mesoarchean assembly and stabilization of the eastern kaapvaal

Mesoarchean assembly and stabilization of the eastern Kaapvaal

craton: A structural-thermochronological perspective

Blair Schoene,1,2 Maarten J. de Wit,3 and Samuel A. Bowring1

Received 26 January 2008; revised 16 May 2008; accepted 21 July 2008; published 17 October 2008.

[1] Understanding the construction and stabilizationof Archean continental lithosphere has importantimplications for models of tectonic and thermalregimes in the early Earth, as well as for thesubsequent evolution of continents. We provide newconstraints for the amalgamation and stabilization ofthe eastern Kaapvaal craton in the vicinity of theBarberton greenstone belt circa 3.3–3.1 Ga. Isotopedilution thermal ionization mass spectrometry U-Pbgeochronology and thermochronology are combinedwith mapping and structural analysis around themargins of this belt in order to constrain movementon major shear zones that separate high-gradeorthogneiss terranes from low-grade supracrustalrocks and to relate this movement to the timing ofdifferential exhumation. New and existing data areconsistent with accretion at �3.23 Ga within anoblique subduction plate boundary manifested in theBGB as an asymmetric flower-like structuralgeometry. For >100 Ma following terrane assembly,transform boundary tectonics, manifested partially astranstension on reactivated faults, led to large-scalejuxtaposition of middle to lower crustal basementorthogneiss complexes against upper crustalgreenstone sequences and episodic emplacement ofgranitic batholiths. Stabilization of this portion ofArchean lithosphere involved at least a three-stageprocess: (1) creation of a thick and rigid mantlelithosphere during crustal growth, subduction, andterrane assembly from circa 3.30 to 3.23 Ga,(2) generation of a rigid crust during strike-slip andtranstensional tectonics coupled with migration ofgranites and heat-producing elements into the uppercrust between circa 3.2 and 3.1 Ga, and (3) erosionand removal of the uppermost crust during regionalpeneplanation that lasted until circa 2.9 Ga.Citation: Schoene, B., M. J. de Wit, and S. A. Bowring

(2008), Mesoarchean assembly and stabilization of the eastern

Kaapvaal craton: A structural-thermochronological perspective,

Tectonics, 27, TC5010, doi:10.1029/2008TC002267.

1. Introduction

[2] Archean cratons are characterized by their resistanceto recycling into the mantle. This may be in part explainedby the observation that numerous physical characteristics ofArchean lithosphere are different from younger lithosphere.The most prominent example is that Archean cratoniclithosphere has been shown by numerous tomographicstudies to have a thick and relatively high-velocity mantleroot [Fouch et al., 2004; James et al., 2001; Jordan, 1978,1988; Ritsema et al., 1998; Rudnick and Nyblade, 1999;Simons et al., 1999]. Jordan [1978] suggested that retaininga thick and ‘‘cold’’ buoyant lithosphere requires depletion inheavy elements such as iron, possibly by the removal ofbasaltic melt in a subduction zone setting, creating aphysically thick and relatively buoyant tectosphere thatwill not sink into the underlying asthenospheric mantle[also see Jordan, 1988]. This so-called isopycnic theory isconvenient in that it also explains the observed lowersurface heat flow of some cratons compared to adjacentterranes [Morgan, 1985; Perry et al., 2006] and predicts arelatively cool mantle lithosphere that is capable of preserv-ing ancient diamonds [e.g., Richardson et al., 1984; Shireyet al., 2002; Westerlund et al., 2006].[3] Continental lithosphere generated in convergent

collisional regimes concurrent with terrane amalgamationshould yield thick continental crust relative to adjacentterranes. In reality, Archean cratonic crustal thickness todayis characterized by highly variable thickness [Doucoure andde Wit, 2002; Durrheim and Mooney, 1994; Gupta et al.,2003; James and Fouch, 2002; Nguuri et al., 2001], but onaverage thinner than adjacent younger terranes [Durrheimand Mooney, 1994; Assumpcao et al., 2002; Clitheroe et al.,2000; Perry et al., 2002]. If cratonic mantle lithosphere isconstructed of mantle lithosphere and asthenosphere depletedduring subduction zone magmatism and accreted duringcollision, then thinner crust can be explained by surfaceerosion to maintain isostatic equilibrium with a positivelybuoyant mantle lithosphere [Perry et al., 2002]. However,an important observation is that in cratonic settings low-grade supracrustal rocks are commonly juxtaposed againstmiddle to lower crustal basement orthogneiss [e.g., de Witand Ashwal, 1997; Van Kranendonk et al., 2007], in what isoften termed ‘‘dome-and-keel’’ structural geometry [e.g.,Marshak et al., 1992]. This is inconsistent with crustalthinning by isostatic forces and erosion alone. This obser-vation, in concert with the heterogeneity of crustal thickness

TECTONICS, VOL. 27, TC5010, doi:10.1029/2008TC002267, 2008ClickHere

for

FullArticle

1Department of Earth, Atmospheric and Planetary Sciences,Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.

2Now at Departement de Mineralogie, Universite de Geneve, Geneva,Switzerland.

3African Earth Observatory Network and Department of GeologicalSciences, University of Cape Town, Rondebosch, South Africa.

Copyright 2008 by the American Geophysical Union.0278-7407/08/2008TC002267$12.00

TC5010 1 of 27

http://dx.doi.org/10.1029/2008TC002267

observed in some cratons, suggests that postassembly mod-ification of the crust is an important mechanism for itsstabilization, and that the processes and timing of stabiliza-tion may vary widely. Determining a high-resolution time-scale for the tectonic and thermal evolution of cratonic crustis essential therefore to determine the relative importance ofdifferent processes leading to lithospheric stabilization.[4] The Kaapvaal craton in southern Africa (Figure 1) is

ideal for the study of Archean lithospheric evolution, in partbecause many aspects of its physical and chemical proper-ties are well understood from seismic [Fouch et al., 2004;James et al., 2001; Nguuri et al., 2001] and xenolith [Belland Moore, 2004; Carlson et al., 2000; Pearson et al.,1995; Schmitz and Bowring, 2003a; Shirey et al., 2001]studies. This work has shown that there is a �250–300 kmthick chemically depleted mantle root beneath continentalcrust that varies from 32 to 50 km thick within the craton,adjacent to Proterozoic lithosphere with significantly thin-ner mantle and consistently >40 km of crust. The oldest andbest exposed basement rocks in the craton are located ineastern South Africa and Swaziland, in the vicinity of theBarberton greenstone belt (BGB; Figures 1 and 2). Inaddition to hosting significant economic concentrations of

gold, this area has become a classic locality for understand-ing Archean tectonics, sedimentology and geobiology sincethe work of Viljoen and Viljoen [1969]. The apparentresistance to deformation of this region for some 3 Gamarks it as a prime target for understanding how it becamepart of stable cratonic lithosphere. The amalgamation ofthe eastern Kaapvaal craton occurred between about 3.7 and3.1 Ga, culminating at circa 3.23 Ga during the collision andamalgamation of two or more microcontinents [de Rondeand de Wit, 1994; de Ronde and Kamo, 2000; de Wit, 1982;Kamo and Davis, 1994; Lowe, 1994]. Stabilization of thecrust may have began circa 3.2–3.1 Ga during the intrusionof large subhorizontal granitic batholiths and a transitiontoward dominantly strike-slip and transtensional tectonics[Anhaeusser, 1983; de Ronde et al., 1994; de Wit et al.,1992; Jackson et al., 1987; Lowe and Byerly, 1999a;Westraatet al., 2005], which was followed by the erosion of at least10 km of crust during regional peneplanation [e.g., de Wit etal., 1992]. However, the period between 3.2 and 3.1 Gafollowing initial amalgamation is poorly understood, despiteits importance in initiating stabilization of this portion of thecraton. In particular, the timing and genetic relationshipbetween granitic magmatism and strike-slip tectonics and

Figure 1. Geologic map of exposed Archean rocks in the Kaapvaal craton, modified from Schmitz et al.[2004]. Included are the major subdivisions of the craton and the distribution of the major Neoarcheandepositional basins. Position of Figure 2 is outlined by black box.

TC5010 SCHOENE ET AL.: EVOLUTION OF EAST KAAPVAAL CRATON

2 of 27

TC5010

their importance in modifying the postassembly circa3.23 Ga crustal architecture remain speculative.[5] This paper summarizes a structural, geochronological,

and thermochronological study aimed at constructing ahigh-precision thermotectonic timeframe that documentsthe relationship between supracrustal sequences and granit-oid and orthogneiss complexes near the margins of the BGBimmediately prior to terminal cratonization. We have usedstructural analysis on several key shear zones that separatehigh-grade orthogneiss from lower-grade greenstone beltrocks in order to characterize the kinematics of the finalperiod of deformation in this area. High-precision isotopedilution thermal ionization mass spectrometry (ID-TIMS)U-Pb zircon geochronology by using new zircon chemicalabrasion techniques [Mattinson, 2005] is employed onorthogneisses and crosscutting syntectonic intrusions inshear zones to bracket the timing of deformation anddetermine crystallization ages with less than million yearprecision and accuracy. U-Pb apatite thermochronology incombination with existing thermochronological data con-strain the relative exhumation history of different crustalblocks between circa 3.45 and 3.1 Ga within a tectonicframework established by structural geology and geochro-nology. The results help differentiate the effects of circa3.23 Ga transpressional assembly from subsequent strike-slip and transtensional modification of the crust circa 3.2–3.1 Ga, and the role of each process in defining the final

architecture of the Mesoarchean craton which has beenrelatively stable for circa 3 Ga.

2. The 3.6–3.1 Ga Evolution of the Eastern

Kaapvaal Craton

[6] The present-day character of the eastern Kaapvaalcraton is the result of a complex and protracted evolutionduring the Mesoarchean, followed by �3 Ga of relativestability. Of primary importance for this study is the tectonicdevelopment circa 3.23–3.10 Ga, which began with aperiod of crustal assembly that is inferred to have largelyinfluenced the present-day architecture of the crust, and issummarized below (Figure 2).[7] This study is concerned with the Barberton Green-

stone Belt (BGB) and the adjacent basement orthogneissterranes. The stratigraphy of the BGB is locally variable andcharacterized by a diverse set of unconformities, structuralrepetitions and deformation patterns. It is broadly subdi-vided into three main units: The Onverwacht Group (circa3.55–3.25 Ga), the Fig Tree Group (circa 3.25–3.22 Ga),and the Moodies Group (3.23–? Ga) [Anhaeusser, 1969,1976, 1983; Condie et al., 1970; de Wit, 1982; de Wit et al.,1987b; Eriksson, 1980; Armstrong et al., 1990; Heubeckand Lowe, 1994a, 1994b; Lowe and Byerly, 1999a; Viljoenand Viljoen, 1969] (Figure 2). The Sandspruit and Thee-

Figure 2. Geologic map of the Barberton greenstone belt and adjacent orthogneiss basement rocks, withfeatures from the text labeled. Position of maps in Figures 3, 6, and 9 are outlined in black boxes. Thickblack lines indicate positions of important faults and shear zones. PSZ, Phophonyane Shear Zone; HSZ,Honeybird Shear Zone; SSZ, Steynsdorp Shear Zone; SIFS, Saddleback-Inyoka fault system. Mapcompiled from this study, Anhaeusser [1983], de Ronde et al. [1994], de Wit [1982], Lowe and Byerly[1999a], and Wilson [1982].


3 of 27

TC5010

spruit formations structurally make up the lowermost unitsof the Onverwacht Group. The former has an abundance ofkomatiites and the latter comprises mafic to silicic fine-grained metasediment and volcaniclastic rocks. The upperOnverwacht Group consists of komatiitic to basaltic meta-volcanics and interlayered siliciclastic units, cherts, banded-iron formations, and tuffs, which are divided into the (frombottom to top) Komati, Hooggenoeg, Kromberg, Mendon,and Weltevreden formations. The overlying Fig Tree Groupconsists of fine-grained shales, siltstones, cherts and silicicto intermediate volcanics that coarsen upward into chertconglomerates that lie below the coarser-grained quartzosesandstones and conglomerates of the Moodies Group. Thesource, depositional setting, and age of the Moodies Grouplikely differs across the greenstone belt [Lamb and Paris,1988] and even within single stratigraphic sequences[Heubeck and Lowe, 1994a, 1994b; Lowe and Byerly, 1999b].[8] The BGB is bordered on all sides by gneisses and

plutonic rocks that span circa 500 Ma in age (Figure 2). TheAncient Gneiss Complex (AGC), exposed in Swaziland,flanks the SE margin of the belt and contains the oldestrocks yet identified in the Kaapvaal craton. The AGCexamined within this study is composed primarily ofbanded felsic to mafic gneisses that were metamorphosedand deformed during several distinct periods at >3.6 Ga,3.55–3.50 Ga, and circa 3.45 Ga [Hunter et al., 1978;Jackson, 1979, 1984; Hunter et al., 1984; Compston andKroner, 1988; Kroner et al., 1989]. These older rocks wereintruded by a series of tonalites and granodiorites circa3.23 Ga that were subsequently ductilely deformed tovarying degrees [Kroner et al., 1989; Schoene and Bowring,2004; Wilson, 1982]. Orthogneiss and igneous complexesare also juxtaposed against the southern and northwesternmargins of the BGB, namely the circa 3.51 Ga Steynsdorppluton (or Steynsdorp dome [Kisters and Anhaeusser,1995b]), the circa 3.45 Ga Stolzburg, Theespruit, andDoornhoek plutons and the circa 3.23 Kaap Valley andNelshoogte plutons (Figure 2). These rocks are intruded bya series of circa 3.107 Ga granitic batholiths and syenogra-nitic intrusions with variable magmatic to subsolidus fabricsalong all margins of the BGB [Anhaeusser, 1983; Kamo andDavis, 1994; Westraat et al., 2005] except for the SEmargin, where the Pigg’s Peak batholith intruded at circa3.14 Ga [Schoene and Bowring, 2007].[9] We note here that there is potential for confusion

regarding the terminology used historically to describe thesegranitoid rocks flanking the margins of the BGB. This hasresulted from the use of the word ‘‘pluton’’ to describegroups of granitic rocks that are presumed or shown to sharea common igneous origin (either temporally or genetically).Many rocks from a single ‘‘pluton,’’ however, were subse-quently deformed ductilely such that they would now bemore aptly described as gneisses. Throughout this paper, wereplace the historically used term ‘‘pluton’’ with ‘‘gneiss’’ or‘‘complex’’ for groups of rocks that are composed mostly oforthogneisses or a mixture of such orthogneisses andigneous rocks, respectively (e.g., the Stolzburg pluton,described herein as the Stolzburg complex). An exceptionis in the Kaap Valley and Nelshoogte plutons, which are

ubiquitously foliated and/or lineated but not banded orlayered and may be relatively uniform in age within theirmapped boundaries.[10] The relationship between the orthogneiss complexes

and the BGB is similar in many places to that observed inother Archean and younger greenstone belts: greenschist- orlower-grade supracrustal rocks in a series of upright tight toisoclinal antiform/synform pairs separated by shear zonesfrom higher-grade basement rocks [i.e., de Wit and Ashwal,1997; Van Kranendonk et al., 2007]. This structural char-acter is often called ‘‘dome-and-keel’’ structure, and hasbeen used to generate various models that apply uniquely toArchean tectonic processes, or lack thereof (see discussionsby Lin [2005, and references therein] and Marshak et al.[1997, and references therein]). In the BGB, however,contacts between some igneous rocks and supracrustal rocksretain an intrusive relationship (for example, in the Stolzburgcomplex, described in this paper and by Kisters andAnhaeusser [1995a]). Hornblende 40Ar/39Ar data combinedwith P-T estimates show that some circa 3.45 Ga mafic toultramafic volcanics (in the ‘‘keels’’) were never above low-greenschist to lower-amphibolite-grade conditions [Cloete,1991; Lopez Martinez et al., 1992; Xie et al., 1997]. Peakmetamorphic conditions of portions of the Komati andHooggenoeg formations are inferred to have been generatedimmediately following their formation and burial on theseafloor [de Ronde and de Wit, 1994; de Wit, 1982; de Wit etal., 1987a, 1987b, 1992; Lowe, 1999; Williams and Furnell,1979]. In contrast, SWof the BGB (in the ‘‘domes’’), sliversof amphibolites and mafic semipelites within the circa3.45 Ga orthogneiss complexes yield P-T estimates fromupper-amphibolite- to granulite-grade conditions [Dieneret al., 2005;Dziggel et al., 2002; Kisters et al., 2003; Stevenset al., 2002]. Significantly, some slivers of rocks record high-P, low-T metamorphism typical in modern subduction zonesettings [Moyen et al., 2006]. The timing of this metamor-phism and deformation in the rocks SW of the BGB isconstrained by a U-Pb date of circa 3.23 Ga on a synkine-matic intrusion [Dziggel et al., 2005] and by U-Pb sphenedates of circa 3.23 Ga, which provide a minimum estimatefor the timing of metamorphism [Diener et al., 2005;Dziggel et al., 2005]. These data are consistent with earlierinvestigations that inferred the existence of a period ofsubduction-accretion at circa 3.23 Ga. An orthogneissxenolith from the Nelspruit batholith north of the BGBgives ages as old as 3.3 Ga [Kamo and Davis, 1994],suggesting that crustal growth (associated with subduc-tion?), may have in fact begun much earlier.[11] The circa 3.23 Ga period of convergent deformation

is suggested to have generated the dominant NE–SWstructural trend of the BGB during collision and suturingalong the Saddleback-Inyoka fault system (SIFS in Figure 2)[de Ronde and de Wit, 1994; de Wit et al., 1992]. NW–SEdirected thrusting and syntectonic sedimentation of theMoodies Group lithologies has been widely reportedthroughout the belt [de Ronde and de Wit, 1994; de Rondeand Kamo, 2000; de Wit et al., 1992; Heubeck and Lowe,1994b; Jackson et al., 1987; Lamb, 1984; Lowe, 1999].Constraints on the timing of NW–SE thrusting within the


