No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Lake Caprivi: a late Quaternary link between the Zambezi and Middle Kalahari drainage systems
Abstract
Recent geomorphological research in northern Botswana has shown the existence of an extensive palaeolake system adjacent to the Okavango Delta, with links to the Zambezi via the Chobe River at altitudes of 940-945 m and 936 m. This paper investigates the Chobe-Zambezi link and presents evidence in the form of terraces and major sand ridges, for the existence of Lake Caprivi, a palaeolake of c 2000 km2 at 936 m, ponded behind the Mambova Falls at the Chobe-Zambezi confluence. -from Authors
... 2), the Chobe River represents one of the few permanently flowing ('perennial') rivers in Botswana territory. The spectacular fluvial landscape approaching the confluence of the Chobe and Zambezi rivers (Fig. 7.2a) is characterised by a complex of alluvial channels and extensive floodplain wetlands, with bedrock-influenced channels more prominent in the vicinity of the features known as the Mambova Rapids (Clark 1950;Moore and Cotterill 2010) or the Mambova Falls (Shaw and Thomas 1988; Thomas and Shaw 1991). In previous publications, the extensive area (*3000-4000 km 2 ) of alluvial channels and floodplain wetlands upstream of the Mambova Rapids (the preferred name herein) has been referred to by various names, including the Chobe swamps (Nugent 1990; Moore and Cotterill 2010;McCarthy 2013), the Chobe-Zambezi or Chobe floodplain (Shaw and Thomas 1988;Moore et al. 2007), and the Zambezi wetlands (Pricope 2013;Burke et al. 2016). ...
... The spectacular fluvial landscape approaching the confluence of the Chobe and Zambezi rivers (Fig. 7.2a) is characterised by a complex of alluvial channels and extensive floodplain wetlands, with bedrock-influenced channels more prominent in the vicinity of the features known as the Mambova Rapids (Clark 1950;Moore and Cotterill 2010) or the Mambova Falls (Shaw and Thomas 1988; Thomas and Shaw 1991). In previous publications, the extensive area (*3000-4000 km 2 ) of alluvial channels and floodplain wetlands upstream of the Mambova Rapids (the preferred name herein) has been referred to by various names, including the Chobe swamps (Nugent 1990; Moore and Cotterill 2010;McCarthy 2013), the Chobe-Zambezi or Chobe floodplain (Shaw and Thomas 1988;Moore et al. 2007), and the Zambezi wetlands (Pricope 2013;Burke et al. 2016). Hereafter, the broader term Chobe-Zambezi channel-floodplain system is adopted to also encompass the more bedrock-influenced river reaches downvalley of the main area of alluvium and wetlands (Fig. 7.2a). ...
... Nugent 1990; Moore et al. 2007;McCarthy 2013) or reconstructions of Quaternary hydrological and lacustrine fluctuations (e.g. Shaw and Thomas 1988; Thomas and Shaw 1991). Furthermore, although part of the Chobe River lies within the Chobe National Park (located on the Botswana side of the Botswana-Namibia border- Fig. 7.2b) and is a focal point for wildlife populations and associated tourism activities (Fig. 7.2c), there is little specific visitor information regarding the channel-floodplain system and its importance for ecosystem service delivery. ...
In this chapter, the Chobe-Zambezi channel-floodplain system is defined as the fluvially influenced area that is located around and between the Chobe and Zambezi rivers approaching their confluence. This area is located in the ‘Four Corners’ region, the informal term given to the region where the Botswana, Namibia, Zambia and Zimbabwe borders meet. The large-scale structure and medium-term (102–105 years) development of the channel-floodplain system is related to a combination of tectonic activity and climatically-driven changes to flow and sediment supply. The system has developed in a region of subsidence that is related to the East African Rift System. Upstream of the Mambova Rapids, the modern sinuous, alluvial channels are flanked by extensive floodplain wetlands, with crevasse splays, gullies, oxbows, scroll plains, abandoned channels and backwaters (stagnant or slow-flowing, channel-like depressions) all widespread. Collectively, these fluvial landforms create the physical template for shorter-term water, sediment and ecosystem dynamics. A strong flood and drying season dynamic is evident; river stages typically rise from January and peak around April, before subsequently falling again. The Zambezi provides the largest flow volumes, with flow spreading gradually from north to south through a complex system of active and partially active channels and floodplain wetlands towards the Chobe. Along the two rivers, lateral channel migration and extension of splays, gullies and backwaters has been negligible over at least the last 40–50 years, with few new oxbows forming. To the east, both rivers cross the uplifting Chobe fault, with each river forming complexes of steeper, bedrock anabranching channels in the region of the Mambova Rapids. The two rivers ultimately coalesce ~10 km farther downvalley, and continue as the Zambezi River. A longer term (>106 years) developmental model is outlined, which posits that headward retreat of the Victoria Falls, at present located ~80 km downstream of the Chobe fault, will initiate a phase of erosion that will cross the fault in ~1–2 million years’ time. This phase of erosion will initiate deep channel incision, river network reorganisation and wider landscape denudation.KeywordsAlluvial channelBedrock channelChobe RiverDrylandsFloodplain wetlandsZambezi River
... This area is primarily fed by the tropically-sourced Cubango and Cuito Rivers. These are the major tributaries of the endoreic Okavango River, the Kwando/Chobe drainage system and, prior to its Pliocene (Lister 1979) or Early-Mid Pleistocene (Bond 1975;Shaw and Thomas 1988) capture and coastal reorientation, the Upper Zambezi. Some surface water reaches the eastern part of the Middle Kalahari via small rivers sourced in western Zimbabwe, notably the Nata, which flows seasonally into Sua Pan, in the east of the Makgadikgadi system. ...
