Witwatersrand Basin

The Witwatersrand basin on South Africa's Kaapvaal craton is one of the best known of Archean sedimentary basins, and it contains some of the largest gold reserves in the world, accounting for more than 55 percent of all the gold ever mined. Sediments in the basin include a lower flysch-type sequence, and an upper molassic facies, both containing abundant silicic volcanic detritus. The strata are thicker and more proximal on the northwestern side of the basin that is at least locally fault-bounded. The Witwatersrand basin is a composite foreland basin that developed initially on the cratonward side of an Andean arc, similar to retroarc basins forming presently behind the Andes. A continental collision between the Kaapvaal and Zimbabwe cratons 2.7

Andean Basin

Map of the Kaapvaal and Zimbabwe cratons, showing the Great Dike on the Zimbabwe craton, the Witwatersrand and Pongola basins on the Kaapvaal craton, the Bushveld complex, and Limpopo Province separating the two cratons. Other major tectonic elements of southern Africa are also shown.

Gold Rock From Limpopo

Mineralized conglomerate forming gold ore from the Archean Witwatersrand basin on the Kaapvaal craton

(Brooks Kraft/Corbis)

billion years ago caused further subsidence and deposition in the Witwatersrand basin. Regional uplift during this later phase of development placed the basin on the cratonward edge of a collision-related plateau, now represented by Limpopo Province. There are many similarities between this phase of development of the Witwatersrand basin and basins such as the Tarim and Tsaidam, north of the Tibetan Plateau.

The Witwatersrand basin is an elongate trough filled predominantly by 2.8-2.6 billion-year-old clastic sedimentary rocks of the West Rand and Central Rand Groups, together constituting the Witwatersrand Supergroup. These are locally, in the northwestern parts of the basin, underlain by the volcano sedimentary Dominion Group. The structure strikes in a northeasterly direction parallel with, but some distance south of, the high-grade gneissic ter-rane of Limpopo Province. The high-grade metamor-phism, calc-alkaline plutonism, uplift, and cooling in the Limpopo are of the same age as and closely related to the evolution of the Witwatersrand basin. strata dip inward with dips greater on the northwestern margin of the basin than on the southeastern margin. The northwestern margin of the basin is a steep fault that locally brings gneissic basement rocks into contact with Witwatersrand strata to the south. Dips are vertical to overturned at depth near the fault, but only 20° near the surface, demonstrating that this is a thrust fault. A number of folds and thrust faults are oriented parallel to the northwestern margin of the basin.

The predominantly clastic sedimentary fill of the Witwatersrand basin has been divided into the West Rand and the overlying Central Rand Groups, which rest conformably on the largely volcanic Dominion Group. The Dominion Group was deposited over approximately 9,000 square miles (15,000 km2), but it is correlated with many similar volcanic groups along the northern margin of the Kaapvaal craton. The Dominion Group and its correlatives, and a group of related plutons, have been interpreted as the products of Andean arc magmatism, formed above a 2.8 billion-year-old subduction zone that dipped beneath the Kaapvaal craton. The overlying West Rand and Central Rand Groups were deposited in a basin at least 50,000 square miles (80,000 km2). stratigraphic thicknesses of the West Rand Group generally increase toward the fault-bounded northwestern margin of the basin, whereas thicknesses of the Central Rand Group increase toward the center of the basin. Strata of both groups thin considerably toward the southeastern basin margin. The northeastern and southwestern margins are poorly defined, but some correlations with other strata (such as the Godwan Formation) indicate that the basin was originally larger than the present basin. strata originally deposited north of Johannesburg are buried, removed by later uplift, omitted by igneous intrusion, and cut out by faulting.

The West Rand Group consists of southeastward tapering sedimentary wedges that overlie the Dominion Group, and were deposited directly on top of granitic basement in many places. The maximum thickness of the West Rand Group, 25,000 feet (7,500 m), occurs along the northern margin of the basin, and the group thins southeast to a preserved thickness of 2,700 feet (830 m) near the southern margin. Shale and sandstone in approximately equal proportions characterize the West Rand Group, and a thin horizon of mafic volcanics is locally present. This volcanic horizon thickens to 800 feet (250 m) near the northern margin of the basin, but is absent in the south. The West Rand Group contains mature quartzites, minor chert, and sedimentation patterns indicating both tidal and aeolian reworking. Much of the West Rand Group is an ebb-dominated tidal deposit later influenced by beach-swash deposition. More shales are preserved near the top of the group. Overall, the West Rand Group preserves a transition from tidal flat to beach, then deeper water deposition indicates a deepening of the Witwatersrand basin during deposition. Upper formations in the West Rand Group contain magnetic shales and other finegrained sediments suggestive of a distal shelf or epicontinental sea environment of deposition.

