J. metamorphic Geol., 2007, 25, 657–682

doi:10.1111/j.1525-1314.2007.00722.x

New, high-precision P–T estimates for Oman blueschists: implications for obduction, nappe stacking and exhumation processes P . Y A M A T O , 1 P . A G A R D , 1 B . G O F F E´ , 2 V . D E A N D R A D E , 3 O . V I D A L 3 A N D L . J O L I V E T 1 1 Laboratoire de tectonique, Universite´ Pierre et Marie Curie - Paris 6, CNRS, UMR 7072, 4 place Jussieu, F-75005 Paris, France ([email protected]) 2 Laboratoire de Ge´ologie de l’Ecole Normale Supe´rieure, CNRS, UMR 8538, 24 rue Lhomond, 75005 Paris, France 3 LGCA, Universite´ Joseph Fourier, CNRS, UMR 5025, 1381 rue de la piscine, 38041 Grenoble, France

ABSTRACT

Oman blueschists and eclogites lie below the obduction nappe of the Semail ophiolite in one of the key areas on Earth for the study of plate convergence. Here new metamorphic and tectonic constraints are provided for the central, yet poorly constrained Hulw unit, sandwiched between the low-grade units (10 kbar, <300 C) and the As Sifah eclogites (Pmax  23 kbar; Tmax  600 C).TWEEQU multiequilibrium thermobarometry, using both compositional mapping and spot analyses, and Raman spectroscopy of carbonaceous material yield a high-precision P–T path for the Hulw and As Sheikh units and reveal that they shared a common P–T history in four stages: (i) a pressure decrease from 10– 12 kbar, 250–300 C to 7–9 kbar, 300–350 C; (ii) almost isobaric heating at 8–10 kbar from 300– 350 C to 450–500 C; (iii) a pressure decrease at moderate temperatures (450–500 C); and (iv) isobaric cooling at 5–6 kbar from 450–500 to 300 C. No significant pressure or temperature gap is observed across the upper plate–lower plate discontinuity to the north and west of the Hulw unit.The combination of tectonic and P–T data constrains the stacking chronology of the three main metamorphic units comprising the Saih Hatat window (i.e. the Ruwi-Quryat, the Hulw-As Sheikh and the Diqdah-As Sifah units). These results strengthen the view that the tectonic and metamorphic data are conveniently accounted for by a simple, N-vergent continental subduction of the passive Arabian margin below the obduction nappe along a cold P–T gradient. Key words: exhumation; HP-LT metamorphism; multi-equilibrium thermobarometry; obduction; Oman blueschists.

INTRODUCTION

Oman constitutes one of the world’s most spectacular areas for the study of plate convergence, with both well-preserved obduction remnants (the famous Semail ophiolite sequence; e.g. Coleman, 1981; Boudier et al., 1988; Nicolas, 1989; Searle & Cox, 1999) and high pressure–low temperature (HP-LT) blueschists and eclogites typical of subduction zones (Goffe´ et al., 1988; El-Shazly et al., 1990; Saddiqi et al., 2006). These latter HP-LT rocks occur as an extensive nappe stack below the ophiolite, in the Saih Hatat window (e.g. Goffe´ et al., 1988; Searle et al., 1994; Jolivet et al., 1998; Fig. 1), with P–T conditions ranging from those of very low-grade blueschists (<300 C lawsonite, carpholite-kaolinite) to those of eclogites (Fig. 2; >12 kbar for 580 C, El-Shazly et al., 2001; possibly >20 kbar for 520 C, Warren & Waters, 2006). This HP-LT metamorphism was classically considered to have formed as a result of the obduction processes initiated at c. 95 Ma (Lippard, 1983; Michard et al., 1983; Goffe´ et al., 1988; Searle & Cox,  2007 Blackwell Publishing Ltd

1999). The structure and genesis of the blueschists, however, have received much attention, new interpretations and conflicting scenarios in the recent years (e.g. El-Shazly et al., 2001; Breton et al., 2004; Gray et al., 2004a,b, 2005a; Searle et al., 2004, 2005). Whereas Goffe´ et al. (1988) and Searle et al. (2004) related the HP-LT metamorphic imprint to the NEvergent subduction of the leading edge of the Arabian continental margin below the ophiolite and exhumation patterns to reverse movements taking place along the subduction plane, El-Shazly et al. (2001) and Breton et al. (2004) related the HP-LT imprint to the existence of an additional, NE-dipping intracontinental subduction isolating a North Muscat microplate. On the other hand, Gray and co-workers (e.g. Gregory et al., 1998; Gray et al., 2000, 2004a,b; Gray & Gregory, 2003) suggested that the HP event involved a precocious intracontinental subduction (c. 130–90 Ma) dipping to the SW under the Arabian passive margin, based on the distribution of metamorphic ages and mapping observations. Part of the dispute lies in the boundaries between structural (and thus metamorphic) sub-units and on 657

658 P. YAMATO ET AL.

55°

Thrust

UP-LT discontinuity 58˚35

W

Q

Stretching lineation (sense of shear)

EU

co

N

RA

llis

ion

SIA

58˚40

Ruwi

23˚35

30°

30°

2 km

F

Persian Gulf

25°

(Makra

n)

subd

uctio

250 km

YS

Z

SZ AW

AK

Om

an

n

SH

Yiti

SZ

Yenkit

23˚30 aud

As Sheikh

?

el A

Hulw

Jeb

Rija

HSZ

bu D

? ?

As Sifah

? 23˚25

Diqdah

? ?

Ruwi-Quryat major unit

? Progressive transition Ruwi-Quryat to Hulw?

Ruwi-Yiti unit

Mayh unit

Diqdah unit

mafic bodies

Bawshar-Khayran unit

Hulw unit

LAS

Eocene

Saih Hatat unit + Quryat

As Sheikh unit

UAS

As Sifah unit

Ophiolite

Fig. 1. Simplified geological map of the metamorphic units of the Saih Hatat (modified after Le Me´tour et al., 1986a,b; Goffe´ et al., 1988; Jolivet et al., 1998; Miller et al., 2002; Searle et al., 1994, 2004; Saddiqi et al., 2006). Stretching lineations and shear senses are from the literature (Michard et al., 1994; Jolivet et al., 1998) and this study.

the interpretation of an upper plate–lower plate discontinuity (UP-LP, hereafter; e.g. Gregory et al., 1998; Miller et al., 1998 and subsequent papers; El-Shazly et al., 2001; Gray et al., 2005a; Searle et al., 2005). Moreover, the large central part of the nappe stack, the Hulw unit, still suffers from unresolved P–T constraints (Fig. 2; Goffe´ et al., 1988; El-Shazly et al., 1990, 2001; Searle et al., 1994, 2004, 2005; Jolivet et al., 1998; Saddiqi et al., 2006), and the validity of the pressure estimates for the highest grade samples was contested by El-Shazly (2001). The aim of this study was to bring new petrological data, together with tectonic observations, in order to shed some light on the stacking of the

HP-LT units and on their geodynamic context. First, high-precision P–T estimates and P–T paths are presented for the poorly constrained, high-variance and strongly retrogressed Hulw metapelites, thanks to the development of new thermobarometric techniques (Berman, 1991; Vidal & Parra, 2000; Beyssac et al., 2002; Vidal et al., 2006). The relationship between the Hulw blueschists and their neighbouring units is established, particularly with the higher grade blueschists and eclogites from As Sifah. Finally the metamorphic history of the Hulw unit is discussed in relation to the framework of the whole Saih Hatat nappe stack and regional scale implications are outlined.  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 659

10

˚C .km

22

-1

5˚C . km

-1

24

20

LAS ?

18

whereas the northern Permian to Jurassic cover units experienced a somewhat lower grade metamorphism and constitute the Ruwi-Quryat major unit (see below). Late Paleocene–Early Eocene sedimentary rocks rest upon the HP-LT rocks of the Ruwi-Quryat unit, while Late Maastrichtian rests uncomformably upon the ophiolite.

Pressure (kbar)

16

-1 .km 5˚C

UAS

14

1 D-Ash

12 HULW

R-Q

4

10

2 6

8

5 3

6

?

1 1

4 2 0

100

200

300

400

500

600

Temperature (˚C) Fig. 2. Compilation of published P–T paths for the different metamorphic units of the Saih Hatat window. P–T paths in grey correspond to those compiled and averaged, for the sake of clarity, for the following units: Ash, As Sheikh; D, Diqdah; LAS, Lower As Sifah; RQ, Ruwi-Quryat Major Unit; UAS, Upper As Sifah. LAS?, contrasted maximum pressure conditions were reported for the LAS unit (El-Shazly et al., 2001; Warren & Waters, 2006). The different P–T–t paths for the Hulw unit from the literature are given in black: 1: Goffe´ et al. (1988); 2: Michard et al. (1994); 3: Searle et al. (1994); 4: Jolivet et al. (1998); 5: El-Shazly et al. (2001); 6: Saddiqi et al. (2006) (the simplified P–T–t paths in grey are after the same authors).

GEOLOGICAL SETTING Lithostratigraphy and overall structure

The Saih Hatat window constitutes the only area in Oman with HP-LT metamorphic index minerals. It is mainly made of quartz mica-schists, calc-shists and mafic schists of continental origin, which are overlain by oceanic foredeep deposits (e.g. Muti, Hawasina formations; e.g. Glennie et al., 1974) sandwiched between the continental margin and the oceanic ophiolite proper (Hanna, 1990). This HP-LT region was the focus of considerable mapping efforts (e.g. Glennie et al., 1974; Le Me´tour et al., 1986a,b; Gregory et al., 1998; Miller et al., 2002) and the lithostratigraphy is well known (e.g. Glennie et al., 1974; Le Me´tour et al., 1986a,b; for recent compilations, see also Miller et al., 2002; Breton et al., 2004; Searle et al., 2004). Proterozoic, Early Ordovician and Permian rocks of the Saih Hatat dome correspond to the most metamorphic units (the so-called lower plate; Miller et al., 1998),  2007 Blackwell Publishing Ltd

Previous P–T estimates and relevant sub-units

Based on metamorphic zoneography, Goffe´ et al. (1988) first distinguished the As Sifah, Hulw, Rija, Mayh Bandar, Yiti and Quryat units and the Mascate nappes (including units such as Ruwi-Al Amarat and the Saih Hatat s.s.). Searle et al. (1994) later proposed to group the Yiti, Al Amarat and Rija units, extended the Mayh Bandar unit to the Wadi Mayh-Al Khuryan unit, and separated the As Sifah eclogites (Lower As Sifah; LAS, hereafter) from the As Sifah glaucophanebearing unit (Upper As Sifah; UAP, hereafter). ElShazly et al. (1990, 2001) and El-Shazly (1994, 1995) defined three regions (I, II & III). Region I corresponds to the Ruwi unit, while region II groups the Al Amarat, Mayh Bandar, Bawshar-Khayran and the Quryat units. Region III corresponds to the core of the Saih Hatat dome, where three zones are distinguished [A, B & C, which respectively correspond to the Hulw (+Rija), Diqdah and As Sifah units], and to what Miller et al. (1998) defined as their Ôlower plateÕ. Jolivet et al. (1998) grouped regions I and II of El-Shazly and co-workers into the ÔRuwi-Quryat major unitÕ characterized by the presence of Fe-Mg carpholite and pyrophyllite, together with incidental chloritoid (Saddiqi et al., 2006). Within the Ôlower plateÕ, they distinguished a garnet-glaucophane bearing Diqdah unit, separated from Hulw by the As Sheikh unit. As a result of the lack of new P–T constraints over the last decade, only minor modifications in the definition of these units were brought in recent papers (e.g. Searle et al., 2004; Saddiqi et al., 2006). For the sake of simplicity, five main metamorphic units are thus considered below and presented in Fig. 1: the upper plate Ruwi-Quryat unit, and the lower plate Hulw, As Sheikh, Diqdah and As Sifah units. A compilation of published P–T paths (Fig. 2) shows that most authors support the P–T estimates for the As Sifah (e.g. Wendt et al., 1993; Searle et al., 1994), Diqdah and RuwiQuryat units, although El-Shazly (2001) suggested that the pressure values obtained are overestimated, particularly for the HP As-Sifah unit (LAS; Fig. 2). In contrast, the central and voluminous Hulw unit, whose P–T conditions are critical for the understanding of the whole nappe stack, is characterized by a wide scatter of P–T estimates (Fig. 2). Goffe´ et al. (1988) proposed an anti-clockwise P–T evolution for the Hulw unit based on carpholite overgrowths after chloritoid, whereas El-Shazly et al. (1990) and ElShazly (1994, 1995) suggested that this unit reached its maximum burial before the thermal peak. Searle et al.

