Author's personal copy

Available online at www.sciencedirect.com

Precambrian Research 160 (2008) 127–141

Palaeoproterozoic to Neoproterozoic growth and evolution of theeastern Congo Craton: Its role in the Rodinia puzzle

B. De Waele a,∗, S.P. Johnson b, S.A. Pisarevsky a

a Tectonics Special Research Centre, School of Earth and Geographical Sciences, The University of Western Australia,35 Stirling Highway, Crawley, WA 6009, Australia

b Institute for Research on Earth Evolution, Japan Agency for Marine-Earth Science and Technology,2-15 Natsushima-cho, Yokosuka 237-0061, Japan

Received 21 December 2005; received in revised form 5 December 2006; accepted 19 April 2007

Abstract

The Central African Cratons comprise various Archaean and Palaeoproterozoic blocks, flanked or truncated by orogenic belts ranging in agefrom Palaeoproterozoic (Rusizian, Ubendian and Usagaran Belts) to Mesoproterozoic (Kibaran and Irumide Belts). These various orogenic systemsmap out the progressive nucleation of the Central African Cratons to form the Congo Craton, which during late Neoproterozoic times participated invarious collisional processes to form part of the Gondwana supercontinent. Subsequently, the opening of the South Atlantic separated a small portionfrom the Congo Craton, which now forms part of the South American cratonic assemblage and is referred to as the Sao Francisco Craton. The originalcontinuity of the Sao Francisco and Congo Craton is supported by similarities in basement ages and craton stabilisation during Eburnean-agedtectonothermal events and the recognition of the original unity of the Aracuai and West Congo Belts and the Sergipane and Oubanguide Belts acrossthe Atlantic. The nucleation of the Congo Craton from its composing cratonic blocks, which include the Angola-Kasai Block, the NE-Congo-UgandaBlock and the Cameroon-Gabon-Congo-Sao Francisco Block to the west and northwest of the Mesoproterozoic Kibaran Belt, and the BangweuluBlock and Tanzania Craton, to the east and southeast, was at the latest completed after peak compressional tectonism in the Kibaran Belt at 1.38 Ga.Late Mesoproterozoic tectonism along the southern margin of this proto-Congo Craton, in a region called the Irumide Belt, marks compressionaltectonism at ca. 1.05–1.02 Ga, which produced extensive reworking along this margin, possibly linked to the participation of the Congo Craton inthe Rodinia Supercontinent. At present, insufficient evidence is available to support or deny the participation of the Congo Craton in Rodinia.

During the early Neoproterozoic, several rifting events occurred along the southern margin of the Congo Craton, in the Lufilian and ZambeziBelts, with localised volcanism and deposition of clastic sequences (Roan and Mwashya Groups), and followed by passive margin sedimentation(Nguba and Kundelungu Group). These sequences also contain large diamictite horizons (Grand and Petit Conglomerat). At ca. 570–530 Ma,convergence with the Kalahari Craton to the south and the Malagasy-Indian Cratons to the East culminated in collisional processes that formed theDamara-Lufilian-Zambezi and the East African Orogens, and led to the formation of Gondwana.© 2007 Elsevier B.V. All rights reserved.

Keywords: Congo Craton; Irumide Belt; Kibaran; Rodinia; Palaeomagnetism; Supercontinent

1. Introduction

The Congo Craton is defined as the amalgamated cen-tral African landmass at the time of Gondwana assembly (ca.550 Ma). Along its southern and eastern margins it comprisesvarious Archaean blocks including the Angola-Kasai Block andthe Tanzania Craton (Fig. 1). These Archaean units have beenextensively affected by Palaeoproterozoic events between 2.2and 1.9 Ga, collectively referred to as the “Eburnean” events,

∗ Corresponding author.E-mail address: bdewaele@bgs.ac.uk (B. De Waele).

and which may also have reworked a cryptic Archaean ter-rane to form the Palaeoproterozoic Bangweulu Block (DeWaele, 2005). By Palaeoproterozoic times, the various blocksof the proto-Congo Craton had stabilised and were subsequentlyaffected by Mesoproterozoic convergent tectonism forming theKibaran, Irumide and Southern Irumide Belts. During the Mid-Neoproterozoic, rift successions and subsequent passive margindeposits developed along the southern margin of the Congo Cra-ton prior to collisional events leading to the amalgamation ofGondwana (Johnson et al., 2005). In this paper, we will reviewthe Palaeo- to Neoproterozoic tectonic evolution of South-Central Africa where (1) Palaeoproterozoic events formedthe Rusizian-Ubendian Belt along western and south-western

0301-9268/$ – see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.precamres.2007.04.020

Author's personal copy

128 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

Fig. 1. Simplified geological map of Sub-Saharan Africa (based on Hanson, 2003). Abbreviations as follows: CKB, Choma Kalomo Block; Gab. Belt, Gabon Belt;LM, Lake Malawi; LT, Lake Tanganyika; LV, Lake Victoria; MB, Magondi belt; NE Kib. Belt, Northeast Kibaran Belt; R. Belt, Ruwenzori Belt; Rus., Rusizian; SF,Sao Francisco Craton; S.M., Southern Malawi; Ub. Belt, Ubendian Belt; Us. B, Usagaran Belt. Note that the Kilimafedha and Sukumaland Greenstone Belts are notdistinguished from the Tanzania Craton.

margins of the Tanzania Craton, and the Usagaran Belt along itssouthern margin; (2) Mesoproterozoic convergence first formedthe Kibaran Belt in between the Tanzania-Bangweulu Blockand the Angola-Kasai Blocks and then the Irumide and South-ern Irumide Belts along the southern margin of the craton; (3)Neoproterozoic rifting events formed the copper-bearing succes-sions of the Central African Copperbelt. These tectonic eventswill be viewed in the context of the supercontinent cycle, and wewill investigate the possible participation of the Central Africanblocks in the Rodinia supercontinent.

2. Make-up of the eastern Congo Craton

2.1. The Angola-Kasai Block

The Angola-Kasai Block comprises two poorly knownregions of central Africa, the Angola Block in southern and cen-tral Angola, and the Kasai Block, straddling the borders betweenAngola, the Democratic Republic of Congo (hereafter DRC) andZambia (Fig. 1). These two blocks are separated by a region of

extensive Phanerozoic cover of the Congo Basin, which togetherwith the scarcity of data for either region, makes correlationimpossible.

The Angola Block comprises a magmatic complex oforthogneisses and a metasedimentary complex of quartziteand schist, affected by granulite facies metamorphism at ca.2.8–2.7 Ga (Trompette, 1994). In the south of the complex, innorthern Namibia and southern Angola, these basement rockscomprise Archaean to Siderian protoliths (2645–2464 Ma, Sethet al., 1998; Delor et al., 2006) extensively affected by migmati-sation events between 2.29 and 1.85 Ga (Cahen et al., 1984; Sethet al., 1998), and accompanied by intrusion of granitoids datedin southern Angola using zircon U–Pb sensitive high-mass reso-lution ion microprobe (SHRIMP) at 2038 ± 28 and 1959 ± 6 Ma(McCourt et al., 2004), 1987–1968 Ma (Delor et al., 2006) and inNamibia by U–Pb or Pb–Pb isotope dilution thermal ionisationmass spectrometry (ID-TIMS) at 1985–1960 Ma. Sedimentationfollows these events with the deposition of the Chela Group, inwhich an ignimbrite yielded a zircon U–Pb SHRIMP age of1790 ± 17 Ma (McCourt et al., 2004).

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 129

The Kasai Block consists of a complex of granitic rocks yield-ing whole-rock Rb–Sr dates of between 3.49 and 3.33 Ga, whichhave been metamorphosed in the granulite facies at ca. 2.8 Ga(Cahen et al., 1984). In northwestern Zambia, Key et al. (2001)reported zircon U–Pb SHRIMP crystallisation ages of 2561 ± 10and 2538 ± 10 Ma for granitic gneisses, providing the only reli-able U–Pb age constraints for the Kasai Block. The same authorsalso reported a zircon crystallisation age of 2058 ± 7 Ma for aporphyritic granite intruding these Archaean gneisses, indicat-ing that an Eburnean-aged event, similar to that affecting theAngola Block to the West, also affected the Kasai Block.

2.2. The Tanzania Craton

The Tanzania Craton is almost entirely constituted of gran-itoids and greenstone belts, the latter of which are referred toin the south (central Tanzania) as the Dodoman, and in thenorth (northern Tanzania, Kenya and southeast Uganda) as theNyanzian and Kavirondian systems (Cahen et al., 1984). Thegranitoids and gneisses yielded whole-rock Rb–Sr dates rangingfrom 2.7 to 2.5 Ga, with some scattered reports of age deter-minations up to 3.12 Ga possibly attesting to Mesoarchaeancomponents (Cahen et al., 1984). Recent zircon U–Pb SHRIMPdeterminations on granulite facies granitoids and gneisses inthe Mozambique Belt of central Tanzania confirmed emplace-ment ages of 2.7 Ga, with xenocrystic components of ca. 3.0 and3.5 Ga (Muhongo et al., 2001; Johnson et al., 2003; Cutten etal., 2006), while 2.7 Ga zircon from an orthogneiss has also beenreported from rocks within the Usagaran Belt to the south (Reddyet al., 2003). U–Pb dating by laser ablation inductively cou-pled plasma mass spectrometry (LA-ICPMS) of igneous zirconsextracted from rhyolites and granitoids within the KilimafedhaGreenstone Belt (eastern Tanzania Craton) indicate that volcan-ism and granitoid intrusion occurred between ca. 2.72–2.71 and2.69–2.65 Ga, respectively (Wirth, 2004). Positive initial εNdwhole rock Sm–Nd isotopes of both the rhyolites and granitoidsindicate derivation from an upper mantle-like source withoutsignificant involvement of older basement, i.e. an oceanic arcsetting. These ages and initial isotopic ratios are very similar tothose obtained for the upper parts of the Sukumaland GreenstoneBelt further west, which have been dated by isotope dilutionthermal ionisation mass spectrometry (ID-TIMS) on zircon atca. 2.65 Ga (Borg and Krogh, 1999), although the lower parts ofthis greenstone belt appear to be significantly older at ca. 2.8 Gabased on a whole-rock Sm–Nd age on mafic volcanics (Manyaand Maboko, 2003). Positive initial εNd whole rock Sm–Nd iso-topes for the basaltic lithologies (Manya and Maboko, 2003) alsosupport a supra-subduction origin for this greenstone belt. Ndisotopic ratios of the Sukumaland metasedimentary lithologiesprovide evidence for an older crustal sedimentary provenanceof ca. 3.0–3.1 Ga (Cloutier et al., 2005).

