Late- and post-orogenic Neoproterozoic intrusions of ...rjstern/pdfs/Jarraretal.pdfsense of...

Precambrian Research 123 (2003) 295–319

Late- and post-orogenic Neoproterozoic intrusions of Jordan:implications for crustal growth in the northernmost

segment of the East African Orogen

G. Jarrara,∗, R.J. Sternb, G. Saffarinia, H. Al-Zubi c

a Geology Department, University of Jordan, 11942 Amman, Jordanb Geosciences Department, University of Texas at Dallas, P.O. Box 830688, 2601 Floyd Rd., Richardson, TX 75083-0688, USA

c Natural Resources Authority, P.O. Box 7, 11118 Amman, Jordan

Received 12 June 2001; received in revised form 25 November 2001; accepted 25 November 2001

Abstract

Approximately 90% of the upper crust exposed in southern Jordan is composed of intrusive rocks of latest Neoproterozoic age,grouped into two major subdivisions: the Aqaba (600–640 Ma) and the Araba (560–600 Ma) complexes. The Aqaba complexcomprises several suites that range in composition from gabbro to high silica granite (45–80% SiO2) and follow a high-Kcalc-alkaline trend. This phase, which started with the emplacement of the Duheila Hornblendic Suite between 640 and 600 Ma,represents the main crust-forming stage in southwest Jordan. The Araba complex is a bimodal alkali-calcic to alkali igneoussuite generated after development of a regional unconformity and deposition of the Saramuj Conglomerate. Mafic members ofthe Aqaba complex, the Duheila Hornblendic Suite, are enriched in LILE relative to the HFSE and are moderately enrichedin REE [(La/Lu)n = 5–11], traits typical of arc basalts. Geochemical modeling suggests derivation by 10–15% melting ofamphibole-bearing spinel lherzolite, possibly above a subduction zone. The granitoids of the Aqaba complex are high in Ba, Srand LREE, have low Y and have steep REE patterns [(La/Lu)n = 20–25]. They have low initial87Sr/86Sr (∼0.70305) and highεNd

values (+2.3 to+5.0). These may have been generated by high degrees of partial melting (10–30%) of subducted oceanic crust,with or without a small proportion of ocean sediments. The mafic and intermediate end members of the Araba complex, the ArabaMafic to Intermediate Suite (48–65% SiO2) comprises quartz-diorites, quartz-monzodiorites, monzodiorites and monzogabbros.The mafic end member of this suite is enriched in LILE compared to the Duheila Hornblendic Suite, although its primitivecompositions have similar REE patterns [(La/Lu)n = 5–15]; and has been emplaced in a within-plate tectonic environment.Geochemical modeling supports formation of the Araba Mafic to Intermediate Suite by 10% partial melting of phlogopite-bearingspinel lherzolite. This hypothesis is supported by low initial87Sr/86Sr (∼0.7035) and highεNd values (+2.4 to+4.2). The felsicend member of the Araba complex, the Humrat–Feinan Suite (68–80% SiO2), is characterized by moderate enrichments inLREE (La/Sm)n ∼ 2, flat HREE patterns(Gd/Lu)n ∼ 1, and strong negative Eu anomalies (Eu/Eu∗ =∼ 0.07–0.53). Thesegranites have geochemical features typical of A-type granite and can be derived by extreme fractional crystallization (∼80%)of the Araba Mafic to Intermediate Suite. The low initial87Sr/86Sr (∼0.7048), highεNd values (+2.5 to+4.8), and spatial andtemporal association with the Araba Mafic Suite support this model. Crustal evolution in the northernmost Arabian NubianShield occurred about the time of terminal collision between east and west Gondwanaland by the addition of magmas derivedfrom mantle-derived melts. These melts changed markedly at about the time of collision, from a unimodal suite of convergent

∗ Corresponding author.E-mail address:[email protected] (G. Jarrar).

0301-9268/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved.doi:10.1016/S0301-9268(03)00073-1

296 G. Jarrar et al. / Precambrian Research 123 (2003) 295–319

margin magmas characterized by deep (garnet-controlled) fractionation, to a bimodal suite of rift-related magmas characterizedby shallow (feldspar-controlled) fractionation.© 2003 Elsevier Science B.V. All rights reserved.

Keywords:Pan-African; Arabian Nubian Shield; Post-collision granitoids; Jordan

1. Introduction

The East African Orogen, referred to hereafter asEAO, is a major suture along which the Gondwanasupercontinent was assembled (Stern, 1994; Unrug,1996). The tectonic evolution of the EAO lasted from900 to 550 Ma and included opening and closing of theMozambique Ocean and terminal collision betweenolder crustal blocks of East and West Gondwana at600± 50 Ma ago. The Arabian Nubian Shield (ANS)formed during this episode and provides an outstand-ing example of the formation of juvenile continentalcrust by plate tectonic processes. The accretion of ju-venile island arcs and micro-continental fragments isan important part of this process, but of equal impor-tance to our understanding of how this tract of conti-nental crust formed is how the intrusive rocks formed.Granite-granodiorite batholiths form approximately65% of the exposed ANS (Stoeser, 1986) and are espe-cially abundant in the north. Because Jordan containsthe northernmost outcrops of the ANS, understand-ing the evolution of Jordanian granitoids and relatedrocks promises to provide important insights intohow continental crust forms and evolves. Late- andPost-orogenic Neoproterozoic Intrusions of Jordan(LPONIJ) also provide a largely unrecognized ana-logue for the development of Mesoproterozoic anoro-genic granitoids common in North America andEurope (Windley, 1993).

Basement exposures in Jordan form the northern ex-tension of the Midyan terrain (Fig. 1), which can betraced farther west into Israel and Sinai. Neoprotero-zoic basement in Jordan crops out east and northeast ofAqaba and along the eastern shoulder of Wadi Araba asfar north as the southern end of the Dead Sea, coveringan area of about 1400 km2 (Bender, 1974). This com-plex is comprised of igneous, sedimentary and meta-morphic rocks and is exclusively of Neoproterozoicage (Brook et al., 1990; Ibrahim and McCourt, 1995;Jarrar et al., 1991, 1992, 1993; Jarrar, 1998). Meta-morphic units are 750–800 Ma old (Jarrar, 1985) and

can be correlated with units in Israel and Sinai (Kröneret al., 1990), while igneous rocks range in age from640 to 540 Ma (Jarrar et al., 1983; Jarrar, 1985; Brooket al., 1990). Almost 90% of the exposed basement inJordan is made up of intrusives with granitoids (in thesense ofStreckeisen, 1976) forming the vast majority.

This overview paper compiles and synthesizes ex-isting geological, geochemical and geochronologicaldata on the LPONIJ. We address the petrogeneticevolution of these granitoids and the implications ofLPONIJ for our understanding of crust formation inthe northern part of the East African Orogen.

2. Geological setting

The LPONIJ intrude high-grade Pan-Africanmetasediments (upper amphibolite–lower granulitefacies) that have been metamorphosed at 750–800 Ma(Jarrar, 1985). The Janub Metamorphic Suite has notyet been dated (cf.Table 1). However, cross-cuttingfield relationships with country rocks of the Aqabacomplex suggest a similar age to that of Abu-BarqaMetamorphic Suite (750–800 Ma). Nevertheless, theJanub Suite, which is of mainly greenschist fa-cies suite, represents a higher crustal level than theAbu-Barqa Suite. On the basis of detailed geologicalmapping, petrographic and geochronological data,plutonic rocks exposed in southern Jordan have beensubdivided into the Aqaba and Araba complexes(McCourt and Ibrahim, 1990; Ibrahim and McCourt,1995). The two complexes are separated by a regionalunconformity overlain by the Saramuj Conglomerateand which is dated in Jordan (Jarrar et al., 1993)and elsewhere in the ANS at about 600 Ma. Thisunconformity marks the beginning of an extensionaltectonic phase that lasted until about 545 Ma. TheAqaba and Araba complexes have been subdividedinto several suites (Fig. 1). Table 1is a modified andupdated hierarchy of events in southern Jordan inthe Neoproterozoic. Several plutons inIbrahim and

G. Jarrar et al. / Precambrian Research 123 (2003) 295–319 297

Fig. 1. The distribution of Neoproterozoic intrusions of Jordan in relation to the Arabian Nubian Shield. Modified afterBender (1974),Ibrahim and McCourt (1995), andJarrar et al. (1992, 1993).


Table 1Lithostratigraphic units and the main rocks suites of the Neoproterozoic in southwest Jordan

The last column is a concise description of field and petrographic aspects. The Aqaba and Araba complexes are separated by a regionalunconformity at about 600 Ma. The ages in italics given for the suites are U–Pb zircon ages (Jarrar, 1985; Jarrar et al., 1991, 1992,1993) while the underlined ones are based on Rb–Sr isochrons (Brook et al., 1990). The data in this table is based in first place on thelithostratigraphic classification ofMcCourt and Ibrahim (1990); first author’s own work on basement complex in southwest Jordan.


McCourt’s (1995)scheme have been placed in theAraba complex rather than the Aqaba complex.

The Duheila Hornblendic Suite (DHS) is 611±9 Ma according to K–Ar dating (Lenz et al., 1972).The rocks of this suite show petrographic and fieldsimilarities to the metagabbros and metadiorites ofthe Shahmon metabasites (north of Elat), which havebeen dated by single-zircon evaporation method at650–640 Ma (Kröner et al., 1990). We consider the lat-ter to be a more reliable estimate for the emplacementage of the DHS since the K–Ar systematics of theregion were reset about 600 Ma (Cosca et al., 1999).On the basis of these arguments, the DHS should beplaced below the Rahma Suite and above the Januband Abu-Barqa Metamorphic Suites.

