ROBERT J. STERN , PETER R. JOHNSON , ALFRED ...rjstern/pdfs/ANSOph...aGeosciences Department,...

Precambrian Ophiolites and Related RocksEdited by Timothy M. KuskyDevelopments in Precambrian Geology, Vol. 13 (K.C. Condie, Series Editor) 95© 2004 Elsevier B.V. All rights reserved.

Chapter 3

NEOPROTEROZOIC OPHIOLITES OF THEARABIAN-NUBIAN SHIELD

ROBERT J. STERNa, PETER R. JOHNSONb, ALFRED KRÖNERc ANDBISRAT YIBASd

aGeosciences Department, University of Texas at Dallas, Box 830688,Richardson, TX 57083-0688, USAbSaudi Geological Survey, P.O. Box 54141, Jiddah 21514, Saudi ArabiacInstitut für Geowissenschaften, Universität Mainz, 55099 Mainz, GermanydPulles Howard and de Lange, Environmental and Water Quality Management,P.O. Box 861, Auckland Park 2006, South Africa

Ophiolites of mid-Neoproterozoic age are abundant in the Arabian-Nubian Shield(ANS) of NE Africa and Arabia. ANS ophiolites range in age from 690 to 890 Ma and littera region that is 3000 km N-S and> 1000 km E-W. In the northern ANS, ophiolites occuras nappe complexes marking suture zones between terranes. Although dismembered andaltered, all of the diagnostic components ofophiolites can be found: harzburgite, cumulateultramafics, layered as well as higher level gabbro and plagiogranite, sheeted dikes, and pil-lowed basalt. Allochthonous mafic-ultramafic complexes in the southern ANS, in Ethiopiaand Eritrea, are interpreted as ophiolites, but are more deformed and metamorphosed thanthose in the north. Reconstructed ophiolitic successions have crustal thicknesses of 2.5to 5 km.

The ANS ophiolitic mantle was mostly harzburgitic, containing magnesian olivines andspinels that have compositions consistent with extensive melting. Cr# for spinels in ANSharzburgites are mostly> 60, comparable to spinels from modern forearcs and distinctlyhigher than spinels from mid-ocean ridges andbackarc basin peridotites. ANS ophiolitesare often associated with a thick (1–3 km) sequence of cumulate ultramafic rocks, whichdefine a transition zone between seismic and petrologic Mohos. These cumulates are dom-inated by dunite, with subordinate pyroxene-rich lithologies. Cumulate ultramafics tran-sition upwards into layered gabbro. Several crystallization sequences are inferred fromANS transition zones and cumulate gabbro sections. In all samples studied, olivine andspinel crystallized first, followed (in order of decreasing abundance) by cpx-plag, cpx-opx-plag, and opx-cpx-plag. ANSophiolitic lavas mostly define asubalkaline suite char-acterized by low K and moderate Ti contents, that has both tholeiitic and calc-alkalineaffinities and includes a significant, althoughsubordinate, amount of boninites. The lavasare fractionated (mean Mg#= 55) but have higher abundances of Cr (mean= 380 ppm)and Ni (mean= 135 ppm) than would be expected for such a low Mg#. The ANS ophioliticlavas include both LREE-depleted and LREE-enriched varieties, but as a group are slightly

DOI: 10.1016/S0166-2635(04)13003-X

96 Chapter 3: Neoproterozoic Ophiolites of the Arabian-Nubian Shield

LREE-enriched: mean (Ce/Yb)n ∼ 2.2 and Sm/Nd∼ 0.30. On a variety of discriminationdiagrams, the lavas plot in fields for MORB, BABB, arc tholeiite, and boninite. ANS lavascluster around Ti/Zr= 97, indicating that Ti-bearing phases did not precipitate early. Ndisotopic compositions indicate derivation from a long-depleted mantle source, withεNd(t)∼ +5 to +8. Mineral and lava compositions are consistent with the hypothesis that mostANS ophiolites formed in ‘suprasubduction zone’ (SSZ) settings, and the high Cr# of ANSophiolitic harzburgites suggests a forearc environment. Geochemical studies of deep watersediments deposited on ANS ophiolites are needed to better characterize and understandthe Neoproterozoic ocean where ANS ophiolites formed.

1. INTRODUCTION

The Arabian-Nubian Shield (ANS) in NE Africa and W. Arabia is the largest tract ofjuvenile continental crust of Neoproterozoic age on Earth (Patchett and Chase, 2002). Thiscrust was generated when smaller terranes of arc and back arc crust were generated withinand around the margins of a large oceanictract known as the Mozambique Ocean, whichformed in association with the breakup of Rodinia∼ 800–900 Ma (Stern, 1994). Oceanicplateaus may also have been accreted (Stein and Goldstein, 1996). These crustal fragmentscollided as the Mozambique Ocean closed, forming arc-arc sutures, composite terranes,the Arabian-Nubian Shield (ANS; Fig. 1), and the larger collisional belt known as the EastAfrican Orogen (Stern, 1994; Kusky et al., 2003). The Arabian-Nubian Shield was caughtbetween fragments of East and West Gondwanaland as these collided at about 600 Ma(Meert, 2003) to form a supercontinent variously referred to as Greater Gondwanaland(Stern, 1994), Pannotia (Dalziel, 1997) or just Gondwanaland. The ANS was subsequentlyburied by Phanerozoic sediments but has been exposed by uplift and erosion on the flanksof the Red Sea in Oligocene and younger times.

Several lines of evidence support the idea that the ANS is juvenile Neoproterozoic crust,including non-radiogenic initial Sr and radiogenic initial Nd isotopic compositions for a

Fig. 1. Location of the Arabian-Nubian Shield and location of ophiolites and related rocks withinit. (A) Political and modern geographic features of the region. B= Bahrain, Dj= Djibouti,Is = Israel, Jrdn= Jordan, K= Kuwait. Location of Figs. 2A and B shown in dashed rectan-gles. (B) Location of Precambrian basement exposures, crustal types, and ophiolites and ophioliticrocks (shown in black). Abbreviations for some of the better studied ophiolites follow. Saudi Ara-bia: H= Halaban, JT= Jebel Tays; JU= Jebel Uwayjah; JE= Jebel Ess; AA= Al ‘Ays (Wask);BT = Bi’r Tuluhah; A = Arjah; DZ = Darb Zubaydah; BU= Bi’r Umq; Th = Thurwah; JN= JebelNabitah; T= Tathlith. Egypt: F= Fawkhir; Br= Barramiya; Gh= Ghadir; AH= Allaqi-Heiani;G = Gerf. Sudan: OSH= Onib-Sol Hamed; Hs= Hamisana; AD= Atmur-Delgo; K = Keraf;R = Rahib; M= Meritri; Os = Oshib; Kb= Kabus (Nuba Mts); I= Ingessana; Kk= Kurmuk.Eritrea: Hg= Hagar Terrane. Ethiopia: Zg= Zager Belt; DT= Daro Tekli Belt; Bd= Baruda;TD = Tulu Dimtu; Y = Yubdo; A= Adola; My = Moyale. Kenya: S= Sekerr, B= Baragoi. TheBi’r Umq-Nakasib suture is defined by the Bi’r Umq-Thurwah-Meritri-Oshib ophiolite belt.

1. Introduction 97

(A)

(B)


wide range of igneous rocks. The ANS can be isotopically defined as the region in NEAfrica and Arabia where Nd-model ages approximate crystallization ages (Stern, 2002).These indications that abundant juvenile continental crust and mantle lithosphere weregenerated during Neoproterozoic time in the region has been confirmed by Nd and Srisotopic studies of samples of mafic lower crust and mantle lithosphere brought up asxenoliths in Tertiary lavas from Saudi Arabia, which also indicate that the lower crust andlithospheric mantle of the region formed during Neoproterozoic time (Henjes-Kunst et al.,1994; McGuire and Stern, 1993).

Ophiolites and ophiolitic rocks are remarkably abundant in the Arabian-Nubian Shield.They are scattered across most of the ANS, over a distance of∼ 3000 km from thefarthest north (Jebel Ess) almost to the equator, and from Rahib in the west to JabalUwayjah (45◦E) in the east, encompassing an area of about two million square kilome-ters (Fig. 1). Ophiolites are particularly well studied in Arabia (see companion paper byJohnson et al. (2004). If ophiolites are the remains of oceanic lithosphere, then the ANSis a massive graveyard of Neoproterozoic oceanic lithosphere. The abundance of ophio-lites is a further indication that ANS crust and lithosphere were produced by processessimilar to those of modern plate tectonics. Their abundance has made it difficult to definethe orientation of sutures solely from the distribution of ophioliticrocks (Church, 1988;Stern et al., 1990). This is further complicated by the fact that not all ANS mafic-ultramaficcomplexes formed in a seafloor spreading environment—some appear to be roots of is-land arcs, such as Darb Zubaydah in Arabia (Quick and Bosch, 1989)—and others areautochthonous layered intrusions, such as Dahanib in Egypt (Dixon, 1981). Care must beexerted to avoid misidentification, but even the most conservative estimates indicate thatthere is a remarkable abundance of ophiolites in the ANS.

Ophiolites were first recognized in the region by Rittmann (1958), but were otherwiseignored until the pioneering study of Bakor et al. (1976). This was followed by a flurry offield, petrological, and geochemical studies throughout the 1980’s. This level of activity hasdecreased significantly through the 1990’s and into the 21st century. This review has threeobjectives. First, it is intended to summarize the most important observations of this firstphase of studying ANS ophiolites. Second, because the ophiolites of the Arabian-NubianShield are so common and so well-exposed, it is hoped that this example will provide abasis of what is expected to be preserved when a major episode of juvenile crust formationassociated with modern-style plate tectonics occurs. Finally, it is hoped that this overviewwill stimulate a resurgence of ANS ophiolite studies.

Note that the Arabic word for mountain is variously spelled ‘Jebel’ (Egypt), ‘Gebel’(Sudan), or ‘Jabal’ (Arabia). For simplicity, we use the ‘Jebel’spelling throughout thiscontribution.

2. OUTCROP PATTERNS

We define the ANS as the northern, juvenile part of the Neoproterozoic EAO. Morerestrictive definitions have the shield ending in the south with the southernmost contiguous

3. Crustal Structure 99

outcrops of basement exposed around the Red Sea. The Arabian-Nubian Shield as definedhere includes basement outliers well to the south in Ethiopia and Kenya. For the purposes ofthis presentation, the ANS is that part of the East African Orogen characterized by juvenileNeoproterozoic crust and is also where ophiolites and ophiolitic rocks are encountered(Fig. 1).

