in Cpx from a Type I lherzolite (G.33). Abbreviations as in Fig. 2. (Niu, 2004; Figs. 3, 5 and 15). These results suggest that the studied basal peridotites represent a fragment of the Tethyan oceanic mantle, which was detached along the oceanic fracture zones (e.g., seafloor spreading- related faults) and exposed along transform or detachment faults (Fig. 16). The Sarami basal peridotites were underlain by the metamorphic sole (mafic Tethyan oceanic crust and various sedimentary rocks) Fig. 9. Ca, Al and Mg distribution maps of coarse Cpx grains in Type II lherzolite (G.27) (a−c) and in Type I lherzolite (G.39) (d−f). Warmer colors show higher concentrations than cooler colors. Note that Cpx is replaced in part by hornblende. Some relics of Cpx have still survived in core of hornblende plates (a). Abbreviations as in Fig. 2. M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 1255 Fig. 10. CI-normalized REE patterns (a, c) and PM-normalized trace-element patterns (b, d) for Cpxs in Sarami basal peridotites. Cpxs display convex-upward REE patterns with highly depleted LREE, and differ from spoon-shaped REE patterns of Cpx from the small clinopyroxenite lens in Sarami hydrous peridotites (a, c). Cpxs are enriched in fluid-mobile elements (e.g., B, Li, Cs and Pb) and depleted in HFSE (e.g., Ta, Nb, Th and Zr) + U (b, d). Fizh Type I lherzolites (Lrz) and Type II lherzolites in northern Oman ophiolite are after Takazawa et al. (2003). Data of abyssal peridotites from Coast Range Ophiolite in western California is obtained from Jean et al. (2010). Normalized PM values are from McDonough and Sun (1995), and CI-normalized values are from Anders and Grevesse (1989). during the early intraoceanic thrusting stage of the Oman ophiolite emplacement (e.g., Ishikawa et al., 2005 and references therein). The occurrences of younger boninites (Ishikawa et al., 2002) and SSZ-related peridotites (e.g., Arai et al., 2006; Tamura and Arai, 2006) are not in contradiction with our results, which support the polygenetic origin of the Oman ophiolite. A possible transition from spreading ridge to arc setting observed in the Oman ophiolite (e.g., Ishikawa et al., 2002) has been also reported in other Tethyan ophiolites, like ophiolites in Albania (e.g., Dilek et al., 2008) and the Coast Range ophiolite in California (Hirauchi et al., 2008; Jean et al., 2010). However, in Sarami and Wuqbah blocks, the SSZ influence is possibly represented by small andesitic basalt extrusives in the upper crust, and the presence of boninites or SSZ-type peridotites has not been reported. We propose that fertile abyssal peridotites along Wadi Sarami were exposed along Fig. 11. Trace-element characteristics of Opx in Sarami basal peridotites. (a, c) CI-normalized REE patterns showing highly depleted LREE. (b, d) PM-normalized trace-element patterns showing enrichment in fluid-mobile elements (e.g., B, Li, Cs, Rb and Pb) and depletion in Ta, Th, Zr and U. Fizh Type I lherzolites (Lrz) and Type II lherzolites in northern Oman ophiolite are after Takazawa et al. (2003). Normalized values of CI and PM are the same as those used in Fig. 10. 1256 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 12. The CI-normalized REE patterns (a) and PM-normalized trace-element patterns (b) for Sarami hornblendes in basal peridotites. Hornblendes display convex-upward REE patterns with highly depleted LREE, like their precursor Cpxs REE patterns (a). Hornblendes are enriched in fluid-mobile elements (e.g., B, Li, Cs and Pb) and depleted in Ta, Nb, Th, Zr and U, like their precursor Cpxs (b). Trace elements of hornblendes are compared with those of Cpxs in Sarami lherzolites (Lrz) and harzburgites (Hrz). Normalized values of CI and PM are the same as those used in Fig. 10. oceanic fracture zones, and possibly represent the main mantle peridotite in the central Oman ophiolite. Other abyssal peridotites from significantly remote locations (e.g., Fizh block in northern Oman Fig. 13. Fractional melting models and calculated melts in equilibrium with Cpxs for the Sarami basal peridotites. (a) CI-normalized REE of Cpxs following non-modal fractional melting in the spinel field. The calculation method and parameters used are from Johnson et al. (1990) and Sano and Kimura (2007), respectively. The best fit for the HREE in Cpxs suggests 1–5% melting for Type II lherzolite, 5–10% melting for Type I lherzolite and ~15% melting for harzburgite in the spinel field. The fractional melting model at 4% fractional melting in the garnet field followed by 10% melting in the spinel field is after Hellebrand et al. (2002). Cpxs in abyssal peridotites (Johnson et al., 1990; Johnson and Dick, 1992) are plotted for comparison. (b) CI-normalized REE patterns for calculated melts in equilibrium with Sarami Cpxs. Cpx/melt partition coefficients used are obtained from Hart and Dunn (1993) and Stosch (1982). The REE in melts calculated to be in equilibrium with Cpx from Oman dunites and harzburgites were drawn in straw yellow and dark gray fields, respectively (Kelemen et al., 1995). Normal mid ocean ridge basalt (N-MORB) (Hofmann, 1988; Sun and McDonough, 1989) and depleted mid ocean ridge basalt (D-MORB) from the Mid-Atlantic Ridge (Frey et al., 1993) are used for comparison. The CI values are from Anders and Grevesse (1989). ophiolite) may have been modified by re-melting and metasomatism under arc (SSZ) setting that formed boninites and/or SSZmantle type peridotites (e.g., Arai et al., 2006; Tamura and Arai, 2006) during intra-oceanic collapse and closure of a seafloor spreading ridge (e.g., Dilek et al., 2008). It is well known that the evolution of ophiolites is rather different between the Wadi Tayin massif (south Oman) and Fizh massif (north Oman) (Nicolas et al., 1988, 2000; Godard et al., 2000; Takazawa et al., 2003; Arai et al., 2006; Tamura Fig. 14. Partial melting for the Sarami peridotites based on mineral chemistry. (a) Dy−Ti in Cpxs. Dry melting is residual clinopyroxene compositions during dry melting (incremental batch melting at 0.1% increments) of a MORB source; amphibole melting is melting of a MORB-depleted source in the presence of amphibole and hydrous melting is melting in the presence of fluids (Bizimis et al., 2000). Fields for Cpx in supra-subduction zone (SSZ) (Bizimis et al., 2000 and references therein) and Cpx in abyssal peridotites (Johnson et