2.3. The Bangweulu Block

The Bangweulu Block forms a crystalline unit of grani-toids and coeval volcanic rocks, unconformably overlain by acontinental clastic succession of fluvial and lacustrine conglom-

erates, quartzites and siltstones called the Mporokoso Group,with minor volcanic intercalations near the base (Fig. 2). Theblock adjoins the Archaean Tanzania Craton along the Palaeo-proterozoic Ubendian Belt (Fig. 1), in which granulite faciesmetamorphism, constrained through zircon U–Pb and Ar–Ardating at ca. 2.0 Ga (Ring et al., 1997; Boven et al., 1999), wasinterpreted to mark collisional processes between the Bang-weulu and Tanzania Blocks. Despite this, lithologies on theBangweulu Block yielded dates no older than 1.9 Ga (Breweret al., 1979; Schandelmeier, 1980, 1983), indicating that the ter-rain underwent significant crustal reworking during Ubendiantectonism. Rainaud et al. (2003) reported a significant compo-nent of xenocrystic zircons with U–Pb SHRIMP crystallisationages of ca. 3.2 Ga in a tuff within the Neoproterozoic LufilianBelt to the West, and proposed that metasediments in the Zam-bian and DRC Copperbelt region, and possibly the BangweuluBlock, are underlain by a Mesoarchaean terrane they termed theLikasi Terrane. One granite gneiss, which forms part of the base-ment within the Irumide Belt along the southern margin of theBangweulu Block, yielded a zircon U–Pb SHRIMP crystallisa-tion age of 2726 ± 36 Ma, and so far represents the only directevidence of Archaean crust in the region (De Waele, 2005; DeWaele et al., 2006a,b).

3. Palaeoproterozoic stabilisation of the proto-CongoCraton

A system of Palaeoproterozoic belts can be traced fromthe south-east margin of the Tanzania Craton (Usagaran Belt),passing between the Bangweulu Block and Tanzania Craton(Ubendian Belt), and northwards into the MesoproterozoicKibaran Belt (Rusizian Rise), marking an important tectonother-mal event that could record the amalgamation of the TanzaniaCraton and the Bangweulu Block (Fig. 1). Similar high-gradethermal events have been reported for the Kasai and AngolaBlocks (Cahen et al., 1984), but because of a lack of reliabledata neither of these regions are further discussed below.

3.1. The Usagaran Belt

The Usagaran Belt is comprised of granitoid gneisses,metasedimentary rocks and high-pressure eclogite facies rocksthat occupy the southern and southeastern corner of the Tanza-nia Craton (Fig. 1). The Konse Group (formally Konse Series)is comprised of various greenschist-grade metasediments andmetavolcanics that directly overlie the Tanzania Craton and theadjacent Isimani Suite (Mruma, 1995). They are tentatively cor-related with the Ndembera volcanics that have a zircon U–PbSHRIMP age of 1921 ± 14 Ma (Sommer et al., 2005). The Isi-mani Suite lies to the east of the Konse Group and is comprisedof upper amphibolite, granulite and eclogite grade metasedimen-tary gneisses, orthogneisses and mafic rocks with N-MORB-likegeochemical compositions (Moller et al., 1995). Detrital zirconanalyses of the metasedimentary gneisses indicate they weresourced from the Tanzania Craton and a 2.6–2.4 Ga sourceregion similar to that of reworked rocks within the East AfricanOrogen (Collins et al., 2004) and an orthogneiss within the

Author's personal copy

130 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

Fig. 2. Simplified geological map of the region outlined in Fig. 1. Abbreviations are as follows: inset political boundary map: DRC, Democratic Republic of Congo;MAL, Malawi; MOZ, Mozambique; TAN, Tanzania; ZIM, Zimbabwe; main map: Chp, Chipata Terrane; C.I, Chewore Inliers; C-R, Chewore-Rufunsa Terrane;HGM, Hook Granite Massif; KnG, Kanona Group; KsF, Kasama Formation; LG, Lusaka Granite; L-N, Luangwa-Nyimba Terrane; LTN, Lake Tanganyika; MfG:Mafingi Group; MrG, Manshya River Group; M.S.Z, Mwembeshi Shear Zone; NgG, Ngoma Gneiss; N.S.Z, Nyamadzi Shear Zone; P-S, Petauke-Sinda Terrane.Southern Irumide Belt divisions after Mapani et al. (2001) and Johnson et al. (2006).

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 131

sequence has been zircon-dated by SHRIMP at 2705 ± 11 Ma(Reddy et al., 2003). Most importantly, the Isimani Suite con-tains N-MORB-type mafic rocks that have been metamorphosedunder eclogite facies conditions of 750 ◦C and 1.8 GPa (Molleret al., 1995) and which were quickly cooled and exhumed to theamphibolite facies (Collins et al., 2004). This subduction-relatedtectonic event is robustly dated at 1999.5 ± 1.4 Ma by ID-TIMSon monazite (Moller et al., 1995) and at 1999.1 ± 1.1 Ma bySHRIMP on zircon (Collins et al., 2004). The Isimani Suitewas exhumed to mid crustal levels by 1996 ± 2 Ma and wasexposed at the surface by 1991 ± 2 Ma (Collins et al., 2004).The age of the post-tectonic Kidete Granite, dated by zirconU–Pb SHRIMP at 1877 ± 7 Ma (Reddy et al., 2003), confirmsthat tectonometamorphism was complete by this time.

The Usagaran Belt to the south and southeast is dominatedby granitoid orthogneisses that were partially derived from thereworking and recycling of the Tanzania Craton. Few of thesegranitoid gneisses have been precisely dated but the majorityhave whole rock Sm–Nd isotopes consistent with the mixingof mantle-derived material with Archaean crust of the Tanza-nian Craton. Post-tectonic volcanics of the Ndembera Seriesand various post-tectonic granitoids have been dated using zir-con U–Pb SHRIMP at 1921 ± 16, 1910 ± 11, 1824 ± 17 and1817 ± 12 Ma (Sommer et al., 2005) indicating that Usagaranconvergent-margin tectonometamorphism was complete by ca.1920 Ma. However, the similarity in ages of crustal componentson either side of this supposed suture zone, i.e. Tanzania Cra-ton and Isimani Suite to the west of the eclogitic rocks, and theWestern Mozambique Belt to the east, led Reddy et al. (2003)to suggest that only a relatively narrow, Red Sea-type oceanicbasin was closed and subducted, leading to a similar pre-riftconfiguration of crustal components.

3.2. The Ubendian Belt

The Ubendian Belt forms an elongate stretch of granulite–amphibolite facies gneisses and metasedimentary rocks inbetween the Tanzania Craton and the Bangweulu Block, and isexposed in northern Malawi, along the Zambia–Tanzania border,and along the shores of Lake Tanganyika (Fig. 1). The belt hasbeen subdivided into various elongated blocks, which are boundby NW–SE oriented faults or shear zones, and which have con-trasting lithologies and structural features (McConnell, 1972;Daly, 1988; Daly et al., 1989; Lenoir et al., 1994; Theunissenet al., 1996). The structural trends within these blocks gener-ally follow a NW–SE trend, and developed under amphibolitefacies conditions (Sklyarov et al., 1998). In some blocks, an E–Wtrending structure associated with granulite facies metamor-phism has been reported, which is truncated by, and thus predatesthe NW–SE shear zones. Local occurrences of eclogite-faciesassemblages have also been reported (Sklyarov et al., 1998). TheUbendian Belt is therefore considered to have undergone a two-stage tectonic evolution, which, although diachronous, could beclosely linked to the development of the more easterly UsagaranBelt (Daly, 1988; Lenoir et al., 1994). The earliest deforma-tion occurred between 2.1 and 2.0 Ga, coeval with collisionaltectonics in the Usagaran Belt, and resulted in granulite-grade

metamorphism, preserved in the southern part of the Uben-dian Belt (Lenoir et al., 1994; Ring et al., 1997). A secondevent occurred at around 1.86 Ga, and involved the emplace-ment of numerous granitoids (Lenoir et al., 1994), and resultedin the exhumation of high-pressure granulites and eclogites forwhich a 40Ar–39Ar barroisite cooling age of 1848 ± 6 Ma wasreported (Boven et al., 1999). Widespread dextral and sinistralstrike-slip shearing has been interpreted as being: (1) Palaeo-proterozoic, Daly (1988) suggested that east–west thrusting inthe Usagaran Belt was linked to shear zone development in theUbendian Belt; (2) Late Mesoproterozoic, coeval with the Iru-mide Orogeny (Ring, 1993); (3) Late Neoproterozoic and anintra-continental response to the East African Orogeny (Ring etal., 2002).

4. Mesoproterozoic tectonic record on the CongoCraton

4.1. The Kibaran Belt

The Kibaran Belt forms a linear NE–SW oriented terraneof amphibolite-grade rocks in between the Tanzania Cra-ton/Bangweulu Block to the south and east, and the Kasai/NECongo–Uganda Block to the west and north (Fig. 1). Therecognition of a basement culmination in the eastern DRC,referred to as the Rusizian (Fig. 1) (Gerards, 1971; Cahen etal., 1984; Lavreau, 1985), and partial truncation by the Ceno-zoic rift of Lake Tanganyika, led to the subdivision of thebelt into the Kibaran Belt s.s. of DRC, and the NE KibaranBelt of eastern DRC, Rwanda, Burundi and northwestern Tan-zania (Fig. 1) (Tack et al., 1994). In the past, four separatemagmatic events were considered to have occurred during theevolution of the Kibaran Belt, based on whole-rock Rb–Sr data(Klerkx et al., 1984; Fernandez-Alonso et al., 1986; Klerkx,1987). However, subsequent zircon U–Pb geochronological datahave refuted these earlier subdivisions, and have increasinglydemonstrated that both parts of the belt underwent a broadlyparallel development. Both sections of the Kibaran Belt con-tain at least two metasedimentary successions, the oldest ofwhich was intruded by S-type granitoids in the north and byintermediate-felsic I-type plutons in the south both of which aredated between 1.38 and 1.37 Ga (Tack et al., 2002; Kokonyangiet al., 2004a). In the Kibaran Belt s.s. this magmatic phase hasbeen linked to supra-subduction magmatism (Kokonyangi etal., 2004a,b, 2005, 2006) while in the NE Kibaran Belt thismagmatic/thermal event has been linked to extensional detach-ment/collapse based on the noted absence of a compressionalphase in the structural record (Klerkx, 1987; Klerkx et al., 1993;Fernandez-Alonso and Theunissen, 1998). A second tectonicevent affected both sedimentary successions, and is recordedin the Kibaran Belt s.s., with P–T conditions of 740–780 ◦Cand 6.0–6.5 kbar (Kokonyangi et al., 2004a). A zircon U–PbSHRIMP age of 1079 ± 14 Ma for a syn-kinematic granitoidassociated with this event, was interpreted by Kokonyangi etal. (2004a,b) to date this compressional tectonic phase. In theNE Kibaran Belt, limited shear-bounded A-type magmatism hasbeen recorded in the Kabanga-Musongati mafic and ultramafic

Author's personal copy

132 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

alignment (KM hereafter), for which for one intrusion a previousbulk zircon U–Pb age determination of ∼1275 Ma (Tack et al.,1994) has recently been revised by a zircon U–Pb SHRIMP ageof 1205 ± 19 Ma for the same intrusion (Tack et al., 2002). Thislatter age could indicate a genetic link between the emplace-ment of the KM along lateral shear zones in the NE KibaranBelt and compressional tectonics recorded in the Kibaran Belts.s. by Kokonyangi et al. (2004a,b). Also in the NE KibaranBelt, post-kinematic Sn-bearing granitoids were emplaced alongthese ca. 1200 Ma shear zones, which have been recently datedby SHRIMP at 987 ± 6 Ma (Tack, pers. comm.) refining earlierage estimations of ∼1.0 Ga by the Rb–Sr method (Cahen andLedent, 1979; Lavreau and Liegeois, 1982).

