persistent fault controlled basin formation since the ......basement fault reactivation - the kibara...

PergamonJournal of African Earth Sciences, Vol. 26, No.3, pp. 347-361, 1998

c 1998 Elsevier Science LtdAll rights reserved. Printed in Great Britain

PI/:S0899-5362(98)00020-7 0899-5362198 $19.00 + 0.00

Persistent fault controlled basin formation since theProterozoic along the Western Branch of the East

African Rift

J. KLERKX, K. THEUNISSEN, and D. DELVAUXDepartment of Geology and Mineralogy, Royal Museum of Central Africa, B-3080

Tervuren, Belgium

Abstract-The Western Branch of the East African Rift System is outlined by elongatesedimentary basins, frequently occupied by Cenozoic rift lakes. The role of theinheritance of the leading rift faults from pre-existing basement structures has oftenbeen invoked. Recent studies in western Tanzania confirm the extent of the northwestorientated Palaeoproterozoic Ubende Belt contribution to the Phanerozoic Rift.Attention is drawn here on the occurrence of different Meso- and Neoproterozoicsedimentary basins that developed along the ductile shear belt as a result of repeatedsinistral wrench fault reactivation. These basins partly overlap each other and typicallybear shallow and weakly evolved sediments. North of the Ubende Belt, theMesoproterozoic Kibara Belt is inferred to have originated as a basin controlled bythe complex termination of the Ubende wrench fault.Phanerozoic rift basins also develop along the northwest orientated Ubende Beltstructure. They display the same elongate shape as the Proterozoic basins. In LatePalaeozoic-Early Mesozoic the Karoo rift basins formed from a dextral lateral shearreactivation of the inherited Proterozoic shear faults. During the first phase ofdevelopment the Lake Tanganyika Basin is believed to bear the same characteristicsas all previous basins along the Ubende Shear Belt, mainly controlled by strike-slipmovements along pre-existing shear faults. The present Lake Tanganyika Basin issubdivided in two sub-basins, separated by the transverse Mahali Shoal, which isan active structure located on the Ubende Shear. The deep lake basin mainlydeveloped outside the Ubende Belt. The northern sub-basin appears to be structurallycontrolled by the reactivation of the Mesoproterozoic sinistral wrench fault terminationof the Ubende shear faults. Structural control of the Palaeoproterozoic basement ishowever unclear for the southern sub-basin of Lake Tanganyika: this part of the riftsegment is flanked by Palaeoproterozoic basement which has not been affected bythe Ubende Shear. © 1998 Elsevier Science Limited.

Resume-La branche occidentale du rift est-africain est caracterisee par une seriede bassins sedimentaires. souvent occupes par des lacs de rift. L'influence desstructures pre-existantes du socle comme heritage des structures principales du rifta ete souvent invoquee. Des etudes recentes en Tanzanie occidentale confirmentI'importance de la contribution des structures paleoproterozoiques orientees NW-SEde la chaine ubendienne aux structures phanerozoiques du rift. L'attention est attiresici sur la presence de bassins sedimentaires d'age meso- a neoproterozoique, qui sesont developpes Ie long de la zone de cisaillement ductile, suite a des reactivationsdecrochantes senestres repetees, Ces bassins se superposent partiellement etcontiennent typiquement des sediments peu evotues, d'eau peu profonde. Au nordde la chaine ubendienne, la chaine rnesoproterozoique Kibarienne est considereecomme issue d'un bassin controle par la termination complexe du svsterne decrochantubendien.Les bassins phanerozoiques se developpent aussi Ie long de la structure NW-SE dela chaine ubendienne. lis ont la merne forme allonqee que les bassins proterozoiques.

Journal of African Earth Sciences 347

J. KLERKX et al.

Au cours du Paleozoique tardif et du Mesozo"ique precoce, les bassins de rift Karoose sont formes par reactivation laterale dextre des failles ductiles heritees duProterozoique. Pendant sa premiere phase de developpernent, Ie bassin du LacTanganyika presents les memes characteristiques que tous les bassins anterieurssitues Ie long de la chaine cisaillante ubendienne et principalement controles par desmouvements Ie long de failles pre-existantes, Le bassin actuel du Lac Tanganyikaest subdivise en deux sous-bassins, separes par Ie seuil transverse des Mahali.Celui-ci est une structure active localisee sur un cisaillement ubendien. Le bassinlacustre profond s'est principalement developpe en dehors de la chaine ubendienne.Le sous-bassin nord apparalt comme controlle structuralement par la reactivation dela terminaison des cisaillements ubendiens mesoproterozolques. Le contr61e structuraldu socle paleoproterozoique est cependant peu clair pour Ie sous-bassin sud du LacTanganyika. Cette partie du rift est bordee de socle paleoproterozolque non affectepar Ie cisaillement ubendien. © 1998 Elsevier Science Limited.

(Received 3 December 1997: revised version received 13 February 1998)

INTRODUCTIONContinental rifts avoid cratonic regions whilepreferentially following pre-existing basementweaknesses. In the East African continent theyare generally associated with belts developingduring the Neoproterozoic Pan-African evolution.Although this is the case for the Eastern RiftBranch (Mozambique Belt), it does not apply forthe Western Branch of the East African Rift witha northwest trending southern segment and amore northeast trending northern segment (insetFig. 1).

West of the Tanzanian Archean Craton theWestern Branch of the East African Rift overliesthe Palaeoproterozoic Ubende Belt (McConnell,1972). The rift inherited a large part of itsnorthwest trending fault pattern. The WesternBranch switches to a north-south trend as soonas it enters the domain of the MesoproterozoicKibara Belt, immediately north of the UbendeBelt. In this domain the rift develops at an angleto the regional trend of the Kibara folding (Klerkxet et., 1987) and the "leading basementweakness" followed by the rift in this region isless evident. More problematic is the basementstructure that contributed to the formation ofthe southern and most recent sub-basin of theLake Tanganyika Rift Zone.

The Western Branch characteristically bearselongate and narrow sedimentary basins,frequently occupied by deep rift lakes. Gravitysurvey and drilling were carried out only in theRukwa Basin (Peirce and Lipkov, 1988; Wescottet et., 1991), and seismic reflection surveys wereperformed in the Rukwa, Tanganyika and MalawiBasins (Sander and Rosendahl, 1989; Spechtand Rosendahl, 1989; Morley et el., 1992;Kilembe and Rosendahl, 1992). Thesegeophysical studies provided a wealth of

348 Journal ofAfrican Earth Sciences

information on the present intracontinental riftevolution. In summary, Rosendahl (1987)introduced the "half-graben" concept in the EastAfrican Rift and Morley et al. (1990) emphasisedon the high abundance of overlapping "transfer"zones, on the weak expression and the reducedoccurrence of large scale cross faults and onthe low extension of the upper crust. Variouspapers reported different extensional stress fieldconditions for the rifting (Kazmin, 1980;Chorowicz and Mukonki, 1980; Tiercelin et al.,1988; Delvaux et al., 1992; Ring et el., 1992).In the widely applied half-graben hypothesis theauthors do not consider the present riftgeometry to be as strongly controlled by prerift basement structure as this paper proposes.On the other hand, the interpretation of Morleyet al. (1990) strongly reflects similaritiesbetween the general Phanerozoic rifting andsome evolutionary characteristics of thebasement prior to the rifting.

Daly et al. (1989, 1991) suggested that in theLate Palaeozoic the sedimentation and tectonicsof the Karoo basins along the southern segmentof the Western Rift Branch, were controlled bydextral strike-slip reactivation of the Ubendestructures. The deep Rukwa Rift Basin with itsthick sedimentary infill, is believed to have beeninitiated in that early stage, as do the Kalemieand north Malawi Basins (Fig. 1), which allextend along a northwest trend, parallel to theUbende Belt leading structure (Delvaux, 1991).