4 of 27

TC5010

BGB are provided by U-Pb dating of circa 3.23 Gaintrusions and volcanic rocks that bracket the generationof folds and unconformities in the Fig Tree and lowerMoodies Groups that developed during shortening in thecentral and western portion of the BGB [de Ronde and deWit, 1994; Kamo and Davis, 1994; de Ronde and Kamo,2000].[12] Following NW–SE convergence and accretion, there

was a transition toward NE–SW strike-slip to transtensionalfaulting and sedimentary deposition through at least thecentral and NW areas of the BGB [de Ronde and de Wit,1994; de Ronde and Kamo, 2000; de Ronde et al., 1991;Heubeck and Lowe, 1994a, 1994b; Jackson et al., 1987;Lowe, 1999]. Strike-slip and extensional kinematics havebeen suggested to be either the result of gravitationalcollapse immediately following circa 3.23 Ga accretion[Kisters et al., 2003] or �100 Ma later contemporaneouswith circa 3.11 Ga granitic intrusions [de Ronde and de Wit,1994; Jackson et al., 1987; Kamo and Davis, 1994;Westraatet al., 2005], though few temporal constraints exist for thisevent; de Ronde et al. [1991] bracketed movement on onedominantly strike-slip fault near Barberton to have occurredbetween 3126 ± 21 and 3080 ± 18 Ma by U-Pb dating of asynkinematic porphyry and postkinematic rutile growth.Rocks inferred to be circa 3.11 Ga associated with theMpuluzi batholith at the southern end of the Stolzburgcomplex intruded synkinematically into an extensionaltranscurrent shear zone [Westraat et al., 2005]. Schoeneand Bowring [2007] used U-Pb apatite and sphene thermo-chronology to show that circa 3227 Ma orthogneisses of theKaap Valley pluton and the Pigg’s Peak inlier (Figure 2)experienced dramatically different cooling histories, andthat the basement to the SE in Swaziland experiencedpostintrusion middle to lower crustal residence followedby rapid cooling from �3140 to 3130 Ma concomitant withintrusion of the Pigg’s Peak batholith. They inferred thatrapid cooling SE of the BGB may have been related to localor regional exhumation along the gneiss-greenstone contact.These data suggest that the scale and effect of strike-slip totranstensional movement throughout the belt may have beenlarger than previously inferred from structural observations.Therefore, measurements that define the kinematics andtiming of circa 3.23–3.10 Ga tectonism along all marginsof the belt are crucial in understanding the final structuraland thermal stabilization of this portion of the Kaapvaalcraton. Here, we report on such work along four of thesemargins.

3. Field Locales, Kinematic Analysis,

and Geochronology

[13] This section summarizes the field observations fromthe selected localities and the samples taken forU-Pb ID-TIMSgeochronology and thermochronology. U-Pb zircon datawas collected primarily using chemical abrasion techniques[Mattinson, 2003, 2005], though some air abrasion data[Krogh, 1982] are also presented. In general, the chemicalabrasion data resulted in a high percentage of concordantdata, and only concordant zircon data (±0.5% concordance)

were used for 207Pb/206Pb age interpretation. More dis-cordant apatites were also used for cooling information,though 207Pb/206Pb dates provide only a minimum age.Sample descriptions and locations, detailed geochronologicalresults and their interpretations are given in the auxiliarymaterial.1 A summary of these dates and interpretations aregiven in Table 1. All geochronological uncertainties [Jaffey etal., 1971; Renne et al., 1998; Mattinson, 2000; Min et al.,2000; Begemann et al., 2001; Schoene et al., 2006] in thispaper are reported at the 95% confidence level and ignoredecay constant and tracer calibration uncertainties. See theauxiliary material for more analytical details.

3.1. Southeast Margin: Pigg’s Peak Inlier

3.1.1. Previous Work[14] The Pigg’s Peak inlier (Figure 3) is the only expo-

sure in the region where rocks of the AGC are in directcontact with any portion of the BGB. Along this contact isan �1.5 km wide zone of mylonitic AGC and othermetaigneous rocks, here termed the Phophonyane ShearZone, PSZ. No detailed lithologic or structural descriptionof the PSZ has been published. Compston and Kroner[1988] presented a U-Pb SHRIMP data set on zircons froma sample of the lithologically complex gneiss from the inlierSE of the PSZ. Apparently concordant U-Pb dates rangingfrom circa 3.6 to 3.0 Ga are found in the zircons, butmultiple growth zones, complex Pb loss and negativediscordance in the analyses prevent a straightforward inter-pretation of the results. An in situ weighted-mean207Pb/206Pb date of 3644 ± 4 Ma from that study is theoldest date that has previously been published fromthe Kaapvaal craton. Kroner et al. [1989] presented furtherU-Pb SHRIMP dates on two leucocratic granitic gneisssamples from this inlier. These analyses are generallydiscordant with an upper intercept age of 3179 ± 11 Ma.Schoene and Bowring [2007] reported U-Pb zircon, sphene,and apatite data from a granodioritic gneiss sample withinthe inlier (AGC01-4; Figure 3). They conclude that the rock,which gives a 207Pb/206Pb age of 3226.1 ± 0.7 Ma, had aprotracted cooling history between 3.23 and 3.05 Ma, whichmay be due to circa 3.14–3.13 Ga exhumation along thePSZ.3.1.2. Mapping and Structural Analysis[15] Mapping, structural analysis, and geochronology

from the Pigg’s Peak inlier are directed toward understand-ing the age relationships of the different intrusions, deter-mining the timing and kinematics of movement along thePSZ, and relating that to the regional thermal structure in theMesoarchean. Figure 3 shows a lithologic and structuralmap of the Pigg’s Peak inlier, as well as a structural crosssection. Rocks of the inlier consist of a complicated series ofigneous and metaigneous rocks that we broadly categorizeinto several mappable units, described below. The inlier iscrosscut in the SE by the medium- to coarse-grainedundeformed Pigg’s Peak granite, which gives a 207Pb/206Pb

1Auxiliary material data sets are available at ftp://ftp.agu.org/apend/tc/2008tc002267. Other auxiliary material files are in the HTML.


5 of 27

TC5010

Table

1.SummaryofU-PbID

-TIM

SDating

Sam

plea

Unitb

Rock

Description

Mineralc

Dated

MSWDe

Number

of

Single-G

rain

Analysesf

Number

of

AnalysesUsedg

Interpretation/Significance

PhophonyaneInlier

AGC01-5

AGC

porphyritictonalitic

gneiss

z3662.8

±0.5

0.3

12AA

+10CA

3oldestdatefrom

Kaapvaalcraton

EKC02-20

AGC

fine-grained

foliated

granite

z3545.6

±0.5

0.9

7CA

3igneouscrystallization

EKC02-8

N/A

apliticsyntectonic

dike

z7AA

+4CA

0inherited/discordantgrains

EKC02-10

N/A

granitic

pegam

titicsyntectonic

dike

z3223.4

±1.9

1.0

7AA

+3CA

2postdates

mostlocalshearingin

PSZ

Steynsdorp

Complex

EKC02-40

Steynsdorp

gneiss

tonalite/granodiorite

z3517.6

±0.7

0.6

4CA


a3160–3190

97

coolingbelow

�400�C

EKC02-38

N/A

apliticdike

z3103.7

±0.8

5CA

1crystallization/deform

ationofdike

EKC02-46

N/A

pegmatitic

dike

z3103.7

±0.8

5CA


ationofdike

EKC02-47

N/A

pegmatitic

dike

z3102.4

±0.8

6CA


ationofdike

WKC00-88

Vlakplaatsgd.

undeform

edgranodiorite

z3229.8

±0.6

1.9

7CA


a3200–3230

66

coolingbelow

�400�C

HoneybirdShearZone

KPV99-90

Stentorbasem

ent

tonalitefrom

banded

gneiss

z3258.3

±0.3

1.0

9CA


KPV99-89

Stentororthogneiss

granitic

orthogneiss

z3106.0

±0.5

1.2

7CA

3igneouscrystallization/deform

ation

BS04-1

N/A

pegmatitic

granitic

dike

z�3100

6CA


ation

BS04-2

N/A

pegmatitic

granitic

dike

z3104.6

±1.3

5CA


ation

Stolzburg

Complex

EKC03-3

Stolzburg

pluton

undeform

edTrondjemite

z3455.9

±0.5

0.1

6CA


a3263–3401(+�3700)

13

11

coolingbelow

�400�C

KPV99-96

Stolzburg

gneiss

deform

edbanded

Trondjemite

z3455.5

±0.6

0.6

4CA


a3092–3101

64

coolingbelow

�400�C

EKC03-11

N/A

deform

edgranitic

dike

z3212.5

±0.8

0.8

3CA


a3082–3100

54

coolingbelow

�400�C

EKC03-9

Nelshoogte

pluton

foliated

tonalite

z3236.0

±0.5

1.0

4CA


a3201–3225

64

coolingbelow

�400�C

KPV99-94

Nelshoogte

pluton

foliated

tonalite

z3236.3

±0.4

0.1

4CA


a>3166

22

coolingbelow

�400�C

aBrief

sample

descriptionsgiven

intext.See

TextS2formore

substantial

descriptions,locations,mineral

characteristics,andinterpretations.

bN/A,notapplicable.

cMineral

dated:z,

zircon;a,

apatite.

dMostsignificantdate(s)

from

thissample.Alldates

are

207Pb/206Pbdates.

eMSWD,meansquareofweighteddeviates[York,1969],forweightedmeandatecalculations.

f Number

ofsingle-grain

analysesdone.

AA,airabraded;CA,chem

ical

abraded.See

TextS1foranalyticaltechniques.

gNumber

ofanalysesusedforageinterpretationgiven

infootnote

d,either

asaweightedmeandate(w

iththeMSWD)adatefrom

asingle

analysis,orthenumber

ofapatitedates

considered

significant.

Reasonsforexclusionmay

includediscordance,inheritance,low

radiogenic/commonPbratios,etc.

See

auxiliary

materialfordiscussion.


6 of 27

TC5010

date of 3140.3 ± 0.3 Ma [Schoene and Bowring, 2007]. TheNW margin of the inlier is juxtaposed against apparentlylower-grade metasediment of the BGB, though this contactis concealed, and outcrop of the metasediment is poor nearthe contact. Where observed, this sediment is low-gradequartzite/sandstone ± highly altered micas. The PSZ is heredefined as the 1–1.5 km wide band of NW dippingmylonitic metaigneous rocks within in the inlier that paral-lels the contact with the BGB and runs NE–SW (Figure 3).[16] Rocks outside the PSZ (SE of the dashed gray line in

Figure 3) preserve evidence for multiple episodes of super-

imposed ductile deformation and igneous intrusion. Theseearly deformations are represented by a gneissose foliationand short-wavelength (<1 m) folding defined by lithologicbanding on all scales, as well as flattened mineral aggre-gates of quartz and feldspar and aligned biotite. Tight toisoclinal folds are in places transposed in moderatelydeveloped doubly plunging and/or sheath folds that definean irregular fold-hinge pattern on a stereographic projection(Figure 3c). Tight folds are often but not always axial planarwith the average local foliation. These structures are well-exposed in road-cuts within the bimodal gneiss near ‘‘the

Figure 3. (a) Geologic map of the Pigg’s Peak inlier and the Phophonyane Shear Zone (PSZ). Samplelocations from this study and Schoene and Bowring [2007] (AGC01-4, BS04-5) with 207Pb/206Pb ages(in Ma) are shown; outcrop photos and concordia plots for U-Pb data are given in Figures 4 and 5. Thickgray dashed line represents the SE transitional boundary of the PSZ. (b) Stereonet with structural datafrom within the PSZ (within �1.5 km from inferred lithologic contact). (c) Stereonet for structural datafrom outside the PSZ. See text for discussion. (d) Structural cross section, no vertical exaggeration. Finelines within unit patterns indicate approximate orientation of foliation.


7 of 27

TC5010

falls’’ on Figure 3. Gneissose foliation is folded on largerscales (tens to hundreds of meters) throughout the field area.Because of the varying scales of banding in this unit (cm to100 m), it is represented in Figure 3 mostly by undifferen-tiated banded gneiss. In places within the PSZ, however, thebanding is thick enough to be mapped as separate litholo-gies, such as the porphyritic tonalitic gneiss in Figure 3(e.g., sample AGC01-5; Figure 4). A fine-grained granite(sample EKC02-20; Figure 4) intruded after at least part ofthe foliation and folding in the banded gneiss had devel-oped, and then subsequently developed a foliation. Thisfine-grained granitic gneiss is in turn cut by undifferentiatedgranitic, granodioritic, and monzogranitic rocks which havepoorly to moderately developed foliation defined bystretched quartz and feldspar and aligned mafic minerals.These three lithologic packages (banded gneiss, graniticgneiss, and undifferentiated younger (meta)igneous rocks)are all cut by a series of thin quartz-feldspar pegmatitic andaplitic dikes that are variably sheared within the PSZ andlargely undeformed outside the PSZ.

[17] The SE margin of the PSZ, demarcated by the thickdashed gray line in Figure 3, is transitional over a distanceof �100–200 m. This transition is characterized by thegradual rotation of foliation toward NE–SW striking, NWdipping foliation, which is best observed in the NE portionof the field area. Foliation within the PSZ, though stilldefined by lithologic banding and aligned or flattenedminerals, is more finely banded and mylonitic and moreconsistent on all scales compared to outside the PSZ.Ubiquitous NE trending shallowly dipping (<20�) quartzand feldspar stretching lineations in nearly all lithologies arepresent (Figure 3b). Isoclinal folds are axial planar to thefoliation defined by lithologic banding and flattened oraligned minerals. Fold hinges within the PSZ are subparallelto stretching lineations (Figures 3b and 4). Therefore, thefinite strain ellipse from deformation within the PSZ isinterpreted to be uniform within the PSZ, with the X-Y axes(X > Y > Z) within the plane of foliation and the x axisparallel to stretching lineations and fold axes. Within the X-Zand X-Y plane, mineral elongation tails around most rigid

Figure 4. Field photos and concordia plots for the Ancient Gneiss Complex (AGC) in the Phophonyaneshear zone. Photos are of typical mafic-silicic banded gneiss (mapped as undifferentiated banded gneissin Figure 3). AGC01-5 represents a more homogeneous, tonalitic unit of the AGC than pictured, andEKC02-20 is the fine-grained granitic gneiss (not pictured) crosscutting the banded AGC. Concordia plotfor AGC01-5 contains data from both air-abraded (shaded circles) and chemical-abraded (open circles)zircons. Note Brunton compass for scale in both photos. Sample locations are shown in Figure 3.


8 of 27

TC5010

objects such as feldspar megacrysts and boudins are sym-metric, though some examples within the X-Z plane showsigmoidal shape consistent with dextral shear. In the X-Yplane, mantled feldspars are symmetric and show elongationparallel to stretching lineations (parallel to the x axis).[18] The PSZ is cut by variably sheared syntectonic

pegmatitic granitic and aplitic dikes. These dikes are iden-tical in lithology to those described outside the PSZ.However, while these dikes remain undeformed outsidethe PSZ, within the shear zone, many of them containfoliation and quartz and feldspar stretching lineations thatare parallel to those in the host mylonites that they crosscut.Thus, they intruded after the majority of foliation develop-ment within the host rocks but before complete terminationof deformation. Many of the hundreds of dikes present areoffset along foliation planes and are well exposed in threedimensions. Those observed in the X-Z plane (perpendicu-lar to foliation, parallel to lineations) record dextral move-ment and, where offset occurred, minerals lineations infoliation planes are parallel to mineral stretching lineationsin the host rock (Figure 5). When observed in the X-Yplane, the dikes show undulating shear folding in bothdirections. Two of these dikes were sampled for geochro-nology (EKC02-8 and EKC02-10; Figure 5 and Table 1).

[19] The consistency in the orientations of foliation,lineations, and fold hinges within the PSZ suggests thatshearing in these rocks overprinted all earlier phases ofdeformation recorded in the SE portion of the inlier outsideof the PSZ. Therefore, these small- and large-scale struc-tures can be used to interpret the kinematics of movementwithin the PSZ without deconvolving the earlier history indetail, and this is done in section 4.1.