4.2. The Irumide Belt

The Mesoproterozoic Irumide Belt is situated along thesouthern margin of the Palaeoproterozoic Bangweulu Block(Figs. 1 and 2), and is comprised of a deformed basement,folded metasedimentary units and voluminous granitoid intru-sions. Exposed parts of deformed basement have yieldedU–Pb SHRIMP crystallisation ages ranging from 2049 ± 6 to1927 ± 10 Ma (Armstrong et al., 1999; Rainaud et al., 1999,2002, 2003; De Waele, 2005; De Waele et al., 2006a,b), withone granite gneiss in the southwestern most part of the IrumideBelt yielding an Archaean age of 2726 ± 36 Ma (Fig. 2) (DeWaele, 2005; De Waele et al., 2006a,b). Sm–Nd data on two ofthese basement gneisses yielded TDM model ages of 3.1–3.2 Ga,indicating a significant crustal residence (De Waele, 2005; DeWaele et al., 2006b). This basement is unconformably, and inplaces structurally, overlain by a metasedimentary successionof shallow marine quartzites and pelites with minor conglomer-atic horizons, referred to as the Manshya River/Kanona Group(Fig. 2) (De Waele and Mapani, 2002). Detrital zircon U–Pbage data for several quartzites from the Manshya River/KanonaGroups (Rainaud et al., 2003; De Waele and Fitzsimons, 2004,2007; De Waele, 2005) and a quartzite from near the base of theMporokoso Group on the Bangweulu Block to the north (Fig. 2),indicate the maximum age of deposition for these successions tobe ca. 1.8 Ga, while discrete water-lain volcanic units within theManshya River Group in the northeastern part of the Irumide Beltyielded direct constraints of deposition between 1879 ± 13 and1856 ± 4 Ma (De Waele and Fitzsimons, 2004, 2007; De Waele,2005). The TDM model ages of the volcanic units range between2.8 and 2.2 Ga indicating a mixture of juvenile and reworkedcrustal material in their generation (De Waele, 2005; De Waeleet al., 2006b). The Palaeoproterozoic basement and overlyingPalaeoproterozoic supracrustal units were locally intruded bygranitoids with a crust-dominated geochemical character withcrystallisation ages between 1664 ± 4 and 1551 ± 33 Ma, andTDM model ages between 3.2 and 2.8 Ga indicating significantcrustal recycling involved in their genesis (De Waele, 2005;De Waele et al., 2006b). The recognition of a small basin ofsupermature fluvial sandstones and minor siltstones on the Bang-weulu Block, with maximum age of deposition at 1434 ± 14 Ma,based on the youngest concordant U–Pb SHRIMP age of detri-tal zircon, and detrital patterns similar to those recorded in the

Mporokoso Group, indicates that the ∼1.6 Ga magmatic eventwas succeeded by a second cycle of sedimentation resulting inthe reworking of the Mporokoso Group (De Waele, 2005; DeWaele and Fitzsimons, 2004, 2007). Shear-controlled intrusionof A-type granitoids between 1119 ± 20 and 1087 ± 11 Ma (zir-con U–Pb SHRIMP, Ring et al., 1999) within the Ubendian Beltand K–Ar biotite ages as old as 1050 Ma along the LuongoFold Belt (Daly, 1986) (Fig. 2) herald the stirring of tecton-ism along the southern margin of the Bangweulu Block. Peakmetamorphism in the Irumide Belt is constrained at between1021 ± 16 and 1018 ± 5 Ma through U–Pb SHRIMP dating oflow Th/U zircon rim overgrowths (De Waele, 2005). This tec-tonic event was accompanied by the intrusion of numeroushigh-K granitoids between 1055 ± 13 and 1003 ± 31 Ma (DeWaele et al., 2003, 2006a; De Waele, 2005) with the bulk ofintrusions being contemporaneous with peak metamorphism atca. 1020 Ma. The geochemistry of these Irumide-age plutonsis dominated by crustal signatures, while Sm–Nd isotopic dataindicate TDM crustal residence ages of up to 3.3 Ga confirmingtheir recycled nature (De Waele et al., 2006b). The overall geo-chemical and Sm–Nd isotopic characteristics of all magmaticsuites recognised in the Irumide Belt are remarkably similar,suggesting that a broadly homogenous and similar crustal sourcehas been tapped for the generation of each of these magmas. Thedata therefore suggest that a cryptic Archaean terrane of fairlyuniform composition underlies the Irumide Belt, while the old-est TDM model ages of 3.3 Ga give some idea of its possibleage. Together with the 3.2 Ga xenocrystic zircon evidence pre-sented by Rainaud et al. (2003), this increasingly supports theexistence of a cryptic ca. 3.2 Ga basement terrane in the regionpossibly both underlying the Bangweulu Block and the IrumideBelt (Fig. 1). This cryptic basement may well represent the enig-matic block that collided with the Tanzania Craton between 2.0and 1.8 Ga.

4.3. The Southern Irumide Belt

The Southern Irumide Belt (SIB) is a relatively new term(Johnson et al., 2005, 2006, 2007b) used to describe litholo-gies that occur to the south of, and that are distinct from, themonotonous granitoids of the Irumide Belt (s.s.). The contactbetween the two belts is obscured by Permo-Triassic grabensand, in places the SIB has been strongly overprinted by Neopro-terozoic tectonometamorphism, where structural fabrics mergewith the Zambezi and Mozambique Belts (Figs. 1 and 2). InZambia, this belt has been subdivided into several distinct litho-tectonic terranes (Mapani et al., 2001; Johnson et al., 2006,2007b), some of which are bounded by discrete shear zones. Thewesternmost terrane, the Chewore-Rufunsa Terrane (Johnsonet al., 2006), is comprised of a wide variety of mafic to fel-sic gneisses and metavolcanic rocks that have calc-alkalinechemistries and trace element, Rare Earth Element (REE) andNd-isotope signatures that are consistent with them havingformed in a continental-margin-arc setting (Oliver et al., 1998;Johnson and Oliver, 2000, 2004; Johnson et al., 2006, 2007b).Magmatic activity in the arc is constrained between ca. 1095and 1040 Ma (Johnson and Oliver, 2004; Johnson et al., 2005)

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 133

although a distinct marginal basin ophiolite (the Chewore Ophi-olite) has been recognised in the Chewore Inliers of northernZimbabwe (Fig. 2) and dated at ca. 1393 Ma (Oliver et al.,1998). Geological investigations of the terranes that lie to theeast, the Luangwa-Nyimba, Petauke-Sinda and Chipata terranes(Fig. 2) are at a rudimentary level, but preliminary geochem-istry and lithofacies analyses suggest that these terranes mayrepresent the components of an accreted island arc complex ora continuation of the Chewore-Rufunsa continental-margin-arc(Mapani et al., 2001; Johnson et al., 2006, 2007b). Reconnais-sance SHRIMP geochronology of granitoids and felsic gneissesand migmatites within these terranes indicates that Mesopro-terozoic magmatism occurred in a similar time frame to thatin the Chewore-Rufunsa Terrane, ca. 1075–1010 Ma but thatthere are significant older Palaeoproterozoic (ca. 1.9 Ga) andyounger Pan African (ca. 0.7–0.5 Ga) components (Johnson etal., 2006, 2007b). SHRIMP dating of metamorphic zircon rimsand TIMS dating of metamorphic monazite extracted from high-temperature (>850 ◦C), low-pressure (<5 kbar) granulite facieslithologies from all the SIB terranes yielded ages between ca.1080 and 1046 Ma (Goscombe et al., 2000; Schenk and Appel,2001; Johnson et al., 2006, 2007b) which is contemporaneouswith the peak of magmatic activity in these terranes and consis-tent with magma loading of the crust (Schenk and Appel, 2001).At present there is no convincing evidence that any of these SIBterranes underwent late Mesoproterozoic compressional tecton-ism.

4.4. Southern Malawi and the Lurio Foreland

Directly to the east of the SIB, Southern Malawi and northernMozambique are dominated by Late Mesoproterozoic to EarlyNeoproterozoic calc-alkaline granitoids and gneisses. In South-ern Malawi these lithologies have yielded Pb-evaporation agesof between 1040 and 999 Ma, have positive to mildly negativeinitial εNd values and have been interpreted to have formed in acontinental-margin-arc setting (Kroner et al., 2001). In the LurioForeland of northern Mozambique similar calc-alkaline gneisseshave been dated by the SHRIMP and Pb-evaporation techniquesbetween ca. 1148 and 1009 Ma (Costa et al., 1994; Kroner et al.,1997; Kroner, 2001). These lithologies lack inherited Archaeanand Palaeoproterozoic zircons, and show positive to mildly neg-ative initial εNd values and have been interpreted to have formedin a mature island-arc-type setting (Jamal and De Wit, 2004).These rocks also lack any evidence for Mesoproterozoic colli-sional tectonism. Collins and Pisarevsky (2005) have suggestedthat the Lurio Foreland lithologies may represent a northerlyextending peninsula of the Kalahari Craton; however, consider-ing the ages and tectonic affinity of these lithologies they couldequally be related to the continental-margin-arc rocks of the SIB.

5. Neoproterozoic divergence

Neoproterozoic divergence along the southern margin ofthe Congo Craton is recorded within the Zambezi and Lufil-ian Belts as two-distinct rift-related volcano-sedimentary andpassive margin sequences (Figs. 2 and 3a–c). The first rift to

drift phase occurred between 880 and 820 Ma (Johnson et al.,2007a) and is coincident with the deposition of the extensivecopper-bearing Roan strata of the Lufilian Belt. The secondcycle lacks voluminous volcanic deposits but was initiated atca. 765 Ma (Key et al., 2001) and is comprised of the Mwashya,Nguba and Kundelungu strata of the Lufilian Belt including twoglobal diamictite horizons. This event records the formation anddevelopment of an extensive passive margin.