The ANS can be usefully subdivided into northern and southern halves. North of theBi’r Umq-Nakasib suture, which extendsNE from the Oshib and Meritri ophiolites inSudan and continues across the Red Sea in Arabia through the Thurwah and Bi’r Umqophiolites, the ophiolite belts trend approximately E-W. It is relatively easy to identifystructures developed during ophiolite obduction in this region. Greenschist-facies meta-morphism is characteristic for these ophiolites, and diagnostic features, such as pillowedbasalts, are well preserved. This is shown in Table 1, which lists ophiolites with a Penrose-type succession; these are from the northern ANS or from the Bi’r Umq-Nakasib suture.Such excellent preservation largely reflects thefact that these northern ophiolites are rela-tively undisturbed by steep N-S structures developed during terminal collision between Eand W Gondwanaland, which are pervasive farther south (Fig. 2A). Many ophiolites in thenorthern ANS are, however, disrupted by NW-trending strike-slip faults and shear zonesof the Najd fault system (Sultan et al., 1988). Remote sensing has proven to be an effec-tive way to map the distribution of spectrally distinct lithologies, especially serpentinitesand amphibole-bearing mafic rocks, providing a quantitative, if indirect, assessment of thedistribution and abundance of disrupted ophiolites in the basement of Egypt (Sultan et al.,1986). A more detailed presentation of the field relations and structure of Arabian Shieldophiolites is presented by Johnson et al. (2004).

It is a significantly greater challenge to identify ophiolites to the south of the Bi’rUmq-Nakasib suture. This area was closer to and thus more intensely affected by the end-Neoproterozoic terminal collision, such that structures related to ophiolite obduction aretransposed or obliterated (Abdelsalam and Stern, 1996). Basement structures dip steeply(Fig. 2B), units are intensely deformed and shuffled by high-angle thrusting and sub-horizontal shearing, and metamorphism is typically amphibolite-facies (Yihunie, 2002).Diagnostic features and emplacement fabrics for units that might originally have been ophi-olites are not common. Purists would hesitate to identify the linear mafic-ultramafic com-plexes of the southern ANS as ophiolites, but theassociation of harzburgitic ultramaficsin association with MORB-like and even boninitic mafic units as well as the regional as-sociation of southern ANS mafic-ultramafic complexes to the abundant and unequivocalophiolites of the northern ANS makes it likely that these mostly represent ophiolites indifferent stages of preservation.

3. CRUSTAL STRUCTURE

The best preserved ophiolites in the northern ANS contain all or most of the com-ponents of complete ‘Penrose’ophiolites (Table 1), including pillowed basalts, gabbros,


Table 1. ANS ophiolites with a “Penrose-type” assemblage

Name Country ReferenceDarb Zubaydah Saudi Arabia Quick (1990)Bi’r Tuluhah Saudi Arabia Pallister et al. (1988)Wadi Khadra Saudi Arabia Quick (1991)Halaban Saudi Arabia Al-Saleh et al. (1998)Ess Saudi Arabia Pallister et al. (1988)Al ‘Ays (Wask) Saudi Arabia Bakor et al. (1976)Tharwah Saudi Arabia Nassief et al. (1984)Bi’r Umq Saudi Arabia Shanti (1983)Tathlith Saudi Arabia Pallister et al. (1988)Oshib-Ariab belt Sudan Abdel-Rahman (1993)Arbaat Sudan Abdelsalam and Stern (1993)Atmur-Delgo Sudan Harms et al. (1994),

Schandelmeier et al. (1994)Sol Hamed Sudan Fitches et al. (1983)Wadi Onib Sudan Hussein et al. (1983)Gebel Gerf Egypt Zimmer et al. (1995)Wadi Ghadir Egypt El-Bayoumi (1983)Fawkhir Egypt El-Sayed et al. (1999)

and tectonized harzburgites. Several ANS ophiolites have sheeted dike complexes (such asGhadir, Onib, and Ess) but these are not always reported. Even the well-preserved ophio-lites are faulted, folded, and otherwise disrupted, so that reconstructing a complete ophio-lite pseudostratigraphy is difficult and equivocal. Nevertheless, three such reconstructionsof ANS ophiolite crustal structure are shown onFig. 3. These reconstructions differ in therelative abundances of volcanics, pillowed lavas, sheeted dikes, and gabbro but all suggestthat the oceanic crust represented by ANS ophiolites was generally in the range of 2.5to 6 km thick. As discussed in the next section, ANS ophiolites were generated and em-

Fig. 2. Remote sensing images of ANS ophiolites, showing the different outcrop patterns of ophi-olites in the northern (A) and southern (B) parts of the Arabian-Nubian Shield. Locations shownin Fig. 1. (A) Allaqi-Heiani Suture along the Egypt-Sudan border. Image is approximately 300 kmacross and N is towards the top of image. Dashed line approximates trace of Allaqi-Heiani Suture.Note that the general E-W structure of the ophiolite belt, which formed during suturing of the SEDesert and Gabgaba terranes is only disrupted by younger N-S structures (developed during terminalcollision between East and West Gondwanaland)of the N-S Hamisana Shear Zone. This outcroppattern indicates the ophiolites and associated accretionary structures are subhorizontal. NASA as-tronaut photograph (S32-74-100). (B) Landsat TM image of basement units in N. Eritrea and E.Sudan, showing dominance of complex deformation related to terminal collision between East andWest Gondwanaland, resulting in∼ N-S structures. Terrane names are modified after Drury andFilho (1998). Scene is about 90 km across.

3. Crustal Structure 101


Fig. 3. Reconstructed crustalsections of ANS ophiolites. Al ‘Ays(Al Wask), Saudi Arabia (Bakoret al., 1976); Bi’r Umq, Saudi Arabia (Al-Rehaili and Warden, 1980); Onib, Sudan (Kröner et al.,1987).

placed relatively early in the history of the ANS and EAO. Fragments from dismemberedophiolites are common in the ANS, and it becomes more difficult to interpret these as oncebeing allochthonous pieces of oceanic crust as these fragments become more deformed andmetamorphosed. Suffice it to say that not all ultramafic rocks in the shield are—or were—parts of ophiolites. There are many layered igneous intrusions containing non-ophioliticultramafics and gabbros and there are many examples of non-ophiolitic pillowed lavas.Nevertheless, the association of harzburgitic ultramafics and low-K tholeiitic metabasaltargues strongly that these once belonged to a coherent ophiolite.

4. Age 103

Fig. 4. Ages of ophiolites in the Arabian-Nubian Shield, binned at 25 million year intervals. Neo-proterozoic time (544 to 1000 Ma) is subdivided into Tonian (1000–850 Ma), Cryogenic (850 to∼ 600 Ma) and Neoproterozoic III (∼ 600–544 Ma), after Knoll (2000). Age range when Ara-bian-Nubian Shield was tectonically and magmatically active (870 Ma to end of Neoproterozoic) isalso given. Ophiolite ages are U-Pb and Pb-Pb zircon ages and Sm-Nd ages from (Stacey et al., 1984;Claesson et al., 1984; Pallister et al., 1988; Kröner et al., 1992; Zimmer et al., 1995; Worku, 1996).Mean age forophiolites±1 standard deviation is given.

4. AGE

ANS ophiolites have been reliably dated using U-Pb zircon techniques (Stacey et al.,1984; Pallister et al., 1988) and Pb-Pb zircon evaporation techniques (Kröner et al., 1992;Zimmer et al., 1995) on zircons separated from gabbros and plagiogranites. Other ageshave been generated using Sm-Nd mineral and whole-rock techniques (Claesson et al.,1984; Zimmer et al., 1995; Worku, 1996). These results give age ranges of 694± 8 Mafor the youngest ANS ophiolite (Urd/Halaban; Stacey et al., 1984) to 870± 11 Ma for theoldest (Thurwah; Pallister et al., 1988). A mean of 781 Ma (1 standard deviation= 47 Ma)is obtained for 16 robust ophiolite ages (Fig. 4). Ophiolites formed during the first halfof the time period encompassed by tectonic and magmatic activity of the Arabian-NubianShield. There is no obviously systematic geographic variation in the distribution of ANS


ophiolite ages; the oldest ophiolite is in the middle of the ANS and the youngest is near theeastern edge of ANS exposures.

5. OPHIOLITE COMPONENTS

5.1. Harzburgites

Where the original protolith can be identified, the mantle peridotites associated withANS ophiolites are predominantly tectonized harzburgites, lherzolite being rarely reported.Studies of ANS ophiolite fabrics are at a very early stage, but harzburgites and dunites ofthe Thurwah ophiolite have been tectonized athigh temperatures (Nassief et al., 1984),and a similar high-temperature ductile fabric is reported from the Sol Hamed ophiolite(Fitches et al., 1983). The harzburgites represent residual mantle after extensive melting,whereas the dunites and wehrlites are cumulates or reflect melt-wallrock interactions.

Harzburgites are mostly altered (Figs. 5A, B), but relict olivines and pyroxenes havebeen analyzed by electron microprobe for 5 ophiolites. Harzburgite associated with theAl Ays ophiolite contains olivine of Fo91–94, orthopyroxene of En89–91, and clinopyroxeneof En48–53, Fs2.2–3.0, Wo44–49.2(Chevremont and Johan, 1982a; Ledru and Auge, 1984).Harzburgite from the nearby Hwanet ophiolite has orthopyroxene of En88.9–91and clinopy-roxene of En48.9–50.3, Fs2.3–3.4, Wo46.3–48.9(Chevremont and Johan, 1982a). Harzburgiteassociated with the Ess ophiolite contains olivine of Fo91–93, orthopyroxene of En88–92,and diopsidic clinopyroxene of En49–52, Fs2.4–3.1, Wo44.6–48.3(Al-Shanti, 1982). Nassief etal. (1984) identified a mantle sequence that is up to 20 km thick for the Thurwah ophiolite,and harzburgite from this consists of 70–90% olivine (Fo89.5–93.4), 15–30% orthopyrox-ene (En90–92) and< 1% each of chromite and clinopyroxene. Mouhamed (1995) arguedon the basis of CIPW normative compositions of Muqsim serpentinites along the Allaqi-Heiani suture (Fig. 2) that these were originally harzburgites. Olivines from the Ingessanaophiolite are Fo91–97(Price, 1984). These compositions are at the magnesium-rich end ofperidotites, as shown on Fig. 6.

