Elsevier

Earth-Science Reviews

Volume 224, January 2022, 103867
Earth-Science Reviews

Review article
Provenance of Kalahari Sand: Paleoweathering and recycling in a linked fluvial-aeolian system

https://doi.org/10.1016/j.earscirev.2021.103867Get rights and content

Abstract

We here review what is known about the dunefields and fluvial systems of the Kalahari Basin in terms of geological setting and Quaternary dynamics and set out what has been hypothesized about the provenance of Kalahari sand so far. Previous work has tackled this problem by applying a limited number of techniques (mostly sediment textures and heavy minerals) to parts of the large dryland. The generally highly quartzose mineralogy of aeolian dunes and their compositional variability have been only broadly evaluated and several sedimentological issues have thus remained controversial, including the relative role played by fluvial processes versus aeolian reworking of older sediments and weathering controls. This reveals a need for a systematic provenance study that considers the entire basin. For this reason, here we combine original petrographic, heavy-mineral, and detrital-zircon geochronology data with previously published clay-mineral, geochemical, and geochronological information to present the first comprehensive provenance study of the vast Kalahari sand sea.

Our multi-proxy dataset comprises 100 samples, collected across the Kalahari Basin from 11°S (NW Zambia) to 28°S (NW South Africa) and from 15°E (Angola) to 28°30′W (Zimbabwe). Kalahari aeolian-dune sand mostly consists of monocrystalline quartz associated with durable heavy minerals and thus drastically differs from coastal dunefields of Namibia and Angola, which are notably richer in feldspar, lithic grains, and chemically labile clinopyroxene. The western Kalahari dunefield of southeastern Namibia is distinguished by its quartz-rich feldspatho-quartzose sand, indicating partly first-cycle provenance from the Damara Belt and Mesoproterozoic terranes. Within the basin, supply from Proterozoic outcrops is documented locally. Composition varies notably at the western and eastern edges of the sand sea, reflecting partly first-cycle fluvial supply from crystalline basements of Cambrian to Archean age in central Namibia and western Zimbabwe. Basaltic detritus from Jurassic Karoo lavas is dominant in aeolian dunes near Victoria Falls.

Bulk-sediment petrography and geochemistry of northern and central Kalahari pure quartzose sand, together with heavy-mineral and clay-mineral assemblages, indicate extensive recycling via aeolian and ephemeral-fluvial processes in arid climate of sediment strongly weathered during previous humid climatic stages in subequatorial Africa. Distilled homogenized composition of aeolian-dune sand thus reverberates the echo of paleo-weathering passed on to the present landscape through multiple episodes of fluvial and aeolian recycling.

Intracratonic sag basins such as the Kalahari contain vast amounts of quartz-rich polycyclic sand that may be tapped by rivers eroding backwards during rejuvenation stages associated with rift propagation. Such an event may considerably increase the sediment flux to the ocean, fostering the progradation of river-fed continental-embankments, as documented by augmented accumulation rates coupled with upward increasing mineralogical durability in the post-Tortonian subsurface succession of the Zambezi Delta.

The Central Kalahari is not a true desert. It has none of the naked, shifting sand dunes that typify the Sahara and other great deserts of the world. In some years the rains may exceed twenty — once even forty — inches, awakening a magic green paradise.” Mark Owens, Cry of the Kalahari.

Introduction

The intracratonic Kalahari sag basin hosts several dunefields that, largely inactive at present, represent the largest sand sea on Earth (Fig. 1). The compositional signatures of such a vast expanse of aeolian sand and their provenance have not been systematically studied so far, and yet encrypted in them lies a bounty of information on the geological, geomorphological, and environmental history of the region. Formed as a consequence of the multistep break-up of Gondwana, the Kalahari Basin presently occupies the core of southern Africa, which experienced dynamic uplift in the Cenozoic and is currently cut across by the southwestward-propagating East African rift system (Haddon and McCarthy, 2005; De Wit, 2007). Complex landscape evolution during the Pleistocene and Holocene was punctuated by a high-frequency alternation of arid and humid climatic stages and consequent repeated changes in hydrology, drainage patterns, and interaction of fluvial and aeolian processes (Burrough et al., 2009a; Hürkamp et al., 2011; Moore et al., 2012; Matmon et al., 2015). Decrypting the Kalahari sedimentary archive is an essential step to improve our understanding not only of the evolution of tropical southern Africa but also of the interplay between tectonic and climatic forces that mould the Earth's surface. Clarifying the control exerted by key climate variables on arid landscapes can, in turn, help test the robustness of numerical models simulating dunefield dynamics and improve model simulations that are used to predict the impact of future climate change on aeolian-dune remobilisation (e.g., Thomas et al., 2005; Mayaud et al., 2017; Vainer et al., 2021a).

