Fig 1 - uploaded by Nils Lenhardt
Content may be subject to copyright.
Source publication
The time of deposition of the Pretoria Group between 2.32 and 2.06 Ga on South Africa's Kaapvaal Craton was characterized by the first major increase in atmospheric oxygen. It was accompanied by the extrusion of significant thicknesses of volcanic deposits, namely the Bushy Bend lavas of the Timeball Hill Formation, the Hekpoort Formation and the M...
Contexts in source publication
Context 1
... in turn succeeded by black pyritic shales of the Lower Timeball Hill Formation (Eriksson et al., 1994a,b, 1995). In the eastern part of the Transvaal Basin, the Bushy Bend Member consists of a highly altered tuff, only a few metres thick (Reczko et al., 1995b). The 300–830 m thick Hekpoort Formation in the Transvaal basin crops out over an area of ca. 100,000 km 2 . Within the Griqualand West basin and as part of the Postmasburg Group, which is the western equivalent of the Pretoria Group (Button, 1973; Tankard et al., 1982; Beukes, 1983), is the Ongeluk Formation, correlated with the Hekpoort through similar ages: 2224 ± 21 Ma (whole-rock Rb–Sr isochron; Walraven and Martini, 1995) for the Hekpoort Formation and 2222 ± 12 Ma (whole rock Pb–Pb isochron; Cornell et al., 1996) for the Ongeluk Formation, respectively. Together with the comagmatic Ongeluk Formation of the Griqualand West basin (Eriksson and Reczko, 1995) the tholeiitic andesites of the Hekpoort Formation are thought to have originally covered ca. 500,000 km 2 of the Kaapvaal Craton (Cornell et al., 1996). In most parts of the Transvaal basin, the Hekpoort Formation sharply overlies the subaerially emplaced Boshoek conglomerates and sandstones with a sharp contact (Cheney, 1996), and is unconformably succeeded by the Dwaalheuwel continental sandstones (Eriksson et al., 1993). In the central-southern parts of the basin the Hekpoort Formation sharply overlies mudrocks of the Timeball Hill Formation, and is similarly overlain by Strubenkop Formation mudrocks (Fig. 1 and Table 1). In the Griqualand West basin, the basaltic lava flows of the Ongeluk Formation overlie glacigenic rocks of the Makganyene Formation and are characterized by evidence for deposition in a subaqueous setting along the submerged western margin of the Kaapvaal Craton (Grobler and Botha, 1976; Gutzmer et al., 2003). The Hekpoort Formation dips gently to the N–NW at ca. 15 ◦ and is characterized by a complex interplay of subaerial lava flows, pyroclastic deposits, and their reworked counterparts (Oberholzer, 1995) representing deposition in a subaerial palaeoenvironment (Button, 1973; Res, 1993; Eriksson and Reczko, 1995; Oberholzer, 1995). Although there has been no regional study so far to estab- lish the proportions of lava flows and volcaniclastic detritus in the Hekpoort Formation, Sharpe et al. (1983) suggested that lava flows predominate. In the east of the preserved Transvaal basin, Button (1973) found that volcaniclastic rocks only made up approximately 10% of the formation, and that they were concentrated towards its base. In the central-southern part, however, Oberholzer and Eriksson (2000) found that volcaniclastic rocks present a significant portion of the lithology of the formation. The predominantly argillaceous Silverton Formation (Fig. 1) has a basin-wide extent and varies in total thickness from ca. 2000 m in the east of the Transvaal Basin to several hundred metres in the west (Button, 1973). Erosional removal characterizes most of the southern and central preserved occurrences (Eriksson et al., 2002). The Silverton Formation is characterized by various mudrocks (approximately 80%), locally significant volcanic rocks (the Machadodorp Member), and minor carbonates, cherts and sandstones (Schreiber, 1991). According to Catuneanu and Eriksson (1999) the Silverton shales reflect the transgressive systems tract of an epeiric sea. A lowermost, westerly arenaceous facies association is ascribed to a braid-deltaic and turbidity current deposition, whereas the predominant argillaceous deposits are interpreted as sub-storm wave base pelagic sediments (Eriksson et al., 2002, 2008). The lower contact of the formation is gradational, locally with mudstones interlayered with cm- to dm-thick beds of immature sandstone resting on the arenaceous Daspoort Formation (Fig. 3). The upper contact with the overlying arenaceous Magaliesberg Formation is gradational and upward-coarsening (Button, 1973; Schreiber, 1991; Van der Neut, 1990). Furthermore, gradational contacts between the lower Boven Shale Member, the medial Machadodorp Volcanic Member and the upper Lydenburg Shale Member are observed within the Silverton Formation. The three units average 250 m, 300 m, and 1250 m in thickness, respectively, the greatest thicknesses being encountered in the east of the Transvaal basin with the lavas being restricted to that portion of the depository (Schreiber, 1991; Eriksson et al., 2002). The up to 90 m thick Bushy Bend Lava Member is characterized by a sharp contact with the conglomerates of the underlying Rooihoogte Formation and a gradational upper contact with the Lower Shale Member of the Timeball Hill Formation. The lithology of the Bushy Bend Lava Member is dominated by lava with subordinate tuffs and clastic sedimentary rocks. The lavas vary from fine crystalline to amygdaloidal. Single lava flows with thicknesses varying between 0.3 and 12 m can be distinguished by their chilled and undulating flow bottoms (Eriksson et al., 1994a,b); rounded and deformed amygdales below and above the flow contacts enhance the discrimination of individual flows. Amygdales are filled with zeolite, chlorite, quartz and/or calcite (Eriksson et al., 1994b). Locally, flows are separated by beds of several metres-thick laminated tuffs with interbedded siltstone (see Fig. 3, borehole C). Phenocrysts consist of plagioclase and amphibole pseudomor- phously developed after clinopyroxene (Eriksson et al., 1994b), and the phenocrysts are set in a fine-grained matrix of similar material. The lavas are epidotized and sericitized, and in addition show extensive veining and/or brecciation with veinlets filled by calcite and, sporadically, quartz. The only geochemical data for the Bushy Bend lavas have been provided by Eriksson et al. (1994b), Reczko et al. (1995b) and Coetzee (2002). The first two research groups report high K contents that are interpreted as either high primary K contents or, alternatively, related to a syn- or post-extrusive phase of K- metasomatism. Coetzee (2002) classifies the Bushy Bend lavas by means of trace elements as subalkaline to alkaline basalts and favours post-exstrusive K-metasomatism as the cause for the high K content. In addition to the main occurrence of the predominant lavas near Potchefstroom (where they are only known from boreholes) several authors have described analogous volcanic rocks in other parts of the basin that are likely to be correlates of the Bushy Bend Lava Member. ...