In this present paper attention is drawn to theoccurrence in parts of the Phanerozoic riftedzone, of fault controlled and locally elongateProterozoic sedimentary basins, which all displayvery shallow level evolution. The geometry ofthese basins is largely inherited from thenorthwest trending structures of the Proterozoic

Persistent fault controlled basin formation since the Proterozoic along the Western Branch

GEOLOGICALOUTLINE

.:.:r····

8"-<"""I;'~'---,,coZe<:w\;; Lfl-lr'-->'>;,,",

~

TANZANIAN eRA: ON

EAST AFRICAN RIFT,/ ,/ /-.. /"

..... \

;s=~.~-::::::::::::::::~:~(a~"""'"

,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/

,/ ,/ ,/ ,/

[L] I2-10

DII~ 12[~-=il3

!>:]I4[-·-.--.: 15~ 16

/ /

/ /

/ /

/ /

,/ ,/ ,/ ,/

,/ ,/ ,/ ,/

./ ,/ ,/ ,/

/ / / / / / / / / ~ / / ~ / / ~ / /~/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ /r-------i////////// ////~v/

I!

28°

Figure 1. Location and geological outline of the region. In between cratonic blocks (the Bangweulu Block in the westand the Tanzanian Archean craton in the east) the Phanerozoic rift follows the regional trend of the Ubende ShearBelt. Inset shows the location of the area in the Western Branch of the East African Rift. 1: Palaeoproterozoic eastnortheast - west-southwest orientated Usagara Belt; 2·10: Palaeoproterozoic northwest-southeast orientated UbendeShear Belt with indication of the different fault bounded terranes: 2: Mahali; 3: Ufipa; 4: Ubende; 5: Mbozi; 6:Wansisi; 7: Katuma; 8: Lupa; 9: Ukenju; 10: Nyika (in Malawi); 11: Mesoproterozoic northeast-southwest trendingKibara Fold Belt; 12: Mesoproterozoic sedimentary basins along the northwest orientated Ubende Belt; 13:Neoproterozoic sedimentary basins along the northwest orientated Ubende Belt; 14: Karoo outcropping area; 15:Cenozoic deposits; 16: Rungwe Quaternary volcanics.

basement (Theunissen et al., 1996). Even basinswhich are oblique to the general trend of theUbende Belt are considered to be controlled bybasement fault reactivation - the Kibara Basinat its northern end in particular is interpreted asa Mesoproterozoic sinistral wrench faulttermination setting of the Ubende Shear Zone(Fernandez-Alonso and Theunissen, in press).

The tectonic control of the Lake TanganyikaBasin in its present stage of evolution is muchless evident. In contrast with the pre-Cenozoic

evolutionary stages, which are all characterisedby shallow basins, the Cenozoic basins are deepwater basins and developed mainly outside theUbende Belt itself. The Lake Tanganyika Basincan be subdivided into a northern and southernsub-basin, separated by the Mahali Shoal andaccommodation zone, which is an activetransverse structure located upon the UbendeShear Zone. The northern sub-basin appears tobe controlled by a Mesoproterozoic sinistralwrench fault termination. The southern sub-basin


J. KLERKX et al.

is orientated at a minor angle to the Ubendetrend. It is flanked on both sides by Palaeoproterozoic basement which is not affected bythe Ubende Shear and which is overlain byweakly deformed, non-metamorphosed andshallow level Proterozoic sediments.

WRENCH FAULT REACTIVATIONS IN THEPALAEOPROTEROZOIC UBENDE SHEAR BELT

With a northwest structural trend, thePalaeoproterozoic Ubende Shear Belt of westernTanzania extends for more than 500 km withan average width of 50 km (Fig. 1). To the westthe belt is bounded by the PalaeoproterozoicBangweulu Block (overlain by the Mesoproterozoic Mporokoso Beds, Fig. 1), while tothe east the belt is flanked by the TanzanianArchean Craton in the northern part and by thePalaeo proterozoic Usagara Belt in the southernpart (Fig. 1). The boundaries between both thesePalaeoproterozoic fold belts and the TanzanianCraton are frequently masked by the latePalaeoproterozoic granites (Lenoir et al., 1994).In an elongate and block shaped shear pattern(McConnell, 1967) the pre-existing terranes(numbered 2-10 on Fig. 1) of the Ubende Belthave been set in the northwest trending shearzone, best outlined by the highly sheared terraneboundaries. In a general Palaeoproterozoic platetectonic model (Daly et et., 1985; Daly, 1986)the northwest trending Ubende Belt has beenproposed as a lateral accretion zone, while themore west-east trending Usagara Belt evolvedas a frontal accretion event on the TanzanianCraton. Recent studies in the region proposedthat an early Palaeoproterozoic oblique suturepreceded the main lateral shear deformation ofthe Ubende Belt (Theunissen et et., 1996).Although the northwest orientated blockpattern, bounded by linear faults was presentsince the early stages, its present outline is aresult of persistent boundary reactivations inthe Proterozoic, as well as the Phanerozoic(Daly et al., 1989; Theunissen et el., 1996).This characteristic has been described as a"perennial taphrogenic structure" (McConnell,1972) and as a "long-lived fundamentalstructural weakness" (Sutton and Watson,1986). It is also often cited as a typicalexample of repeated reactivation of steeplyinclined shear fault zones.

The Palaeoproterozoic ductile shear beltThe Ubende Belt is composed of high-grademetamorphic sequences. They comprize maficand ultramafic granulite facies and rare eclogiterocks (Sklyarov et et., in press), which arepreserved in the Mbozi and Ubende terranes (4and 5 on Fig. 1). These terranes represent deeplysubducted ophiolites, exhumed to the subsurfaceduring the later Palaeoproterozoic and right lateralshear deformation (Boven et et., in prep.). Thisphase of pronounced lateral shear has imposedto the entire Ubende Belt its characteristicnorthwest orientated fabric. Ductile shearing isvariably distributed over the terranes. It is mostintensely expressed along and associated withsteeply dipping fold limbs along northwesttrending corridors. The latter constitute theboundaries of the distinct terranes of the UbendeBelt (Fig. 1) (Theunissen et al., 1996).

Proterozoic reactivationsAlong the northwest trending and steeplyinclined fold limbs and more particularly wherethese limbs coincide with terrane boundaries,the ductile dextral shear zones are frequentlyoverprinted by mylonite sequences, which areconcordantly interleaved in the ductile fabric. Insharp contrast with the ductile fabric, thesemylonites display consistent sinistral strike-slipdeformational characteristics in amphibolite-togreenschist-facies conditions (Theunissen et el.,1996). These mylonites reflect a pluriphasereactivation of the ductile shear zones. Only inthe southernmost part of the Ubende Belt has aNeoproterozoic age been obtained for themylonites (Theunissen et al., 1992). Geochronological evidence for such Neoproterozoicreactivation is lacking for the northern part.However, relative ages for the repeated sinistralstrike-slip deformational regimes can be deducedfrom structural styles and setting shown by thenearby Meso- and Neoproterozoic sediments(Theunissen et al., 1992, 1996).

Phanerozoic rift faultingIn the high-grade metamorphic Ubende Belt,northwest orientated ductile shear zones withinterleaved ductile-brittle mylonites are welldeveloped in between different terranes andconstitute prominent mechanical anisotropies.These structures have been repeatedlyreactivated during the Phanerozoic and have

Figure 2. Consistent sedimentary basin formation along the northwest Ubende Shear Belt. (A) During the Mesoproterozoic:ltieso, Mbala, Ukinga and Mbeya sedimentary basins. (B) Neoproterozoic sedimentary basins: Bukoba and Buanji. (C) Karoo(Permo- Triassic) basins: Kelemie, Rukwa and north Malawi. (0) Cenozoic basins covering Rukwa and part of Kalemie with theformation of deep lakes Tanganyika and north Malawi.


~<riiii'

~....Ql'§,C')

g~ci6'c,

~S'03III

a'~

'"S·C')<b

S<b

~Cbc;~0'IIIc-~S<b

~'"Cbs~~g.