An adamellite granite that outcrops in the area ofUm Saiyala as part of the Rahma Suite (Fig. 1D)has been dated by U–Pb on zircons at 626± 3 Ma(Jarrar, 1985). This adamellite has been transformedinto a granitic gneiss (lineation as defined by elon-gated feldspars trends ENE;Jarrar, 1985) shortly af-ter its intrusion since it is present as xenoliths in theUm Saiyala porphyritic granodiorite (Table 1, RahmaSuite). The latter granodiorite also contains blasto-mylonite xenoliths. Moreover, several E–W mylonitezones were found affecting Aqaba granitoids, whichare in turn cut by a variety of dikes (Hassuneh, 1994).Noteworthy is that these mylonites did not affect thegranitoids of the younger Araba complex granites. TheYutum Suite has strong chemical affinities with Arabacomplex granitoids, but it is classified with the Aqabacomplex on the basis of field evidence—it is dis-sected by the same dike generation like other suites ofthe Aqaba complex—and has a zircon age of 608 Ma(Jarrar, 1985).

The Feinan, Humrat, and Mubarak alkali feldspargranites are grouped and renamed as the Humrat–Feinan Suite, while the Qunaia and Um Rachel monzo-diorites and monzogabbros together with the GhuweirMafics (Jarrar et al., submitted for publication) arecalled Araba Mafic to Intermediate Suite (AMIS). Thatis the mafic and intermediate members of the ArabaAlkaline Bimodal Plutonic Suite. It occurs in smalloutcrops along the eastern shoulder of Wadi Araba inthe area of Wadi Qunaia, where it intrudes the SaramujConglomerate at the southeast corner of the Dead Sea(Jarrar et al., 1993), along Wadi Ghuweir in the Feinanarea forming the so-called Ghuweir Mafics (Jarrar

et al., submitted for publication), and in Wadi UmRachel, about 80 km north of Aqaba. This suite hasbeen dated by U/Pb and Rb/Sr methods at 595–572 Ma(Jarrar et al., 1993; Jarrar, 1985; Jarrar et al., submitt-ed and unpublished data).

The Humrat–Feinan Suite is the felsic end mem-ber of the Araba Alkaline Plutonic Bimodal Suite andconsists predominantly of reddish microperthitic alkalifeldspar and quartz with occasional biotite. Volumet-rically these granites are insignificant. Rb/Sr isochronages for both the Araba granites and the mafic rockscluster around 570 Ma. However, U–Pb zircon agesfor the mafic rocks give ages as old as 595 Ma. Sinceboth mafic and felsic end members are cut by thesame generation of dikes, i.e. the dolerites (548 Ma;Jarrar et al., 1992; Jarrar, 2001), and intrude the Sara-muj Conglomerate, it is reasonable to treat them ascontemporaneous.Table 1includes a summary of themain field and petrographic characteristics and agesof the various suites.

3. Geochemistry

The database used in this study is compiled (254 fulland partial chemical analyses) from the author’s ownunpublished and published data; unpublished M.Sc.theses ofAbdel-Fattah (1985),El-Hilu (1995),Al-Zubi(1996), Habboush (1996), Ibrahim and McCourt(1995), and Brook et al. (1990). Chemical and iso-topic data for Um Rachel Mafic and intermediaterocks (part of Araba Mafic to Intermediate Suite) wereobtained by the first author while spending a sabbati-cal leave at the Geosciences Department, UT Dallas,TX. Methods of analysis are detailed inJarrar et al.(submitted to International Journal of Earth Science).

3.1. Major elements

Table 2contains ranges and averages of the vari-ous suites and the data are presented graphically inFigs. 2–13. A full listing of the raw data is availablefrom the first author upon request. In the forthcomingsections the abbreviation LPONIJ or the Aqaba com-plex (including DHS and the Suites of Intermediateand felsic intrusives) and Araba complex (ArabaMafic to intermediate Suite and Humrat–Feinan Suite)will be used.

300G

.Ja

rrar

et

al./P

reca

mb

rian

Re

sea

rch1

23

(20

03

)2

95

–3

19

Table 2Chemical data for the investigated rocks expressed in range and mean for the different suites

Aqaba complex Araba complex

Duheila HornblendicSuite (10)

Rahma Suite (18) Darba Suite (31) Rumman Suite (38) Urf Suite (6) Yutum Suite (59) Araba Mafic toIntermediate Suite (31)

Humrat–FeinanSuite (63)

Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean Range Mean

SiO2 44.90 55.49 49.30 69.50 75.37 72.44 65.19 75.86 70.80 54.50 79.60 65.16 70.51 71.93 71.12 63.10 76.68 70.79 48.00 64.60 54.59 68.00 79.69 75.78TiO2 0.56 2.17 1.30 0.13 0.42 0.22 0.01 0.77 0.28 0.19 1.38 0.82 0.21 0.38 0.31 0.10 0.74 0.36 0.62 2.46 1.16 0.04 0.42 0.15Al2O3 9.31 18.11 14.70 12.79 16.38 15.13 13.76 15.64 14.82 11.70 17.15 14.35 14.49 15.39 14.99 12.41 16.11 14.66 14.12 21.90 17.19 11.48 16.42 12.51Fe2O3 2.06 3.72 2.80 0.87 2.60 1.88 0.36 4.33 2.04 1.40 11.60 5.13 1.89 2.94 2.19 0.46 4.78 2.07 3.11 11.83 7.60 0.40 2.72 1.30MnO 0.04 0.17 0.10 0.03 0.17 0.07 0.03 0.33 0.10 0.03 0.25 0.11 0.04 0.07 0.05 0.03 0.19 0.06 0.06 0.60 0.14 0.01 0.16 0.07MgO 2.83 15.20 8.80 0.21 0.71 0.43 0.01 3.11 1.13 0.22 5.20 2.29 0.41 0.76 0.61 0.00 3.39 1.00 2.16 8.98 4.08 0.03 0.86 0.19CaO 6.72 11.22 9.30 0.25 2.42 1.43 0.26 4.45 1.84 0.74 9.20 4.21 1.68 1.83 1.74 0.25 4.44 1.67 2.72 8.54 5.80 0.15 3.20 0.94Na2O 1.13 3.80 2.50 3.40 5.42 4.41 2.70 6.34 4.20 3.94 5.70 4.60 4.12 5.41 4.83 3.51 5.42 4.61 2.39 6.21 3.91 3.27 5.00 4.37K2O 0.56 1.91 1.10 2.14 5.05 3.34 1.73 5.28 3.24 1.62 5.32 2.82 2.84 4.03 3.55 2.58 5.79 4.42 0.76 4.04 2.41 3.41 5.69 4.19P2O5 0.05 0.34 0.20 0.03 0.13 0.06 0.03 0.14 0.06 0.04 0.72 0.39 0.07 0.13 0.10 0.01 0.32 0.15 0.06 1.35 0.49 0.01 0.42 0.06LOI 1.12 1.86 1.50 0.13 1.51 0.59 0.68 2.11 1.11 0.08 0.91 0.52 0.13 0.43 0.28 0.05 0.74 0.41 0.27 6.38 2.32 0.08 1.53 0.57

Cs 0.90 1.20 1.00 1.10 1.10 1.10 0.10 6.40 2.74 1.20 5.10 2.12Rb 12 58 35 35 135 86 67 262 115 31 142 82 77 120 96 63 298 144 15 156 74 70 221 142Ba 198 482 288 169 1180 796 30 958 560 469 1930 883 699 1232 892 12 1810 595 160 2200 808 20 717 227Sr 140 694 430 90 887 525 15 573 324 62 680 443 295 777 644 9 703 256 316 1673 708 11 194 56Pb 10 10 10 11 165 75 12 98 43 3 38 15 16 23 20 2 28 12 5 25 14 4 170 23

U 1 2 1 2 3 2 0 1 0 0 2 1 0 2 1 3 9 4Th 1 2 1 1 11 5 5 11 8 0 17 11 3 14 9 0 20 8 0 13 5 10 41 21Zr 54 150 98 14 215 99 6 115 59 106 981 353 89 148 134 62 194 136 36 295 132 59 652 243Hf 1 1 1 3 5 4 4 6 5 5 7 6 3 8 5 4 6 5 1 7 5 3 11 6Nb 1 22 8 3 13 6 5 27 11 2 21 13 2 9 5 3 46 16 2 29 16 9 65 25Y 13 38 23 2 18 10 7 34 13 12 116 26 8 21 12 13 61 19 8 44 22 13 82 32

La 12.5 48.0 30.6 4.0 33.8 20.1 8.0 17.0 12.5 17.0 140.0 47.2 17.0 23.0 20.0 16.0 57.0 34.6 8.5 61.5 25.9 10.0 99.0 37.3Ce 31.5 71.5 47.3 22.0 70.8 48.0 20.0 46.0 33.1 40.0 266.0 109.2 36.0 58.0 45.8 44.0 86.0 66.2 17.5 141.0 60.2 32.0 234.0 78.0Pr 5.2 8.6 6.5 4.9 9.2 7.2 2.2 18.0 7.4 5.0 13.1 8.7Nd 23.0 31.5 26.0 7.0 28.0 20.2 16.0 16.0 16.0 16.0 137.0 47.3 16.0 22.0 19.2 19.0 170.0 64.4 9.0 77.5 28.6 16.0 47.5 29.5Sm 5.2 6.9 6.1 1.0 5.4 3.5 3.0 3.0 3.0 4.0 6.0 5.0 3.0 4.0 3.2 4.0 5.0 4.5 1.6 13.8 5.9 3.0 9.7 5.7Eu 1.4 1.8 1.6 0.9 1.3 1.0 0.6 3.8 1.8 0.0 1.4 0.4Gd 4.7 6.1 5.4 1.4 2.6 2.1 1.8 12.2 5.3 3.0 11.3 5.1Tb 0.8 1.1 0.9 0.2 0.4 0.3 0.3 1.7 0.8 0.5 1.0 0.7Dy 4.0 5.6 4.7 1.3 2.1 1.7 1.6 8.3 3.8 2.3 10.3 4.5Ho 0.7 1.1 0.9 0.2 0.3 0.3 0.3 1.7 0.7 0.5 1.1 0.8Er 2.0 2.8 2.3 0.6 1.0 0.8 0.8 4.2 1.9 1.2 3.4 2.2Tm 0.2 0.4 0.3 0.0 0.1 0.1 0.0 0.6 0.2 0.1 0.6 0.4Yb 1.6 4.7 2.8 0.5 1.4 1.0 1.0 1.0 1.0 1.0 2.0 1.8 2.0 2.0 2.0 0.8 4.1 2.1 1.0 8.2 2.8Lu 0.2 0.3 0.3 0.1 0.1 0.1 0.1 0.5 0.3 0.2 1.4 0.5

REE 93 190 136 44.1 157 106 45.1 349 145 74.8 442 177

V 160 500 299 15 27 22 37 216 117 150 245 180 5 41 17Cr 20 1600 352 4 11 6 19 274 82Ni 5 197 95 2 15 8 10 175 49 5 10 5Cu 25 110 70 5 5 5 7 21 11 2 59 21 30 224 77 1 130 12Zn 90 115 104 12 43 32 12 74 33 41 150 94 60 145 111 0 154 41Ga 15 24 19 14 24 19 19 19 19 17 21 19 19 23 21 16 18 17 18 22 20 12 23 18Sn 1 8 3 1 2 2 1 2 1 1 3 2

The number of analyses for each suite is given in parenthesis. A complete listing of the chemical data is obtainable from the first author upon request.