... During peak flows, the impediment of the ridge results in an annual backflood up to 20 km westwards along the Chobe (Thomas and Shaw 1991). Evidence for periods of extreme flooding during the Quaternary is derived from erosional notches cut along the hill slope at 935 m above mean sea level (amsl), from calcretized alluvial terraces at 932-934 m amsl (Shaw and Thomas 1988), and from shorelines bounding the backflooding zone (Burrough and Thomas 2008). It is hypothesized that Zambezi flow made a significant contribution to the surface water of the Middle Kalahari during these periods of high discharge and back-ponding (Shaw and Thomas 1988;Burrough and Thomas 2008), providing an important link between these two sizeable hydrological systems during the Quaternary. ...
... Evidence for periods of extreme flooding during the Quaternary is derived from erosional notches cut along the hill slope at 935 m above mean sea level (amsl), from calcretized alluvial terraces at 932-934 m amsl (Shaw and Thomas 1988), and from shorelines bounding the backflooding zone (Burrough and Thomas 2008). It is hypothesized that Zambezi flow made a significant contribution to the surface water of the Middle Kalahari during these periods of high discharge and back-ponding (Shaw and Thomas 1988;Burrough and Thomas 2008), providing an important link between these two sizeable hydrological systems during the Quaternary. ...
The interior southern African basin (Kalahari) is a remarkable region, with a complex and dynamic environmental history and a long record of utilization by human populations during the late Quaternary. Paleoenvironmental reconstructions are beginning to provide a spatially detailed record of landscape and hydrological dynamics in the Kalahari, with a strong chronometric underpinning for records of environmental extremes. Theories concerning the distribution of early people in the landscape place great importance on the temporal dynamics of water availability, and may be particularly relevant in the Kalahari where there is significant evidence of hydrologic/climatic-driven landscape change. High amplitude environmental variability during MIS 6-2 is evidenced by periods of dune building within currently stabilized dunefields
and the intermittent existence of large lacustrine systems such as Megalake Makgadikgadi
that remain all but ephemerally dry under present-day conditions. That the wider Kalahari was, at times, a key resource for Stone Age populations is evident from the extensive occurrence of stone tools, most notably in association with the fluvial networks and lake basins of the Okavango-Chobe-Zambezi system. Today, these riparian
corridors link the semiarid
desert region to the southern subtropics and, in the past, drove environmental change in the Kalahari, potentially impacting the occupation and dispersal of hominins within the interior southern African basin.
... 4. The fine fraction of the basal unit shows a slightly better resemblance of elemental ratios to mafic sources (although still mostly felsic) than the lower unit (Table S7). Accompanied with the similarity of REE ratios between the basal unit and Zambezi-type sediments ( Figure 5, Table S7), these characteristics support the suggested westward orientation of the ancient Zambezi tributaries (Shaw & Thomas, 1988), that drain ample amounts of the Lufilian Belt and the Karoo basalts in the present (Figure 2). ...
... Based on our data, neither of these terranes can be ruled out and different studies suggest both cratons as possible source rocks. The lack of west-bank tributaries in the Gwayi River ( Figure 2) and the nature of fluvial deposits in NW Zimbabwe that were suggested to result from original westwards flow of streams towards the Okavango Basin (Shaw & Thomas, 1988), support the Zimbabwe Craton being the main source for sediments that were deposited during the Pliocene at OGN. Two additional observations support the Kasai Craton being the more probable source: (a) Geochemical provenance study showing that deeply buried sediments beneath the Okavango Delta were partially derived from Angola (Huntsman-Map ila et al., 2005) where the Kasai Craton is exposed (b) The southerly drainage system that operated throughout the Kalahari at the same time (Vainer, Erel, et al., 2018) could have delivered sediments from the Kasai Craton that is located to the north of the site. ...