This article considers what is known about the Kalahari Basin and its hydrological systems and dunefields (Fig. 2), including an overview of their Quaternary history. The geology of the region is first outlined in the wider context of southern Africa (Fig. 3), before reviewing what was currently known about the provenance of the dunefields and potential fluvial feeder systems. To date, much of the information on provenance has been inferred from likely palaeowind directions (e.g., Thomas, 1987) or by applying a limited number of techniques (mostly sediment textures and heavy minerals) to some parts of this vast basin. Several sedimentological issues have thus remained controversial, including the relative role played by fluvial processes versus aeolian reworking and the origin of weathering. Quantitative petrographic data were obtained only on a few aeolian-dune sands in the north and west, and detrital-zircon ages only on fluvial sands in the north (Gärtner et al., 2014; Garzanti et al., 2014a).

For these reasons, we present new results from bulk-petrography, heavy-mineral, and detrital-zircon U–Pb geochronology analyses on 100 aeolian-dune and river-bar sands collected across 17 degrees of latitude from Zambia to South Africa and over 13 degrees of longitude from Angola to Zimbabwe. A set of statistical techniques was applied to this multi-proxy dataset to adequately illustrate the compositional variability of aeolian sand across the Kalahari Basin, reveal meaningful mineralogical patterns, identify the original sediment sources, and gain insight into sand dispersal pathways. In particular, this paper investigates inheritance from past climatic conditions, buffering of environmental signals through linked fluvial and aeolian systems, and progressive compositional homogenization and concentration of most durable minerals acquired through multiple cycles of erosion, transport, deposition, and diagenesis. The new provenance data are integrated and reviewed in terms of what is known about fluvial-aeolian interactions, chemical weathering, and drainage evolution in the Kalahari. Understanding the complexities of sediment transport systems, and particularly how sediment-routing connectivity regulates the transmission of environmental signals from source areas to depositional sinks over spatial and temporal scales, is essential for a realistic interpretation of the stratigraphic record (Romans et al., 2016; Allen, 2017; Caracciolo, 2020).

Section snippets

Geology of southern Africa

Southern Africa was amalgamated through multiple tectono-magmatic events dating back to the Archean and culminated with the Neoproterozoic Pan-African orogeny (Fig. 3; Hanson, 2003). The Archean core consists of the Kaapvaal and Zimbabwe Cratons, welded by the Limpopo Belt. The Kaapvaal Craton, progressively amalgamated between 3.7 and 2.7 Ga, was stabilized by 2.6 Ga, and eventually intruded by the Bushveld Complex at 2.06 Ga (Eglington and Armstrong, 2004). The Zimbabwe Craton, comprising

The Kalahari Basin

The intracratonic Kalahari sag basin comprises the largest continuous sand sea on Earth, which extends for over 2.5 × 106 km2 (Fig. 1). The interior of the Kalahari is an elevated plateau with flat topography (average altitude 1200 m), delimited by relatively steep escarpments down to the Atlantic Ocean in the west and to the Indian Ocean in the east. Kalahari Group sediments — including basal gravel and conglomerate, sandstone with calcrete, and unconsolidated sand —stretch north from the

Overview of previous work on the provenance of Kalahari sands

The provenance of Kalahari sands has not been investigated thoroughly by a multi-technique approach so far. Previous surveys recognized the highly quartzose composition of aeolian sands, but their compositional variability has been only broadly evaluated, and the origin of Kalahari dunefields was mostly ascribed to either reworking of older sediments (e.g., Du Toit, 1954; Baillieul, 1975; Thomas, 1987) or dominant fluvial processes (e.g., De Ploey et al., 1968; Verboom, 1974; Moore and Dingle,