Context 2
... the fountain, while the vesic- ulating lava was still slightly plastic (Mueller and White, 1992). The extensive distribution of the fluidal-clast breccia, at least 220 km along strike, and indicating a close proximity to a volcanic source (e.g., Kokelaar and Durant, 1983; Kokelaar, 1986), suggests the presence of several vents between Carolina in the South and Burgersfort in the North. In subaerial settings, fountaining commonly occurs at vents along fissures (Macdonald, 1972; Wilson and Head, 1981). A similar scenario may be expected in subaqueous settings (Simpson and McPhie, 2001; Scott et al., 2002). The eruption must have been sufficiently strong to create a steam envelope in which ballistic material had time to be deposited, but not voluminous enough to develop thick pyroclastic flow deposits with heat retention structures and welding (Mueller et al., 2000). The stratified lapilli tuff exhibits a change in the fragmentation process from magmatic to hydrovolcanic, attributable to significant ingress of water into the eruption column. They probably formed due to the collapse of a fountaining subaqeous eruption and its steam envelope, either complete or only at the margins, producing hot, high-concentration density currents (c.f., Lowe, 1982; White, 2000). The overlying sheet lavas most likely resulted from effusion that accompanied fountaining and may indicate fluctuations in the magma discharge rate (Simpson and McPhie, 2001). In particular, relative to fountaining episodes, the lavas were probably generated during periods of reduced discharge (e.g., Griffiths and Fink, 1992). An alternative explanation is given by Head and Wilson (2003) who propose that the volatile build-up at the top of the reservoir would leave a complementary volatile-depleted magma below. Thus, after the volatile-rich layer was discharged during the hawaiian-style eruption event, it could be followed by a very volatile-depleted effusive phase, and the vesicle-poor lavas overlying pyroclastic-rich cones might therefore be a distinctive signature of this eruption setting. The Machadodorp lavas are unlikely to be fountain-fed lavas, given the non-welded character of the associated fluidal-clast breccia (cf., Simpson and McPhie, 2001). The dominance of sheet lavas over pillow lavas affirms the prediction that the relatively high effusion rates associated with hawaiian-type eruptions would lead to lava flows that would be characterized by lobate sheets, rather than pillows (e.g., Head et al., 1996; Gregg and Fink, 1995; Head and Wilson, 2003). In general, the Machadodorp Volcanic Member can thus be interpreted as the deposits of several seamounts (e.g., Keating et al., 1987; Wessell and Lyons, 1997; Schmidt and Schmincke, 2000) aligned along a fissure, probably within an extensional environment (Button, 1973; Reczko et al., 1995b) (Fig. 9). The volcanic facies constituting seamounts often overlie deep water sediments and/or are interstratified with sedimentary material deposited as suspension fallout during volcanism (Fisher, 1984). According to Head and Wilson (2003) landforms anticipated from these eruptions might include cones surrounding the vent with rim crests within a few metres of the vent, possible lava ponds within the cone, and an apron of pyroclastic deposits surrounding the vent. The cone and the flanking deposits should consist of interlayers of pyroclastic flows and lava flows with sheet flow morphology, rather than pillow lava morphology dominating. At greater radial distances from the vent, one would predict that flows and layers of agglutinated pyroclasts would dominate proximally in the cone, giving way to bedded pyroclastics and interlayered lava flows distally. Seamounts, although primarily associated with divergent plate boundaries (e.g. Buck et al., 1998; Macdonald, 1998; Perfit and Chadwick, 1998; Head et al., 1996), have also been related to back-arc, arc, and hot spot volcanism (Corcoran, 2000). These subaqueous features can vary from 0.05 to 10 km in thickness and attain diameters as large as 100 km (Corcoran, 2000). Their depths are generally recognized to be approximately 200–1000 m and less, depending on magma composition and volatile content (e.g., Kokelaar, 1986; Bonatti and Harrison, 1988; Gill et al., 1990; Oshima et al., 1991; Heikinian et al., 1991; Binard et al., 1992; Wright, 1996, 1999; White, 1996; Kano, 1998; Fiske et al., 1998, 2001; Hunns and McPhie, 1999). The sporadic tuffs in the west of the basin contrast with the eastern Machadodorp Lava Member (Eriksson et al., 1998) and most certainly did not arise from the same eruption mechanisms. However, to date they have not been studied in detail. The lithostratigraphy, interpreted depositional environments and sequence stratigraphic framework proposed for the Pretoria Group have been investigated in detail in published work (e.g., Eriksson et al., 1991, 2001, 2006; Eriksson and Reczko, 1995; Catuneanu and Eriksson, 1999; Moore et al., 2001) (Fig. 1). Geody- namic control inferred for the evolution of the Pretoria Group basin encompasses two cycles of prerift uplift – subsequent mechanical rifting – long lived thermal subsidence; in this model, two thick predominantly argillaceous successions ascribed to epeiric marine deposition (respectively, the Timeball Hill and younger Silverton Formations) are separated by thinner fluvial (alluvial) deposits and volcanic units related to the uplift and rifting phases (Catuneanu and Eriksson, 1999; Eriksson et al., 2002, 2008) (Fig. 1). Glacia- tion during Pretoria Group sedimentation is inferred to have been related to global palaeo-atmospheric and geodynamic events and shows no obvious relationship with the evolution of the basin itself (Eriksson et al., 2001). Eriksson et al. (2001) suggested that a generally weak spa- tial relationship of the lower Pretoria rifting cycle as evidence for volcanism supported a role for cratonic-scale plate tectonic processes in the evolution of this first basinal geodynamic cycle. In this paper, this postulate can be expanded upon: the limited extent of the Bushy Bend lavas in the south of the Pretoria depository and the relatively poorly developed and discontinu- ous, mostly tuffaceous lithologies at the equivalent stratigraphic level around the rest of the basin, attest to the likelihood that volcanism accompanying the first rifting event was short-lived, with both subaerial and subaqueous eruptions scattered around the preserved basin margin. Basaltic–andesitic fissure eruptions accompanied by fumarolic activity are thus inferred within the basin at the Rooihoogte–Timeball Hill transition, with limited subaerial explosive eruptions probably having occurred in the basin’s hinterland. The limited scope of volcanism accompanying the rifting within the first cycle of Pretoria basin evolution is thus confirmed in this paper. The Kaapvaal craton at this time in its evolution has no known plate tectonic associations with larger supercontinental plates nor are plate collisions known which might have enabled rifting related to plate tectonic processes in general (e.g., Eriksson et al., 2011). The cause of this first rifting cycle might thus indeed lie in sub-cratonic magmatic–thermal processes, even though the surficial volcanic expression thereof was limited. In contrast, the second rifting cycle postulated for the Pretoria Group basin has a strong association with widespread and large scale volcanism of the Hekpoort–Ongeluk flood basalts, which might well have been plume-related (e.g., Cornell and Schütte, 1995; Reczko et al., 1995b; Eriksson et al., 2001, 2006). Reczko et al. (1995b) have argued for a model encompassing genetic processes within a replenished, fractionated, tapped, assimilated (RFTA; Arndt et al., 1993) magma chamber, due to the geochemistry and preserved geometry of the inferred comagmatic Hekpoort–Ongeluk volcanics. Ongoing debate on the chronology of the two flood basalts (e.g., Moore et al., 2001; current paper under review by these authors), one (Hekpoort) subaerial and the other (Ongeluk) demonstrably essentially subaqueous might place the latter in a stratigraphic position preceding Hekpoort eruption. However, any such debate aside, a major and widespread plume- related volcanic episode would still remain related to this second cycle of rifting in Pretoria Group evolution. The scale of this second cycle volcanism and its importance in influencing the upper part of the Pretoria Group basin-fill (Daspoort-Silverton-Magaliesberg epeiric sea succession; cf. Eriksson et al., 2002, 2008) is underlined by the Machadodorp volcanism, which is related in this paper to extension, the development of an extensive fissure and a set of inferred seamounts, probably related to hot spot volcanism as the Ongeluk–Hekpoort plume possibly waned. Globally, the period ca. 2.5–2.0 Ga is seen by some work- ers to have been characterized by the occurrence of ultramafic intrusions, dyke swarms, and layered mafic complexes, indicating a continental break-up (Heaman, 1997). This viewpoint is one coloured by thinking related to the inferred extended breakup of the Kenorland supercontinent from ca. 2.45 to 2.2 Ga (e.g. Aspler and Chiarenzelli, 1998); however, supercontinentality should not be seen as a universally applicable state of the Palaeoproterozoic Earth (e.g., Eriksson et al., 2011). In addition, Condie et al. (2009) argue in favour of a global-scale magmatic shutdown in the ca. 2.45–2.2 Ga period, which provides an alternative model for the apparently extended breakup of Kenorland. That the deposition of the Pretoria Group sediments was accompanied by the extrusion of significant thicknesses of volcanic rocks in this same time period, also suggesting that rifting was active at this time (Eriksson and Reczko, 1995), stresses its global importance against this back- ground of variable views of Earth’s geodynamic evolution in the Palaeoproterozoic. Although the ...