3.2. Southern Margin: Steynsdorp Antiform

3.2.1. Previous Work[20] The Steynsdorp antiform, along the southern margin

of the BGB (Figures 2 and 6), consists of the dome-shapedtonalitic-trondhjemitic-granodioritic Steynsdorp gneiss (pre-viously called the Steynsdorp pluton; we use the termSteynsdorp dome here to refer to its geometry [Kistersand Anhaeusser, 1995b]) and the structurally overlyinglowermost units of the BGB: the Sandspruit (arguablypresent), Theespruit, Komati, and Hooggenoeg formations[Anhaeusser, 1983; Kroner et al., 1996; Viljoen and Viljoen,1969] (Figure 6). Amphibolite lenses within the Steynsdorpgneiss are the only indication of its metamorphic grade,mafic units of the Theespruit formation have both amphib-olite and actinolite-talc assemblages [Kisters and

Figure 5. Field photos and concordia plot for syntectonic dikes from within the Phophonyane shearzone, Pigg’s Peak inlier. Field photos show dikes crosscutting foliation of banded orthogneiss of theAGC; viewed within �35� of the X-Z plane of the finite strain ellipse. The dikes contain weak foliationparallel to the host rocks and are sheared, indicating right-lateral dip-slip offset. EKC02-8 crosscutsEKC02-10; the older dike has strong stretching lineations parallel to those in the host rock. Fieldnotebook is for scale in both photos. Sample locations are shown in Figure 3.


9 of 27

TC5010

Anhaeusser, 1995b], whereas the Komati and overlyingHooggenoeg formations are likely greenschist facies, withcommonly observed actinolite-chlorite and talc schists[Viljoen and Viljoen, 1969; Kroner et al., 1996]. No quan-titative thermobarometry has yet been published on theserocks. The southern margin of the gneiss is intruded by the

Mpuluzi granite, which has been dated by U-Pb ID-TIMSon zircon and sphene at 3107 +4/�2 Ma [Kamo and Davis,1994]. The igneous protolith of the Steynsdorp gneiss hasbeen dated by U-Pb in zircon both by ID-TIMS, giving anupper intercept of 3509 +8/�7 Ma [Kamo and Davis, 1994],and by SHRIMP, giving an upper intercept at 3509 ± 4 Ma

Figure 6. (a) Geologic map of the Steynsdorp antiform, compiled from this study, Anhaeusser [1983],Kroner et al. [1996], and Kisters and Anhaeusser [1995b]. Structural data are from this study only; seeKisters and Anhaeusser [1995b] for further structural detail. Structural symbols are the same as those inFigure 3, except that here, strike and dip symbols represent foliation defined by compositional bandingand aligned planar minerals of undifferentiated origin; thick dark lines represent faults. Sample locationsare shown with 207Pb/206Pb ages (in Ma; z, zircon; a, apatite; s, sphene). Concordia plots for U-Pb dataare in Figure 7. Ages for the Dalmein pluton and the Mpuluzi batholith are from Kamo and Davis [1994].(b) Structural data from this study only plotted on a stereonet. (c) Schematic structural cross section fromX-X0. The zone of maximum strain (gradational but hundreds of meters wide), the Steynsdorp shear zone(SSZ), is shown in the vicinity of the contact between the Steynsdorp gneiss and the TheespruitFormation. Below, rock names representing degree of strain are given.


10 of 27

TC5010

Figure 7


11 of 27

TC5010

[Kroner et al., 1996]. Zircons from a felsic mylonitic tuffwithin the Theespruit Formation were dated by U-PbSHRIMP, and these analyses form a discordia with an upperintercept of 3544 ± 3 Ma [Kroner et al., 1996]. A sample ofthe Steynsdorp gneiss was also collected for ID-TIMS datingin this study (EKC02-40; Figure 7).[21] The structural history of the Steynsdorp antiform

was studied in detail by Kisters and Anhaeusser [1995b].Contacts between the Steynsdorp gneiss and the TheespruitFormation are sharp, mylonitic and clearly faulted. The keystructural observations in that study are (1) foliation in theSteynsdorp gneiss is defined by compositional banding andplanar minerals such as biotite, and this fabric representssolid-state deformation; (2) foliation in both the Steynsdorpgneiss and the overlying BGB units are parallel to thecontact between them (i.e., hemispherical), and this foliationbecomes weaker toward the contact with the KomatiFormation, interpreted as a fault; (3) lineations in both theSteynsdorp gneiss and the BGB units, defined by elongatedamphiboles or stretched quartz and feldspar trend uniformlyto the NE throughout the antiform; (4) lineations are alsosubparallel to fold axes of mesoscale isoclinal folds (wave-lengths of several centimeters to tens of meters) defined bylithologic banding within the Theespruit Formation andSteynsdorp gneiss; and (5) a vertical crenulation cleavage,which is axial planar to the Steynsdorp dome, is present inthe Theespruit formation. Kisters and Anhaeusser [1995b]use these observations to suggest the Steynsdorp antiformwas the result of subhorizontal E–W shortening super-imposed upon a buoyantly rising Steynsdorp dome.[22] The timing of the formation of the Steynsdorp anti-

form is constrained by the intrusion of the 3216 +2/�1 MaDalmein pluton [Kamo and Davis, 1994], located immedi-ately to the west. However, 207Pb/206Pb zircon evaporationdates of the undeformed Vlakplaats granodiorite, whichcrosscuts the Komati Formation and its foliation withinthe antiform (Figure 6), give a mean of 3450 ± 3 Ma, whichwas interpreted to place a minimum age on development ofthe antiform [Kroner et al., 1996]. We resampled thisintrusion for the present study (WKC00-88; Figure 7).3.2.2. Mapping and Structural Analysis[23] Field mapping and sample collection focused mainly

on the Steynsdorp gneiss and the contact between it and theTheespruit Formation. A map with lithologic contacts basedprimarily on Kroner et al. [1996], with additional lithologicdetail and new structural data from this study, is presented inFigure 6. Overall, our observations agree well with thosereported in previous studies [Kisters and Anhaeusser,1995b; Kroner et al., 1996; Viljoen and Viljoen, 1969],

though a few points not reported or emphasized in thosestudies will receive focus here, in that they lead us to aricher kinematic interpretation.[24] First, contact-parallel, concentric foliation in the dome

becomes less evident toward its center, where foliation isweak to nonexistent. Lineations defined by stretched quartz,feldspar and amphibole are still present in the core of thegneiss dome (Figure 6). There is a sharp decrease in theamount of strain across the contact between the Theespruitand Komaati formations, and undeformed semicircular ocelliand pillow lavas are preserved in the Hooggenoeg Formation.[25] Second, an important characteristic of the sheared

contact between the Steynsdorp gneiss and the overlyingTheespruit Formation is that asymmetric kinematic indicatorswithin tens of meters of the contact show evidence fornoncoaxial shear. This consists of sigmoidal garnet porphyr-oblasts in the Theespruit Formation within several meters ofthe contact, �15 examples of sigmoidal feldspar megacrystsin the orthogneiss (symmetric mantled porphyroclasts aredominant), and sheared and offset granitic dikes (Figure 7).[26] The direction of offset for noncoaxial shear can be

determined by these asymmetric features. Along the westernmargin of the Steynsdorp gneiss, mantled garnet and feld-spar porphyroclasts give dextral offset when observed in aplane perpendicular to foliation and parallel to lineation(Figures 7a and 7b). Dextral shears in offset dikes along thismargin are consistent with this interpretation, and indicatethat simple shear was important locally (Figure 7c). Similarto the late dikes described in the PSZ, these crosscut thefoliation but contain a similar lineation to that of the hostgneiss and in places are offset along foliation planes,sometimes brittley (samples EKC02-38, EKC02-46, andEKC02-47). The northern margin of the Steynsdorp gneisscontains asymmetric mantled feldspar megacrysts that indi-cate a dominantly normal sense of motion (top down to theNE) when observed in a plane perpendicular to foliation andparallel to lineations. No macroscopic indications of non-coaxial shear are evident in the center of the Steynsdorpgneiss dome, though the rock is strongly lineated. Thecontact on the eastern margin of the pluton is not as wellexposed, though lineations are parallel to those throughoutthe antiform. Along the southern margin of the Steynsdorppluton, the crosscutting Mpuluzi batholith is well exposed.This portion of the batholith is fine-grained and in placescontains shallow dipping foliation defined by aligned biotite(Figure 6). Some localities contain strongly to moderatelydeveloped quartz stretching lineations that are subparallel tothe lineations observed in the Steynsdorp antiform.

Figure 7. Field photos and concordia plots for U-Pb apatite and zircon data from the Steynsdorp antiform. See Figure 6for sample localities. (a) Mantled garnet porphyroblasts from the western margin of the Steynsdorp dome indicating right-lateral shear in subvertical foliation. Viewed in the X-Z plane of the finite strain ellipse. Photo was taken within theTheespruit Formation several meters from the contact with the Steynsdorp gneiss (near sample EKC02-46; Figure 6). Tip ofpencil is for scale. (b) Example of mantled feldspar porphyroclast from the Steynsdorp gneiss, with dextral shear senseindicated. Viewed in the X-Z plane of the finite strain ellipse. Photo was taken on western margin of the dome, within 10 mfrom contact with Theespruit Formation. (c) Typical granitic dike on the western margin of the dome, crosscutting foliation,but sheared and folded, with NE trending fold axes and stretching lineations.


12 of 27

TC5010

3.3. Northwest Margin: Honeybird Shear Zone

3.3.1. Previous Work[27] A series of orthogneiss and amphibolite exposures

located along the Kaap river NE of Barberton clearly exposea shear zone, here called the Honeybird Shear Zone (HSZ).The HSZ separates quartzites and metaconglomerates of theMoodies Group from several generations of orthgneisses,including a gneissose portion of the Stentor pluton. Severalstudies detail the sedimentology and structural geology ofthis portion of the Moodies Group and nearby localities[Anhaeusser, 1972, 1976; de Ronde and de Wit, 1994; Frippet al., 1980; Gay, 1969; Heubeck and Lowe, 1994b; Loweand Byerly, 1999a]. Fripp et al. [1980] interpreted thecontact to be a ductile, south dipping thrust fault. However,de Ronde and de Wit [1994] noted that the footwall of thissouth dipping shear zone is part of the Stentor pluton, whichwas dated at 3107 ± 5 Ma �10 km away from the Honey-bird locality [Kamo and Davis, 1994], and suggest that thefabrics in the HSZ record late transtension in the area.

Dziggel et al. [2006] report quantitative P-T data as well asstructural data and interpretations for the rocks in and nearthe HSZ. They find there is a transition, at least in someplaces sharp, from amphibolite to greenschist grade rocks ator near the roughly E–W trending contact between orthog-neiss and schists of the Onverwacht group. Shear senseindicators suggest orthogneiss-up, paragneiss-down move-ment along high-strain zones located near these contacts[Dziggel et al., 2006]. Although that study provides nogeochronological constraints on the timing of metmorphismor deformation, they infer that both were related to circa3230 Ma terrane collision and orogen parallel gravitationalcollapse [e.g., Kisters et al., 2003]. Our purpose in studyingthe area was to work out the relative sequence of intrusionsand to help resolve the kinematics and timing of movementon the HSZ.3.3.2. Field Descriptions of the HSZ[28] Our investigation of the HSZ focused on rocks that

are clearly exposed along the Kaap river, off R38 NE ofBarberton, on the dirt road to the old Honeybird rail depot

Figure 8. Concordia plots and structural data from the Honeybird Shear Zone (HSZ). See Figure 2 andauxiliary material for location and text for discussion.


13 of 27

TC5010

(called the ‘‘Honeybird siding’’ by Dziggel et al. [2006,Figure 2]). The footwall of the Honeybird Shear Zone iscomposed of two main rock types. The older is a bandedgneiss with alternating tonalitic to granodioritic lithologiesand also sparse amphibolitic layers (sample KPV99-90;Figure 8). The gneissosity in this unit is defined bycompositional banding, flattened quartz and feldspar, andthe planar alignment of biotite. This unit is cut by a morehomogeneous orthogneiss of granodioritic to granitic com-position. About 1 km north of the road crossing the river,only the homogeneous granitic gneiss is present, and itstrikes �E–W and dips 30–60� to the south (sampleKPV99-89; Figure 8 and Table 1). This granitic gneiss iscut by late pegmatitic dikes. Farther south (closer to theMoodies contact), the older banded gneiss is present andboth it and the granitic gneiss develop a gneissic foliationwith well-developed quartz and feldspar stretching linea-tions that trend and plunge to the SW (Figure 8). Inaddition, several generations of pegmatitic dikes crosscutthe mylonitic foliation. The older generations are variablysheared into parallelism with the host rock and have parallellineations as well. Younger generations of dikes, thatcrosscut the foliation are foliated yet unlineated, and areoffset semiductilely (samples BS04-1 and BS04-2; Figure 8).Many of these dikes are offset with a normal shear sense(top to the south) along foliation planes, though quality 3Dviewing is difficult in this area, so this remains a tentative

conclusion. The contact with the Moodies Group in thehanging wall of the HSZ is several hundred meters to thesouth, but is not exposed. Though published maps ofthe area [Anhaeusser, 1972; Dziggel et al., 2006] showOverwacht Group rocks between orthogneiss and MoodiesGroups in this area, these units are covered and/ornonexistent in the observed section. Lower-grade Moodiesconglomerates are exposed �1 km to the SW in the EurekaSyncline.

3.4. Southwest Margin: Stolzburg Complexand Nelshoogte Pluton

[29] Most of the recent published work on the granitoidsflanking the BGB has been focused along the SW margin ofthe belt, particularly within the Stolzburg complex andsurroundings (see section 2; Figure 9). Kamo and Davis[1994] reported an age of 3460 +5/�4 Ma for the oldestphase of the Stolzburg complex (from the same locality asKPV99-96), and also dated two near-concordant zirconswith 207Pb/206Pb dates of circa 3237 and 3255 Ma in ayounger foliated phase. One concordant sphene analysisfrom that sample gave a 207Pb/206Pb date of circa 3201 Ma.Kisters et al. [2003] suggest exhumation of these rocksoccurred along a midcrustal detachment that runs within theBGB directly north of the northern margin of the Stolzburgcomplex (Figure 9), thereby juxtaposing those rocks against

Figure 9. Geologic map of the Stolzburg area. See Figure 2 for location. Sample locations and207Pb/206Pb dates are shown (in Ma; z, zircon; a, apatite; s, sphene), and concordia plots for U-Pb dataare in Figures 10 and 11. Data from the literature: one asterisk, Kamo and Davis [1994]; two asterisks,Diener et al. [2005]; and three asterisks, Dziggel et al. [2005]; Ar/Ar data from Lopez-Martinez et al.[1992]. See text for additional data from the literature. Thick, light gray normal fault indicates the Komatifault, suggested as a circa 3.22 Ga midcrustal detachment by Kisters et al. [2003] and Diener et al.[2005]. Map compiled from Anhaeusser [1983], de Wit [1982], Kisters and Anhaeusser [1995a], andWestraat et al. [2005].


14 of 27

TC5010

the lower-grade rocks of the greenstone belt. Our samplingstrategy in this area aimed to place constraints on the<600�C exhumation history of the footwall rocks of theproposed detachment, in order to help determine the timingof juxtaposition of the BGB with the higher-grade rocks tothe south.[30] We collected three sets of samples for zircon and

apatite geo/thermochronology. EKC03-3 was collected fromthe NE margin of the Stolzburg complex where undeformedrocks of the Stolzburg gneiss intrude into amphibolites thatare in contact with garnet-sericite schists of the TheespruitFormation (Figure 10). The other samples were collectedfrom the central Stolzburg Complex where highly deformedbanded gneiss (KPV99-96; Figure 11 and Table 1) containsgreenstone xenoliths, both of which are cut by a set ofdeformed dikes (EKC03-11; Figure 11 and Table 1). Finally,samples EKC03-9 and KPV99-94 were collected from thecirca 3236 Ma Nelshoogte pluton to confirm its age andconstrain its thermal history (Figure 10 and Table 1).

4. Kinematic and Geochronologic

Interpretations

[31] This study has focused on four well-exposed areasthat separate the low-grade supracrustal rocks of theBarberton greenstone belt (BGB) from the generallyhigher-grade orthogneisses that surround the belt. Thissection provides kinematic interpretations based on thestructural data from this and other studies, with the timingof movement in the examined shear zones provided by ournew U-Pb geochronology (summarized in Table 1 andFigure 12). This is coupled with evidence for differentialexhumation given new and published U-Pb and 40Ar/39Arthermochronology.