660 P. YAMATO ET AL.

(1994) estimated the metamorphic peak for Hulw at around 7–8 kbar, 380–420 C and El-Shazly (2001) revised his earlier estimates for the Hulw unit to lower pressure values (6.5–8.5 kbar; 400–460 C). Jolivet et al. (1998) attributed a clockwise retrograde path to the Hulw unit, with pressure and temperature peaks at 11–12 kbar and 280–320 C, and proposed, for the As-Sheikh and Diqdah units, a common P–T evolution at 12–15 kbar and 540 C. This petrological study therefore tried to estimate reliable P–T conditions for the Hulw unit, and answer such questions as: (i) why is this unit so retrogressed? (ii) what is the P–T contrast with the adjacent Mayh and As Sheikh units? (iii) what is the pressure difference with the As Sifah unit? Geodynamic setting: timing of HP-LT metamorphism v. obduction

One of the major unresolved issues for the Oman HPLT rocks is the age of the metamorphic peak, particularly for the As Sifah eclogites (see discussions by Warren et al., 2003, 2005; Gray et al., 2004b, 2005a; Searle et al., 2004, 2005; for a historical perspective, see Saddiqi et al., 2006). A compilation of radiometric ages, particularly numerous for the As Sifah unit, is presented in Fig. 3. There is a wide scatter in Ar/Ar and K/Ar ages on white mica between c. 125 and 70 Ma (Fig. 3b), hence before and after the start of obduction at c. 95 Ma (Tilton et al., 1981; Hacker, 1994; Hacker et al., 1996; Searle & Cox, 1999, 2002). The scatter is notably less for the Hulw unit (77– 98 Ma; Fig. 3a). Fission track dating yields comparably tighter constraints, suggesting that the exhuming HP-LT units reached 260 and 100 C (closure temperature on zircon and apatite, respectively; e.g. Wagner & Van den Haute, 1992; Tagami et al., 1998) at 66–70 and 45–48 Ma, respectively (Fig. 3a). Gray et al. (2004a,b, 2005a) considered that these Ar ages post-date the HP-LT event, in part because they correspond to cooling ages at 350–400 C rather than to crystallization ages of the As Sifah eclogite (600– 650 C). This is also supported by Sm/Nd ages obtained on garnet (110 ± 9 and 119 ± 13 Ma) by the same authors (Gray et al., 2004b). In their view, following Montigny et al. (1988) and El-Shazly & Lanphere (1992), HP-LT metamorphism precedes the obduction of the Neotethys onto the Arabian margin and implies the early subduction, at 130–95Ma, of an exotic continental microplate (e.g. Gregory et al., 1998; Gray et al., 2004a).

In contrast, Searle et al. (1994) and El-Shazly et al. (2001) discounted white mica Ar ages on the assumption of excess argon, as in a number of other settings worldwide (e.g. Arnaud & Kelley, 1995; Sherlock & Kelley, 2002), and based on concordant ages obtained by two independent methods (Rb/Sr: El-Shazly et al., 2001; U/Pb on zircon: Warren et al., 2003, 2005; Gray et al., 2004b), Searle & Cox (2002), Searle et al. (2004) and Warren et al. (2005) suggested that the HP event effectively occurred around 85–80 Ma, hence after the onset of obduction at c. 95 Ma. They also argued that, according to closure temperatures, Sm/Nd ages should be less than U/Pb ages (Fig. 3) and suggested that the older ages obtained by Gray et al. (2004b) could be accounted for by residual Nd in garnet (Searle et al., 2005). Although at present the link between HP-LT metamorphism and obduction cannot be regarded as certain (i.e. further dates are needed to clarify the exact timing of HP-LT metamorphism), in our view this second interpretation is the most plausible and HP-LT metamorphism was most likely an obduction-related, short-lived process. This interpretation has the advantage, over that of a now vanished exotic microplate (Gray et al., 2004a), that it can be tested on geological grounds and that it is not incompatible with available stratigraphic constraints on the foredeep sediments overthrust by the obducted ophiolite and the passive character of the Arabian margin (Scott, 1990; Breton et al., 2004; Searle et al., 2005), the chronology of ophiolite emplacement (Boudier et al., 1988; Nicolas, 1989; Hacker, 1994; Hacker et al., 1996) and the triggering of obduction (Agard et al., 2007). Tectonic patterns

The tectonic units of the Saih Hatat HP-LT dome are mainly differentiated on the basis of metamorphic constraints (see above) and on structural grounds. The overall structure is quite complicated because of an intense deformation marked by spectacular sheath folds, boudinage, shear zones of variable extent (e.g. Jolivet et al., 1998; Miller et al., 2002; Searle et al., 2004) and the interference of successive generations of structures. The present-day controversy over the structures of the Saih Hatat dome (e.g. Searle et al., 1994, 2004; Gray et al., 2004a, 2005a) lies more on the interpretation of these structures, than on the structures themselves. Some of the most conflicting and/or ill-constrained aspects are given below.

Fig. 3. Compilation of the radiometric ages for the HP-LT units of the Saih Hatat dome. (a) Spatial distribution of the ages (same legend as for Fig. 1 for the metamorphic units). Ages in bold correspond to those of the Hulw and As Sheikh units. Key to brackets: AA: Ar/Ar; KA: K/Ar; FT: Fission tracks; RS: Rb/Sr; SN: Sm/Nd; UPb; U/Pb; A: Apatite; Cross: Crossite; E: Epidote; G: Garnet; Gln: Glaucophane; M: Muscovite; O: Omphacite; P: Phengite; Pg: Paragonite; WR: Whole rock; Z: Zircon. Other symbols: *plateau age; **total gas age; tf: total fusion age. (b) Histogram of radiometric ages for Oman blueschists and eclogites. Data from the literature, after Montigny et al. (1988), El-Shazly & Lanphere (1992), Searle et al. (1994), Saddiqi et al. (1995, 2006), Poupeau et al. (1998), Miller et al. (1999), El-Shazly et al. (2001), Warren et al. (2003, 2005), Gray et al. (2004a,b). (c) Average closure temperatures (Tc) for the various radiometric methods used in Oman.  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 661

(a)

58˚35

58˚40

58˚45

Ruwi

23˚35

79 +/- 0.3 (UPb/Z; 5 ages) 82 +/- 1 (UPb/Z) 110 (SN/G) 79 (RS/O-P) 78 (RS/WR-P) 78 (RS/E-P; 2 ages) 112 (KA/P) 79 (RS/WR-P) 109 (KA/P) 77 (RS/E-P; 2 ages) 88 (KA/P) 141 (AA/P)* 82-90 (AA/P) <133 (AA/P) 80 (KA/WM) 117 (AA/P)* 67.5 (KA/Cross) <115 (AA/P) 67 (KA/Gln) <108 (AA/P) 101 (AA/P)* 101 (AA/P)* 96 (AA/M) As Sheikh <96 (AA/P; 2 ages) <92 (AA/P) 88 (AA/P)* 90 (AA/P)* 83 (AA/P)** 89.5 (AA/P)* <83 (AA/P) 86 (AA/P)* 67 (FT/Z) 84 (AA/P)* 55 (FT/A)

80 (AA/WM)* 95 (KA/WM)

69 (TF/Z) Yiti Yenkit

70 (AA/M)

23˚30 78 (KA/WM) 76 (AA/M) 91 (AA/MP)* 472 (AA/M)

76 (AA/P)* 73.5 (AA/P)tf

58˚50

72 (AA/P)tf

<146 (AA/P)

82 (AA/MP) 239 (KA/WM) 70 (FT/Z) 45 (FT/A)

82-98 (AA/M)

23˚25 67 (FT/Z) 69 (FT/Z)

102 (AA/P)** 82 (AA/MP)* 89 (AA/M) 122 (AA/M)

83 (AA/P)* 82 (AA/P)* 79-80 (AA/M-P) Diqdah 66 (FT/Z) 48 (FT/A)

103 (KA/WM) 69 (FT/Z)

(b)

HP Exhumation Intense top to the NE-shearing HP Older ages due to Ar Excess

Contrasted interpretations

Hulw / As Sheikh (~260°C)

12

119 (AA/P)* 107 (KA/P-Pg) 94 (AA/P)* 81-84 (AA/P)

119 (SN/G) 101 (AA/M) 102 (AA/P)* 111 (AA/P)* 107 (AA/P)* 104 (AA/P)** 131 (KA/WM) 102 (KA/Gln) 111 (KA/P) 89 (KA/P) 79 (KA/Cross) 76 (KA/Cross)

Montigny et al., 1988 e.g. Miller et al., 1999 Gray et al., 2004 e.g. Searle et al.,1994, 2004 El-Shazly et al., 2001

Hulw / As Sheikh (350-400°C)

Radiometric Ages

Number of ages

10 8 6 4 2 0 65

70

4

Others 4 time constraints

(c)

FT/Z Tc ~ 260˚C

75

80

85

3

2

sediments sealing the obduction stack and resumption of extension

KA Tc ~ 250-350˚C

90

95

Age (Ma)

1 Continental subduction

3 of the Arabian margin

HP-LT metamorphism (?)

AA/WM (all ages) AA (plateau only) Tc ~ 350-400˚C

 2007 Blackwell Publishing Ltd

100 105 110 115 120 125 130 135 140 145

youngest sediments below

2 the obduction nappe

RS + UPb Tc (RS) ~500˚C Tc (UPb/ Z) > 850˚C

Beginning of

1 ophiolite emplacement

SN/G Tc > 600-650˚C

662 P. YAMATO ET AL.

The interpretation of the upper plate–lower plate discontinuity

Gregory et al. (1998) and Miller et al. (1998, 2002) highlighted the major tectonic discontinuity separating the Ruwi-Quryat major unit (their so-called upper plate; UP) from underlying units (their lower plate; LP). Whether this discontinuity (more or less equivalent to the Hulw shear zone; Searle et al., 1994) is rooted in the Moho (Gray et al., 2000) is highly controversial (Searle et al., 2004, 2005; Gray et al., 2005a).

proposed that each metamorphic unit (LAS, UAS, Diqdah, As Sheikh and Hulw) is separated from the other by localized extensional shear zones. Miller et al. (2002) suggested that the shear zone between Hulw and the deeper units is located below, and truncated by, the UP-LP discontinuity. If this holds true, a clear P–T contrast should be observed on each side of the 2 km wide stretch of upper plate located to the east of Hulw (Jebel Abu Daud; Fig. 1), as well as across the UP-LP discontinuity. This will be discussed in the light of our new P–T estimates. NEW TECTONIC OBSERVATIONS

Exhumation patterns and shear senses

They are clearly top to the NE as per Jolivet et al. (1998) and dominantly top to the NE, yet more coaxial, as Michard et al. (1989, 1994) and Miller et al. (2002). Early S-directed shear senses were reported by Michard et al. (1994) but contested by Miller et al. (2002). Searle et al. (1994) reported the existence of several shear zones with top to the SW transport direction in the upper plate (from S to N: Hulw, Al Wudya, Al Khuryan and Yenkit shear zones; Fig. 1). In contrast, Jolivet et al. (1998) mapped extensional shear zones corresponding to unit boundaries. In any case, pressure gaps across the various shear zones are unclear.

Considerable structural analysis of the Saih Hatat already exists in the literature, and the reader is referred to previous mappings and detailed observations by Le Me´tour et al. (1986a,b) and Miller et al. (2002), and to additional reports by Michard et al. (1994), Searle et al. (1994, 2004) and Jolivet et al. (1998). The aim of our study was mainly to accurately assess the position of our samples in the nappe pile (cf. schematic 3D diagram showing the main structures in the northern Saih Hatat window in Fig. 4). In addition, complementary tectonic observations are reported below, with particular emphasis on stretching and boudinage, whose importance for the nappe stacking, in our view, was somewhat neglected so far.

The problematic, central Hulw unit

Shear senses

The structural relationship between the central Hulw unit, generally outlined as a separate unit by previous authors (rather paradoxically, because of its imprecise P–T constraints), and other neighbouring units is unclear. The Mayh unit, for example, was either related to the upper plate units or to the Hulw unit (Goffe´ et al., 1988; Saddiqi et al., 2006). Jolivet et al. (1998)

Ductile deformation is rather evenly distributed within the Hulw unit, even if some domains are more strongly sheared because of lithological contrasts. Shear senses, as reported by Jolivet et al. (1998), are uniformly striking top to N020–N030E on average, and deformation is mostly non-coaxial and accompanied by NNE-vergent drag folds. New kinematic data are

Fig. 4. 3D diagram across the Saih Hatat dome (same unit legends as for Fig. 1). The location of the main studied samples is shown (white boxes). Numbers in the stars correspond to the location of Figs 5 & 6.  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 663

(a) SSW

NNE

~10 m NNE

SSW

LP

UP

~10 m r ban

d

Sample m96.B Prl

Shea

(b)

Phg

Phg

d

Cl

500 µm

Qtz

Fig. 5. (a) Deformation observed near the UP-LP discontinuity (see Fig. 4 for location). The contrast of deformation style between the upper plate and the lower plate upper plate is well marked. Upper plate is mainly characterized by brittle behaviour associated with folding whereas many ductile shear bands coupled with boudinage affect the lower plate. (b) Back-scattered electron image (15 kV) showing chloritoid (Cld), Phengite (Phg) and Pyrophyllite (Prl) in a shear band just below the UP-LP discontinuity (sample m96.B).

shown in Fig. 1. The deformation observed at the roof of the Hulw unit, below the UP-LP discontinuity, is consistent with the internal deformation of the unit (Fig. 5a). In order to track the P–T conditions of this deformation, one sample (d04.B) was chosen from an N020 shear band within the core of the Hulw unit. The latest ductile deformation increment is marked by PrlPhg recrystallization in shear bands north of sample m73B2 (Figs 1 & 5b). Large-scale boudinage

Numerous decimetre- to kilometre-sized metabasite boudins (i.e. Saiq 2 Formation, Le Me´tour et al., 1986a,b or Sq1dL Formation, Miller et al., 2002) wrapped in reddish Sq1L limestones are found in the Hulw unit. One of the largest boudinaged mafic layers is visible on the map of Fig. 1 and detailed in Fig. 6. These metamorphosed volcanics (El-Shazly, 1994;  2007 Blackwell Publishing Ltd