Context 3
... expected in subaqueous settings (Simpson and McPhie, 2001; Scott et al., 2002). The eruption must have been sufficiently strong to create a steam envelope in which ballistic material had time to be deposited, but not voluminous enough to develop thick pyroclastic flow deposits with heat retention structures and welding (Mueller et al., 2000). The stratified lapilli tuff exhibits a change in the fragmentation process from magmatic to hydrovolcanic, attributable to significant ingress of water into the eruption column. They probably formed due to the collapse of a fountaining subaqeous eruption and its steam envelope, either complete or only at the margins, producing hot, high-concentration density currents (c.f., Lowe, 1982; White, 2000). The overlying sheet lavas most likely resulted from effusion that accompanied fountaining and may indicate fluctuations in the magma discharge rate (Simpson and McPhie, 2001). In particular, relative to fountaining episodes, the lavas were probably generated during periods of reduced discharge (e.g., Griffiths and Fink, 1992). An alternative explanation is given by Head and Wilson (2003) who propose that the volatile build-up at the top of the reservoir would leave a complementary volatile-depleted magma below. Thus, after the volatile-rich layer was discharged during the hawaiian-style eruption event, it could be followed by a very volatile-depleted effusive phase, and the vesicle-poor lavas overlying pyroclastic-rich cones might therefore be a distinctive signature of this eruption setting. The Machadodorp lavas are unlikely to be fountain-fed lavas, given the non-welded character of the associated fluidal-clast breccia (cf., Simpson and McPhie, 2001). The dominance of sheet lavas over pillow lavas affirms the prediction that the relatively high effusion rates associated with hawaiian-type eruptions would lead to lava flows that would be characterized by lobate sheets, rather than pillows (e.g., Head et al., 1996; Gregg and Fink, 1995; Head and Wilson, 2003). In general, the Machadodorp Volcanic Member can thus be interpreted as the deposits of several seamounts (e.g., Keating et al., 1987; Wessell and Lyons, 1997; Schmidt and Schmincke, 2000) aligned along a fissure, probably within an extensional environment (Button, 1973; Reczko et al., 1995b) (Fig. 9). The volcanic facies constituting seamounts often overlie deep water sediments and/or are interstratified with sedimentary material deposited as suspension fallout during volcanism (Fisher, 1984). According to Head and Wilson (2003) landforms anticipated from these eruptions might include cones surrounding the vent with rim crests within a few metres of the vent, possible lava ponds within the cone, and an apron of pyroclastic deposits surrounding the vent. The cone and the flanking deposits should consist of interlayers of pyroclastic flows and lava flows with sheet flow morphology, rather than pillow lava morphology dominating. At greater radial distances from the vent, one would predict that flows and layers of agglutinated pyroclasts would dominate proximally in the cone, giving way to bedded pyroclastics and interlayered lava flows distally. Seamounts, although primarily associated with divergent plate boundaries (e.g. Buck et al., 1998; Macdonald, 1998; Perfit and Chadwick, 1998; Head et al., 1996), have also been related to back-arc, arc, and hot spot volcanism (Corcoran, 2000). These subaqueous features can vary from 0.05 to 10 km in thickness and attain diameters as large as 100 km (Corcoran, 2000). Their depths are generally recognized to be approximately 200–1000 m and less, depending on magma composition and volatile content (e.g., Kokelaar, 1986; Bonatti and Harrison, 1988; Gill et al., 1990; Oshima et al., 1991; Heikinian et al., 1991; Binard et al., 1992; Wright, 1996, 1999; White, 1996; Kano, 1998; Fiske et al., 1998, 2001; Hunns and McPhie, 1999). The sporadic tuffs in the west of the basin contrast with the eastern Machadodorp Lava Member (Eriksson et al., 1998) and most certainly did not arise from the same eruption mechanisms. However, to date they have not been studied in detail. The lithostratigraphy, interpreted depositional environments and sequence stratigraphic framework proposed for the Pretoria Group have been investigated in detail in published work (e.g., Eriksson et al., 1991, 2001, 2006; Eriksson and Reczko, 1995; Catuneanu and Eriksson, 1999; Moore et al., 2001) (Fig. 1). Geody- namic control inferred for the evolution of the Pretoria Group basin encompasses two cycles of prerift uplift – subsequent mechanical rifting – long lived thermal subsidence; in this model, two thick predominantly argillaceous successions ascribed to epeiric marine deposition (respectively, the Timeball Hill and younger Silverton Formations) are separated by thinner fluvial (alluvial) deposits and volcanic units related to the uplift and rifting phases (Catuneanu and Eriksson, 1999; Eriksson et al., 2002, 2008) (Fig. 1). Glacia- tion during Pretoria Group sedimentation is inferred to have been related to global palaeo-atmospheric and geodynamic events and shows no obvious relationship with the evolution of the basin itself (Eriksson et al., 2001). Eriksson et al. (2001) suggested that a generally weak spa- tial relationship of the lower Pretoria rifting cycle as evidence for volcanism supported a role for cratonic-scale plate tectonic processes in the evolution of this first basinal geodynamic cycle. In this paper, this postulate can be expanded upon: the limited extent of the Bushy Bend lavas in the south of the Pretoria depository and the relatively poorly developed and discontinu- ous, mostly tuffaceous lithologies at the equivalent stratigraphic level around the rest of the basin, attest to the likelihood that volcanism accompanying the first rifting event was short-lived, with both subaerial and subaqueous eruptions scattered around the preserved basin margin. Basaltic–andesitic fissure eruptions accompanied by fumarolic activity are thus inferred within the basin at the Rooihoogte–Timeball Hill transition, with limited subaerial explosive eruptions probably having occurred in the basin’s hinterland. The limited scope of volcanism accompanying the rifting within the first cycle of Pretoria basin evolution is thus confirmed in this paper. The Kaapvaal craton at this time in its evolution has no known plate tectonic associations with larger supercontinental plates nor are plate collisions known which might have enabled rifting related to plate tectonic processes in general (e.g., Eriksson et al., 2011). The cause of this first rifting cycle might thus indeed lie in sub-cratonic magmatic–thermal processes, even though the surficial volcanic expression thereof was limited. In contrast, the second rifting cycle postulated for the Pretoria Group basin has a strong association with widespread and large scale volcanism of the Hekpoort–Ongeluk flood basalts, which might well have been plume-related (e.g., Cornell and Schütte, 1995; Reczko et al., 1995b; Eriksson et al., 2001, 2006). Reczko et al. (1995b) have argued for a model encompassing genetic processes within a replenished, fractionated, tapped, assimilated (RFTA; Arndt et al., 1993) magma chamber, due to the geochemistry and preserved geometry of the inferred comagmatic Hekpoort–Ongeluk volcanics. Ongoing debate on the chronology of the two flood basalts (e.g., Moore et al., 2001; current paper under review by these authors), one (Hekpoort) subaerial and the other (Ongeluk) demonstrably essentially subaqueous might place the latter in a stratigraphic position preceding Hekpoort eruption. However, any such debate aside, a major and widespread plume- related volcanic episode would still remain related to this second cycle of rifting in Pretoria Group evolution. The scale of this second cycle volcanism and its importance in influencing the upper part of the Pretoria Group basin-fill (Daspoort-Silverton-Magaliesberg epeiric sea succession; cf. Eriksson et al., 2002, 2008) is underlined by the Machadodorp volcanism, which is related in this paper to extension, the development of an extensive fissure and a set of inferred seamounts, probably related to hot spot volcanism as the Ongeluk–Hekpoort plume possibly waned. Globally, the period ca. 2.5–2.0 Ga is seen by some work- ers to have been characterized by the occurrence of ultramafic intrusions, dyke swarms, and layered mafic complexes, indicating a continental break-up (Heaman, 1997). This viewpoint is one coloured by thinking related to the inferred extended breakup of the Kenorland supercontinent from ca. 2.45 to 2.2 Ga (e.g. Aspler and Chiarenzelli, 1998); however, supercontinentality should not be seen as a universally applicable state of the Palaeoproterozoic Earth (e.g., Eriksson et al., 2011). In addition, Condie et al. (2009) argue in favour of a global-scale magmatic shutdown in the ca. 2.45–2.2 Ga period, which provides an alternative model for the apparently extended breakup of Kenorland. That the deposition of the Pretoria Group sediments was accompanied by the extrusion of significant thicknesses of volcanic rocks in this same time period, also suggesting that rifting was active at this time (Eriksson and Reczko, 1995), stresses its global importance against this back- ground of variable views of Earth’s geodynamic evolution in the Palaeoproterozoic. Although the syn-Transvaal Supergroup deformation on the Kaapvaal Craton is generally regarded by structural geologists as being extensional, producing small, deep pull-apart basins which acted as depositories for the Transvaal Supergroup (e.g. Eriksson et al., 1996; Bumby et al., 1998), the sheet-like nature of the entire volcano-sedimentary pile is more indicative of thermal subsidence, rather than deposition in a fault-bounded trough. Therefore, it seems more likely that the actual process of syn-Pretoria Group subsidence was transitional between ...