5.1. Rift cycle 1—the Roan and Zambezi supracrustal riftbasins

The onset of rifting is recorded in the Zambezi Belt bythe eruption of a 2500 m thick pile of felsic volcanics andvolcanoclastics, the Kafue Rhyolite and Nazingwe Forma-tions (Mallick, 1963; Smith, 1963). Three volcanic units fromdifferent parts of the volcanic stratigraphy have been datedby the SHRIMP at ca. 880 Ma (Fig. 3b) (Johnson et al.,2007a). The geochemical characteristics combined with neg-ative εNd(t) isotope ratios indicate that they were generated bythe mixing/assimilation of juvenile material with older base-ment gneisses (Johnson et al., 2007a), a scenario consistentwith continental thinning and extension. Although currentlyundated, potential felsic volcanic equivalents have been iden-tified from the Roan Group of the Lufilian Belt around theLuwishi Dome (Porada, 1989), as part of the R.A.T. (Cailteuxet al., 1994) and from the lower Roan rocks from southeastShaba in the Democratic Republic of Congo (Lefebvre, 1989).A granitoid gneiss with similar geochemical compositions to theKafue and Nazingwe Formation volcanics (Katongo et al., 2004)has been dated by zircon U–Pb SHRIMP at ca. 883 ± 10 Ma(Fig. 3a) (Armstrong et al., 1999, 2005). The rift to drift sed-imentary sequences in both the Zambezi and Lufilian Beltsrest unconformably upon these volcanics/basement gneisses(Mallick, 1966) indicating that there had been uplift and tiltingshortly after, or accompanying, volcanism/plutonism (Fig. 3band c). It is apparent that uplift and the formation of a dynamictopography, especially in the Copperbelt Region, was critical forcontrolling local oxidising–reducing conditions thus allowingthe deposition of the extensive syn-sedimentary Cu–Co miner-als (Binda, 1994; Cailteux et al., 2005; Sutton and Maynard,2005). The lithofacies assemblage of the Zambezi and LufilianBelt sedimentary sequences are remarkably similar (Fig. 3a–c)but the most striking feature are the presence of thousands ofrandomly oriented, isolated, variably metamorphosed but non-deformed gabbro, mafic and ultramafic blocks (Vrana et al.,1975) that occur within the upper marble sequence (the Chetaand Muzuma formations in the Zambezi supracrustal sequenceand the Bancroft Group in the Lufilian Belt; Fig. 3a–c). In theLufilian Belt this marble unit also contains large clasts of theunderlying sedimentary succession that along with the maficunits can be interpreted as olistostromes (Johnson et al., 2005). Inthe Lufilian Belt these mafic blocks display continental-within-plate and E-MORB-type chemistries (Tembo et al., 1999) whilstin the Zambezi Belt they have mid-ocean-ridge (N-MORB)chemistries and isotopic signatures (John, 2001; John et al.,2003, 2004b). In the Lufilian Belt they have been interpreted

Author's personal copy

134 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

Fig. 3. Summary diagram illustrating and comparing the various rift-related sedimentary sequences deposited along the southern margin of the Congo Craton (a, theCopperbelt region of Zambia; b, Lusaka region north of Mpande Dome; c, Lusaka region, south of Mpande Dome). Age data description of units are summarised inJohnson et al. (2007a). Abbreviations as follows: D/C, disconformity; KRF, Kafue Rhyolite Formation; MHG, Munali Hills Granite; N.C, Nega Conglomerates; NF,Nazingwe Formation; NgG, Ngoma Gneiss; T, Thrust; U/C, unconformity; volcs, Mesoproterozoic volcanics related to the Mpande Gneiss and Munali Hills Granite.

as dykes or sills intruded during the initial stages of continentalrifting but in the Zambezi Belt they are interpreted to be ves-tiges of oceanic crust either tectonically emplaced during PanAfrican collision tectonics (John et al., 2003) or as an ophioliticmelange at the time of deposition of the marble units (Johnson etal., 2005). The lack of age data for any of the mafic units makestheir interpretation difficult, but it is not beyond reason that allthe mafic bodies are tectonically related and represent disruptedand boudinaged syn-rifting sills/dykes (Johnson et al., 2007a).

In the Zambezi Belt the sediments are intruded by variousmetaluminous, monogranitic plutons (Katongo et al., 2004), twoof which have been dated at ca. 820 Ma (Fig. 3b and c) (Hansonet al., 1988; Johnson et al., 2007a); thus providing the youngestage for the upper part of the rift succession. The geochemicaland isotopic signatures of these granitoids indicate that theyoriginated by the complete recycling of sialic continental crust,but their tectonic significance is unclear (Johnson et al., 2007a).Such granitoids have yet to be identified in the Lufilian Belt.

5.2. Rift cycle 2—the Mwashya-Nguba and Kundelungu riftbasin

From 820 Ma there appears to have been a cessation in sed-imentation until the extrusion of the Luakela volcanics of theMwashya Sub-group at 765 ± 5 Ma (Key et al., 2001) (Fig. 3a),heralding renewed rifting in the Lufilian Belt (Porada andBerhorst, 2000). The Mwashya Sub-group is comprised of minormafic volcanics and felsic tuffs (Fig. 3a) and a series of coarseclastics with abundant olistostromes (Wendorff, 2005a,b), fol-

lowed by shallow-water argillites and carbonates (Porada andBerhorst, 2000). This group is directly overlain by a diamic-tite horizon known as the Grand Conglomerat, which is takento be the base of the Nguba Group (Fig. 3a). This diamictitehas a well-constrained age through volcanic horizons that occurin the overlying strata (735 ± 5 Ma zircon U–Pb SHRIMP age)and in the underlying Mwashya Sub-group (765 ± 5 Ma zirconU–Pb SHRIMP age, see Key et al., 2001). The remainder of theNguba Group is comprised of an extensive and rapidly deepen-ing passive margin sequence. The overlying Kundelungu Groupcontains a diamictite at its base known as the Petit Conglomerat,for which there are no direct age constraints; however, it has beensuggested it may correlate with the global Marinoan Glacialdeposited at around 635 Ma (Robb et al., 2002; Condon et al.,2005), which provides a tentative upper age limit for the sed-imentary basin. Subduction of oceanic crust and ocean basinclosure is dated by the age of eclogite facies metamorphism inthe Zambezi Belt between 659 and 595 Ma (John et al., 2004a)with orogenesis culminating in collision between the Congoand Kalahari Cratons at 530 and 520 Ma (data summarised inJohnson et al., 2005) during the assembly of Gondwana.

6. Palaeomagnetic constraints for the position of Congo

The Precambrian palaeomagnetic database of the Congo Cra-ton and its predecessors is relatively poor. A search in the latestversion of the Global Palaeomagnetic Database (Pisarevsky,2005) produced only few reliable data after filtering out poleswith large age uncertainty and low reliability. Poles with

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 135

Table 1Precambrian palaeomagnetic poles from Congo-Sao Francisco Craton

Object Age (Ma) Pole A95 (◦) Q Reference

◦N ◦E

CongoNyanzian Lavas, Kenya 2690–2670 14 150 6 IIIIIII 7 Meert et al. (1994b)Kisii Series, Kenya 2531 ± 3 7 345 13 II-I-I 4 Brock et al. (1972) and Pinna et al. (2000)Proterozoic Granites, Kenya 2476 ± 50 61 210 15 −I-III 4 Patel and Raja (1979)Nyabikere Massif, Burundi 980–920 43 137 11 I-I-I-I 4 Meert et al. (1994a)Gagwe Lavas, Tanzania 795 ± 7 25 93 10 III-I-I 5 Meert et al. (1995) and Deblond et al. (2001)Mbozi Complex, Tanzania 755 ± 25 46 325 9 III–III 6 Meert et al. (1995)Sinyai Metadolerite, Kenya 547 ± 4 29 139 4 III–III 6 Meert and Van der Voo (1996)

Sao FranciscoIlheus dykes 1011 ± 24 30 100 4 III-I-I 5 D’Agrella-Filho et al. (1990) and Renne et al. (1990)Olivenca dykes, normal ∼1035 16 107 8 III-I-I 5 D’Agrella-Filho et al. (1990) and Renne et al. (1990)Itaju de Colonia ∼1055 8 111 10 III-III 6 D’Agrella-Filho et al. (1990) and Renne et al. (1990)Olivenca dykes, Reverse 1078 ± 18 −10 100 9 III-I-I 5 D’Agrella-Filho et al. (1990) and Renne et al. (1990)

Q is the quality factor after Van der Voo (1990) and ranges from zero to seven, with the later representing the highest-quality data.

Q > 3 (Van der Voo, 1990) are shown in Table 1. A groupof Archaean–Early Palaeoproterozoic poles from Tanzania andKenya (Brock et al., 1972; Patel and Raja, 1979; Meert et al.,1994b), although scattered, indicates low to medium palaeo-latitudes for Congo at between 2.6 and 2.4 Ga. Unfortunatelythere are no coeval poles west of the Kibaran Belt to testthe pre-Kibaran configuration of Congo’s building blocks. The1.1–1.0 Ga time period is constrained by palaeomagnetic datafrom the Sao Francisco Craton, which is traditionally treatedas a western extension of Congo owing to the similarity ofArchaean and Palaeo- to Mesoproterozoic rocks and bound-ing late Neoproterozoic mobile belts (Fig. 1) (Trompette, 1994;Teixeira et al., 2000; Barbosa and Sabate, 2004; da Silva etal., 2005). Four palaeopoles from Sao Francisco (D’Agrella-Filho et al., 1990) together with a less reliable palaeomagneticpole from the ∼950 Ma Nyabikere massif in Burundi (Meert etal., 1994a) generate a slightly curved Apparent Polar WanderPath (APWP) that is anchored by hornblende Ar–Ar dates of1078 ± 18 and 1011 ± 24 Ma for the Olivenca Dykes and IlheusDykes, respectively (Fig. 4f) (Renne et al., 1990). A compari-son of this APWP with the coeval Laurentian “Keweenawan”APWP track led Pisarevsky et al. (2003) to suggest that it wasmore likely for the Congo Craton not to form part of Rodinia,although other options were still considered possible.