al., 1990; Johnson and Dick, 1992) are used for comparison. (b) Relationship between spinel Cr# and Cpx Yb (ppm). Lherzolites and harzburgites underwent fractional melting less than 10% and 17% melting, respectively. The degree of partial melting based on spinel compositions is after Hellebrand et al. (2001). Data of abyssal peridotites were compiled from the literature (Dick and Bullen, 1984; Johnson et al., 1990; Johnson and Dick, 1992; Dick and Natland, 1996; Ross and Elthon, 1997; Hellebrand et al., 2001, 2002). The Sarami basal peridotites show wide range of partial melting degrees, and are similar in compositions to abyssal peridotites. M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 1257 5.3. Origin of mantle heterogeneity It is well known that the Earth's mantle is heterogeneous based on volcanic rock studies (e.g., Schilling, 1985; Hofmann, 1997; Brandl et al., 2012). But, only few petrological studies have been conducted on abyssal peridotites to refer to mantle heterogeneities (Dick et al., 1984; Michael and Bonatti, 1985; Hellebrand et al., 2002; Brunelli et al., 2006). We observed a lithological change from Type II lherzolites to harzburgites through Type I lherzolites or from Type I lherzolites to harzburgites on various scales from few meters to few hundreds of meters (b 0.4 km, Khedr et al., 2013). This lithological variety is probably related to small-scale (b 1 m to few tens of meters) to middle scale (b0.4 km) mantle heterogeneities. This change is in accordance with the compositional change of pyroxenes and spinel (Tables 1−3; Figs. 6 and 7) with a distance from the base (the sole contact). Fig. 15. Estimation of the degree of partial melting based on bulk-rock chemistry. (a) Dy (ppm) versus Ti (ppm) relations. MORB melting field shows residual peridotite compositions during anhydrous melting (incremental batch melting at 0.1% increments) of a MORB source (Bizimis et al., 2000). Refertilization hydrous melting trend exhibits residual peridotite compositions during refertilization-hydrous melting of the previously depleted source (residue of 9% anhydrous melting) calculated using the model of Bizimis et al. (2000) and numbers along the lines indicating percent melting (Barth et al., 2008). The partial-melting degrees are b10% melting for lherzolites and b20% for harzburgites, in consistent with those obtained from Cpx chemistry. (b) Yb (ppm) versus Cr (ppm) relations. FMM is fertile MORB mantle (Pearce and Parkinson, 1993 and references therein). Melting trend is after Pearce and Parkinson (1993). Abyssal peridotites from Pacific and Indian Ocean ridge are after Niu (2004). and Arai, 2006; Hanghøj et al., 2010). Finally, the tectonic evolution and origin of peridotites in the central Oman ophiolite (MORB type) are different with those of peridotites in Wadi Fizh (SSZ type + MORB type) (e.g., Arai et al., 2006; Tamura and Arai, 2006). 5.2. Thermometry of basal peridotites The two-pyroxene thermometer of Wells (1977) yields equilibrium temperatures of 817 ± 21 °C and 841 ± 61 °C for Type I and Type II lherzolites, respectively. These temperatures are in the same range to those calculated using the two-pyroxene thermometer of Brey and Kohler (1990): approximately 744 ± 45 °C for Type I and ~825 ± 91 °C for Type II lherzolites under 15 kbar. These results are consistent with the equilibrium temperature calculated using Fe−Mg exchange between olivine and spinel (YCr and ln KD) after the modal of Evans and Frost (1975), giving on average 780 ± 72 °C and 800 ± 95 °C °C for Type I and II lherzolites, respectively. There is no significant difference in equilibrium temperature between Type I and II lherzolites. The existence of hornblende replacing Cpx in both Types I and II lherzolites and tremolite replacing Opx (Fig. 9), possibly points to moderate temperatures of about 700 to 800 °C (in the amphibolite facies) (e.g., Evans, 1977; Ohara and Ishii, 1998; Khedr and Arai, 2010 and references therein). This is consistent with low to moderate equilibrium temperature (755 to 878 °C) of abyssal peridotites from the Coast Range in California (Hirauchi et al., 2008). Hence, the late alteration and metasomatism have possibly started from less than 800 °C to lower temperature, forming low-T serpentines (b400 °C) (lizardite and chrysotile) in Sarami lherzolites during obduction of the Oman ophiolite (e.g., Evans, 2004; Khedr and Arai, 2010, 2011). 5.3.1. Petrogenesis of basal peridotites: are Cpxs in lherzolites trapped melt? In Al-Qala area (Wuqbah block), the Cpx modal amount shows a significant increase from 0 to about 240 m above the amphibolites (Fig. 7) and a sudden drop in the harzburgitic domain above 240 m. Such a variation was not observed in Al-Khabt and Al-Baks areas: this is probably related to their specific location within the hinge of complex folds (Fig. 1; Nicolas et al., 1988, 2000). The mineral–chemical compositions show a general upward depletion, a decrease in Na2O and Al2O3 in pyroxenes, and an increase of Mg# in pyroxenes and Cr# of spinel (Fig. 7). These observations, associated with overlapping of Cr# and partial-melting degrees of both lherzolite types (see the next section), lead us to consider two different sources for Type I and Type II lherzolites, and to link Type I lherzolites with harzburgites through magmatic processes involving an increasing degree of partial melting. In the northern Fizh block, Takazawa et al. (2003) suggested that Fizh Type II lherzolites could be formed from Fizh Type I lherzolites by refertilization of a MORB melt. However, their samples are characterized by a weakly spoon-shaped REE pattern of Cpx, which we did not observe in our peridotite Cpxs (Figs. 10a, c and 13a). Moreover, structural data showed that a failing ridge system was located near the base of the Fizh massif (Nicolas et al., 2000): a geotectonic context that could allow refertilization of lithospheric mantle by MORB melts. Refertilization of Sarami Type I lherzolites by a MORB-type melt to form Type II would lead to Ti enrichment in spinel (>0.2 wt.% of TiO2) (Dick and Bullen, 1984; Bonatti et al., 1992) and to a decrease in olivine Mg# at a given spinel Cr# (Batanova et al., 1998), which is not the case for Type II lherzolites; spinel is low in Ti (TiO2 b 0.09 wt.