4.1. Pigg’s Peak Inlier

[32] The Pigg’s Peak inlier in NWSwaziland (Figures 2–5;section 3.1) records complicated early intrusion and defor-

mation at circa 3.66–3.23 Ga within the AGC. The oldestunit sampled in this study, a porphyritic tonalite gneiss(Figure 4), also gives the oldest age in the Kaapvaal craton,with a 207Pb/206Pb date of 3662.8 ± 0.5 Ma. The fine-grained granitic gneiss that cuts the porphyritic gneiss andthe bimodal gneisses (Figure 3) was originally mapped asPigg’s Peak granite [Wilson, 1979, 1982]. It is in factdistinct texturally from the circa 3140 Ma Pigg’s Peakgranite [Schoene and Bowring, 2007] and is also consider-ably older, dated at 3545.6 ± 0.5 Ma (Figure 4). These datesfrom the AGC are consistent with zircon ion-microprobedates reported by Compston and Kroner [1988] and Kroneret al. [1989], though our samples show a less heterogeneousspread in dates that may be due to actual differences inzircon growth, a bias introduced by the chemical abrasionmethod (e.g., which may have preferentially dissolvedyounger high-U metamorphic rims), or that Pb loss in theprevious studies was masked by the relatively large errorsassociated with ion microprobe U-Pb analysis.[33] Evidence for deformation before circa 3.23 Ga in the

AGCwaserasedintheNEstriking,NWdippingPhonphonyaneShear Zone (PSZ). This zone separates the amphibolite-grade basement orthogneiss of the Ancient Gneiss Complex(AGC) from what are likely subgreenschist grade sandstoneand siltstone of the BGB to the NW. The most prominentfeatures in the PSZ include the consistency of the NEstriking, NW dipping foliation and gneissose layering andthe NE trending and plunging linear features, such asstretching lineations and fold hinges (Figure 3, section 3.1).The finite strain ellipse is oriented with the X-Y axes withinthe plane of foliation and the X direction parallel tostretching lineations. The X-Z plane of the finite strainellipse is often inferred to also indicate the direction ofmovement within the shear zone (i.e., parallel to lineations).Several studies suggest that this assumption can be mislead-ing, in that natural examples and numerical modeling oftranspressional and transtensional systems show that the X-Zplane can be perpendicular to the transport direction depend-ing on the relative magnitude of pure and simple shear [Lin

Figure 10. Concordia diagrams from the northern margin of the Stolzburg complex and the Nelshoogtepluton. See Figure 9 and auxiliary material for sample localities.


15 of 27

TC5010

and Jiang, 2001; Lin et al., 2007; Passchier, 1998; Sullivanand Law, 2007; Tikoff, 1997]. The type of shear in the PSZwas undoubtedly noncoaxial and heterogeneous on a varietyof scales given the rheological differences resulting fromdifferent lithologies (e.g., zones of boudinage), as well asthe rotation of a complex nonuniaxial fold hinge distributionrecorded outside the PSZ into the subparallel hinge patternsobserved within the PSZ (Figure 3). These rotated featuresand the mylonitic foliation with uniform lineation patternsdeveloped in the PSZ suggest that finite strain was high; therarity of asymmetric shear sense indicators suggest thatfinite strain was primarily that of flattening. The shearsense indicators that were observed, however, indicate

right-lateral shear parallel to stretching lineations and foldaxes (section 3.1). A glimpse into the incremental shearhistory of the PSZ is given by late synkinematic dikes thatare offset dextrally subparallel to lineations, suggesting thatthe paucity of asymmetric features in the host rock was aresult of large noncoaxial strain. Furthermore, Lin et al.[2007] argue that lineations in planes that approach simpleshear strain are akin to slickenslide formation in brittleregimes and clearly indicate transport direction. Thus,lineations on foliation planes that clearly offset dikes(Figure 5) provide strong evidence that NE–SW right-lateral shear dominated the PSZ. Therefore, the kinematicsof the PSZ were right-lateral strike-slip motion with a

Figure 11. Field photos and concordia plots for samples from the central Stolzburg complex. See Figure 9and auxiliary material for sample localities. (a) Sample location of orthogneiss KPV99-96. This sample wasthe oldest generation from this outcrop. Note field book for scale. (b) Sample location of late syntectonicdike EKC03-11. Dark rock is a large greenstone xenolith. Aplitic dikes intrude parallel and discordant tofoliation and have foliation parallel to the host rock. Hammer and tape roll are for scale.


16 of 27

TC5010

component of dip-slip extensional shearing (i.e., transten-sion). Such a normal displacement in the PSZ is consistentwith the abrupt change in metamorphic gradient (highergrade in the SE) across the NW dipping structure.[34] Timing of NE directed transtension in the PSZ is

constrained locally by the crystallization age of a latesyntectonic dike (sample EKC02-8) of 3223.4 ± 1.9 Ma(Figure 5). As mentioned above, these dikes record the late

history of the shear zone, but have lineation patternsidentical to other rocks in the PSZ, suggesting their crys-tallization constrains the termination of a longer sequence ofright-lateral shear in the PSZ. This timing for deformationcontrasts the results of Schoene and Bowring [2007] basedon U-Pb sphene and apatite thermochronology, which showthat a period of rapid cooling and 7–15 km of exhumationof the AGC coincided with the intrusion of the Pigg’s Peak

Figure 12. Summary diagram of structural, geochronological, and thermochronological data for thekinematics, distribution, and timing of important tectonothermal events in the BGB between circa 3.45and 3.10 Ga. BGB, Barberton greenstone belt; PSZ, Phophonyane Shear Zone; SSZ, Steynsdorp ShearZone; HSZ, Honeybird Shear Zone. Numbers in parentheses represent data from the literature: 1, Lopez-Martinez et al. [1992]; 2, Kamo and Davis [1994]; 3, de Ronde et al. [1991]; 4, Layer et al. [1998]; 5,Layer et al. [1992]; 6, de Ronde and Kamo [2000]; 7, Westraat et al. [2005]; 8, Schoene and Bowring[2007]; all other data from this study.


17 of 27

TC5010

granite at circa 3140 Ma (see section 2). Therefore, our datarequire that either the zircons from EKC02-8 are inheritedor that the younger period of exhumation was not due topenetrative shear in the PSZ. If the latter is true, then offsetwas accommodated along a reactivated portion of the faultfarther NW than sample EKC02-8 (Figure 3). Given that theanalyzed zircons showed no resorption and there was noevidence for inheritance (e.g., >3.2 Ga zircons of the hostrock), the zircons likely crystallized in the dike. Thegeochronologic and thermochronologic data sets thereforeimply a complicated history of movement along this struc-tural boundary that involved movement circa 3.22 Ga attemperatures in excess of 600�C, following by reactiva-tional exhumation at circa 3.14 Ga to below 400�C in themiddle crust. If it is safe to assume that the contact betweenthe BGB on the north and AGC on the south parallels thefoliation patterns, then the overall offset between the tworequires a large component of normal faulting over time.Detailed thermobarometry is needed to further quantifythese first-order P-T-t estimates. Such work would helpverify if, for example, higher-grade metasediments (e.g.,amphibolite grade) exist in contact with the orthogneisses,as has been well-documented in the Stolzburg Complex[Kisters et al., 2003; Diener et al., 2005] and is likely true inthe Steynsdorp Complex. This would thereby increase thelateral distance over which offset could have occurredwithin the PSZ.

4.2. Steynsdorp Complex

[35] The structural observations described in section 3.2,when coupled with the analysis presented by Kisters andAnhaeusser [1995b], allow for interpretation of thekinematics of the formation of the Steynsdorp Complex.Kisters and Anhaeusser [1995b] interpret the complex tohave formed during NW–SE subhoriztonal compression:the rheological contrast between the buoyantly risingSteynsdorp Gneiss and the overlying metasedimentsresulted in foliation that parallels the concentric contactbetween the two lithologies. Because the amount of strainalso decreases away from this contact, we term the zonefrom the Theespruit-Komati contact to slightly below theTheespruit-Steynsdorp contact the Steynsdorp Shear Zone(SSZ; Figure 6).[36] An aspect not emphasized by Kisters and

Anhaeusser [1995b] is the importance of offset along theSSZ. Several observations, however, indicate that signifi-cant exhumation occurred along this contact. First is theapparent contrast in metamorphic grade of mafic rocksbetween the Steynsdorp Gneiss (containing amphibolites),the overlying Theespruit formation (talc-chlorite-actinoliteschists), and the Komati formation (highly altered, but low-grade ultramafic volcanics). Quantitative thermobarometrywith geochronology, however, is necessary to calculate anabsolute amount of vertical displacement as a function ofspace. Second, asymmetric shear indicators within the SSZgive consistent hanging wall to the NE shear sense (seesection 3.2; Figure 7). This suggests that in fact along thewestern margin of the SSZ, the faulting is predominantlydextral strike slip, whereas on the northern margin it is

dominated by normal sense shear. The direction of transportimplied by these features is parallel to the regional NE–SWstretching lineations in all lithologies (and notably notconcentric, as would be expected by simple diapiric riseof the gneiss dome). Thus the Z axis of the finite strainellipse trends uniformly to the NE, though its plunge rotates,as does the trend and plunge of the x axis. This implies thatlocal stresses dictated the resulting strain along the PSZ,rather than the regional NW–SE compression that isimplied by crenulation cleavages in the metasediments aswell as the ‘‘hinge collapse’’ of the anticline in the NEsection of the field area [Kisters and Anhaeusser, 1995b].The faulted contact between the Theespruit and Komatiformations also corresponds with a significant lowering inthe amount of strain and potentially the metamorphic gradeof the rocks to the north. This implies that this contact mayhave contributed significantly to the overall offset.[37] The kinematics of the SSZ are consistent with the

interpretation that NE–SW extension along the SSZ wasimportant during both the formation of the lineation andfoliation patterns and also its vertical, domal exhumation,not unlike those described in metamorphic core complexesin younger orogens [e.g., Davis and Lister, 1988; Lister etal., 1984; Saltzer and Hodges, 1988; Whitney et al., 2004].These systems are common where orogen-parallel extensionand dome formation locally dominate the kinematics ofdeformation. In some core complex systems, horizontalcontractional features perpendicular to the extension direc-tion are also observed (i.e., y axis contraction [Fletcher andBartley, 1994]), not unlike the NE–SW striking, verticallydipping crenulation cleavage in the Theespruit Formation[Kisters and Anhaeusser, 1995b]. Modeling the Steynsdorpantiform as a core complex is convenient in that all folia-tions and lineations observed can be generated simulta-neously. However, such a model implies a geneticrelationship between extension and dome formation, whichis difficult to establish in this area based on the fieldrelationships alone. Therefore, geochronology and thermo-chronology are used to provide further constraints.[38] The timing of formation of the predominant foliation

and lineation patterns in the Steynsdorp antiform is in partbracketed by the intrusion of the undeformed Vlakplaatsgranodiorite (sample WKC00-88; Figure 7), dated here at3229.8 ± 0.6 Ma, which crosscuts the fabrics of the KomatiFormation. This is over 200 Ma younger than the previousestimate for the age of this granodiorite (circa 3450 Ma[Kroner et al., 1996]), and therefore drastically lowers theage limit on the timing of formation of the Steynsdorpantiform to one that is more consistent with existing tectonicmodels for BGB. However, a series of syntectonic dikeswith similar kinematic indicators to the host SSZ (Figure 7;samples EKC02-38, EKC02-46, and EKC02-47) intrudedthe SSZ circa 3104 Ma. This suggests that while the majorgeometry of the antiform formed circa 3230 Ma, the SSZwas locally reactivated with similar kinematic sense duringintrusion of the dikes. The dikes are, in places, also brittelyoffset at the contact between the Steynsdorp gneiss andTheespruit Formation, suggesting that brittle offset alongthis margin was important at lower temperatures as well.


18 of 27

TC5010

Moderate foliation and locally well-developed lineationswithin the circa 3107 Ma Mpuluzi batholith at the southerncontact with the Steynsdorp pluton are consistent with itsintrusion into a tectonically active setting.[39] Vertical exhumation of the Steynsdorp gneiss (i.e.,

dome formation) is constrained in part by U-Pb apatitethermochronology. U-Pb apatite dates from the Vlakplaatsgranodiorite in the hanging wall of the SSZ are 3200–3230 Ma whereas those from the footwall in the Steynsdorpgneiss (sample EKC02-40; Figure 7) are circa 3160–3185 Ma. Younger apatite dates from the Steynsdorp gneissare therefore either a result of (1) continued localizedextensional exhumation along the SSZ, (2) slow thermalequilibration following exhumation, or (3) from resetting bythe Mpuluzi batholith, whose contact is �1.5 km from thesample locality. However, because the igneous protolith forthe Steynsdorp gneiss is circa 3518 Ma, it is clear that circa3.23 Ga reheating and/or exhumation occurred within theSteynsdorp gneiss at that time, and that dome formationlikely occurred at temperatures above 400 – 500�C.Therefore, all the above structural geochronological andthermochronological data are consistent with generation ofthe Steynsdorp dome during orogen parallel extension alongthe SSZ, and therefore supportive of a midcrustal corecomplex model for its formation at or immediately priorto circa 3.230 Ga.

4.3. Southwest Margin of the BGB

[40] The SW margin between the BGB and the Stolzburgcomplex exposes the highest-grade rocks near the BGB yetdescribed [Diener et al., 2005; Dziggel et al., 2002; Kisterset al., 2003; Stevens et al., 2002; Moyen et al., 2006]. Oursamples of the Stolzburg complex record very differentthermal and deformational histories. One sample from theinterior of the complex (sample KPV99-96) is of highlydeformed gneiss whereas that from the NE margin of thecomplex (sample EKC03-3) is undeformed and preservesintrusive contacts into the base of the Theespruit Formationof the BGB (Figure 9). Both samples give identical crys-tallization ages of 3455.5 ± 0.6 and 3455.9 ± 0.5 Ma for thedeformed and undeformed samples, respectively (Figures 9,10, and 11). Ductile deformation of KPV99-96 in the centralportion of the Stolzburg complex largely predates the age ofa crosscutting dike (sample EKC03-11, Figure 11), thatcrystallized at 3212.5 ± 0.8 Ma. The fact that the centralStolzburg complex is deformed whereas the northernboundary of the pluton is not suggests that either there isa significant structural break between these two locales orthat they responded differently to circa 3.2 Ga regionaldeformation. This question is further clarified by the U-Pbapatite thermochronology, as follows.[41] U/Pb apatite dates in the undeformed northern mar-

gin of the Stolzburg complex (sample EKC03-3) are allolder than 3.27 Ga and should yield a minimum coolingtime for the rock through �400–500�C [Cherniak et al.,1991; Chamberlain and Bowring, 2001; Schoene andBowring, 2007]. Retrograde P-T conditions for metasedi-ments <100 m from the sample locality are 6.1 ± 2.7 kbarand 569 ± 42�C and peak metamorphic conditions in rocks

nearby are �7.5 kbar and 550�C [Diener et al., 2005].Taking our apatite data at face value suggests that peakmetamorphism occurred at >3.4 Ga and that the apatite datarecord cooling and/or cooling and low-T metamorphicovergrowth or Pb loss. However, U-Pb sphene data reportedbyDiener et al. [2005] give an upper intercept date of 3229 ±25 Ma (on 12 LA-ICPMS U-Pb analyses) that they suggestestimates peak metamorphism. It is difficult to reconcilethose conclusions with our data because apatite should becompletely reset at 3230 Ma under the reported metamor-phic conditions given the assumed Pb diffusion parameters.Application of the simple diffusion model of Schoene andBowring [2007] shows that for apatite dates to fall roughlybetween 3.40 and 3.26 Ga by resetting at 3.23 Ga, the rockwould have to be heated, for example, from <400 to 560�C(the lower end of peak temperatures, giving a maximumtime estimate) and cooled again in less than �1.5 Ma.Given the P-T path calculated for these rocks by Diener etal. [2005], this would imply a minimum burial-exhumationrate of�2.4 cm/a assuming instantaneous conductive heating/cooling in the host rocks. This would correspond to valuesamong the highest exhumation rates observed in modernorogenic belts [e.g., Burbank, 2002]. Considering that thesample in question here is undeformed, this seems unlikely.A simpler explanation is that the sphene dated by Dieneret al. [2005] grew below its closure temperature [e.g.,Aleinikoff et al., 2002; Frost et al., 2001] and that thisportion of the BGB actually maintains a signature of circa3.45 Ga metamorphism synchronous with intrusion of theStolzburg, Theespruit and Doornhoek plutons, as waspreviously inferred [de Wit et al., 1983, 1987b; Kistersand Anhaeusser, 1995a]. Metamorphism at circa 3.4 Ga isalso reflected by 40Ar/39Ar dates of metamorphic amphib-oles of >3.4 Ga from within the Komati Formation [LopezMartinez et al., 1992], structurally <2 km from the presentlocality (but on the hanging wall of the proposed detach-ment of Kisters et al. [2003]) (Figure 9). A sphene date ofcirca 3.45 Ga was reported from the Theespruit pluton(Figure 9) [Kamo and Davis, 1994], which is consistentwith it having resided well below the closure temperature ofsphene (�600�C) since its intrusion. However, other occur-rences of circa 3.2 Ga sphene have been reported as well,for example, from the circa 3.45 Ga Doornoek pluton[Kamo and Davis, 1994] and metamorphic rocks withinthe central Stolzburg complex [Dziggel et al., 2005]. Inaddition, several of our 207Pb/206Pb apatite cooling dates arealso much older than the age of the intrusion; only one ofthese is concordant and is circa 3.7 Ga. Apatite U/Pb datesthat retain pre-high-temperature histories have never beenreported and are unlikely given our understanding of Pbdiffusion in apatite. The data presented here in combinationwith data from the literature thus clearly show a that a betterunderstanding what controls the recorded dates of thermo-chronometers is necessary and that taking nominal closuretemperatures at face value may sometimes be misleading orambiguous.[42] U/Pb apatite dates from the central Stolzburg com-

plex (samples KPV99-96 and EKC03-11; Figure 11) ofcirca 3080–3100 Ma can either record circa 3.1 Ga resetting


19 of 27

TC5010

or exhumation and cooling of these rocks synchronous withgranitoid intrusion circa 3.1 Ga. Argument for the lattercomes from the fact that the circa 3107 Ma Boemanskopsyenite and Mpuluzi batholith are >5 km from the samplelocality (Figure 9), and are therefore unlikely to have resetthe observed circa 3.1 Ga apatite dates from initial 3.23 Gacooling. By comparison, the apatite from the Steynsdorpsamples (EKC02-40 and WKC00-88) are <2 km from theMpuluzi batholith and still largely retain the 3.2 Ga coolingsignature in the U-Pb systematics. Westraat et al. [2005]show that the circa 3.1 Ga Boesmanskop syenite and nearbyMpuluzi granite intruded in a transcurrent extensionalsetting, within an inferred regional-scale dextral transpres-sional system. A similar history may explain the exhuma-tion of the southern/central Stolzburg complex, whichwould reconcile the potentially disparate cooling historiesobserved in the central and northern portions of the com-plex. But this would require the existence of a major zone ofoffset within the (poorly exposed) interior of the Stolzburgcomplex rather than north of its contact with the Theespruitformation (i.e., the Komaati fault as suggested by Kisters etal. [2003]; Figure 9), though such a structure has not beenidentified in the field. Such a model may be oversimplified,however, given the preservation of (inferred circa 3.23 Ga)high-P, low-T metamorphism in the Stolzburg complex[Moyen et al., 2006], which is difficult to do if the rocksresided in the middle-lower crust for 100 Ma. Thus, thepotentially contradicting U-Pb apatite and sphene datesshow that either the Stolzburg complex had an extremelyspatially variable thermal history or that our knowledge ofthe kinetics of metamorphic reactions and volume diffusionin metamorphic rocks or thermochronometers is insuffi-cient. Further studies in this area involving a combinationof P-T and T-t data will help resolve these issues and betterassess the history of this important Archean terrane.