G.

Jarra

re

ta

l./Pre

cam

bria

nR

ese

arch

12

3(2

00

3)

29

5–

31

9301

Fig. 2. Harker type variation diagrams for selected major and trace elements. Symbols as inFig. 4.


Fig. 3. Silica distribution in the Aqaba and Araba complexes.

The LPONIJ cover a wide of range of compositionsand exhibit geochemical variation spanning the entirerange from gabbro to high silica granites (45–80%SiO2) (Fig. 2). The Aqaba complex does not show aclear compositional gap while the intrusions of Arabacomplex display a distinctly bimodal distribution(Fig. 3), similar to the one shown by the sub-volcanicand volcanic suites of the Araba complex (Jarraret al., 1992).

Mineralogy and geochemistry places the granitoidsinto three major classes in the sense ofBarbarin(1999); namely: amphibole-rich calc-alkaline grani-toids; the K-rich and K-feldspar porphyritic calc-alk-aline granitoids; and the peralkaline and alkalinegranitoids. Aqaba complex granitoids belong tothe amphibole-rich calc-alkaline granitoids and theK-feldspar porphyritic calc-alkaline granitoids types

Fig. 4. Q′-ANOR classification diagram (Streckeisen and LeMaitre, 1979) using normative compositions, Q′ = Q/(Q + Or+ Ab + An), ANOR = An/(Or + An) × 100.

where the latter dominates, while Araba complexgranitoids belong to the peralkaline and alkalinegranitoidal type. It must however, be stressed thatthe peralkaline granitoids are represented by rhyo-lites (Jarrar et al., 1992; Jarrar, 1992). K-feldsparporphyritic calc-alkaline granitoids is thought to formeither during periods of relaxation between culmi-nation periods within a collision event, or transitionfrom a compressional regime to a tensional regime(e.g. Bonin, 1990). They are abundant in orogenicbelts related to continental collision, in particularduring the final stages (Barbarin, 1999).

In the Q′-ANOR diagram ofStreckeisen and LeMaitre (1979), the Aqaba complex rocks are classi-fied as gabbro, quartz-diorite, quartz-monzodiorite,monzodiorite, quartz-monzonite, granodiorite, mon-zogranite, syenogranite and alkali feldspar granite


Fig. 5. Ab-or-an normative classification diagram afterBarker (1979). Symbols as inFig. 4.

(Fig. 4). On the other hand, the Araba complex isdominated by quartz- and monzodiorites, monzogab-bro, syenogranite and alkali feldspar granite. On thenormative ab-an-or plot (Barker, 1979) the LPONIJcluster in the fields of granite, granodiorite, tonaliteand trondhjemite (Fig. 5).

In the TAS diagram (Fig. 2), the LPONIJ form acoherent trend that passes from gabbro to alkalinegranite, straddling the boundary between alkalic and

Fig. 6. K2O vs. SiO2 diagram. Dividers are fromRickwood (1989).Symbols as inFig. 4.

subalkalic rocks as defined byIrvine and Baragar(1971). The Araba Mafic to Intermediate Suite andthe Yutum Suite, the youngest of the Aqaba complex,plot clearly in the alkaline field. In contrast, granitesof the Humrat–Feinan Suite of the Araba complex plotin the subalkaline field. The LPONIJ define a typicalcalc-alkaline trend on the standard AFM diagram (notshown) and occupy the same field defined by ANScalc-alkaline granitoids (Ramsay et al., 1986). Fur-thermore, they follow the typical calc-alkaline ratherthan the trondhjemitic trend on the Na2O–K2O–CaOtriangle (Barker and Arth, 1976) (not shown). Note-worthy also is that the mafic end member of theAqaba complex (the DHS) is more primitive (MeanMg# = 59) and calc-alkaline while the AMIS is morefractionated (Mean Mg#= 51) and follows a tholei-itic fractionation trend (not shown). On the K2O–SiO2plot (Rickwood, 1989) the LPONIJ cluster mainlyin the high-K field (Fig. 6). The mafic and felsic ex-treme compositions show a wider range of variationextending from medium-K to shoshonitic fields.

In the major element discrimination plot for gran-ites (SiO2 > 68%; Sylvester, 1989; diagram notshown) the LPONIJ plot in the calc-alkaline field(mostly members of the Aqaba complex) and thefield for alkaline and highly fractionated calc-alkaline


Fig. 7. A/NK vs. A/CNK Shand diagram. Symbols as inFig. 4.

granites (mostly Yutum Suite and Araba complex).Few samples plot in the field for alkaline granites. Inthe Peacock diagram (not shown), LPONIJ defines acalc-alkali trend (Aqaba complex) and an alkali-calcicto alkali trend (Araba complex). The Humrat–FeinanSuite does not follow either of the trends. On theShand diagram (Fig. 7) the LPONIJ are metalumi-nous, peraluminous, and peralkaline. The Rahmaand Rumman suites of the Aqaba complex show thestrongest peraluminous character, with A/CNK up to1.38. These are similar to post-collisional, stronglyperaluminous granites as defined bySylvester (1998).The Humrat–Feinan Suite shows a peralkaline affin-ity, although alkali pyroxenes and amphibole are rare

in these granites. On the other hand, the peralka-line rhyolites of the Aheimir Suite (Table 1) containabundant aegirine.

3.2. Trace elements

The investigated rocks are characterized by the fol-lowing ranges in incompatible element ratios: K/Rb=(117–623); K/Ba = (12–3160); Zr/Hf = (2–59);La/Yb = (3–46); Zr/Nb = (1–148); Zr/Y = (1–85).The average K/Rb value of 282 is closer to those ofthe upper continental crust ratios of 250 (Taylor andMcLennan, 1985). Sr and Ba behave incompatibly inthe mafic rocks and compatibly in the granitic rocks(Fig. 2). On log–log plots of Rb versus Sr and Ba ver-sus Sr (Fig. 8) the trends are non-linear. It is also clearthat the bulk of the granitoids of Aqaba complex haveRb/Sr between 0.1 and 1 (the exception being the Yu-tum Suite), while those of the Araba complex haveRb/Sr> 1. The Rb/Sr of the mafic parts of both com-plexes cluster around 0.1, typical for continental trond-hjemite and quartz-diorites (Coleman and Peterman,1975). Ba/Sr are predominantly greater than one foralmost all granitoids. On a triangular Sr–Rb–Ba dia-gram (Fig. 9), the granitoids define two trends: normal,low Ba–Sr granitoids, which trend with increasing sil-ica towards the Rb corner (mainly the Araba complexgranitoids and Yutum Suite); and high Ba–Sr grani-toids (all other suites of the Aqaba complex), whichevolve towards the Ba corner. The high Ba–Sr group isabundant in Archean granitoids and post-Cretaceousto Recent orogenic volcanic and plutonic rocks butrare among Proterozoic granitoids (Tarney and Jones,1994).

Complete rare earth element data sets are availablefor the DHS and the Rahma Suite from the Aqabacomplex, and the Humrat–Feinan and Araba Mafic toIntermediate Suites from the Araba complex. Partialdata, however, are available for La, Sm, Nd and Yb formost suites. The chondrite-normalized REE patternsare presented inFig. 10. The DHS rocks are mod-erately enriched in REE (

∑REE = 100–168 ppm).

The lowest and highest concentrations are recorded inone of the primitive compositions (JK-23; Mg#= 63)and the most fractionated (JK-5; Mg#= 41), respec-tively. The REE patterns of the DHS (Fig. 10) areLREE-enriched [(La/Lu)n = 4.5–11.3] and show amodestly negative Eu anomaly (Eu/Eu∗ = 0.72–0.90).


Fig. 8. Rb vs. Sr and Ba vs. Sr. The vectors indicate the influence of fractionation of various mineral phases on the evolution of theinvestigated rocks; melts governed by fractionation of a specific mineral will evolve in the direction of the arrowhead. Symbols as inFig. 4.

Fig. 9. Ba–Rb–Sr triangular plot. Symbols as inFig. 4.

The REE pattern slope steepens and the Eu anomalydeepens with fractionation.

Granitoids of the Aqaba complex (represented bythe Rahma Suite) are characterized by moderate abun-dances of REE (

∑REE = 96–151 ppm) and steep

REE patterns (Fig. 10). These granitoids are highlyenriched in LREE [(La/Lu)n = 20–25] yet lack a sig-nificant Eu anomaly (Eu/Eu∗ = 0.97–1). The aboveREE patterns are almost identical to that of the por-phyritic granite (625 Ma) of the Timna complex (Beythet al., 1994), and thus seems to be characteristic oflate-orogenic granitic melts in the northernmost ANS,i.e. those that formed prior to development of the re-gional unconformity∼600 Ma.