Warren & Waters, 2005, 2006) correspond to an antiformal, N-trending sheath fold axis sheared off along its limbs. The existence of synthetic shear bands points to an intense combination of flattening and shearing (Fig. 6a). The shear bands located within the limbs of this fold (Fig. 6b) contain glaucophane and/or greenschist facies minerals (i.e. epidote-chlorite-albite). This major sheath fold dragging the rocks thus operated prior to and during the main top to the NNE deformation. The strongly foliated matrix limestones are folded into a series of disharmonic folds (Fig. 6c,d). The same rock types are found in As Sifah and Diqdah units, supporting the view that all units from the lower plate belong to the same terrane (Le Me´tour et al., 1986a,b; Miller et al., 1999; Searle et al., 2004). However, in the As Sifah unit, small, decimetre- to hectometre-sized mafic boudins are dispersed in the strongly folded, reddish, matrix limestone, whereas hectometre- to kilometre-sized elongated, continuous mafic slices and boudins are found in the Diqdah unit. Our interpretation is that the boudinage gradient increases from Hulw to As Sifah (Fig. 6e). It should be added that the Diqdah unit is highly retrogressed, in line with earlier reports suggesting the presence of a major shear zone in the vicinity of the As SheikhDiqdah units (Searle et al., 1994; Jolivet et al., 1998; As Sheikh shear zone; Gray et al., 2004a). Contrast in deformation style between UP and LP and deformation chronology

The deformation style is markedly different in the UP, where there is far less stretching, shear movement and boudinage (except in the sheath fold hinges of southern Wadi Mayh; Miller et al., 2002, their fig. 9; Searle et al., 2004, their fig. 10). Deformation is also much more coaxial in the UP (Miller et al., 1998, 1999, 2002), with only scattered ductile to brittle shear zones (Searle et al., 1994) formed at low metamorphic temperatures (350 C; Jolivet et al., 1998). Overall, finite deformation is less intense than in the LP, as demonstrated by Gray et al. (2005b). Within the more or less continuous deformation process affecting the Saih Hatat unit, the following chronology can be outlined: Step 1: Flattening, stretching and boudinage at depth, marked by shear zones within and between the units and early folds, affected the LP units. This deformation did not affect the Ruwi-Quryat major unit. Step 2: The major NNE-vergent deformation stage emplacing the regional fold-nappe along the UP-LP discontinuity (Gregory et al., 1998) was marked by atype, sheath folds affecting both the upper and lower plates, and late recumbent folding, particularly visible in the upper plate Hijam and Amdeh formations, associated with a S-dipping foliation. Step 3: A penetrative crenulation, with late axial planes dipping WSW, later affected all units.

664 P. YAMATO ET AL.

SSW

NNE

b d

N

~2 km c (a)

S

N

E

W Sq1L

metabasite SqdL (b)

(c)

~50 m

N S Sq1L

SqdL

(d)

~50 m SSW

NNE

(e)

Ruw

i-

yat Qur

H u lw -A

s

S h e ik h

UP LP D iq d a h

A s S ifa h

~10 km Fig. 6. (a) Schematic representation of large-scale boudinage in the Hulw unit (see Fig. 4 for location; b, c and d refer to the photographs below, same unit legend is used). (b) Shear bands from the stretched upper limb of the metabasite boudin. (c, d) Disharmonic folding in the reddish Sq1L limestones wrapping the metabasite boudins. (e) Schematic cross-section of the Saih Hatat window highlighting the deformation contrast on both sides of the UP-LP discontinuity.  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 665

ANALYTICAL TECHNIQUES Sampling

Many samples, mainly metapelites (+minor metabasites) and dark, carbonaceous-rich limestones, were collected across the entire Saih Hatat window, with 58˚35

58˚40

Ruwi

23˚35

W Q

special emphasis on the Hulw unit (Fig. 7). Twenty samples were studied by EPMA (see below) and 12 thin sections were selected for detailed thermobarometric calculations. Mineral occurrences for the latter samples are presented with GPS coordinates in Table 1. Among the compositional maps acquired, four were selected according to their location: two 58˚45

58˚50

2 km P–T estimates samples

F

RSCM samples

YS Z

SZ AW

AK SZ

Yiti

d51 d27

23˚30

Yenkit

m78.B3

d29 d52

m73.B2

m30.Bb2

d04.A

d10.B

As Sifah

?

d.72

d46

m29.C2

d69.2

d12

m77.B3

?

d48

m31.A2

d03.B1

23˚25

aud

m82.A4 el A

m71

m96.B

Jeb

HSZ

Rija m13.D1

d04.B

Hulw

bu D

?

d53.B

?

As Sheikh

d50

d40.2 m45.A2

m55.B2

d11.2

m27.A1

Diqdah

d32.2 d51 937f

d41.A

?

? Progressive transition Ruwi-Quryat to Hulw?

?

Fig. 7. Sampling localities (same legend for the metamorphic units as for Fig. 1). Samples with a white dot and a white font label were studied with the electron microprobe for P–T estimates (see Figs 8–10). Samples with a black star and black font label were used for the Raman spectroscopy of carbonaceous material (see Fig. 11).

Table 1. Parageneses of the samples from the Hulw and As Sheikh units (and corresponding GPS coordinates) on which conventional microprobe spot analyses (s) or compositional mapping (m) were performed. Ab: albite; Car: carpholite; Cc: calcite; Chl: chlorite; Cld: chloritoid; Phg: Phengite; Pg: paragonite; Qz: quartz; Rt: rutile. Sample

analyse

Qz

Cc

Phg

Chl

Cld

d29 d04.B d11.2 m29.C2 m30.Bb2 m55.B2 937f d41.A m45.A2 m77.B3 m78.B3 m82.A4

s s m s m+s s s s m s s m

x x x x x x x x x x x x

x x

x x x x x x x x x x x x

x x x x x x x x x x x x

x

 2007 Blackwell Publishing Ltd

x

Car

Pg

x

x

Ab

Rt

x x x x

x x x x x

x x x

x

Area Mayh Bandar Hulw W Hulw W Hulw Hulw Hulw Hulw SE As Sheikh S As Sheikh S As Sheikh N As Sheikh N As Sheikh N

GPS coordinates (N .¢.¢¢, E .¢.¢¢) 23.29.55 23.26.30 23.24.36 23.25.22 23.25.51 23.23.47 23.23.40 23.23.35 23.25.24 23.28.55 23.29.09 23.28.16

58.37.34 58.34.23 58.34.48 58.41.36 58.40.24 58.41.37 58.42.38 58.45.09 58.45.33 58.45.13 58.45.37 58.45.06

666 P. YAMATO ET AL.

 2007 Blackwell Publishing Ltd

Fig. 8. Results obtained through the compositional mapping of sample d11.2. (a) Ternary diagram for phengite (Ms: Muscovite; Cel: Celadonite; Prl: Pyrophyllite) and chlorite (Am: Amesite; Clin: Clinochlore; Dph: Daphnite; Sud: Sudoite). Red ellipses in dotted lines correspond to the range of values for all the samples. (b) Thin-section photomicrograph under cross-polars. Scale is shown in (c). (c) Back-scattered image of the area shown in (b). (d) Map of XMg values in chloritoid. (e) Map of Si content in phengite (Phg). Frames: families (A-H) distinguished on the basis of textural–chemical criteria. See text for details. (f) Map of XMg in chlorite (Chl). Frames: families (I–XI) distinguished on the basis of textural–chemical criteria. (g) Plots of Si content v. XMg for all the pixels of the map (top: for chlorite; bottom: for phengite). Points and groups corresponding to the families selected from the compositional map are indicated. (h) Results obtained with the selected families and details on how we defined the P–T paths shown in Figs 9 & 10 (see text for details). (h1) Results are shown with their INTERSX error-bars; [I,D]* reads as: result of the calculation performed with chlorite I and phengite D in equilibrium with a certain chloritoid composition (*). (h2) P–T envelope of the results (keyed to symbols in Fig. 9). (h3) Definition of the most reliable P–T path (based on the smallest standard deviations and the highest number of NRI; Table 2). (h4) Simplified overview of the P–T results obtained for this sample.

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 667

 2007 Blackwell Publishing Ltd

from the Hulw unit (d11.2 from the W, m30Bb2 from the core), and two from the S and N of the As Sheikh unit (m45.A2 and m82.A4, respectively). The Raman spectroscopy of carbonaceous material was performed on 21 samples chosen to document a possible gap of peak temperatures across the UP-LP discontinuity. Analytical methods Electron probe microanalysis

For mineral chemistry, two different types of electron probe microanalyses (EMPA) are acquired: conventional spot analyses and compositional maps (De Andrade et al., 2006). Analyses were both performed with a Cameca SX100 instrument (at Camparis, Univ. Paris 6; a Cameca SX50 was also used for additional conventional analyses). Classical analytical conditions were adopted for spot analyses [15 kV, 10 nA, wavelength-dispersive spectroscopy (WDS) mode], using Fe2O3 (Fe), MnTiO3 (Mn, Ti), diopside (Mg, Si), CaF2 (F), orthoclase (Al, K), anorthite (Ca) and albite (Na) as standards. Acquisition time for compositional mapping needs to be much shorter than that for conventional spot analyses (the 376 · 1350 pixel map of Fig. 8 corresponds to 507 600 analyses). In order to optimize counting times, acquisition time for compositional mapping of 5 lm point spacing grids was set at 100 or 200 ms (see De Andrade et al., 2006 for details concerning statistical checks). Operating conditions were 15 kV accelerating voltage, 100 nA beam current, 5 lm beam size and data acquisition in WDS mode. During the same microprobe session, standard spot analyses were also performed along profiles crossing the major minerals of the map (e.g. Chl, Cld, Phg), which were then used to calibrate the compositional maps (see De Andrade et al., 2006). TWEEQU multi-equilibrium thermobarometry

Mineral parageneses of the dominantly metapelitic Hulw samples are poor, containing mostly phengite, chlorite and in fewer cases, chloritoid ± albite ± paragonite (Table 1). In order to derive P–T estimates for these high-variance samples, the multiequilibrium approach of Berman (1991) was used as adapted by Vidal & Parra (2000), who defined several chlorite-phengite end-members (see also Vidal et al., 2001, 2005; Parra et al., 2002a, 2005). The TWEEQU 2.02 software (Berman, 1991) and its associated database JUN92 were used, complemented by thermodynamic properties for Mg-amesite, Mg-sudoite, Mg-celadonite for chlorite and phengite solid-solution models (25.min and 26.sln) of Vidal & Parra (2000). For phengite, the assym/25 solid-solution model is used. For chlorite, both V2001 and VP209 models are used. Only P–T estimates realized both with V2001 and

668 P. YAMATO ET AL.

VP209, however, are considered in equilibrium (P–T estimates obtained, with only one of them discarded). The temperature (rT) and pressure (rP) scatter were calculated with INTERSX (Berman, 1991). INTERSX was used only when the total number of the reactions are in the screen field (generally 2– 20 kbar; 200–650 C). Only the first value of INTERSX is taken into account. The minerals are then considered to be in equilibrium if these first values of rP or rT are lower than 10% of the P–T estimates, respectively. In order to keep only the most reliable values other P–T estimates are rejected, even if the last value given by INTERSX is correct. The different assemblages and end-members used in our study, the number of independent reactions (NIR) considered (a higher confidence is accorded to the assemblages with a high NIR), together with several typical output plots from TWEEQU, are shown in the Appendix. Estimation of peak temperatures (via RSCM)

The Raman spectroscopy of carbonaceous material (RSCM) was calibrated as a geothermometer (±50 C) in the range 330–650 C (Beyssac et al., 2002). Relative, inter-sample uncertainties on temperature can be much smaller, around 10–15 C (Beyssac et al., 2004). As the degree of organization of the carbonaceous material is irreversible, temperatures deduced from the Raman spectra represent peaktemperature conditions reached by the rocks. RSCM was performed on thin sections of graphite-bearing schists oriented perpendicular to the foliation by focusing the laser beam beneath a transparent crystal. Raman spectra were obtained with a Renishaw INVIA Reflex Raman micro-spectrometer at the Laboratoire de Ge´ologie of the Ecole Normale Supe´rieure, Paris, France. Spectra were excited at room temperature with the 514.5-nm line of a 20-mW Ar Spectra Physics laser through a LEICA 100· objective (NA 0.90). The laser beam is depolarized before the microscope using a 1/4 k waveplate. The laser spot on the surface had a diameter of approximately 1 lm and a power of 1 mW, which should be low enough to avoid any spectral change or sample destruction caused by light absorption and local temperature increase (e.g. Beyssac et al., 2003). Light was dispersed by a holographic grating with 1800 grooves mm)1. A spectral resolution of about 1.4 cm)1 was obtained by measuring a neon lamp emission. The spectrometer is calibrated for every session by measuring the position of the neon lamp emission and/or a silicon wafer. The dispersed light was collected by a RENCAM CCD detector. Confocality was achieved by setting the entrance slit into the spectrometer to 11 lm and selecting via the software few relevant rows on the CCD creating a ÔvirtualÕ pinhole. The depth resolution of this confocal configuration is <2 lm. The synchroscanmode from 700 to 2000 cm)1 was selected in

order to avoid step-like mismatches between neighbouring spectral windows probably occurring in samples with intense and uneven background and to maximize the signal-to-noise ratio. Acquisition duration was 60 s divided in three 20-s substractive runs. We recorded at least 10 spectra for each sample to take into account the CM heterogeneities. The program Peak Fit 4.0 was then used to process the spectra. MINERAL CHEMISTRY