The positions of the Congo-Sao Francisco Craton at800–750 Ma is constrained by the 795 Ma Gagwe and 748 MaMbozi poles (Meert et al., 1995; Deblond et al., 2001). Accord-ing to these data, the Congo Craton remained near the equator,but rotated at about 90◦ between 800 and 750 Ma. More recently,a primary palaeomagnetic pole has been reported from the765 Ma Luakela Volcanics in NW Zambia (Wingate et al., 2005)which is very similar to the 748 Ma Mbozi pole. However, asecondary Luakela remanence of uncertain age resembles the795 Ma Gagwe remanence, casting some doubts on the assumedprimary nature of the Gagwe remanence. The 547 Ma pole fromthe Sinyai Dolerite in Kenya suggests that the assembly of proto-Gondwana was complete by that time (Meert and Van der Voo,1996).

7. Did the Congo Craton participate in Rodinia?

Geological evidence indicates that the southern margin ofthe Congo Craton underwent various compressional tectono-magmatic events during the time period associated with Rodiniaamalgamation; however, this data does not lend itself to a sin-gle or simple interpretation. The contact between the Irumideand Southern Irumide belts is obscured by Phanerozoic Karoograbens and a large Pan-African-aged crustal mega shear, theMwembeshi Shear Zone (Fig. 2). Most importantly, both beltsrecord different tectonothermal histories. Peak high-temperaturelow-pressure metamorphism in the Southern Irumide Beltappears to be contemporaneous with calc-alkaline continental-margin-arc magmatism between ca. 1090 and 1040 Ma (Johnsonet al., 2005, 2006, 2007b). The low-pressure nature of thesemetamorphic assemblages (<4.5 kbar) and lack of compres-sional field structures (folds and thrusts) indicate that there waslittle, if any, compressional tectonism associated with this event(Goscombe et al., 2000; Johnson et al., 2006, 2007a). In con-trast, tectonometamorphism in the Irumide Belt was dominatedby crustal thickening processes (∼7–8 kbar) as documentedby extensive isoclinal folding and thrusting, with large scalereworking of basement gneisses without the addition of juvenilecrust (De Waele, 2005; De Waele et al., 2006b). The peak ofthis magmatic and metamorphic episode post-dates that in theSouthern Irumide Belt by ca. 30 million years (i.e. ca. 1020 Ma,De Waele, 2005) and appears to have been initiated once mag-matism had ceased in the Southern Irumide Belt. Our favouredinterpretation of these events is that the Southern Irumide Beltdid not develop on the Congo Craton margin but represents theactive margin of a separate microcontinental block or craton,along which extensive oceanic crust was subducted ultimatelyleading to the collision of this terrane/craton with the CongoCraton margin at ca. 1020 Ma.

Although detailed palaeomagnetic information is not of afine-enough scale to determine actual plate velocities, the APWPprovided by the few poles in the time period being considered(1100–750 Ma) is adequate to show that (1) the Congo Craton

Author's personal copy

136 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

Fig. 4. Various Palaeogeographic reconstructions illustrating the two possible tectonic models proposed in the main text. Reconstructions were made using thePLATES and GMT software. Craton abbreviations as follows: A, Australia; Am, Amazonia; B, Baltica; C, Congo-Sao Francisco Craton; I, India; K, Kalahari; L,Laurentia; M, Mawson Craton (Antarctica); S, Siberia; WA, West Africa. Palaeomagnetic data used for these reconstructions are tabulated in Pisarevsky et al. (2003)and Pisarevsky and Natapov (2003). (a and e) 1050 Ma; (b and f) 1020 Ma; (c and g) 880 Ma; (d and h) 760 Ma.

was not stationary, (2) at the time of collision ca. 1020 Ma, thepotential position of the craton within Rodinia can be assessedusing the Olivenca and Ilheus palaeomagnetic poles and (3)between ca. 800 Ma (Gagwe pole) and ca. 750 Ma (Mbozi pole)

the craton underwent a significant 90◦ spin, suggesting thatduring this time the Congo Craton was separate from Rodiniain order to allow such a rotation. With these constraints inmind, several options remain possible, including a detached

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 137

and attached configuration with respect to Rodinia. These twocontrasting tectonic scenarios will now be discussed in detail.

7.1. The Congo Craton as an independent craton

The predominance of Late Mesoproterozoic supra-subduction related rocks in the Southern Irumide Belt and theLurio Foreland indicate that prior to collisional tectonism inthe Irumide Belt at ca. 1020 Ma, the southern margin of theCongo Craton faced an open ocean dominated by juvenile tomature island arcs with possible microcontinental fragments(Fig. 4a). If we presume that the Congo Craton developedindependently from Rodinia, then Irumide orogenesis may havebeen the result of accretion of these arcs/microcontinents to theCongo margin (Fig. 4b), a scenario not too dissimilar to theOrdovician-Silurian arc accretion/collision events along easternLaurentia, namely the Grampian and Appalachian events (e.g.Oliver, 2001; Murphy and Keppie, 2005). Continental-margin-arc magmatism in Southern Malawi dated between 1040 and999 Ma (Kroner et al., 2001), indicates that subduction-relatedmagmatism continued after arc-accretion (Fig. 4b) supportingthe hypothesis that Irumide orogenesis was not related to thecollision of this margin with Rodinia. The Gagwe palaeomag-netic pole dated at ca. 795 Ma indicates that the Congo Cratonhad drifted from mid- to equatorial-latitudes (Fig. 4c) and thensometime before ca. 755 Ma, the Mbozi Complex palaeomag-netic pole indicates that, the craton had undergone a significant90◦ rotation (Fig. 4d). This data is most compatible with afreely moving, Rodinia-independent Congo Craton between800 and 750 Ma but says nothing about the pre-800 positionof the craton. Rifting at ca. 880 Ma and again at ca. 765 Mato form the Roan-Zambezi and Mwashya-Nguba-Kundelungurift basins, respectively finally resulted in full continentalseparation by ca. 750 Ma, with the production of new oceaniccrust and allowing sufficient space for the 90◦ rotation of thecraton documented by the palaeomagnetic data (Fig. 4d).

7.2. The Congo Craton as part of Rodinia

The four palaeomagnetic poles from the Sao FranciscoCraton generate a slightly curved APWP that is anchored at1078 ± 18 and 1011 ± 24 Ma by the Olivenca and Ilheus dykes,respectively (Renne et al., 1990) (Fig. 4f). This APWP is similarto the Keeweenawan track of Laurentia (Pisarevsky et al., 2003)suggesting a common motion between the two and could beused to suggest that the Congo was an integral part of Rodinia.A comparison between the two APWP’s allows the possibility ofconstraining the Congo-Sao Francisco Craton’s position withinRodinia at the time of collision. Since both APWP’s have a lowcurvature, both polarity options (poles and antipoles) are per-missible (Fig. 4f). If one polarity option is accepted (position Ain Fig. 4f), the northern Congo-Sao Francisco margin is juxta-posed against Australia-Antarctica and causes overlap with theKalahari Craton (Fig. 4e and f) with the southern margin of theCongo Craton facing an open ocean. If the alternative polarityoption is accepted (position B in Fig. 4f) then the Irumide Beltis juxtaposed against the Rodinian margin but the craton itself

lies directly on top of West Africa. Trompette (1994, 1997) pro-posed the existence of a single West Africa-Amazonia-Rio de laPlata mega Craton in the Mesoproterozoic and Neoproterozoic.However, through applying minor movements between the WestAfrica and Amazonia Cratons a permissible fit can be proposedbetween the Congo and Amazonia Cratons (compare Fig. 4a andFig. 4g). If we accept this reconstruction then it is possible thatthe Southern Irumide Belt may represent a continental-margin-arc developed on the Amazonia Craton margin during oceanclosure (Fig. 4e). The Lurio Foreland arcs may have devel-oped as intra-oceanic arcs between the Congo and AmazoniaCratons but equally they may represent a northerly extendingpeninsula of the Kalahari Craton as suggested by Collins andPisarevsky (2005) (Fig. 4e). This position could also accountfor the post 1020 Ma supra-subduction magmatism recorded inSouthern Malawi.

The formation of the Roan Rift Basin at ca. 880 Ma wouldrepresent the earliest record of attempted break-up of Rodinia(Fig. 4g). At present, rifting of this age has not been recognised inany of the other Rodinian blocks (e.g. Li et al., 2004). Althoughnot conclusive, it is most likely that rifting at ca. 880 Ma neverresulted in the production of oceanic crust (Johnson et al., 2006,2007a,b), and we suggest that break-up of the Congo Cratonfrom Rodinia occurred during Mwashya rifting (ca. 765 Ma), ata similar time to other recorded Rodinia break-up events (sum-marised in Pisarevsky et al., 2003). In order for the Congo-SaoFrancisco Craton to reach equatorial latitudes and a rotationalaspect as constrained by the Mbozi Complex pole at ca. 755 Ma,10 million years after rifting (compare Fig. 4g with Fig. 4h),potentially unrealistic plate motions would be required.

8. Conclusions

Geological evidence indicates that the southern margin ofthe Congo Craton underwent compressional tectonics at ca.1020 Ma, which could be used to infer that the combined Congo-Sao Francisco Craton collided with, and became an integralpart of Rodinia. If the Congo-Sao Francisco Craton indeeddid form part of Rodinia, then two potential positions can beconstrained using available palaeomagnetic data. A careful con-sideration of these two positions either shows incompatibilitieswith geological evidence or overlap in position of other (betterconstrained) cratons, and we consider it unlikely that the Congo-Sao Francisco Craton ever formed part of Rodinia. The availablepalaeomagnetic data are also compatible with the scenario thatthe Congo-Sao Francisco Craton existed as an independent cra-ton that was not associated with Rodinia at all. We proposethat accretion of juvenile to mature island arcs and microcon-tinental fragments along the margin of the independent CongoCraton, could account for the formation of the 1020 Ma com-pressional Irumide Orogen. The current data do not allow usto unequivocally determine whether the Congo-Sao FranciscoCraton formed an integral part of the Rodinia supercontinent orwhether it acted as an independent craton.

More high-quality precisely dated palaeomagnetic polesfrom the Congo-Sao Francisco Craton could help resolve thisuncertainty. As one can see from Table 1, it would be espe-

Author's personal copy

138 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

cially helpful to obtain new palaeomagnetic data for the time of“stable” Rodinia around 900–800 Ma.

Acknowledgements

Reconstructions were made using the PLATES software fromthe University of Texas at Austin, and GMT software of Wesseland Smith. We are grateful to Grahame Oliver, Victor Bagdadseand Bruce Eglington for their constructive reviews. This is TSRCmanuscript number 363 and a contribution to ICGP 440.

References

Armstrong, R.A., Robb, L.J., Master, S., Kruger, F.J., Mumba, P.A.C.C., 1999.New U–Pb age constraints on the Katangan sequence, Central AfricanCopperbelt. In: Tembo, Francis, De Waele, B., Nkemba, Simon (Eds.), Pro-ceedings of the IGCP 2nd Fieldmeeting, vol. 418. Geological Society ofZambia, Kitwe, Zambia, pp. 48–49.