The lower West Rand Group records subsidence of the Witwatersrand basin since the sediments grade vertically from beach deposits to a distal shallow marine facies. This transition means that the water was becoming deeper during deposition, showing that there was active subsidence of the basin. Since there is a lack of coarse, immature, angular conglomerate and breccia-type sediments, the subsidence was probably accommodated by gentle warping and flexure, not by faulting. A decreasing rate of subsidence and/or a higher rate of clastic sediment supply is indicated by the progressively shallower-water facies deposits in the upper West Rand Group. Numerous silicic volcanic clasts in the West Rand Group indicate that a volcanic arc terrane to the north was contributing volcanic detritus to the Witwatersrand basin. Additionally, the presence of detrital ilmenite, fuchsite, and chromite indicate that an ultramafic source such as an elevated greenstone belt was also contributing detritus to the basin.

The Central Rand Group was deposited conformably on top of the West Rand Group and attains a maximum preserved thickness of 9,500 feet (2,880 m) northwest of the center of the basin, and north of the younger Vredefort impact structure. Sediments of the Central Rand Group consist of coarsegrained graywackes and conglomerates along with subordinate quartz sandstone interbedded with local lacustrine or shallow marine shales and siltstones. The conglomerates are typically poorly sorted and have large clasts with well-rounded shapes, while the smaller pebbles have angular to subangular shapes. Paleocurrent indicators show that the sediments pro-graded into the basin from the northwestern margin in the form of a fan-delta complex. This is economically important because numerous goldfields in the Central Rand Group are closely associated with major entry points into the basin. Some transport of sediments along the axis of the basin is indicated by paleocurrent directions in a few locations. A few volcanic ash (tuffaceous) horizons and a thin mafic lava unit are found in the Central Rand Group in the northeast part of the basin. The great dispersion of unimodal paleocurrent directions derived from most of the Central Rand Group indicates that these sediments were deposited in shallow-braided streams on coalescing alluvial fans. The paleorelief is estimated at 20 feet (6 m) in areas proximal to the source, and 1-2 feet (0.5 m) in more distal areas. Some of the placers in the Central Rand Group have planar upper surfaces, commonly associated with pebbles and heavy placer mineral concentrations, which may be attributed to reworking by tidal currents. Clasts in the conglomerates include vein quartz, quartz aren-ite, chert, jasper, silicic volcanics, shales and schists, and other rare rocks.

The Central Rand Group contains a large amount of molassic-type sediments disposed as sand and gravel bars in coalesced alluvial fans and fluvial systems. The West Rand/Central Rand division of the Witwatersrand basin into a lower flysch-type sequence and an upper molasse facies is typical of foreland basins. Extensive mining of paleoplacers for gold and uranium has enabled mapping of the dendritic paleodrainage patterns and the points of entry into the basin to be determined. The source of the Central Rand sediments was a mountain range located to the northwest of the basin, and this range contained a large amount of silicic volcanic material.

The growth of folds parallel to the basin margin during sedimentation and the preferential filling of synclines by some of the mafic lava flows in the basin indicate that folding was in progress during Central Rand Group sedimentation. Deformation of this kind is diagnostic of flexural foreland basins, and studies show that the depositional axis of the basin migrated southeastward during sedimentation, with many local unconformities related to tilting during flexural migration of the depositional centers.

The Witwatersrand basin exhibits many features characteristic of foreland basins, including an asymmetric profile with thicker strata and steeper dips toward the mountainous flank, a basal flysch sequence overlain by molassic-type sediments, and thrust faults bounding one side of the basin. Com-pressional deformation was in part syn-sedimentary and associated folds and faults strike parallel to the basin margins, and the depositional axis migrated away from the thrust front with time, as in younger foreland basins. Stratigraphic relationships within the underlying Dominion Group, the presence of silicic volcanic clasts throughout the stratigraphy, and minor lava flows within the basin suggest that the foreland basin was developed behind a volcanic arc, partly preserved as the Dominion Group. Sediments of the West Rand Group are interpreted as deposited in an actively subsiding foreland basin developed adjacent to an Andean margin and fold thrust belt.

Deformation in Limpopo Province and the northern margin of the Kaapvaal craton are related to a collision between the northern Andean margin of the Kaapvaal craton with a passive margin developed on the southern margin of the Zimbabwe craton that began before 2.64 billion years ago, when Venters-dorp rifting, related to the collision, commenced. It is possible that some of the rocks in the Witwatersrand basin, particularly the molasse of the Central Rand Group, may represent erosion of a collisional plateau developed as a consequence of this collision. The plateau would have been formed in the region between the Witwatersrand basin and Limpopo Province, a region characterized by a deeply eroded gneiss terrane. A major change in the depositional style occurs in the Witwatersrand basin between the Central Rand and West Rand Groups, and this break may represent the change from Andean arc retroarc foreland basin sedimentation to collisional plateau erosion-related phases of foreland basin evolution.