34°

34°

i

TANZ~IAN CRATON

TANZAl'JIAN CRATON

UBENnE terranes- CENOZOIC!

C] UBENPE terranes- NEOPROTEROZOIC

32°

32°

30°

30°

!

rrGANYIKA

._+------ --.._.+--- "---------.,,..--+---I

CENOZOIC

- -!--, II I

,,13ANGWEUIj,u BLOC

D

BNEOPROTEROZOIC

4°~·--·--·· ... ..-\-n----\---c-

6° 1---- «. I

c<-:r;'

It:Q J

"" /

*'// i,,} '\

8°l::,. L':== f _-~I i,,,

8° 1_

6° ~.... " ...... _ ... ~

10°,-' ...1--'--__--:..

28°

100L!;:- -=-~------.L.L---~28° '>'

34°

c>i~

J'<t>!.~I «J.( ",'I

«{ftc 0~~,,~,,<,

~UBE~~DE terranesKARbo

34°

TANZ~IAN CRATON

IUBENDEterranesMESOPROTEROZOIC

Ii

-i

-----i----I!

TANZANI~ CRATON

~6 I~ft.

'1

..

32°

32°

---_._.-+_._-~--~ ._--+------_...., !

I

I------·-..-· ..·..·---·..·..·1·----·..···--·-,..·

I

30°

30°

AMESOPROTEROZOIC

_~ANGWEULU ~LOCK~_._------ --------

C

KAROO

4 0 J- m •••_._,,_• .. . .. . - \ II \;' I

c<-:r;

t:Q

"" J*' /// ,

8°L L ._.~_. I-i-.f-"

" -71BANGWE~U ELOCK

... I / !(Mporokoso)

", )8°~.--_-

10°)2~8-;;0 -:;;;:- =_L,.l__j@

10°",":'::- -L...L__....':::28°

~'0ssa.Q,l>

~§2;'s~~~

J. KLERKX et al.

largely contributed to the rift fault geometry.Some major northwest trending and brittle riftfaults are associated with dextral strike-slipreactivation: the Ufipa Fault in the north(Chorowicz et el., 1988) and the LivingstoneFault in the south (Wheeler and Karson, 1989;Wheeler and Rosendahl, 1994). However, theseauthors did not mention precise age constraintsfor the strike-slip movements. Delvaux et at.(1992) and Ring et at. (1992) have shown acomplex succession of dextral strike-slip andnormal fault movements along these northwesttrending faults, concluding that the stress fieldwas not stable and that the fault kinematicschanged several times from the Late Palaeozoicto the present times.

More characteristic are the very deep riftbasins, flanked by the superimposed, pluriphaseProterozoic and Phanerozoic faults, oftenrepresenting master faults of these basins (Figs1 and 2). The lack of stratigraphical time markersremains a major handicap in unravelling thecomplexity of rift faulting and its role in theformation of the succeeding basins.

Rifting also developed outside the Ubende Belt,as evidenced by the Lake Tanganyika Basin. Inthe south Lake Tanganyika sub-basin onlysinistral wrench fault reactivation in the nearbyUbende Belt is assumed to control the rift faults.In the northern sub-basin the weakest structuresare inferred to belong to the wrench faulttermination of the Ubende Belt (FernandezAlonso and Theunissen, in press).

PROTEROZOIC SEDIMENTARY BASINS ALONGTHE UBENDE BELT

Very low-grade metamorphic and shallow levelsedimentary sequences, assigned to differentProterozoic units, are variably distributed uponand along the high-grade metamorphic UbendeBelt (Fig. 1). Inside the belt these sedimentsoccur as narrow and elongate strips or patches,flanked by sheared basement. The sedimentsare more extended along the eastern andwestern boundaries of the belt, where theyappear in Meso- and Neoproterozoic settings.It is however in the northern part of the UbendeBelt that the sedimentary sequences are mostdeveloped, particularly within the Mesoproterozoic Kibara Belt (Figs 1 and 2A). Alongthe contact between the Kibara Belt and theTanzanian Craton, the Mesoproterozoicsediments are partly overlain by Neoproterozoicshallow level sedimentary and volcanicsequences (Tack, 1995).


As a rule, Neoproterozoic sediments overlapMesoproterozoic deposits in many places. Thisis the case along the northwest trending wrenchfault zone itself and its northwestern termination.It is assumed that basin forming processes inthe Proterozoic concentrated in the same areaby repeated reactivation of the same basementstructures.

Basins inside the Ubende BeltPatches (not outlined on the figures) of nonmetamorphosed and weakly consolidatedProterozoic sediments locally overly the highgrade Ubende sequences. They have beendescribed and mapped as the Ifume and SanyikaFormations along the Ufipa Shear Zone (Halligan,1962; Smirnov et a/., 1973). They consist ofpolymictic conglomerates, sandstones andsiltstones with thicknesses generally notexceeding a few hundreds of metres, exceptfor the Mongwe conglomerate (400 rn) . Thesequences are gently folded or tilted and containnumerous northwest striking faults and fractures.These Proterozoic, very shallow, arenaceoussediments testify for a tectonic activity alongthe inter-terrane boundary faults (Ufipa andRukwa) of the Ubende Belt.

In the southern part of the belt, the Ukwamacataclasites are attributed to the Mesoproterozoic (Harpum and Harris, 1958). Theserocks occur in narrow, elongate and northwesttrending troughs of sheared greenschist-faciesmetasediments and volcanics. They sharplycontrast with the surrounding Ubendian Beltmeta-anorthosites (Theunissen et a/., 1996). TheUkwama sequences may be considered asequivalent to the badly outcropping MbeyaRange unit (Fig. 2A), which is correlated to theMesoproterozoic Ukinga sediments (Klerkx andNanyaro, 1988; Marobhe, 1989).

The Proterozoic sediments within the northernpart of the Ubende Belt are weakly deformedand have less well outlined depocentres (lfurne,Sanyika and Mongwe). In the south they aresheared and confined to narrow and northwesttrending elongate troughs (Ukwama and Mbeya).

Basins along the margins of the Ubende BeltShallow arenitic and arenopelitic sediments occurover an extended area along the eastern (Ukinga,Buanji) and the western (Mbala) boundaries ofthe Ubende Belt (Fig. 2A, B).

The Ukinga-Buanji BasinIn the southeast the Kipengere Mountain Range(Fig. 2) is composed of low-grade metamorphic


and predominant pelitic sediments withsubordinate arenaceous horizons. The sedimentsunconformably overly the Ubendian crystallinebasement rocks and have been identified as theMesoproterozoic Ukinga Group (Harpum andHarris, 1958). Highly sheared along theirwestern boundary with the meta-anorthositicUbendian Belt, these sediments are almostundeformed in their eastern part (Klerkx andNanyaro, 1988; Theunissen et al., 1996). TheseUkinga Beds have been associated with theare no-pelitic schistose rocks outcropping in theMbeya Range (Marobhe, 1989). The northernpart of the Ukinga Basin is unconformablyoverlain by the Neoproterozoic Buanji Group,consisting of shales with interbedded quartzites,succeeded by limestones and topped byandesitic lavas (Harpurn and Harris, 1958;Maloney and Downie, 1972; Piper, 1975).Geophysical studies (Marobhe, 1989) estimateda 4.7 km sedimentary thickness for theUkinga-Buanji Basin.