The structural depression that occupies the Okavango Basin in southern Africa comprises a depo-centre within the intracratonic Kalahari Basin where sediments of the Cenozoic Kalahari Group have accumulated. The Okavango Basin has been formed due to stretching and subsidence at an area of diffused deformation, southwestwards to the main East African Rift System (EARS). Sediments from two full Kalahari Group sequences, located on opposite sides of the Gumare Fault that forms a major fault within the Okavango Basin, were studied to determine their provenance and chronology. Terrestrial Cosmogenic Nuclide (TCN) 26 Al/ 10 Be burial dating was used to constrain a chronostratigraphical framework, and Pb, Sr, and Nd isotopic ratios combined with geochemical and sedimentological analyses were applied to track the source areas of the sediments.Results indicate the following sequence of basin filling: (a) Accumulation between ca. 4-3 Ma during which the currently down-thrown (southern) block received a mixture of sediments mostly from the Choma-Kalomo, Ghanzi-Chobe, and Damara terranes, and possibly from the Lufilian Belt and/or Karoo basalts during earlier stages of deposition. Simultaneously, the up-thrown (northern) block received sediments from more distant Archean sources in the Zimbabwe and/or Kasai cratons, (b) Hiatus in sedimentation occurred at both sites between ca. 3-2 Ma, (c) Sediments on both sides of the Gumare Fault share a similar source (Angolan Shield) with minor distinct contributions to the downthrown block from the Kasai Craton and local sources input to the upthrown block, and (d) Regional distribution of aeolian sand since at least 1 Ma. The change in source areas is attributed to rearrangements of the drainage systems that were probably linked to vertical crustal movements on the margins of the Okavango Basin. The tectonically induced morphodynamics controlled the landscape evolution of the endorheic basin where vast lakes, wetlands and salt pans have developed through time. K E Y W O R D S cosmogenic nuclides dating (26 Al/ 10 Be), incipient rifting, intracratonic morphodynamics, Okavango Basin, provenance analyses (Pb, Sr, Nd isotopes)
... As similar dates occur at both higher and lower strandline levels these are therefore considered to , Shaw and Thomas (1988), Shaw (1993, 2002). The 41 ka wet interval is in general agreement with isotopic data from southern Botswana (e.g. ...
... While the three phases are represented by duricrusted strandlines at present heights above sea level of 943-945 m , 936 m and 924-910 m, evidence from TL dates suggests that more than one elevation was occupied at least during the two later palaeo-lake events. The last major palaeo-lake event recorded here is comparable to existing 14 C dates at c 41 000 BP mainly from the western margin of the MSB (Cooke, 1975;Cooke and Verstappen, 1984;Thomas, 1988). The range of strandline heights with similar dates may result from palaeo-lake levels moving rapidly and frequently forming transgressive/regressive cycles (Eugster and Maglione, 1978) during sustained drying. ...
This thesis presents the results of a multidisciplinary study of sediments, with an emphasis on geochemistry, with the following objectives : 1) improve our understanding of geochemical processes occurring as a result of erosion, transport and deposition in two rift b asin in their early stage of development; 2) reconstruct environments in these two regions of the rift during the Late Pleistocene and Holocene and 3) define sedimentary and geochemical criteria to allow us to discriminate tectonic setting and climate change in nascent and early stage rift basins. The two basins chosen to represent early stagerift development were 1) the Makgadikgadi-Okavango-Zambesi basin located in NW Botswana and 2) the Mpulungu Basin, located at the south-western extremity of Lake Tanganyika. The geochemical study of sediments and water of the Okavango Delta revealed the presence of elevated arsenic in the groundwater occuring as a result of the dissolution of oxides under reducing conditions. The presence of elevated arsenic in the sediments in linked to organic rich clays, deposited under lacustrine conditions during periods of higher rainfall and possible impoundment of the river by the Thamalakane Fault. The geochemical results were used as a proxy fro climate change in the Mpulungu basin. The results indicate a remarkable excursion which coincides with the Younder Dryas event. This environmental event appears to be the most important event in southern lake Tanganyika during the period 23-3 ka, which is characterized by the transport of weathered material into the basin due to a change in vegetation cover in the catchment area.