Olivine compositions provide insights into the tectonic setting of ophiolites becausemagmagenesis in different tectonic settings reflects differing extents of melting. Becauseresidual olivines become increasingly magnesian as melting progresses, residual mantleshould have olivines that are more magnesian than the Fo88 of undepleted ‘pyrolitic’ up-per mantle (Fig. 6). Bonatti and Michael (1989) suggest that mantle melting ranges fromnearly zero for undepleted continental peridotites to about 10–15% melting for rifted mar-gins to 10–25% melting associatedwith mid-ocean ridge (MOR) peridotites to 30% forperidotites recovered from forearcs, which generally form during the early stages in theevolution of the associated subduction zone (Bloomer et al., 1995). Mantle peridotites fromback-arc basins were not available when this diagram was originally generated, but sincethat time mantle peridotites from the Mariana Trough active back-arc basin in the westernPacific have been studied (Ohara et al., 2002). These harzburgites have olivine composi-tions that are indistinguishable from MOR harzburgites and are distinctly less magnesian

5. Ophiolite Components 105

Fig. 5. Outcrop photographs of ANSophiolites. (A) Barramiya, Egypt. Light areas are talc-carbonatealteration of harzburgitic ultramafics, darker areas are serpentinized ultramafics. White house forscale. (B) Thurwah, Saudi Arabia. Light areas are talc-carbonate alteration of harzburgitic ultra-mafics, darker areas are serpentinized ultramafics. Seated geologist (far left) for scale. (C) Carbon-ated ultramafics of the Bi’r Umq ophiolite (Saudi Arabia) thrust south over metasediments of theMahd Group. Thrust contact is dashed. Three geologists climbing ridge for scale. (D) N-dipping lay-ered gabbros and cumulate ultramafics on the south side of the Gerf ophiolite, SE Egypt/NE Sudan.Ridge is about 100 m tall. (E) Layered gabbros ofthe Onib ophiolite, NE Sudan. Light layers areanorthositic, darker layers are richer in mafic minerals. Card (∼ 15 cm long) for scale. (F) Pillowedbasalts, Wadi Zeidun (near Fawkhirophiolite), Egypt. Hammer for scale.


Fig. 5. (Continued.)

than those of forearc peridotites. In this context, the Mg-rich nature of ANS ophiolite peri-dotite olivines is best interpreted to indicate that these are residual after extensive melting,similar to that observed for forearc peridotites.

Inferences about the extent of melting based on olivine compositions are supportedby abundant compositional data for spinels. Spinel resists alteration better than olivine,may be economically important, and thus are relatively well studied for ANS ophiolites(Fig. 7). Progressive melting of peridotites depletes Al relative to more refractory Cr inresidual spinels, such that the Cr# of spinels (= 100Cr/Cr+ Al) increases with melting


Fig. 6. Composition of olivines in peridotites from various tectonic settings, modified after Bonattiand Michael (1989). The tectonic settings are arranged in an orderin which melt depletion increasesto the right. Note that the field for olivines from peridotites of 4 ANS ophiolites are Fo91 or are moremagnesian. These compositions are most consistent with formation in a forearc setting.

(Dick and Bullen, 1984). As is the case for Fo content of harzburgitic olivines, the increasein Cr# of harzburgitic spinels reflects increasing degree of melting. Degree of melting isrelated in a general way to tectonic setting, at least for Cenozoic peridotites (Fig. 7A).Spinels from MOR peridotites generally have Cr#< 50 (although Barnes and Roeder,2001 report that a subordinate proportion of MOR peridotites have Cr# up to 80). A limiteddataset for spinels from back-arc basin peridotites indicates that these experienced extentsof melting similar to MOR and thus have spinel Cr# that are similar to those of MORperidotites (Ohara et al., 2002). Spinels in forearc harzburgites generally have higher Cr#


Fig. 7. Composition of spinels in ANS ophiolitic peridotites compared with those in modern peri-dotites. Data are plotted on 100Cr/Cr+ Al (Cr#) vs. 100Mg/Mg+ Fe (Mg#) diagram, modified afterDick and Bullen (1984). (A) Spinels from ANS ophiolitic peridotites, location and size of rectangledetermined by means and standard deviations listed in Table 1. (B) Fields defined by spinel composi-tions for likely ophiolitic analogues of Cenozoic age, modified afterBloomer et al. (1995). Note thatspinels from peridotites from mid-ocean ridges and back-arc basin spreading axes characteristicallyhave Cr#< 60, whereas spinels from forearc peridotites and boninites Cr#> 40. The vast majorityof ANS ophiolitic spinels have high Cr#, suggesting these formed in a forearc setting.


Table 2. Mean compositions of spinels in ANS harzburgites

Locality Mean Cr# 1 Std. Dev. Mean Mg# 1 Std. Dev.Sudan

Onib/Sol Hamed 70.4 17.6 56.4 11.9Oshib 71.9 9.2 65.1 5.4Rahib 73.4 11.7 56.5 22.3Ingessana 77.5 7.0 57.6 14.7

EgyptVarious 52.3 5.8 56.6 4.9

Saudi ArabiaEss/Al ‘Ays 67.2 10.0 60.6 9.9Nabitah 78.3 4.9 64.7 4.5Al Amar 62.9 14.2 52.4 5.4Thurwah 79.8 6.4 56.0 2.8Bi’r Umq 52.6 7.8 61.0 4.1Tuluhah/Arjah 72.4 4.4 70.2 8.0Halaban 76 6.8 67.4 2.9

EthiopiaAdola 91.4 2.5 63.1 7.6Moyale 58.2 2.8 64.4 2.6

KenyaSekerr 88.1 1.3 68.7 7.9Baragoi 82.6 1.1 56.7 5.1

Data sources: Onib-Sol Hamed, Sudan (N = 60: Abdel-Rahman, 1993; Hussein, 2000; Price, 1984); Oshib,Sudan (N = 15: Abdel-Rahman, 1993); Rahib, Sudan (N = 15: Abdel-Rahman, 1993); Ingessana-Kurmuk,Sudan (N = 97: Abdel-Rahman, 1993; Price, 1984); Egypt (various; 4 averages for spinels in harzburgite:Ahmed et al., 2001); Ess-Al ‘Ays, Saudi Arabia (N = 81: Al-Shanti and El-Mahdy, 1988; Al-Shanti, 1982;Chevremont and Johan, 1982a; Chevremont and Johan, 1982b; Ledru and Auge, 1984); Nabitah, Saudi Ara-bia (N = 9: Al-Shanti and El-Mahdy, 1988); Bi’r Umq, Saudi Arabia (N = 20: LeMetour et al., 1982); Thurwah,Saudi Arabia (N = 9: Al-Shanti and El-Mahdy, 1988); Tuluhah/Arjah, Saudi Arabia (N = 14: Al-Shanti andEl-Mahdy, 1988); Halaban, Saudi Arabia (N = 8: Al-Shanti and El-Mahdy, 1988); Adola, Ethiopia (N = 10:Bonavia et al., 1993); Moyale, Ethiopia (N = 32: Berhe, 1988); Sekerr, Kenya (N = 123: Price, 1984); Baragoi,Kenya (N = 50: Berhe, 1988).

(up to 80) and spinels from boninites typically have Cr# of 70–90 (Fig. 7B), consistentinferences from olivine compositions that these are manifest residues and products of thehighest degree of melting found for post-Archean igneous rocks.

Table 2 summarizes spinel compositions from peridotites (mostly harzburgites) associ-ated with 16 ANS ophiolites. These are mostly for spinels in harzburgites, but where thisdata was unavailable, compositions of podiform chromite were used. The data indicate thatANS peridotitic spinels have Cr# that are mostly> 60 (Egypt, Bi’r Umq, and Moyale be-ing notable exceptions). Spinelsfrom the Sekerr ophiolite (mean Cr#= 88) and the Adolaultramafic complex (mean Cr#= 91) are remarkably rich in chromium. Overall, the Cr#


of ANS ophiolitic peridotites appear most similar to those of modern forearc peridotites(Fig. 7B), although there is a hint of a regional gradient, from higher Cr# in the south tolower Cr# in the north. Podiform chromites from the Adola complex in southern Ethiopiaare associated with harzburgites; Bonavia et al. (1993) did not explicitly identify this aspart of an ophiolite, but they argued from the absence of phase layering and low Rh/Ir thatthe rocks did not form as part of a layered intrusion. These data supports an interpretationthat the Adola mafic-ultramafic complex represents a highly metamorphosed, deformedophiolite. This inference is supported by the interpretation of Yibas et al. (2003) that manymafic-ultramafic complexes in southern Ethiopia are ophiolites.

5.2. Transition Zone and Gabbros

The transition zone lies between the petrologic Moho (defined at the base of the cumu-late section and the top of the tectonized peridotites) and the seismic Moho (defined asthe top of cumulate ultramafic section and the base of gabbros). Several ANS ophioliteshave well-preserved transition zones characterized by interlayered pyroxenite, wehrlite,lherzolite, dunite, and/or chromite at the base (Fig. 5D) grading upwards into sections thatare increasingly dominated bygabbro. Other ophiolites have thin to non-existent transi-tion zones (e.g., Fawkhir; El-Sayed et al.,1999). The Onib ophiolite is characterized byan unusually thick (2–3 km) transition zone of interlayered cumulateultramafics, podi-form chromites, and layered gabbros (Fig. 3; Hussein et al., this volume; Kröner et al.,1987). The ultramafic rocks of the Sol Hamedophiolite, NW of Onib, also seem to repre-sent a similar crust-mantle transition zone, and are 80% dunite (with interbedded chromi-tites), with lesser proportions of wehrlite, harzburgite, and pyroxenite (Fitches et al., 1983;Price, 1984). Similarly, the Oshib ophiolite has a 2-km thick transition zone, which gradesupwards from harzburgite tectonite throughdunite and wehrlite followed by pyroxeniteand cumulate gabbro (Abdel-Rahman, 1993). The transition zone of the Thurwah ophiolitecontains about 1 km of cumulate dunite, lherzolite, and pyroxenite, in similar proportions(Nassief et al., 1984). Each rock layer is 1–10 m thick and is developed in the upwardsequence dunite-lherzolite-pyroxenite. TheEss ophiolite containsa cumulate peridotitesection that is about 400 m thick and consistsof wehrlite and dunite (Shanti and Roobol,1979).

Significant bodies of dunite, often associated with podiform chromite, exist at the baseof some gabbro sections (El-Bayoumi, 1983). Olivines in dunites from the Thurwah ophio-lite are significantly more Fe-rich (Fo86.5–89.5; Nassief et al., 1984) than olivines of typicalANS harzburgites. The Al Ays ophiolite contains over 350 lensesof podiformchromititewithin the dunite unit (Bakor et al., 1976). Chromites associated with dunites are often sig-nificantly more Cr-rich (Cr#= 65–85) than those associated with harzburgite (Cr#∼ 50;Ahmed et al., 2001); the reason for this is not understood. Some chromites contain in-clusions of olivine (Fo97–99) and diopside, along with hydrous phases, such as amphibole(edenite-tremolite) and phlogopite (Ahmed etal., 2001). The primary mineralogy of theseinclusions is remarkably preserved, and studying inclusions in chromites and other resis-tant minerals is a promising avenue for understanding ANS ophiolites.