The mineralogy of the Hulw unit is rather monotonous and poor compared with units such as As Sifah or Diqdah, where garnet and amphibole are commonly found. The typical Hulw paragenesis is quartz-phengite-chlorite ± chloritoid ± paragonite, albite, pyrophyllite and/or epidote (see Table 1). Tables 2 and 3 show some representative analyses from the different samples. A brief description of the main minerals is given below. Chlorite

Chlorite analyses showing an oxide sum <83 wt% or >89.5 wt% were excluded (e.g. Rimmele´ et al., 2004). Analyses are also rejected if K2O > 0.1 wt% and/or if (Na2O + K2O + CaO) >0.5 wt%. Only the analyses expressed as a linear combination of the amesite (Am), clinochlore (Clin), daphnite (Dph) and sudoite (Sud) end-members (Vidal et al., 2001, 2005) and meeting the six chemical criteria reported by Vidal & Parra (2000) were retained for multi-equilibrium calculations. Chlorite analyses cluster in a relatively confined area in the ternary diagram of Fig. 8a. The range of tschermak substitution variation (along the Am-Clin joint) is thus fairly restricted, and sudoite rarely exceeds 25%. In most cases, Sud-rich chlorite is texturally younger. Evaluation of the amount of ferric iron

The problem of the Fe3+, particularly in chlorite, was addressed by Vidal et al. (2001, 2005, 2006). Despite relatively constant compositions in the ternary diagram of Fig. 8a, chlorite often presents strikingly different colours under the microscope (brownish, reddish-brown, dark or pale green), which suggest variable Fe3+ contents. Besides, temperatures calculated with TWEEQU in the MASH system for chlorite-quartz (Vidal et al., 2001), without incorporating Fe3+ in chlorite and using only reaction (1) below, are too high (i.e. >650 C) for a number of chlorite from the Hulw and As Sheikh units, but become reasonable adding Fe3+ (see below). In order to estimate the quantity of Fe3+ within chlorite, the method described by Vidal et al. (2001, 2005) was used. For a constant pressure (10 kbar), Fe3+ is added in increments of 1% for each chlorite. The Fe3+ content for chlorite is considered reliable  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 669

Table 2. Representative analyses of the main mineral species from the Hulw and As Sheikh units in weight percents and their structural formulas. Mineral sample an

Cld m30Bb2 n13

Cld 937f o15

Chl m29.C2 m73

Chl m29.C2 m97

Chl m29.C2 n64

Chl m30Bb2 m55

Chl m30Bb2 o49

Chl m30Bb2 o51

Chl m30Bb2 Chl-z2B*

Chl d11.2 Chl-II*

Chl d11.2 Chl-IV*

Chl d11.2 Chl-VIII*

Chl d11.2 Chl-XI*

Chl d04. B q28

Chl d04. B q53

Chl M78B3 s75

Chl m78B3 s88

SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2 O R Structural Si Ti Al Fetot Mn Mg Ca Na K XMg nOx

24.45 0.04 40.58 26.15 0.20 1.38 0.02 0.03 0.02 92.86 formula 1.01 0.00 1.98 0.91 0.01 0.09 0.00 0.00 0.00 0.09 12

23.73 0.01 38.04 29.29 0.18 0.87 0.02 0.01 0.00 92.15

25.23 0.05 20.11 29.54 0.12 10.12 0.01 0.06 0.00 85.23

26.07 0.03 21.05 27.60 0.27 9.93 0.11 0.00 0.04 85.10

24.36 0.07 19.86 31.78 0.33 10.15 0.00 0.03 0.02 86.60

24.08 0.03 22.07 30.99 0.04 8.52 0.12 0.03 0.01 85.88

23.50 0.04 22.99 32.27 0.05 8.65 0.02 0.02 0.02 87.56

24.12 0.05 22.32 31.36 0.08 9.38 0.04 0.01 0.05 87.41

22.95 0.00 22.71 31.38 0.09 8.66 0.18 0.06 0.04 86.07

25.02 0.00 23.16 27.02 0.07 13.06 0.06 0.08 0.04 88.50

24.43 0.00 22.95 27.84 0.09 11.58 0.09 0.11 0.04 87.12

25.08 0.00 22.85 25.49 0.07 13.16 0.08 0.09 0.04 86.86

25.46 0.00 22.51 26.62 0.07 12.72 0.07 0.09 0.05 87.56

25.21 0.13 20.47 29.86 0.27 11.89 0.01 0.00 0.02 87.86

25.31 0.00 21.26 29.66 0.25 12.25 0.00 0.03 0.00 88.76

24.87 0.00 22.43 28.56 0.09 11.93 0.12 0.01 0.04 88.06

25.43 0.00 21.42 27.46 0.12 12.54 0.12 0.00 0.02 87.10

1.01 0.00 1.91 1.04 0.01 0.06 0.00 0.00 0.00 0.05 12

2.80 0.00 2.63 2.74 0.01 1.68 0.00 0.01 0.00 0.38 14

2.86 0.00 2.72 2.53 0.02 1.62 0.01 0.00 0.01 0.39 14

2.71 0.01 2.60 2.95 0.03 1.68 0.00 0.01 0.00 0.36 14

2.68 0.00 2.89 2.88 0.00 1.41 0.01 0.01 0.00 0.33 14

2.58 0.00 2.97 2.96 0.00 1.41 0.00 0.00 0.00 0.32 14

2.64 0.00 2.87 2.87 0.01 1.53 0.00 0.00 0.01 0.35 14

2.56 0.00 2.98 2.93 0.01 1.44 0.02 0.01 0.01 0.33 14

2.63 0.00 2.87 2.37 0.01 2.05 0.01 0.02 0.00 0.46 14

2.63 0.00 2.91 2.50 0.01 1.85 0.01 0.02 0.01 0.42 14

2.66 0.00 2.86 2.27 0.01 2.08 0.01 0.02 0.01 0.48 14

2.70 0.00 2.81 2.36 0.01 2.01 0.01 0.02 0.01 0.46 14

2.72 0.01 2.60 2.69 0.02 1.91 0.00 0.00 0.00 0.41 14

2.69 0.00 2.67 2.64 0.02 1.94 0.00 0.01 0.00 0.42 14

2.65 0.00 2.82 2.55 0.01 1.90 0.01 0.00 0.01 0.43 14

2.73 0.00 2.71 2.46 0.01 2.00 0.01 0.00 0.00 0.45 14

Mineral sample an

Phg m29.C2 m66

45.56 SiO2 0.34 TiO2 30.01 Al2O3 FeO 4.75 MnO 0.00 MgO 2.37 CaO 0.07 0.41 Na2O 8.71 K2 O R 92.21 Structural formula Si 3.17 Ti 0.02 Al 2.46 Fetot 0.28 Mn 0.00 Mg 0.25 Ca 0.01 Na 0.05 K 0.77 0.47 XMg nOx 11

Phg m29.C2 m68

Phg m30Bb2 m34

Phg m30Bb2 m42

Phg m30Bb2 m48

Phg m30Bb2 Phg-z1C*

Phg m30Bb2 Phg-z2D*

Phg d11.2 Phg-C*

Phg d11.2 Phg-F*

Phg d11.2 Phg-G*

Phg d11.2 Phg-H*

Phg d04.B q32

Phg m78.B3 s74

Phg m78.B3 s92

50.14 0.22 29.13 2.50 0.00 2.25 0.05 0.28 9.98 94.53

48.92 0.14 31.83 1.51 0.00 1.59 0.04 0.32 9.65 94.01

47.64 0.11 31.17 1.76 0.00 1.38 0.15 0.39 9.68 92.27

48.44 0.11 33.61 1.18 0.02 1.32 0.02 0.58 9.76 95.05

47.23 0.00 33.31 1.37 0.02 1.25 0.06 0.57 9.52 93.34

47.84 0.00 32.16 1.26 0.02 1.08 0.07 0.36 9.25 92.04

48.64 0.00 32.24 1.84 0.03 1.47 0.03 0.44 9.83 94.52

49.01 0.00 31.98 1.93 0.03 1.40 0.03 0.46 9.60 94.44

50.33 0.00 30.82 2.50 0.03 1.62 0.04 0.35 10.10 95.78

50.18 0.00 29.88 2.81 0.03 1.82 0.03 0.33 9.10 94.17

47.73 0.00 32.16 2.12 0.01 1.61 0.01 0.50 10.12 94.26

48.96 0.05 31.95 2.21 0.00 1.73 0.09 0.46 9.49 94.93

49.14 0.04 31.43 2.40 0.01 1.76 0.10 0.41 9.43 94.72

3.36 0.01 2.30 0.14 0.00 0.22 0.00 0.04 0.85 0.62 11

3.27 0.01 2.51 0.08 0.00 0.16 0.00 0.04 0.82 0.65 11

3.26 0.01 2.51 0.10 0.00 0.14 0.01 0.05 0.85 0.58 11

3.20 0.01 2.62 0.07 0.00 0.13 0.00 0.07 0.82 0.66 11

3.19 0.00 2.65 0.08 0.00 0.13 0.00 0.07 0.82 0.61 11

3.26 0.00 2.58 0.07 0.00 0.11 0.01 0.05 0.80 0.60 11

3.25 0.00 2.54 0.10 0.00 0.15 0.00 0.06 0.84 0.58 11

3.27 0.00 2.51 0.11 0.00 0.14 0.00 0.06 0.82 0.56 11

3.33 0.00 2.40 0.14 0.00 0.16 0.00 0.04 0.85 0.53 11

3.36 0.00 2.36 0.16 0.00 0.18 0.00 0.04 0.78 0.53 11

3.21 0.00 2.55 0.12 0.00 0.16 0.00 0.06 0.87 0.57 11

3.25 0.00 2.50 0.12 0.00 0.17 0.01 0.06 0.80 0.58 11

3.27 0.00 2.47 0.13 0.00 0.17 0.01 0.05 0.80 0.56 11

*Analyses corresponding to a family selected from compositional mapping (see Fig. 8). nOx: number of oxygen used to calculate the structural formula.

when the difference between the temperatures given by reactions (1), (2), (3) and (4) below for the chloritequartz assemblage (calculated with TWEEQU) is minimum. 2Clin þ 3Sud ¼ 4Mg-Am þ 7Qtz þ 4H2 O

ð1Þ

16Dph þ 15sud ¼ 20FeAm þ 6Clin þ 35aQtz þ 20H2 O ð2Þ 4Dph þ 6sud ¼ 5FeAm þ 3MgAm þ 14aQtz þ 8H2 O ð3Þ  2007 Blackwell Publishing Ltd

4Dph þ 5MgAm ¼ 5FeAm þ 4Clin

ð4Þ 3+

estiIt should be noted that the quantity of Fe mated at 10 kbar does not change significantly over the pressure range 5–20 kbar (i.e. range of interest for the Hulw and As Sheikh units). Adding 10–15% of Fe3+ is generally enough and the quantity added rarely exceeds 25% (Table 3). Analyses performed at the synchrotron have yielded similar amounts of Fe3+ within the chlorite for Hulw (Mun˜oz et al., 2006). Consequently, P–T conditions were estimated in this study using recalculated chlorite compositions with Fe3+. Using the empirical Kd relationship given by

670 P. YAMATO ET AL.

Table 3. Fe3+ estimates for chlorite from the Hulw and As Sheikh units.

XFe3+ Si Ti Al Fe2+ Fe3+ Mn Mg Ca Na K Sum Si XMg Avg T (1,2,3)

Chl m29.C2 m73

Chl m29.C2 m97

Chl m29.C2 n64

Chl m30Bb2 m55

Chl m30Bb2 o49

Chl m30Bb2 o51

Chl m30Bb2 Chl-z2B*

Chl d11.2 Chl-II*

Chl d11.2 Chl-IV*

Chl d11.2 Chl-VIII*

Chl d11.2 Chl-XI*

Chl d04.B q28

Chl d04.B q53

Chl m78.B3 s75

Chl m78.B3 s88

0.23 2.74 0.00 2.57 2.07 0.62 0.01 1.64 0.00 0.01 0.00 85.23

0.21 2.81 0.00 2.67 1.96 0.52 0.02 1.59 0.01 0.00 0.01 85.10

0.16 2.66 0.01 2.56 2.44 0.46 0.03 1.65 0.00 0.01 0.00 86.60

0.07 2.66 0.00 2.87 2.66 0.20 0.00 1.40 0.01 0.01 0.00 85.88

0.07 2.56 0.00 2.95 2.73 0.21 0.00 1.40 0.00 0.00 0.00 87.56

0.04 2.62 0.00 2.86 2.74 0.11 0.01 1.52 0.00 0.00 0.01 87.41

0.09 2.54 0.00 2.96 2.64 0.26 0.01 1.43 0.02 0.01 0.01 86.07

0.07 2.61 0.00 2.85 2.19 0.17 0.01 2.03 0.01 0.02 0.00 88.50

0.06 2.61 0.00 2.89 2.34 0.15 0.01 1.84 0.01 0.02 0.01 87.12

0.07 2.65 0.00 2.85 2.10 0.16 0.01 2.07 0.01 0.02 0.01 86.86

0.14 2.67 0.00 2.78 2.00 0.33 0.01 1.98 0.01 0.02 0.01 87.56

0.19 2.67 0.01 2.56 2.14 0.50 0.02 1.88 0.00 0.00 0.00 87.86

0.06 2.68 0.00 2.65 2.47 0.16 0.02 1.93 0.00 0.01 0.00 88.76

0.08 2.63 0.00 2.80 2.33 0.20 0.01 1.88 0.01 0.00 0.01 88.06

0.15 2.69 0.00 2.67 2.07 0.36 0.01 1.98 0.01 0.00 0.00 87.10

2.74 0.44 271

2.81 0.45 226

2.66 0.40 371

2.66 0.35 343

2.56 0.34 499

2.62 0.36 445

2.54 0.35 533

2.61 0.48 475

2.61 0.44 452

2.65 0.50 389

2.67 0.50 337

2.67 0.47 345

2.68 0.44 481

2.63 0.45 429

2.69 0.49 322

See text for details. Avg T (1, 2, 3) refers to the temperature at which reactions (1) to (3), given in the text, intersect at a given pressure with a scatter < 20C.