The Mbala BasinWest of the Ubende Belt (Fig. 2A) shallow andalmost undeformed sandy and pelitic Mbalasediments outcrop along the southern flanks ofLake Tanganyika. Page (1960) considered thesesequences as equivalent to the NeoproterozoicBukoba Beds. Unrug (1984) proposed aMesoproterozoic age and correlated them to theMporokoso Beds along the Bangweulu Plateauin adjacent Zambia (Figs 1 and 2A). The MbalaBeds unconformably overly the weakly deformedand almost unmetamorphosed volcano-plutoniccomplex of Kate-Kipili of late Palaeoproterozoicage (Lenoir et al., 1994). The Mbala Beds alongthe eastern boundary display local shearingassociated with sinistral strike-slip deformation,which is best evidenced in the shallow levelKate-Kipili Palaeoproterozoic basement rocks(Theunissen et el., 1996). The Mbala sequenceslithologically resemble the Bukoba sandstoneunit, which extends southeast of the Kibara Belt(Fig. 1). Commonly attributed to the Neoproterozoic, the Bukoban sandstones have beenrecently interpreted as being Mesoproterozoic(Tack et el., 1992; Tack, 1995).

The Itiaso-Bukoba BasinNorth of the Ubende Shear Belt (Figs 1 and 2A,B) lies a vast region of weakly deformedsedimentary sequences, known as theMesoproterozoic Itiaso Group and theNeoproterozoic Bukoba Group (Halligan, 1962).The Itiaso sediments are predominantly

composed of dark gray shales with subordinatequartzites, while the Bukoba units are composedof a thick succession of sandstones and shales,succeeded by limestones and basalts. Halligan(1962) reported that the Itiaso Beds are severalthousand meters thick and were deposited in anarrow tectonic trough, confined to thewesternmost part of the region, near the LakeTanganyika border. The Itiaso Beds are alignedalong northwest-southeast trending upright foldsand folding is increasingly complex towards thewest while dying out to the east. TheNeoproterozoic Bukoba sedimentsunconformably overly the Itiaso Beds. They arevery weakly folded and again show moredeformation to the west.

TECTONIC CONTROL OF PROTEROZOICBASINS

Proterozoic sediments are preferentiallydeposited along the borders of the Ubende Beltare less developed within the belt. TheMesoproterozoic sediments are more deformedthan the Neoproterozoic ones. Mesoproterozoicbasins seem to have nucleated along reactivatedUbende shear faults. The same mechanism ofbasin formation prevails during the Neoproterozoic.

Mesoproterozoic basins along the Ubende BeltWhile the age of the Mbala Beds is questionable,the ltiaso and Ukinga sediments occurring alongthe eastern boundary of the Ubende Belt aredefinitively of Mesoproterozoic age (Fig. 2A).

The Ukinga and Mbeya GroupsThe Ukinga and Mbeya Groups have beenrelated to strike-slip reactivation along thenorthwest orientated Ubende basement (Klerkxand Nanyaro, 1988). While the basin asymmetryand deformational style of the Ukingasedimentary beds were recently interpreted toreflect sinistral strike-slip deformation at theorigin of the Mesoproterozoic basin, sinistraltranspression was identified as Neoproterozoic(Theunissen et el., 1996). The Ukwama andMbeya Range highly sheared metasediments areassumed to represent equivalents of the UkingaBeds. In contrast with the latter, the MbeyaRange metasediments evolve inside the wrenchfault zone and not along its boundary. Thesenarrow troughs are bounded on both sides bygreenschist-facies and retrograde sinistral strikeslip mylonites, locally intercalated in the basindeposits.


J. KLERKX et al.

The Mbala BedsThe Mbala Beds trend north-northwest on anelevated ridge upon the Kate-Kipili basement.Along the eastern border, the sediments showsinistral transpressional structures whichappear as reverse faulting (Theunissen et et.,1996).

The Itiaso GroupThe Itiaso Group sediments are intenselyfolded and metamorphosed. They areunconformably overlain by the only weaklydeformed and non-metamorphic BukobanGroup (Halligan, 1962). The evolution of theItiaso-Bukoba sedimentary basins along theUbendian controlled strike-slip basins wassuggested by Theunissen (1988a).

The above mentioned sedimentary basins havetheir deepest and most deformed parts locatedalong the sinistral strike-slip faults, outlined byretrograde mylonites which mark the basinboundary faults. The deformational regime ofthe highly sheared sediments is related to sinistralstrike-slip. The deformation systematically diesout away from the boundary fault zone of thebasin.

The Mesoproterezoic Kibara BasinDifferent from the above Mesoproterozoic basinsalong the Ubende Belt, the Kibara Basindeveloped northwest of the Ubende Belt (Figs 1and 2A). Considering that the Kibara and IrumideBelts have comparable evolutionary trends,Klerkx and Nanyaro (1988), Daly et al. (1989)and Theunissen (1988b) suggested that theseMesoproterozoic orogenic belts must haveoriginated from the reactivation of the UbendeBelt. At least the eastern Kibara Belt has beenclearly demonstrated to reflect a completelyintracontinental orogenic evolution and moreparticularly has largely preserved its extensionalbasin fabric (Klerkx et al., 1987). A majorproblem in linking the Kibara evolution with thePalaeoproterozoic Ubende reactivation is thepersistent sinistral strike-slip shown by the latter,while many Kibara models imply an early dextralstrike-slip for creating the Kibara Basin.

In the pre-Kibara basement, structuralweaknesses were created at the northwesterntermination of the Ubende Belt during thePalaeoproterozoic dextral shear. In the KibaraBelt Fernandez-Alonso and Theunissen (in press)have shown the presence of detachment styleextensional tectonics with detachment rootedto the south in a persistent sinistral strike slipzone.


Neoproterozoic sedimentary basinsPatchy distributed arenaceous Ifume, Sanyikaand Mongwe Beds (not represented on Fig. 2)are considered to be Neoproterozoic. These bedsare composed of very shallow sediments withweak deformational features.

Neoproterozoic sediments are morewidespread along the Ubende Belt than theMesoproterozoic sediments and they partlyoverlap the latter beds. They are systematicallyless deformed than the Mesoproterozoicsediments and are more distant from the UbendeBelt boundary. Like the Mesoproterozoic beds,they are affected by sinistral strike-slip alongtheir northwest orientated contact with eithersheared Mesoproterozoic beds or with sinistralstrike-slip mylonites.

The Neoproterozoic basins are partlysuperimposed on the Mesoproterozoic basins.There is evidence of repeated reactivations andsimilar basin formation conditions all over theProterozoic. Development, of basins and theirsubsequent deformation is attributable to sinistralwrench faulting. This inheritance of Mesoproterozoic basins during Neoproterozoicsedimentation is also very clearly evidencedalong the southern boundary of the Kibara Belt.The Mesoproterozoic rhomb-shaped basins withvery shallow sedimentary and volcanic depositsoriginate in the reactivated external domain ofthe belt near the contact with its cratonicforeland (Theunissen, 1988b; Tack et el., 1992).

KAROO BASINSKaroo basins correspond with an importantperiod of basin formation in eastern and southernAfrica, from Permian to Triassic (McKinlay,1965; Delvaux, 1991). In the Western RiftBranch this Karoo sedimentation occurs innorthwest trending basins (Fig. 2C), as did theirProterozoic precursors. However two differencesappear: (1) the elongate and narrow basins occurlargely upon, and not along the Ubende Belt;and (2) in the south, the Karoo basins developwith a northeast-southwest trend, outside theUbende terranes.

Karoo deposits along the northwest UbendetrendSurface outcrops of Karoo deposits occur asisolated patches along a northwest trend ingrabens and half grabens, tilted to the east(McKinlay, 1965). Karoo deposits covered alarger area than their present outcrop pattern(Figs 1 and 2C). Karoo sediments in the Rukwa


34°32°30°

KAROO

28°

8°t--~~-----+--".c---+----'

Figure 3. Karoo basins which have developed along a northwest trend and areconsidered to behave as strike-slip basins along northwest reactivated (inversion)Ubende Palaeoproterozoic structures.