Modal abundances of olivine decrease upsection as the abundance of especially clinopy-roxene increases; correspondingly dunites are upwardly succeeded by wehrlites or lher-zolites and then pyroxenites. Cumulate pyroxenites are dominated by diopside. Cumulatepyroxenites of the Onib transition zone are low in TiO2 (0.01–0.06%) and rich in Cr (1500–3000 ppm; Kröner et al., 1987). Gabbro bodies up to 1 km thick are interleaved with theultramafic cumulates at Igariri, part ofthe Oshib ophiolite (Abdel-Rahman, 1993). ANSophiolites with well-developed transition zones may have formed by fast seafloor spread-ing, whereas those that do not have thick transition zones may represent seafloor producedby slow spreading (Dilek et al., 1998).

Gabbros are ubiquitous and important components of ANS ophiolites. Original igneoustextures are common but metamorphic recrystallization under greenschist to amphibo-lite facies conditions is ubiquitous. ANS ophiolitic gabbros are mostly pyroxene gabbro;olivine gabbro is much less common. Clinopyroxene generally dominates over orthopyrox-ene. Where ophiolitic gabbros are well preserved and studied, they are commonly layered,at least in part. Phase layering, evidenced by melanogabbro and anorthositic gabbro cou-plets are typically∼ 5 cm thick and can be traced for meters. Layered metagabbros madeup of alternating plagioclase (An63–83)- and amphibole-rich layers are part of the Ess ophi-olite (Shanti, 1983). Plagioclase compositions in Sol Hamed gabbros change from An70–85

at the base to more sodic compositions upsection (Fitches et al., 1983). Amphibole maybe uralitized pyroxene (El-Bayoumi, 1983). Similar uralitized clinopyroxene gabbros arereported for the Al Ays (al Wask) ophiolite. Upwards through the cumulate sequences,from the base of the transition zone into the layered gabbros, the sequence of rocks in-dicates a crystallization sequence of olivine± chromite-clinopyroxene-plagioclase (Price,1984), olivine± chromite-clinopyroxene-orthopyroxene-plagioclase (Nassief et al., 1984),or, less commonly, olivine± chromite-orthopyroxene-clinopyroxene-plagioclase (Abdel-Rahman, 1993).

High level gabbros are more massive, lack layering and other evidence of crystal accu-mulation, and commonly include pegmatitic gabbro and isolated bodies of plagiogranite.ANS plagiogranites are high in SiO2 (70–77%) and low in K2O (0.04–1.9%: Shanti, 1983;Abdel-Rahman, 1993). These mostly plotin the field of trondhjemite and tonalite on anormative Ab-An-Or diagram (Shanti, 1983). The high level gabbros are often intruded bymafic dikes, which represent the base of the sheeted dike complex.

5.3. Sheeted Dykes

Sheeted diabase dikes are common components of ANS ophiolites. Where observed theytypically transition downwards into the high-level gabbros and grade upwards into pillowedbasalts. Sheeted dikes are reported from thefollowing ophiolites: Rahib (Abdel-Rahmanet al., 1990), Ess (Shanti and Roobol, 1979), Sol Hamed (Fitches et al., 1983), Inges-sana (Price, 1984), Ghadir (El-Bayoumi, 1983), Gerf (Zimmer et al., 1995), and Thurwah(Nassief et al., 1984). Sheeted dykes for many other ANS ophiolites are not identified or arepoorly developed (e.g., Al Ays and Fawkhir: Bakor et al., 1976; El-Sayed et al., 1999). It


is not yet clear whether this represents differential preservation of diagnostic dike-on-diketextures or represents originaldifferences in crustal sections.

5.4. Pillowed Basalts and Other Volcanics

Lavas with well-preserved pillow structures are diagnostic parts of ANS ophiolites(Fig. 5F). These are especially important because lava compositions provide valuable cluesabout tectonic setting and the nature of melt generation. This section thus concentrates onthe chemical composition of ANS ophiolitic lavas.We recognize that the greenschist-faciesmetamorphism that these basalts have suffered probably disrupted their chemical compo-sition, but it is likely that these effects may cancel each other out (e.g., Mg gain in onelava, Mg loss in another). For the purpose ofunderstanding the compositional variabil-ity of ANS ophiolitic lavas, we compiled available chemical data for these. A total of200 samples were used in the compilation, including 37 analyses from Egypt (El-Sayedet al., 1999; Stern, 1981; Zimmer et al., 1995), 52 from Sudan (Price, 1984; Abdel-Rahman, 1990, 1993; Harms et al., 1994; Hussein, 2000), 32 from Arabia (excludingDarb Zubaydah; Al-Shanti, 1982; Bakor et al., 1976; Kattan, 1983; Nassief et al., 1984;Shanti, 1983), and 29 from the Sekerr ophiolite, Kenya (Price, 1984). An additional50 samples from Eritrea (Woldehaimanot, 2000) and Ethiopia (Wolde et al., 1996;Woldehaimanot and Behrmann, 1995) were used to compare the metavolcanic sequencesof suspected ophiolites in these countries. These data are summarized in Table 3.

Table 3. Mean composition ofANS ophioliticpillowed basalts

Egypt Arabia Sudan Ethiopia& Eritrea

SekerrDPL

SekerrUPL

ANSmean

N 37 32 52 50 16 13 200SiO2 (%) 51± 4 50± 3 49± 3 49± 4 46± 2 47± 2 49± 4TiO2 1.3± 0.4 1.0± 0.5 1.1± 0.6 0.7± 0.7 1.4± 0.3 2.7± 0.7 1.2± 0.7FeO∗ 9.8± 2.4 9.4± 2.2 10.0± 1.8 11.2± 3.0 9.9± 1.0 11.2± 0.9 10.2± 2.3MgO 6.2± 2.0 6.9± 1.7 7.5± 2.2 9.7± 5.9 8.4± 3.7 4.6± 1.2 7.6± 3.8K2O 0.18± 0.13 0.28± 0.31 0.15± 0.21 0.43± 0.8 0.14± 0.09 0.48± 0.35 0.26± 0.46Mg# 52± 8 56± 7 56± 9 58± 14 58± 8 42± 7 55± 11Sr (ppm) 147± 144 143± 47 191± 114 192± 205 257± 135 466± 85 199± 161Ba 84± 81 25± 21 70± 44 72± 102 99± 86 259± 241 85± 112Y 33± 13 21± 9 24± 7 13± 10 25± 7 33± 7 23± 12Zr 87± 36 65± 44 70± 37 42± 42 111± 20 171± 15 76± 49V 320± 90 230± 75 287± 75 84± 67 222± 40 227± 19 234± 108Cr 295± 150 326± 212 356± 238 467± 591 621± 535 160± 154 380± 392Ni 110± 56 99± 76 113± 84 155± 246 304± 305 69± 81 135± 174

Data sources: Egypt (El-Sayed et al., 1999; Stern, 1981;Zimmer et al., 1995), 52 from Sudan (Abdel-Rahmanet al., 1990; Price, 1984; Abdel-Rahman, 1993; Harms et al., 1994; Hussein, 2000); Arabia (excluding DarbZubaydah; Al-Shanti, 1982; Bakor et al., 1976; Kattan, 1983; Nassief et al., 1984; Shanti, 1983); Kenya (Price,1984); Eritrea (Woldehaimanot, 2000) and Ethiopia (Wolde et al., 1996; Woldehaimanot and Behrmann, 1995).


ANS ophiolitic lavas are dominantly basalts, with a mean of 49% SiO2 (1 standarddeviation= 4% SiO2). There are some andesitic and even a few dacitic samples identi-fied as part of the ophiolitic sequence (e.g., Fawkhir, Egypt). These lavas average 1.2%(±0.7%) TiO2. ANS ophiolitic pillow basalts define a low-K suite, with mean K2O =0.26%. These major element characteristics are similar to a wide range of oceanic lavas,including MORB, some intra-oceanic arc lavas, and back-arc basin basalt (BABB). Ophi-olitic ANS lavas are often fractionated, with Mg# (= 100Mg/Mg+ Fe)= 55± 11. Mg#for basaltic melt in equilibrium with mantleperidotite is expected to be in the range 65–70.The mean of 135 ppm Ni (±174 ppm) and 380 ppm Cr (±392 ppm) is higher than would beexpected for a mean Mg#= 55. Eritrean and Ethiopia suspected ophiolitic lavas typicallycontain less TiO2, Y, and Zr and have higher K2O and Mg#, Cr and Ni than most other ANSophiolitic lavas. Upper pillow lavas from the Sekerr ophiolite are also distinct, with muchhigher TiO2, Y, and Zr along with lower Mg#, Cr, and Ni than most other ANS ophioliticsequences. The Sekerr UPL sequence may represent an alkalic succession, something thatis rarely found for other ANS ophiolitics.

ANS ophiolitic lavas are mostly tholeiitic but also include some calc-alkaline examples(Fig. 8). There is no obvious geographic variation. Samples from Kenya and Eritrea aredominated by tholeiites whereas those from Ethiopia are predominantly calc-alkaline. Mostsamples from Sudan, Egypt, and Arabia are tholeiitic, but with a significant proportion ofcalc-alkaline representatives as well.

REE data for 67 samples from the ANS (including several from suspected ophiolitesin Ethiopia and Eritrea) indicate that both LREE-enriched and LREE-depleted varietiesexist. A simple way to see the extent to which samples are LREE-enriched or LREE-depleted is by use of the chondrite normalized Ce/Yb ratio, or (Ce/Yb)n. If (Ce/Yb)n > 1,the REE pattern is generally LREE-enriched (and the higher the ratio, the greater theLREE-enrichment). Similarly, if (Ce/Yb)n < 1, the sample is LREE-depleted. The mean(Ce/Yb)n for ANS ophiolitic lavas is 2.18, but with a very large standard deviation of 3.07.This largely results from one unusual sample or analysis (MV33 with (Ce/Yb)n = 21.9;El-Sayed et al., 1999), and omitting this sample lowers the mean (Ce/Yb)n to 1.89± 1.88.Another way to summarize the REE patterns is to look at Sm/Nd. The chondritic value,taken to approximate the bulk Earth, is∼ 0.325, so values greater than this are LREE-depleted and values less than this are LREE-enriched. The mean for ANS ophiolitic lavas(0.30± 0.13) again indicates modest LREE-enrichment. Including data for Eritrea andEthiopia lowers the mean Sm/Nd to 0.29± 0.10.