Paleoplacers in the Witwatersrand basin have yielded more than 850 million tons of gold, dwarfing all the world's other gold placer deposits put together. Many of the placer deposits (called reefs in local terminology) preserved detrital gold grains on erosion surfaces, along foreset beds in cross-laminated sandstone and conglomerate, in trough cross-beds, in gravel bars, and as detrital grains in sheet sands. Most of the gold is located close to the northern margin of the basin in the fluvial channel systems. Some of the gold flakes in more distal areas were trapped by stromatolite-like filamentous algae, and some appear to have even been precipitated by types of algae, although it is more likely that these are fine, recrystallized grains trapped by algal filaments. Besides gold more than 70 other ore minerals are recognized in the Witwatersrand basin; most are detrital grains, and others are from metamorphic fluids. The most abundant detrital grains include pyrite, uraninite, brannerite, arsenopyrite, cobalt-ite, chromite, and zircon. Gold-mining operations in the Witwatersrand employ more than 300,000 people and have led to the economic success of South Africa.


The Zimbabwe craton is a classic granite greenstone terrane. In 1971 Clive W. Stowe, in a Ph.D. dissertation and publication from the University of London, proposed a division of the Zimbabwean (then Rho-desian) craton into four main tectonic units. His first unit includes remnants of older gneissic basement in the central part of the craton, including the Rhodes-dale, Shangani, and Chilimanzi gneissic complexes. Stowe's second (northern) unit includes mafic and ultramafic volcanics overlain by a mafic/felsic volcanic sequence, iron formation, phyllites, and conglomerates of the Bulawayan Group all overlain by sandstones of the Shamvaian Group. The third, or southern, unit consists of mafic and ultramafic lavas of the Bulawayan Group, overlain by sediments of the Shamvaian Group. The southern unit is folded about east-northeast axes. Stowe defined a fourth unit in the east, including remnants of schist and gneissic rocks, enclosed in a sea of younger granitic rocks. Stowe was a leader in recognizing complex structures in greenstone belts, stating in his dissertation and 1974 paper that the Selukwe greenstone belt "appears to be part of an imbricated and overturned lower limb of a large recumbent fold, resting alloch-thonously on a gneissic basement." In 1979 John F. Wilson, a geologist from the Geological Survey of Rhodesia (now Zimbabwe), proposed a regional correlation between the greenstone belts in the craton. His general comparison of the compositions of the upper volcanics in the greenstone belts resulted in a distinction between the greenstone belts located in the western part of the craton from those in the eastern section. The greenstone belts to the west of his division are composed of dominantly calc-alka-line rock suites including basalt, andesite, and dacite flows and pyroclastic rocks. This western section includes bimodal volcanic rocks consisting of tho-leiite and magnesium-rich pillow basalt and massive flows, with some peridotitic rocks alternating with dacite flows, tuffs, and agglomerates. The eastern section of the Zimbabwe craton is characterized by pillowed and massive tholeiitic basalt flows and less abundant magnesium-rich basalts and their meta-

morphic equivalents. The eastern section contains a number of phyllites, banded iron formations, local conglomerate, and grit and rare limestone. Wilson identified an area of well-preserved 3.5 billion-year-old gneissic rocks and greenstones in the southern part of the province, and named this the Tokwe segment. He suggested that this may be a "mini-craton," and that the rest of the Zimbabwe craton stabilized around this ancient nucleus.

Despite these early hints that the Zimbabwe cra-ton may be composed of a number of distinct ter-ranes, much of the work on rocks of the Zimbabwe craton has been geared toward making lithostrati-graphic correlations between these different belts, and attempting to link them all to a single supergroup-style nomenclature. Many workers attempted to pin the presumably correlatable 2.7 billion-year-old stratigraphy of the entire Zimbabwe craton to an unconformable relationship between older gneissic rocks and overlying sedimentary rocks exposed in the Belingwe greenstone belt. More recently, Timothy Kusky, Axel Hoffman, and others have emphasized that the sedimentary sequence unconformably overlying the gneissic basement may be separated from the mafic/ultramafic magmatic sequences by a regional structural break, and that the presence of a structural break in the type of stratigraphic section for the Zimbabwe craton casts doubt on the significance of any lithostratigraphic correlations across Stowe's divisions of the craton.

Was this article helpful?

0 0
The Basic Survival Guide

The Basic Survival Guide

Disasters: Why No ones Really 100 Safe. This is common knowledgethat disaster is everywhere. Its in the streets, its inside your campuses, and it can even be found inside your home. The question is not whether we are safe because no one is really THAT secure anymore but whether we can do something to lessen the odds of ever becoming a victim.

Get My Free Ebook

Post a comment