Trough (Fig. 2C) accumulated up to 3.5 km atthe base of the basin (Wescott et al., 1991).They are also present in the northern LakeMalawi Basin (Specht and Rosendahl, 1989) andin the Kalemie Basin in Zaire (Cahen andLepersonne, 1978). The Karoo event in theKalemie, Rukwa and north Lake Malawi Troughs(Figs 2C and 3) appears as a series of deep basinsflanked by Proterozoic mylonites (Rukwa andLivingstone Faults in Tanzania). Lake Tanganyikais probably underlain by Permo-Triassicsediments where that prominent fault zonecrosscuts the lake (Ding and Rosendahl, 1989).These Karoo basins occur inside and along theUbende Belt, which appears as a basement highin between the basins.

Considering the Rukwa Trough as arepresentative basin for the Karoo sedimentation,it is evident that the Lupa Fault acted as themain boundary fault, flanking the deepest partof the active basin, and located along theeasternmost sheared sequences of the UbendeBelt (Fig. 3). Inversion of the Neoproterozoicsinistral transpression structure (Theunissen etal., 1996) probably contributed to the KarooBasin formation (Daly et al., 1989). Karoo

sedimentation preferentially occurred in theeastern part of the rift segment.

Other Karoo depositsIn southwest Tanzania the Karoo Beds weredeposited in the Ruhuhu Basin, which lies atright angles to the Ubende trend and is closeto the Usagara structure (Fig. 1). The basincontains a total of 3 km of sediments rangingfrom Late Carboniferous to Early Triassic(Kreuser and Markwort, 1990). The initialdepocentres resulted from crustal downwarping, which was possibly associated withsome faulting (Wopfner and Kaaya, 1992).However, the main basin boundary faultsdeveloped after the deposition of the Karoosediments and were active during the MiddleJurassic. Vertical displacements along themajor faults reached up to 1000 m and tiltedthe sediments to the southeast (Kreuser andSemkiwa, 1987).

The Karoo Beds also occur in the poorlyinvestigated Maniamba-Metangula Basin, whichextends parallel to and more to the south of theRuhuhu Basin (Verniers et al., 1989). It is notclear whether Karoo sediments underlie the


J. KLERKX et al.

Cenozoic beds of the Usangu flats and/or theBuhoro Trough (Marobhe, 1989).

The origin and the extension of the northeastorientated Karoo basins along the Usagarastructural grain is not well documented. It ispossible that the basins originated as extensionalbasins perpendicular to the main northwesttrending strike-slip zone (Lupa-Livingstone). TheCretaceous basins in the central African shearzone in west Africa originated in a similar way(Fairhead and Green, 1989).

Geodynamic setting of Karoo basinsThe Permo-Triassic deformation and associatedKaroo basin formation in the Ubende Belt andadjacent areas are typical intracontinental basinswhich were formed due to a combination ofcollisional processes at the southern margin ofGondwana, and extensional processes at itsnorthern margin.

During the Permo-Triassic period, the Pacificmargin of Gondwana was in a collisional setting,while the Tethyan margin was in an extensionalcondition. In the present-day co-ordinates of thesouthern margin of Gondwana, orogenic collisionoccurred in the Cape Fold Belt where it generateda north-south compression during Permian-EarlyTriassic times (Halbich, 1983). To the northeast,Early Permian basins developed along a northsouth trend coincides with the present EastAfrica coast and southwestern Madagascar.These basins belong to a major rift system whichwas linked to a large extensional fault thatpenetrated deep into Gondwana, starting fromits southern margin (de Wit et el., 1988;W6pfner, 1994). The northeast trending SelousBasin in southeast Tanzania marks the terminationof this major marine embayment. This basin hasmarine fauna with a Tethyan affinity (Kreuser,1984). Daly et al. (1991) revealed contractionaldeformation in the Central Congo Basin usingseismic reflection profiles from the CuvetteCentrale. They relate this compressionaldeformation to other areas of east and centralAfrica that experienced strike-slip deformation andrifting during the Late Permian-Early Triassic.

Preliminary fault-kinematic observations andpalaeostress reconstruction along the Lupa Fault,Kanda Fault and Namwele coal field, has revealedan episode of dextral strike-slip faulting alongthe Ubende Belt (Delvaux et al., 1998). Thisdeformation is associated to a north-southhorizontal principal compression. It has affectedthe Permian Karoo Beds, but pre-dates the MidMiocene lateritic peneplain, which is welldeveloped in this area.

356 Journal of African Earth Sciences

In summary, the Karoo period in the WesternRift segment is characterised by two majoralignments of tectono-sedimentary basins:northwest and northeast. Extension is likely tohave been larger than that which is actuallypreserved, probably underlying the Cenozoic riftdepressions. The northwest trending basinsdeveloped over the Ubende Shear Belt with theirleading faults located on the most prominentPalaeoproterozoic Shear Boundary Zone (Fig.2C). The basins probably developed in atranspressional context. The northeast trendingbasins developed in the PalaeoproterozoicUsagara Belt (Fig. 1) in an extensional totranstensional context.

CENOZOIC BASINSCenozoic rift basins in the western segment (Fig.20) are well outlined by the elongate and verydeep lakes as Tanganyika and Malawi and bythe shallow Lake Rukwa. The two deep lakeshave developed outside the Ubende Belt,whereas Lake Rukwa overlies a thicksedimentary pile including the Karoosuccessions. Elevated basement blocks separatethe Cenozoic depocentres, some of which areclearly related to basement reactivation(Theunissen et et., 1996). The Buhoro Basin hasdeveloped along a northeast trend, at right angleto the northwest trending rift basins.

The Rukwa BasinAbove the 3.5 km Karoo sedimentary pile, 7.5km of sediments accumulated in the RukwaBasin (Fig. 2C) (Morley et al., 1992) mostly ofLate Cenozoic age (Westcott et el., 1991;Mbede, 1993). Like the Karoo Beds, thesesediments thicken towards the Lupa BoundaryFault (Dypvik et et., 1990; Morley et al., 1992),implying that the latter is actively involved inthe recent basin formation. A recentlyestablished local seismic network in the MbeyaRange has indicated active faulting along theLupa Fault (Delvaux and Hanon, 1993; Camelbeekand Iranga, 1996). That a large proportion of the7.5 km Late Cenozoic sedimentary pile is of fluvialprovenance is an indication that the depositionof sediments occurred in shallow waterconditions over a long period. These conditionsstill prevail at the present time in the RukwaBasin. This is in sharp contrast with the nearbylakes, where deep water conditions prevail.

Kilembe and Rosendahl (1992) proposed apull-apart origin for the Cretaceous-CenozoicRukwa Basin. They inferred a northwest-


I' 1. //:~ ("jI \ i-----,---"------\:-------- ···_-t.,.----------..

... // \ !J , !

I i, l iI

6° ---",-"-,_",,.,__

Kalemiebasin

":\(

\ ,go . +_~~_-_--..-~-..-

: I -,

'I I

_________________1__ ~~_~~ _Proto-Tanganyika 1

6°

!\! )~ \

! l" \

go ------ ----------t----~---~\

_____ J;'~'Proto-Tanganyika 2

32°

Figure 4. Model evolutionary stages of the Tanganyika Basin. During an early stage, Lake Tanganyika started(Proto- Tanganyika 1) upon the northwest orientated Ubende Belt structure with the development of the KalemieBasin. During a second stage the lake propagated to the north (Proto- Tanganyika 2) as a result of the reactivated(Peleeooroterozoic) horsetail (wrench fault termination), formed the north Kigoma Basin.

southeast extensional direction for the basin.Morley et al. (1992) suggested a more east-westorientated principal extension stress field.Kinematic and stress analyses (Delvaux et al.,1992; Ringet al., 1992) made clear that themain direction of extension varied during theLate Cenozoic evolution.

In the Cenozoic rift pattern Rosendahl et al.(1992) proposed that the Kalemie-Rukwa-northMalawi Basins are all flanked by the mainUbendian Shear Fault Zone, and that they aredown-to-the-east tilted half-grabens. TheKalemie Basin bears no water cover, the RukwaBasin has a very shallow lake (max. 15 rn) andthe northern part of Lake Malawi has a waterdepth of up to 400 m. In between the Kalemieand Rukwa Basins (Fig. 2C) the Mahali Shoalcuts the Cenozoic Lake Tanganyika. Thestructure develops upon an Ubendian terrane andseismic profiling shows a complex faulted horstand graben morphology (Back, 1994). TheMahali Shoal structure had a prominent role inthe evolution of Lake Tanganyika.