Context 4
... in turn succeeded by black pyritic shales of the Lower Timeball Hill Formation (Eriksson et al., 1994a,b, 1995). In the eastern part of the Transvaal Basin, the Bushy Bend Member consists of a highly altered tuff, only a few metres thick (Reczko et al., 1995b). The 300–830 m thick Hekpoort Formation in the Transvaal basin crops out over an area of ca. 100,000 km 2 . Within the Griqualand West basin and as part of the Postmasburg Group, which is the western equivalent of the Pretoria Group (Button, 1973; Tankard et al., 1982; Beukes, 1983), is the Ongeluk Formation, correlated with the Hekpoort through similar ages: 2224 ± 21 Ma (whole-rock Rb–Sr isochron; Walraven and Martini, 1995) for the Hekpoort Formation and 2222 ± 12 Ma (whole rock Pb–Pb isochron; Cornell et al., 1996) for the Ongeluk Formation, respectively. Together with the comagmatic Ongeluk Formation of the Griqualand West basin (Eriksson and Reczko, 1995) the tholeiitic andesites of the Hekpoort Formation are thought to have originally covered ca. 500,000 km 2 of the Kaapvaal Craton (Cornell et al., 1996). In most parts of the Transvaal basin, the Hekpoort Formation sharply overlies the subaerially emplaced Boshoek conglomerates and sandstones with a sharp contact (Cheney, 1996), and is unconformably succeeded by the Dwaalheuwel continental sandstones (Eriksson et al., 1993). In the central-southern parts of the basin the Hekpoort Formation sharply overlies mudrocks of the Timeball Hill Formation, and is similarly overlain by Strubenkop Formation mudrocks (Fig. 1 and Table 1). In the Griqualand West basin, the basaltic lava flows of the Ongeluk Formation overlie glacigenic rocks of the Makganyene Formation and are characterized by evidence for deposition in a subaqueous setting along the submerged western margin of the Kaapvaal Craton (Grobler and Botha, 1976; Gutzmer et al., 2003). The Hekpoort Formation dips gently to the N–NW at ca. 15 ◦ and is characterized by a complex interplay of subaerial lava flows, pyroclastic deposits, and their reworked counterparts (Oberholzer, 1995) representing deposition in a subaerial palaeoenvironment (Button, 1973; Res, 1993; Eriksson and Reczko, 1995; Oberholzer, 1995). Although there has been no regional study so far to estab- lish the proportions of lava flows and volcaniclastic detritus in the Hekpoort Formation, Sharpe et al. (1983) suggested that lava flows predominate. In the east of the preserved Transvaal basin, Button (1973) found that volcaniclastic rocks only made up approximately 10% of the formation, and that they were concentrated towards its base. In the central-southern part, however, Oberholzer and Eriksson (2000) found that volcaniclastic rocks present a significant portion of the lithology of the formation. The predominantly argillaceous Silverton Formation (Fig. 1) has a basin-wide extent and varies in total thickness from ca. 2000 m in the east of the Transvaal Basin to several hundred metres in the west (Button, 1973). Erosional removal characterizes most of the southern and central preserved occurrences (Eriksson et al., 2002). The Silverton Formation is characterized by various mudrocks (approximately 80%), locally significant volcanic rocks (the Machadodorp Member), and minor carbonates, cherts and sandstones (Schreiber, 1991). According to Catuneanu and Eriksson (1999) the Silverton shales reflect the transgressive systems tract of an epeiric sea. A lowermost, westerly arenaceous facies association is ascribed to a braid-deltaic and turbidity current deposition, whereas the predominant argillaceous deposits are interpreted as sub-storm wave base pelagic sediments (Eriksson et al., 2002, 2008). The lower contact of the formation is gradational, locally with mudstones interlayered with cm- to dm-thick beds of immature sandstone resting on the arenaceous Daspoort Formation (Fig. 3). The upper contact with the overlying arenaceous Magaliesberg Formation is gradational and upward-coarsening (Button, 1973; Schreiber, 1991; Van der Neut, 1990). Furthermore, gradational contacts between the lower Boven Shale Member, the medial Machadodorp Volcanic Member and the upper Lydenburg Shale Member are observed within the Silverton Formation. The three units average 250 m, 300 m, and 1250 m in thickness, respectively, the greatest thicknesses being encountered in the east of the Transvaal basin with the lavas being restricted to that portion of the depository (Schreiber, 1991; Eriksson et al., 2002). The up to 90 m thick Bushy Bend Lava Member is characterized by a sharp contact with the conglomerates of the underlying Rooihoogte Formation and a gradational upper contact with the Lower Shale Member of the Timeball Hill Formation. The lithology of the Bushy Bend Lava Member is dominated by lava with subordinate tuffs and clastic sedimentary rocks. The lavas vary from fine crystalline to amygdaloidal. Single lava flows with thicknesses varying between 0.3 and 12 m can be distinguished by their chilled and undulating flow bottoms (Eriksson et al., 1994a,b); rounded and deformed amygdales below and above the flow contacts enhance the discrimination of individual flows. Amygdales are filled with zeolite, chlorite, quartz and/or calcite (Eriksson et al., 1994b). Locally, flows are separated by beds of several metres-thick laminated tuffs with interbedded siltstone (see Fig. 3, borehole C). Phenocrysts consist of plagioclase and amphibole pseudomor- phously developed after clinopyroxene (Eriksson et al., 1994b), and the phenocrysts are set in a fine-grained matrix of similar material. The lavas are epidotized and sericitized, and in addition show extensive veining and/or brecciation with veinlets filled by calcite and, sporadically, quartz. The only geochemical data for the Bushy Bend lavas have been provided by Eriksson et al. (1994b), Reczko et al. (1995b) and Coetzee (2002). The first two research groups report high K contents that are interpreted as either high primary K contents or, alternatively, related to a syn- or post-extrusive phase of K- metasomatism. Coetzee (2002) classifies the Bushy Bend lavas by means of trace elements as subalkaline to alkaline basalts and favours post-exstrusive K-metasomatism as the cause for the high K content. In addition to the main occurrence of the predominant lavas near Potchefstroom (where they are only known from boreholes) several authors have described analogous volcanic rocks in other parts of the basin that are likely to be correlates of the Bushy Bend Lava Member. Key (1983) found a 1 m-thick pyritic bed in the lower Timeball Hill Formation in eastern Botswana, where the sulphides are associated with feldspathic tuffs. Klop (1978) found a chert interbed in the Gopane area of the southwestern Transvaal basin in the basal pyritic black shales of the Timeball Hill Formation, which contains possible altered glass shards. In the Pretoria region, Van der Neut (1990) described a mudclast conglomerate immediately underlying the Timeball Hill black shales, which he interpreted as reworked and weathered products of basaltic rocks by means of XRF and XRD analyses. In the east of the Timeball Hill basin, a several dm-thick weathered tuff occurs within the basal black shales, and extends for approximately 60 km along strike, north of Carolina (Eriksson et al., 1994b; Reczko et al., 1995b). The Hekpoort/Ongeluk volcanism marks a major volcanic event within the Transvaal Supergroup. Although limited regional data (Button, 1973; Sharpe et al., 1983; Engelbrecht, 1986) indicate that volcaniclastic rocks within the Hekpoort Formation are subordinate to basaltic–andesitic lava flows, in the preserved Transvaal Basin (TB) the work of Oberholzer (1995) and Oberholzer and Eriksson (2000) shows that the formation in the southwestern part of the ...