Methods

In this provenance study, we have analysed 57 samples of aeolian dunes collected across the vast Kalahari sand sea in the frame of diverse research projects (Stone and Thomas, 2008; Matmon et al., 2018; Burrough et al., 2019; Stone et al., 2019; Wittmann et al., 2020; Vainer et al., 2021a). Another 43 sand samples collected from exposed sandbars or dry riverbeds in Angola, Botswana, Zambia, Zimbabwe, Namibia and South Africa, and previously studied with similar and complementary methodological

Mineralogy of river sands

The main rivers that drain into the Kalahari Basin are sourced in the humid regions of Angola and Zambia in the north. The Caculuvar and Mucope tributaries of the Cunene River, draining the western edge of the Kalahari Erg in Angola (Fig. 1A), carry pure quartzose sand with a few K-feldspar grains. The extremely poor, zircon-rich tHM suite includes epidote, tourmaline, and minor andalusite, staurolite and rutile (Table 1). The Okavango and Cuando Rivers, which are also sourced in Angola and

Mineralogy of aeolian-dune sands

Dune sand is quartz-rich over most the vast Kalahari Basin: out of the 57 studied samples, 31 are pure quartzose, 12 quartzose, and 10 quartz-rich feldspatho-quartzose (Fig. 5). Throughout the northern Kalahari, in Angola, northeastern Namibia, Caprivi Strip, northern Botswana and western Zambia, sand consists virtually exclusively of monocrystalline quartz commonly showing rounded to subrounded outline and abraded overgrowths (Fig. 4L). Aeolian sand in the Caprivi Strip contains abundant red

Ages of detrital zircons

Five age ranges recur among the analysed samples (Fig. 6), corresponding to main orogenic episodes in southern Africa (Hanson, 2003; Dirks et al., 2009; Andersen et al., 2016, Andersen et al., 2018): I) Limpopo (2.5–2.8 Ga, Neoarchean; peak at 2720 Ma); II) Eburnean (1.8–2.05 Ga, Orosirian; peak at 1893 Ma); III) Sinclair-Kibaran (1.2–1.4 Ga, Ectasian; peak at 1312 Ma); IV) Namaqua-Irumide (1.0–1.1 Ga, Stenian; peak at 1056 Ma); V) Damara-Lufilian (0.45–0.65 Ga, Ordovician-Cryogenian; peak at

Provenance of Kalahari aeolian-dune sand

In most of the Kalahari Basin, including the NWK and NEK dunefields in Angola, Namibia and Zambia, much of Botswana, and part of the EK dunefield in Zimbabwe, dune sand is dominated by monocrystalline quartz associated with very poor tHM suites including durable ZTR minerals, staurolite, and kyanite (Fig. 8, Fig. 9, Fig. 10). Such a homogenized mineralogical signature reveals multiple recycling of older quartzose sandstones through geological time.

A local exception is the quartz-rich

Fluvial-aeolian interactions and multistep recycling

The mineralogical composition of aeolian dunes and its variability across a sand sea reflect the relative importance of fluvial and aeolian processes and the degree of their interplay. Sand seas largely fed by river systems are typically characterized by partly first-cycle detritus including various amounts of diverse types of rock fragments, feldspars and heavy minerals, generally allowing identification of a single dominant source, as for the Namib Erg (Garzanti et al., 2012). The opposite

Paleoweathering in the Kalahari

The extent to which a sediment has been subjected to chemical weathering, integrated over a series of sedimentary cycles, can be evaluated by combining evidence from petrographic, heavy-mineral, clay-mineral, and geochemical data.