In the Ufipa Plateau, between the Tanganyikaand Rukwa Basins, the Proterozoic mylonites inthe Ubende Belt were reactivated several times

(Theunissen et al., 1996). They were reactivatedfirst during the Karoo times, controlling theNamwele Coal Field (McConnell, 1946;Theunissen et el., 1996; Delvaux et aI, 1998).They were again reactivated in the LateQuaternary along the Kanda Fault, flanked by aQuaternary depocentre (Vittori et al., 1997).

Whereas it is quite clear that Cenozoic activerifting processes occur along the northwesttrending Ubende Belt with basins filled with veryshallow water sediments, most obviousCenozoic rift features appear as deep andelongate lakes. Most surprising these deeptroughs do not overly the PalaeoproterozoicUbendian terranes (Fig. 2D) and moreover theirnorth-northwest regional trend is slightly obliqueto the northwest trend along which the aboveshallow level rift basins form.

For the deep lakes Sander and Rosendahl(1989) and Specht and Rosendahl (1989)proposed a model of asymmetric half-grabenswith opposite facing polarities, separated byaccommodation zones as outlined by seismicprofiling. However, with the same geophysicalinformation, Morley (1988) proposed a moresimple model with half-grabens as well as full

Journal ofAfrican Earth Sciences 357

J. KLERKX et al.

grabens. Both models identified the Mahali Shoal,a northwest trending structure cross cutting LakeTanganyika and reflecting a strike-slipdeformational setting as also demonstrated bythe high resolution seismic study (Back, 1994).The Mahali Shoal is a seismically activePrecambrian horst (Wohlenberg, 1969) and focalmechanisms indicate extension oblique to therift trend (Kebede and Kulhanek, 1991).

North of the Mahali Shoal, the northTanganyika Basin enters the MesoproterozoicKibara domain where it is considered as aProterozoic sinistral shear fault terminationsetting. The northeast-southwest trendingUbwari Shoal in the lake is considered as abasement high joining the arcuate western borderfault of the Kigoma Basin (Morley et al., 1990).The Cenozoic north Tanganyika Basin appearsas a "horsetail" termination on a localised shearfault reactivated Ubende structure, evidencedby the Mahali Shoal.

Early stage of the Lake Tanganyika BasinDextral strike-slip movement along the northwesttrending Mahali Fault and its horsetailprolongation toward the Ubwari is suggested tobe the first tectonic control of the Cenozoic basindevelopment in the Tanganyika area (Fig. 4).Transtensive dextral movement along the MahaliUbwari Fault created the Kigoma Basin as adown-to-the-west half-graben, the arcuatewestern border fault acting as the major fault(Proto-Tanganyika 1, Fig. 4).

However, the transfer zone between the LupaFault and the Mahali Fault also played a role atthat time. It is obvious that the Mahali Fault hasshifted to the west, compared to the Lupa Fault.The en echelon transfer between the two faultsappears to occur through the transverseNamwele Fault Zone. This fault zone is alignedalong a west-northwest trend and appears tohave been active during the Karoo, as shownby small outcrops of Karoo sediments. It isconsequently suggested that during the Karoo,the Rukwa Basin was joined to the Kalemie Basinby a small Namwele Basin, continuing into thepresent lake (Fig. 3).

During the Cenozoic, the northwest trendingLupa Fault was reactivated. To the northwest,reactivation was deflected by the Mahali Block(a massive core of less sheared Ubende rocks).The Lupa Fault then swinged to the east alongthe Namwele Depression in a zig-zag patterntowards Lake Tanganyika (a left stepping jumpbetween Lake Tanganyika and Lake Rukwa;Nelson et et., 1992). In this concept, the Proto-


Tanganyika Basin consists of two basins (Fig.4) separated by the Mahali Shoal. The northernKigoma Basin is bordered to the west by thearcuate horsetail fault, a continuation of theMahali Fault. To the south, the smaller KalemieBasin develops as an en echelon pull-apart, incontinuity with the Rukwa Basin. A connectionpossibly existed between the Kalemie and theRukwa Basins through the Namwele transferzone. An hydrological connection between thesebasins is likely to have existed during the lasthumid period in the Early Holocene (Delvaux etal., 1998).

It is obvious that during this early stage thebasin formation was absolutely controlled by thereactivated Proterozoic structures. Mechanismswere very similar to those controlling the Karoobasin development. The only difference is thatthe northwest trending structures were notfollowed over their entire length, but weredeviated along the horsetail faults associatedwith strike-slip faults.

A similar process had already been activeduring Neoproterozoic. This close control of basinformation and evolution is lost during later stagesof the evolution of the Tanganyika Basin.

DISCUSSION AND CONCLUSIONSThe Palaeoproterozoic Ubende Shear Belt is bestconsidered as an early Palaeoproterozoic plateboundary outlined by deep seated ophioliteremnants which have been set in a subsequentand younger Palaeoproterozoic dextral lateralshear belt with amphibolite-facies conditions(Theunissen et al., 1996). Since the Palaeoproterozoic, the shear zone has been repeatedlyreactivated along northwest trending shearcorridors. The repeated wrench faulting regimecreated different Meso- and Neoproterozoicsedimentary basins, as well as the earlyPhanerozoic basins. Only in the Late Cenozoicrifting did this leading control diminish, althoughthe structure remained active.

Reactivation under different tectonicconditions resulted in three main periods of basinformation:

i) During the Meso- and Neoproterozoic,reactivation of the main northwest trending shearfaults under transpressive conditions, resultedin local and isolated strike-slip related (pull-apart)basins of restricted size and relatively shallow.The major basins were formed along the outerborders of the shear belt. The Lupa Fault,corresponding to the eastern limit of the UbendeBelt, was the main active fault.


ii) The Karoo period corresponded to a majorphase of basin formation in eastern and southernAfrica with major reactivation of the Lupa Faultcreating large basins, perfectly aligned along anorthwest trend. These basins formed inaccordance to the strike-slip movement alongthe leading fault and they are of pull-apart type.Although the major basin (the Rukwa Basin)accumulated several km of sediments, there areno indications that these sediments weredeposited in deep water conditions. Tectonicconditions were transtensive all over that period.

iii) During the Cenozoic, there were depositioncenters located along two major alignments. TheLupa Fault was again reactivated, with theRukwa Basin and its southwards prolongationin the northern Malawi Basin acting as depositioncentres again. Meanwhile, the northerly enechelon prolongation of the Lupa Fault, whichwas probably already active during the Karooperiod, was reactivated. A northern basin formedby using the Proterozoic horsetail fault, whichbecame the boundary fault of the stilltranstensive Proto-Tanganyika northern basin.

iv) During the later evolution of the TanganyikaBasin, different tectonic conditions prevailed: thenarrow link between basins and basementstructures was lost, the basin widened towardsnorth and south, sometimes oblique to preexisting structures. Sub-basins such as thenorthern Rusizi Basin appear to be of extensionaltype. Most probably during this last period thebasin evolved to deep water conditions.

ACKNOWLEDGMENTSThe authors are grateful to Peter Bowden, Editorin-Chief of the Journal of African Earth Sciences,for his suggestion to prepare this special issue.The autohrs especially thank the Ministry ofWater, Energy and Minerals (S. Bugaisal, MADINI(P. Kenyunko) and the Department of Geologyof the University of Dar-es-Salaam for theirsupport in the field investigations. This paperbenefitted from the comments and suggestionsof H. Dypvik, E. Kilembe and E. Mbede. Thiswork is a contribution to projects IGCP400"Geodynamics of Continental Rifting" and INTAS93-134 "Continental Rift Tectonics andSedimentary Basin Evolution".