%) and olivine is high in Mg# in Sarami Type II lherzolites (Table 1; Fig. 6). Magmatic Cpx crystallized from MORB melt may display a flat REE pattern with a gentle slope of LREE (e.g., Batanova et al., 1998; Akizawa et al., 2012). In contrast, Cpxs in Sarami and Wuqbah (Al-Qala site) basal peridotites are free of evidence of melt refertilization in texture, e.g., wormy or thin veins of interstitial Cpx around porphyroclast grains, and in chemistry (Figs. 10 and 13), e.g., relative enrichment of HFSE in Cpx (Jean et al., 2010), relative enrichment of LREE or MREE to HREE in Cpx with spoon-, flat-, hump- or S-shaped REE patterns (Batanova et al., 1998; Godard et al., 2000; Hellebrand et al., 2002; Seyler et al., 2003, 2007). Our Cpxs are highly depleted in LREE and typically similar to those in residual abyssal peridotites (Johnson et al., 1990; Hellebrand et al., 2002) (Fig. 13a); they are also enriched in fluid-mobile element (e.g., B, Li, Cs, Rb and Pb) relative to HFSE (Ta, Nb, Th and Zr) (Fig. 10b, d). Similarly, Sarami Opxs are highly depleted in LREE relative to HREE, and are enriched in fluid-mobile elements relative to HFSE, suggesting no effect of melt refertilization (Fig. 11). Moreover, the hypothetical melts in equilibrium with Cpxs in Type I and II lherzolites are different with depleted MORB (D-MORB) (Frey et al., 1993) and normal MORB (N-MORB) (Hofmann, 1988; Sun and McDonough, 1989), respectively, (Fig. 13b). Hence, we exclude the refertilization origin of Sarami lherzolites (Types I and II) by these melts. These lherzolites are far from 1258 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Fig. 16. Schematic illustration showing the tectonic setting of basal peridotites from Wadi Sarami, central Oman ophiolite. The Sarami lherzolites represent abyssal peridotites along oceanic fracture, which is consistent with the inferred ridge axes in the east of the Sarami block and mantle diapiric apex in the Wuqbah block (Nicolas et al., 1988, 2000). The Type II lherzolites represent a remnant of asthenospheric materials trapped at the base of oceanic lithosphere mantle (Type I) during detachment and obduction. They lie at the segment end in the space that is slightly far from the mantle diapiric apex in the Wuqbah block. Lherzolites and harzburgites have been exposed on the surface along oceanic thrusting faults that were formed during extension movements. equilibrium with mid-ocean ridge basalt (MORB), like residual peridotites from mid-ocean ridge and Oman peridotites (Kelemen et al., 1995). Geologic structure suggests that Wuqbah and Sarami massifs were located off-ridge, rather far from the spreading axis, and Sarami block is close to a mantle diapir in Wuqbah massif (Nicolas et al., 1988, 2000) (Fig. 1). In this configuration, it would be difficult to explain the origin and behavior of MORB-type melts. In addition, HREE are compatible with Na- and Al-rich Cpx in spinel lherzolites at relatively high pressures (Blundy et al., 1998; Hellebrand et al., 2002). The HREE, YbN, Al2O3 and Na2O contents in Cpx of Type II lherzolites are higher than those of Type I lherzolites (Table 2; Figs. 6 and 7), suggesting that the former have been possibly formed under higher pressure and temperature than the latter. Thus, we infer that Type I lherzolites, which display lithological and chemical characteristics common to abyssal peridotites, were formed at the base of the oceanic lithosphere during Neo–Tethyan expansion. The Type II lherzolites (with Cpx enriched in Na, Ti and Al, Opx enriched in Al and spinel Cr# b 0.15; see Tables 2 and 3; Figs. 6 and 7), on the other hand, likely represent a remnant of asthenospheric materials trapped by the base of oceanic lithosphere mantle during detachment and obduction (Fig. 16). We suggest possibly two melting series and sources for Type I and II lherzolites because they show overlap in partial melting degrees (see below) and in chemistry of pyroxenes and olivines as well as spinels (Figs. 6 and 7; see Khedr et al., 2013). Consequently, we exclude the derivation of Type II lherzolites from Type I by melt refertilization. Type II lherzolites are exposed only in the base of the Sarami block, while Type I are exposed in the base of both Sarami and Wuqbah blocks toward the Wuqbah mantle diapir (Fig. 1). The final scarcity of asthenospheric materials in Al-Qala (Wuqbah block) can be explained by the proximity to the diapiric zone, which led to a high melting degree in the Wuqbah block relative to the Sarami block (Fig. 16). In contrast, the relative abundances of almost unmolten asthenospheric remnants (Type II) at the base of the Sarami block are likely due to a significant distance from the diapiric apex in the Wuqbah block (Nicolas et al., 1988, 2000). In addition, type II lherzolites are close to the segment end affected by low degree of partial melting, where they lie in the space (off-axis) between mid-ocean ride center in the east of the Sarami block and a diapiric zone in the Wuqbah block (Nicolas et al., 2000) (Fig. 16). 5.3.2. Variation of partial melting degrees To examine primary mantle characteristics, only porphyroclast cores that were least affected by secondary processes (e.g., low-P subsolidus recrystallization, metasomatism and low-T alteration) were analyzed. The studied peridotites have a wide range of spinel Cr# (0.04−0.53; see Fig. 6d), suggesting variations of partial melting degrees (Jaques and Green, 1980; Dick and Bullen, 1984; Arai, 1994). The HREE in Sarami Cpxs (Fig. 13a) are used to determine the degree of melt extraction and the nature of melting conditions (e.g., Johnson et al., 1990; Dick and Natland, 1996; Hellebrand et al., 2001). They M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 suggest varying extents of melt extraction by near-fractional melting under polybaric condition for Sarami peridotites. The best fit of Cpx HREE in Type II lherzolites, Type I lherzolites and harzburgites requires 1–5%, 3–10% and 10–15% melting mainly in the spinel field, respectively (Fig. 13a). This is consistent with partial-melting degrees inferred from olivine (Fo) versus spinel (Cr#) (Fig. 6d). Moreover, the Cpx HREE feature in Sarami lherzolits is also a coincidence with the model where melting started in the garnet field (~4% melting) and was followed by ~10% in the spinel field (e.g., Hellebrand et al., 2002) (Fig. 13a). The studied lherzolites possibly require melt extractions from the source containing a small amount