Context 5
... deposition of the Pretoria Group (Transvaal Supergroup) on South Africa’s Kaapvaal Craton marks a radical change in environment during the Palaeoproterozoic (Eriksson et al., 2006). What was initially a shallowly submerged epeiric continental platform on which cyanobacteria flourished and formed the thick stromatolitic carbonate beds of the underlying Chuniespoort Group, became covered by large volumes of Pretoria Group clastic sediment. After a period of uplift (80 My according to Eriksson et al. (2001), possibly up to ca. 200 My according to Mapeo et al., 2006) following the end of chemical (carbonate and BIF) sedimentation, a clastic epi- continental sea formed on the craton, into which sediments from various sources, mainly of Palaeoproterozoic age with minor inputs from Archaean provenances (Mapeo et al., 2006) were deposited. Deposition of clastic sediment was accompanied by the extrusion of significant thicknesses of volcanic deposits, thereby largely hinder- ing the further growth of stromatolites. Studies on pyrites in marine shales of the Rooihoogte and Timeball Hill Formations at the base of the Pretoria Group suggest that the first major increase in global atmospheric oxygen may have occurred during the deposition of these formations (Bekker et al., 2004). Alternating mudstone and quartzose sandstone formations characterize the volcano-sedimentary succession of the Pretoria Group (Schreiber, 1991), in which three main volcanic units are identified. These volcanic units comprise the Bushy Bend Lava Member of the Timeball Hill Formation, the Hekpoort Formation, and the Machadodorp Member of the Silverton Formation, in ascending stratigraphic order. In addition, thin tuff and tuffaceous shale beds are scattered throughout the Silverton and Vermont Formations, to a lesser extent in the Nederhorst and Houtenbek Formations, and are locally found at the base of the Magaliesberg Formation (Reczko, 1994). A generalized lithostratigraphic profile for the Pretoria Group is presented in Fig. 1. The Pretoria Group comprises 14 formations, starting with the basal Rooihoogte Formation and ending with the Houtenbek Formation at the top where it is unconformably capped by the Rooiberg Group lavas of the Bushveld Large Igneous Province (Lenhardt and Eriksson, 2012). The upper five formations are only preserved in the east of the Transvaal Basin, with a single western equivalent lacking any con- tinuous outcrops in between (Eriksson et al., 1998, 2001). The depositional age of the onset of Pretoria Group sedimentation is partially constrained by a Re/Os age of 2322 ± 16 Ma derived from mudstones from the lower part of the Timeball Hill Formation (Hannah et al., 2004) and a SHRIMP U–Pb detrital zircon age of 2324 ± 17 Ma from the Timeball Hill Formation (Dorland et al., 2004), near the base of the group. The timing of the cessation of Pretoria Group sedimentation is unknown as yet. A minimum age is given by the extrusion of the extensive Rooiberg Group lavas (2061 ± 2 Ma; Walraven, 1997) at the top of the Transvaal succession, that represent the oldest part of the Bushveld Large Igneous Province (Ernst and Bell, 2010). The depositional palaeoenvironment of the Pretoria Group is thought to have been either an epeiric marine setting (Willemse, 1959; Visser, 1969; Button, 1973, 1986; Button and Vos, 1977; Eriksson et al., 2001, 2006) or an intracratonic basin with short- lived marine incursions (e.g., Crockett, 1972). Thin, lenticular diamictites found within the Timeball Hill Formation (Eriksson and Altermann, in press) reflect fluvioglacial to predominantly glaciomarine periglacial deposits (Visser, 1971; De Villiers and Visser, 1977; Eriksson et al., 1993; Evans et al., 1997). The Palaeoproterozoic glaciation, however, is a complex issue with highly divergent views – such as those supporting the “Snowball Earth hypothesis” (e.g., Kirschvink, 1992; Hoffman et al., 1998) and those diametrically opposed to this (e.g., Williams, 2004; Young, 2004). It is beyond the purposes of this paper to become involved in a debate as yet essentially unresolved. Eriksson et al. (1991) proposed a continental rift tectonic setting with half-graben development for the Pretoria Group. This model encompasses a polycyclic rift stage, followed by thermal subsidence, with expansion of the basin during Silverton Formation times (Eriksson and Reczko, 1995). The rift hypothesis has been applied by Von Gruenewald and Harmer (1993) for the Pretoria Group volcanic rocks, which are thought to be precursors to the succeeding Bushveld Complex magmatism. More recently, a geodynamic model encompassing two second-order rifting-thermal subsidence cycles has been espoused (Catuneanu and Eriksson, 1999; Eriksson et al., 2001, 2006). This article briefly reviews the existing literature on the Pretoria Group volcanism in conjunction with new or as yet unpublished field and minor new geochemical data (a major basin-wide geochemical study of the entire Pretoria Group, including its volcanic units was performed by Reczko, 1994), and examines the emplacement of the three main volcanic units within the Pretoria Group; by studying the various eruptional, transport and depositional processes inferred, we will attempt an explanation for the tectonic control of the volcanism. This will be related to geodynamic models for Pretoria Group basin evolution. A hiatus of between 80 My (based on minimum age of Chuniespoort Group of 2432 ± 31 Ma (Trendall et al., 1990), and maximum age of Pretoria Group of 2316 ± 7 Ma (Hannah et al., 2004; Eriksson and Reczko, 1995; Eriksson et al., 1995, 2006) and 200 My (based on detrital zircon data of major sandstones in the Pretoria Group (Timeball Hill, Daspoort and Magaliesberg formations); Mapeo et al., 2006) marks the transition from the underlying Chuniespoort Group to the deposition of the mudstone–sandstone, lavas and pyroclastics with intercalated subordinate conglomerate, diamictite and carbonate beds of the Pretoria Group of the Transvaal Basin (Dankert and Hein, 2010) (Table 1). Erosion has widely reduced the area that was formerly covered by the Pretoria Group and it is now mainly confined to an area extending from eastern Botswana (where it is correlated with the Segwagwa Group; Mapeo et al., 2006) around the Transvaal Basin, including the smaller Potchefstroom sub-basin along its southern margin (Reczko et al., 1995a), to the eastern escarpment of the Mpumalanga Province. Correlated successions occur within the small Kanye basin of Botswana and in the Griqualand West basin in the Northern Cape Province of South Africa where it is known as the Postmasburg Group (Fig. 2a). In the centre of the Transvaal basin, the Pretoria Group is covered by rocks of the Bushveld Complex, and is only exposed along the margins thereof in a number of floor- and roof- attached (Hartzer, 1995) “fragments”, lying to the northeast and northwest of Pretoria, and in a basin-marginal main outcrop zone (Fig. 2b). The Pretoria ...
Citations
... Department of Mineral and Energy Affairs, 2018). Ages of the basement and cover were obtained from Bangert et al. (1999), Poujol and Anhaeusser (2001), Kositcin and Krapež (2004), Lenhardt et al. (2012) and Gumsley et al. (2020). The red box, marked Fig. 2, shows the present study area. ...
... Precambrian Pretoria Group (2.3-2.1 Ga) shales and quartzites dominate the regional geology (Lenhardt et al., 2012). Bands of dolerite (diabase sills formed during the early Jurassic because of the break-up of Gondwana, ) and quartzite form ridges and high ground since they are more resistant to erosion (Norman and Whitfield, 2006). ...
Inpalaeoenvironmentalreconstructions,fewstudieshavefocusedoncomparingthepresenceofpollenandphytolithsinsurfacesedimentstothelocalvegetation.Notwithstandinginherentdifferentialpollenandphytolithpreservation,production,anddispersal,whichaffecttheirrecoveryandinterpretationinsedimentarchives,thereremainsaneedtoexploremodernpollen-vegetation,phytolith-vegetationandpollen-phyto-lithrelationshipstoimprovepalaeoenvironmentalreconstructions.Wecarriedoutpalynologicalandphyto-lithanalysisonsixsurfacesedimentsampleslinkedtothebotanicalsurveyoffivevegetationsurveyplotsatGustavKlingbielNatureReserve,MpumalangaprovinceinnortheasternSouthAfrica,tounderstandtherela-tionshipsbetweenthemodernpollenandphytolithassemblagesinthesurfacesedimentsandthecontem-poraryvegetationofthestudyarea.CorrespondenceAnalysis(CA)wasundertakentoexaminethefingerprintofthelocalvegetationinthemodernpollen-phytolithassemblageandevaluatewhetherthephy-tolithandpollenassemblagesweretrackingeachother.CAillustratedthatpollenandphytolithproxiesarenottrackingthesameprimaryvegetationsignals;modernpollenassemblagesbestrepresentthecompo-nentsofforestandwetlandvegetation.Incontrast,modernphytolithassemblagesbestrepresentgrasslandvegetation.Ourstudyshowsthatdifferentialpollenandphytolithpreservation,productionanddispersalsig-nificantlyaffectsedimentrecordsmorethananticipated,thuscombiningmulti-proxies(e.g.,phytoliths,pol-len)providesamoreaccuratebasisfortheinterpretationoffossilspectrainpalaeoenvironmentalreconstructions.