... As similar dates occur at both higher and lower strandline levels these are therefore considered to , Shaw and Thomas (1988), Shaw (1993, 2002). The 41 ka wet interval is in general agreement with isotopic data from southern Botswana (e.g. ...
... While the three phases are represented by duricrusted strandlines at present heights above sea level of 943-945 m , 936 m and 924-910 m, evidence from TL dates suggests that more than one elevation was occupied at least during the two later palaeo-lake events. The last major palaeo-lake event recorded here is comparable to existing 14 C dates at c 41 000 BP mainly from the western margin of the MSB (Cooke, 1975;Cooke and Verstappen, 1984;Thomas, 1988). The range of strandline heights with similar dates may result from palaeo-lake levels moving rapidly and frequently forming transgressive/regressive cycles (Eugster and Maglione, 1978) during sustained drying. ...
... The straight, subparallel pattern of east-bank Gwai tributaries evolved upon a linear Kalahari dune landscape. Lack of west-bank tributaries and fluvial deposits in NW Zimbabwe suggests that these streams drained originally westwards toward Botswana, and were captured progressively by headward erosion of the Gwai after establishment of the modern course of the Zambezi (Shaw and Thomas, 1988). In its final part, the Gwai cuts across a ridge of Paleoproterozoic basement (Hanson, 2003), where more youthful tributaries are sourced. ...
... After entering the Okavango Rift (Modisi et al., 2000;Kinabo et al., 2007), the Cuando River (here named Kwando, Linyanti and next Chobe) deviates sharply eastward along the Linyanti and Chobe Faults, following a tectonic depression that hosts large swamps and once large paleolakes (Shaw and Thomas, 1988). The depression, continuing into Zambia where it includes the low-gradient Kasaya and Ngwezi Rivers, has favoured the recent capture of the Kwando by the Zambezi. ...
This study presents an integrated mineralogical–geochemical database on fine-grained sediments transported by all major rivers of southern Africa, including the Zambezi, Okavango, Limpopo, Olifants, Orange and Kunene. Clay mineralogy, bulk geochemistry, Sr and Nd isotopic signatures of river mud, considered a proxy of suspended load, are used to investigate the influence of source-rock lithology and weathering intensity on the composition of clay and silt produced in subequatorial to subtropical latitudes.
Depletion in mobile alkali and alkaline-earth metals, minor in arid Namibia, is strong in the Okavango, Kwando and Upper Zambezi catchments, where recycling is also extensive. Element removal is most significant for Na, and to a lesser extent for Sr. Depletion in K, Ca and other elements, negligible in Namibia, is moderate elsewhere. The most widespread clay minerals are smectite, dominant in muds derived from Karoo or Etendeka flood basalts, or illite and chlorite, dominant in muds derived from metasedimentary rocks of the Damara Orogen or Zimbabwe Craton. Kaolinite represents 30–40% of clay minerals only in Okavango and Upper Zambezi sediments sourced in humid subequatorial Angola and Zambia. After subtracting the effects of recycling and of local accumulation of authigenic carbonates in soils, the regional distribution of clay minerals and chemical indices consistently reflect weathering intensity primarily controlled by climate.
Bulk geochemistry identifies most clearly volcaniclastic sediments and mafic sources in general, but cannot discriminate the other sources of detritus in detail. Instead, Sr and Nd isotopic fingerprints are insensitive to weathering, and thus mirror faithfully the tectonic structure of the southern African continent. Isotopic tools thus represent a much firmer basis than bulk geochemistry or clay mineralogy in the provenance study of mudrocks.
... , 1973 ; Butzer , 1984 ) . In the Makgadikgadi Basin , elevated lake levels have been proposed from 32 – 27 ka based on a set of cursory OSL ages 2 from the Moremaoto diatom beds along the Boteti River , and radiocarbon dates of 31 , 300 ( Shaw and Thomas , 1988 ) , 30 , 600 and 27 , 200 cal yr BP ( Shaw et al . , 1988 ) from neighbouring river terraces . ...
... identified at Alexandersfontein ( Butzer et al . , 1973 ; Butzer , 1984 ) . In the Makgadikgadi Basin , elevated lake levels have been proposed from 32 – 27 ka based on a set of cursory OSL ages 2 from the Moremaoto diatom beds along the Boteti River , and radiocarbon dates of 31 , 300 ( Shaw and Thomas , 1988 ) , 30 , 600 and 27 , 200 cal yr BP ( Shaw et al . , 1988 ) from neighbouring river terraces . A subsequent phase of high lake levels , referred to as the ' Lake Thamalakane phase ' , has been identified at 20 , 400 – 16 , 000 cal yr BP ( Shaw and Thomas , 1996 ) . Diatom - rich organic sediments from the presently dry bed of nearby Lake Ngami have been dated to 40 . 5 ka , and 21 , 500 – 17 , ...