of garnet, first, and then from spinel-bearing source. It is well known that combination of spinel Cr# with Cpx HREE in peridotites can be used as indicators for the degree of partial melting (e.g., Hellebrand et al., 2001, 2002). Moreover, the behaviors of some major (Al, Ca) and trace (e.g., HREE, Cr, Ti, Sc, Ni, Cr, Co) elements are almost the same during partial melting and subsequent processes. These elements are nearly immobile during serpentinization and sea floor weathering (e.g., Niu, 2004; Paulick et al., 2006) and can be used as markers of mantle magmatic processes. In our samples, the variations of spinel Cr# with Cpx Yb (ppm), and of Dy with Ti in Cpx, as well as the bulk rock major and trace elements with MgO (Figs. 3 and 4), suggest that Sarami basal peridotites are residues after less than 10% of melt extraction (around 7%) for lherzolites, and less than 20% (mainly~15%) for harzburgites (Fig. 14). Sc and Yb, known to be highly compatible with garnet and incompatible with Cpx and spinel, show a negative correlation with MgO (Fig. 4). This indicates that melting occurred at low pressures, mainly in the spinel stability field, for our peridotites (e.g., Niu, 2004). These results are in accordance with all our data (bulk-rock trace and major elements, Cpx trace elements, etc; see Figs. 13, 14 and 15), which point to a continuous partial melting trend to form harzburgites from Type I lherzolites (1−20% of partial melting), while Type II lherzolites seem to have been hardly melted and their Cpx shows fertile characteristics with high amount of incompatible elements. Variations of partial melting would explain the large heterogeneities observed in our samples. The extent and variation of mantle melting depend on variations of, among others, mantle potential temperature (Langmuir et al., 1992; Niu et al., 1997), which may be related to the spreading rate (Niu, 1997; Niu and Hékinian, 1997) and a distance from the spreading ridge or mantle diapiric zones. The variation of melting degrees in Sarami and Wuqbah basal peridotites is possibly related to small-scale variations in mantle temperature. Finally, the variations of partial melting degrees observed here are responsible for the severely mantle heterogeneity. The mantle heterogeneity is significantly high in the base of the mantle section relative to its upper part. 5.3.3. Sub-solidus cooling of peridotites Some mineral compositions in Sarami basal peridotites have been modified during sub-solidus cooling at the shallow-level mantle or during/after emplacements. This sub-solidus modification of some minerals is possibly one factor causing small-scale mantle heterogeneity at a shallow mantle depth. Cpx may have been exsolved as lamellae or blebs from residual Opx porphyroclasts (Fig. 2c) and vice versa, which are common in abyssal peridotites (e.g., Hellebrand et al., 2002). In addition, some Cpx porphyroclasts in our samples include exsolution lamellae of Al-rich spinels (Fig. 8). Al-rich pyroxenes are usually markers of higher temperature and pressure conditions (e.g., Obata and Dickey, 1976), and the presence of Al-rich spinel exsolutions within pyroxene probably reflects the drop of temperature and pressure during exhumation. Occurrences of Al-spinel lamellae are common only in Type I lherzolites Cpx (Fig. 8), indicating longer duration of cooling in comparison to Type II lherzolites. The absence of this kind of lamellae in harzburgites Cpx is probably due to their Al-depleted composition relative to the lherzolite composition. The Na content in Cpx depends on the pressure and degree of partial melting (Blundy et al., 1995; Takazawa et al., 2003), but the Al content in Cpx depends on the degree of partial melting and equilibrium 1259 temperatures (Blundy et al., 1995; Takazawa et al, 2003; Khedr et al., 2010). In the lherzolitic Cpx porphyroclasts, the decrease of Al and Na contents from core to rim without change of Ti content is a consequence of sub-solidus re-equilibration. This process also explains the low Al and Na contents in recrystallized fine Cpx grains associated with Na-bearing hornblende (Khedr et al., 2013; see Table 2 and Fig. 6). The large scatter of Al2O3, Na2O and Cr2O3 contents in Cpx in one Type II lherzolite (G. 27) and Type I lherzolite (G.33), associated with unchangeable compositions of spinel (e.g., Cr#) and olivine (e.g., Fo), can also be attributed to incomplete equilibration by sub-solidus cooling (Table 2; Fig. 7), and are also typical features of abyssal peridotites (e.g., Hellebrand et al., 2002). Cpx porphyroclast-core compositions show a negative correlation between Mg# and Al2O3 or TiO2 in Cpx (Table 2; Khedr et al., 2013), suggesting an increasing degree of partial melting from the most primitive lherzolite to harzburgite with only limited sub-solidus effect. 5.4. Late stage metasomatism and post-melting refertilization 5.4.1. Metasomatized basal peridotites Fluid metasomatism is expected to affect on basal peridotites in ophiolites by precipitating hydrous phases like amphiboles and serpentines (e.g., Tatsumi et al., 1986; Peacock, 1990; Bizimis et al., 2000; Khedr et al., 2010). In our samples, hornblendes replacing precursor Cpxs, which were possibly formed during the invasion of hydrous fluids and preferential reaction with Cpx, are observed in both lherzolite types but are absent in harzburgites (Fig. 9), showing a limited extent of the late metasomatism. The extent of such metasomatism in the Oman ophiolite was only over few meters to few hundred of meters from the base. This is in favor of metasomatism related to the formation of amphibolite during the detachment stage. We suggest that Sarami peridotites have been subjected to metasomatism starting at detachment stage (pargasitic hornblende) and flourishing during the ophiolite emplacements (formation of low-T serpentines, tremolites). As shown in Figs. 5 and 10, general levels and shapes of REE patterns of whole rocks are slightly different with those of their corresponding Cpxs. The whole REE values of peridotites are lower than those of their Cpxs because of the abundance serpentines and olivines (poor in REE). The CI-normalized REE patterns of basal peridotites show a spoon shape (Fig. 5a, c) with inflection at La and Ce, in contrast to depleted LREE patterns of their Cpxs (Fig. 10a, c). The former may reflect post-melting processes