Vidal & Parra (2000), the amount of Fe3+ in phengite was estimated (e.g. Vidal et al., 2006). It turns out, however, that the recalculation of phengite yields, within error, the same P–T estimates. Phengite

Only phengite analyses which could be expressed as a linear combination of muscovite (Ms), pyrophyllite (Prl), celadonite (Cel), paragonite and biotite and meeting the six chemical criteria reported by Vidal & Parra (2000) were retained for multi-equilibrium calculations. All phengite analyses are shown in the Ms-Cel-Prl ternary diagram of Fig. 8a,g. These analyses show a minor Prl content and an evolution trend along the Ms-Cel Si(Mg,Fe)Al)2 tschermak substitution line, from values of 50% Ms-rich to the almost pure Ms end-member. On textural grounds, the Ms-rich phengite is always a product of late crystallization (see Fig. 8). Chloritoid

Chloritoid [Cld: (Fe, Mn, Mg)Al2SiO5(OH)2], which is nearly always retrogressed and oxydized, is found still fresh in some samples (e.g. d11.2; 937 f, m30.Bb2). These samples show XMg values roughly between 0.05 and 0.15. In some cases, XMg values increase from 0.05 to 0.15 from the core to the border, as shown in Fig. 8d (see below). This trend, mainly observed at the rim of the crystal, testifies to the fact that temperature increased during chloritoid growth. Fe-Mg carpholite

Carpholite [Car: (Fe,Mn,Mg)Al2Si2O6(OH,F)4], does not occur within the Hulw unit s.s., but is

found immediately north of it, in the Mayh unit, and in the Ruwi-Quryat major unit (Goffe´ et al., 1988). Tiny relic crystals hosted in quartz have an XMg of 0.3 and were replaced in the matrix by radial chloritoid aggregates with an XMg of 0.07 (sample d29). INSIGHTS FROM COMPOSITIONAL MAPS Mineral selection and strategy for P–T calculations

One example of a compositional map is shown here and described in detail (Fig. 8). The sample (d11.2) comes from the SW of the Hulw unit, and contains large chloritoid crystals lying in (and wrapped by) the chlorite-phengite-rich schistosity (+ additional quartz, calcite and oxides). The compositional map comprises 376 · 1350 pixels, which corresponds to 507 600 analyses 5 lm apart (1.5 day acquisition time). The quantification of the map has been realized thanks to the conventional spot analysis profiles shown in Fig. 8c, after removing biased analyses not fulfilling the above-mentioned criteria. This compositional map readily shows that: • Chloritoid is zoned (Fig. 8d) and shows an increase of XMg, from 0.05 to 0.15, during growth. • Phengite with a high Si (>3.35) and early phengite, oblique with respect to the schistosity, are located close to chloritoid (Fig. 8e). Phengite with a lower Si-content lies along and within the schistosity. • Colourless to brownish chlorite is also localized mainly in the schistosity (Fig. 8f). • The late, pale green chlorite located in the vicinity of a large oxide (upper part of Fig. 8f) has a high XMg value. In this study, P–T conditions were calculated after defining chemical–textural families. In many recent studies dealing with metapelites, equilibrium was  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 671

Table 4. Examples of calculations performed with TWEEQU (version 2.02., Berman, 1991). Minerals involved in the paragenesis Cld*

Chl

%Fe

3+

Phg

Others

Chl Mod.

m29.C2 (spot analyses) – m97 21

m68

–

–

m73

23

m66

–

–

n64

16

n63

–

m30Bb2 (spot analyses) 0.06 m55 7

m42

–

0.09

o51

4

m34

–

–

o49

7

m48

–

m30Bb2 (compositional map) 0.10 Chl-z2B 9 Phg-z2D

–

–

Phg-z1C

–

d11.2 (compositional map) 0.10 I 14

D

–

0.13

IV

7

B

–

0.12

VIII

6

E

–

–

X

7

F

d04.B (spot analyses) – q53 6

–

Ab,Prg

–

19

q32

–

m78.B3 (spot analyses) – s88 15

s92

–

–

s74

–

Chl-z2B

q28

s75

9

8

P–T calculations P

Pdev

T

Tdev

V2001 VP209 V2001 VP209 V2001 VP209

11.5 11.4 9.6 9.5 9.4 9.3

0.9 0.1 0.5 0.1 0.9 0.5

238 242 273 275 374 379

21 1 13 1 18 11

V2001 VP209 V2001 VP209 V2001 VP209

7.9 7.8 9.1 8.9 6.2 6.2

0.3 0.4 0.7 0.2 0.1 0.3

345 347 438 445 455 455

11 10 25 6 3 7

V2001 VP209 V2001 VP209

7.6 7.5 5.5 5.5

0.5 0.5 0.0 0.5

472 475 469 466

13 10 0.5 10

V2001 VP209 V2001 VP209 V2001 VP209 V2001 VP209

9.2 9.1 10.2 10.0 8.6 8.4 6.4 6.4

0.7 0.9 0.6 0.8 0.8 0.6 0.3 0.5

327 330 376 383 442 447 469 469

26 33 19 26 28 24 12 17

V2001 VP209 V2001 VP209

6.8 6.8 5.6 5.6

0.6 0.3 0.4 0.1

491 493 346 344

7 8 8 2

V2001 VP209 V2001 VP209

9.1 9.1 8.1 8.1

0.6 0.2 0.6 0.3

338 341 452 456

11 9 22 8

The corresponding analyses are given in Table 3. Same mineral abbreviations as for Table 1. *XMg value of chloritoid.

thought to be achieved only among minerals in contact and devoid of reaction features (Vidal & Parra, 2000; Trotet et al., 2001; Parra et al., 2002b; Rimmele´ et al., 2004). However, in high strained rocks (such as those from the Hulw unit), even though equilibrium was locally realized among minerals at a certain stage, they may have been later separated by progressive deformation, and assumptions of textural equilibrium alone may rest upon dubious grounds (not to mention that textures are 3D, not 2D as thin sections). For this reason, chemical–textural families are based on the recognition of distinct phyllosilicate compositions and on a critical evaluation of their textural location in the map (Fig. 8g). P–T estimates are then performed using the average chemical composition of each family. As an example, highly substituted phengite compositions A–D (Fig. 8e) are located close to the chloritoid (and referred to as group a, hereafter, for the sake of discussion). Phengite analyses E to H (group b below) have a lower Si content (particularly G and H) and correspond to later phengite from the  2007 Blackwell Publishing Ltd

schistosity. Concerning chlorite, a first group of Si-rich compositions, corresponding to chlorite located near the chloritoid, can be isolated (group 1: I to VII). Later chlorite compositions VIII to XI can be subdivided into two groups (group 2: chlorite VIII-X; group 3: chlorite XI), by recognizing that chlorite XI is a late pressure-shadow crystallization cutting across the schistosity. A clear chronological trend, from group 1 to group 3, is thus evidenced for chlorite. Reliable P–T equilibrium conditions are obtained when using group a phengite compositions (except A) with those of chlorite from group 1. All these minerals are moreover closely associated in space (lower half of the compositional map). Similarly, some of the phengite compositions from group b (F and E) are in equilibrium with group 2 chlorite compositions (and all are mainly in the upper part of the compositional map). P–T estimates with their uncertainties are shown in Fig. 8h. In order to refine the P–T path, a higher weight is given to estimates based on a high number of independent reactions (see the four steps of Fig. 8h; Table 4). Phengite A (the most Si-rich) could not be equilibrated with any chlorite, perhaps because the chlorite of that stage is not present in this part of the thin section or reacted out since. More puzzling, the late, muscovite-rich phengite analyses G and H, too, could not be equilibrated with any selected chlorite, even with chlorite XI. Finally, it should be noted that one chlorite from group 1 (V) could also equilibrate with some group b phengite compositions (E & F): such (rather rare) P–T estimates, lacking textural and chemical significance, were systematically rejected (Fig. 8, h1). Finally, compositional maps provide further evidence that minerals which crystallized at contrasted temperatures and pressures can be preserved in the same thin section, as already shown by Vidal et al. (2006) for example. Textural observations coupled with chemical information are thus a powerful tool for the determination of a P–T path. P–T estimates: maps v. spots

For the sake of comparison, both compositional maps over a selected area and spot analyses spanning the whole thin section were acquired for sample m30Bb2, taken from the core of the Hulw unit. Two compositional maps, located at both ends of a tiny, centimetrescale shear band, were acquired during the same microprobe session (Fig. 9a). Zone 1 compositional map (Z1, 600 · 200 pixels) contains chlorite, phengite, chloritoid, quartz and oxides (some are chloritoid pseudomorphs), while zone 2 compositional map (Z2, 320 · 166 pixels) contains only phengite and chlorite. Figure 9b,c demonstrates that for sample m30Bb2, the P–T estimates obtained from the compositional map analyses are similar to those calculated from conventional spot analyses. This result underlines the

672 P. YAMATO ET AL.

Thin-section m30.B.b2 (a)

z2

Chl

z1 Phg Chl

Phg Cld

500µm

12

12

12

10

10

10

8

8

8

6

6

6

4

T (˚C)

P (kbar)

300

400

4

T (˚C) 300

500

400

4

Results from point analyses

12

12

12

10

10

10

8

8

8

6

6

6

T (˚C) 300

400

500

4

T (˚C) 300

400

500

T (˚C) 300

500

(c)

4

T˚ RAMAN = 460˚C

Results from compositional maps

400

500 T˚ RAMAN = 460˚C

P (kbar)

(b)

4

T (˚C) 300

400

500

Fig. 9. Comparison of the thermobarometric estimates obtained by compositional mapping and conventional spot analyses for sample m30.Bb2. (a) Microphotograph showing the shear band where the two compositional maps were obtained (cross-polar light). (b) P–T estimates obtained with the families selected from the compositional map (see Fig. 8 and text). (c) P–T estimates obtained with conventional spot analyses (see text for details). Key to symbols: stars: Ctd-Chl-Phg; squares: Ctd-Chl; dots: ChlPhg; triangles: Chl-Ab-Prg. Calculations were performed with two distinct chlorite solution models (Vidal et al., 2001, 2005): VP209 (filled symbols) or V2001 (open symbols). Calculations were either performed with Fe-Amesite (black symbols), or without (grey symbols).

 2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 673

HULW UNIT

10

12

10

m29.C2 14

? 12

8

8

8

6

6

6

6

4

4

4

4

300

14

12

400

m55.B2 937f

500

?

10

200

300

400

500

m45.A2 m82.A4

300

400

500

200

m77.B3

400

500

12

12

12

10

10

10

8

8

AS SHEIKH UNIT

14

8

300

m78.B3

14

14

8

200

?