Variations in the nature and extent of southern Africa's winter rainfall zone (WRZ) have the potential to provide important information concerning the nature of long-term climate change at both regional and hemispheric scales. Positioned at the interface between tropical and temperate systems, southern Africa's climate is influenced by shifts in the Intertropical Convergence Zone, the westerlies, and the development and position of continental and oceanic anticyclones. Over the last glacial–interglacial cycle substantial changes in the amount and seasonality of precipitation across the subcontinent have been linked to the relative dominance of these systems. Central to this discussion has been the extent to which the region's glacial climates would have been affected by expansions of Antarctic sea-ice, equatorward migrations of the westerlies, more frequent/intense winter storms and an expanded WRZ. This paper reviews the developing body of evidence pertaining to shifts in the WRZ, and the evolution of ideas that have been presented to explain the patterns observed. Dividing the region into three separate axes, along the western and southern margins of the continent and across the interior into the Karoo and the Kalahari, a range of evidence from both terrestrial sites and marine cores is considered, and potential expansions of the WRZ expansions are explored. Despite the limitations of many of the region's proxy records, a coherent pattern has begun to develop of a significantly expanded WRZ during phases of the last glacial period, with the best-documented being between 32–17 ka. While more detailed inferences will require the recovery and analysis of longer and better-dated records, this synthesis provides a new baseline for further research in this key region.
... Several authors have suggested that the Zambezi transferred water into the early Makgadikgadi system, with the water level regulated by overflow to the Zambezi Valley at Victoria Falls. Subsequent tectonism then altered the configuration of the system, by downwarping dong the Zambezi axis, andtor isolation of the Makgadikgadi by movement of Thomas and Shaw (1988) noted that fossil alluvium is extensive in the Caprivi Strip of Namibia, between the Chobe and Zambezi rivers, reaching 1050m at the Kafue watershed to the north of the Zambezi. However no landforms at the 940-945m level are encountered at the Chobe-Zambezi confluence, suggesting that deformation has been active in this area. ...
Between 1992, when Botswana and Namibia agreed to appoint a a Joint Team of Technical Experts to determine the disputed boundary between the two countries around Kasikili/Sedudu Island on the basis of that Treaty and of the applicable principles of international law, and 1999 when the International Court of Justice reached a Judgement in favour of Botswana with respect to the dispute, I was actively involved as a member of the Botswana Sedudu Task Force, which had set up by the Office of the President under the supervision of the Permanent Secretary for Political Affairs, Mr. Molosiwa Louis Selepeng. The Botswana Team Leader was the late Prof. Ian Brownlee, C.B.E., Q.C., deputized by the Hon, Abednego Tafa, who was then our Deputy Attorney General.
In the above context it largely fell on my shoulders to undertake historical research, including the discovery of past documents and map evidence, relevant to the case. This assignment ultimately involved extensive archival research in Namibia, South Africa, Zambia, the United Kingdom and the United States, as well as the Botswana archives and fieldwork in the Chobe District. My expectation is that my contributions as well as those of the other members of the Task Force are documented and should become available to interested members of the public through the appropriate declassification protocols over time. It is also my intent to donate private papers to the Botswana National Archives.
Public papers relevant to the case, as drafted by Botswana and Namibia, as well as the final Judgement itself, have in the meantime been published online by the website of the International Court of Justice, available @ https://www.icj-cij.org/case/98.
I have here shared from the site Botswana’s lead document. i.e. “Case Concerning Kasikili/Sedudu Island (Botswana/Namibia) Memorial of the Republic of Botswana, Vol. 1”, which outlines the core of the Botswana case, including chapters on history and map evidence with which this author was especially engaged as part of a wider collective effort.
... In the literature, two main factors contribute to increase the concentration of these elements in a hydro-sedimentary environment: (i) a significant input of ions from rivers or from hydrothermalism (Irkec and Ünlü, 1993;Galán and Pozo, 2011), and (ii) an evaporative system (Helgeson et al., 1969). Tectonic activity has been obviously operating in the study area, changing the watershed and basin geometry forming and reshaping various waterbodies in the Kalahari Basin (Shaw and Thomas, 1988;Burrough and Thomas, 2008;Vainer et al., 2021) that are clearly attributed to the result of such tectonic effects (Moore, 1999;Ringrose et al., 2005;McCarthy, 2013). Alongside the influence of neotectonics on the landscape evolution in the Chobe Enclave (Mokatse et al., 2022b), hydrothermal springs of low temperature have been reported in northern Botswana (Niles, 2012;Mukwati et al., 2018). ...