... Tuffs are noticeably present just above the Duitschland-Timeball Hill contact at profiles in the far west (1) and east (7) of the transect (Figs. 2 and 4). Their relationship to the Bushy Bend lavas, which is commonly assigned to the very top of the Rooihoogte Formation (Coetzee, 2001) or lowermost Timeball Hill Formation (Eriksson et al., 1995;Lenhardt et al., 2012), remains currently uncertain. On a basinal scale, it seems chronostratigraphic correlation is not in line with sequence stratigraphic correlations of the sediments. ...
The stratigraphic correlation of the ca. 2353 ± 18 Ma to 2316 ± 7 Ma Duitschland and Rooihoogte formations, Transvaal Supergroup, South Africa, becomes critically important when viewed in the light of the Great Oxidation Event (GOE), since both formations record the transition from mass-independent (MIF-S) to mass-dependent fractionation of sulfur isotopes (MDF-S) indicative of the transition to an oxygenated atmosphere. At present, discrepancies exist as to whether the Duitschland and Rooihoogte formations are two distinct formations , formed in an older and younger depositional basin, respectively, or alternatively were deposited contemporaneously in the same basin at different positions to shorelines of the basin. A decoupled depositional history would indicate an oscillating oxygenation trajectory of the planet, whereas a contemporaneous deposi-tion of the two formations would point towards a sudden increase of atmospheric oxygen. To investigate the controversy of a contemporaneous versus a consecutive depositional relationship, this study presents a sedi-mentological investigation of 11 stratigraphic profiles intersecting the Duitschland to lower Timeball Hill and Rooihoogte to lower Timeball Hill formations. The 11 profiles are correlated between the Transvaal Supergroup in the Transvaal area of South Africa and the adjoining Kanye area in Botswana and show that the Duitschland and Rooihoogte formations are laterally correlative. The sedimentological observations are reinforced by similar major and trace element data as well as Sr-Nd isotope compositions measured in four scientific drill cores, intersecting both the Duitschland and the Rooihoogte formations. Consequently, the documented disappearance of MIF-S in these two formations should be regarded as a single-step atmospheric change, thereby removing the best supporting evidence for a protracted and dynamic oxygenation trajectory. Furthermore, bulk sediment T DM (Nd) of the Duitschland and Rooihoogte formations consistently display Archean crustal residence ages in the interval between 2.9 Ga and 3.3 Ga for most samples. Geochemical discrimination functions and zircon distribution patterns suggest a collisional tectonic setting and measured bulk sediment Nd isotope data combined with literature zircon Hf isotope data indicate a potential collision between a Rae-type craton and the Kaapvaal craton around 2.5 Ga.
... These deposits are considered to have formed by rapid suspension fallout from high particle concentration flows typifying subaqueous eruptions (e.g. Lenhardt et al., 2012). The occurrence of accretionary lapilli is generally interpreted as an indication of proximity to the volcanic vent, with the lapilli interpreted to have originated from phreatomagmatic eruptions associated with magma-water interactions (e.g. ...
... The accretionary lapilli tuff beds are interpreted as phreatomagmatic fallout deposits, where the accretionary lapilli form where water vapour enables adhesion of the ash grains (Lenhardt et al., 2012). They typically do not form in the presence of excessive water (i.e. ...
... where tephra grains are entrained in water) and are considered to be a proximal facies. These conditions occur in subaqueous environments where the hot grains are isolated from the surrounding water due to steam envelopes that, in essence, create water-free sub-environments (Lenhardt et al., 2012). Occurrence of accretionary lapilli tuffs proximal to the vent areas is further evidence of a steam rich cupola. ...
The greenschist to lower-amphibolite facies Proterozoic (Orosirian: ca. 1883–1848 Ma) Aillik Group in Labrador, Canada, comprises a well-preserved example of subaqueous to intermittently subaerial volcanism and sedimentation. The group consists of interlayered sedimentary, volcaniclastic, pyroclastic and bimodal volcanic units that exhibit well-preserved depositional and structural relationships. Cumulatively, these lithofacies significantly demonstrate the evolution of Paleoproterozoic volcanism during the assembly of the supercontinent Nuna. A minimum stratigraphic thickness of 14.9 km of the Aillik Group is exposed along the coast of Labrador and is divided into two successions. The western stratigraphic succession consists of subaerial to subaqueous sandstone, siltstone, conglomerate, resedimented tuffaceous sandstones, and tuffaceous conglomerate with lesser rhyolite flows and felsic tuff. The eastern stratigraphic succession is dominated by felsic tuff and rhyolite, lesser pillowed to massive mafic lavas, and in situ and resedimented mafic tuff. Mafic rocks are basaltic, whereas felsic rocks comprise high-silica rhyolites and tuffs. Primary subaqueous pyroclastic deposits are indicative of explosive activity, an observation also supported by the presence of accretionary lapilli and an abundance of lithic fragments in epiclastic deposits. The volcano-sedimentary lithofacies of the Aillik Group are consistent with that observed in modern intra-arc or arc-rift settings, and the succession is interpreted to have formed in a localized back-arc rifting environment within an overall complex convergent margin.
... The lack of evidence that shows that magmas had interaction with water, such as peperites and hyaloclastites (definitive indicators for subaqueous environments), reinforces the subaerial nature of explosive volcanism (McPhie et al., 1993). However, it is essential to notice that the observed set of structures and textures are not exclusive of subaerial volcanism despite being often interpreted (Walker, 1992;Gregg and Fink, 1995;Robins et al., 2010;Lenhardt et al., 2012). ...
Proterozoic Silicic Large Igneous Provinces (SLIP) are part of the geological inventory since they record the planet's thermal evolution. The Amazonian Craton hosted the poorly understood Uatumã SLIP, which formed ca. 1.88 Ga years ago. The so-called Uatumã magmatism represents the volcanic products of SLIP. This magmatism also refers to specific stratigraphic units, such as the Uatumã, Iriri, and Iricoumé groups. The Aruri and Salustiano formations within the Iriri Group represent an important SLIP stratigraphy segment transitioning from effusive to explosive volcanism. However, this interval is still undisclosed and lacks a thorough comparison of the Iriri Group stratigraphy. From field observations, detailed petrographic description, new geochemistry data on the Aruri and Salustiano formations, and a compilation of geochemistry data on the Uatumã SLIP, we present the litho-stratigraphy for the two formations and a comparison to the other levels of the SLIP stratigraphy. Our data show a predominance of effusive over explosive volcanism and a set of intermediate rocks that intrude the sequence. The Aruri Formation contains six lithofacies organized as interbedded facies representing coherent rhyolitic lava flows and volcaniclastic rocks (tuffs), which are intruded by massive dacite dykes. The Salustiano Formation contains three lithofacies that comprehend coherent rhyolite flows and ignimbrites. Whole-rock geochemistry data indicate that the felsic facies rocks represent rhyolites to alkali-rhyolites (>70% SiO2) of the high-K calc-alkaline to shoshonite series with metaluminous to a peraluminous character. The rocks are considered as formed in transitional settings from a mature arc to an extensional regime. The comparison to the remainder of the Iriri Group corroborates an interpretation of their origin as subaerial volcanism with coeval effusive and explosive phases in a caldera-type volcanic system and epiclastic rocks deposited in a distal position concerning the volcanic center. The whole-rock geochemistry is similar to the other levels of the Uatumã SLIP and confirms the large extent of the ca. 1.88 Ga years ago in the Amazonian Craton.