REFERENCESBack, S. 1994. Vergleichende Interpretation reflexions

seismischer Profile im bereich der Kavala IslandAkkomodation Zone, Tanganyika Rift, Tanzania. M. Sc.Dissertation. Universitat Kiel, Germany.

Cahen, L. et Lepersonne, J. 1978. Svrrthe se desconnaissances relatives au Groupe (anciennement Serie)de la Lukuga [Perrnien du Zairel. Musee Royal AfriqueCentrale, Tervuren (Belgique), Annales serie in 8°, SciencesGeoloqiques 82, 115-152.

Camelbeeck, T. and Iranga, M. D. 1996. Deep crustalearthquakes and active faults along the Rukwa trough,Eastern Africa. Geophysical Joumallnternational124, 612630.

Chorowicz, J. et Mukonki, M. B. 1980. Lineaments anciens,zones transformantes recentss et qeotectonique des tossesdans I'est Africain, dapres la te le de tection et lamicrotectonique. Rapport Annuel 1979, pp143-146.Departement de Geologie et Mineraloqie , Muses Royalede I' Afrique Centrale, Tervuren, Belgium.

Chorowicz, J., Roggieri, I., Rudant, J. P., Tamain G. andYanda, P. 1988. Structural survey survey of the Ufipablock (Karema area, Lake Tanganyika, Tanzania). In: Imageanalysis, geological control and radiometric survey ofLandsat TM data in Tanzania, the UNESCO/lUGS GARSProgram (Edited by Lavreau, J. and Bardinet, C.) MuseeRoyalAfrique Centrale, Tervuren (Belgique), Annales seriein 8°, Sciences Geoloqiques 96, 5-11.

Daly, M. C. 1986. Crustal shear zones and thrust belts:their geometry and continuity in Central Africa. RoyalSociety London Philosophical Transactions A317, 111128.

Daly, M. C., Klerkx, J. and Nanyaro, J. T. 1985. Proterozoicstrike-slip accretion in the Ubendian belt of south-westTanzania. Terra Cognita 5, 257.

Daly, M. C., Chorowicz, J. and Fairhead, J. D. 1989. Riftbasin evolution in Africa: the influence of reactivated steepbasement shear zones. In: Inversion Tectonics (Edited byCooper, M. A. and Williams, G. D.) Geological SocietySpecial Publication 44, 309-334.

Daly, M. C., Lawrence, S. R., Kimun'a, D. and Binga, M.1991. Late Paleozoic deformation in central Africa: a resultof distant collision? Nature 350, 605-607.

Delvaux, D. 1991 . The Karoo to Recent rifting in the westernbranch of the East-African Rift System: A bibliographicalsynthesis. Rapport Annuel 1989-1990, pp63-83.Departernent de Geologie et Mineraloqie, Musee Royalede I' Afrique Centrale, Tervuren, Belgium.

Delvaux, D. and Hanon, M. 1993. Neotectonics of the Mbeyaarea, SW Tanzania. Rapport Annuel 1991-1992, pp8797. Departement de Geologie et Mineraloqie, Muses Royalede I'Afrique Centrale, Tervuren, Belgium.

Delvaux, D., Levi, K., Kajara, R. and Sarota, J. 1992.Cenozoic paleostress and kinematic evolution of the Rukwa- North Malawi rift valley (East African rift system). BulletinCentres Recherches Exploration Production Elf-Aquitaine16, 383-406.

Delvaux, D., Kervyn, R., Vittori, E., Kajara, R. S. A. andKilembe, E. 1998. Late Quaternary tectonic activity andlake level change in the Rukwa Rift Basin. In: Tectonics,sedimentation and volcanism in the East African RiftSystem (Edited by Delvaux, D. and Khan, M. A.) JournalAfrican Earth Sciences 26, 399-423.

Ding, Z.-D. and Rosendahl, B. R. 1989. The preliminaryresults of ESP experiment in northern Lake Malawi. EOS,Transactions American Geophysical Union 70, 1211.

Dypvik, H., Nestby, H., Rudent, F., Aagaard, P., Johansson,T., Msindai, J. and Massay, C. 1990. Upper Paleozoicand Mesozoic sedimentation in the Rukwa-TukuyuRegion, Tanzania. Journal African Earth Sciences 11,437-456.

Fairhead, J. D. and Green, C. M. 1989. Controls on rifting inAfrica and the regional tectonic model for Nigeria and EastNiger rift basins. Journal African Earth Sciences 8, 231250.

Journal ofAfrican Earth Sciences 359

J. KLERKX et al.

Fernandez-Alonso, M. and Theunissen, K. In press. Airbornegeophysics and geochemistry provide new insights in theintracontinental evolution of the Mesoproterozoic Kibarabelt (Central Africa). Geological Magazine.

Halbich, I. W. 1983. A tectogenesis of the Cape Fold Belt.Special Publication Geological Society South Africa 12,165-175.

Halligan, R. 1962. The Proterozoic rocks of westernTanganyika. Bulletin Geological Survey Tanganyika 34,34p.

Harpum, J. R. 1970. Summary of the geology of Tanzania.Part V: Structure and geotectonics of the Precambrian.Mineral Ressources Division Tanzania Memoir 1, 58p.

Kazmin, V. 1980. Transform faults in the East African riftsystem. In: Geodynamic Evolution of the Afro-Arabian RiftSystem. Accademia Nazionale dei Lincei, Roma. Atti deiconvegni Lincei 47,65-73.

Kebede, F. and Kulhanek, O. 1991. Recent seismicity of theEast African Rift System and its implications. Physics EarthPlanetary Interiors 68, 259-273.

Kilembe, E. A. and Rosendahl, B. R. 1992. Structure andstratigraphy of the Rukwa Basin. Tectonophysics 209,143-158.

Klerkx, J. and Nanyaro, J. T. 1988. The Ukinga Group inSouthwestern Tanzania: a shear belt of Middle Proterozoicage? IGCP Project N°255 Newsletter 1, 37-41.

Klerkx, J., l.ieqeois, J.-P., Lavreau, J. and Claessens, W.1987. Crustal evolution of the northern Kibara belt, Easternand Central Africa. In: Proterozoic Lithospheric Evolution(Edited by Kroner, A.) American Geophysical UnionGeodynamic Series 17, 217-233.

Kreuser, T. 1984. Source rocks analysis in two Karoo basinsof coastal Tanzania. Journal Petroleum Geology 7, 4754.

Kreuser, T. and Semkiwa, P. M. 1987. Geometry anddepositional history of a Karoo (Permian) Coal basin(Mchuchuma-Ketewaka) in SW-Tanzania. Neues JahrbuchGeologie Peleontoloqie Montatshefte 1987, 69-98.

Lenoir, J.-L., l.ieqeois. J.-P., Theunissen, K. and Klerkx, J.1994. The Paleoproterozoic Ubendian shear belt inTanzania: geochronology and structure. Journal AfricanEarth Sciences 19, 169-184.

Maloney, R. and Downie, C. 1972. Palynomorphs from theBuanji series. Bukoban System, and their stratigraphicalimpications. In: Paleoecology ofAfrica and the surroundingislands, Cape Town, 7, 18p.

Marobhe, I. M. 1989. Interpretation of aerogeophysicalanomalies of south-western Tanzania. Geological SurveyFinland, Bulletin 350, 72p.

Mbede, E. I. 1993. Tectonic development of the Rukwa Riftbasin in SW Tanzania. Berliner GeowissenschaflicheAbhandlungen A 152, 92p.

McConnell, R. B. 1946. The geology of the NamweleMkomolo Coalfield. Imperial Institute Bulletin 44, 227-251,330-354.

McConnell, R. B. 1967. The East-African rift system. Nature215,578-581.

McConnell, R. B. 1972. Geological development of the riftsystem of eastern Africa. Geological Society AmericaBulletin 83, 2549-2572.