Mafic to intermediate intrusives of the Araba com-plex, the Araba Mafic to Intermediate Suite, generally


Fig. 10. Chondrite-normalized REE plots for investigated intrusions. Normalizing values are afterSun and McDonough (1989). Symbolsas in Fig. 4. The stippled area on the plot of Rahma Suite is the range of partial melt derived by 10–30% partial melting of (a 99.3%MORB+ 0.7% deep sea sediment) source (Drummond et al., 1994). The solid line in the same plot corresponds to the REE compositionof this source.

contain higher abundances of rare earth elementscompared to the Aqaba complex, although the formershows a wide range of REE abundances (

∑REE =

45–350 ppm). The Araba Mafic to Intermediate Suitehas LREE enriched patterns that nevertheless are lesssteep than those of mafic intrusives of the Aqabacomplex [(La/Lu)n = 5–15]. They also exhibit aEu/Eu∗ ∼ 0.46–1.24. These REE patterns are similarto those of mafic members of the late-NeoproterozoicBimodal Suite in the region (Stern and Gottfried,1986; Beyth et al., 1994).

Felsic intrusives of the Araba complex (the Humrat–Feinan Suite) have REE abundances comparableto those of the Araba Mafic to Intermediate Suite(∑

REE = 113–261 ppm). The REE patterns aredistinct from those for rocks of the Aqaba complex.Araba complex granitoids show a monotonous slop-

ing LREE pattern [(La/Sm)n ∼ 2] and almost flatHREE pattern [(Gd/Lu) =∼ 1] with a pronouncedEu negative anomaly [Eu/Eu∗ ∼ 0.07–0.53]. SimilarREE patterns are reported for A-type granites fromother parts of the ANS (Stern and Gottfried, 1986;Beyth et al., 1994; Moghazi et al., 1998), and are char-acteristic features of A-type granites the world over(Whalen et al., 1987; Rogers and Greenberg, 1990).

MORB-normalized multi-element plots (normaliz-ing values afterPearce, 1983) are shown inFig. 11.The spider diagrams indicate an overall enrichmentin large ion lithophile elements (LILE) compared toMORB. DHS displays the least enriched pattern, withmoderate negative anomalies at Nb and Hf. The felsicto intermediate suites of the Aqaba complex showminor to moderate Nb anomalies and occasionallyminor anomalies of Sr, P and Ti. The Araba Mafic to


Fig. 11. MORB normalized multi-element spider diagrams for the different investigated suites. The three plots in each diagram representminimum, average, and maximum from bottom to top, respectively. The averages only are shown for Aqaba granitoids. Normalizing valuesare afterPearce (1983). Symbols as inFig. 4.

Fig. 12. Fe2O3/MgO vs. (Y+Zr+Nb+Ce) diagram. The differentgranite fields are fromWhalen et al. (1987). Symbols as inFig. 4.

Intermediate Suite displays minor Nb and Zr anoma-lies. On the contrary, the Humrat–Feinan Suite showssignificant relative depletions of Sr, Ba, P and Ti anda minor depletion at Nb. These features are typical forA-type granites.

A plot of HFSE versus FeO∗/MgO (Whalen et al.,1987) is shown inFig. 12. The Humrat–Feinan Suiteplot in the fields of A-type granites and fractionatedI-type granites. On the tectonic discrimination dia-grams (Fig. 13), most granitoids of the Aqaba com-plex plot exclusively in the field of volcanic arc gran-ites. However, the Yutum Suite rocks, the youngestof the Aqaba complex, extend to the within-platefield. The granitoids of the Araba complex, on theother hand, plot mostly in the field of within-plategranites. Both complexes cluster in the field definedfor post-collisional granites byPearce (1996). TheDuheila Hornblendic Suite and the Araba Mafic to


Fig. 13. Nb vs. Y and Rb vs. (Nb+ Y) tectonic discriminationdiagrams for granites afterPearce et al. (1984). VAG = volcanicarc granites, Syn-COLG= syncollisional granites, WPG= withinplate granites, ORG= ocean ridge granites. Symbols as inFig. 4.

Intermediate Suite plot in the fields of arc lavas andwithin-plate basalts, respectively, on several tectonicdiscrimination plots for mafic rocks, not shown here(Jarrar et al., 1993; Jarrar et al., submitted for publi-cation; Jarrar, 2002).

3.3. Sr and Nd isotopic compositions

Pertinent Rb/Sr and Sm/Nd isotopic data on theinvestigated rocks are given inTables 3 and 4. The87Sr/86Sr initial ratios and theεSr are calculated for95 samples with87Rb/86Sr < 3 out of a total of 141samples in our database. Samples with87Rb/86Sr <

3 are excluded because their initial ratios are sus-ceptible to magnified errors; this omits data for theHumrat–Feinan Suite. U/Pb lead zircon ages have beenused for the calculation when available because Rb/Srages of granites from this region are known to besystematically 10–20 Ma younger.Fig. 14 is a his-togram of initial ratios for individual samples of bothAqaba and Araba complexes. The Aqaba complexhas a mean initial87Sr/86Sr of 0.70305± 0.00012,whereas the Araba complex (AMIS only) averages0.70357±0.00009.Mittelfeldt and Reymer (1986)di-vided the Sinai granitoids into three groups on the ba-sis of their Sr initial ratios: (1) a low initial ratio group(87Sr/86Sri ≤ 0.7036), (2) an intermediate initial ratio

Fig. 14. Histogram of the87Sr/86Sr initial ratios for Aqaba andAraba complexes; only samples with87Rb/86Sr < 3 are used.Initial ratios were calculated for individual samples using isochronages and U–Pb zircon ages when available.


Table 3Rb–Sr isotopic data for the Jordanian late- and post-orogenic intrusions

Sample ID Rb (ppm) Sr (ppm) 87Rb/86Sr 87Sr/86Sr Sri E Srt

Rahma Suite87/112 (615) 35 790 0.1290 0.70470 0.70357 −3.0787/113 39 912 0.1250 0.70460 0.70350 −3.9987/114 40 867 0.1340 0.70472 0.70355 −3.4087/115 35 887 0.1150 0.70459 0.70358 −2.8887/116 42 848 0.1440 0.70474 0.70348 −4.3787/117 39 944 0.1190 0.70461 0.70357 −3.1087/119 45 611 0.2140 0.70575 0.70387 1.2687/120 44 630 0.2020 0.70557 0.70380 0.2087/121 43 611 0.2060 0.70557 0.70376 −0.30JOR-13 49 690 0.2043 0.70550 0.70371 −1.08JOR-33 A 40 688 0.1669 0.70548 0.70402 3.29JOR-33 E 33 647 0.1481 0.70590 0.70460 11.6087/125 (610) 84 542 0.4460 0.70741 0.70353 −3.7087/126 85 788 0.3130 0.70624 0.70352 −3.8887/127 86 560 0.4450 0.70729 0.70342 −5.2887/128 92 556 0.4760 0.70744 0.70330 −6.9887/129 90 546 0.4780 0.70760 0.70344 −4.9587/130 96 600 0.4620 0.70748 0.70346 −4.6887/131 89 575 0.4470 0.70732 0.70343 −5.1087/61 (600) 74 547 0.3920 0.70715 0.70380 −0.08

Darba Suite87/62 71 573 0.3560 0.70678 0.70373 −0.9687/63 83 517 0.3000 0.70593 0.70336 −6.2387/64 84 522 0.4680 0.70774 0.70374 −0.9487/65 78 540 0.4180 0.70728 0.70370 −1.4087/66 75 530 0.4040 0.70712 0.70366 −1.9787/67 78 514 0.4400 0.70751 0.70375 −0.8087/68 79 529 0.4430 0.70732 0.70353 −3.8787/69 78 537 0.4180 0.70752 0.70394 2.0187/77 (589) 84 704 0.3440 0.70599 0.70310 −10.1487/78 77 732 0.3050 0.70582 0.70326 −7.90

Rumman Suite87/79 79 761 0.3000 0.70593 0.70341 −5.7487/80 80 760 0.3080 0.70601 0.70342 −5.5687/81 101 378 0.7770 0.70989 0.70336 −6.4087/82 96 363 0.7770 0.70986 0.70333 −6.8387/83 94 360 0.7700 0.70986 0.70339 −5.9987/84 98 643 0.4420 0.70722 0.70351 −4.3687/85 101 642 0.4570 0.70749 0.70365 −2.3287/86 91 642 0.4120 0.70695 0.70349 −4.6287/109 (589) 107 577 0.5360 0.70803 0.70353 −4.0787/110 108 572 0.5490 0.70796 0.70335 −6.6287/111 105 571 0.5310 0.70787 0.70341 −5.7587/87 (620) 98 591 0.4800 0.70749 0.70325 −7.57S7/S8 90 618 0.4220 0.70706 0.70333 −6.4087/89 90 636 0.4100 0.70684 0.70321 −8.01

Urf Suite87/90 86 616 0.4030 0.70690 0.70334 −6.2867/91 89 621 0.4140 0.70692 0.70326 −7.3887/92 100 658 0.4370 0.70718 0.70332 −6.5887/93 83 625 0.3850 0.70676 0.70336 −6.0187/94 88 615 0.3650 0.70642 0.70319 −8.33


Table 3 (Continued)

Sample ID Rb (ppm) Sr (ppm) 87Rb/86Sr 87Sr/86Sr Sri E Srt

87/95 86 611 0.4080 0.70695 0.70334 −6.2087/99 120 295 1.1750 0.71372 0.70333 −6.3887/106 111 678 0.4720 0.70748 0.70331 −6.7167/143 60 712 0.2410 0.705510 0.70338 −5.6887/144 80 770 0.2990 0.705970 0.70333 −6.43JOR-40 (625) 111 111 2.8880 0.729150 0.70341 −5.22JOR-114 A 97 107 2.6277 0.726780 0.70336 −5.93JB-19 91 111 2.3976 0.722190 0.70082 −42.0087/70 (623) 77 771 0.2890 0.706090 0.70352 −3.5987/71 104 716 0.4200 0.707390 0.70366 −1.6687/72 79 707 0.3230 0.706410 0.70354 −3.3487173 87 825 0.3060 0.706220 0.70350 −3.8987174 78 771 0.2940 0.706060 0.70345 −4.6587/75 118 717 0.4750 0.707680 0.70346 −4.4987/76 86 688 0.3640 0.706740 0.70351 −3.8387/96 81 640 0.3670 0.706560 0.70330 −6.7687/97 88 625 0.4060 0.707050 0.70344 −4.7387/98 83 645 0.3710 0.706620 0.70332 −6.42