Armstrong, R.A., Master, S., Robb, L.J., 2005. Geochronology of the NchangaGranite, and constraints on the maximum age of the Katanga Supergroup,Zambian Copperbelt. J. Afr. Earth Sci. 42, 32–40.

Barbosa, J.S.F., Sabate, P., 2004. Archean and Paleoproterozoic crust of the SaoFrancisco Craton, Bahia, Brazil: geodynamic features. Precambrian Res.133, 1–27.

Binda, P.L., 1994. Stratigraphy of the Zambian Copperbelt orebodies. In: Kam-punzu, Ali Basira, Lubala, R.T. (Eds.), Neoproterozoic Belts of Zambia,Zaire and Namibia. J. Afr. Earth Sci., 251–264.

Borg, G., Krogh, T., 1999. Isotopic age data of single zircons from theArchaean Sukumaland Greenstone belt, Tanzania. J. Afr. Earth Sci. 29, 301–312.

Boven, A., Theunissen, K., Sklyarov, E., Klerkx, J., Melnikov, A., Mruma, A.,Punzalan, L., 1999. Timing of exhumation of a high-pressure granulite ter-rane of the Palaeoproterozoic Ubende belt (west Tanzania). PrecambrianRes. 93, 119–137.

Brewer, M.S., Haslam, H.W., Darbyshire, P.F.P., Davis, A.E., 1979. Rb–Sr agedeterminations in the Bangweulu block Luapula Province, Zambia. Report.Institute of Geological Sciences, London, May.

Brock, A., Raja, P.K.S., Vise, J.B., 1972. The Paleomagnetism of the Kisii Series,Western Kenya. Geophys. J. Roy. Astronom. Soc. 28 (2), 129–137.

Cahen, L., Ledent, D., 1979. Precisions sur l’age, la petrogenese et al positionstratigraphique des “granites a etain” de l’est de l’Afrique centrale. Bull.Soc. Belg. Geol. 88, 33–49.

Cahen, L., Snelling, N.J., Delhal, J., Vail, J.R., Bonhomme, M., Ledent, D.,1984. The Geochronology and Evolution of Africa. Oxford University Press,Oxford, p. 512.

Cailteux, J., Binda, P.L., Katekesha, W.M., Kampunzu, A.B., Intiomale, M.M.,Kapenda, D., Kaunda, C., Ngongo, K., Tshiauka, T., Wendorff, M., 1994.Lithostratigraphic correlation of the Neoproterozoic Roan Supergroup fromShaba (Zaire) and Zambia in the Central African copper–cobalt metallogenicprovince. J. Afr. Earth Sci. 19, 265–278.

Cailteux, J.H.L., Kampunzu, A.B., Lerouge, C., Kaputo, A.K., Milesi, J.P.,2005. Genesis of sediment-hosted stratiform copper–cobalt deposits, centralAfrican Copperbelt. J. Afr. Earth Sci. 42, 134–158.

Cloutier, J., Stevenson, R.K., Bardoux, B., 2005. Nd isotopic, petrologic andgeochemical investigation of the Tulawaka East gold deposit, TanzanianCraton. Precambrian Res. 139, 147–163.

Collins, A.S., Pisarevsky, S.A., 2005. Amalgamating eastern Gondwana:the evolution of the Circum-Indian orogens. Earth Sci. Rev. 71, 229–270.

Collins, A.S., Reddy, S.M., Buchan, C., Mruma, A., 2004. Temporal constraintson Palaeoproterozoic eclogite formation and exhumation (Usagaran OrogenTanzania). Earth Planet. Sci. Lett. 224, 175–192.

Condon, D., Zyu, M., Bowring, S., Wang, W., Yang, A., Jin, Y., 2005. U–Pbages from the Neoproterozoic Doushantuo Formation China. Science 308,95–98.

Costa, M., Cadoppi, P., Sacchi, R., Fanning, C.M., 1994. U–Pb “SHRIMP”dating of zircons from Mozambique Gneiss. Boll. Soc. Geol. Ital. 113,173–178.

Cutten, N.H.C., Johnson, S.P., De Waele, B., 2006. Protolith ages and timing ofmetasomatism related to formation of whiteschists at Mautia Hill Tanzania:implications for the assembly of Gondwana. J. Geol. 114, 683–698.

D’Agrella-Filho, M.S., Pacca, I.G., Renne, P.R., Onstott, T.C., Teixeira, W.,1990. Paleomagnetism of Middle Proterozoic (1.01 to 1.08 Ga) mafic dykesin southeastern Bahia State-Sao Francisco Craton, Brazil. Earth Planet. Sci.Lett. 101, 332–348.

da Silva, L.C., McNaughton, N.J., Armstrong, R.A., Hartmann, L.A., Fletcher,I.R., 2005. The Neoproterozoic Mantiqueira Province and its Africanconnections: a zircon-based U–Pb geochronologic subdivision for theBrasiliano/Pan-African systems of orogens. Precambrian Res. 136, 203–240.

Daly, M.C., 1986. The tectonic and thermal evolution of the Irumide belt, Zam-bia. PhD Thesis, University of Leeds, Leeds, p. 326.

Daly, M.C., 1988. Crustal shear zones in central Africa. A kinematic approachto Proterozoic tectonics. Episodes 11, 5–11.

Daly, M.C., Chorowizc, J., Fairhead, J.D., 1989. Rift basin evolution in Africa:the influence of reactivated steep basement shear zones. In: Cooper, M.A.,Williams, G.D. (Eds.), Inversion Tectonics. Geological Society of London,p. 44.

De Waele, B., 2005. The Proterozoic geological history of the Irumide belt,Zambia. PhD Thesis, Curtin University of Technology, Perth, p. 468.

De Waele, B., Fitzsimons, I.C.W., 2004. The age and detrital fingerprint of theMuva Supergroup of Zambia: molassic deposition to the southwest of theUbendian belt. In: Abstracts Volume of Geoscience Africa 2004, Johannes-burg, South Africa, pp. 162–163.

De Waele, B., Wingate, M.T.D., Fitzsimons, I.C.W., Mapani, B.S.E., 2003. Unty-ing the Kibaran knot: A reassessment of Mesoproterozoic correlations insouthern Africa based on SHRIMP U-Pb data from the Irumide belt. Geology31 (6), 509–512.

De Waele, B., Fitzsimons, I.C.W., 2007. The nature and timing of Palaeoprotero-zoic sedimentation at the southeastern margin of the Congo Craton. Prec.Res. 159, 95–116.

De Waele, B., Mapani, B., 2002. Geology and correlation of the central Irumidebelt. J. Afr. Earth Sci. 35, 385–397.

De Waele, B., Kampunzu, A.B., Mapani, B.S.E., Tembo, F., 2006a. The IrumideBelt of Zambia. J. Afr. Earth Sci. 46, 36–70.

De Waele, B., Liegeois, J.P., Nemchin, A.A., Tembo, F., 2006b. Isotopic andgeochemical evidence of Proterozoic episodic crustal reworking within theIrumide Belt of south-central Africa, the southern metacratonic boundary ofan Archaean Bangweulu Craton. Precambrian Res. 148, 225–256.

Deblond, A., Punzalan, L.E., Boven, A., Tack, L., 2001. The Malagarazi Super-group of SE Burundi and its correlative Bukoban Supergroup of NWTanzania: Neo- and Mesoproterozoic chronostratigraphic constraints fromAr–Ar ages on mafic intrusive rocks. J. Afr. Earth Sci. 32, 435–449.

Delor, C., Lafon, J.M., Rossi, P., Cage, M., Pato, D., Chevrel, S., Le Matour, J.,Matukov, D., Sergeev, S., 2006. Unravelling Precambrian crustal growth ofcentral west Angola: Neoarchaean to Siderian inheritance, main Orosirianaccretion and discovery of the “Angolan” Pan African Belt. In: Abstract ofthe 21st Colloquium of African Geology, Maputo, Mozambique, 3–5 July.

Fernandez-Alonso, M., Theunissen, K., 1998. Airborne geophysics and geo-chemistry provide new insights in the intracontinental evolution of theMesoproterozoic Kibaran belt (Central Africa). Geol. Mag. 135 (2),203–216.

Fernandez-Alonso, M., Lavreau, J., Klerkx, J., 1986. Geochemistry andgeochronology of the Kibaran granites in Burundi Central Africa: impli-cations for the Kibaran Orogeny. Chem. Geol. 57, 217–234.

Gerards, J., 1971. L’anticlinal de Mugombwa (Rwanda); un exemple de discor-dance majeure entre le Burundien et son socle. The Mugombwa Anticline(Rwanda); an example of a major discordance between the Burundian and itsbasement. Rapport Annuel pour l’annee 1970 du Departement de Geologieet de Mineralogie du Musee royal de l’Afrique centrale. Museum for CentralAfrica, Tervuren, Belgium, pp. 22–26.

Goscombe, B., Armstrong, R.A., Barton, J.M., 2000. Geology of the CheworeInliers, Zimbabwe: constraining the Mesoproterozoic to Palaeozoic evolu-tion of the Zambezi belt. J. Afr. Earth Sci. 30 (3), 589–627.

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 139

Hanson, R.E., 2003. Proterozoic geochronology and tectonic evolution ofsouthern Africa. In: Yoshida, M., Windley, Brian, F., Dasgupta, S. (Eds.),Proterozoic East Gondwana: Supercontinent Assembly and Breakup, vol.206. Geological Society of London Special Publication, London, pp.427–463.

Hanson, R.E., Wilson, T.J., Wardlaw, M.S., 1988. Deformed batholiths in the PanAfrican Zambezi belt, Zambia: age and implications for regional Proterozoictectonics. Geology 16, 1134–1137.

Jamal, D., De Wit, M., 2004. U–Pb geochronology and Sm–Nd data from LurioBelt, NE-Mozambique; significance for crustal evolution. In: Abstract Vol-ume of the 20th Colloquium of African Geology, BRGM, Orleans, France,pp. 204–205.

John, T., 2001. Subduction and continental collision in the Lufilian Arc-Zambezibelt orogen: a petrological, geochemical, and geochronological study ofeclogites and whiteschists (Zambia). PhD Thesis, University of Kiel, Kiel,p. 78.

John, T., Schenk, V., Haase, K., Scherer, E., Tembo, F., 2003. Evidence fora Neoproterozoic ocean in south-central Africa from mid-oceanic-ridge-type geochemical signatures and pressure–temperature estimates of Zambianeclogites. Geology 31 (3), 243–246.