such as late stage metasomatic refertilization, and the latter reflect sub-ridge mantle melting processes (e.g., Niu, 2004). In contrast to the conservative elements (Ca, Al, Ti, Sc, Ni, Cr, Co and HREE; see Figs. 3 and 4), the bulk LREE (La, Ce, Pr and Nd), Cs, Sr and Ba (fluid mobile elements) concentrations show erratic or scattered correlations with MgO (Table 1), suggesting secondary addition of these mobile elements. The addition of LREE to the Sarami peridotites during the post-melting stage through a reaction with influxed fluids after melting is consistent with the LREE enrichment in Oman peridotites (mainly harzburgites) after Monnier et al. (2006) (Fig. 5). Moreover, the multi-element patterns (Fig. 5b and d) of bulk peridotites display positive anomalies at Cs and Sr, and HFSE depletion. Mobile elements (e.g., Rb, Cs, Ba, Sr and LREE) may be added through fluid metasomatism during hydrothermal alteration as well as seafloor weathering at or near the spreading axis, and/or during ophiolite emplacement onto the metamorphic sole (Fig. 5) (e.g., Hellebrand et al., 2002; Niu, 2004; Ishikawa et al., 2005); they are possibly concentrated along silicate grain boundaries (e.g., Hiraga et al., 2004; Khedr et al., 2010). This is supported from the mass balance calculation of peridotites based on the modal and mineral chemical compositions that indicates the presence of excess Cs, Sr and Ba in the measured concentrations (Table 1) over the mass calculated ones. The fluid expelled from the amphibolite-facies slab (= Tethyan oceanic crust) underlying the mantle section of the Oman ophiolite is enriched in LREE (e.g., La and Ce), Rb, Ba, B, K, Sr, Li, Be and Pb (e.g., Ishikawa et al., 2005), and possibly migrated upward to metasomatize the overlying basal peridotites. A small portion of the amphibolite-derived fluid 1260 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 was allowed to ascend into the overlying Sarami basal peridotites through porous (channel or fracture) flow (e.g., Ishikawa et al., 2005) to cause their enrichment of fluid mobile elements. Finally, the post-melting metasomatism can cause chemical change of overlaying peridotite compositions, and is responsible for their incompatible-element enrichment. The aforementioned results are consistent with the trace-element chemistry of hornblendes replacing Cpxs in our lherzolites (Figs. 9 and 12). Hornblende REE patterns closely fit those of their precursor Cpxs, with higher HREE than MREE–LREE (Figs. 5, 10 and 12). These hornblendes show relative enrichment in fluid-mobile elements (e.g., B, Li, Cs, Pb, Rb and Ba) and depleted in HFSE (e.g., Ta, Nb, Th and Zr) + U (Fig. 12b), similar to those in their precursor Cpxs (Figs. 5 and 10). Incompatible elements are readily mobilized in aqueous solutions during metasomatism or serpentinization processes, and experimental studies verified that slab-derived fluids are depleted in HFSE and enriched in both LREE and LILE (Tatsumi et al., 1986; Kessel et al., 2005; Marocchi et al., 2007). However, the total chemical overlap both in LILE and LREE observed between Cpxs and hornblendes suggests that the metasomatic agent involved was extremely depleted in all trace elements so that the composition of Cpx was inherited to hornblende during hydration. The depleted nature of the fluid is compatible with the petrographical nature of the metamorphic sole in this region, which points to a metasedimentary origin with a protolith mainly composed of SiO2-rich sand with minor aluminous phases (Python, Takeshita and Arai, unpublished data). Thus, metasomatism in some samples likely comes from direct reaction between fluids released from the metamorphic sole and overlaying basal peridotite. The basal peridotites (Type II lherzolites) have been metasomatized by influxed fluids to form hydrous peridotites (Fig. 2e, f), which were recognized in the base of Oman mantle sections in a small scale (few tens of centimeters). 5.4.2. Origin of the clinopyroxenitic lens Pyroxenite veins and late interstitial clinopyroxenes have been described in some abyssal peridotites (e.g., Kempton and Stephens, 1997; Seyler et al., 2001; Dantas et al., 2007). The Sarami hydrous peridotites contain a small clinopyroxenite lens (Fig. 2e and f) at the base of Type II lherzolites, providing small-scale (b 5 cm) mineral and chemical heterogeneities at the base of Oman mantle section. This lens is possibly a part or fragment broken from clinopyroxenite bands, layers or veins within Sarami basal lherzolites during deformation and emplacement. The origin of pyroxenite veins in abyssal peridotites is still a matter of debates. Dantas et al. (2007) stated that the abyssal mantle pyroxenite veins were formed from melts that correspond to incremental melt fractions produced during fractional decompression melting of a normal MORB (N-MORB) mantle source. The crystallization of pyroxenites, and possibly the melt segregation event itself, took place close to the asthenosphere–lithosphere boundary (Dantas et al., 2007). The Cpx from the clinopyroxenite lens is lower in REE concentrations (ΣREE = 0.7−1.0 ppm) than Cpxs in basal peridotites and exhibits flat-shape REE patterns with inflection at La and Ce (Figs. 10a, 13b). This is a coincidence with the depleted character of clinopyroxenite lens: Cpx is poor in REE, Al, Na, Ti and Cr and spinel is high in Cr# (0.4–0.6) relative to the host Sarami peridotites (Fig. 6a; Supplementary Data 2) and Cpx of pyroxenite veins in abyssal peridotites from south Indian ridge (Dantas et al., 2007). Therefore, this clinopyroxenite lens was possibly formed in a shallow-mantle level (at low-P) relative to pyroxenite veins from the Indian ridge (Dantas et al., 2007). The calculated melts in equilibrium with Cpx in lherzolites (Types I and II) (Fig. 13b) are different from the calculated melt in equilibrium with Cpx in clinopyroxenite lens (low in REE), which is comparable to the calculated melt in equilibrium with Cpx in Oman dunites (Kelemen et al., 1995) (Fig. 13b). These results refer to different origins between Sarami lherzolite Cpx (residual origin) and Cpx in the clinopyroxenite lens. The clinopyroxenite lens has been possibly formed from fractional crystallization of interstitial incremental melt that formed during fractional decompression melting of a normal MORB mantle source (e.g., Dantas et al., 