... It is noteworthy that both fluvial and aeolian sands along the final Chobe tract of the Cuando River get notably enriched in kyanite. This reveals mixing with sand originally fed by the upper Zambezi and reworked by the Cuando from the toe of the alluvial fan previously built by the Zambezi across the Okavango rift, between Lake Liambezi in the west and the Chobe depression in the east (Lake Caprivi of Shaw and Thomas, 1988). ...
Moustafa (2021) addressed four main comments in his discussion “On the structures of Abu Roash area and inverted basins of northern Egypt - Comments on “Surface expression of the Syrian Arc Kattaniya inverted basin in the Abu Roash area, northeast Western Desert, Egypt: Structural style and tectonic history” by Sayed et al. (2020)”. Two of these comments criticized the new structural mapping of Abu Roash outcrops and the proposed model for El Ghigiga-El-Hassana Folds based on the early geologic mapping of the Abu Roash area by Moustafa (1988). His other comments were concerning the NE-orientation and end of tectonic inversion of the extensional Jurassic -Lower Cretaceous basins in North Egypt including the Kattaniya basin. Except for two constructive memos on drafting errors in the regional geologic map of Abu Roash, the registered comments were inaccurate and can be refuted with the results of the comprehensive regional and integrated subsurface geologic mapping on the North Western Desert basins, the previous works of Moustafa and his collaborators and the figures included in his discussion as we will present in this reply. It was documented that the inversion time, intensity, and type (orthogonal versus oblique or compressional versus transpressional/transtensional) differ from one basin to another in the northern part of Egypt. The geologic structures of Abu Roash helped to define a transpressional (oblique) inversion tectonic model that started at the Santonian and ended by the Latest Senonian. This tectonic model represents the northeastern segment of the Kattaniya basin and does not generalize to the whole basin or other basins in North Egypt.
... It is noteworthy that both fluvial and aeolian sands along the final Chobe tract of the Cuando River get notably enriched in kyanite. This reveals mixing with sand originally fed by the upper Zambezi and reworked by the Cuando from the toe of the alluvial fan previously built by the Zambezi across the Okavango rift, between Lake Liambezi in the west and the Chobe depression in the east (Lake Caprivi of Shaw and Thomas, 1988). ...
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.
... More complete material is required to determine the affinities of this snail. Frankel, 1964, and Ngami is after Shaw & Thomas, 1988, Thomas & Shaw, 1991, and Shaw, 1985. ...
Fossil freshwater molluscan shells and steinkerns have been found at Simanya, on the Southwestern Bank of the Cubango River in Northern Namibia. The occurrence of fossils resembles those reported from other sites in the « Grès Polymorphes » subunit of the Kalahari System of Central Africa. The aim of this paper is to describe and interpret the Simanya fossils and to discuss their strati-graphic and palaeoenvironmental contexts. The silicified deposits (chert, chalcedony) in which the snails occur were extensively used by prehistoric peoples for manufacturing of stone tools.
... A smaller paleolake is associated with an elevation of 930 m a.s.l. and referred to as Paleolake Thamalakane (PLT). Shaw and Thomas (1988) proved the existence of a paleolake in the ZR, which they named Paleolake Caprivi (PLC). The PLC probably corresponds to the PLT stage. ...
Recent exploration has revealed that deep-seated and large groundwater reservoirs in Africa’s intracontinental basins can be regarded as an additional strategic resource for supply of drinking water. The origin, genesis and recharge of these groundwater reservoirs, however, are still poorly understood. A multidisciplinary approach involving remote sensing, geophysical surveys and hydraulic investigations, as well as hydrochemical and isotope studies, was pursued to gain better insight into the genesis and the potential of a recently discovered lower Kalahari aquifer (LKA) located in the Zambezi Region (Namibia). The study shows that regional tectonic activity associated with the propagation of the Okavango Rift Zone had a tremendous impact on the drainage evolution and hydrogeological setting of the region. Furthermore, there is geomorphological evidence that the LKA—prior to tectonic subsidence and burial—was part of a paleochannel of the upper Zambezi River. Hydraulic continuity could be confirmed by geochemical evolution down the flow path. Cation exchange combined with dissolution of calcite progressively produces alkalinity and sodium and consumes calcium in the north–south direction. Comparison of stable isotope content of the LKA with modern rainfall indicates that the recharge occurred under cooler climate conditions. Analysis of ¹⁴C concentrations and ³⁶Cl/Cl ratios show that the age of the groundwater exceeds 100 ka and is hence older than presumed. It is concluded that the assessment of the sedimentology, tectonic structures and geochemistry are key factors for understanding both the paleoclimatic and modern recharge processes of deep-seated aquifer systems.