Yutum Suite87/1 (608) 124 196 1.8240 0.718320 0.70250 −18.3187/3 124 197 1.0700 0.718360 0.70908 75.1687/5 121 201 1.7440 0.717810 0.70269 −15.7087/6 120 187 1.8680 0.718840 0.70264 −16.3487/7 118 196 1.7380 0.717070 0.70200 −25.4787/8 132 170 2.2610 0.722110 0.70251 −18.29JOR-30 163 165 2.8700 0.727800 0.70291 −12.4887/13 93 375 0.7160 0.708910 0.70270 −15.5087/14 91 385 0.6840 0.708720 0.70279 −14.2687/15 91 376 0.7060 0.708930 0.70281 −13.9867/19 90 388 0.6750 0.708780 0.70293 −12.3087/20 92 377 0.7030 0.708930 0.70283 −13.6187/23 101 472 0.6220 0.708300 0.70291 −12.5987/24 93 444 0.5980 0.708030 0.70285 −13.4787/25 95 450 0.6090 0.708100 0.70282 −13.8387/26 98 206 1.3810 0.714800 0.70283 −13.7487/28 80 717 0.3240 0.705860 0.70305 −10.54JOR-38 92 389 0.6865 0.709300 0.70335 −6.32JOR-97 102 373 0.7945 0.710620 0.70373 −0.87

Araba Mafic to Intermediate SuiteGW-4 (572) 42 882 0.1422 0.704720 0.70356 −3.90GW-8 65 876 0.2071 0.705210 0.70352 −4.45GW-19 47 628 0.2119 0.705320 0.70359 −3.44GW-21 15 452 0.0899 0.704290 0.70356 −3.94RZ-1 (590) 55 599 0.2670 0.705660 0.70341 −5.72RZ-7 48 638 0.2159 0.705350 0.70353 −4.04RZ-8 46 657 0.2034 0.705360 0.70364 −2.41RZ-9 49 657 0.2157 0.705360 0.70355 −3.78RZ-10 111 443 0.7299 0.709900 0.70376 −0.82

The sample numbers with forward slash are fromBrook et al., 1990.The samples with the prefix (JOR) are fromJarrar (1985).The samples with the prefix (GW) are fromJarrar et al. (submitted for publication).The samples with the prefix (RZ) are from this study.The italicized numbers are ages for which the initial ratios and epsilon values have been calculated.Initial ratios have been calculated for samples with87Rb/86Sr < 3.


Table 4Sm–Nd isotopic data for some of the investigated suites

Sample ID Sm (ppm) Nd (ppm) 143Nd/144Nd 147Sm/144Nd TDM (Ga) ε Nd (t Ma)

Yutum Suite (Brook et al., 1990) 608 Ma87/5 3.52 23.20 0.512385 0.0916 0.81 3.5987/7 3.35 22.21 0.512334 0.0912 0.87 2.6287/24 4.44 27.51 0.512429 0.0975 0.79 3.9987/28 5.80 33.45 0.512465 0.1048 0.79 4.12

Darba Suite (Brook et al., 1990) 600 Ma87/62 3.16 16.32 0.512438 0.1169 0.93 2.57

Urf Suite (Brook et al., 1990) 620 Ma87/70 2.79 16.36 0.512414 0.103 0.85 3.4187/93 2.66 17.42 0.512407 0.0923 0.78 4.1287/94 2.66 17.17 0.512318 0.0935 0.9 2.2987/99 4.25 22.04 0.512456 0.1166 0.9 3.1587/108 3.23 20.66 0.512408 0.0946 0.8 3.9687/143 3.56 22.49 0.512387 0.0958 0.83 3.4687/144 3.17 20.42 0.512427 0.0937 0.77 4.4

Rumman Suite (Brook et al., 1990) 600 Ma87/109 3.46 20.24 0.512445 0.1032 0.81 3.7687/111 3.17 21.34 0.512457 0.0898 0.71 5.0287/86 6.29 34.83 0.512427 0.1092 0.88 2.95

Rahma Suite (Brook et al., 1990) 615 Ma87/115 4.32 27.45 0.512387 0.0951 0.83 3.4587/121 3.18 17.56 0.512406 0.1093 0.91 2.7

Humrat–Feinan Suite (Brook et al., 1990; Jarrar et al., submitted for publication) 560 Ma87/30 2.56 15.89 0.512456 0.0972 0.75 3.9287/37 3.81 17.02 0.512524 0.1354 0.98 2.5187/43 3.98 20.70 0.51248 0.1161 0.86 3.0487/60 2.12 14.72 0.51239 0.087 0.77 3.37GW-25 4.50 27.00 0.51253 0.1008 0.7 4.76GW-32 6.60 42.40 0.512473 0.0941 0.74 4.13GW-35 3.90 16.40 0.512654 0.1438 0.86 4.1

Araba Mafic to Intermediate Suite (this study) 570 and 590 MaGW-19 9.11 44.69 0.512556 0.1232 0.83 3.76GW-21 9.72 49.63 0.512561 0.1183 0.78 4.21RZ-7 4.98 24.31 0.51248 0.1238 0.96 2.42RZ-9 5.22 25.53 0.512526 0.1236 0.88 3.33RZ-10 4.79 25.76 0.512474 0.1123 0.86 3.17

UCSD standard of143Nd/144Nd is 0.511841 for the data fromBrook et al., 1990and 0.511859 for the data fromJarrar et al. (submittedfor publication)to the International Journal of Earth Science and this study.

group (87Sr/86Sri = 0.704–0.705), and (3) a high ini-tial ratio group (87Sr/86Sri ≥ 0.705). Both Aqaba andAraba complexes fall within the first group of the Sinaigranitoids and the granitoids of the Eastern Desert ofEgypt (Moghazi et al., 1998).

Nd isotopic data indicate a remarkably homoge-nous source for all LPONIJ (Table 4). Araba andAqaba complexes have indistinguishableεNd(t) val-

ues (+2.3 to +5.0, mean= +3.5; Fig. 15A). TheLREE-enriched nature of all LPONIJ for which wehave Nd isotopic is reflected in their moderately low147Sm/144Nd (0.087–0.1438, mean= 0.106). RobustNd model ages can be calculated from all 29 sam-ples listed inTable 4, using the algorithm ofNelsonand DePaolo (1984). This yields a narrow range inmodel ages, from 0.71 to 0.98 Ga (mean= 0.83 Ga;

Fig. 15. Histogram of Nd isotopic dataεNd(t) and TDM given in Table 4. Note the tight clustering of data, including mafic and felsicsuites, for both Aqaba and Araba complexes.

Fig. 16. εNd vs. εSr plot for some suites for which Nd and Sr data are available. Mantle array afterO’Nions et al. (1979).


Fig. 15B). Samples for which both initial Nd and Srdata are available and useful have been plotted on theεNd–εSr diagram (Fig. 16). All of the samples plot inthe mantle array and within the field occupied by thegranitoids of the Arabian Nubian Shield (Duyvermanet al., 1982; Stern and Kröner, 1993). The positiveεNd and low Sr initial ratios (<0.704) preclude anyimportant contribution of pre-Neoproterozoic con-tinental crust. These data clearly support modelswhereby the Jordanian Neoproterozoic granitoidsbelong exclusively to the juvenile terrains of theANS.

4. Petrogenesis

4.1. Duheila Hornblendic Suite

This is the oldest intrusive suite in the Aqaba com-plex, comprising diorites and gabbros and containingabundant amphibole. It has a calc-alkaline characterand is enriched in LILE including K, Rb, Sr, Ba,La, Ce and Th relative to the HFSE (Zr, Nb and Y).Spider diagrams for these rocks, in particular thenegative Nb anomalies, suggest a subduction-relatedenvironment. Some compositions approximate primi-tive magmas (Mg# 63, Ni and Cr contents of 190 and1600 ppm, respectively). A couple of samples haveMg# as high as 75 but these likely to be amphibolecumulate rocks. Enrichments of LILE may have beencaused by the release of enriched fluids from the sub-ducted slab into the overlying mantle wedge. REEpatterns(La/Lu)n = 4.5 for the primitive composi-tions excludes the presence of residual garnet in thesource area. Based on the above lines of evidence,Jarrar (2002)demonstrated that this suite could havebeen produced by 10–15% modal batch meltingof amphibole-bearing spinel lherzolite in the man-tle wedge above a Neoproterozoic subduction zone(Fig. 17A). The REE patterns of the DHS are indistin-guishable from those of some island-arc basalt (IAB)(Wilson, 1989; Gill, 1981), supporting the conclu-sion that these melts formed in a subduction-relatedtectonic setting. Intra-suite elemental variations andthe presence of local cumulate hornblendite suggestthat the fractionation of amphibole, plagioclase andpyroxene played important roles in the evolution ofthis suite.

4.2. Granitoids of the Aqaba complex

Aqaba complex granitoids show compositionalcontinuities in terms of major and trace elementsgeochemistry (Fig. 2) and dominate the basementoutcrops volumetrically. These granitoids, especiallythe tonalites and granodiorites, are characterizedby abundant amphibolite xenoliths (Jarrar, 1995)and amphibole-rich dioritic enclaves (Habboush,1996). Low initial 87Sr/86Sr (∼0.7033) and posi-tive εNd values (+2.3 to +5.0) permit fractionationof mantle-derived magma to yield these granitoids.Aqaba complex granitoids have REE patterns that donot permit them to be generated by fractionation ofthe DHS, but it can be generated by partial meltingof a mafic source, in which amphibole and/or garnetis present as residual phases (Hanson, 1978). Addi-tionally, we would expect fewer granitoids and moremafic cumulate bodies if fractional crystallizationfrom a mantle-derived mafic melt was responsible.While over two-thirds of the ANS is dominated bygranitoids (Harris et al., 1990), more than 80% ofthe Jordanian part (this study) is made of the Aqabacomplex granitoids. This is difficult to reconcilewith a fractionation model for these granitoids, un-less the mafic complement is largely sequestered inthe lower crust, as suggested byMcGuire and Stern(1993).