A variety of discriminant diagrams can be applied to examine the tectonic affinities ofANS ophiolitic lavas. A plot of V vs. Ti (Fig. 9) shows that most samples have Ti/V be-tween 20 and 50, and plot in fields defined by mid-ocean ridge and back-arc basin basalts.This includes most of the samples from Egypt, Sudan, and Sekerr, Kenya, downstream pil-low basalts. A substantial portion of the ophiolitic basalts have Ti/V< 20 and so plot in thefield of island arc tholeiites or calc-alkaline basalts. This includes a subordinate proportionof samples from Egypt and Sudan. No field forboninitic rocks is shown on these diagrams,but boninites have very low Ti and V and plot near the origin. Eritrea and Ethiopia samples


Fig. 8. Magmatic affinities of ANS ophiolitic pillow lavas and suspected ophiolitic metavolcanicsfrom Eritrea and Ethiopia. (A) TiO2 vs. FeO∗/MgO. (B) SiO2 vs. FeO∗/MgO. Field boundaries afterMiyashiro (1975).

have low Ti/V and some have very low Ti contents, supporting suggestions that these areboninitic. Only a subset of the upstream pillow lavas from the Sekerr ophiolite plot in thefield of within-plate basalts.


Fig. 9. Ti-V discrimination diagram for basalts, after Shervais (1982). IAT= arc tholeiite, MORB &BABB = mid-ocean ridge basalt and back-arc basin basalt, WPB= within-plate basalt. Calc-alkalinebasalts have low Ti concentrations and a wide range of Ti/V and plot in the grey field.

A plot of Cr vs. Y shows similar mixed tectonic affinities for ANS ophiolitic lavas(Fig. 10). Samples from Egypt, Sudan, and downstream pillow lavas from the Sekerr ophi-olite mostly plot in the MORB field (note that there is no field for back-arc basin basalts onthis diagram). Samples from Arabia plot in the MORB field and in the field for arc basaltsor even on the low-Y side of the field for arc basalts. Samples from Eritrea and Ethiopiaplot within the field for arc basalts and to the low-Y side of the arc field. Again, only a sub-set of the upstream pillow lavas from the Sekerr ophiolite plot in the field of within-platebasalts.

Finally, Zr-based discriminant diagrams (Fig. 11) show the mixed affinities of ANSophiolitic lavas. On a plot of Ti vs. Zr (Fig.11A), the samples are remarkably coher-ent, clustering about a mean Ti/Zr= 97± 41 (excluding Ethiopia and Eritrea), decreasingslightly to Ti/Zr = 95± 41 when data from Ethiopia and Eritrea are included. In contrastto Fig. 8, a calc-alkaline trend is not seen on the Ti-Zr diagram. Fig. 11A suggests that theophiolitic lavas are simply related by differing degrees of melting and fractionation, withSekerr ophiolite samples representing low degrees of melting (and/or extensive fraction-ation), whereas samples plotting near the origin (most samples from Ethiopia-Eritrea andsome samples from Arabia, Sudan, and Egypt) are relatively high degree melts. Most of the


Fig. 10. Cr-Y discrimination diagram for basalts (Pearce, 1982). Note that there is no field forback-arc basin basalts.

Sekerr upstream pillow lavas and a sample from Al Ays (sample 5 of Bakor et al., 1976),a sample from the Uogame basalts of Eritrea (ER54 of Woldehaimanot, 2000), and OPL/2from Onib (Abdel-Rahman, 1993) lie along the extension of the MORB field in a regionwhere within-plate basalts plot. Zr/Y vs. Zr systematics reveal similar affinities and alsoshows that a large proportion of samples plot to the low-Zr side of fields defined by arcbasalts, MORB, and within-plate basalts. This is the area where boninites plot, using thesummary data of Crawford et al. (1989), and samples from Arabia, Sudan, and especiallyEthiopia and Eritrea plot in this field.

5.5. Sediments

Pelagic sediments that rest immediately on the uppermost part of ANS ophiolites includedolomite and ribbon chert (e.g., Al Ays: Bakor et al., 1976; Hussein, 2000). These lime-stones are inferred to be similar to modern pelagic carbonates, whereas the cherts arethought to be due to silica-rich emanations without the involvement of radiolaria. Metased-


Fig. 11. Zr-based discriminantdiagrams for ANS ophiolitic basalts. (A) Ti vs. Zr (Pearce and Cann,1973). Field labels: I= arc tholeiites, II= MORB, III = calc-alkaline basalt. A fourth field, situatedwithin I but obscured by data points is defined by all three fields. No field for within-plate lavas isshown, but these plot along the upper right extension of the MORB field. (B) Zr/Y vs. Zr (Pearce andNorry, 1979). Gray field shows composition of boninites, from Table 1-1 of Crawford et al. (1989).

iments overlying the Ess ophiolite include shale and minor conglomerate (Shanti andRoobol, 1979). Ophiolites in the Central EasternDesert of Egypt are overlain by sedimentsthat include diamictite, tuffaceous siltstones, and banded iron formation (Stern, 1981;Sims and James, 1982). Future ophiolite studiesshould take care to carefully character-ize the nature of the sedimentary sequence immediately above the pillowed lavas.


All too little attention has been paid to the nature and composition of sediments thatwere deposited on ANS ophiolites, probably because investigations of these ophioliteshave mostly been led by petrologists and igneous geochemists. Becausethe Neoproterozoicwas an exceptional period in Earth history, characterized by extensive—perhaps global—glaciation, unparalleled excursions in seawaterstable isotopic compositions, and explosiveradiation of metazoa (Evans, 2000), sediments of this age are of increasing interest to theglobal geoscientific community. Efforts to understand the Neoproterozoic ‘Snowball Earth’have concentrated to date on shallow water sequences, with very little of the record pre-served in deep-sea sediments examined. Because sediments deposited on ANS ophiolitesshould preserve an excellent record of the composition of deep water of the Neoproterozoicocean, these should become an increasing focus of research in the near future.

5.6. Alteration and Obduction-Related Metamorphism

The ultramafic rocks associated with ANS ophiolites are generally highly altered, butit is often not known whether this alteration occurred before, during, or after emplace-ment. These rocks are largely converted to serpentinite or to mixtures of serpentine, talc,tremolite, magnesite, chlorite, magnetite, and carbonate. These rocks are variously called‘talc-carbonate schists’, ‘Barramiya rocks’, or ‘listwaenite’. Strictly speaking, listwaen-ite should be reserved for rocks that are fuchsite-quartz-carbonate lithologies derivedfrom ultramafic rocks by potassic and carbonate metasomatism (Halls and Zhao, 1995),but common usage in the ANS is more general. Silicified serpentinites are sometimescalled ‘birbirites’ (Augustithus, 1965). Bastaand Kader (1969) reported that lizardite isthe main constituent of Egyptian serpentinites, whereas Akaad and Noweir (1972) iden-tify antigorite as volumetrically dominant. The talc-carbonate rocks mainly consist ofmagnesite (± dolomite) and talc. The origin of the carbonate alteration fluids remainsto be elucidated, but Stern and Gwinn (1990) argued on the basis of C and Sr isotopicstudies that carbonate intrusions in the Eastern Desert of Egypt—which could be re-lated to the carbonatizing fluids affecting ANS ultramafic rocks—are mixtures of mantle-derived and remobilized sedimentary carbonate. Certainly the prevalence of carbonatealteration of ANS ophiolitic ultramafics suggests a tremendous flux of CO2-rich fluidsfrom the mantle during middle and late Neoproterozoic time (Newton and Stern, 1990;Stern and Gwinn, 1990). In contrast, Surour and Arafa (1997) argued that the ‘ophicar-bonates’ of the Ghadir ophiolite are reworked oceanic calcites that formed after it wasobducted.

Regardless of how carbonatization of the ophiolitic ultramafics occurred, it has eco-nomic implications. A spatial and genetic relationship has been observed between carbon-atized ultramafics, subsequent granite intrusions, and gold mineralization. Apparently thecarbonatization preconcentrates gold up to 1,000 times that in the original ultramafic rocks,and interaction with hydrothermal systems associated with granite intrusions may furtherconcentrate gold (Cox and Singer, 1986).

Thrust contacts are documented at thebase of some, but not all, ANS ophiolites(Fig. 5C). Metamorphic soles of ANS ophiolites have been studied along the west-

6. Isotopic Data 119

ern part of the Allaqi-Heiani ophiolite belt in SE Egypt, where El-Naby and Frisch(1999) inferred temperatures of up to 700◦C and pressures up to 8 kbar. Other thrustsat the base of the ophiolites are associatedwith no obvious thermal effects. Chlo-ritites are developed around the peripheries of someultramafic masses (Takla, 1991;Takla et al., 1992). The metamorphic sole of the Halaban ophiolite consists of tightly foldedamphibolites, hornblendites, gneisses, magmatirestites, rodingites and serpentinized ultra-mafics, all of which are cut by granitic dikes associated with partial melting (Al-Salehet al., 1998). The metamorphic sole of the Halaban ophiolite in Arabia is divided intoa western segment that is dominated by greenschist to amphibolite facies metasedimen-tary rocks, structurally overlain by higher grade amphibolite-facies meta-igneous rocks.Such inverted metamorphic zonation is common in sub-ophiolitic complexes (Al-Saleh etal., 1998). Most amphibolites and meta-sediments in the latter segment have experiencedvarying degrees of calcium metasomatismdue to the release of excess calcium duringserpentinization, and the original assemblages have often been rodingitized. The Halabanophiolitic sole includes exotic blocks of serpentinite and metamorphosed ultramafic rock,interpreted to be derived from the basal peridotites of the Halaban Ophiolite that becamefragmented as thrusting progressed and were later incorporated within the underlying am-phibolites.

6. ISOTOPIC DATA

There has been a modest amount of isotopic work conducted on ANS ophiolites, in-cluding two ophiolites in Arabia, a concentration of work around the Gerf ophiolite in SEEgypt, and data for the Adola mafic-ultramafic complex. The most reliable isotopic data forANS ophiolites comes from Sm-Nd isotopic systematics. This is because alteration makesit difficult to rely on Sr and Pb isotopic compositions, whereas Sm and Nd are relativelyimmobile and corrections for in situ radiogenic growth is simple if the age is known. Theseresults are summarized in Table 4, which shows that all ANS ophiolites studied to datehave strongly positiveεNd (+5.0 to +7.7). This indicates that these melts were generatedfrom a long-depleted (high Sm/Nd) mantle source. The Nd isotopic data indicate an as-thenospheric source and a juvenile, ensimatic setting. There is a hint that the mantle source

Table 4. Neodymium isotopic composition of ANS ophiolites

Ophiolite Age (Ma) εNd(t) ReferenceEss (2) 780 6.9 Claesson et al. (1984)Al ‘Ays (Wask) 743 7.6 Claesson et al. (1984)Harga Zarqa (10) 750 7.6± 0.5 Zimmer et al. (1995)Heiani (7) 750 7.7± 0.8 Zimmer et al. (1995)Gerf (20) 750 6.8± 0.7 Zimmer et al. (1995)Adola 789 5.0 Worku (1996)

Numbers in parentheses refer to number of samples used to calculate mean value.


for the northern ophiolites may have been more—or longer—depleted than those farthersouth, but the data at present are too sparse for this to be anything more than a suggestion.