4.4. Northwest Margin of the BGB

[43] Along the northern margin of the BGB, the southdipping Honeybird Shear Zone (HSZ) separates amphibolitegrade orthogneiss in the footwall from schists of subgreens-chist grade in the hanging wall (Figure 2, also see map inthe work by Dziggel et al. [2006]). A banded gneiss in thefootwall (sample KPV99-90; Figure 8) crystallized at3258.3 ± 0.3 Ma, and is the oldest date of granitoid gneissimmediately north of the BGB, though xenoliths from theadjacent Nelspruit batholith gave U-Pb zircon dates of circa3.3 Ga [Kamo and Davis, 1994]. This gneiss sample may bethe plutonic equivalent of felsic tuffs from within theFig Tree Group, which contain zircons that give a weightedmean U-Pb date of 3258 ± 3 Ma [Byerly et al., 1996],implying a spatial proximity of the two at that time. Theexistence of these two rocks suggests that in fact thegeneration of continental crust and perhaps subductionnorth of the BGB began well before 3.23 Ga. The stronglyfoliated and lineated granitic gneiss of the Stentor pluton(sample KPV99-89; Figure 8) in this area intruded at 3106.0 ±0.5 Ma, and both the older and younger phases havestrongly developed NE–SW stretching lineations thatplunge shallowly to the SW (section 3.3; Figure 8). A late

synkinematic dike (sample BS04-2; Figure 8) from the HSZcontained a single noninherited concordant zircon with anage of 3104.6 ± 1.3 Ma. These two data require thatsignificant ductile deformation occurred between 3106 and3104 Ma. Macroscopic kinematic indicators within thegneiss samples (though sparse) suggest top to the southwestmovement (involving a combination of dextral and normalmovement), which is consistent with the sudden jump inmetamorphic grade into the hanging wall metasedimentschists. Dziggel et al. [2006] suggest both sinistral andnormal component of shear several kilometers to the west ofour study area. Such a change in displacement along strikemay reflect the rotating orientation of the orthogneisscontact as a means of generating complex stress fields. Inany case, because both sets of shallowly plunging lineationsare structurally related [Dziggel et al., 2006] and containedwithin the circa 3106 Ma Stentor orthogneiss, the HSZ wasactive during or immediately after the emplacement of itsgranitic igneous protolith, in addition to or in place of circa3.23 Ga, as was concluded by Dziggel et al. [2006].Therefore, it is possible that the retrograde path of high-T,medium-P metamorphism recorded in these rocks may havebeen similarly related to 3.1 Ga exhumation.[44] In contrast, thermochronology from the circa 3.23 Ga

Kaap Valley and Nelshoogte plutons, also along the NWmargin of the BGB (Figure 2), is consistent with theirintrusion or rapid ascent into the middle or upper crust.U-Pb sphene and apatite data [Kamo and Davis, 1994;Schoene and Bowring, 2007] and 40Ar/39Ar hornblendedata [Layer et al., 1992] from the Kaap Valley plutondocument rapid cooling after intrusion circa 3227 Ma[Kamo and Davis, 1994; Schoene et al., 2006] (Figure 10).To the SW, our U-Pb apatite data from the 3236.2 ± 0.3 MaNelshoogte pluton (sample EKC03-9) range between 3225and 3200 Ma, consistent with slightly slower postintrusioncooling compared to the Kaap Valley pluton. 40Ar/39Arbiotite and muscovite from those plutons give dates from3.1 to 3.0 Ga that were interpreted to reflect a combinationof slow cooling and partial resetting by the circa 3.107 GaMpuluzi batholith [Layer et al., 1992, 1998]. In light of thedata presented in this study, the 40Ar/39Ar dates mayalso represent middle to upper crustal residence and slowcooling of these rocks during relaxation of isothermsfollowing circa 3.14–3.10 Ga tectonism and granitebatholith intrusion.

5. The 3.23–3.10 Ga Tectonothermal History

of the BGB

[45] The period of circa 3.23 Ga subduction accretionwas likely an important formative event in the creation ofthe present architecture of the BGB. The discovery of thecirca 3.258 Ga Stentor orthogneiss in combination withcirca 3.3 Ga xenoliths reported north of the BGB [Kamoand Davis, 1994], however, suggest that this period ofcrustal genesis was much longer lived than previouslyinferred (Figure 12). In addition, our findings from fourlocalities that flank the margins of the BGB supportprevious studies showing that fault zones demarcating the


20 of 27

TC5010

contacts between orthogneisses and supracrustal BGBrocks, as well as many thrust faults within the BGB supra-crustal sequences, are dominated or overprinted by strike-slip and/or extensional movement instead of contraction [deRonde and de Wit, 1994; de Ronde and Kamo, 2000; deRonde et al., 1991; Heubeck and Lowe, 1994a, 1994b; Loweet al., 1999]. This could indicate that following the 3.30–3.23 Ga period of continent building that ended in circa3.23 Ga compressional tectonism, there was a switch to adominantly strike-slip and/or extensional tectonic regime.Two end-member models have been proposed for themechanism of this transition. The first is that it was linkedto overthickening of the crust during collision-accretioncirca 3.23 Ga [Kisters et al., 2003; Diener et al., 2005;Dziggel et al., 2006], resembling orogenic collapseobserved in modern orogens worldwide. Thus, the collapseis manifested as extension along low-angle detachments andthe resulting gravitational unloading of unusually buoyantorthogneiss complexes that produces a diapiric inversion ofthe crust along steeply dipping normal faults. This modelwas generated for other cratonic settings [e.g., Marshak etal., 1997] and is convenient in that it helps explain theapparent ‘‘dome-and-keel’’ structure that exemplifies manygreenstone belts around the world [e.g., de Wit and Ashwal,1997; Van Kranendonck et al., 2007]. The other end-member model involves tectonic quiescence following3.23 Ga terrane accretion and a transition to strike-slipand transtensional tectonics that is concurrent and relatedto the intrusion of the prolific granitoid batholiths circa3.10 Ga [e.g., de Ronde and de Wit, 1994; Kamo and Davis,1994]. Decipheringwhether either of thesemodels is accuraterequires linking the kinematics and timing of deformationwith block exhumation on a regional scale.[46] If the dominant crustal architecture of the BGB at

present was the result of the orogenic collapse model ofKisters et al. [2003], one would expect a transition toextensional tectonics to follow immediately after contrac-tional deformation that occurred between 3230 and 3227 Mawithin the Onverwacht and Fig Tree Group rocks through-out SW BGB [de Ronde and de Wit, 1994; de Ronde andKamo, 2000; Kamo and Davis, 1994] and in the adjacentbasement complexes south of the BGB [e.g., Dziggel et al.,2005; Stevens et al., 2002]. In addition, considerable verti-cal exhumation of flanking orthogneiss terranes (i.e., cool-ing) along the bounding extensional faults would beexpected at that time. However, the majority of the datafrom this study (section 4, Figure 12) require that (1) zonesof extension and contraction were active synchronously atcirca 3.23 Ga and (2) large crustal blocks were exhumedfrom the middle/lower crust to the upper crust duringgranite batholith intrusion circa 3.14–3.10 Ga, in additionto or instead of at circa 3.23 Ga (Figure 12). In fact,suggesting an overall switch to extensional and strike-sliptectonics after circa 3.23 Ga may be slightly misleading,given that both types of kinematics likely existed duringcirca 3.23 Ga accretion. However, the importance of verticalexhumation and cooling of crustal blocks at circa 3.14–3.10 Ga is more easily accommodated if an importantextensional component existed at that time, which is

supported by field and geochronological data. Therefore,an orogenic collapse and diapiric episode at circa 3.23 Gamay or may not have occurred, but there is little geochro-nologic evidence for it, and its existence would be partiallyobscured by the drastic modification of crustal architecturecirca 3.14–3.10 Ga.[47] We suggest that the near contemporaneous existence

of contractional, extensional and strike-slip kinematicscombined with widespread magmatism circa 3.23 Ga ismost consistent with a model involving dextral obliquesubduction-accretion in which the BGB and its boundingblocks accommodated strain in such a way that resulted inan asymmetric flower structure geometry. This mechanismcan accommodate the simultaneous occurrence of multiplemodes of deformation without appealing to major changesin the regional tectonic regime [e.g., Holdsworth et al.,1998]. In such an environment, the specific kinematics on afault plane would be dictated by the local stresses related byrigid bodies such as plutons or gneiss domes and thepreexisting orientation of fault surfaces (Figure 12). Forexample, in an overall dextral transpressional systemaccommodated along NE–SW faults, transtension may infact be expected along a fault like the PSZ with a moreeasterly strike as small dilational transcurrent zone (Figure 12).This interpretation is compatible with previous modelsregarding the localized timing of compression or inferredsubduction and accretion (see references above) and obliqueconvergence [de Ronde and de Wit, 1994]. Sedimentologyin the Moodies Group is consistent with deposition ofconglomerates into a wide array of depositional settingscontemporaneous with the NW–SE shortening in the belt[Heubeck and Lowe, 1994a, 1994b; Lamb, 1984; Lamb andParis, 1988; Lowe and Byerly, 1999b]. For example,Heubeck and Lowe [1994b] showed that Moodies deposi-tion in the western BGB, north of the Saddleback-Inyokafault system (Figures 2 and 9), was initiated in extensionalbasins, some of which were then converted to contractionaldepositional settings in intermontane basins. Additionally,clast composition and paleocurrent directions are variablewithin the Moodies Groups across the belt [Heubeck andLowe, 1994a, 1999; Jackson et al., 1987; Compston andKroner, 1988], indicating a potentially complex tectonicsetting for their deposition. Therefore, while there is goodevidence for thickening of the crust in a convergrenttectonic regime circa 3.23 Ga, there is little evidence thatthe thickened crustal column resulted in gravitational insta-bility such as diapirism on a regional scale that postdatedcrustal thickening. We suggest that if such a processoccurred, it may be recorded in the Kaap Valley or Nel-shoogte plutons, as they are the only studied orthogneissbodies that have a thermal history consistent with such amodel. Exhumation and core complex formation circa3230 Ma in the Steynsdorp dome clearly indicates thatbuoyancy driven processes were important, as in otherorogenic belts, though in this case either predates or issynchronous with contraction in the belt.[48] New and published data indicate that the majority of

differential exhumation recorded by juxtaposed metamor-phic blocks along the boundaries of the BGB occurred


21 of 27

TC5010

during granite batholith intrusion and strike-slip to trans-tensional tectonics circa 3.14–3.10 Ga. During this timereactivation of older structures partitioned strain into local-ized zones and acted to help preserve the older geologichistories of individual tectonic blocks (Figure 12). Asduring 3.23 Ga transform accretion, it is possible thattranspressive kinematics would be recorded in some local-ities, depending on the orientation of reactivated faults.Westraat et al. [2005] argue that the circa 3.1 Ga Mpuluzibatholith intruded in a dilational jog of an overall (oblique-dextral) compressive stress field. However, we argue thatbased on the orientation of many observed shear zones (e.g.,the PSZ and HSZ, Figure 2, with low-grade rocks in thehanging wall) and the amount of necessary vertical exhu-mation to account for the cooling patterns (e.g., 7–15 km inthe PSZ; Figure 12 [Schoene and Bowring, 2007]), thatsignificant amounts of normal offset were also required atthis time. If strike-slip motion was dominant, it requires thatrocks traveled large lateral distances as well, potentiallyjuxtaposing blocks of crust with vastly different sedimen-tological, structural and thermal histories. Similarly long-lived transform boundary accretionary events have beensuggested in more modern orogens, for example the north-ern Cordillera in North America [e.g., Johnson, 2001; Irvinget al., 1996] and the Alpine fault system in New Zealand[e.g., Walcott, 1998], and both involve significant amountsof vertical displacement in addition to 1000s of kilometersof lateral transport. In this case, ‘‘dome-and-keel’’ formationin the BGB need not be the result of a single and instan-taneous regional-scale event but instead may be bestexplained by a series of processes having affected a nowdeeply eroded, long-lived transform boundary.[49] The interaction between magmatism and deforma-

tion is of great importance for the generation and subse-quent evolution of the continents [Axen et al., 1998;Hollister and Crawford, 1986; Hollister and Andronicos,2006; Hutton, 1988; Klepeis et al., 2003; Mahan et al.,2003; Miller and Paterson, 2001; Pavlis, 1996]. Thus, thecoincidence of 3.2–3.1 Ga magmatism and transpressionalto transtensional deformation within the BGB is in manyways typical of more modern orogenic settings and zones ofcrustal building. Several mechanisms can be invoked tocause the coincidence of strike-slip and transtensionaltectonics and granitic batholith intrusion between 3.14 and3.10 Ga in the BGB: crustal thickening and extended lowercrustal incubational heating over 100 Ma [e.g., Bird, 1991;Kusznir and Matthews, 1988; Sandiford et al., 2004; Zhaoand Morgan, 1987], lower crustal delamination [Bird, 1979;Meissner and Mooney, 1998], or simply by further fieldplate interactions given the abundance of circa 3.1 Gaactivity throughout the central craton [de Wit et al., 1992;Moser et al., 2001]. Deciphering between these differentmechanisms will involve a systematic treatment of circa3.14–3.10 Ga granite geochemistry and intrusive mecha-nisms on a larger regional scale, which requires furtherwork and is beyond the scope of this paper. In reality, theBGB is likely a very small portion of a potentially muchlarger orogenic system. Further work north and south of theBGB in the Nelspruit migmatite terranes described by

Viljoen and Viljoen [1969] and Robb and Anhaeusser[1983] and to the south in the AGC and southern Swazilandmay help answer these questions as well as further definethe structural geometries circa 3.23–3.10 Ga. The newconstraints provided in this contribution and from previouswork form a solid foundation on which further regionalmodels can be built.