Distilled multicyclic sand of northern Kalahari aeolian dunes and of the Okavango, Cuando, and upper Zambezi Rivers is composed dominantly of quartz with strongly depleted tHM suites including zircon, tourmaline, rutile, staurolite, and kyanite but virtually no garnet

Drainage integration as a driver of provenance change

Sand seas such as the Kalahari or the Great Nafud and Rub'Al Khali in Arabia are huge reservoirs of quartz-rich polycyclic sand trapped in the continental interiors. In the hyperarid climate of the tropics, precipitation and runoff can be so scarce that ephemeral rivers are unable to carry sediment across and beyond these vast rift-related sags, where sand is dumped and multiply reworked while remaining largely untapped for tens or even hundreds of million years. An analogous ancient case is

Conclusions

Aeolian dunes of the Kalahari are homogeneously pure quartzose in Angola, Zambia, and over much of Botswana and parts of Zimbabwe and Namibia, where they also contain a few K-feldspar grains and strongly depleted heavy-mineral assemblages dominated by ZTR minerals and including common staurolite in Botswana and kyanite in Zambia, but virtually no garnet. Composition varies markedly only at the western and eastern edges of the erg, ranging from feldspar-rich feldspatho-quartzose and

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Several samples from Kalahari aeolian dunes were obtained thanks to the generosity of Sallie Burrough, Tom Kemp, and Matt Telfer; some fluvial samples were provided by Hella Wittmann, Hanna Haedke, Armanda Trindade, Dario Fornara, Lindani Ncube, and Helena Johanna Van Niekerk. Most petrographic and heavy-mineral analyses were made by Giuditta Radeff, Marta Padoan, and Mara Limonta. Marta Padoan helped substantially with the compilation of geochronological ages and Martin Rittner with

References (227)

  • S.L. Burrough et al.

    Late Quaternary lake-level fluctuations in the Mababe Depression: Middle Kalahari palaeolakes and the role of Zambezi inflows

    Quat. Res.

    (2008)
  • S.L. Burrough et al.

    Mega-Lake in the Kalahari: a Late Pleistocene record of the Palaeolake Makgadikgadi system

    Quat. Sci. Rev.

    (2009)
  • S.L. Burrough et al.

    Late Quaternary hydrological dynamics in the Middle Kalahari: forcing and feedbacks

    Earth Sci. Rev.

    (2009)
  • S.L. Burrough et al.

    Implications of a new chronology for the interpretation of the Middle and Later Stone Age of the upper Zambezi Valley

    J. Archaeol. Sci. Rep.

    (2019)
  • L. Caracciolo

    Sediment generation and sediment routing systems from a quantitative provenance analysis perspective: Review, application and future development

    Earth Sci. Rev.

    (2020)
  • O. Catuneanu et al.

    The Karoo basins of south-central Africa

    J. Asian Earth Sci.

    (2005)
  • O. Crouvi et al.

    Active sand seas and the formation of desert loess

    Quat. Sci. Rev.

    (2010)
  • H. De Carvalho et al.

    Geochronological review of the Precambrian in western Angola: links with Brazil

    J. Afr. Earth Sci.

    (2000)
  • A. Eglinger et al.

    Tectono-metamorphic evolution of the internal zone of the Pan-African Lufilian orogenic belt (Zambia): Implications for crustal reworking and syn-orogenic uranium mineralizations

    Lithos

    (2016)
  • B.M. Eglington

    Evolution of the Namaqua–Natal Belt, southern Africa—a geochronological and isotope geochemical review

    J. Afr. Earth Sci.

    (2006)
  • E. Garzanti

    From static to dynamic provenance analysis – Sedimentary petrology upgraded

  • E. Garzanti

    Petrographic classification of sand and sandstone

    Earth Sci. Rev.

    (2019)
  • E. Garzanti et al.

    Heavy-mineral concentration in modern sands: implications for provenance interpretation

  • E. Garzanti et al.

    Petrology of the Namib sand sea: long-distance transport and compositional variability in the wind-displaced Orange Delta

    Earth Sci. Rev.

    (2012)
  • E. Garzanti et al.

    Provenance and recycling of Arabian desert sand

    Earth Sci. Rev.

    (2013)
  • E. Garzanti et al.

    Provenance versus weathering control on the composition of tropical river mud (southern Africa)

    Chem. Geol.

    (2014)
  • E. Garzanti et al.

    The modern Nile sediment system: Processes and products

    Quat. Sci. Rev.

    (2015)
  • E. Garzanti et al.