The model ages are about 200 million years olderthan the crystallization age, and approximates the ageof the metamorphic rocks in the region, allowing theinterpretation that the LPONIJ formed by melting of∼800 Ma sediment and/or crust. We do not favor thisinterpretation, because of the isotopic homogeneityshown by all LPONIJ and because mafic and felsicmembers have similar Nd isotopic compositions. Thehomogeneity shown by LPONIJ may be a regionalfeature. Stern (2002)commented on the similar-ity of Nd isotopic compositions of basement rocksfrom Jordan, Israel, and Sinai, and suggested thatthe homogeneous and slightly older Nd model agesof Sinai–Israel–Jordan basement compared to that ofE. Egypt and Sudan implies a major Neoproterozoiccrustal boundary in the region, possibly the isotopicsignature of the Midyan terrain. Regardless of thesedifferences, the crust of this region must be regardedas a juvenile addition to the crust from the mantle ofNeoproterozoic age.


Fig. 17. (A) REE chondrite-normalized plot for the modeled partial melt of the parental magma of the Duheila Hornblendic Suite (Jarrar,2002). (B) Primitive mantle-normalized multi-element spider diagram for the modeled partial melt of the Araba Mafic to Intermediate Suitecompared with the most primitive composition of the suite (GW-23) (Jarrar et al., submitted for publication). (C) Primitive mantle-normalizedmulti-element spider diagram for the modeled parental magma of the Humrat–Feinan Suite compared with the most primitive graniticcomposition of the suite (GW-34) (Jarrar et al., submitted for publication). (D) Primitive mantle-normalized multi-element spider diagramfor the most fractionated sample of the Humrat–Feinan Suite (GW-26) compared with the modeled fractionated melt from the Araba Maficto Intermediate Suite (Jarrar et al., submitted for publication).

Partial melting of water-saturated amphibolite isa plausible mechanism for the formation of Aqabacomplex granitoids. REE patterns argue for the pres-ence of residual garnet and/or amphibole in thesource. The most primitive members of the Aqabagranitoids plot in the field of tonalities, granodioritesand trondjhemites (Fig. 4). Springer and Seck (1997)demonstrated that partial melting of basaltic rocksat P = 5–15 kbar under water saturated conditionsprogressively generates granites, granodiorites andfinally at highest degrees of partial melting tonalite,bypassing trondhjemite, while under water-deficientconditions granodiorite is bypassed and trondhjemiteis formed instead (Best and Christiansen, 2001). The

trend defined by the LPONIJ (Fig. 5) favors the partialmelting of basaltic rocks under water-saturated condi-tions. Experimental investigations (Rapp et al., 1991)demonstrate that tonalitic parental magmas with steepREE patterns can be generated by partial melting ofbasalts at pressures≥16 kbar, leaving an hornblendeeclogitic residue.Drummond et al. (1994)modeledthe melting of subducted oceanic crust and sedimentsto yield a parental magma for a quartz-diorite. Select-ing a source composed of 99.3% MORB and 0.7%deep sea sediments, they found that 10–30% meltingshould yield tonalitic–granodiorite melt in equilib-rium with a refractory assemblage of 30% hornblende,30% garnet and 40% clinopyroxene. The modeled


REE patterns for the assumed source and the resul-tant partial melts obtained by these authors are shownon the Rahma Suite REE plot (Fig. 10). A good fitbetween the modeled partial melts (10–30%) and theRahma Suite patterns suggests that the MORB partialmelting model is appropriate.Drummond et al. (1994,and references therein)list the following geochemicalcharacteristics of felsic melts derived from meltinga garnet amphibolite eclogite slab source: Al2O3 >

15% at 70% SiO2, Yb < 1.9 ppm, Y < 15 ppm,Sr > 600 ppm, Sr/Y> 10, La/Yb> 20, K/Rb< 550,and Nb< 10–11 ppm. These features are similar tothose of relatively primitive Aqaba complex gran-itoids. Based on geochemical modeling, tectonicobservations, experimental petrology and numericalheat transfer models,Peacock (1990)andDrummondand Defant (1990)found that subduction of young(<30 Ma) and hot oceanic crust is necessary for slabmelting to take place. Alternatively, melting of amphi-bolite facies juvenile crust could also yield these gran-itoids.Furnes et al. (1996)suggested partial melting ofamphibolitic deep crust to generate the older granites(700–750 Ma) in Wadi El-Imra district, Central East-ern Desert of Egypt. Likewise,Moghazi et al. (1998)advocated, on the basis of Pb–Sr–Nd isotopic data,the partial melting of an amphibolitic juvenile lowercrust as a source for the granodiorite–monzograniteassociation of Wadi Kid area, southeast Sinai.

We conclude that melting of young oceanic crustof the Mozambiquean ocean is a plausible mecha-nism for generating Aqaba complex granitoid parentalmelts. Since, the vast majority of the LPONIJ aregranitoids that consist primarily of quartz, alkalifeldspar, plagioclase, and variable amounts of am-phibole and biotite, the Rb, Sr, and Ba variationsare useful for estimating the role of these mineralsin the evolution of the LPONIJ. Rb (and to a lesserextent, Zr, Nb, Y, and LREE) is the only trace ele-ment, which behaves incompatibly through the wholecompositional range (Fig. 2). Consequently, the vari-ation within the complex itself and among varioussuites can be explained by fractionation of variousproportions of amphibole, biotite and feldspars, assuggested by the Rb, Sr, Ba vector plots (Fig. 8).The vectors drawn for the various minerals displaythe role of plagioclase and amphibole fractionation inmafic members and K-feldspar fractionation in felsicmembers.

4.3. Araba Mafic to Intermediate Suite

The Araba Mafic to Intermediate Suite spans abroad compositional range in silica (48–64.6%) andMg# (38–66) (Fig. 2). They are characterized by mod-erate to high concentrations of K2O (0.75–4.08 wt.%),TiO2 (0.62–2.46 wt.%), and P2O5 (0.06–1.35 wt.%)(Table 2). Although they are enriched in the incompat-ible trace elements (LILE, HFSE), they show negativeNb anomalies (Fig. 11). All of these features indi-cate that Araba Mafic to Intermediate Suite parentalmagmas were alkaline and derived from melting ofenriched mantle. Experimental work demonstratesthat K2O metasomatism of the mantle overlying asubduction zone can produce potassic phases like phl-ogopite or K-richterite (Schmidt, 1996). Turner et al.(1996) suggested that phlogopite-rich lithosphericmantle was the source for post-collisional potassicmagmas in Tibet.

Our model for the Araba Mafic Suite involves thegenerating alkaline mafic magma by batch modalmelting of LILE-enriched, phlogopite-bearing litho-spheric mantle. This metasomatism may have accom-panied the dehydration melting of the subducted slabto give rise to the Aqaba complex granitoids. The man-tle source composition is taken as phlogopite-bearingspinel lherzolite with the following mineralogy: oliv-ine:orthopyroxene:clinopyroxene:spinel:phlogopite=65:20:10:4:1.Jarrar et al. (submitted for publication)carried out the modeling and found that a primitivesample (GW-23; Mg#= 61, 175 ppm Ni, 274 ppmCr) of the Ghuweir Mafics shows an excellent fit with10% melting of that source (Fig. 17B). The composi-tion of a sample with Mg#= 66 is not different fromthe modeled sample and it also shows lower Ni andCr concentrations of 42 and 103 ppm, respectively.Consequently, sample GW-23 best approximates theprimary magma composition of the AMS. On thebasis of the above discussion, we conclude that theparental mafic magmas of the Araba Mafic to Inter-mediate Suite were derived by 10% partial melting ofphlogopite-bearing spinel lherzolite mantle source.

A drop in Mg# from 66 to 38, accompaniedby drastic changes in the chemistry of this suite(Table 2), is best explained by fractional crystalliza-tion. Jarrar et al. (submitted for publication)carriedout major and trace element modeling and con-cluded that separating 79% of a gabbroic assemblage


(olivine:pyroxene:plagioclase= 21:16:42) from theleast evolved magma best explains the intra-suitevariations in major and trace elements. Nevertheless,the observed LREE concentrations are higher thanthe modeled concentrations. The generally high con-tents of REE in the most evolved samples have beenattributed to the accumulation of apatite, as evidencedby excellent correlation between total REE and theP2O5 contents.

4.4. The Humrat–Feinan Suite

A-type granitic rocks of the Humrat–Feinan Suiteare thought to be comagmatic with the Araba Maficto Intermediate Suite. This conclusion is supportedby almost undistinguishableεNd values: +2.5 to+4.8 and+2.4 to +4.2, respectively, for the felsicand mafic end members of the Araba Alkaline Bi-modal Suite. To test the possibility that the Humrat–Feinan Suite evolved via fractional crystallization ofthe Araba Mafic to Intermediate Suite,Jarrar et al.(submitted for publication)used major elementsleast square modeling and found that fractional crystal-lization of a gabbroic assemblage (olivine:pyroxene:plagioclase:magnetite:melt= 12:19:37:8:23) couldproduce the most primitive of the A-type granites(represented by sample GW-34; Mg#= 38). Traceelement modeling supports this result. There is goodagreement between the calculated melt and the ob-served compositions (Fig. 17C), which supports thehypothesis that the A-type granitic melt was pro-duced by extensive fractional crystallization of amantle-derived mafic magma. In order to account forthe high silica end members of the Feinan–HumratA-type granites, extensive fractional crystallization ofK- and Na-feldspars, amphibole, and small amountsof accessory magnetite, zircon, apatite, and allaniteare required. Geochemical modeling of the majorelements shows that the most evolved granitic sam-ple (GW-26; Mg#= 10;) can be produced from themost primitive granite (GW-34; Mg#= 39) by theseparation of the mineral assemblage (amphibole:Na-feldspar:K-feldspar:magnetite:apatite:melt= 6:44:17:1.5:0.5:31) (Jarrar et al., submitted for publication).This model was tested by the Rayleigh fractiona-tion modelling of the trace elements and excellentagreement was found between observed and modeledcompositions (Fig. 17D). An alternative interpretation

is that A-type granitic melts of the Humrat–FeinanSuite formed by melting of underplated mafic crust(Frost and Frost, 1997), where AMS equivalents arethe mafic underplates.