6. Implications for Stabilization of the Eastern

Kaapvaal Craton

[50] The period of strike-slip to transtensional tectonics atcirca 3.2–3.1 Ga is the last penetrative deformational eventdocumented in the vicinity of the BGB and may be relatedto the final stabilization of this segment of lithosphere. Twomajor modifications of the compositional and structuralarchitecture of this portion of the crust likely occurredduring this period: (1) crustal extension associated withtranstensional faulting may have led to an overall thinnercrust, and (2) perhaps more importantly, granitic magmagenerated in the lower crust intruded as subhorizontal planarsheets in the middle to upper crust. This transition paved theway for erosion and peneplanation of at least 5–10 km ofupper crust prior to shallow marine deposition of theTransvaal Supergroup on the western margin of the BGBat circa 2.6 Ga, and probably by 2.9 Ga before the marinetransgressions of the lower Witwatersrand basin andPongola sequences to the west and south, respectively[e.g., de Wit, 2007].[51] If circa 3.14–3.10 Ga crustal modification resulted

in thinner crust, as may be implied by the significantcooling recorded by U-Pb thermochronology, it is predict-able that such a process would create a lithosphere that isstructurally stable over long intervals (Figure 13). Rheolog-ical profiles of materials as a function of pressure andtemperature show that a material’s intrinsic strength gener-ally increases as a function of depth until it passes from thebrittle into the ductile flow regime, at which point thestrength decreases dramatically with depth [Brace andKohlstedt, 1980; Kusznir and Karner, 1985; Ranalli,1995; Ranalli and Murphy, 1987]. Therefore, a crust ofany single composition and mineralogy will be stronger ifits thickness is near the brittle-ductile transition, which isusually in the middle to lower crust, depending on thecomposition of the crust and the geotherm.[52] The transport of granitic magma into the upper crust

can help stabilize it in several ways in addition to generatinga strong rheological column. First, the placement of hori-zontally extensive sheets of granitic material can mechan-ically strengthen the crust, perhaps by crosscuttingpreexisting weaknesses with rigid material [Davidson etal., 1994; Karlstrom and Williams, 1998; Pavlis, 1996].Perhaps more important is the redistribution of the radio-genic heat producing elements (HPEs) from the lower crustinto higher crustal levels. This mechanism can act tostabilize lower crust by lowering its long-term heat produc-tion and also by creating opportunity to remove HPEs fromthe crustal column entirely through erosion of the uppercrust [Sandiford and McLaren, 2002; Sandiford et al., 2002;


22 of 27

TC5010

Morgan, 1985]. To the west of our field area, in the centralKaapvaal craton, exhumation and peneplanation of circa3.1 Ga granites and their subsequent burial beneath theWitswatersrand foreland basin occurred during a relativelyshort amount of time, constrained by the deposition ofDominion Group volcanics circa 3.09–3.07 Ga [Armstronget al., 1991; Robb et al., 1992]. This process acted toremobilize HPEs upward in the crust, which in turn hadan important effect on the final distribution of heat in the crustthere [Hart et al., 1981; Nicolaysen et al., 1981; Johnson etal., 2006]. Timing of upper crustal exhumation is less wellconstrained in the eastern craton, but deposition of the craton-wide Transvaal Supergroup at circa 2.6 Ga adjacent to theBGB and deposition of the Pongola Supergroup in southernSwaziland at circa 2.9 Ga [e.g., Gutzmer et al., 1999; Striket al., 2007] provide some constraints for upper crustalpeneplanation. Thus, there are a number of positive feed-backs between extension and granitic magmatism that couldhave contributed toward a stable cratonic crust that is separatefrom the generation of a cratonic lithospheric mantle throughsubduction and accretion [Jordan, 1988; Schmitz et al., 2004;Grove and Parman, 2004].[53] Our observations in concert with previous investiga-

tions allow us to pose a three-stage model for stabilizationof the eastern Kaapvaal craton: (1) subduction, mantlemelting and block collision/accretion at circa 3.23 Ga ledto depletion, accretional thickening and stabilization of themantle lithosphere [de Wit et al., 1992; Jordan, 1988;Parman et al., 2004], (2) extension, crustal thinning, andlower crustal melting circa 3.14–3.10 Ga led to a morestable crustal rheological structures, which was furtherinduced by (3) erosion and removal of HPE-rich uppercrust. This sequence of processes gave the lithosphericcolumn an inherent strength that has isolated it from furtherhorizontal deformation for circa 3.0 Ga, despite its recentproximity to the margin of the African plate.[54] The timing of stabilization is apparently different

between the eastern, western, and northern Kaapvaal craton,in that most of the western and northern craton (Kimberly

and Pietersburg blocks, respectively, see Figure 1) experi-enced several episodes of compression and extension be-tween 3.1 and 2.7 Ga (see summaries of de Wit et al. [1992],Schmitz et al. [2004], Thomas et al. [1993], and Johnson etal. [2006]). Generation and apparent stabilization of themantle lithosphere in the vicinity of Kimberley (Figure 1) isconstrained by circa 2.9 Ga Re-Os model ages fromkimberlite-borne mantle peridotite xenoliths and diamondinclusions [Carlson et al., 1999; Richardson et al., 1984,2001], which Schmitz et al. [2004] argued was the result ofmantle growth during convergence and block accretion atcirca 2.9 Ga between the Witswatersrand and Kimberleyblocks (Figure 1). However, significant crustal modificationoccurred after circa 2.9 Ga block collision across largeportions of the central craton, most notably that of the circa2.71 Ga Ventersdorp extensional episode. The effects of thisevent are evidenced as flood volcanism, rift basin subsi-dence (Figure 1) and also as ultrahigh temperature meta-morphism of the lower crust [Armstrong et al., 1991; Burkeet al., 1985; de la Winter, 1976; de Wit and Tinker, 2004,Schmitz and Bowring, 2003b; Tinker et al., 2002]. This finalreactivation of the western Kaapvaal craton brings upseveral important points. The first is that the BGB andnorthern AGC were largely unaffected by these events,illustrating that craton stabilization is diachronous, affectingdifferent portions of lithosphere at different times. Second,it is an interesting observation that extensional riftingassociated with the Ventersdorp event, not compression,was the final event to modify substantial portions ofcratonic crust to the west of the areas described in thispaper. Third, the preservation of peridotic diamond ages ofcirca 2.9–3.2 Ga from lithospheric mantle may imply adecoupling between mantle and crust at circa 2.7 Ga[Doucoure and de Wit, 2002; Schmitz and Bowring,2003b]. If our model for the eastern Kaapvaal craton iscorrect, it also implies decoupling between the crust andmantle in the eastern Kaapvaal craton during circa 3.1 Gacrustal (but not mantle) thinning, if crustal thinning was anecessary process in addition to HPE redistribution. This

Figure 13. Cartoon illustrating how lower crustal melting and crustal thinning can result in a strongerlithospheric column, as occurred in the eastern Kaapvaal craton circa 3.2–3.1 Ga. Note that structuresand therefore the mechanism of crustal reworking are not shown. Black box is a histogram indicating therelative concentration of heat-producing elements (HPEs) U, Th, and K within the crust before and aftergranite production and migration upward in the crust [Sandiford et al., 2002]. Unitless s1-s3 representsthe differential stress required for failure of a rock (i.e., higher s1-s3 corresponds to a stronger crust),plotted as a function of depth in the lithosphere, assuming a quartzofeldspathic crust and peridotiticmantle, both with homogeneous compositions, modified from Ranalli and Murphy [1987].


23 of 27

TC5010

striking similarity between the history of events recorded inthe eastern and the west central parts of the Kaapvaal cratonthat potentially led to their temporally distinct stabilizationssuggests a general applicability of this model for cratoniclithosphere growth and preservation.

[55] Acknowledgments. This work benefited from very helpful com-ments and reviews at various stages of the manuscript from K. Burke,C. Heubeck, K. Hodges, S. Kamo, A. Kisters, K. Klepeis, A. Kroner,R. Parrish, and one anonymous reviewer. This work was supported in partby NSF EAR9526702. M.D.W.’s research in the BGB is supported by theNational Research Foundation of South Africa. This is AEON contribution19. Field work in the PSZ was made possible in part by the kindness of thePhophonyane Falls Lodge and Nature Reserve.

ReferencesAleinikoff, J. N., R. P.Wintsch, C. M. Fanning, andM. J.

Dorais (2002), U-Pb geochronology of zircon andpolygenetic titanite from the Glastonbury Complex,Connecticut, USA: An integrated SEM, EMPA,TIMS, and SHRIMP study, Chem. Geol., 188, 125–147, doi:10.1016/S0009-2541(02)00076-1.

Anhaeusser, C. R. (1969), The stratigraphy, structure,and gold mineralization of the Jamestown and ShebaHills areas of the Barberton Mountain Land,Ph.D. thesis, 332 pp., Univ. of Witwatersrand,Johannesburg, South Africa.

Anhaeusser, C. R. (1972), The geology of theJamestown hills area of the Barberton MountainLand, South Africa, Trans. Geol. Soc. S. Afr., 75,225–263.

Anhaeusser, C. R. (1976), The geology of the ShebaHills area of the Barberton Mountain Land, SouthAfrica, with particular references to the Eurekasyncline:, Trans. Geol. Soc. S. Afr., 79, 253–280.

Anhaeusser, C. R. (1983), Contributions to the Geologyof the Barberton Mountain Land, Spec. Publ. Geol.Soc. S. Afr., 9, 223 pp.

Armstrong, R. A., W. Compston, M. J. de Wit, and I. S.William (1990), The stratigraphy of the 3.5 – 3.2 GaBarberton Greenstone Belt revisited: A single zirconion microprobe study, Earth Planet. Sci. Lett., 101,90–106, doi:10.1016/0012-821X(90)90127-J.

Armstrong, R. A., W. Compston, E. A. Retief, I. S.Williams, and H. J. Welke (1991), Zircon ionmicroprobe studies bearing on the age and evolu-tion of the Witwatersrand triad, Precambrian Res.,53, 243 –266, doi:10.1016/0301-9268(91)90074-K.

Assumpcao, M., D. James, and A. Snoke (2002), Crustalthicknesses in SE Brazilian Shield by receiver func-tion analysis: Implications for isostatic compensa-tion, J. Geophys. Res., 107(B1), 2006, doi:10.1029/2001JB000422.

Axen, G. J., J. E. Selverstone, T. Byrne, and J. M.Fletcher (1998), If the strong crust leads, will theweak crust follow?, GSA Today, 8, 1 – 8.

Begemann, F., K. R. Ludwig, G. W. Lugmair, K. Min,L. E. Nyquist, P. J. Patchett, P. R. Renne, C.-Y.Shih, I. M. Villa, and R. J. Walker (2001), Callfor an improved set of decay constants for geochro-nological use, Geochim. Cosmochim. Acta, 65,111–121, doi:10.1016/S0016-7037(00)00512-3.

Bell, D. R., and R. O. Moore (2004), Deep chemicalstructure of the southern African mantle fromkimberlite megacrysts, S. Afr. J. Geol., 107, 59–80, doi:10.2113/107.1-2.59.

Bird, P. (1979),Continental delaminationand theColoradoPlateau, J. Geophys. Res., 84, 7561–7571.

Bird, P. (1991), Lateral extrusion of lower crust fromunder high topography, in the isostatic limit, J. Geo-phys. Res., 96, 10,275 – 10,286, doi:10.1029/91JB00370.

Brace, W. F., and D. L. Kohlstedt (1980), Limits onlithospheric strength imposed by laboratory experi-ments, J. Geophys. Res. , 85 , 6248 – 6252,doi:10.1029/JB085iB11p06248.

Burbank, D. W. (2002), Rates of erosion and theirimplications for exhumation, Mineral. Mag., 66,25– 52, doi:10.1180/0026461026610014.

Burke, K., W. S. F. Kidd, and T. Kusky (1985), Isthe Ventersdorp rift system of southern Africarelated to a continental collision between the

Kaapvaal and Zimbabwe Cratons at 2.64 Ga ago?,Tectonophysics, 115, 1 – 24, doi:10.1016/0040-1951(85)90096-4.

Byerly, G. R., A. Kroner, D. R. Lowe, W. Todt, andM. M. Walsh (1996), Prolonged magmatism andtime constraints for sediment deposition in the earlyArchean Barberton greenstone belt: Evidence fromthe Upper Onverwacht and Fig Tree groups,Precambrian Res., 78, 125 – 138, doi:10.1016/0301-9268(95)00073-9.

Carlson, R. W., D. G. Pearson, F. R. Boyd, S. B. Shirey,G. Irvine, A. H. Menzies, and J. J. Gurney (1999),Re-Os systematics of lithospheric peridotites:Implications for lithosphere formation and preser-vation, in Proceedings of the VIIth InternationalKimberlite Conference, edited by J. J. Gurneyet al., pp. 99–108, Red Roof Design, Cape Town,South Africa.

Carlson, R. W., et al. (2000), Continental growth, pre-servation, and modification in Southern Africa,GSA Today, 10, 1 – 8.

Chamberlain, K. R., and S. A. Bowring (2001),Apatite-feldspar U-Pb thermochronometer: Areliable mid-range (�450�C), diffusion controlledsystem, Chem. Geol., 172, 173–200, doi:10.1016/S0009-2541(00)00242-4.

Cherniak, D. J., W. A. Lanford, and F. J. Ryerson(1991), Lead diffusion in apatite and zircon usingion implantation and Rutherford Backscatteringtechniques, Geochim. Cosmochim. Acta, 55,1663–1673, doi:10.1016/0016-7037(91)90137-T.

Clitheroe, G., O. Gudmundsson, and B. L. N. Kennett(2000), The crustal thickness of Australia, J. Geo-phys. Res., 105, 13,697 – 13,713, doi:10.1029/1999JB900317.

Cloete, M. (1991), An overview of metamorphism inthe Barberton greenstone belt, in Two Cratons andan Orogen—Excursion Guidebook and ReviewArticles for a Field Workshop through Selected

Archean Terranes of Swaziland, South Africa andZimbabwe, L. D. Ashwal, pp. 84 –98, Johannes-burg, IGCP project 280, Dep. of Geol., Univ. ofthe Witwatersrand, Johannesburg, South Africa.

Compston, W., and A. Kroner (1988), Multiple zircongrowth within early Archean tonalitic gneiss fromthe Ancient Gneiss Complex, Swaziland, EarthPlanet. Sci. Lett., 87, 13 – 28, doi:10.1016/0012-821X(88)90061-1.

Condie, K. C., J. E. Macke, and T. O. Reimer (1970),Petrology and geochemistry of early Precambriangraywackes from the Fig Tree Group, South Africa,Geol. Soc. Am. Bull., 81, 2759–2776, doi:10.1130/0016-7606(1970)81[2759:PAGOEP]2.0.CO;2.

Davidson, C., S. M. Schmidt, and L. S. Hollister(1994), Role of melt during deformation in the deepcrust, Terra Nova, 6, 133–142, doi:10.1111/j.1365-3121.1994.tb00646.x.

Davis, G. A., and G. S. Lister (1988), Detachmentfaulting in continental extension: Perspectives fromthe southwestern U.S. Cordillera,, in Processes in

Continental Lithospheric Deformation, edited byS. P. Clark Jr., B. C. Burchfiel, and J. Suppe, Spec.Pap. Geol. Soc. Am., 218, 133–159.

de la Winter, H. R. (1976), A lithostratigraphic classi-fication of the Ventersdorp succession, Trans. Geol.Soc. S. Afr., 79, 31– 48.

de Ronde, C. E. J., and M. J. de Wit (1994), Tectonichistory of the Barberton Greenstone Belt, SouthAfrica: 490 million years of Archean crustal evolu-tion, Tectonics, 13, 983 – 1005, doi:10.1029/94TC00353.

de Ronde, C. E. J., and S. Kamo (2000), An Archeanarc-arc collisional event: A short-lived (ca. 3 Myr)episode, Weltevreden area, Barberton greenstonebelt, South Africa, J. Afr. Earth Sci., 30, 219 –248, doi:10.1016/S0899-5362(00)00017-8.

de Ronde, C. E. J., S. Kamo, D. W. Davis, M. J. de Wit,and E. T. C. Spooner (1991), Field, geochemicaland U-Pb isotopic constraints from hypabyssal fel-sic intrusions within the Barberton greenstone belt,South Africa: Implications for tectonics and thetiming of gold mineralization, Precambrian Res.,49, 261–280, doi:10.1016/0301-9268(91)90037-B.

de Ronde, C. E. J., M. J. de Wit, and E. T. C. Spooner(1994), EarlyArchean (>3.2Ga) Fe-oxide-rich, hydro-thermal discharge veins in the Barberton greenstonebelt, South Africa,Geol. Soc. Am. Bull., 106, 86–104.

de Wit, M. J. (1982), Gliding and overthrust nappetectonics in the Barberton greenstone belt, J. Struct.Geol., 4, 117 – 136, doi:10.1016/0191-8141(82)90022-0.

de Wit, M. J. (2007), The Kalahari Epeirogeny andclimate change: Differentiating cause and effectfrom core to space, S. Afr. J. Geol., 110, 367 –392, doi:10.2113/gssajg.110.2-3.367.

de Wit, M. J., and L. D. Ashwal (Eds.) (1997),Greenstone Belts, Oxford Monogr. Geol. Geo-phys., vol. 35, 809 pp., Clarendon, Oxford, U.K.

de Wit, M. J., and J. Tinker (2004), Crustal structuresacross the central Kaapvaal craton from deep-seismicreflection data, S. Afr. J. Geol., 107, 185 – 206,doi:10.2113/107.1-2.185.

de Wit, M. J., R. E. P. Fripp, and I. G. Stanistreet(1983), Tectonic and stratigraphic implications ofnew field observations along the southern part ofthe Barberton greenstone belt, Spec. Publ. Geol.Soc. S. Afr., 9, 21 –29.

de Wit, M. J., R. A. Armstrong, R. J. Hart, and A. H.Wilson (1987a), Felsic igneous rocks within the3.3 –3.5 Ga Barberton greenstone belt: High crustallevel equivalents of the surrounding tonalite-trondhjemite terrain, emplaced during thrusting,Te c t o n i c s , 6 , 5 2 9 – 5 4 9 , d o i : 1 0 . 1 0 2 9 /TC006i005p00529.

de Wit, M. J., R. A. Hart, and R. J. Hart (1987b), TheJamestown ophiolite complex, Barberton mountainbelt; a section through 3.5 Ga oceanic crust, J. Afr.Earth Sci., 6, 681 – 730, doi:10.1016/0899-5362(87)90007-8.

de Wit, M. J., C. Roering, R. J. Hart, R. A. Armstrong,C. E. J. de Ronde, R. W. E. Green, M. Tredoux,E. Peberdy, and R. A. Hart (1992), Formation ofan Archean continent, Nature, 357, 553 – 562,doi:10.1038/357553a0.