... A smaller paleolake is associated with an elevation of 930 m a.s.l. and referred to as Paleolake Thamalakane (PLT). Shaw and Thomas (1988) proved the existence of a paleolake in the ZR, which they named Paleolake Caprivi (PLC). The PLC probably corresponds to the PLT stage. ...
Zusammenfassung
Semi-fossile Grundwasserleiter sind im Gegensatz zu fossilen Aquiferen noch an den hydrologischen Kreislauf angebunden und werden zu Teilen erneuert. In Afrika bilden diese Aquifere eine wichtige Quelle für die Trinkwasserversorgung und sind noch nicht vollständig erschlossen. Im nördlichen Teil des Kalahari-Beckens werden aktuell zwei in größeren Tiefen entdeckte poröse Grundwasserleiter hydrogeologisch erkundet: der Ohangwena II-Aquifer (KOH-2) im Grenzgebiet zwischen Angola und Namibia und der Lower-Kalahari-Aquifer (LKA) in der Zambezi-Region im Nordosten Namibias. Die hydrogeologische Ausprägung des KOH-2 wird bestimmt durch nachgewiesene sedimentäre Strukturen in Form eines Paläo-Deltas, die des LKA durch teilweise rezente tektonische Prozesse entlang einer Grabenbruchzone und einhergehenden Flussablenkungen durch rückschreitende Erosion. Hydrochemische und -isotopische Untersuchungen ergaben Hinweise, dass im LKA ein ursprünglich salziger Grundwasserkörper seit dem späten Pleistozän langsam aussüßt. Die Exploration solcher semi-fossiler Grundwassersysteme ist zwingend an die Erforschung geologisch-tektonischer und sedimentologischer Strukturen gebunden.
... Subtle tectonic movements may also have impacted upon the low-gradient drainage network in the past and led to the diversion of flow from the Zambezi River towards the Palaeo-Makgadikgadi system (e.g., Nugent, 1990). Alternatively, as occurs today at times of enhanced flood, high discharges in the Zambezi and Chobe rivers may have caused a backing up of water behind the Chobe/Zambezi confluence and the reversal of flow towards the Palaeo-Makgadikgadi system (Shaw and Thomas, 1988;Thomas and Shaw, 1992;Burrough and Thomas, 2008). ...
This study utilises geochemical provenancing of silcrete raw materials, in combination with chaîne opératoire analyses, to explore lithic procurement and behavioural patterns in the northern Kalahari Desert during the Middle Stone Age (MSA). New data from the sites of Rhino Cave, Corner Cave, and ≠Gi in northwest Botswana, combined with earlier results from White Paintings Shelter, reveal that the long distance transport of silcrete for stone tool manufacture was a repeated and extensively used behaviour in this region. Silcrete was imported over distances of up to 295 km to all four sites, from locations along the Boteti River and around Lake Ngami. Significantly, closer known sources of silcrete of equivalent quality were largely bypassed. Silcrete artefacts were transported at various stages of production (as partially and fully prepared cores, blanks, and finished tools) and, with the exception of ≠Gi, in large volumes. The import occurred despite the abundance of locally available raw materials, which were also used to manufacture the same tool types. On the basis of regional palaeoenvironmental data, the timing of the majority of silcrete import from the Boteti River and Lake Ngami is constrained to regionally drier periods of the MSA. The results of our investigation challenge key assumptions underlying predictive models of human mobility that use distance–decay curves and drop-off rates. Middle Stone Age peoples in the Kalahari appear to have been more mobile than anticipated, and repeatedly made costly choices with regard to both raw material selection and items to be transported. We conclude that (i) base transport cost has been overemphasised as a restrictive factor in predictive models, and (ii) factors such as source availability and preference, raw material quality, and potential sociocultural influences significantly shaped prehistoric landscape use choices.