5. Implications for crustal evolution andfuture research

The crust of SW Jordan appears to have mostlyformed by magmatic additions from the mantle to thecrust. These magmas share common features through-out the∼80 million years of crustal evolution theyrecord, especially with respect to isotopic compositionof Sr and Nd and in being a high-K suite in general.Nevertheless, these magmatic additions tracked re-gional tectonics, with subduction-related (Aqaba com-plex) magmas dominating prior to terminal collisionbetween E and W Gondwana and rift-related (Arabacomplex) magmas dominating afterwards.

We favor a model of slab melting to generate Aqabacomplex felsic to intermediate melts. Nevertheless, thedominant ultimate source for all LPONIJ was man-tle with initial 87Sr/86Sr0̃.7030–0.7035 andεNd(t) ∼+3.5, and this did not change significantly during the80 Ma of crustal evolution. The isotopic similarity be-tween Aqaba and Araba complex melts that stronglyfavors, if not requires, similar mantle source regions. Itis very difficult to call upon slab melting for the formerand mantle melting for the latter. We also recognizethat a clear geological relationship between a subduc-tion zone and Aqaba complex granitoids remains to beestablished: we cannot identify other elements of in-ferred 600–625 Ma subduction zones, such as pairedmetamorphic belts, accretionary prism, etc. We rec-ognize that the linear arrangement of plutonic beltsassociated with sub-parallel dike swarms is more con-sistent with a rift setting than it is with an arc setting.We are also aware that the escape tectonic phase ofthe East African Orogen, thought to be associated withonset of terminal collision between E and W Gond-wana, was already underway by about 630 Ma (Stern,1994), about the time that Aqaba complex magmasformed. Terminal collision implies that true subduc-tion at this has ended, although it is conceivable thatother subduction zones formed around the periphery ofthe newly amalgamated supercontinent. These prob-lems are promising directions for future research.


It is also clear that magmatic evolution happened atdifferent crustal levels for Aqaba and Araba complexgranitoids, with Aqaba complex granitoids controlledby garnet- or hornblende-fractionates or residue andAraba complex granitoids controlled by feldsparfractionates or residues. The lower-pressure regimeevidenced by Araba complex melts is expected forthin, rifted crust, where faults provide conduits to thesurface and space for growth of magma chambers isreadily available. Likewise, a higher-pressure regimeis consistent with melt generation and evolution dur-ing the collisional phase, either via melting in thesubduction zone itself or by melting at the base ofthickened crust. Progressive shallowing of sites ofmelt equilibrium observed in the LPONIJ thus maymake up a “standard mode” of melt evolution in thelate stages of a collision.

This overview of LPONIJ indicates that thereis much to be learned about the evolution of thecontinental crust from studying East African Orogenintrusive rocks, particularly where robust geochrono-logical, geochemical, and isotopic data sets are avail-able. In the case of LPONIJ, several answerablequestions remain. What is the significance of the∼600 Ma unconformity and the Saramuj Conglom-erate? Are these largely paleoclimatic (i.e. Varanger-Marinoan glaciation) or tectonic indicators (i.e.beginning of orogenic collapse)? How was the tran-sition of magma genetic processes seen in the con-trasting chemistries of Aqaba complex and Arabacomplex granitoids accomplished? The unconformityis younger than the youngest member of the Aqabacomplex (Yutum Suite), butBeyth et al. (1994)placedthe transition from calc-alkaline to alkali granite ig-neous activity in the northernmost about 625–610 Ma,before emplacement of the Yutum Suite. REE data forthe Yutum Suite is needed to better assess when andhow the transition occurred. Why are the∼610 Masanukitoid melts which controlled development ofthe Timna complex in southern Israel not yet rec-ognized in the LPONIJ? Why is there no evidenceof the Najd Shear Zone in LPONIJ outcrops, inspite of the fact that Najd deformation and LPONIJmagmatism was contemporaneous and cospatial? Fi-nally, can the belts of LPONIJ be traced across areconstructed Dead Sea transform into correlatives inSinai? Again, fruitful avenues of future research areindicated.

6. Conclusions

The main conclusions drawn from this study are:

1. The Neoproterozoic basement complex in south-ern Jordan is dominated by high-K calc-alkalinepost-collisional granitoids and subordinate maficrocks, emplaced at about 640–560 Ma in a metaig-neous and metasedimentary sequence of green-schist and amphibolite facies metamorphism oflow pressure type dated at about 750–800 Ma.

2. The intrusions are divided into two complexesthat are separated by an intra-Precambrian uncon-formity at about 600 Ma. Granitoids of the older,Aqaba complex show characteristics of typicalvolcanic arc granites while those of the youngerAraba complex display features of A-type gran-ites. Both show isotopic signatures consistent withtheir derivation from mantle-derived magmas.

3. The mafic end members of both complexes are vol-umetrically insignificant. The Duheila HornblendicSuite of the Aqaba complex could be generated by10–15% melting of an amphibole-bearing spinellherzolite, possibly in the mantle wedge above asubduction zone. The Araba Mafic could be de-rived by 10% partial melting of a phlogopite bear-ing spinel lherzolite from a metazomatized mantlewedge.

4. Granitoids of the Aqaba complex comprise tona-lites, granodiorites, some trondhjemite, and gran-ites. These could be derived mainly by 10–30%melting of subducted oceanic crust with minor seasediments, or perhaps mafic lower crust. Granitesof the Araba complex were derived from the ArabaMafic Suite by extensive fractional crystallization.

5. Crust formation in the northernmost EAO was lar-gely accomplished by juvenile magmatic additions,with the region of important melt equilibrium shal-lowing in the transition from collision to rifting.

Acknowledgements

The authors thank the conveners of the SpecialSession at the GSA meeting in November 2000 on the“Evolution of the East African and Related Orogensand the Assembly of Gondwana,” Robert Tucker,Timothy Kusky and Robert Stern for the opportunity


to present this work as part of the session, and to theGSA International Division for partial support to par-ticipate in the meeting. The manuscript benefited fromthe critical comments and constructive suggestionsmade by Diane Smith and Michael Gilbert.

References

Abdel-Fattah, K.A.Y., 1985. Geology, Petrology and Geochemistryof the Basement Rocks, Southeast of Al-Aqaba, Jordan.Unpublished M.Sc. Thesis, Kuwait University, 260 pp.

Al-Zubi, H.A., 1996. Geology, Petrology and Geochemistry ofthe Granitoids of Wadi Es-Sabil Area, Southwest Jordan.Unpublished M.Sc. Thesis, University of Jordan, 173 pp.

Barbarin, B., 1999. A review of the relationships between granitoidtypes, their origin and their geodynamic environments. Lithos46, 605–626.

Barker, F., 1979. Trondhjemites, Dacites, and Related Rocks.Elsevier, New York.

Barker, F., Arth, J.G., 1976. Generation of trondjhemitic-tonaliticliquids and Archean Bimodal trondjhemite-basalt suites.Geology 4, 596–600.

Bender, F., 1974. Geology of Jordan. Gebrueder Borntraeger,Berlin, 196 pp.

Best, M., Christiansen, E., 2001. Igneous Petrology. Blackwell,Oxford.

Beyth, M., Stern, R.J., Altherr, R., Kröner, A., 1994. The latePrecambrian Timna igneous complex, southern Israel: evidencefor comagmatic-type sanukitoid monzodiorite and alkali granitemagma. Lithos 31, 103–124.

Bonin, B., 1990. From orogenic to anorogenic setting: evolution ofgranitic suites after a major orogenesis. Geol. J. 25, 261–270.

Brook, M., Ibrahim, K.M., McCourt, W.J., 1990. Newgeochronological data from the Arabian Shield area ofsouthwest Jordan. In: Proceedings of the 3rd JordanianGeological. Conference, Amman, Jordan, pp. 361–394.

Coleman, R.G., Peterman, Z.E., 1975. Oceanic plagiogranites. J.Geophys. Res. 80, 1099–1108.

Cosca, M.A., Shimron, A., Caby, R., 1999. Late Precambrian meta-morphism and cooling in the Arabian Nubian Shield: petrologyand40Ar/39Ar geochronology of metamorphic rocks of the Elatarea (southern Israel). Precambrian Res. 98, 107–127.

Drummond, M.S., Defant, M.J., 1990. A model for trondjhemite–tonalite–dacite genesis and crustal growth via slab melting:Archean to modern comparisons. J. Geophys. Res. 95, 503–521.

Drummond, M.S., Allison, D.T., Wesolowski, D.J., 1994. Igneouspetrogenesis and tectonic setting of the Elkahatchee quartzdiorite, Alabama appalachian: implications for penobscotianmagmatism in the eastern blue ridge. Am. J. Sci. 291, 173–236.

Duyverman, H.J., Harris, N.B.W., Hawkesworth, C.J., 1982.Crustal accretion in the Pan-African: Nd and Sr isotope evidencefrom the Arabian Shield. Earth Planet. Sci. Lett. 59, 315–326.

El-Hilu, M.M., 1995. Mineralogy, Geochemistry and Petrogenesisof the Humrat Granite in the Area of Quweira, South Jordan.Unpublished M.Sc. Thesis, University of Jordan, 112 pp.

Frost, C.D., Frost, B.R., 1997. Reduced rapikivi-type granites: thetholeiite connection. Geology 25, 647–650.

Furnes, H., El-Sayed, M.M., Khalil, S.O., Hassanen, M.A., 1996.Pan-African magmatism in the Wadi El-Imra district, CentralEastern Desert, Egypt: geochemistry and tectonic environment.J. Geol. Soc. Lond. 153, 705–718.

Gill, J.B., 1981. Orogenic Andesites and Plate Tectonics. Springer-Verlag, New York, 390 pp.

Habboush, M.A., 1996. Petrography, Geochemistry and Petro-genesis of the Quartz Diorite in Wadi A-Muhtadi. UnpublishedM.Sc. Thesis, University of Jordan, 119 pp.

Hanson, G.N., 1978. The application of trace elements to thepetrogenesis of igneous rocks of granitic composition. EarthPlanet Sci. Lett. 38, 26–43.