Diener, J. F. A., G. Stevens, A. F. M. Kisters, andM. Poujol (2005), Metamorphism and exhumationof the basal parts of the Barberton greenstone belt,South Africa: Constraining the rates of Mesoarcheantectonism, Precambrian Res., 143, 87 – 112,doi:10.1016/j.precamres.2005.10.001.

Doucoure, C. M., and M. J. de Wit (2002), Temporalvariation in rigidity and mechanical behavior of old


24 of 27

TC5010

thick continental lithosphere, S. Afr. J. Geol., 105,39– 50, doi:10.2113/1050039.

Durrheim, R. J., and W. D. Mooney (1994), Evolutionof the Precambrian lithosphere: Seismological andgeochemical constraints, J. Geophys. Res., 99,15,359–15,374, doi:10.1029/94JB00138.

Dziggel, A., G. Stevens, M. Poujol, C. R. Anhaeusser,and R. A. Armstrong (2002), Metamorphism of thegranite-greenstone terrane south of the BarbertonGreenstone Belt, South Africa: An insight into thetectono-thermal evolution of the ’lower’ portions ofthe Onverwacht Group, Precambrian Res., 114,221–247, doi:10.1016/S0301-9268(01)00225-X.

Dziggel,A., R.A.Armstrong,G. Stevens, andL.Nasdala(2005), Growth of zircon and titanite duringmetamorphism in the granitoid-gneiss terrane southof the Barberton greenstone belt, South Africa,Mineral. Mag., 69, 1019 – 1036, doi:10.1180/0026461056960305.

Dziggel, A., S. Knipfer, A. F. M. Kisters, and F. M.Meyer (2006), P-T and structural evolution duringexhumation of high-T, medium-P basement rocksin the Barberton Mountain Land, South Africa,J. Metamorph. Geol., 24, 535 –551, doi:10.1111/j.1525-1314.2006.00653.x.

Eriksson, K. A. (1980), Transitional sedimentationstyles in the Moodies and Fig Tree Groups,Barberton Mountain Land, South Africa: Evidencefavouring an Archean continental margin,Precambrian Res., 12, 141 – 160, doi:10.1016/0301-9268(80)90027-3.

Fletcher, J. M., and J. M. Bartley (1994), Constrictionalstrain in a non-coaxial shear zone: Implications forfold and rock fabric development, central Mojavemetamorphic core complex, California, J. Struct.Geol. , 16(4), 555 – 570, doi:10.1016/0191-8141(94)90097-3.

Fouch, M. J., D. E. James, J. C. VanDecar, S. van derLee, and K. S. Group (2004), Mantle seismic struc-ture beneath the Kaapvaal and Zimbabwe Cratons,S. Afr. J. Geol., 107, 33 – 44, doi:10.2113/107.1-2.33.

Fripp, R. E. P., D. A. Van Nierop, M. J. Callow, P. A.Lilly, and L. U. de Plessis (1980), Deformation inpart of the Archaean Kaapvaal Craton, SouthAfrica , Precambrian Res. , 13 , 241 – 251,doi:10.1016/0301-9268(80)90006-6.

Frost, B. R., K. R. Chamberlain, and J. C. Schumacher(2001), Sphene (titanite): Phase relations and role asa geochronometer, Chem. Geol., 172, 131 – 148,doi:10.1016/S0009-2541(00)00240-0.

Gay, N. C. (1969), The analysis of strain in the BarbertonMountain Land, Eastern Transvaal, using deformedpebbles, J. Geol., 77, 377–396.

Grove, T. L., and S. W. Parman (2004), Thermal evolu-tion of the Earth as recorded by komatiites, EarthPlanet. Sci. Lett, 219, 173 – 187, doi:10.1016/S0012-821X(04)00002-0.

Gupta, S., S. S. Rai, K. S. Prakasam, D. Srinagesh,B. K. Bansal, R. K. Chadha, K. Priestley, andV. K. Gaur (2003), The nature of the crust insouthern India: Implications for Precambrian crustalevolution, Geophys. Res. Lett., 30(8), 1419,doi:10.1029/2002GL016770.

Gutzmer, J., N. Nhleko, N. J. Beukes, A. Pickard, andM. E. Barley (1999), Geochemistry and ion microp-robe (SHRIMP) age of a quartz porphyry sill in theMozaan Group of the Pongola Supergroup: Implica-tions for thePongola andWitwatersrand supergroups,S. Afr. J. Geol., 102, 139–146.

Hart, R. J., L. O. Nicolaysen, and N. H. Gale (1981),Radioelement concentrations in the deep profilethrough Precambrian basement of the Vredefortstructure, J. Geophys. Res., 86, 10,639 – 10,652,doi:10.1029/JB086iB11p10639.

Heubeck, C., and D. R. Lowe (1994a), Depositionaland tectonic setting of the Archean MoodiesGroup, Barberton Greenstone Belt, South Africa,Precambrian Res., 68, 257 – 290, doi:10.1016/0301-9268(94)90033-7.

Heubeck, C., and D. R. Lowe (1994b), Late syndeposi-tional deformation and detachment tectonics in the

Barberton Greenstone Belt, South Africa, Tectonics,13, 1514–1536, doi:10.1029/94TC01809.

Heubeck, C., and D. R. Lowe (1999), Sedimentary pet-rography and provenance of the Archean MoodiesGroup, Barberton Greenstone Belt, in GeologicEvolution of the Barberton Greenstone Belt, SouthAfrica, edited by D. R. Lowe and G. R. Byerly,Spec. Pap. Geol. Soc. Am., 329, 259–286.

Holdsworth, R. E., R. A. Strachan, and J. F. Dewey(Eds.) (1998), Continental Transpressional andTranstensional Tectonics, Geol. Soc. Spec. Publ.,135, 354 pp.

Hollister, L. S., and C. L. Andronicos (2006), Forma-tion of new continental crust in Western BritishColumbia during transpression and transtension,Earth Planet. Sci. Lett., 249, 29 –38, doi:10.1016/j.epsl.2006.06.042.

Hollister, L. S., and M. L. Crawford (1986), Melt-enhanced deformation: A major tectonic process,Geology, 14, 558 – 561, doi:10.1130/0091-7613(1986)14<558:MDAMTP>2.0.CO;2.

Hunter, D. R., F. Barker, and H. T. Millard (1978), Thegeochemical nature of the Archean Ancient GneissComplex and granodiorite suite, Swaziland: A pre-liminary study, Precambrian Res., 7, 105 – 127,doi:10.1016/0301-9268(78)90030-X.

Hunter, D. R., F. Barker, and H. T. Millard (1984),Geochemical investigations of Archean bimodaland Dwalile metamorphic suites, Ancient GneissComplex, Swaziland, Precambrian Res., 24, 131 –155, doi:10.1016/0301-9268(84)90055-X.

Hutton, D. H. W. (1988), Granite emplacementmechanisms and tectonic controls: Inferences fromdeformation studies, Trans. R. Soc. EdinburghEarth Sci., 79, 245–255.

Irving,E., P. J.Wayne,D. J. Thorkelson, andP. Schiarizza(1996), Large (1000 to 4000 km) northward move-ments of tectonic domains in the northern Cordillera,83 to 45 Ma, J. Geophys. Res., 101(B8), 17,901 –17,916, doi:10.1029/96JB01181.

Jackson, M. P. A. (1979), High-strain deformation ofthe Ancient Gneiss Complex in the Mankayanearea, Swaziland: A preliminary account, paperpresented at 18th Congress, Geol. Soc. S. Afr.,Johannesburg.

Jackson, M. P. A. (1984), Archean structural styles inthe Ancient Gneiss Complex of Swaziland, south-ern Africa, in Precambrian Tectonics Illustrated,edited by A. Kroner and R. Greiling, pp. 1 – 18,Schweizerbart’sche, Stuttgart, Germany.

Jackson,M. P.A., K.A.Eriksson, andC.W.Harris (1987),EarlyArcheanforedeepsedimentationrelatedtocrustalshortening: A reinterpretation of the BarbertonSequence, southern Africa, Tectonophysics, 136,197–221, doi:10.1016/0040-1951(87)90025-4.

Jaffey, A. H., K. F. Flynn, L. E. Glendenin, W. C.Bentley, and A. M. Essling (1971), Precision mea-surement of half-lives and specific activities of 235Uand 238U, Phys. Rev. C, 4, 1889–1906.

James, D. E., and M. J. Fouch (2002), Formation andevolution of Archean cratons: Insights from South-ern Africa, in The Early Earth: Physical Chemicaland Biological Development, edited by C. Ebinger,M. Fowler, and C. J. Hawkesworth, Geol. Soc.Spec. Publ., 199, 1 – 26.

James, D. E., M. J. Fouch, J. C. VanDecar, S. van derLee, and K. S. Group (2001), Tectospheric structurebeneath southern Africa, Geophys. Res. Lett., 28,2485–2488, doi:10.1029/2000GL012578.

Johnson, M. R., C. R. Anhaeusser, and R. J. Thomas(2006), The Geology of South Africa, 691 pp., Geol.Soc. of S. Afr., Johannesburg.

Johnson, S. T. (2001), The Great Alaskan TerraneWreck: Reconciliation of paleomagnetic and geolo-gical data in the northern Cordillera, Earth Planet.Sci. Lett., 193(3 – 4), 259–272, doi:10.1016/S0012-821X(01)00516-7.

Jordan, T. H. (1978), Composition and development ofthe continental tectosphere, Nature, 274, 544–548,doi:10.1038/274544a0.

Jordan, T. H. (1988), Structure and formation of thecontinental tectosphere, in Oceanic and Continental

Lithosphere: Similarities and Differences, edited byM. A. Menzies and K. G. Cox, pp. 11–37, OxfordUniv. Press, London.

Kamo, S., and D. W. Davis (1994), Reassessment ofArchean crustal development in theBarbertonMoun-tain Land, South Africa, based on U-Pb dating, Tec-tonics, 13, 167–192, doi:10.1029/93TC02254.

Karlstrom, K. E., and M. L. Williams (1998), Hetero-geneity of the middle crust: Implications forstrength of continental lithosphere, Geology, 26,815 – 818 , do i :10 .1130 /0091 -7613(1998)026<0815:HOTMCI>2.3.CO;2.

Kisters, A. F. M., and C. R. Anhaeusser (1995a),Emplacement features of Archean TTG plutonsalong the southern margin of the Barberton green-stone belt, South Africa, Precambrian Res., 75, 1 –15, doi:10.1016/0301-9268(95)00003-N.

Kisters,A.F.M., andC.R.Anhaeusser (1995b),The struc-tural significanceoftheSteynsdorpplutonandanticlinewithin the tectonomagmatic framework of the Barber-tonMountain Land, S. Afr. J. Geol., 98, 43–51.

Kisters, A. F. M., G. Stevens, A. Dziggel, and R. A.Armstrong (2003), Extensional detachment faultingand core-complex formation in the southernBarberton granite-greenstone terrain, South Africa:Evidence for a 3.2 Ga orogenic collapse,Precambrian Res., 127, 355 – 378, doi:10.1016/j.precamres.2003.08.002.

Klepeis, K. A., G. L. Clarke, and T. Rushmer (2003),Magma transport and coupling between deforma-tion and magmatism in the continental lithosphere,GSA Today, 13 , 4 – 11, doi:10.1130/1052-5173(2003)013<0004:MTACBD>2.0.CO;2.

Krogh, T. E. (1982), Improved accuracy of U-Pb zirconages by the creation of more concordant systemsusing an air abrasion technique, Geochim. Cosmo-chim. Acta, 46, 637 – 649, doi:10.1016/0016-7037(82)90165-X.

Kroner, A., W. Compston, and I. S. William (1989),Growth of early Archean crust in the AncientGneiss Complex of Swaziland as revealed by singlezircon dating, Tectonophysics, 161, 271 – 298,doi:10.1016/0040-1951(89)90159-5.

Kroner, A., E. Hegner, J. I. Wendt, and G. R. Byerly(1996), The oldest part of the Barberton granitoid-greenstone terrain, South Africa: Evidence for crustformation between 3.5 and 3.7 Ga, PrecambrianRes., 78, 105 – 124, doi:10.1016/0301-9268(95)00072-0.

Kusznir, N., and G. Karner (1985), Dependence of theflexural rigidity of the continental lithosphere onrheology and temperature, Nature, 316, 138–142,doi:10.1038/316138a0.

Kusznir, N., and D. H. Matthews (1988), Deep seismicreflections and the deformational mechanics of thecontinental lithosphere, J. Petrol., Special Litho-sphere Issue, 63–87.

Lamb, S. H. (1984), Structures on the eastern margin ofthe Archaean Barberton greenstone belt, northwestSwaziland, in Precambrian Tectonics Illustrated,edited by A. Kroner and R. Greiling, pp. 19–39,E. Schweizerbart’sche, Stuttgart, Germany.

Lamb, S. H., and I. Paris (1988), Post-OnverwachtGroup stratigraphy in the SE part of the ArcheanBarberton greenstone belt, J. Afr. Earth Sci., 7,285 –306, doi:10.1016/0899-5362(88)90074-7.

Layer, P.W.,A.Kroner, andD.York (1992), Pre-3000Mathermal history of the Archean Kaap Valley pluton,South Africa, Geology, 20, 717–720, doi:10.1130/0091-7613(1992)020<0717:PMTHOT>2.3.CO;2.

Layer, P. W., M. Lopez-Martinez, A. Kroner, andD. York (1998), Thermochronometry and paleo-magnetism of the Archean Nelshoogte Pluton,South Africa, Geophys. J. Int., 135, 129–145,doi:10.1046/j.1365-246X.1998.00613.x.

Lin, S. (2005), Synchronous vertical and horizontaltectonism in the Neoarchean: Kinematic evidencefrom a synclinal keel in the northwestern Superiorcraton, Canada, Precambrian Res., 139, 181–194,doi:10.1016/j.precamres.2005.07.001.

Lin, S., and D. Jiang (2001), Using along-strike varia-tion in strain and kinematics to define the move-


25 of 27

TC5010

ment direction of curved transpressional shearzones: An example from northwestern SuperiorProvince, Manitoba, Geology, 29(9), 767 – 770,doi:10.1130/0091-7613(2001)029<0767:UASVIS>2.0.CO;2.

Lin, S., D. Jiang, and P. F. Williams (2007), Importanceof differentiating ductile slickenslide striations fromstretching lineations and variation of shear directionacross a high-strain zone, J. Struct. Geol., 29, 850–862, doi:10.1016/j.jsg.2006.12.006.

Lister, G. S., G. Banga, and A. Feenstra (1984), Meta-morphic core complexes of Cordilleran type in theCyclades, Aegean Sea, Greece, Geology, 12, 221–225, doi:10.1130/0091-7613(1984)12<221:MCCOCT>2.0.CO;2.

Lopez Martinez, M., D. York, and J. A. Hanes (1992),A 40Ar/39Ar geochronological study of komatiitesand komatiitic basalts from the Lower OnverwachtVolcanics: Barberton Mountain Land, South Africa,Precambrian Res., 57, 91 –119, doi:10.1016/0301-9268(92)90095-6.

Lowe, D. R. (1994), Accretionary history of the ArcheanBarberton greenstone belt (3.55–3.22 Ga), southernAfrica,Geology, 22, 1099–1102.

Lowe, D. R. (1999), Geologic evolution of the Barber-ton Greenstone Belt and vicinity, in Geologic Evo-lution of the Barberton Greenstone Belt, SouthAfrica, edited by D. R. Lowe and G. R. Byerly,Spec. Pap. Geol. Soc. Am., 329, 287–312.

Lowe, D. R., and G. R. Byerly (Eds.) (1999a), Geolo-gic Evolution of the Barberton Greenstone Belt,South Africa, Spec. Pap. Geol. Soc. Am., 329.

Lowe, D. R., and G. R. Byerly (1999b), Stratigraphy ofthe west-central part of the Barberton GreenstoneBelt, in Geologic Evolution of the Barberton

Greenstone Belt, South Africa, edited by D. R.Lowe and G. R. Byerly, Spec. Pap. Geol. Soc.Am., 329, 1 –36.