Os isotope data on chromites from the AlAys ophiolite have recently been reported(Walker et al., 2002); these straddle the line defined for the evolution of the primitivemantle.

7. DISCUSSION

Ophiolites of mid-Neoproterozoic age (690 to 890 Ma; mean= 781± 47 Ma) areabundant in NE Africa and Arabia. Ophiolites encompass an area of 3000 km N-S and> 1000 km E-W. These ophiolites are in various stages of dismemberment and alteration,but all of the diagnostic components can be found, including harzburgite, cumulate ultra-mafics, layered as well as higher level gabbro and plagiogranite, sheeted dikes, and pil-lowed basalt. Reconstructions of a few ophiolitic successions indicate a crustal thicknessof 2.5 to 6 km. Many ANS ophiolites mark suture zones where smaller terranes coalesced;these suture zones indicate the location of fossil subduction zones. Some ANS ophioliteswere emplaced while still hot enough to metamorphose underlying rocks, while otherswere emplaced cold. This gross tectonic setting is simplest to explain if ANS ophiolitesgenerally were located on the hanging wall of a convergent plate margin and were em-placed when buoyant crust entered the subduction zone. Such an event could cause thesubduction zone to fail, suturing the two terranes at the same time that the ophiolite wasemplaced (Cloos, 1993). This describes a forearc setting for such ophiolites, an interpreta-tion that finds increasing favor in the scientific community (Shervais, 2001).

Mineral and lava compositions as well as limited isotopic data are consistent with thehypothesis that most ANS ophiolites formed in ‘suprasubduction zone’ (SSZ) settings.Most ANS ophiolites have the hallmarks of forearc ophiolites. Harzburgites are the mostcommon type of mantle peridotite, and these contain magnesian olivines and spinels withcompositions that indicate large extents of melting. Limited data for relict olivines inharzburgites show these to be significantly more Mg-rich than peridotites recovered frommodern mid-ocean ridges and similar to olivines in harzburgites recovered from forearcs.Limited data for olivines from backarc basinperidotites are indistinguishable from MORBperidotites, arguing against a back-arc basin setting. This interpretation is consistent withspinel compositions. Cr# for spinels in ANS harzburgites are mostly> 60, again most likethose recovered from modern forearcs and distinctly higher than those from mid-oceanridges and the admittedly sparse database for backarc basin peridotites.

ANS ophiolites are often associated with a thick (1–3 km) sequence of cumulate ul-tramafic rocks, which define a transition zonebetween the seismic and petrologic Mohos.These cumulates consist of a very high proportion of dunite but there are also a lot ofpyroxene-rich lithologies. Chromites associated with dunites are often more Cr-rich thanthose in the underlying spinels. These cumulate ultramaficstransition upwards into layeredgabbro. Together these cumulate sequences indicate that the extensional magmatic sys-tems represented by ANS ophiolites experienced significant fractionation. Several crystal-

7. Discussion 121

lization sequences are inferred from ANS transition zones and cumulate gabbro sections.Olivine and spinel are always first to crystallize: especially cpx-plag and cpx-opx-plag;opx-cpx-plag is rare. These are crystallization sequences (respectively) C, B, and A ofPearce et al. (1984), identified as hallmarksof SSZ ophiolites. This sequence contrasts withthe crystallization sequence olivine-plagioclase-clinopyroxene which Pearce et al. (1984)interpreted to be diagnostic of ophiolites that formed at true mid-ocean ridges. The lattercrystallization sequence has not been reported for ANS ophiolites. The overall sense ofANS ophiolites inferred from the transition zone and the gabbroic section is that of large,fractionating magma chambers that were repeatedly tapped and recharged but which gen-erally evolved from a system dominated by primitive mafic magmas to ones dominated byhighly evolved magmatic liquids.

ANS ophiolitic lavas mostly define a subalkaline suite characterized by low K andmoderate Ti contents that is moderately fractionated. They reveal both tholeiitic and calc-alkaline affinities and include a significant if subordinate proportion of boninites. We findlittle in the structure or composition ofANS ophiolites to support the hypothesis thatANS crustal growth entailed widespread involvement of oceanic plateaus or large ig-neous provinces, although rare examples of within-plate lavas are identified. ANS ophi-olitic lavas are quite fractionated (mean Mg#= 55) but have higher abundances of Cr(mean= 380 ppm) and Ni (mean= 135 ppm) than would be expected for this relativelylow Mg#. ANS ophiolitic lavas include both LREE-depleted and LREE-enriched varieties,but as a group are slightly LREE-enriched: mean (Ce/Yb)n ∼ 2.2 and Sm/Nd∼ 0.30. On avariety of discrimination diagrams, ANSophiolitic lavas plot in fields for MORB, BABB,and arc tholeiites, along witha significant proportion of lavas with strong boninitic affini-ties. ANS lavas cluster reasonably tightly around Ti/Zr= 97, indicating that Ti-bearingphases did not precipitate early. These mixed subalkaline characteristics are characteristicof SSZ ophiolitic lavas, and the presence of boninitic lavas in particular supports a forearcorigin.

There is a tremendous amount of work that needs to be done on ANS ophiolites. Most ofthe information that we have on ANS ophiolites was collected in the 1980s and early 1990sand the field is ripe for new perspectives and for detailed studies using modern techniques.There have been few modern geochemical studies. Most of the trace element data sum-marized in this overview was generated with XRF techniques. We need to analyze ANSophiolitic basalts using modern plasma analytical techniques. ModernICP-MS techniquesresult in much better precision and accuracyfor many more elements, including Nb andTa, which are essential to understand tectonic setting of igneous rocks. Interestingly, one ofthe few ANS ophiolites that have been studied with modern geochemical techniques—theGerf ophiolite—has strong trace element affinities to true MORB.

The geochronologic and isotopic database is limited and additional data are needed ifwe are to understand when the crust and mantlelithosphere represented by these ophioliteswere generated as well as when a given ophiolite was emplaced. Limited Sm-Nd isotopicdata indicate derivation from depleted asthenospheric mantle. Isotopic data however donot resolve tectonic setting. There are no Hf isotopic data for these ophiolites, and suchanalyses should provide a valuable perspective on the evolution of the Lu-Hf isotopic sys-


tem in the Neoproterozoic mantle. We also need to better understand the composition ofprimary magmatic liquids; compositions of olivines and spinels in ANS harzburgites in-dicate unusually high degrees of melting so primary liquids should be very primitive butlava compositions are very fractionated. One way to resolve this apparent contradictionwould be to analyze clinopyroxenes in ANS ophiolitic harzburgites using an ion micro-probe to determine compositions of REE and other trace elements in equilibrium liquids.This would allow a more explicit linkage between the composition of ophiolitic lavas andthe extent of melting in the associated mantle section to be examined. Another very usefulstrategy would be to identify and study inclusions in spinels, which have recently beendiscovered to preserve primary magmatic phases.

One of the most interesting unresolved problems concerns the nature of the carbonatealteration of especially the ultramafics. Where did all this CO2 come from? Is there anyrelationship between this alteration and gold mineralization in the ANS? Several workerssuggest a link between more depleted, boniniticophiolites and gold mineralization; is thererelationship between gold mineralization in the ANS and ophiolite type? Modern regionsof seafloor spreading often are associated with hydrothermal vents and associated mineral-ization and biota, are such associations preserved in ANS ophiolites? Such studies may beespecially important because ANS ophiolites may have formed at a time when a variety ofindependent geologic observations suggest that the surface of the earth was covered withice (Evans, 2000), and life may have been restricted to such vents. Similarly, the associ-ation of banded iron formations overlying ANS ophiolites (Sims and James, 1982) mayreflect interactions between deep water and hydrothermal activity during a snowball earthepisode.

Another aspect of ANS ophiolites worthy of study is a possible relationship betweeneconomic chromite deposits and lava compositions. In some ophiolites (e.g., Zambales,Philippines) the affinity of the magmatic section is a reliable guide to the grade of associ-ated chromite deposits (Evans and Hawkins, 1989).

8. CONCLUSIONS

The last three decades of research on ANSophiolites allows us to sketch the broadoutlines of how these complexes formed, but most of this work was completed a decadeor more ago. New research initiatives that focus on ANS ophiolites are needed and thesepromise to be rewarding. A few examples of what needs to be done are presented below.

ANS ophiolites change style on either side of the Bi’r Umq-Nakasib suture zone. Arethese correlative? We are not certain that the allochthonous mafic-ultramafic complexes ofEritrea and Ethiopia should be considered asophiolites, although we have argued for this.Compositions of spinels in harzburgite are a powerful tool for understanding the magmaticevolution and tectonic setting of ophiolites, but we have too few analyses for ophioliticharzburgites from Egypt, Eritrea, and Ethiopia. Similarly, we need more mineral chemicaldata for primary ophiolite phases (olivine, clinopyroxene, plagioclase). We need new cam-paigns of petrologic and geochemical studies using the most modern analytical techniques

Acknowledgements 123

(ICP-MS, etc.) to determine the diagnostictrace element contents of ANS ophiolitic lavas.Some modern analytical techniques, such as using ion probe analyses of clinopyroxenes toindependently infer trace element compositions of equilibrium liquids, have not yet beenapplied. More isotopic studies, especially Nd, Hf, and Os, are needed to understand theevolution of the mantle source region for these lavas. Similarly, we need more geochrono-logical constraints on the age of ANS ophiolites. Ion probe ages of zircons would be es-pecially useful because this best identifies inherited components but this technique has notbeen applied to ANS ophiolites. Precise determination of ophioliteage will also be criti-cal for studies of deep water marine sequences immediately overlying the pillow lavas, aresearch direction that is needed to understand global change during Neoproterozoic time.

We should identify the most complete ANS ophiolites and study them in detail. Interdis-ciplinary study of the best-preserved ophiolites by teams of structural geologists, petrolo-gists, geochemists, and geochronologists focusing on the ophiolite itself in tandem withsedimentologists, geochemists, and paleontologists studying the overlying sedimentarysuccession should be encouraged. Efforts should be made to identify ancient hydrother-mal vent deposits and investigate the associated fossilized biota.