John, T., Schenk, V., Mezger, K., Tembo, F., 2004a. Timing and PT evolu-tion of whiteschist metamorphism in the Lufilian Arc–Zambezi belt Orogen(Zambia): implications for the assembly of Gondwana. J. Geol. 112, 71–90.

John, T., Scherer, E.E., Haase, K., Schenk, V., 2004b. Trace element fractionationduring fluid-induced eclogitization in a subducting slab: trace element andLu–Hf–Sm–Nd isotope systematics. Earth Planet. Sci. Lett. 227, 441–456.

Johnson, S.P., Oliver, G.J.H., 2000. Mesoproterozoic oceanic subduction, island-arc formation and the initiation of back-arc spreading in the Kibaran beltof central, southern Africa: evidence from the ophiolite terrane, Cheworeinliers, northern Zimbabwe. Precambrian Res. 103, 125–146.

Johnson, S.P., Oliver, G.J.H., 2004. Tectonothermal history of the Kaourera Arc,northern Zimbabwe: implications for the tectonic evolution of the Irumideand Zambezi Belts of south central Africa. Precambrian Res. 130, 71–97.

Johnson, S.P., Cutten, H.N.C., Muhongo, S., De Waele, B., 2003. Neoarcheanmagmatism and metamorphism of the western granulites in the centraldomain of the Mozambique belt, Tanzania: U–Pb SHRIMP geochronologyand PT estimates. Tectonophysics 375, 125–145.

Johnson, S.P., Rivers, T., De Waele, B., 2005. A review of the Mesoproterozoicto early Palaeozoic magmatic and tectonothermal history of south-centralAfrica: implications for Rodinia and Gondwana. J. Geol. Soc. Lond. 162,433–450.

Johnson, S.P., De Waele, B., Liyungu, A.K., 2006. U–Pb SHRIMP geochronol-ogy of granitoid rocks in eastern Zambia: terrane subdivision ofthe Mesoproterozoic Southern Irumide Belt. Tectonics 25, TC6004,doi:10.1029/2006TC001977.

Johnson, S.P., De Waele, B., Tembo, F., Evans, D., Iizuka, T., Tani, K., 2007a.Geochronology of the Zambezi Supracrustal sequence, southern Zambia: arecord of Neoproterozoic divergent processes along the southern margin ofthe Congo Craton. J. Geol. 115, 355–374.

Johnson, S.P., De Waele, B., Tembo, F., Katongo, C., Tani, K., Chang, Q., Iizuka,T., Dunkley, D., 2007b. Geochemistry, geochronology and isotopic evolutionof the Chewore-Rufunsa terrane, Southern Irumide Belt: A Mesoproterozoic,continental-margin-arc. J. Petr. 48, 1411–1441.

Katongo, C., Koller, F., Klotzli, U., Koeberl, C., Tembo, F., De Waele, B.,2004. Petrography, geochemistry and geochronology of granitoid rocks inthe Neoproterozoic–Paleozoic Lufilian-Zambezi belt, Zambia: implicationsfor tectonic setting and regional correlation. J. Afr. Earth Sci. 40, 219–244.

Key, R.M., Liyungu, A.K., Njamu, F.M., Somwe, V., Banda, J., Mosley, P.N.,Armstrong, R.A., 2001. The western arm of the Lufilian Arc in NW Zam-bia and its potential for copper mineralisation. J. Afr. Earth Sci. 33 (3/4),503–528.

Klerkx, J., 1987. Crustal evolution of the northern Kibaran Belt, eastern andcentral Africa. In: Kroner, A. (Ed.), Proterozoic Lithospheric Evolution.American Geophysical Union and the Geological Society of America, pp.217–233.

Klerkx, J., Liegeois, J.P., Lavreau, J., Theunissen, K., 1984. Granitoideskibariens precoces et tectonique tangentielle au Burundi: magmatisme

bimodal lie a une distention crustale. In: Klerckx, J., Michot, J. (Eds.),African Geology, a Volume in Honour of L. Cahen. Royal Museum forCentral Africa, Tervueren, pp. 29–46.

Klerkx, J., Theunissen, K., Tack, L., 1993. The Mesoproterozoic Kibaran belt; aresult of strike-slip related intraplate extension and inversion. In: Anonymous(Ed.), Seventh Meeting of the European Union of Geosciences; Abstract Sup-plement. Terra Abstracts, vol. 5, Suppl. Blackwell Scientific Publications,Oxford International, p. 239.

Kokonyangi, J., Armstrong, R.A., Kampunzu, A.B., Yoshida, M., Okudaira,T., 2004a. U–Pb zircon geochronology and petrology of granitoids fromMitwaba (Katanga, Congo): implications for the evolution of the Mesopro-terozoic Kibaran belt. Precambrian Res. 132, 79–106.

Kokonyangi, J., Kampunzu, A.B., Armstrong, R., Yoshida, M., Okudaira, T.,2004b. Mesoproterozoic evolution of the Kibaride (Congo, Central Africa):geotectonic implications. In: Proceedings of the 32nd International Geolog-ical Congress, Florence, Italy, pp. 875–1875.

Kokonyangi, J., Kampunzu, A.B., Poujol, M., Okudaira, T., Yoshida, M.,Shabeer, K.P., 2005. Petrology and geochronology of Mesoproterozoicmafic–intermediate plutonic rocks from Mitwaba (D.R. Congo): implica-tions for the evolution of the Kibaran belt in central Africa. Geol. Mag. 142,109–130.

Kokonyangi, J., Kampunzu, A.B., Armstrong, R., Yoshida, M., Okudaira,T., Arima, M., Ngulube, D.A., 2006. The Mesoproterozoic Kibaride belt(Katanga, SE D.R. Congo). J. Afr. Earth Sci. 46, 1–35.

Kroner, A., 2001. The Mozambique belt of East Africa and Madagascar:significance of zircon and Nd model ages for Rodinia and Gondwana super-continent formation and dispersal (Du Toit Memorial Lecture 1999). SouthAfr. J. Geol. 104 (2), 151–166.

Kroner, A., Sacchi, R., Jaeckel, P., Costa, M., 1997. Kibaran magmatism and Pan-African granulite metamorphism in northern Mozambique: single zirconages and regional implications. J. Afr. Earth Sci. 25 (3), 467–484.

Kroner, A., Willner, A.P., Hegner, E., Jaeckel, P., Nemchin, A., 2001. Singlezircon ages, PT evolution and Nd isotopic systematics of high-grade gneissesin southern Malawi and their bearing on the evolution of the Mozambiquebelt in south eastern Africa. Precambrian Res. 109, 257–291.

Lavreau, J., 1985. Le Groupe de la Rusizi (Rusizien du Zaıre, Rwanda etBurundi) a la lumiere des connaissances actuelles. Annual Report 1983-84, Department of Geology and Mining, Royal Museum for Central Africa,Tervueren, pp. 111–119.

Lavreau, J., Liegeois, J.P., 1982. Granites a etain et granito-gneiss burundiensau Rwanda (region de Kibuye): age et signification. Ann. Soc. Geol. Belg.105, 289–294.

Lefebvre, J.J., 1989. Depositional environment of copper–cobalt mineraliza-tion in the Katangan sediments of Southeast Shaba, Zaire. In: Boyle, R.W.,Brown, A.C., Jefferson, C.W., Jowett, E., Kirkham, C.R.V. (Eds.), Sediment-hosted Stratiform Copper Deposits, vol. 36. Geological Association ofCanada, pp. 401–426.

Lenoir, J.L., Liegeois, J.-P., Theunissen, K., Klerkx, J., 1994. The Palaeopro-terozoic Ubendian shear belt in Tanzania: geochronology and structure. J.Afr. Earth Sci. 19 (3), 169–184.

Li, Z.X., Evans, D.A.D., Zhang, S., 2004. A 90◦ spin on Rodinia: possiblecausal links between the Neoproterozoic supercontinent, superplume, truepolar wander and low-latitude glaciation. Earth Planet. Sci. Lett. 220 (3/4),409–421.

Mallick, D.I.J., 1963. The stratigraphy, structure and metamorphism of the pre-Karoo rocks of the Mpande Dome and adjacent areas, N. Rhodesia. PhDThesis, Leeds, Leeds, p. 200.

Mallick, D.I.J., 1966. The stratigraphy and structural development of the MpandeDome, Southern Zambia. Trans. Geol. Soc. South Afr. 69, 211–230.

Manya, S., Maboko, M.A.H., 2003. Dating basaltic volcanism in the Neoar-chaean Sukumaland Greenstone Belt of the Tanzanian craton using theSm–Nd method: implications for the geological evolution of the Tanzaniacraton. Precambrian Res. 121, 35–45.

Mapani, B.S.E., Rivers, T., Tembo, F., Katongo, C., 2001. Terrane mapping inthe eastern Irumide and Mozambique belts: implications for the assemblyand dispersal of Rodinia. In: McCourt, S. (Ed.), Proceedings of the IGCP4th Fieldmeeting, vol. 418. University of Durban-Westville, Durban, SouthAfrica.

Author's personal copy

140 B. De Waele et al. / Precambrian Research 160 (2008) 127–141

McConnell, R.B., 1972. Geological development of the rift system of easternAfrica. Bull. Geol. Soc. Am. 83, 2549–2572.

McCourt, S., Armstrong, R.A., Kampunzu, A.B., Mapeo, R.B.M., Morais, E.,2004. New U–Pb SHRIMP ages on zircons from the Lubango region, south-west Angola: insights into the Proterozoic evolution of south-western Africa.In: Proceedings of the 20th Colloquium of African Geology, Orleans, France,pp. 438–439.

Meert, J.G., Van der Voo, R., 1996. Paleomagnetic and 40Ar/39Ar study of theSinyai Dolerite, Kenya: implications for Gondwana assembly. J. Geol. 104,131–142.

Meert, J.G., Hargraves, R.B., Van der Voo, R., Hall, C.M., Halliday, A.N., 1994a.Paleomagnetic and 40Ar/39Ar studies of late Kibaran intrusives in Burundi,east Africa; implications for late Proterozoic supercontinents. J. Geol. 102(6), 621–637.

Meert, J.G., Van der Voo, R., Patel, J., 1994b. Paleomagnetism of the LateArchean Nyanzian System, western Kenya. Precambrian Res. 69, 113–131.

Meert, J.G., Van der Voo, R., Ayub, S., 1995. Paleomagnetic investigation ofthe Late Neoproterozoic Gagwe lavas and Mbozi complex, Tanzania and theassembly of Gondwana. Precambrian Res. 74, 225–244.

Moller, A., Appel, P., Mezger, K., Schenk, V., 1995. Evidence for a 2 Ga sub-duction zone. Eclogites in the Usagaran belt of Tanzania. Geology 23,1067–1070.