10

8

200

Pressure (kbar)

12

m30.Bb2 14

T˚ RAMAN ~ 460˚C

10

T˚ RAMAN ~ 445˚C

HULW UNIT

Pressure (kbar)

12

14

T˚ RAMAN ~ 460˚C

14

d11.2

T˚ RAMAN ~ 520˚C

T˚ RAMAN ~ 495˚C

d04.B

? 6

6

6

6

4

4

4

4

200

300

400

500

200

Temperature (˚C)

300

400

500

200

Temperature (˚C)

300

400

500

200

Temperature (˚C)

300

400

500

Temperature (˚C)

AS SHEIKH UNIT

Fig. 10. Summary of the P–T estimates for the Hulw and As Sheikh units. Peak temperature obtained by RSCM of the closest sample are also reported. Symbols as for Fig. 9.

validity of our method. Compositional maps, however, yield only a part of the overall P–T path compared with spot analyses. This probably stems from the fact that compositional maps only encompass restricted areas of the thin section, and therefore only a part of the overall mineral variety. On the other hand, the P–T evolution appears to be much better assessed by the continuous, 2D chemical–textural information provided by compositional maps (as shown above for the d11.2 map). Spot analyses complemented by compositional maps on key areas of the thin section thus appears to be a powerful approach. OVERVIEW OF THE P–T RESULTS TWEEQU multi-equilibrium calculations

P–T conditions were estimated as described above for each thin section. Examples of TWEEQU calculations are presented in Table 4 and all the results are summarized in Fig. 10. These P–T estimates and the textural observations on which they rely (e.g. Fig. 8) lead us to propose that the P–T evolution of the Hulw and As Sheikh units is definitively clockwise and follows four major stages:  2007 Blackwell Publishing Ltd

Table 5. Peak temperatures obtained by RSCM. Area

Sample

Diqdah Unit Diqdah Unit Hulw N Hulw NE Hulw NW Hulw W Hulw W Hulw core Hulw core Hulw core Hulw SW Hulw SW Hulw SE UP* UP* UP* UP* UP UP UP UP

d46 d48 m71 m73 m13 d03 d04 m31 d69 d72 d10 d12 m27 d31 d32 d40 d50 d53 d27 d51 d52

GPS coordinates (N .¢.¢¢, E .¢.¢¢) 23.25.22 23.26.13 23.27.48 23.27.37 23.27.56 23.26.00 23.26.30 23.25.47 23.25.01 23.25.35 23.24.27 23.24.45 23.23.44 23.23.33 23.23.34 23.24.21 23.29.22 23.28.26 23.30.39 23.31.13 23.29.39

58.45.47 58.45.23 58.39.36 58.41.06 58.36.02 58.33.30 58.34.23 58.39.57 58.37.49 58.37.56 58.34.39 58.34.53 58.41.39 58.44.28 58.44.09 58.43.33 58.44.22 58.36.39 58.36.46 58.42.51 58.37.20

Sp

Average T (C)

SE

SD

10 10 15 10 12 10 15 13 14 10 14 10 13 10 10 10 13 10 12 11 10

542 585 485 483 434 447 494 457 495 511 522 515 446 495 444 473 452 426 373 363 371

6 7 6 6 3 3 5 5 6 6 7 8 2 3 1 5 8 4 5 2 3

20 23 21 19 9 8 19 17 23 20 27 25 9 10 4 16 27 13 18 8 9

SD: standard deviation; SE: standard error, which is the standard deviation divided by SQRT(n); Sp: number of spectral acquisitions. *Samples belonging to the part of the upper plate located between the As Sheikh and Hulw units (Jebel Abu Daud; Fig. 1). All theses samples contain also quartz, calcite, oxides and in some cases phengite.

674 P. YAMATO ET AL.

58˚35 23˚35

58˚40

58˚45

58˚50

2 km

23˚30

<400˚C

>530˚C 475+/- 30˚C 23˚25

Fig. 11. Map of the peak temperatures calculated by the RSCM method (see text and Table 5; same legend for the metamorphic units as for Fig. 1). The thick dotted line underlines the area of maximum temperatures for the Hulw unit revealing the hotter stretched core of the Hulw mafic megaboudin.

1 A possible equilibration stage at high-pressure conditions and very low temperature. The highest pressure values were obtained (11–12 kbar; m29.C2, m55.B2, m45.A2, m82.A4) for relatively low temperatures (250–300 C) using Chl-Phg equilibria. On the basis of chemical–textural observations, these parageneses correspond to early crystallizations. Two possibilities can then be envisaged. The first one is that the thermodynamic database is not well constrained for lowtemperature phengite, leading to an overestimate of pressure. The second is that these values are true, implying a first pressure decrease at relatively low temperature, from 11–12 kbar, 250–300 C to 7– 9 kbar, 300–350 C. 2 An almost isobaric heating at 8–10 kbar. Indeed, an increase of the temperature from 300 C to 400– 450 C is observed in all samples (except for thin section d04.B) in this pressure range. This temperature increase takes place in the chloritoid stability field, whose overgrowths show increasing XMg values with time. This is the best constrained part of the P–T path with up to six independent reactions (Table 4).

3 An isothermal pressure decrease at 450–500 C: this part of the P–T path is constrained by Chl-Phg or chlorite-albite-paragonite (Chl-Ab-Pg) assemblages. In this part of the metamorphic path, chloritoid no longer appears to re-equilibrate with chlorite and phengite. 4 An almost isobaric cooling at 5–6 kbar. This part of the P–T path is obtained from sample d04.B, located in an intensively strained and highly retrogressed shear zone along Wadi Mayh. RSCM results

The detailed results of the maximum temperatures obtained from the Raman spectra of carbonaceous material are presented in Table 5. Average values for each sample are plotted on a map in Fig. 11. The main results can be summarized as follows: 1 There is a good correlation between this independent method and TWEEQU multi-equilibrium P–T estimates. The difference between the results obtained by these two methods is always lower than the error bars for each method (i.e. 30–50 C; Fig. 10).  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 675

5 No sharp decrease of maximum temperatures is observed across the UP-LP discontinuity between the Hulw and the Mayh units (434 C v. 426 C across the HSZ; Fig. 11). Similarly, no sudden decrease of maximum temperatures seems to occur between the eastern part of the Hulw unit and the NS-trending UP klippen of Jebel Abu Daud (Figs 1 & 11; where relatively high values of peak temperatures are even obtained; e.g. 495 C), which separates the Hulw and As Sheikh units. 6 Somewhat higher temperatures are found in the core and to the SW of the Hulw unit (480–520 C; Fig. 11, thick dotted line), approximately along the sheath fold axis of the central metabasite refolded boudin. Given the antiformal structure of the Saih Hatat, this trend thus consistently coincides with the location of the deepest parts of the Hulw unit.

24

LAS ?

18

?

16

16

14

1

?

?

10

8

8

6

6

4

4

2

2

66-70 Ma

HULWAS SHEIKH

75-85 Ma

45-48 Ma

T (˚C) 300

400

500

600

0 0

100

200

300

400

(Diqdah)

R-Q 10

0 200

-1 km . 5˚C

D 12

Tmax

12

UAS

(Hulw)

14

?

Tmax

Tmax lower value

Tmax higher value

20

Average Tmax

P (kbar)

(a)

10

˚C

.km

(b) 22

-1

5˚C .km 1

P (kbar)

2 RSCM results indirectly confirm the amount of Fe3+ present in chlorite. Because the temperatures of chlorite formation would be higher without adding Fe3+, temperatures obtained with TWEEQU multi-equilibrium thermobarometry would otherwise be higher than the peak temperatures obtained by the RSCM method. 3 The Diqdah Unit suffered higher peak temperatures (>530 C). The two analysed samples (d46 & d48) yield maximum temperatures of 542 and 585 C, respectively. These values are close to those documented for As Sifah (Fig. 2). 4 Low peak temperatures (<400 C) are exclusively found to the north, in the Bandar Kayhran unit. Maximum temperatures decrease slightly towards the north (across the AKSZ, AWSZ; Fig. 1).

500

T (˚C) 600

Fig. 12. P–T evolution for the metamorphic units of the Saih Hatat window. (a) Comparison between P–T estimates of the Hulw unit (plain grey) and those of the As Sheikh unit (striped grey). The thick black line corresponds to the simplified P–T path for the Hulw-As Sheikh unit. Average values of the RSCM results are also shown. (b) Position of this new P–T path (thick black line) within the other metamorphic units of the Saih Hatat window. Same unit abbreviations as for Fig. 2. RSCM results are shown as thick, dark grey vertical segments. The most likely age constraints are also shown (see Discussion for details).  2007 Blackwell Publishing Ltd

676 P. YAMATO ET AL.

DISCUSSION Interpretation of the P–T results New constraints for the Hulw–As Sheikh unit

TheP–T conditions based on complementary and concordant estimates from compositional maps and spot analyses (Fig. 10) reveal that the P–T evolution of the Hulw unit starts with a first equilibration stage along a cold metamorphic gradient of 7 C km)1, at 11–12 kbar, 250–300 C. After a minor pressure drop to 7–9 kbar, 300–350 C, isobaric heating to 450 C ensued. The Hulw unit is clearly clockwise and follows the simplified path shown on Fig. 12a. The pressure decrease (for temperatures lower than 350 C) at the end of the P–T path is constrained by the presence of pyrophyllite (±kaolinite) in shear bands (in samples just north of m73B2, Fig. 1) and by the stability of aragonite (Jolivet et al., 1998). This P–T path fits well the maximum temperature obtained by RSCM (Figs 10 & 11). ÔPeakÕ conditions at 9–10 kbar/450 C are consistent with the absence of garnet and with the lack of jadeite in quartzofeldspathic micaschists. Figure 12a also demonstrates that the new P–T estimates for the As Sheikh unit (striped patterns) are identical to those of the Hulw unit. This result suggests that the As Sheikh and Hulw units share a common P–T history, both of them extending continuously below the upper plate stretch of Fig. 1. Comparison with previous estimates

This P–T path is compared with available P–T constraints for the other surrounding metamorphic units in Fig. 12b. A first observation is that this new, P–T path crosses the range of the earlier, rather scattered estimates for the Hulw unit (thin lines in Fig. 12b). Our study demonstrates that the peak metamorphic pressure for the Hulw unit was equal to that of the upper plate units. Moreover, the maximum temperatures estimated for the Hulw–As Sheikh unit are consistent with the estimates of El-Shazly (2001) obtained from the study of the interlayered, cofacial metabasites. Hence, contrary to earlier reports (Goffe´ et al., 1988; El-Shazly et al., 1990; Jolivet et al., 1998), no significant pressure gap exists across the UP-LP discontinuity, at least to the west and north of the Hulw unit (note that it is not necessarily the case, however, for the eastern part where low-P rocks from the RuwiQuryat unit are juxtaposed against the As Sheikh– Diqdah–As Sifah units). On the other hand, our P–T estimates indirectly confirm the significant pressure gap between the As Sifah and Hulw–As Sheikh units (10 kbar). It is also noted that isobaric cooling is in good agreement with the suggested retrograde paths for the lower plate units (e.g. El-Shazly et al., 1997; El-Shazly & Sisson, 2004).

We note that the early part of the P–T path showing high-pressure conditions at very low temperature (11–12 kbar, 250–300 C; Fig. 12a) is consistent or not with the metamorphic gradient for the deeper As Sifah units, depending on whether the latter is deduced from the high or low Pmax estimates, respectively (Fig. 12b). Such cold, prograde P–T conditions have already been reported from a number of convergent settings (e.g. Ernst, 1993; Goffe´ et al., 1994; Okay, 2002; Agard et al., 2005; Tsujimori et al., 2006). Maximum (RSCM) temperatures lower than 400 C at the north of the upper plate (Fig. 11) are consistent with the occurrence of carpholite-pyrophyllite (±chloritoid) assemblages in the BandarKhuryan and Mayh units (Goffe´ et al., 1988). The other RSCM results of this study, which point to higher temperatures for Diqdah than for the Hulw-As Sheikh units, strengthen the view that the Diqdah and As Sheikh units have a distinct P–T evolution and a different P–T path (Fig. 12b). The Diqdah unit thus probably corresponds, together with the upper As Sifah unit, to a transitional unit between the Hulw– As Sheikh units and the lower As Sifah unit (for higher P). Implications for the structure of the nappe stack Three distinct metamorphic unit groups?

Combining constraints from the literature and the P–T estimates from Fig. 12b, three main metamorphic unit groups can be inferred for the Saih Hatat window: from top to bottom, one finds the Ruwi–Quryat major unit (RQ, hereafter), the Hulw (+As Sheikh) unit and the As Sifah (+Diqdah) unit. The contrast between the first two concerns the deformation style, whereas the constrast between the latter two is essentially a matter of pressure (depth). Looking at the transition between the RQ and Hulw units, no sharp RSCM temperature gap is observed across the UP-LP discontinuity, either to the north of Hulw (426 C v. 434 C; Fig. 11) or along the stretch of UP to the east (473 C v. 475–480 C; Fig. 11). A slight RSCM temperature gradient is suspected towards the north, near the Al Wudya shear zone (AWSZ; Fig. 11; Searle et al., 1994), but the number of RSCM temperatures in this area is yet insufficient to arrive at a conclusion. The lack of a sharp temperature or pressure gap suggests that the vertical displacement across the UP-LP is relatively minor and/or that maximum temperatures for the lowest UP units are greater than expected (Goffe´ et al., 1988). Another interpretation could be that the UP-LP discontinuity is a diffuse strain zone corresponding more or less to the Mayh unit (Fig. 1). In this line of thought, the Mayh unit would constitute a transitional domain, such as the one suggested by Jolivet et al. (1998) to the SE of the Hulw unit (Fig. 1), between the Hulw and BandarAl Khuryan units.  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 677

Fig. 13. (left) Tentative geodynamic model evolution for the metamorphic units of the Saih Hatat window. See Discussion for details. (right) Hypothetical time evolution (and stacking chronology) of the three main units along their P–T–t paths, inferred from their respective structural position, available radiometric data and the respective shape of the P–T paths.