    Dynamic uplift, recycling, and climate control on the petrology of passive-margin sand (Angola)

    Sediment. Geol.

    (2018)
  • E. Garzanti et al.

    Congo River sand and the equatorial quartz factory

    Earth Sci. Rev.

    (2019)
  • E. Garzanti et al.

    Provenance of Thal Desert sand: Focused erosion in the western Himalayan syntaxis and foreland-basin deposition driven by latest Quaternary climate change

    Earth Sci. Rev.

    (2020)
  • S.M. Glynn et al.

    U-Pb zircon geochronology of the Dete-Kamativi Inlier, NW Zimbabwe, with implications for the western margin of the Archaean Zimbabwe Craton

    Precambrian Res.

    (2020)
  • B. Goscombe et al.

    Assembly of central Gondwana along the Zambezi Belt: metamorphic response and basement reactivation during the Kuunga Orogeny

    Gondwana Res.

    (2020)
  • N.D. Greber et al.

    New high precision U-Pb ages and Hf isotope data from the Karoo large igneous province; implications for pulsed magmatism and early Toarcian environmental perturbations

    Res. Geochem.

    (2020)
  • I.G. Haddon et al.

    The Mesozoic–Cenozoic interior sag basins of Central Africa: the Late Cretaceous–Cenozoic Kalahari and Okavango basins

    J. Afr. Earth Sci.

    (2005)
  • W.S. Hall et al.

    Igneous and detrital zircon U-Pb and Lu-Hf geochronology of the late Meso-to Neoproterozoic northwest Botswana rift: Maximum depositional age and provenance of the Ghanzi Group, Kalahari Copperbelt, Botswana and Namibia

    Precambrian Res.

    (2018)
  • R.E. Hanson et al.

    Mesoproterozoic intraplate magmatism in the Kalahari Craton: a review

    J. Afr. Earth Sci.

    (2006)
  • U.S. Hargrove et al.

    Tectonic evolution of the Zambezi orogenic belt: geochronological, structural, and petrological constraints from northern Zimbabwe

    Precambrian Res.

    (2003)
  • W.W. Hay

    Detrital sediment fluxes from continents to oceans

    Chem. Geol.

    (1998)
  • M.H.T. Hipondoka et al.

    Chronology of sand ridges and the Late Quaternary evolution of Etosha Pan, Namibia

    Geomorphology

    (2014)
  • K. Holmgren et al.

    Paleoclimatic significance of the stable isotopic composition and petrology of a Late Pleistocene stalagmite from Botswana

    Quat. Res.

    (1995)
  • K. Holmgren et al.

    Persistent millennial-scale climatic variability over the past 25,000 years in Southern Africa

    Quat. Sci. Rev.

    (2003)
  • S. Holzkämper et al.

    Late Pleistocene stalagmite growth in Wolkberg Cave, South Africa

    Earth Planet. Sci. Lett.

    (2009)
  • P. Huntsman-Mapila et al.

    Cryptic indicators of provenance from the geochemistry of the Okavango Delta sediments, Botswana

    Sediment. Geol.

    (2005)
  • S.E. Jackson et al.

    The application of laser ablation-inductively coupled plasma-mass spectrometry to in situ U–Pb zircon geochronology

    Chem. Geol.

    (2004)
  • J. Jacobs et al.

    The Kalahari Craton during the assembly and dispersal of Rodinia

    Precambrian Res.

    (2008)
  • Y. Al-Hajri et al.

    Scales of transient convective support beneath Africa

    Geology

    (2009)
  • P.A. Allen

    Sediment Routing Systems. The Fate of Sediments from Source to Sink

    (2017)
  • S. Andò et al.

    Metamorphic grade of source rocks revealed by chemical fingerprints of detrital amphibole and garnet

  • T.A. Baillieul

    A reconnaissance survey of the cover sands in the Republic of Botswana

    J. Sediment. Petrol.

    (1975)
  • G. Bertolini et al.

    Controls on Early Cretaceous desert sediment provenance in south-west Gondwana, Botucatu Formation (Brazil and Uruguay)

    Sedimentology

    (2020)
  • Cited by (32)

    View all citing articles on Scopus
    View full text