2007) in a later stage during conductive cooling under low pressure relative to the Sarami lherzolites and pyroxenite veins after Dantas et al. (2007). It contains Al-, Ti-poor spinel rather than plagioclase, suggesting its crystallization in the spinel stability field (Seyler et al., 2001; Dantas et al., 2007), being in accordance with the later stage formation. The trace-element compositions of hornblendes associated with clinopyroxenite lens or veinlet (Fig. 2e, f) are different with those of hornblendes replacing residual Cpxs in lherzolites (Fig. 9). The former hornblendes are enriched in LREE, Th, U, Nb, Ta and Pb relative to the later ones (Fig. 12); this is possibly attributed to the difference in chemistry of precursor Cpxs. The in-situ Cpx chemistry of this magmatic veinlet is depleted in HREE but enriched in LREE relative to those of residual Cpxs (Figs. 10, 12 and 13) and pyroxenite veins in abyssal peridotites (e.g., Dantas et al., 2007). This result suggests that the parent melt for this veinlet was probably issued from a strongly depleted source (low HREE) that melted under LREE–LILE enriched fluid influx (Figs. 10 and 12), or this veinlet was possibly reacted with LREE-enriched melts or fluids at a low melt/rock ratio, similar to a LREE-enriched melt in Wadi Fizh, north Oman (Takazawa et al., 2003). The pyroxenite and gabbro dykes in the Oman mantle sections form off-axis in the lithospheric mantle (Kelemen et al., 1995 and references therein). This is in agreement with the occurrence of clinopyroxenite lens in the off-axis mantle part represented by Type II lherzolites in the base of the Sarami block (Fig. 16). Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.gr.2013.05.010. Acknowledgments We are grateful to Durair A'Shaikh and Ali Al-Rajhi in the Ministry of Commerce and Industry, Sultanate of Oman for their help during our staying in Oman. We thank S. Ishimaru, N. Akizawa, H. Negishi and M. Miura for their assistance during our field expedition to Oman (2010−2012). We are grateful to S. Umino, T. Mizukami, T. Morishita and K. Abbou-Kebir for their beneficial discussions. We are grateful to Y. Dilek, H. Moghadam and an anonymous reviewer for their beneficial comments. We thank M. Santosh for his editorial handling of this manuscript. References Akizawa, N., Arai, S., Tamura, A., 2012. Behavior of MORB magmas at uppermost mantle beneath a fast-spreading axis: an example from Wadi Fizh of the northern Oman ophiolite. Contributions to Mineralogy and Petrology 164, 601–625. Anders, E., Grevesse, N., 1989. Abundances of the elements: meteoritic and solar. Geochimica et Cosmochimica Acta 53, 197–214. Anderson, D.L., 2006. Peculations on the nature and cause of mantle heterogeneity. Tectonophysics 416, 7–22. Arai, S., 1994. Characterization of spinel peridotites by olivine-spinel compositional relationships: review and interpretation. Chemical Geology 113, 191–204. Arai, S., Kadoshima, K., Morishita, T., 2006. Widespread arc-related melting in the mantle section of the northern Oman ophiolite as inferred from detrital chromian spinels. Geological Society of London 163, 869–879. Barth, M.G., Mason, P.R.D., Davies, G.R., Drury, M.R., 2008. The Othris Ophiolite, Greece: a snapshot of subduction initiation at a mid-ocean ridge. Lithos 100, 234–254. Batanova, V.G., Suhr, G., Sobolev, A.V., 1998. Origin of geochemical heterogeneity in the mantle peridotite from the Bay of Islands ophiolite, Newfoundland, Canada: ion probe study of clinopyroxene. Geochimica et Cosmochimica Acta 62, 853–866. Bizimis, M., Salters, V.J.M., Bonatti, E., 2000. Trace and REE content of clinopyroxenes from supra-subduction zone peridotites. Implications for melting and enrichment processes in island arcs. Chemical Geology 165, 67–85. Blundy, J.D., Falloon, T.J., Wood, B.J., Dalton, J.A., 1995. Sodium partitioning between clinopyroxene and silicate melts. Journal of Geophysical Research 100, 15501–15515. Blundy, J., Robinson, J.A.C., Wood, B., 1998. Heavy REE are compatible in clinopyroxene on the spinel lherzolite solidus. Earth and Planetary Science Letters 160, 493–504. Bonatti, E., Peyve, A., Kepezhinskas, P., Kurentsova, N., Seyler, M., Skolotnev, S., Udintsev, G., 1992. Upper mantle heterogeneity below the Mid-Atlantic Ridge, 0°–15°N. Journal of Geophysical Research 97, 4461–4476. Boudier, F., Nicolas, A., 1995. Nature of the Moho transition zone in the Oman ophiolite. Journal of Petrology 36, 777–796. Boudier, F., Ceuleneer, G., Nicolas, A., 1988. Shear zones, thrusts and related magmatism in the Oman ophiolite: initiation of thrusting on an oceanic ridge. Tectonophysics 151, 275–296. M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 Bougault, H., Dmitriev, L., Schilling, J.G., Sobolev, A., Joron, J.L., Needham, H.D., 1988. Mantle heterogeneity from trace elements: MAR triple junction near 14°N. Earth and Planetary Science Letters 88, 27–36. Brandl, P.A., Beier, C., Regelous, M., Abouchami, W., Haase, K.M., Garbe-Schönberg, D., Galer, S.J.G., 2012. Volcanism on the flanks of the East Pacific Rise: quantitative constraints on mantle heterogeneity and melting processes. Chemical Geology 298–299, 41–56. Brey, G.P., Kohler, T., 1990. Geothermobarometry in four-phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378. Brunelli, D., Seyler, M., Cipriani, A., Ottolini, L., Bonatti, E., 2006. Discontinuous melt extraction and weak refertilization of mantle peridotites at the Vema lithospheric section (Mid-Atlantic Ridge). Journal of Petrology 47, 745–771. Dantas, C., Ceuleneer, G., Gregoire, M., Python, M., Freydier, R., Warren, J., Dick, H.J.B., 2007. Pyroxenites from the Southwest Indian Ridge, 9–16o E: cumulates from incremental melt fractions produced at the top of a cold melting regime. Journal of Petrology 48, 647–660. Dick, H.J.B., Bullen, T., 1984. Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contributions to Mineralogy and Petrology 86, 54–76. Dick, H.J.B., Natland, J.H., 1996. Late stage melt evolution and transport in the shallow mantle beneath the East Pacific Rise. In: Gillis, K., Mevel, C., Allan, J. (Eds.), Proceedings of the Ocean Drilling Program: Scientific Results: Ocean Drilling Program, vol. 147, pp. 103–134. Dick, H.J.B., Fisher, R.L., Bryan, W.B., 1984. Mineralogic variability of the uppermost mantle along mid-ocean ridges. Earth and Planetary Science Letters 69, 88–106. Dilek, Y., Furnes, H., Shallo, M., 2008. Geochemistry of the Jurassic Mirdita Ophiolite (Albania) and the MORB to SSZ evolution of a marginal basin oceanic crust. Lithos 100, 174–209. Evans, B.W., 1977. Metamorphism of Alpine peridotite and serpentinite. Annual Review of Earth and Planetary Sciences 5, 397–447. Evans, B., 2004. The serpentinite multisystem revisited: chrysotile is metastable. International Geology Review 46, 479–506. Evans, B.W., Frost, B.R., 1975. Chrome-spinel in progressive metamorphism: a preliminary analysis. Geochimica et Cosmochimica Acta 39, 959–972. Frey, F.A., Walker, N., Stakes, D., Hart, S.R., Nielsen, R., 1993. Geochemical characteristics of basaltic glasses from the AMAR and FAMOUS axial valleys, Mid-Atlantic Ridge (36°–37°N): petrogenetic implications. Earth and Planetary Science Letters 115, 117–136. Godard, M., Jousselin, D., Bodinier, J.-L., 2000. Relationships between geochemistry and structure beneath a paleo-spreading centre: a study of the mantle section in the Oman ophiolite. Earth and Planetary Science Letters 180, 133–148. Hanghøj, K., Kelemen, P.B., Hassler, D., Godard, M., 2010. Peridotites from the Wadi Tayin massif, Oman ophiolite; major and trace element geochemistry, and Os isotope and PGE systematics. Journal of Petrology 51, 201–227. Hart, S.R., Dunn, T., 1993. Experimental cpx/melt partitioning of 24 trace elements. Contributions to Mineralogy and Petrology 113, 1–8. Hellebrand, E., Snow, J.E., Dick, H.J.B., Hofman, A.W., 2001. Coupled major and trace elements as indicators of the extent of melting in mid-ocean-ridge peridotites. Nature 410, 677–681. Hellebrand, E., Snow, J.E., Hoppe, P., Hofmann, A.W., 2002. Garnet-field melting and late-stage refertilization in residual abyssal peridotites from the Central Indian Ridge. Journal of Petrology 43, 2305–2338. Hiraga, T., Anderson, A.M., Kohlstedt, D.L., 2004. Grain boundaries as reservoirs of incompatible elements in the Earth's mantle. Nature 427, 699–703. Hirauchi, K., Tamura, A., Arai, S., Yamaguchi, H., Hisada, K., 2008. Fertile abyssal peridotites within the Franciscan subduction complex, central California: possible origin as detached remnants of oceanic fracture zones located close to a slow-spreading ridge. Lithos 105, 319–328. Hofmann, A.W., 1988. Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90, 297–314. Hofmann, A.W., 1997. Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229. Ishii, T., Robinson, P.T., Maekawa, H., Fiske, R., 1992. Petrological studies of peridotites from diapiric serpentinite seamounts in the Izu–Ogasawara–Mariana forearc, LEG125. In: Fryer, P., Pearce, J.A., Stokking, L.B. (Eds.), Proceedings of the Ocean Drilling Program: Scientific Results: Ocean Drilling Program, vol. 125, pp. 445–485. Ishikawa, T., Nagaishi, K., Umino, S., 2002. Boninitic volcanism in the Oman ophiolite: implications for thermal condition during transition from spreading ridge to arc. Geology 30, 899–902. Ishikawa, T., Fujisawa, S., Nagaishi, K., Masuda, T., 2005. Trace element characteristics of the fluid liberated from amphibolite-facies slab: inference from the metamorphic sole beneath the Oman ophiolite and implication for boninite genesis. Earth and Planetary Science Letters 240, 355–377. Jaques, A.L., Green, D.H., 1980. Anhydrous melting of peridotite at 0–15 kb pressure and the genesis of tholeiitic basalts. Contributions to Mineralogy and Petrology 73, 287–310. Jean, M.M., Shervais, J.W., Choi, S.-H., Mukasa, S.B., 2010. Melt extraction and melt refertilization in mantle peridotite of the Coast Range ophiolite: an LA–ICP–MS study. Contributions to Mineralogy and Petrology 159, 113–136. Johnson, K.T.M., Dick, H.J.B., 1992. Open system melting and the temporal and spatial variation of peridotite and basalt compositions at the Atlantis II F. Z. Journal of Geophysical Research 97, 9219–9241. Johnson, K.T.M., Dick, H.J.B., Shimizu, N., 1990. Melting in the oceanic upper mantle: an ion microprobe study of diopside in abyssal peridotites. Journal of Geophysical Research 95, 2661–2678. 1261 Kelemen, P.B., Shimizu, N., Salters, V.J.M., 1995. Extraction of mid-oceanic-ridge basalt from the upwelling mantle by focused flow of melt in dunite channels. Nature 375, 747–753. Kempton, P.D., Stephens, C.J., 1997. Petrology and geochemistry of nodular websterite inclusions in harzburgite, Hole 920D. In: Karson, J.A., Cannat, M., Miller, D.J., Elthon, D. (Eds.), Proceedings of the Ocean Drilling Program: Scientific Results: Ocean Drilling Program, vol. 153, pp. 321–331. Kessel, R., Schmidt, M.W., Ulmer, P., Pettke, T., 2005. Trace element signature of subduction-zone fluids, melts and supercritical liquids at 120–180 km depth. Nature 437, 724–727. Khedr, M.Z., Arai, S., 2010. Hydrous peridotites with Ti-rich chromian spinel as a lowtemperature forearc mantle facies: evidence from the Happo-O'ne metaperidotites (Japan). Contributions to Mineralogy and Petrology 159, 137–157. Khedr, M.Z., Arai, S., 2011. Petrology and geochemistry of chromian spinel-bearing serpentinites in the Hida Marginal Belt (Ise area, Japan): characteristics of their protoliths. Journal of Mineralogical and Petrological Sciences 106, 255–260. Khedr, M.Z., Arai, S., Tamura, A., Morishita, T., 2010. Clinopyroxenes in high-P metaperidotites from Happo-O'ne, central Japan: implications for wedge-transversal chemical change of slab-derived fluids. Lithos 119, 439–456. Khedr, M.Z., Arai, S., Python, M., 2013. Petrology and chemistry of basal lherzolites above the metamorphic sole from Wadi Sarami, central Oman ophiolite. Journal of Mineralogical and Petrological Sciences 108, 13–24. Kogiso, T., Tatsumi, Y., Nakano, S., 1997. Trace element transport during dehydration processes in the subducted oceanic crust: 1. Experiments and implications for the origin of ocean island basalts. Earth and Planetary Science Letters 148, 193–206. Langmuir, C.H., Klein, E.M., Plank, T., 1992. Petrological systematics of mid-ocean ridge basalts: constraints on melt generation beneath ocean ridges. In: Morgan, J.P., Blackman, D.K., Sinton, J.M. (Eds.), Mantle Flow and Melt Generation at Midocean Ridges: Geophysical Monograph, 71. American Geophysical Union, Washigton, D.C., pp. 183–280. Lippard, S.J., Shelton, A.W., Gass, I.G., 1986. The ophiolite of northern Oman. Geological Society Memoir, vol. 11. Blackwell, Oxford 178. Marocchi, M., Hermann, J., Morten, L., 