[https://www.sciencedirect.com/science/article/abs/pii/S0895981120306386]
[Free access until Feb 17, 2021 - https://authors.elsevier.com/c/1cK1U3BkFSPhpI]
... 200 m thick Boven Shale Member, which is best developed in the eastern part of the sub-basin (Eriksson et al., 2006). The upper Lydenburg Shale Member occurs with considerable thickness (1 250 m on average) across much of the sub-basin (Lenhardt et al., 2012). ...
... The volcaniclastic material directly underlying the AIB may be related to the ca. 2 200 Ma Machadodorp Member whose lavas have been identified ~1.5 m above the AIB. This member can indeed be rich in various volcaniclastic rocks, frequently in its lower part (Lenhardt et al., 2012). ...
... The albite crystals observed in the varioles possibly resulted from albitisation/Ca-remobilisation of primary plagioclase during metamorphic alteration of the AIB. Rocks from the Pretoria Group were affected by burial metamorphism up to the lower greenschist facies (e.g., Button 1973;Lenhardt et al., 2012;Mapeo et al., 2006;Humbert et al., 2017Humbert et al., , 2018Humbert et al., , 2020, with albite/An-poor plagioclase as the stable plagioclase variety. The predominance of secondary calcite within amygdales suggests that Ca from the plagioclase has been remobilized to form it, which supports the hypothesis of albitisation with local diffusion (i.e., without involving metasomatism). ...
Numerous Mesoproterozoic alkaline intrusions belonging to the Pilanesberg Alkaline Province are present within the Transvaal subbasin of the Kaapvaal craton. The Pilanesberg Complex is the best-known example; it represents one of the world's largest alkaline complexes, and is associated with NW-SE trending dyke swarm that extends from Botswana to the SW of Johannesburg. We here present a petrological and geochemical study of a thin mafic sill (here referred to as alkaline igneous body, AIB), which intrudes the ca. 2200 Ma Silverton Formation close to the southernmost part of the Pilanesberg dyke swarm. It has only been observed in cores from a borehole drilled close to Carletonville. The AIB is hypocrystalline, containing randomly oriented elongated skeletal kaersutite crystals and 6 - 8 mm varioles mainly composed of radially oriented acicular plagioclase. These two textures are related to undercooling, probably linked to the limited thickness (70 cm) of the AIB coupled with a possible shallow emplacement. Ar-Ar dating of the kaersutite gives an age of ca. 1400 Ma, similar to the age of Pilanesberg Complex. However, the AIB is an alkaline basaltic andesite and is thus notably less differentiated than the Pilanesberg Complex and some of its associated dykes, such as the Maanhaarrand dyke, for which we provide whole-rock geochemical data. Literature data indicate that the Pilanesberg dyke swarm also contains mafic hypabyssal rocks suggesting a link between the dyke swarm with the AIB. The AIB is characterized by strongly negative εNd and εHf, which cannot be related to crustal contamination, as shown by positive Ti and P anomalies, and the absence of negative Nb-Ta anomalies in mantle-normalised trace element diagrams. The AIB magma was presumably derived from a long-lived enriched, probably lithospheric mantle reservoir. The AIB thus provides important information on the magma source of the Pilanesberg Alkaline Province.
... The ca. 2.3 Ga restricted volcanic activity of the Bushy Bend lavas at the transition from the Rooihoogte Formation to the Timeball Hill Formation (Eriksson et al., 2001) possibly coincides with this cycle. The Bushy Bend Member volcanism was, however, short-lived and locally very restricted to the south of the Pretoria depository (Eriksson et al., 1994;Lenhardt et al., 2012). Nevertheless, eventually it led to the formation of the Timeball Hill epeiric sea that covered the Transvaal sub-basin (Eriksson et al., 2001). ...
... In the south of the Transvaal sub-basin, the preserved thickness of Hekpoort Formation lavas is in excess of 1100 m Humbert et al., 2018a, and references therein). The 800 m thickness that can be found in the west of the basin, thins out towards the northeast where less than 50 m are found (Button, 1973;Lenhardt et al., 2012). Concomitant with the north-easterly thinning, a reduction in the number of individual lava flows can be observed (Button, 1973). ...
... In the Mooikloof area, east of Pretoria, small, approximately east-west striking synclines and anticlines paralleling the strike of the margin to the BC have been mapped out (Robinson, 2015). Deposition of the entire formation generally took place during rifting and subsidence of the Kaapvaal Craton (Eriksson et al., 2001(Eriksson et al., , 2006Lenhardt et al., 2012). ...
The Palaeoproterozoic Hekpoort Formation of the Pretoria Group is a lava-dominated unit that has a basin-wide extent throughout the Transvaal sub-basin of South Africa. Additional correlative units may be present in the Kanye sub-basin of Botswana. The key characteristic of the formation is its general geochemical uniformity. Volcaniclastic and other sedimentary rocks are relatively rare throughout the succession but may be dominant in some locations. Hekpoort Formation outcrops are sporadic throughout the basin and mostly occur in the form of gentle hills and valleys, mainly encircling Archaean domes and the Palaeoproterozoic Bushveld Complex (BC). The unit is exposed in the western Pretoria Group basin, sitting unconformably either on the Timeball Hill Formation or Boshoek Formation, which is lenticular there, and on top of the Boshoek Formation in the east of the basin. The unit is unconformably overlain by the Dwaalheuwel Formation. The type-locality for the Hekpoort Formation is the Hekpoort farm (504 IQ Hekpoort), ca. 60 km to the west-southwest of Pretoria. However, no stratotype has ever been proposed. A lectostratotype, i.e., the Mooikloof area in Pretoria East, that can be enhanced by two reference stratotypes are proposed herein. The Hekpoort Formation was deposited in a cratonic subaerial setting, forming a large igneous province (LIP) in which short-termed localised ponds and small braided river systems existed. It therefore forms one of the major Palaeoproterozoic magmatic events on the Kaapvaal Craton.
... During the pre-rift uplift the Timeball Hill marine sediments were partly eroded and overlain during the syn-rift stage by continental clastic sediments (braided stream, alluvial fan, lacustrine) belonging to the Boshoek, Dwaalheuwel, Strubenkop and Daspoort formations (Eriksson et al., 2001). In addition, up to 1000 m thick successions of flood basalts and related pyroclastic rocks formed at 2.222 Ga (Cornell et al. 1996;Lenhardt et al., 2012). These are exposed in the Hekpoort Formation (between Boshoek and Dwaalheuwel formations in the Transvaal basin). ...
... According to Zeh et al. (2020) sandstones and metapelites of the Houtenbek formation deposited at < 2068 ± 7 Ma, and the detritus mainly derived from the weathering of igneous sources formed at 2.22 and 2.12 G by the reworking of older continental crust, as indicated by εHf t values ranging from 0 to −12. The post-Magaliesberg sediments are overlain by the Dullstroom lavas (located close to the town Dullstroom, Fig. 2), locally interlayered by clastic sedimentary rocks (Eriksson et al., 1994;Lenhardt et al., 2012), forming the oldest unit of the Rooiberg Group. Subsequently, the different formations of the Pretoria Group (from Silverton to post-Magaliesberg), and volcanic rocks of the Rooiberg Group were intruded by (ultra)mafic rocks and granites of the Bushveld Complex. ...