Lowe, D. R., G. R. Byerly, and C. Heubeck (1999),Structural divisions and development of the west-central part of the Barberton Greenstone Belt, inGeologicEvolutionof theBarbertonGreenstoneBelt,

South Africa, edited by D. R. Lowe and G. R. Byerly,Spec. Pap. Geol. Soc. Am., 329, 37–82.

Mahan, K. H., J. M. Bartley, D. S. Coleman, A. F.Glazner, and B. S. Carl (2003), Sheeted intrusionof the synkinematic McDoogle pluton, SierraNevada, California, Geol. Soc. Am. Bull., 115,1570–1582.

Marshak, S., F. Alkmim, and H. Jordt-Evangelista(1992), Proterozoic crustal extension and the gen-eration of dome-and-keel structure in an Archaeangranite-greenstone terrane, Nature, 357, 491–493,doi:10.1038/357491a0.

Marshak, S., D. Tinkham, F. Alkmim, H. Brueckner,and T. Bornhorst (1997), Dome-and-keel provincesformed during Paleoproterozoic orogenic collapse –core complexes, diapirs, or neither?: Examples fromthe Quadrilatero Ferrıfero and the Penokean orogen,Geology, 25(5), 415 – 418, doi:10.1130/0091-7613(1997)025<0415:DAKPFD>2.3.CO;2.

Mattinson, J. M. (1994), Uranium decay constantuncertainties and their implications for high-resolution U-Pb geochronology, Geol. Soc. Am.Abstr. Programs, 77, A-221.

Mattinson, J. M. (2000), Revising the ‘‘goldstandard’’ – the uranium decay constants of Jaffeyet al., 1971, Eos Trans. AGU, 81(19), Spring Meet.Suppl., Abstract V61A-02.

Mattinson, J. M. (2003), CA (chemical abrasion)-TIMS: High-resolution U-Pb zircon geochronologycombining high-temperature annealing of radiationdamage and multi-step partial dissolution analysis,Eos Trans. AGU, 85(47), Fall Meet. Suppl.,Abstract V22E-06.

Mattinson, J. M. (2005), Zircon U-Pb chemical-abrasion(‘‘CA-TIMS’’) method: Combined annealing andmulti-step dissolution analysis for improved preci-sion and accuracy of zircon ages, Chem. Geol.,220, 47–56, doi:10.1016/j.chemgeo.2005.03.011.

Meissner, R., andW.D.Mooney (1998),Weakness of thelower continental crust:Acondition for delamination,

uplift, and escape, Tectonophysics, 296(1 – 2), 47–60, doi:10.1016/S0040-1951(98)00136-X.

Miller, R. B., and S. R. Paterson (2001), Influence oflithological heterogeneity, mechanical anisotropy,and magmatism on the rheology of an arc, NorthCascades, Washington, Tectonophysics, 342, 351 –370, doi:10.1016/S0040-1951(01)00170-6.

Min, K., R. Mundil, P. R. Renne, and K. R. Ludwig(2000), A test for systematic errors in 40Ar/39Argeochronology through comparison with U-Pbanalysis of a 1.1 Ga rhyolite, Geochim. Cosmochim.A c t a , 64 , 7 3 – 98 , d o i : 1 0 . 1 0 16 / S 00 16 -7037(99)00204-5.

Morgan, P. (1985), Crustal radiogenic heat productionand the selective survival of ancient continentalcrust, Proc. Lunar Planet. Sci. Conf., 15th, Part 2,J. Geophys. Res., 90, suppl., C561–C570.

Moser, D. E., R. M. Flowers, and R. J. Hart (2001),Birth of the Kaapvaal tectosphere 3.08 billion yearsago, Science , 291 , 465 – 468, doi:10.1126/science.291.5503.465.

Moyen, J.-F., G. Stevens, and A. F. M. Kisters (2006),Record of mid-Archaean subduction from meta-morphism in the Barberton terrain, South Africa,Nature, 442, 559–562, doi:10.1038/nature04972.

Nguuri, T. K., J. Gore, D. E. James, S. J. Webb,C. Wright, T. G. Zengeni, O. Gwavava, J. A. Snoke,and Kaapvaal Seismic Group (2001), Crustal struc-ture beneath southern Africa and its implications forthe formation and evolution of the Kaapvaal andZimbabwe cratons, Geophys. Res. Lett., 28, 2501–2504, doi:10.1029/2000GL012587.

Nicolaysen, L. O., R. J. Hart, and N. H. Gale (1981),Vredefort radioelement profile extended to supra-crustal strata at Carletonville, with implicationsfor continental heat flow, J. Geophys. Res., 86,10,653–10,662, doi:10.1029/JB086iB11p10653.

Parman, S. W., T. L. Grove, J. C. Dann, and M. J. deWit (2004), A subduction origin for komatiites andcratonic lithospheric mantle, S. Afr. J. Geol., 107,107–118, doi:10.2113/107.1-2.107.

Passchier, C. W. (1998), Monoclinic model shear zones,J. Struct. Geol., 20(8), 1121–1137, doi:10.1016/S0191-8141(98)00046-7.

Pavlis, T. L. (1996), Fabric development in syn-tectonicintrusive sheets as a consequence of melt-dominated flow and thermal softening of the crust,Tectonophysics, 253, 1 – 31, doi:10.1016/0040-1951(95)00049-6.

Pearson, D. G., R. W. Carlson, S. B. Shirey, F. R. Boyd,and P. H. Nixon (1995), Stabilization of Archeanlithospheric mantle: A Re-Os isotope study of peri-dotite xenoliths from the Kaapvaal craton, EarthPlanet. Sci. Lett., 134, 341 – 357, doi:10.1016/0012-821X(95)00125-V.

Perry, H. K. C., D. W. S. Eaton, and A. M. Forte(2002), LITH5.0: A revised crustal model forCanada based on Lithoprobe results, Geophys.J. Int. , 150, 285 – 294, doi:10.1046/j.1365-246X.2002.01712.x.

Perry, H. K. C., C. Jaupart, J.-C. Mareschal, andG. Bienfait (2006), Crustal heat production in theSuperior Province, Canadian Shield, and in NorthAmerica inferred from heat flow data, J. Geophys.Res., 111, B04401, doi:10.1029/2005JB003893.

Ranalli, G. (1995), Rheology of the Earth, 413 pp.,Chapman and Hall, New York.

Ranalli, G., andD. C.Murphy (1987), Rheological strati-ficationof the lithosphere,Tectonophysics,132, 281–295, doi:10.1016/0040-1951(87)90348-9.

Renne, P. R., D. B. Karner, and K. R. Ludwig (1998),Absolute ages aren’t exactly, Science, 282, 1840–1841, doi:10.1126/science.282.5395.1840.

Richardson, S. H., J. J. Gurney, A. J. Erlank, and J. W.Harris (1984), Origin of diamonds in old enrichedmantle, Nature, 310, 198 – 202, doi:10.1038/310198a0.

Richardson, S. H., S. B. Shirey, J. W. Harris, and R. W.Carlson (2001), Archean subduction recorded byRe-Os isotopes in eclogitic sulfide inclusions inKimberly diamonds, Earth Planet. Sci. Lett., 191,257–266, doi:10.1016/S0012-821X(01)00419-8.

Ritsema, J., A. A. Nyblade, T. J. Owens, C. A. Langston,and J. C. VanDecar (1998), Upper mantle seismicvelocity structure beneath Tanzania, east Africa:Implications for the stability of cratonic lithosphere,J. Geophys. Res., 103, 21,201–21,213, doi:10.1029/98JB01274.

Robb, L. J., and C. R. Anhaeusser (1983), Chemicaland petrogenetic characteristics of the Archaeantonalite-trondhjemite gneiss plutons in theBarberton Mountain Land, in Contributions to theGeology of the Barberton Mountain Land, editedby C. R. Anhaeusser, Spec. Publ. Geol. Soc. S.Afr., 9, 103 –116.

Robb, L. J., D. W. Davis, S. L. Kamo, and F. M. Meyer(1992), Ages of altered granites adjoining theWitwatersrand Basin with implications for theorigin of gold and uranium, Nature, 357, 677 –680, doi:10.1038/357677a0.

Rudnick, R. L., and A. A. Nyblade (1999), Thethickness and heat production of Archean litho-sphere: Constraints from xenolith thermobarometryand surface heat flow, in Mantle Petrology: FieldObservations and High Pressure Experimenta-tion: A Tribute to Francis R. (Joe), edited byY. Fei, C. Bertka, and B. O. Mysen, Spec. Publ.Geochem. Soc., 6, 3 –12.

Saltzer, S. D., and K. V. Hodges (1988), The MiddleMountain shear zone, southern Idaho: Kinematicanalysis of an early Tertiary high-temperaturedetachment, Geol. Soc. Am. Bull., 100, 96 –103.

Sandiford, M., and S. McLaren (2002), Tectonic feed-back and the ordering of heat producing elementswithin the continental lithosphere, Earth Planet.Sci. Lett., 204, 133 – 150, doi:10.1016/S0012-821X(02)00958-5.

Sandiford, M., S. McLaren, and N. Neumann (2002),Long-term thermal consequences of the redistribu-tion of heat-producing elements associated withlarge-scale granitic complexes, J. Metamorph.

Geol., 20, 87 – 98, doi:10.1046/j.0263-4929.2001.00359.x.

Sandiford, M., M. J. Van Kranendonk, and S. Bodorkos(2004), Conductive incubation and the origin ofdome-and-keel structure in Archean granite-green-stone terrains: A model based on the eastern PilbaraCraton, Western Australia, Tectonics, 23, TC1009,doi:10.1029/2002TC001452.

Schmitz, M. D., and S. A. Bowring (2003a), constraintson the thermal evolution of continental lithospherefrom U-Pb accessory mineral thermochronometryof lower crustal xenoliths, southern Africa, Contrib.Mineral. Petrol., 144, 592–618.

Schmitz, M. D., and S. A. Bowring (2003b), Ultrahigh-temperature metamorphism in the lower crustduring Neoarchean Ventersdorp rifting and magma-tism, Kaapvaal Craton, southern Africa, Geol. Soc.Am. Bull., 115, 533 –548.

Schmitz, M. D., S. A. Bowring, M. J. de Wit, andV. Gartz (2004), Subduction and terrane collisionstabilized the western Kaapvaal craton tectosphere2.9 billion years ago, Earth Planet. Sci. Lett., 222,363–376, doi:10.1016/j.epsl.2004.03.036.

Schoene, B., and S. A. Bowring (2004), Rates andmechanisms of magma emplacement in the Usutuintrusive suite: Implications for ca. 3.2 Ga assemblyof the Kaapvaal craton, Geol. Soc. Am. Abstr. Pro-grams, 36, 46.

Schoene, B., and S. A. Bowring (2007), Determiningaccurate temperature-time paths in U-Pb thermo-chronology: An example from the SE Kaapvaalcraton, southern Africa, Geochim. Cosmochim.

Acta, 71, 165–185, doi:10.1016/j.gca.2006.08.029.Schoene, B., J. L. Crowley, D. C. Condon, M. D.

Schmitz, and S. A. Bowring (2006), Reassessingthe uranium decay constants for geochronologyusing ID-TIMS U-Pb data, Geochim. Cosmochim.Acta, 70, 426–445, doi:10.1016/j.gca.2005.09.007.

Shirey, S. B., R. W. Carlson, S. H. Richardson, A. H.Menzies, J. J. Gurney, D. G. Pearson, J. W. Harris,and U. Wiechert (2001), Archean emplacement ofeclogitic components into the lithospheric mantleduring formation of the Kaapvaal Craton, Geophys.


26 of 27

TC5010

Res. Let t . , 28 , 2509 – 2512, doi :10.1029/2000GL012589.

Shirey, S. B., J. W. Harris, S. H. Richardson, M. J.Fouch, D. E. James, P. Cartigny, P. Deines,and F. Viljoen (2002), Diamond genesis, seismicstructure, and the evolution of the Kaapvaal-Zimbabwe craton, Science, 297, 1683 – 1686,doi:10.1126/science.1072384.

Simons, F. J., A. Zielhuis, and R. D. van der Hilst(1999), The deep structure of the Australian conti-nent from surface wave tomography, Lithos, 48,17– 43, doi:10.1016/S0024-4937(99)00041-9.

Stevens, G., G. T. R. Droop, R. A. Armstrong, and C. R.Anhaeusser (2002), Amphibolite facies metamorph-ism in the Schapenburg schist belt: A record of themid-crustal response to�3.23Ga terrane accretion inthe Barberton greenstone belt, S. Afr. J. Geol., 105,271–284, doi:10.2113/1050271.

Strik, G., M. J. de Wit, and C. G. Langereis (2007),Palaeomagnetism of the Neoarchaean Pongola andVentersdorp Supergroups and an appraisal of the3.0 – 1.9 Ga apparent polar wander path of theKaapvaal Craton, southern Africa, PrecambrianRes., 153(1 –2), 96 –115, doi:10.1016/j.precamres.2006.11.006.

Sullivan, W. A., and R. D. Law (2007), Deformationpath partitioning within the transpressional WhiteMountain shear zone, California and Nevada,J. Struct. Geol., 29, 583 – 598, doi:10.1016/j.jsg.2006.11.001.

Thomas, R. J., M. W. von Veh, and S. McCourt (1993),The tectonic evolution of southern Africa: An over-view, J. Afr. Earth Sci., 16, 5 – 24, doi:10.1016/0899-5362(93)90159-N.

Tikoff, B. (1997), Stretching lineations in transpressionalshear zones: And example from the Sierra NevadaBatholith, California, J. Struct. Geol., 19(1), 29–39,doi:10.1016/S0191-8141(96)00056-9.

Tinker, J., M. J. de Wit, and J. Grotzinger (2002),Seismic stratigraphic constraints on Neoarchean-Paleoproterozoic evolution of the Western Marginof the Kaapvaal Craton, South Africa, S. Afr.J. Geol., 105, 107–134, doi:10.2113/105.2.107.

Van Kranendonk, M. J., H. Smithies, and V. Bennett(Eds.) (2007), Earth’s Oldest Rocks, Dev. Precam-brian Geol., vol. 15, 1330 pp., Elsevier, New York.

Viljoen, M. J., and R. P. Viljoen (1969), An introduc-tion to the geology of the Barberton, granite-greenstone terrain, Spec. Publ. Geol. Soc. S. Afr.,9, 1 –20.

Walcott, R. I. (1998), Modes of oblique compression:Late Cenozoic tectonics of the South Island of NewZealand, Rev. Geophys., 36, 1 – 26.

Westerlund, K. J., S. B. Shirey, S. H. Richardson, R. W.Carlson, J. J. Gurney, and J. W. Harris (2006), Asubduction wedge origin for Paleoarchean peridoti-tic diamonds and harzburgites from the Pandakimberlite, Slave craton: Evidence from Re-Os iso-tope systematics, Contrib. Mineral. Petrol., 152,275–294, doi:10.1007/s00410-006-0101-8.

Westraat, J. D., A. F. M. Kisters, M. Poujol, andG. Stevens (2005), Transcurrent shearing, granitesheeting and the incremental construction of thetabular 3.1 Ga Mpuluzi batholith, Barbertongranite-greenstone terrane, South Africa, J. Geol.Soc., 162, 373–388, doi:10.1144/0016-764904-026.

Whitney, D. L., C. Teyssier, and C. S. Siddoway (Eds.)(2004), Gneiss Domes in Orogeny, Spec. Publ.Geol. Soc. Am., 380, 378 pp.

Williams, D. A. C., and R. G. Furnell (1979), Areassessment of part of the Barberton area, Precam-brian Res., 9, 325 – 347, doi:10.1016/0301-9268(79)90011-1.

Wilson, A. C. (1979), Geological map of Swaziland,1:50,000 series, Geol. Surv. and Mines Dep., Govt.of Swaziland, Mbabane.

Wilson, A. C. (1982), Geological map of Swaziland,1:250,000 series, Geol. Surv. and Mines Dep.,Govt. of Swaziland, Mbabane.

Xie, X., G. R. Byerly, and R. E. Ferrell (1997), IIbtrioctehedral chlorite from the Barberton greenstonebelt; crystal structure and rock compositionconstraints with implications to geothermometry,Contrib. Mineral. Petrol . , 126 , 275 – 291,doi:10.1007/s004100050250.

York, D. (1969), Least-squares fitting of a straight linewith correlated errors, Earth Planet. Sci. Lett., 5,320 –324.

Zhao, W., and W. J. P. Morgan (1987), Injection ofIndian crust into Tibetan lower crust; a two-dimensional finite element model study, Tectonics,6, 489 –504, doi:10.1029/TC006i004p00489.

��S. A. Bowring, Department of Earth, Atmospheric

and Planetary Sciences, Massachusetts Institute ofTechnology, Cambridge, MA 02139, USA.

M. J. de Wit, Department of Geological Sciences,University of Cape Town, Rondebosch, 7701, SouthAfrica.

B. Schoene, Departement de Mineralogie, Univer-site de Geneve, Geneve, CH-1205, Switzerland.([email protected])


27 of 27

TC5010

mesoarchean assembly and stabilization of the eastern kaapvaal

Documents