2007. Evidence for multi-stage metasomatism of chlorite-amphibole peridotites (Ulten Zone, Italy): constraints from trace element compositions of hydrous phases. Lithos 99, 85–104. McDonough, W.F., Frey, F.A., 1989. Rare earth elements in upper mantle rocks. In: Lipin, B.R., McKay, G.A. (Eds.), Geochemistry and Mineralogy of Rare Earth Elements: Reviews in Mineralogy, 21, pp. 99–145. McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chemical Geology 120, 223–253. Pearce, N.J.G., Perkins, W.T., Westgate, J.A., Gorton, M.P., Jackson, S.E., Neal, C.R., Chenery, S.P., 1997. A compilation of new and published major and trace element data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter 21, 115–144. Michael, P.J., Bonatti, E., 1985. Peridotite composition from the North Atlantic: regional and tectonic variations and implications for partial melting. Earth and Planetary Science Letters 73, 91–104. Ministry of Petroleum and Minerals, 1992. Geological map of Buraymi, scale 1:250,000. Sultanate of Oman, Mascat. Monnier, C., Girardeau, J., Le Me'e, L., Polvé, M., 2006. Along-ridge petrological segmentation of the mantle in the Oman ophiolite. Geochemistry, Geophysics, Geosystems 7, 11008. http://dx.doi.org/10.1029/2006GC001320. Morishita, T., Ishida, Y., Arai, S., Shirasaka, M., 2005. Determination of multiple trace element compositions in thin (b30 μm) layers of NIST SRM 614 and 616 using laser ablation–inductively coupled plasma–mass spectrometry. Geostandards and Geoanalytical Research 29, 107–122. Nicolas, A., 1989. Structure of Ophiolites and Dynamics of Oceanic Lithosphere. Kluwer, Dordrecht. Nicolas, A., Ceuleneer, G., Boudier, F., Misseri, M., 1988. Structural mapping in the Oman ophiolites: mantle diapirism along an oceanic ridge. Tectonophysics 151, 27–56. Nicolas, A., Boudier, F., Ildefonse, B., Ball, E., 2000. Accretion of Oman and United Arab Emirates ophiolite—discussion of a new structural map. Marine Geophysical Researches 21, 147–179. Niu, Y., 1997. Mantle melting and melt extraction processes beneath ocean ridges: evidence from abyssal peridotites. Journal of Petrology 38, 1047–1074. Niu, Y., 2004. Bulk-rock major and trace element compositions of abyssal peridotites: implications for mantle melting, melt extraction and post-melting processes beneath mid-ocean ridges. Journal of Petrology 45, 2423–2458. Niu, Y., Hékinian, R., 1997. Spreading rate dependence of the extent of mantle melting beneath ocean ridges. Nature 385, 326–329. Niu, Y., Langmuir, C.H., Kinzler, R.J., 1997. The origin of abyssal peridotites: a new perspective. Earth and Planetary Science Letters 152, 251–265. Obata, M., Dickey, J.S., 1976. Phase relations of mafic layers in the Ronda peridotite. Carnegie Institution of Washington Yearbook, 75 562–566. Ohara, Y., Ishii, T., 1998. Peridotites from the southern Mariana forearc: heterogeneous fluid supply in mantle wedge. Island Arc 7, 541–558. Parkinson, I.J., Pearce, J.A., 1998. Peridotites from the Izu–Bonin–Mariana forearc (ODP Leg 125): evidence for mantle melting and melt–mantle interaction in a suprasubduction zone setting. Journal of Petrology 39, 1577–1618. Paulick, H., Bach, W., Godard, M., De Hoog, J.C.M., Suhr, G., Harvey, J., 2006. Geochemistry of abyssal peridotites (Mid-Atlantic Ridge, 15° 20′ N, ODP Leg 209): implications for fluid/rock interaction in slow spreading environments. Chemical Geology 234, 179–210. Peacock, S.M., 1990. Fluid processes in subduction zones. Science 248, 329–337. Pearce, J.A., Parkinson, I.J., 1993. Trace element models for mantle melting: application to volcanic arc petrogenesis. In: Prichard, H.M., Alabaster, T., Harris, N.B.W., Neary, 1262 M.Z. Khedr et al. / Gondwana Research 25 (2014) 1242–1262 C.R. (Eds.), Magmatic Processes and Plate Tectonics: Geological Society, London, Special Publication, 76, pp. 373–403. Ross, K., Elthon, D., 1997. Extreme incompatible trace-element depletion of diopside in residual mantle from south of the Kane F.Z. In: Karson, J.A., Cannat, M., Miller, D.J., Elthon, D. (Eds.), Proceedings of the Ocean Drilling Program: Scientific Results: Ocean Drilling Program, vol. 153, pp. 277–284. Sano, S., Kimura, J.-I., 2007. Clinopyroxene REE geochemistry of the Red Hills peridotite, New Zealand: interpretation of magmatic processes in the upper mantle and in the Moho transition zone. Journal of Petrology 48, 113–139. Schilling, J.-G., 1985. Upper mantle heterogeneities and dynamics. Nature 314, 62–67. Seyler, M., Toplis, M.J., Lorand, J.P., Luguet, A., Cannat, M., 2001. Clinopyroxene microtextures reveal incompletely extracted melts in abyssal peridotites. Geology 29, 155–158. Seyler, M., Cannat, M., Mevel, C., 2003. Evidence for major-element heterogeneity in the mantle source of abyssal peridotites from the Southwest Indian Ridge (52 to 68 E). Geochemistry, Geophysics, Geosystems 4 (2), 9101. http://dx.doi.org/10.1029/ 2002GC000305. Seyler, M., Lorand, J.P., Dick, H.J.B., Drouin, M., 2007. Pervasive melt percolation reactions in ultra-depleted refractory harzburgites at the Mid-Atlantic Ridge, 15o 20/ N: ODP Hole 1274A. Contributions to Mineralogy and Petrology 153, 303–319. Stosch, H.-G., 1982. Rare earth partitioning between minerals from anhydrous spinel peridotite xenoliths. Geochimica et Cosmochimica Acta 46, 793–811. Stracke, A., Bourdon, B., 2009. The importance of melt extraction for tracing mantle heterogeneity. Geochimica et Cosmochimica Acta 73, 218–238. Sun, S.-S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean Basins: Geological Society London Special Publications, 42, pp. 313–345. Takazawa, E., Okayasu, T., Satoh, K., 2003. Geochemistry and origin of the basal lherzolites from the northern Oman ophiolite (northern Fizh block). Geochemistry, Geophysics, Geosystems 4, 1021. http://dx.doi.org/10.1029/2001GC000232. Tamura, A., Arai, S., 2006. Harzburgite–dunite–orthopyroxenite suite as a record of supra-subduction zone setting for the Oman ophiolite mantle. Lithos 90, 43–56. Tatsumi, Y., Hamilton, D.L., Nesbit, R.W., 1986. Chemical characteristics of fluid phase released from lithosphere and origin of arc magmas: evidence from high pressure experiments and natural rocks. Journal of Volcanology and Geothermal Research 39, 293–309. Wells, P.R.A., 1977. Pyroxene geothermometry in simple and complex systems. Contributions to Mineralogy and Petrology 62, 129–139.