Harris, N.B., Schandelmeier, H., Darbyshire, D.P.F., 1990.Pan-African reworked early/middle Proterozoic crust in NEAfrica west of the Nile: Sr and Nd isotope evidence. J. Geol.Soc. Lond. 147, 859–872.

Hassuneh, M.H., 1994. Geological, Petrological and GeochemicalInvestigation of the Janub Metamorphic Suite Rocks in WadiEs-Sabil, Ain El-Hasheem Area, SE Aqaba. Unpublished M.Sc.Thesis, University of Jordan, Amman, 254 pp.

Ibrahim, K.M., McCourt, W.J., 1995. Neoproterozoic graniticmagmatism and tectonic evolution of the northern ArabianShield: evidence from Southwest Jordan. J. Afr. Earth Sci. 20,103–118.

Irvine, T.N., Baragar, W.R.A., 1971. A guide to the chemicalclassification of the common volcanic rocks. Can. J. Earth Sci.8, 523–548.

Jarrar, G., 1985. Late Proterozoic crustal evolution of the ArabianNubian Shield in the Wadi Araba area, SW-Jordan. Geol J.B61, 3–87.

Jarrar, G., 1992. Geochemistry and petrogenesis of analkali-feldspar rhyolite suite from Wadi Museimir, Central WadiAraba, Jordan. Chem. Erde 52, 301–312.

Jarrar, G., 1995. Pan-African amphibolites from SW-Jordan: theirorigin, mineralogy, bulk and mineral chemistry as a pressureand temperature indicator. Chem. Erde 55, 31–45.

Jarrar, G., 1998. Mineral chemistry in dioritic hornblendites fromWadi Araba, SW Jordan. J. Afr. Earth Sci. 26, 285–295.

Jarrar, G., 2001. The youngest Neoproterozoic Mafic Dike Suitein the Arabian Shield mildly alkaline dolerites from SouthJordan—their geochemistry and petrogenesis. Geol. Mag. 138,309–323.

Jarrar, G., 2002. Duheila Hornblendic Suite, Wadi Araba, SWJordan: Geochemistry and Petrogenesis. Dirasat 29 (1), 89–104.

Jarrar, G., Baumann, A., Wachendorf, H., 1983. Age determina-tions in the Precambrian basement of the Wadi Araba area,southwest Jordan. Earth Planet. Sci. Lett. 63, 292–304.

Jarrar, G., Wachendorf, H., Zellmer, 1991. The SaramujConglomerate: evolution of a Pan-African molasse sequencefrom southwest Jordan. N. Jb. Geol. Palaeont. Mh. H6, 335–356.

Jarrar, G., Wachendorf, H., Saffarini, G., 1992. A lateProterozoic Bimodal volcanic/subvolcanic suite from WadiAraba, Southwest Jordan. Precambrian Res. 56, 51–72.

Jarrar, G., Wachendorf, H., Zachmann, D., 1993. A Pan-Africanpluton intruding the Saramuj Conglomerate, South-west Jordan.Geol. Rundsch. 82, 121–135.


Jarrar, G., Manton, W., Stern, R.J. Late Neoproterozoic A-typegranites in the Northernmost Arabian-Nubian Shield formedby fractional crystallization of basaltic melts. Int. J. Earth Sci.,submitted for publication.

Kröner, A., Eyal, M., Eyal, Y., 1990. Early Pan-African evolutionof the basement around Elat, Israel, and the Sinai Peninsularevealed by single-zircon evaporation dating, and implicationsfor crustal accretion rates. Geology 18, 545–548.

Lenz, H., Bender, F., Besang, C., Harre, W., Kreuzer, H., Müller,P., Wendt, I., 1972. The age of the early tectonic events in thezone of Jordan geostructure based on radiometric data. 24thInt. Geol. Cong. Sec. 3, 371–379.

McCourt, W.J., Ibrahim, K.M., 1990. The geology, geochemistryand tectonic setting of the granitic and associated rocks inthe Aqaba and Araba complexes of southwest Jordan. BulletinGeological Mapping Division 10. Natural Resources Authority,Amman, Jordan, 96 pp.

McGuire, A.V., Stern, R.J., 1993. Granulite xenoliths fromwestern Saudi Arabia: the lower crust of the Late PrecambrianArabian-Nubian Shield. Contrib. Miner. Petrol. 114, 395–408.

Mittelfeldt, D.W., Reymer, A.P.S., 1986. Sinai granites: adiscussion of their origin based on petrological and Sr-isotopicconstraints. Isr. J. Earth Sci. 35, 40–50.

Moghazi, A.M., Anderson, T., Oweiss, G.A., EL Bouseily, A.M.,1998. Geochemical and Sr–Nd–Pb isotopic data bearing onthe origin of Pan-African granitoids in the Kid area, southeastSinai, Egypt. J. Geol. Soc. Lond. 155, 697–710.

Nelson, B.K., DePaolo, D.J., 1984. Rapid production of continentalcrust 1.7 to 1.9 b.y. ago: Nd isotopic evidence from the basementof the North American mid-continent. Geol. Soc. Am. Bull.96, 746–754.

O’Nions, R.K., Carter, S.R., Evenson, N.M., Hamilton, P.J., 1979.Geochemical and comsochemical applications of Nd isotopeanalyses. Ann. Rev. Earth Planet. Sci. 7, 11–38.

Peacock, S.M., 1990. Numerical simulation of metamorphicpressure–temperature–time paths and fluid production insubducting slabs. Tectonics 9, 1197–1211.

Pearce, J.A., 1983. The role of the sub-continental lithosphere inmagma genesis at active continental margins. In: Hawkeswoth,C.J., Norry, M.J. (Eds.), Continental Basalts and MantleXenoliths. Shiva Publishing Ltd., Cheshire, pp. 230–249.

Pearce, J.A., 1996. Sources and settings of granitic rocks. Episodes19, 120–125.

Pearce, J.A., Harris, N.B., Tindle, A.G., 1984. Trace elementsdiscrimination diagrams for the tectonic interpretation ofgranitic rocks. J. Petrol. 25, 956–983.

Ramsay, C.R., Stoeser, D.B., Drysdall, A.R., 1986. Guidelinesto classification and nomenclature of Arabian felsic plutonicrocks. J. Afr. Earth Sci. 4, 13–20.

Rapp, R.P., Watson, E.B., Miller, C.F., 1991. Partial melting ofamphibolite eclogite and the origin of Archean trondhjemitesand tonalities. Precambrian Res. 51, 1–25.

Rickwood, P.C., 1989. Boundary lines within petrologic diagramswhich use oxides of major and minor elements. Lithos 22,247–263.

Rogers, J.J.W., Greenberg, J.K., 1990. Late-orogenic, post-oro-genic, and anorogenic granites: distribution by major element

and trace element chemistry and possible origins. J. Geol. 98,291–309.

Schmidt, M.W., 1996. Experimental constraints on recycling ofpotassium from subducted oceanic crust. Science 272, 1927–1930.

Springer, W., Seck, H.A., 1997. Partial fusion of basic granulitesat 5 to 15 kbar: implications for the origin of TTG magmas.Contrib. Miner. Petrol. 127, 30–45.

Stern, R.J., 1994. Arc assembly and continental collision inthe Neoproterozoic East African Orogen: implications for theconsolidation of Gondwana land. Annu. Rev. Earth Planet. Sci.22, 319–351.

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

Stern, R.J., Gottfried, D., 1986. Petrogenesis of a late Precambrian(575–600 Ma) bimodal suite in northeast Africa. Contrib. Miner.Petrol. 92, 492–501.

Stern, R.J., Kröner, A., 1993. Late Precambrian crustal evolutionin the NE Sudan: isotopic and geochronologic constraints. J.Geol. 101, 555–574.

Stoeser, D.B., 1986. Distribution and tectonic setting of plutonicrocks of the Arabian Shield. In: Drysdall, A.R., Ramsay, C.R.,Stoeser, D.B. (Eds.), Felsic Plutonic Rocks and AssociatedMineralization of the Kingdom of Saudi Arabia. J. Afr. EarthSci. 4, 21–46.

Streckeisen, A., 1976. To each plutonic rock its proper name.Earth Sci. Rev. 12, 1–33.

Streckeisen, A., Le Maitre, R.W., 1979. A chemical approximationto the modal QAPF classification of the igneous rocks. NeuesJahrb. Miner. Abh. 136, 169–206.

Sun S.-S., McDonough, W.F., 1989. Chemical and isotopicsystematics of the oceanic basalts: implications for mantlecomposition and processes. In: Saunder, A.D., Norry, M.J.(Eds.), Magmatism in the Oceanic Basins. Geological Society,London, pp. 313–345.

Sylvester, P.J., 1989. Post-collisional alkaline granites. J. Geol. 97,261–280.

Sylvester, P.J., 1998. Post-collisional strongly peraluminousgranites. Lithos 45, 29–44.

Tarney, J., Jones, C.E., 1994. Trace element geochemistry oforogenic igneous rocks and crustal growth models. J. Geol.Soc. Lond. 151, 855–868.

Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: ItsComposition and Evolution. Blackwell, Oxford, 312 pp.

Turner, S., Arnaud, N., Liu, J., Rogers, N., Hawkesworth, C.,Harris, N., Kelly, S., Van Calsteren, P., Deng, W., 1996.Post-collision, shoshonitic volcanism on the Tibetian plateau:implications for the convective thinning of the lithosphere andthe source of ocean island basalts. J. Petrol. 37, 45–71.

Unrug, R., 1996. The assembly of Gondwanaland. Episodes 19,11–19.

Whalen, J.B., Currie, K.L., Chappell, B.W., 1987. A-type granites:geochemical characteristics, discrimination and petrogenesis.Contrib. Miner. Petrol. 95, 407–419.

Wilson, M., 1989. Igneous Petrogenesis. Unwin Hyman, London,466 pp.

Windley, B.F., 1993. Proterozoic anorogenic magmatism and itsorogenic connections. J. Geol. Soc. Lond. 150, 39–50.

Late- and post-orogenic Neoproterozoic intrusions of ...rjstern/pdfs/Jarraretal.pdfsense of...

Documents

Transcript of Late- and post-orogenic Neoproterozoic intrusions of ...rjstern/pdfs/Jarraretal.pdfsense of...