 2007 Blackwell Publishing Ltd

678 P. YAMATO ET AL.

Finite deformation is dominated by penetrative shear and boudinage below the UP-LP discontinuity (Figs 5 & 6) and increases from the Hulw to the As Sifah unit. Approximately 10 kbar (30 km) separates the deepest As Sifah unit (LAS) from the Hulw unit, implying the existence of a major strain zone between them. However, unlike earlier claims (Miller et al., 2002), P–T estimates and RSCM temperatures show that it does not occur below the stretch of upper plate to the east of Hulw unit. Given the deformation gradient, we propose that the Diqdah unit, instead, represents the transition zone between the As Sifah and the overlying Hulw–As Sheikh units. Stacking chronology

Based on their respective structural position, on available radiometric data suggesting that the RQ units were metamorphosed after the Hulw–As Sifah units (Fig. 3), and the shape of their P–T paths (Fig. 12) we tentatively propose a simple chronology for the stacking of these units (Fig. 13). As Hulw and As Sifah enter subduction, only As Sifah is dragged to depths greater than 30–35 km (Fig. 13b). While As Sifah suffers extreme deformation and boudinage along a cold P–T gradient in the eclogite facies, the Hulw unit is detached from the sinking plate at intermediate depths and starts reequilibrating towards higher temperatures (Fig. 13c). This would explain the isobaric heating part of the P–T path documented by our P–T estimates for the Hulw unit (stage 2). The As Sifah unit is then forced back to intermediate depths, close to the Hulw unit (Fig. 13d), and deformation concentrates in the UAS and Diqdah units. The final, pronounced cooling inferred for both AS and Hulw units (Fig. 13e, stage 4 of the Hulw unit P–T path) could result from the juxtaposition with the latest subducted, cold RQ units. Further radiometric constraints are needed to confirm this interpretation. Constraints for obduction and exhumation processes A possible geodynamic frame

An interpretative geodynamic model is proposed in Fig. 13a, from the initial subduction of the Saih Hatat units to the present-day nappe stack. It is implicit that the HP-LT metamorphism and burial of these units took place as the Arabian passive margin was flexured and dragged to the north, below the obduction overthrust, and that As Sifah units represented the distal part of the passive margin (e.g. Oberha¨nsli et al., 1999; Searle et al., 2004). A word of caution and justification is necessary, as any geodynamic interpretation, particularly in Oman, is controversial and fraught with oversimplifications (see Geological Setting, Gray et al., 2005a; Searle

et al., 2005). One of the important conclusions from this study is that the same low P–T gradient prevailed for all Saih Hatat units (7 C km)1; Fig. 12b). This result supports the view of a single, obduction-related subduction process, rather than the existence of an early, independent subduction zone responsible for the As Sifah HP-LT metamorphism, distinct from the one affecting the RQ units (Gregory et al., 1998; Gray et al., 2004b). Besides, no suture zone exists between the As Sifah and Hulw (as recalled by Searle et al., 2004) and the tectonic patterns in both units are strikingly similar (see above). We also note that an alternative scenario in which all lower plate rocks (As Sifah + Hulw) would have been dragged into an early, independent S-vergent subduction zone, would equally fail to explain why the metamorphism of Hulw is lower than that of As Sifah. In our interpretation (Fig. 13), the competition between volume forces generated by the presence of the light continental crust and the slab-pull exerted by the subduction led to an intense stretching at depth within the units. The fact that the As Sifah units correspond to the thinnest (hence less buoyant) parts of the margin could explain why these units reached greater depths (70–80 km) than more the proximal Hulw unit. Continental subduction is known to be physically difficult and transient (from 70 to 85 Ma for Oman), particularly when the continent is attached to a short oceanic slab, as was probably the case (as intra-oceanic obduction initiated close to the Arabian margin; Searle & Cox, 1999). The transient subduction of the Arabian margin and/or the existence of instabilities at depth (e.g. buoyancy, slab-breakoff) forced the As Sifah unit back to intermediate depths (20–25 km), where it was juxtaposed to the Hulw unit (deformation step 1). Age constraints point to a relatively fast, early cooling of the As Sifah unit to such depths (30 C km)1, Saddiqi et al., 2006). The Ruwi Quryat units entered last the subduction wedge (Breton et al., 2004 and references therein; Fig. 13). In this interpretation, the cooling of the As Sifah and Hulw units, by means of the cold RQ units, would have taken place towards the latest stages of effective continental subduction. This would have been more or less coeval with the major deformation stage affecting both the UP and the LP (step 2). Compression v. extension?

Despite the fact that obduction and subduction required major contractional movements, regional-scale, net extension was documented as early as the Campanian–Maastrichtien (Mann et al., 1990; Fournier et al., 2006). The respective contributions of compression and extension to the overall exhumation are unclear: Michard et al. (1994) suggested the existence of late, truly extensional movements, whereas Searle et al. (2004) argued that all the exhumation proceeds  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 679

within compressional boundary conditions (restricting movements along major contacts to relative extensional shear). The final ductile exhumation steps are realized through a penetrative, strongly non-coaxial N030 shear, marked by spectacular sheath folds, clear lineations and shear senses affecting all units (e.g. Jolivet et al., 1998) and fold-nappe development at 75 Ma (see Gray et al., 2005b and references therein). This stage necessarily postdated the syn-convergence juxtaposition of the Hulw and As Sifah units (Fig. 13; Searle et al., 2004), in support of the two-stage exhumation scenario proposed by Michard et al. (1994) or Miller et al. (1999). What remains unresolved is whether contractional boundary conditions prevailed continuously throughout all this second exhumation stage. Given the overall structure of the stacking (e.g. high-P underly low-P units, both across the UP-LP and within the LP), the extent of extensional structures and the development of Ôa-typeÕ folds diagnostic of post-orogenic extension (Jolivet et al., 2004), future work should aim at determining if part of the N030-directed exhumation stage occurred under truly extensional boundary conditions or not. CONCLUSIONS

This petrological study provides a high precision P–T path for the Hulw and As Sheikh units, and demonstrates that they shared a common P–T history (Fig. 12). P–T estimates based on compositional mapping and spot analyses are shown to be complementary and prove to be a powerful tool to decipher the P–T history of metapelitic rocks. In addition, Raman spectroscopy of carbonaceous material yields remarkably consistent peak temperatures. Combining earlier P–T estimates from adjacent units with this new set of P–T data and complementary tectonic observations, we were able to reconstruct the nappe stacking of the Oman HP-LT rocks. Three main metamorphic unit groups can be inferred for the Saih Hatat window: from top to bottom, one finds the Ruwi-Quryat, the Hulw (+As Sheikh) and the As Sifah (+Diqdah) units. Our results strengthen the view that the tectonic and metamorphic data are conveniently accounted for by a simple, N-vergent continental subduction of the passive Arabian margin below the obduction nappe along a cold P–T gradient (7 C km)1). The early subduction of the thinned margin, accompanied by intense stretching and boudinage, allowed for the deep subduction (70 km) of part of the margin (e.g. the As Sifah unit), while the rest reached only intermediate depths (30–35 km) and started to re-equilibrate thermally (e.g. the Hulw unit). All these (lower plate) units were then juxtaposed with the cold, later subducted (Ruwi-Quryat) units from the upper plate. This nappe stack was then strongly deformed and exhumed  2007 Blackwell Publishing Ltd

in a highly non-coaxial, NNE-vergent, and possibly extensional tectonic regime. Our model accounts for the new metamorphic data obtained for the Hulw and As Sheikh units and integrates most earlier observations (e.g. Goffe´ et al., 1988; Michard et al., 1994; Searle et al., 1994, 2004; Jolivet et al., 1998; Miller et al., 1998, 2002; El-Shazly et al., 2001; Gray & Gregory, 2003; Gray et al., 2004a, 2005a,b and others). In our view, this model also partly reconciles the models of Goffe´ et al. (1988), Searle et al. (2004) and Saddiqi et al. (2006) with the recent mapping observations, by Gray and co-workers, documenting a major UP-LP discontinuity. ACKNOWLEDGEMENTS

We thank Dr H. al-Azri and J.-P. Breton for their assistance during our field work in Oman. The help from R. Caron, for his nice thin-sections, M. Fialin and F. Couffignal, during microprobe acquisitions and O. Beyssac, for RSCM spectroscopy, was most appreciated. We also thank B. Hacker and A.K. ElShazly for their reviews and corrections and D. Robinson for his proof-reading. REFERENCES Agard, P., Labrousse, L., Elvevold, S. & Lepvrier, C., 2005. Discovery of Paleozoic Fe-Mg carpholite in Motalafjella, Svalbard Caledonides: a milestone for subduction-zone gradients. Geology, 33, 761–764. Agard, P., Jolivet, L., Vrielinck, B., Burov, E. & Monie´, P., 2007. Plate acceleration: the obduction trigger? Earth and Planetary Science Letters, 258, 428–441. Arnaud, N. O. & Kelley, S. P., 1995. Evidence for excess argon during high pressure metamorphism in the Dora Maira Massif (western Alps, Italy), using an ultra-violet laser ablation microprobe 40Ar-39Ar technique. Contributions to Mineralogy and Petrology, 121, 1–11. Berman, R. G., 1991. Thermobarometry using multi-equilibrium calculations: a new technique, with petrological applications. The Canadian Mineralogist, 29, 833–855. Beyssac, O., Goffe´, B., Chopin, C. & Rouzaud, J. N., 2002. Raman spectra of carbonaceous material in metasediments: a new geothermometer. Journal of Metamorphic Geology, 20, 859–871. Beyssac, O., Goffe´, B., Petitet, J. P., Froigneux, E., Moreau, M. & Rouzaud, J. N., 2003. On the characterization of disordered and heterogeneous carbonaceous materials using Raman spectroscopy. Spectrochimica Acta, 59, 2267–2276. Beyssac, O., Bollinger, L., Avouac, J. P. & Goffe´, B., 2004. Thermal metamorphism in the lesser Himalaya of Nepal determined from Raman spectroscopy of carbonaceous material. Earth and Planetary Science Letters, 225, 233–241. 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. Breton, J. P., Be´chennec, F., Le Me´tour, J., Moen-Maurel, L. & Razin, P., 2004. Eoalpine (Cretaceous) evolution of the Oman Tethyan continental margin: insights from a structural field study in Jabal Akhdar (Oman Mountains). GeoArabia, 9, 1–18.

680 P. YAMATO ET AL.

Coleman, R. G., 1981. Tectonic setting for ophiolite obduction in Oman. Journal of Geophysical Research, 86, 2497–2508. De Andrade, V., Vidal, O., Lewin, E., O’Brien, P. & Agard, P., 2006. Quantification of electron microprobe compositional maps of rock thin sections: an optimized method examples. Journal of Metamorphic Geology, 24, 655–668. El-Shazly, A. K., 1994. Petrology of lawsonite-, pumpellyite- and sodic amphibole-bearing metabasites from north-east Oman. Journal of Metamorphic Geology, 12, 23–48. El-Shazly, A. K., 1995. Petrology of Fe-Mg-carpholite-bearing metasediments from NE Oman. Journal of Metamorphic Geology, 13, 379–396. El-Shazly, A. K., 2001. Are pressures for blueschists and eclogites overestimated? The case from NE Oman. Lithos, 56, 231– 264. El-Shazly, A. K. & Lanphere, M. A., 1992. Two high pressure metamorphic events in NE Oman: evidence from 40Ar/39Ar dating and petrological data. Journal of Geology, 100, 731– 751. El-Shazly, A. K. & Sisson, V. B., 2004. Fluid inclusions in carpholite-bearing metasediments and blueschists from NE Oman: constraints on P-T evolution. European Journal of Mineralogy, 16, 221–233. El-Shazly, A. K., Coleman, R. G. & Liou, J. G., 1990. Eclogites and blueschists from northeastern Oman: petrology and PT evolution. Journal of Petrology, 31, 629–666. El-Shazly, A. K., Worthing, M. A. & Liou, J. G., 1997. Interlayered eclogites, blueschists and epidote amphibolites from NE Oman: a record of protolith compositional control and limited fluid infiltration. Journal of Petrology, 38, 1461–1487. El-Shazly, A. K., Bro¨cker, M., Hacker, B. & Calvert, A., 2001. Formation and exhumation of blueschists and eclogites from NE Oman: new perspecives from Rb-Sr and 40Ar/39Ar dating. Journal of Metamorphic Geology, 19, 233–248. Ernst, W. G., 1993. Metamorphism of Franciscan tectonostratigraphic assemblage, Pacheco Pass area, east-central Diablo Range, California Coast Ranges. Geological Society of America Bulletin, 105, 618–636. Fournier, M., Lepvrier, C., Razin, P. & Jolivet, L., 2006. Postobduction extension in the Oman Mountains and subsequent compression. GeoArabia, 4, 17–40. Glennie, K. W., Boeuf, M. G., Hughes-Clarke, M. H. W., Moody-Stuart, M., Pilaar, W. F. & Reinhardt, B. M., 1974. Geology of the Oman Mountains. Verhandelingen van het Koninklijk Nederlands Geologischs Mijnbouwkundig Genootschap, 31, 1–423. Goffe´, B., Michard, A., Kienast, J. R. & Le Mer, O., 1988. A case of obduction-related high-pressure, low-temperature metamorphism in upper crustal nappes, Arabian continental margin, Oman. Tectonophysics, 151, 363–386. Goffe´, B., Baronnet, A. & Morin, G., 1994. La saliotite, interstratifie´ re´gulier 1:1 cooke´ite/paragonite: nouveau phyllosilicate du me´tamorphisme de haute pression et basse tempe´rature. European Journal of Mineralogy, 6, 897– 911. Gray, D. R. & Gregory, R. T., 2003. Ophiolite obduction and the Samail Ophiolite: the behaviour of the underlying margin. In: Ophiolites and Earth History, Special Publication 218, (eds Dilek, Y. & Robinson, P. T.), pp. 449–466. Geological Society, London. Gray, D. R., Gregory, R. T. & Miller, J. M., 2000. A new structural profile along the Muscat–Ibra transect, Oman: implications for the emplacement of the Samail ophiolite. In: Ophiolites and Oceanic Crust. New Insights From Field Studies and the Ocean Drilling Program, Special Paper 349, (eds Dilek, Y., Moores, E. M., Elthon, D. & Nicolas, A.), pp. 513– 523. Geological Society of America. Gray, D. R., Miller, J. M., Foster, D. A. & Gregory, R. T., 2004a. Transition from subduction- to exhumation-related fabrics in glaucophane-bearing eclogites, Oman: evidence from relative fabric chronology and 40Ar/39Ar ages. Tectonophysics, 389, 35–64.