Economic considerations also favor renewed study of ANS ophiolites. The high gradeof ANS chromites may be rich enough to mine. Given the observation that ANS gold min-eralization often appears to be related tocarbonate alteration of ANS ophiolitic peridotites,we can expect to better understand the former by focused studies of the latter. At presentwe have a very poor understanding of pervasive carbonate alteration of ANS ophioliticperidotites. Research programs should be designed to better understand the age of thisalteration and the origins of these fluids.

ACKNOWLEDGEMENTS

Thanks to T. Tadesse (Ethiopian Geological Survey) and B. Woldhaimanot (U. Asmara)for their thoughts on the ‘ophiolites’ of Eritreaand Ethiopia. Thanks to J. Encarnacion andY. Dilek for thoughtful reviews. This manuscript is dedicated to the memory of Ian Gass,who stimulated so many excellent studies ofANS ophiolites and whose enthusiasm for theophiolites of the Arabian-Nubian Shield has yet to be matched.

REFERENCES

Abdel-Rahman, E.M., et al., 1990. A new ophioliteoccurrence in NW Sudan—constraints on lateProterozoic tectonism. Terra Nova 2, 363–376.

Abdel-Rahman, M., 1993. Geochemical and geotectonic controls of the metallogenic evolution ofselected ophiolite complexes from the Sudan. Berliner geowissenschaftliche Abhandlungen 145,175.

Abdelsalam, M.G., Stern, R.J.,1993. Tectonic evolution of the Nakasib suture, Red Sea Hills, Sudan:evidence for a late Precambrian Wilson cycle. Journal of the Geological Society of London 150,393–404.


Abdelsalam, M.G., Stern, R.J., 1996. Sutures and Shear Zones in the Arabian-Nubian Shield. Journalof African Earth Sciences 23, 289–310.

Ahmed, A.H., Arai, S., Attia, A.K., 2001. Petrological characteristics of podiform chromitites andassociated peridotites of thePan African ophiolite complexes of Egypt. Mineralium Deposita 36,72–84.

Akaad, M.K., Noweir, A.M., 1972. Some Aspects of the Serpentinites and their Associated Deriva-tives along Qift-Quesir Road, Eastern Desert. Annals of the Geological Survey of Egypt 2, 251–270.

Al-Rehaili, M.H., Warden, A.J., 1980. Comparison of the Bi’r Umq and Hamdah ultrabasic com-plexes, Saudi Arabia. Institute ofApplied GeologyBulletin 3 (4), 143–156.

Al-Saleh, A.M., Boyle, A.P., Mussett, A.E., 1998. Metamorphism and40Ar/39Ar dating of the Ha-laban ophiolite and associated units: evidence fortwo-stage orogenesis in the eastern Arabianshield. Journal of the Geological Society of London 155, 165–175.

Al-Shanti, A.M., El-Mahdy, O.R., 1988. Geological Studies and Assessment of Chromite Occur-rences in Saudi Arabia. KACST Project No. AT-6-094.

Al-Shanti, M.M.S., 1982. Geology and mineralization of the Ash Shizm-Jabal Ess area. Ph.D. thesis.King Abdulaziz University, p. 291.

Augustithus, S.S., 1965. Mineralogical and geochemical studies of the platiniferous dunite-birbirite-pyroxenite complex of Yubdo/Birbir, W. Ethiopia. Chemie der Erde 24, 159–196.

Bakor, A.R., Gass, I.G., Neary, C.R., 1976. Jabal al Wask, northwest Saudi Arabia: an Eocambrianback-arc ophiolite. Earth and Planetary Science Letters 30, 1–9.

Barnes, S.J., Roeder, P.L., 2001. The Range of Spinel Compositions in Terrestrial Mafic and Ultra-mafic Rocks. Journal of Petrology 42, 2279–2302.

Basta, E.Z., Kader, Z.A., 1969. The mineralogy of Egyptian serpentinites and talc-carbonates. Min-eralogical Magazine 37, 394–408.

Berhe, S.M., 1988. The Geologic and Tectonic Evolution of the Pan African/Mozambique Belt inEast Africa. Ph.D. thesis. Open University, Milton Keynes, p. 265.

Bloomer, S.H., Taylor, B., MacLeod, C.J., Stern, R.J., Fryer, P., Hawkins, J.W., Johnson, L., 1995.Early Arc Volcanism and the Ophiolite Problem: A Perspective from Drilling in the WesternPacific. In: Taylor, B., Natland, J. (Eds.), Active Margins and Marginal Basins of the WesternPacific. American Geophysical Union, Washington, DC, pp. 1–30.

Bonatti, E., Michael, P.J., 1989. Mantle peridotitesfrom continental rifts to ocean basins to subduc-tion zones. Earth and Planetary Science Letters 91, 297–311.

Bonavia, F.F., Diella, V., Ferrario, A., 1993. Precambrian Podiform Chromitites from Kenticha Hill,southern Ethiopia. Economic Geology 88, 198–202.

Chevremont, P., Johan, Z., 1982a. The Al Aysophiolite complex. BRGM-OF-02-5. Saudi ArabiaDeputy Ministry for Mineral Resources.

Chevremont, P., Johan, Z., 1982b. Wadi al Hwanet-Jabal Iss ophiolite complex. BRGM-OF-02-14.Saudi Arabia Deputy Ministry for Mineral Resources.

Church, W.R., 1988. Ophiolites, structures, and micro-plates of the Arabian-Nubian shield: a crit-ical comment. In: Greiling, S.E.-G.a.R.O. (Ed.), The Pan-African Belt of Northeast Africa andAdjacent Areas. Veiweg, Braunschweig/Wiesbaden, pp. 289–316.

Claesson, S., Pallister, J.S., Tatsumoto, M., 1984. Samarium-neodymium data on two late Proterozoicophiolites of Saudi Arabia and implications for crustal and mantle evolution. Contributions toMineralogy and Petrology 85, 244–252.

References 125

Cloos, M., 1993. Lithospheric buoyancy and collisional orogenesis: Subduction of oceanic plateaus,continental margins, island arcs, spreading ridges, and seamounts. Geological Society of AmericaBulletin 105, 715–737.

Cox, D.P., Singer, D.A., 1986. Grade and tonnage model of porphyry Cu-Au. In: Cox, D.P.,Singer, D.A. (Eds.), Mineral Deposit Models. U.S. Geological Survey Bulletin, 110–114.

Crawford, A.J., Falloon, T.J., Green, D.H., 1989. Classification, petrogenesis and tectonic settingof boninites. In: Crawford, A.J. (Ed.), Boninites and Related Rocks. Cambridge Univ. Press,Cambridge, pp. 1–49.

Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, envi-ronmental speculation. Geological Society of America Bulletin 109, 16–42.

Dick, H.J.B., Bullen, T., 1984. Chromian spinel as a petrogenetic indicator in abyssal and alpine-typeperidotites and spatially associated lavas. Contributions to Mineralogy and Petrology 86, 54–76.

Dilek, Y., Moores, E.M., Furnes, H., 1998. Structure of modern oceanic crust and ophiolites andimplications for faulting and magmatism at oceanicspreading centers. In: Buck, P.T.D., Karson,J.A., Lagabrielle, Y. (Eds.), Faulting and Magmatism at Mid-Ocean Ridges. AGU, Washington,DC, pp. 219–265.

Dixon, T.H., 1981. Gebel dahanib, a late Precambrian layered sill of komatiiticcomposition. Contri-butions to Mineralogy and Petrology 76, 42–52.

Drury, S.A., Filho, C.R.d.S., 1998. Neoproterozoic terrane assemblages in Eritrea: review andprospects. Journal of African Earth Sciences 27, 331–348.

El-Bayoumi, R.M., 1983. Ophiolites and mélange complex of Wadi Ghadir Area, Eastern Desert,Egypt. Bulletin of the Faculty of Earth Sciences, King Abdulaziz University, Jiddah 6, 329–342.

El-Naby, H.A., Frisch, W., 1999. Metamorphic sole of Wadi Haimur-Abu Swayel ophiolite: Implica-tions on late Proterozoic accretion. In: Wall, H.D., Greiling, R.O. (Eds.), Aspects of Pan-AfricanTectonics. Heidelberg, Germany, pp. 9–14.

El-Sayed, M.M., Furnes, H., Mohamed, F.H., 1999. Geochemical constraints on the tectonomagmaticevolution of the late Precambrian Fawakhir ophiolite, Central eastern Desert, Egypt. Journal ofAfrican Earth Sciences 29, 515–533.

Evans, C., Hawkins, J.W., 1989. Compositional heterogeneities in upper mantle peridotites from theZambales Range Ophiolite, Luzon, Philippines. Tectonophysics 168, 23–41.

Evans, D., 2000. Stratigraphic, geochronological, and paleomagnetic constraints upon the Neopro-terozoic climatic paradox. American Journal of Science 300, 347–433.

Fitches, W.R., et al., 1983. The Late Proterozoic ophiolite of Sol Hamed, NE Sudan. PrecambrianResearch 19, 515–533.

Halls, C., Zhao, R., 1995. Listwaenite and related rocks: perspectives on terminology and mineralogywith reference to an occurrence at Cregganbaun, Co. Mayo, Republic of Ireland. MineraliumDeposita 30, 303–313.

Harms, U., Darbyshire, D.P.F., Denkler, T., Hengst, M., Schandelmeier, H., 1994. Evolution of theNeoproterozoic Delgo suture zone and crustal growth in Northern Sudan: Geochemical and radi-ogenic isotope constraints. Geologische Rundschau 83, 591–603.

Henjes-Kunst, F., Altherr, R., Baumann, A., 1994.Evolution and composition of the lithosphericmantle underneath the western Arabian Peninsula; constraints from Sr-Nd isotope systematics ofmantle xenoliths. Contributions toMineralogy and Petrology 105, 460–472.

Hussein, I.M., 2000. Geodynamic evolution of the Pan-African crystalline basement in the northernRed Sea Hills, Sudan, with special emphasis on the Wadi Onib ophiolite and geology to the westof Port Sudan. Joahnnes Gutenberg Universität-Mainz, Mainz.


Hussein, I.M., Kröner, A., Durr, S., 1983. Wadi Onib-A Dismembered Pan-African Ophiolite in theRed Sea Hills of Sudan. Bulletin of the Faculty ofEarth Sciences, King Abdulaziz University,Jiddah 6, 320–327.

Hussein, I.M., Kröner, A., Reischmann, T., this volume. The Wadi Onib mafic-ultramafic complex:a Neoproterozoic supra-subduction zone ophiolite in the northern Red Sea Hills of the Sudan.

Johnson, P.R., Kattan, F.H., Al-Saleh, A.M., 2004.Neoproterozoic ophiolites in the Arabian Shield:Field relations and structure. In: Kusky, T.M.(Ed.), Precambrian Ophiolites and Related Rocks.In: Developments in Precambrian Geology, vol. 13. Elsevier, Amsterdam, pp. 129–162.