The Transvaal Basin in South Africa hosts a 15 km thick pile of sedimentary successions deposited over a period of more than 600 Ma during the Neoarchean to Paleoproterozoic. Presently, little is known about the source of these sediments, as well as about the tectono-magmatic evolution in the hinterland of the Transvaal Basin, preventing detailed geotectonic correlations of the Kaapvaal Craton (KC) with other cratons worldwide. To solve this problem, we present the first systematic study of combined U-Pb and Lu-Hf isotope data of more than 2000 detrital zircons from fourteen formations of the Transvaal Supergroup. These reveal that clastic sedimentary rocks were supplied from sources on and off the present-day KC. Detrital zircons in conglomerates of the Wolkberg and Black Reef formations, maximum deposition ages at 2769±8 and 2618±11 Ma respectively, were mainly supplied from surrounding KC, either from Pietersburg Block Basement (PBB), and/or from eroded sedimentary successions of the Witwatersrand, Pongola and/or Ventersdorp Supergroups. In contrast, clastic sedimentary rocks of the Rooihoogte, Duitschland and Timeball Hill formations (maximum deposition ages at 2353±18 Ma, 2342±18 and 2290±8 Ma, respectively) were predominately supplied from a juvenile Neoarchean terrane (JUNAT) formed at 2570-2500 Ma (εHf2500 Ma = +2 to +9) and intensely reworked at 2400, and to a minor amount from a composite Archean terrane (CAT) emplaced by granitoids between 3540 and 2680 Ma, and affected by crust reworking at 2570-2430 Ma (εHf2.5Ga = -3 to -12) in a Neoarchean to Paleoproterozoic continental arc terrane (NPCAT). Subsequent periodic reworking of JUNAT at 2250-2220 Ma and 2120 Ma is recorded by detrital zircons in sandstones of the overlying Boshoek, Dwaalheuwel, Daspoort, Magaliesberg and post-Magaliesberg formations, having maximum deposition ages at 2243±7, 2242±7, 2240±7, 2080±7, and 2068±7 Ma, respectively. The Archean zircons (age >2650 Ma) in all these formations were mainly supplied from PBB. The new data sets also suggest that the KC was connected to CAT, NPCAT and JUNAT at <2350 Ma. The nearly absence of detrital zircons with ages of 2570-2500 Ma in all formations younger than Boshoek perhaps results from intense reworking of JUNAT during magmatic events at 2400, 2340, 2220, and 2120 Ma, causing loss of the original juvenile character. Paleoproterozoic zircons with ages of 2220 and 2120 Ma in Dullstroom sandstones most likely result from re-deposition of post-Magalisberg sedimentary rocks, and Archean zircons from sources similar to Moodies and Fig Tree sandstones of the Barberton greenstone belt. Comparison of our new data from the Transvaal Basin with such from the Turee Creek and Horseshoe basins in NW-Australia provides no evidence for Kaapvaal-Pilbara Craton connection during the Neoarchean to Paleoproterozoic.
... The two main volcanic units of the Transvaal Supergroup, difference in the paleomagnetic poles reported for the two formations (Evans et al., 1997;Gumsley et al., 2017;Humbert et al., 2017). In the Kanye subbasin in Botswana (not represented in Fig. 1), the Tsatsu Formation is the main volcanic unit, and has been related to either the Hekpoort or the Ongeluk formations (e.g., Lenhardt et al., 2012). The Transvaal Supergroup is mostly covered by younger sediments in the Kanye subbasin. ...
... Underlying the Machadodorp Member is the 200 m-thick in average Boven Shale Member, which is best developed in the eastern part of the basin, and is characterized by high-alumina shales (Eriksson et al., 2006). The upper Lydenburg Shale Member occurs in considerable thickness (1250 m on average) across much of the basin, and is characterized by tuffaceous, high CaO-MgO-MnO shales (Lenhardt et al., 2012). The Silverton Formation conformably overlies the Daspoort Formation and is overlain by the Magaliesberg Formation with a gradual contact (Fig. 2a). ...
... Sporadic lavas and tuffs occur at the same stratigraphic level in the central and western regions of the Transvaal basin . The basal part of the Machadodorp Member is generally formed by a coarsening-upward sequence of pyroclastic rocks, ranging from relatively fine-grained tuffs to tuff-breccias and fluidal-clast breccias (Lenhardt et al., 2012). Recently, Lenhardt et al. (2020) however showed that at the Doornkop Nature Reserve, which is~35 km south of Machadodorp ( Fig. 1), a~10-20 m sheet lava layer occurs at the bottommost of Machadodorp Member and underlies the volcaniclastics rocks. ...
The Neoarchean to Paleoproterozoic (ca. 2.65 – 2.06 Ga) Transvaal Supergroup of the Kaapvaal Craton is dominated by clastic and chemical sedimentary rocks, with few strata-bound igneous units. Two ca. 2.2 Ga igneous units in its upper part (Pretoria Group), the Machadodorp Member of the Silverton Formation and the Mashishing dyke swarm, have so far only been recognized in the eastern part of the Transvaal subbasin. As these units are near-contemporaneous with an early Paleoproterozoic geochemical excursion to higher Th/Nb ratios compared to both older and younger igneous units, they probably mark a change in the geodynamic evolution of the craton. In the present contribution, we provide new data on the igneous units within the Silverton Formation (one lava and 5 sills) from a borehole (BB11) in the south-western part of the Transvaal subbasin. Two of these igneous units are particularly interesting because they show strong geochemical similarities to the Machadodorp Member and Mashishing dyke swarm, which are thus more spatially extensive than previously thought. In combination with existing data, our results further show that the Machadodorp Member corresponds to two separate, possibly not co-magmatic, igneous events, as documented by low- and high-Th basalts. The low-Th lavas show flat mantle-normalized REE and multi-element patterns, along with depleted Nd and Hf isotopic signatures, which are unusual among Precambrian magmatic rocks in the craton. This signature can be explained by partial melting of a depleted spinel peridotite incorporated into the subcratonic lithospheric mantle (SCLM) by a prior subduction event. A failed rifting event can have triggered melting of such a shallow source in a cratonic context. Even though comparable in age, the high-Th ca. 2.2 Ga Mashishing dyke swarm comes from a very different source, likely a part of the SCLM that had experienced subduction-related fluid metasomatism around the Archean - Proterozoic transition. About 150 million years later, the Bushveld Complex possibly exploited the same rifting structures during its intrusion.
... The~2.1 Ga rutile cooling age is significantly older than the timing of granulite-facies metamorphism in the central Limpopo Belt at ca. 2.0 Ga (Holzer et al., 1998(Holzer et al., , 1999Kramers and Mouri, 2011;Schaller et al., 1999) and is also significantly older than estimates for the age of crystallization of the Bushveld Complex in the north western Kaapvaal craton at between 2.05 and 2.06 Ga (Olsson et al., 2010;Scoates and Friedman, 2008). However, the 2.32 to 2.06 Ga Pretoria Group of the Transvaal Supergroup does contain a substantial volume of mantle-derived volcanic rocks (Lenhardt et al., 2012) that may explain the heat source for an amphibolite facies metamorphic event in the SMZ of the Limpopo Belt between 2.3 and 2.1 Ga. The Hekpoort Formation includes over 1000 m and 800 m of basaltic andesite in the southern and western portions of the Transvaal Basin, respectively (Lenhardt et al., 2012). ...
... However, the 2.32 to 2.06 Ga Pretoria Group of the Transvaal Supergroup does contain a substantial volume of mantle-derived volcanic rocks (Lenhardt et al., 2012) that may explain the heat source for an amphibolite facies metamorphic event in the SMZ of the Limpopo Belt between 2.3 and 2.1 Ga. The Hekpoort Formation includes over 1000 m and 800 m of basaltic andesite in the southern and western portions of the Transvaal Basin, respectively (Lenhardt et al., 2012). Slightly higher in the stratigraphy, the Machadodorp Member for the Silverton Formation contains an average thickness of 300 m of lavas and volcanoclastic rocks (Lenhardt et al., 2012). ...
... The Hekpoort Formation includes over 1000 m and 800 m of basaltic andesite in the southern and western portions of the Transvaal Basin, respectively (Lenhardt et al., 2012). Slightly higher in the stratigraphy, the Machadodorp Member for the Silverton Formation contains an average thickness of 300 m of lavas and volcanoclastic rocks (Lenhardt et al., 2012). The mafic and intermediate composition volcanic rocks of the Pretoria Group are interpreted to reflect partial melting of the mantle lithosphere below the Kaapvaal craton (Lenhardt et al., 2012), with the volcanic rocks preserved as erosional remnants over a substantial portion of the interior of the Kaapvaal Craton. ...