Gray, D. R., Hand, M., Mawby, J., Armstrong, R. A., Miller, J. M. & Gregory, R. T., 2004b. Sm-Nd and Zircon U-Pb ages from garnet-bearing eclogites, NE Oman: constraints on High-P metamorphism. Earth and Planetary Sciences Letters, 222, 407–422. Gray, D. R., Gregory, R. T. & Miller, J. M., 2005a. Comment on ÔStructural evolution, metamorphism and restoration of the Arabian continental margin, Saih Hatat region, Oman MountainsÕ. Journal of Structural Geology, 27, 371–374. Gray, D. R., Miller, J. M. & Gregory, R. T., 2005b. Strain and kinematic evolution of a fold-nappe beneath the Samail Ophiolite, Oman. Journal of Structural Geology, 27, 1986– 2007. Gregory, R. T., Gray, D. R. & Miller, J. M., 1998. Tectonics of the Arabian margin associated with the formation and exhumation of high-pressure rocks, Sultanate of Oman. Tectonics, 17, 657–670. Hacker, B., 1994. Rapid emplacement of young oceanic lithosphere: argon geochronology of the Oman ophiolite. Science, 265, 1563–1565. Hacker, B. R., Mosenfelder, J. L. & Gnos, E., 1996. Rapid emplacement of the Oman ophiolite: Thermal and geochronologic constraints. Tectonics, 15, 1230–1247. Hanna, S. S., 1990. The Alpine deformation of the central Oman Mountains. In: The Geology and Tectonics of the Oman Region, Special Publication, 49 (eds Robertson, A. F. H., Searle, M. P. & Ries, A. C.), pp. 341–359. Geological Society, London. Jolivet, L., Goffe´, B., Bousquet, R., Oberha¨nsli, R. & Michard, A., 1998. Detachements in high pressure mountains belts, Tethyan examples. Earth and Planetary Science Letters, 160, 31–47. Jolivet, L., Famin, V., Mehl, C. et al., 2004. Strain localization during crustal-scale boudinage to form extensional metamorphic domes in the Aegean Sea. Geological Society of America Special Paper, 380, 185–210. Le Me´tour, J., Villey, M. & de Gramont, X., 1986a. Geological Map of Quryat, Sheet NF 40-4D. Ministry of Petroleum and Minerals, Sultanate of Oman. Le Me´tour, J., Villey, M. & de Gramont, X., 1986b. Geological Map of Masqat, Sheet NF 40-4A. Ministry of Petroleum and Minerals, Sultanate of Oman. Lippard, S. J., 1983. Cretaceous high pressure metamorphism in NE Oman and its relationship to subduction and ophiolite nappe emplacement. Journal of Geological Society of London, 140, 97–104. Mann, A., Hanna, S. S. & Nolan, S. C., 1990. The post-Campanian tectonic evolution of the central Oman mountains: tertiary extension of the eastern Arabian margin. In: The Geology and Tectonics of the Oman Region, Special Publication, 49 (eds Robertson, A. F. H., Searle, M. P. & Ries, A. C.), pp. 549–564. Geological Society, London. Michard, A., Goffe´, B. & Ouazzani-Touhami, M., 1983. Obduction related high pressure, low temperature metamorphism in upper crustal materials, Muscat, Oman. Terra Cognita, 3, 22–23. Michard, A., Goffe´, B., Saddiqi, O., Oberha¨nsli, R. & Wendt, A. S., 1994. Late Cretaceous exhumation of the Oman blueschists and eclogites: a two-stage extensional mechanism. Terra Nova, 6, 404–413. Miller, J. M., Gray, D. R. & Gregory, R. T., 1998. Exhumation of high-pressure rocks in northeastern Oman. Geology, 26, 235–238. Miller, J. M., Gregory, R. T., Gray, D. R. & Foster, D. A., 1999. Geological and geochronological constraints on the exhumation of a high-pressure metamorphic terrane, Oman. In: Exhumation Processes: Normal Faulting, Ductile Flow and Erosion, Special Publication 154 (eds Ring, U., Brandon, M.T., Lister, G.S. & Willett, S.D.), pp. 241–260, Geological Society, London. Miller, J. M., Gray, D. R. & Gregory, R. T., 2002. Geometry and significance of internal windows and regional isoclinal  2007 Blackwell Publishing Ltd

NEW P–T ESTIMATES FOR OMAN BLUESCHISTS 681

folds in northeast Saih Hatat, Sultanate of Oman. Journal of Structural Geology, 24, 359–386. Montigny, R., Le Mer, O., Thuizat, R. & Whitechurch, H., 1988. K-Ar and 40Ar/39Ar study of metamorphic rocks associated with the Oman ophiolite: tectonic implications. Tectonophysics, 151, 345–362. Mun˜oz, M., De Andrade, V., Vidal, O., Lewin, E., Pascarelli, S. & Susini, J., 2006. Redox and speciation micro-mapping using dispersive X-ray absorption spectroscopy: application to iron in chlorite mineral of a metamorphic rock thin section. Geochemistry Geophysics Geosystems, 7, Q11020, doi: 10.1029/ 2006GC001381. in press. Nicolas, A., 1989. Structures in Ophiolites and Dynamics of Oceanic Lithosphere (Petrology and Structural Geology). Kluwer, Dordrecht. Oberha¨nsli, R., Wendt, A. S. & Goffe´, B., 1999. Detrital chromites in metasediments of the East-Arabian continental margin in the saih Hatat area: constraints for the palaeogeographic setting of the Hawasina and semail basins (Oman mountains). International Journal of Earth Sciences, 88, 13–25. Okay, A. I., 2002. Jadeite-chloritoid-glaucophane-lawsonite blueschists in north-west Turkey: unusually high P/T ratios in continental crust. Journal of Metamorphic Geology, 20, 757– 768. Parra, T., Vidal, O. & Agard, P., 2002a. A thermodynamic model for Fe-Mg dioctahedral K white micas using data from phaseequilibrium experiments and natural pelitic assemblages. Contributions to Mineralogy and Petrology, 143, 706–732. Parra, T., Vidal, O. & Jolivet, L., 2002b. Relation between deformation and retrogression in blueschist metapelites of Tinos island (Greece) evidenced by chlorite-mica local equilibria. Lithos, 63, 41–66. Parra, T., Vidal, O. & Theye, T., 2005. Experimental data on the Tschermak substitution in Fe-chlorite. American Mineralogist, 90, 359–370. Poupeau, G., Saddiqi, O., Michard, A., Goffe´, B. & Oberha¨nsli, R., 1998. Late thermal evolution of the Oman Mountains subophiolitic windows: apatite fission track thermochronology. Geology, 26, 1139–1142. Rimmele´, G., Parra, T., Goffe´, B., Oberha¨nsli, R., Jolivet, L. & Candan, O., 2004. Exhumation paths of high-pressure-lowtemperature metamorphic rocks from the Lycian Nappes and Menderes Massif (SW Turkey): a multi-equilibrium approach. Journal of Petrology, 46, 641–669. Saddiqi, O., Poupeau, G., Michard, A., Goffe´, B. & Oberha¨nsli, R., 1995. Exhumation des roches metamorphiques HP-LT d’Oman: Datation par traces de fission sur zircons. Comptes Rendus de l’Acade´mie des Sciences de Paris, 320, 1071–1077. Saddiqi, O., Michard, A., Goffe´, B., Poupeau, G. & Oberha¨nsli, R., 2006. Fission-track thermochronology of the Oman Mountains continental windows, and current problems of tectonic interpretation. Bulletin of the French Geological Society, 177, 127–134. Scott, R. W., 1990. Chronostratigraphy of the Cretaceous carbonate shelf, southeastern Arabia. In: The Geology and Tectonics of the Oman Region, Special Publication 49 (eds Robertson, A. F. H., Searle, M. P. & Ries, A. C.), pp. 89–108. Geological Society, London. Searle, M. P. & Cox, J., 1999. Tectonic setting, origin and obduction of the Oman ophiolite. Geological Society of America Bulletin, 111, 104–122. Searle, M. P. & Cox, J. S., 2002. Subduction zone metamorphism during formation and emplacement of the Semail Ophiolite in the Oman mountains. Geological Magazine, 139, 241–255. Searle, M. P., Waters, D. J., Martin, H. N. & Rex, D. C., 1994. Structure and metamorphism of blueschist-eclogite facies rocks from the northeastern Oman Mountains. Journal of the Geological Society, London, 151, 555–576.

 2007 Blackwell Publishing Ltd

Searle, M. P., Warren, C. J., Waters, D. J. & Parrish, R. R., 2004. Structural evolution, metamorphism and restoration of the Arabian continental margin, Saih Hatat region, Oman Mountains. Journal of Structural Geology, 26, 451–473. Searle, M. P., Warren, C. J., Waters, D. J. & Parrish, R. R., 2005. Reply to: Comment by Gray, Gregory and Miller on ÔStructural evolution, metamorphism and restoration of the Arabian continental margin, Saih Hatat region, Oman MountainsÕ. Journal of Structural Geology, 27, 375–377. Sherlock, S. C. & Kelley, S. P., 2002. Excess argon in HP-LT rocks: a UV laserprobe study of phengite and K-free minerals. Chemical Geology, 182, 619–636. Tagami, T., Galbraith, R. F., Yamada, R. & Laslett, G. M., 1998. Revised annealing kinetics of fission-tracks in zircon and geological implications. In: Advances in Fission-Track Geochronology (eds Van den Haute, P. & De Corte, F.), pp. 99– 112. Kluwer Academic Publishers, Dordrecht. Tilton, G. R., Hopson, C. A. & Wright, J. E., 1981. Uraniumlead isotopic ages of the Semail ophiolite, Oman, with applications to Tethyan Ridge Tectonics. Journal of Geophysical Research, 86, 2763–2775. Trotet, F., Vidal, O. & Jolivet, L., 2001. Exhumation of Syros and Sifnos metamorphic rocks (Cyclades, Greece):new constraints on the P-T paths. European Journal of Mineralogy, 13, 901–920. Tsujimori, T., Sisson, V. B., Liou, J. G., Harlow, G. E. & Sorensen, S. S., 2006. Very-low-temperature record of the subduction process: a review of worldwide lawsonite eclogites. Lithos, 92, 609–624. Vidal, O. & Parra, T., 2000. Exhumation paths of high-pressure metapelites obtained from local equilibria for chlorite-phengite assemblages. Geological Journal, 35, 139–161. Vidal, O., Parra, T. & Trotet, F., 2001. A thermodynamic model for Fe-Mg aluminous chlorite using data from phase equilibrium experiments and natural pelitic assemblages in the 100– 600 C, 1–25 kbar range. American Journal of Science, 6, 557– 592. Vidal, O., Parra, T. & Vieillard, P., 2005. Experimental data on the Tschermak solid solution in Fe-chlorites: application to natural examples and possible role of oxydation. American Mineralogist, 90, 359–370. Vidal, O., De Andrade, V., Lewin, E., Munoz, M., Parra, T. & Pascarelli, S., 2006. P-T-deformation-Fe3+/Fe2+ mapping at the thin section scale and comparison with XANES mapping. Application to a garnet-bearing metapelite from the Sambagawa metamorphic belt. Journal of Metamorphic Geology, 24, 669–683. Wagner, G. & Van den Haute, P., 1992. Fission-Track Dating. Kluwer Academic Publishers, Dordrecht. Warren, C. J., Parrish, R. R., Searle, M. P. & Waters, D. J., 2003. Dating the subduction of the Arabian continental margin beneath the Semail ophiolite, Oman. Geology, 31, 889– 892. Warren, C. J., Parrish, R. R., Waters, D. J. & Searle, M. P., 2005. Dating the geologic history of Oman’s Semail ophiolite: insights from U-Pb geochronology. Contributions to Mineralogy and Petrology, 150, 403–422. Warren, C. J. & Waters, D. J., 2006. Oxidized eclogites and garnet-blueschists from Oman: P–T path modelling in the NCFMASHO system. Journal of Metamorphic Geology, 24, 783–802. Wendt, A. S., D’Arco, P., Goffe´, B. & Oberha¨nsli, R., 1993. Radial cracks around a-quartz inclusions in almandine: constraints on the metamorphic history of the Oman mountains. Earth and Planetary Sciences Letters, 114, 449–461.

Received 30 September 2006; revision accepted 21 May 2007.

682 P. YAMATO ET AL.

APPENDIX

Figure A1. At the top, parageneses and corresponding NIR (number of independent reactions) using the following end-members with TWEEQU 2.02. For chlorite: Sud, sudoite; Dph, daphnite; Fe-Am, ferro-amesite; Mg-Am, magnesio-amesite; Clin, clinochlore. For chloritoid: Fe-Cld, ferro-chloritoid; Mg-Cld, magnesio-chloritoid. For phengite: Al-Cel: alumino-celadonite; Fe-Cel, ferro-celadonite; Ms, muscovite; Prl, pyrophyllite. In grey, theoretically possible equilibria rarely obtained. In black, frequently used assemblages. *Estimates with too many reactions to be calculated with INTERSX. Below, four examples of P–T plots obtained with TWEEQU are shown for sample m30.Bb2.

 2007 Blackwell Publishing Ltd