McKinlay, A. C. M. 1965. The coal fields and coal resourcesof Tanzania. Geological Survey Tanzania, Bulletin 38, 82p.

Morley, C. K. 1988. Variable extension in Lake Tanganyika.Tectonics 7, 785-801.

Morley, C. K., Nelson, T. L., Patton, T. L. and Muss, S. G.1990. Transfer zones in the East African rift system.American Association Petroleum Geologists Bulletin 74,1234-1253.

Morley, C. K., Cunningham, S. M., Harper, R. M. andWescott, W. A. 1992. Geology and geophysics of theRukwa rift, East Africa. Tectonics 11, 68-81.


Nelson, R. A., Patton, T. L. and Morley, C. K. 1992. Riftsegment interaction and its relation to hydrocarbonexploration in continental rift systems. AmericanAssociation Petroleum Geologists Bulletin 76, 1153-1169.

Page, E. N. 1960. The stratigraphical and structuralrelationship of the Abercorn Sandstones, the Plateau Seriesand basement rocks of the Kawimbe area, AbercornDistrict, Northern Rhodesia. Annual Report 1960-1961,29p. Research Institute of African Geology, University ofLeeds, UK.

Peirce, J. and Lipkov, L. 1988. Structural interpretation ofthe Rukwa rift, Tanzania. Geophysics 53, 824-836.

Piper, J. D. A. 1975. Paleomagnetic correlation ofPrecambrian formations of east-central Africa and theirtectonic implications. Tectonophysics 26, 135-161.

Ring, U., Betzler, C. and Delvaux, D. 1992. Normal vs. strikeslip faulting during rift development in East Africa: theMalawi rift. Geology 20, 1015-1018.

Rosendahl, B. R. 1987. Architecture of continental rifts withspecial reference to East Africa. Annual Review EarthPlanetary Sciences 15, 445-503.

Rosendahl, B. R., Kilembe, E. and Kaczmarick, 1992.Comparison of the Tanganyika, Malawi, Rukwa andTurkana rift zones from analyses of seismic reflection data.Tectonophysics 213, 235-256.

Sander, S. and Rosendahl, B. R. 1989. The geometry ofrifting in Lake Tanganyika, East Africa. Journal AfricanEarth Sciences 8, 323-354.

Smirnov, V., Pentel'kov, V., Tolochov, V., Trifan, M. andZhukov, S. 1973. Geology and Minerals of the CentralPart of the Western Rift. Unpubl. report 333p. Ministry ofCommerce and Industry, Dodoma, Tanzania.

Specht, T. D. and Rosendahl, B. R. 1989. Architecture ofthe Lake Malawi Rift, East Africa. Journal African EarthSciences 8, 355-382.

Sutton, J. and Watson, J. V. 1986. Architecture of thecontinental crust. Royal Society London PhilosophicalTransactions A317, 5-12.

Sklyarov, E. V., Theunissen, K., Melnikov, A. T, Klerkx, J.,Gladkochub, D. P. and Mruma, A. In press.Paleoproterozoic eclogites and garnet pyroxenites of theUbende belt (Tanzania). Swiss Bulletin MineralogyPetrology.

Tack, L. 1995. The Neoproterozoic Malagarazi Supergroupof SE Burundi and its equivalent Bukoban system in NWTanzania: A current review. In: Late Proterozoic Belts inCentral and Southwestern Africa, IGCP Project 302 (Editedby Wendorff, M. and Tack, L): Musee Royal AfriqueCentrale, Tervuren (Belgique), Annales serie in 8°, SciencesGeoloqiques 101,121-130.

Tack, L., Sindayihebura, A. and Cimpaye, D. 1992. TheNkoma 5SE Burundi): an episodically reactivated lowerBurundian (Middle Proterozoic) siliciclastic sequence,locally overlain by a Malagarasian (Upper Proterozoic)sedimentary breccia. IGCP n " 255 Newsletter/Bulletin4,31-43.

Theunissen, K. 1988a. The Ufipa shear zone in the Ubendebelt at Karema (W. Tanzania): A NW oriented left lateralstrike-slip of postulated Kibaran age. Musee Royal AfriqueCentrale, Tervuren (Belgique), Annales serie in 8°, SciencesGeoloqiques 96,63-77.

Theunissen, K. 1988b. Evidence for Upper Proterozoicconjugate strike-slip basins controlled by basementstructures in northwest Tanzania and Burundi. SocietebeIge Geologie Bulletin 97(2, 115-130.

Theunissen, K., Lenoir, J.-L., Lieqeois, J.-P., Delvaux, D. etMruma, A. 1992. Empreinte mozambiquienne majeuredans la chaine ubendienne de Tanzanie sud-occidentale:geochronologie U-Pb sur zircon et contexte structural.Compte Rendus Acedemie Sciences Paris, Serie II 314,1355-1362.


Theunissen, K., Klerkx, J., Melnikov, A. and Mruma, A. 1996.Mechanism of inheritance of rift faulting in the westernbranch of the East African Rift, Tanzania. Tectonics 15,776-790.

Tiercelin, J.-J., Chorowicz, J., Bellon, H., Richert, J.-P.,Mwambene, J. T. and Walgenwitz, F. 1988. East Africanrift system: offset, age and tectonic significance of theTanganyika-Rukwa-Malawi intracontinental fault zone.Tectonophysics 148, 241-252.

Unrug, R. 1984. The Mid-Proterozoic Mporokoso Group ofnorthern Zambia: stratigraphy, sedimentation and regionalposition. Precambrian Research 24, 99-121.

Verniers, J., Jourdan, P. P., Paulis, R. V., Frasca-Spada, L.and de Bock, F. R. 1989. The Karoo graben of Metangula,northern Mozambique. Journal African Earth Sciences 9,137-158.

Vittori, E., Delvaux, D. and Kervyn, F. 1997. Kanda Fault: amajor seismogenic element west of the Rukwa rift (EastAfrica, Tanzania). In: Paleoseismology: using Quaternarygeology to understand past earthquakes (Edited byHancock, P. L. and Michetti, A. L.) Journal Geodynamics24, 139-153.

Wescott, W. A., Krebs,W. K., Engelhardt.D. W. and Cunningham,S. M. 1991. New biostratigraphic age dates from the LakeRukwa rift basin in Western Tanzania. American AssociationPetroleum Geologists Bulletin 75, 1255-1263.

Wheeler, W. H. and Karson, J. A. 1989. Structure andkinematics of the Livingstone Mountains border fault zone,Nyasa (Malawi) Rift, southwestern Tanzania. JournalAfrican Earth Sciences 8, 393-414.

Wheeler, W. H. and Rosendahl, B. R. 1994. Geometry ofthe livingstone Mountains Border Fault, Nyasa (Malawi)Rift, East Africa. Tectonics 13, 303-312.

Wopfner, H. 1994. The Malagasy Rift, a chasm in theTethyan margin of Gondwana. Journal Southeast AsianEarth Sciences 9, 451-461.

Wopfner, H. and Kaaya, C. Z. 1992. Syndepositional riftingin the Ruhuhu and Selous basin - Tanzania. In: Geologyfor Economic Development (Edited by UNESCO) Newsletter9,159-168.

Wohlenberg, J. 1969. Remarks on the seismicity of EastAfrica between 4°N - 12°S and 23°E - 40 0E.

Tectonophysics 8,567-577.

Geological mapsGeological map of sectors of Gondwana, 1: 10 000 000.

1988. Compiled by de Wit, M., Jeffer, Y. M., Bergh, H.and Nicolaysen, L.American Association PetroleumGeologists and University of the Witwatersrand (Eds).

Geological map of Kipengere, Quarter degree sheet 79W;1:125000. 1958. Compiled by Harpum, J. R. and Harris,J. F. Geological Survey Tanganyika, Dodoma.


persistent fault controlled basin formation since the ......basement fault reactivation - the kibara...

Documents