Mruma, A., 1995. Stratigraphy and palaeo-depositional environment of thePalaeoproterozoic volcano-sedimentary Konse Group in Tanzania. J. Afr.Earth Sci. 21, 281–290.

Muhongo, S., Kroner, A., Nemchin, A.A., 2001. Single zircon evaporation andSHRIMP ages for granulite-facies rocks in the Mozambique belt of Tanzania.J. Geol. 109, 171–189.

Murphy, J.B., Keppie, J.D., 2005. The Acadian Orogeny in the NorthernAppalachians. Int. Geol. Rev. 47, 663–687.

Oliver, G.J.H., 2001. Reconstruction of the Grampian episode in Scotland: itsplace in the Caledonian Orogeny. Tectonophysics 332, 23–49.

Oliver, G.J.H., Johnson, S.P., Williams, I.S., Herd, D.A., 1998. Relict 1.4 Gaoceanic crust in the Zambezi Valley, northern Zimbabwe: evidence forMesoproterozoic supercontinental fragmentation. Geology 26, 571–573.

Patel, J.P., Raja, P.K.S., 1979. A palaeomagnetic study of lower Proterozoicgranites from western Kenya. Geophys. J. Roy. Astron. Soc. 58, 279–286.

Pinna, P., Cocherie, A., Thieblemont, D., Jezequel, P., 2000. The Kisii Group ofwestern Kenya: an end-Archaean (2.53 Ga) late orogenic volcano sedimen-tary sequence. J. Afr. Earth Sci. 30, 79–97.

Pisarevsky, S.A., 2005. New edition of the global paleomagnetic database. EOSTrans. 86 (17), 170.

Pisarevsky, S.A., Natapov, L.M., 2003. Siberia and Rodinia. Tectonophysics375, 221–245.

Pisarevsky, S.A., Wingate, M.T.D., Powell, C.M., Johnson, S., Evans, D.A.D.,2003. Models of Rodinia assembly and fragmentation. In: Yoshida, M.,Windley, Brian, Dasgupta, S. (Eds.), Proterozoic East Gondwana: Super-continent Assembly and Breakup, vol. 206. Geological Society of LondonSpecial Publication, pp. 35–55.

Porada, H., 1989. Pan-African rifting and orogenesis in southern to equatorialAfrica and eastern Brazil. Precambrian Res. 44, 103–136.

Porada, H., Berhorst, V., 2000. Towards a new understanding of theNeoproterozoic–Early Palaeozoic Lufilian and northern Zambezi belts inZambia and the Democratic Republic of Congo. J. Afr. Earth Sci. 30,727–771.

Rainaud, C., Armstrong, R.A., Master, S., Robb, L.J., 1999. A fertile Palaeopro-terozoic magmatic arc beneath the Central African Copperbelt. In: Stanley,C.J. (Ed.), Mineral Deposits: Processes to Processing. Balkema, London,pp. 1427–1430.

Rainaud, C.L., Armstrong, R.A., Master, S., Robb, L.J., Mumba, P.A.C.C., 2002.Contributions to the geology and mineralisation of the central African Cop-perbelt. I. Nature and geochronology of the pre-Katangan basement. In:G.S.o. Namibia (Ed.), Proceedings of the 11th IAGOD Quadrennial Sympo-sium and Geocongress. Geological Survey of Namibia, Windhoek, Namibia,p. 5.

Rainaud, C., Master, S., Armstrong, R.A., Robb, L.J., 2003. A crypticMesoarchean terrane in the basement to the central African Copperbelt. J.Geol. Soc. Lond. 160, 11–14.

Reddy, S.M., Collins, A.S., Mruma, A., 2003. Complex high-strain deformationin the Usagaran Orogen, Tanzania: structural setting of PalaeoproterozoicEclogites. Tectonophysics 375, 101–123.

Renne, P.R., Onstott, T.C., D’Agrella-Filho, M.S., Pacca, I.G., Teixeira, W.,1990. 40Ar/39Ar dating of 1.0–1.1 Ga magnetizations from the Sao Franciscoand Kalahari cratons: tectonic implications for Pan-African and Brasilianomobile belts. Earth Planet. Sci. Lett. 101, 349–366.

Ring, U., 1993. Aspects of the kinematic history and mechanisms of superpo-sition of the Proterozoic mobile belts of eastern Central Africa (northernMalawi and southern Tanzania). Precambrian Res. 62, 207–226.

Ring, U., Kroner, A., Toulkeredis, T., 1997. Palaeoproterozoic granulite-faciesmetamorphism and granitoid intrusions in the Ubendian-Usagaran Orogenof northern Malawi, east-central Africa. Precambrian Res. 85, 27–51.

Ring, U., Kroner, A., Layer, P., Buchwaldt, R., Toulkeredis, T., 1999. DeformedA-type granites in northern Malawi, east-central Africa: pre- or syntectonic.J. Geol. Soc. Lond. 156, 695–714.

Ring, U., Kroner, A., Buchwaldt, R., Toulkeridis, T., Layer, P.W., 2002. Shear-zone patterns and eclogite-facies metamorphism in the Mozambique beltof northern Malawi, east-central Africa: implications for the assembly ofGondwana. Precambrian Res. 116, 19–56.

Robb, L.J., Master, S., Greyling, L.N., Yao, Y., Rainaud, C.L., 2002. Contribu-tions to the geology and mineralization of the central African Copperbelt.V. Speculations regarding the ‘snowball earth’ and redox controls onstratabound Cu–Co and Pb–Zn mineralization. In: Proceedings of Geo-congress 2002, Windhoek, Namibia, p. 6.

Schandelmeier, H., 1980. Regionale gliederung des Prakambriums und aspekteder krustentwicklung um Mambwe/nordost-Zambia. PhD Thesis, TechnicalUniversity of Berlin, Berlin, p. 134.

Schandelmeier, H., 1983. The geochronology of post-Ubendian granitoids anddolerites from the Mambwe area, northern province, Zambia. Report 83(1).Institute of Geological Sciences, pp. 40–46.

Schenk, V., Appel, P., 2001. Anti-clockwise P–T path during ultrahigh-temperature (UHT) metamorphism at ca. 1050 Ma in the Irumide Belt ofEastern Zambia. Berichte der Deutschen Mineralogischen Gesellschaft, Bei-hefte zum. Eur. J. Mineral. 13, 161.

Seth, B., Kroner, A., Mezger, K., Nemchin, A.A., Pidgeon, R.T., Okruscha, M.,1998. Archaean to Neoproterozoic magmatic events in the Kaoko belt ofNW Namibia and their geodynamic significance. Precambrian Res. 92 (4),341–363.

Sklyarov, E.V., Gladkochub, D.P., Mruma, A., Theunissen, K., Melnikov, A.I.,Klerkx, J., 1998. Paleoproterozoic eclogites and garnet pyroxenites of theUbende Belt (Tanzania). Schweizerische Mineralogische und Petrographis-che Mitteilungen 78 (2), 257–271.

Smith, A.G., 1963. The geology of the country around Mazabuka and Kafue;explanation of degree sheets 1527, SE quarter and 1528 SW quarter. Reportof the Geological Survey Department of Zambia, Lusaka, 2.

Sommer, H., Kroner, A., Muhongo, S., Hauzenberger, C., 2005. SHRIMP zirconages for post-Usagaran granitoid and rhyolitic rocks from the Palaeopro-terozoic terrain of southwestern Tanzania. South Afr. J. Geol. 108, 247–256.

Sutton, S.J., Maynard, J.B., 2005. A fluid mixing model for copper mineral-ization at Konkola North, Zambian Copperbelt. J. Afr. Earth Sci. 42, 95–118.

Tack, L., Liegeois, J.P., Deblond, A., Duchesne, J.C., 1994. Kibaran A-typegranitoids and mafic rocks generated by two mantle sources in a late orogenicsetting (Burundi). Precambrian Res. 68 (3/4), 323–356.

Tack, L., Fernandez-Alonso, M., Tahon, M., Wingate, M.T.D., Barritt, S., 2002.The “northeastern Kibaran belt” (NKB) and its mineralisations reconsid-ered: new constraints from a revised lithostratigraphy, a GIS-compilation ofexisting geological maps and a review of recently published as well as unpub-lished igneous emplacement ages in Burundi. In: G.S.o. Namibia (Ed.), 11thIAGOD Quadrennial Symposium and Geocongress. Geological Survey ofNamibia, Windhoek, Namibia, p. 6.

Teixeira, W., Sabate, P., Barbosa, J., Noce, C.M., Carneiro, M.A., 2000. Archeanand Paleoproterozoic tectonic evolution of the Sao Francisco craton, Brazil.In: Cordiani, Umberto, G., Milani, E.J., Thomaz Filho, A., Campos, D.A.(Eds.), Tectonic Evolution of South America, Rio de Janeiro, pp. 101–137.

Author's personal copy

B. De Waele et al. / Precambrian Research 160 (2008) 127–141 141

Tembo, F., Kampunzu, A.B., Porada, H., 1999. Tholeiitic magmatism associatedwith continental rifting in the Lufilian belt of Zambia. J. Afr. Earth Sci. 28,403–425.

Theunissen, K., Klerkx, J., Melnikov, A., Mruma, A., 1996. Mechanisms ofinheritance of rift faulting in the western branch of the East African Rift,Tanzania. Tectonics 15, 776–790.

Trompette, R., 1994. Geology of Western Gondwana. A.A. Balkema, Rotterdam,p. 350.

Trompette, R., 1997. Neoproterozoic (∼600) aggregation of Western Gondwana:a tentative scenario. Precambrian Res. 82, 101–112.

Van der Voo, R., 1990. The reliability of paleomagnetic data. Tectonophysics184, 1–9.

Vrana, S., Prasad, R., Fediukova, E., 1975. Metamorphic kyanite eclogites inthe Lufilian Arc of Zambia. Contrib. Mineral. Petrol. 51 (2), 139–160.

Wendorff, M., 2005a. Evolution of Neoproterozoic–lower Palaeozoic LufilianArc, Central Africa: a new model based on syntectonic conglomerates. J.Geol. Soc. Lond. 162, 5–8.

Wendorff, M., 2005b. Sedimentary genesis and lithostratigraphy of Neoprotero-zoic megabreccia from Mufulira, Copperbelt of Zambia. J. Afr. Earth Sci.42, 61–81.

Wingate, M.T.D., Pisarevsky, S.A., De Waele, B., 2005. Paleomagnetism ofthe 765 Ma Luakela Volcanics in NW Zambia and implications for Neo-proterozoic positions of the Congo Craton. In: AGU Fall meeting, SanFrancisco.

Wirth, K.M., 2004. Origin and evolution of the Kilimafedha Greenstone Belt,Eastern Tanzania Craton: evidence from Nd and Pb Isotopes. In: GSA AnnualMeeting, Denver.