Kattan, F.H., 1983. Petrology and Geochemistry of the Tuluhah Belt, North East Arabian Shield.M.S. thesis. King Abdulaziz University, Jiddah.

Knoll, A.H., 2000. Learning to tell Neoproterozoic Time. Precambrian Research 100, 3–20.Kröner, A., et al., 1987. Pan-African Crustal Evolution in the Nubian segment of Northeast Africa. In:

Kröner, A. (Ed.), Proterozoic Lithosphere Evolution. American Geophysical Union, Washington,DC, pp. 235–257.

Kröner, A., Todt, W., Hussein, I.M., Mansour, M., Rashwan, A.A., 1992. Dating of late Proterozoicophiolites in Egypt and Sudan using the single grainzircon evaporation technique. PrecambrianResearch 59, 15–32.

Kusky, T.M., Abdelsalam, M., Tucker, R., Stern, R., 2003. Evolution of The East African and RelatedOrogens, and the Assembly of Gondwana. Special Issue of Precambrian Research 123, 81–344.

Ledru, P., Auge, T., The Al Ays ophiolitic complex; petrology and structural evolution. BRGM-OF-04-15. Saudi Arabia Deputy Ministry for Mineral Resources.

LeMetour, J., Johan, V., Tegyey, M., Relationships between ultramafic-mafic complexes and vol-canosedimentary rocks in the Precambrian Arabian Shield. BRGM-OF-12-15. Saudi ArabiaDeputy Ministry for Mineral Resources.

McGuire, A.V., Stern, R.J., 1993. Granulite xenoliths from western Saudi Arabia: The lower crustof the Late Precambrian Arabian-Nubian Shield. Contributions to Mineralogy and Petrology 114,395–408.

Meert, J.G., 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectono-physics 362, 1–40.

Miyashiro, A., 1975. Classification, Characteristics, and Origin of Ophiolites. Journal of Geology 83,249–281.

Mouhamed, M.F.S., 1995. Geology, Geochemistry and Structure of Gabal Muqsim Area and Envi-rons, South Eastern Desert, Egypt. Forschungszentrum Jülich, Jülich.

Nassief, M.O., Macdonald, R., Gass, I.G., 1984.The Jebel Thurwah upper Proterozoic ophiolitecomplex, western Saudi Arabia. Journal of the Geological Society of London 141, 537–546.

Newton, R.C., Stern, R.J., 1990. A Late Precambrian CO2 “Event”. Geological Society of America,Dallas, p. 190.

Ohara, Y., Stern, R.J., Ishii, T., Yurimoto, H., Yamazaki, T., 2002. Peridotites from the MarianaTrough: First look at the mantle beneath an active backarc basin. Contributions to Mineralogyand Petrology 143, 1–18.

Pallister, J.S., Stacey, J.S., Fischer, L.B., Premo, W.R., 1988. Precambrian ophiolites of Arabia: Ge-ologic settings, U-Pb geochronology, Pb-isotope characteristics, and implications for continentalaccretion. Precambrian Research 38, 1–54.

Patchett, P.J., Chase, C.G., 2002. Role of transform continental margins in major crustal growthepisodes. Geology 30, 39–42.

Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plate boundaries. In:Thorpe, R.S. (Ed.), Andesites. Wiley, Chichester, pp. 525–548.

References 127

Pearce, J.A., Cann, J.R., 1973. Ophiolite origin investigated by discriminant analysis. Earth andPlanetary Science Letters 12, 339–349.

Pearce, J.A., Lippard, S.J., Roberts, S., 1984. Characteristics and tectonic significance of supra-subduction zone ophiolites. In: Kokelaar, B.P., Howell, M.F. (Eds.), Marginal Basin Geology.Geological Society of London, pp. 77–94.

Pearce, J.A., Norry, M.J., 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanicrocks. Contributions to Mineralogy and Petrology 69, 33–47.

Price, R.C., 1984. Late Precambrian Mafic-Ultramafic Complexes in Northeast Africa. Ph.D. thesis.Open University, Milton Keynes, p. 325.

Quick, J.E., 1990. Geology and origin of the LateProterozoic Darb Zubaydah ophiolite, Kingdom ofSaudi Arabia. Geological Society of America Bulletin 102, 1007–1020.

Quick, J.E., 1991. Late Proterozoic transpression of the Nabitah fault system—implications for theassembly of the Arabian shield. Precambrian Research 53, 119–147.

Quick, J.E., Bosch, P.S., 1989. Tectonic history of the northern Nabitah fault zone, Arabian Shield,Kingdom of Saudi Arabia. USGS-TR-08-2. Saudi Arabia Directorate General of Mineral Re-sources.

Rittmann, A., 1958. Geosynclinal volcanism, Ophiolites and Barramiya rocks. The Egyptian Journalof Geology 2, 61–66.

Schandelmeier, H., et al., 1994. Atmur-Delgo suture: a Neoproterozoic oceanic basin extending intothe interior of northeast Africa. Geology 22, 563–566.

Shanti, M., 1983. The Jabal Ess ophiolite complex. Bulletin of the Faculty of Earth Sciences, KingAbdulaziz University, Jiddah 6, 289–317.

Shanti, M., Roobol, M.J., 1979. A late Proterozoic ophiolite complex at Jabal Ess in northern SaudiArabia. Nature 279, 488–491.

Shervais, J.W., 1982. Ti-V plot and the petrogenesis of modern and ophiolitic lavas. Earth and Plan-etary Science Letters 59, 101–118.

Shervais, J.W., 2001. Birth, death,and resurrection; the life cycle of suprasubduction zone ophiolites.Geochemistry, Geophysics, Geosystems 2, 2000GC000080.

Sims, P.K., James, H.L., 1982. Banded iron-formations of late Proterozoic age in the central EasternDesert, Egypt; geologyand tectonic setting.Economic Geology 79, 1777–1784.

Stacey, J.S., Stoeser, D.B., Greenwood, W.R., Fischer, L.B., 1984. U-Pb zircon geochronology andgeological evolution of the Halaban-Al Amar region of the Eastern Arabian Shield, Kingdom ofSaudi Arabia. Journal of the Geological Society of London 141, 1043–1055.

Stein, M., Goldstein, S.L., 1996. From plume head tocontinental lithospherein the Arabian-NubianShield. Nature 382, 773–778.

Stern, R.J., 1981. Petrogenesis andtectonic setting of late Precambrian enzymatic volcanic rocks,Central Eastern Desert of Egypt. Precambrian Research 16, 195–230.

Stern, R.J., 1994. Arc Assembly and continental collision in the Neoproterozoic East African Oro-gen: Implications for the assembly of Gondwanaland. Annual Reviews of Earth and PlanetarySciences 22, 319–351.

Stern, R.J., 2002. Crustal evolution in the East African Orogen: a neodymium isotopic perspective.Journal of African Earth Sciences 34, 109–117.

Stern, R.J., Gwinn, C.J., 1990. Origin of Late Precambrian Intrusive Carbonates, Eastern Desert ofEgypt and Sudan: C, O, and Sr Isotopic Evidence. Precambrian Research 46, 259–272.

Stern, R.J., et al., 1990. Orientation of late Precambrian sutures in the Arabian-Nubian shield. Geol-ogy 18, 1103–1106.


Sultan, M., Arvidson, R.E., Duncan, I.J., Stern, R.J., Kaliouby, B.E., 1988. Extension of the Najdshear system from Saudi Arabia to the central Eastern Desert of Egypt based on integrated fieldand Landsat observations. Tectonics 7, 1291–1306.

Sultan, M., Arvidson, R.E., Sturchio, N.C., 1986. Mapping of serpentinites in the Eastern Desert ofEgypt by using Landsat thematic mapper data. Geology 14, 995–999.

Surour, A.A., Arafa, E.H., 1997. Ophicarbonates: calcified serpentinites from Gebel Moghara, WadiGhadir area, Eastern Desert, Egypt. Journal of African Earth Sciences 24, 315–324.

Takla, M.A., 1991. Chloritites at the contacts of some ophiolitic ultramafics, Eastern Desert, Egypt.Egyptian Mineralogist 3, 151–165.

Takla, M.A., Basta, F.F., Surour, A.A., 1992. Petrology and Mineral Chemistry of Rodingites as-sociating the Pan-African Ultramafics of Sikait-Abu Rusheid area, South Eastern Desert, Egypt.Geology of the Arab World, 491–507.

Walker, R.J., Prichard, H.M., Ishiwatari, A.,Pimentel, M., 2002. The Osmium isotopic compositionof convecting upper mantle deduced from ophiolite chromites. Geochimica et CosmochimicaActa 66, 329–345.

Wolde, B., Asres, Z., Desta, Z., Gonzalez, J.J., 1996. Neoproterozoic zirconium-depleted boniniteand tholeiitic series rocksfrom Adola, southern Ethiopia. Precambrian Research 80, 261–279.

Woldehaimanot, B., 2000. Tectonic setting and geochemical characterization of Neoproterozoic vol-canics and granitoids from the Adobha Belt, northern Eritrea. Journal of African Earth Sci-ences 30, 817–831.

Woldehaimanot, B., Behrmann, J.H., 1995. A study of metabasite and metagranite chemistry in theAdola region (south Ethiopia): Implications for the evolution of the East African orogen. Journalof African Earth Sciences 21, 459–476.

Worku, H., 1996. Geodynamic Development of the Adola Belt (Southern Ethiopia) in the Neopro-terozoic and Its Control on Gold Mineralization. Verlag Dr. Köster, Berlin, p. 156.

Yibas, B., Reimold, W.U., Anhaeusser, C.R., Koeberl, C., 2003. Geochemistry of the mafic rocks ofthe ophiolitic fold and thrust belts of southern Ethiopia: constraints on the tectonic regime duringthe Neoproterozoic (900–700 Ma). Precambrian Research 121, 157–183.

Yihunie, T., 2002. Pan-African deformations in the basement of the Negele area, southern Ethiopia.International Journal of Earth Science 91, 922–933.

Zimmer, M., Kröner, A., Jochum, K.P., Reischmann, T., Todt, W., 1995. The Gabal Gerf complex:A Precambrian N-MORB ophiolite in the Nubian Shield, NE Africa. Chemical Geology 123,29–51.

ROBERT J. STERN , PETER R. JOHNSON , ALFRED ...rjstern/pdfs/ANSOph...aGeosciences Department,...

Documents

Transcript of ROBERT J. STERN , PETER R. JOHNSON , ALFRED ...rjstern/pdfs/ANSOph...aGeosciences Department,...