... There is evidence across southern Africa for at least some of these, although the fragmentation of the record remains a constraint. Climate variability associated with the cooler period around 8.2 kyr (Alley et al., 1997 ) is recorded in other parts of Africa (see sections 4.4 and 4.5 ) and this period corresponds to a preces- Fig. 4.17 Combined luminescence, U/Th and calibrated radiocarbon chronology of Holocene environmental change in the Kalahari, derived from data for the following locations: (A) Gcwihabe (Drotsky ' s) Cave (Wayland, 1944 ;Cooke, 1975Cooke, , 1984Cooke and Verhagen, 1977 ;Shaw and Cooke, 1986 ;Burney et al., 1994 ;Railsback et al., 1994Railsback et al., , 1999Brook, G.A. et al., 1996Brook, G.A. et al., , 1997Brook, G.A. et al., , 1998Robbins et al., 1996 ); (B) Tsodilo Hills (Robbins et al., 1994 ); (C) Ncamasere and Dobe (Helgren and Brooks, 1983 ;Brook, G.A. 1995 ); (D) Okavango Panhandle (Nash et al., 1997(Nash et al., , 2006; (E) Lake Ngami (Shaw, 1985 ;Shaw and Cooke, 1986 ;Robbins et al., 1998 ;Shaw et al., 2003 ;Burrough et al., 2007 ); (F) Mababe Depression (Shaw, 1985 ;Burrough and Thomas, 2008 ); (G) Makgadikgadi Basin (Helgren, 1984 ;Burrough et al., 2009 ); (H) Chobe River (Shaw and Thomas, 1988 ;Burrough and Thomas, 2008 ); (I) Zambia palaeodunes (O ' Connor and Thomas, 1999 ); (J) Zimbabwe palaeodunes (Stokes et al., 1997a; (K) Etosha lunette dunes ; (L) Otavi Cenotes (Brook, G.A. et al., 1997(Brook, G.A. et al., , 1998; (M) Gcwihabe (Drotsky ' s) Cave pollen record (Burney et al., 1994 ); (N) Otjikoto pollen record (Scott et al., 1991 ); (O) Windhoek pollen record (Scott et al., 1991 ); (P) Gaap Escarpment tufas (Butzer et al., 1978 ;Beaumont and Vogel, 1993 ); (Q) Stampriet Aquifer (Heaton et al., 1983 ;Stute and Talma, 1998 ); (R) Auob and Kuruman rivers (Heine, 1982 ;Shaw et al., 1992 ); (S) SW Kalahari palaeodunes (Eitel and Bl ü mel, 1997 ;Stokes et al., 1997a, b ;Thomas et al., 1997Thomas et al., , 1998Bl ü mel et al., 1998 ;Lawson and Thomas, 2002 ;Bateman et al., 2003 ); (T) Kathu Vlei (Beaumont et al., 1984 ); (U) Wonderwerk Cave (Beaumont et al., 1984 ;Butzer, 1984 ); (V) Equus Cave (Scott, 1987 ). synchronous dune activation at Tsodilo Hills and in northern Namibia (Thomas et al., 2000. Dune activity is, however, evident in the southwest Kalahari (Eitel and Bl ü mel, 1997 ;Stokes et al., 1997a,b ;Thomas et al., 1997 ;Bl ü mel et al., 1998 ), with reduced moisture levels identifi ed at Equus Cave (Beaumont et al., 1984 ;Scott, 1987 ;Johnson et al., 1997 ). ...
Introduction Potential climate forcing factors Mediterranean North Africa The Sahara and the Sahel Equatorial Africa Southern Africa Synthesis Directions for future research References
... km. At a lower level than Lake Palaeo-Makgadikgadi, Palaeolake Thamalakane was recogned at 936m but only to the west of the pan system, along the distal end of the Okavango Delta, around Lake Ngami and the Mababe depression and in Caprivi (Thomas and Shaw, 1991; Shaw and Thomas, 1988). Lower, un-named palaeolake levels were also recognised in the Makgadikgadi pans area at 920 m. and 912 m., below which the water body broke up to form smaller separate bodies each of which functioned independently in terms of lower and younger stages of palaeolake levels. ...
... Moderate excursions in the ðD record from both the Congo and the Zambezi delta sediment records suggest both the hydrological system and regions to the east and west of the interior experienced substantial fluctuations in moisture input. High lake stands recorded in the Mababe Depression and in the Parakarungu backflooding zone (29, 25) between the Okavango and Zambezi systems (Shaw and Thomas, 1988;Burrough and Thomas, 2008) are coincident with negative ðD excursions (increased P-E) in both the West (Congo basin) and the east (Zambezi basin) and documented by a short flood pulse through the Okavango at w6 ka (Nash et al., 2006). Shortly after this at 5.0 AE 0.6 ka and again at 3.4e3.9 ...
... The Zambezi River has been divided into three distinct geomorphic units: (1) headwaters to Victoria Falls, (2) Falls to the edge of the Mozambique coastal plains, and (3) the stretch traversing the coastal plains to the Ocean (Wellington, 1955; Moore et al., 2007). It has been suggested that the modern course of the Zambezi River is quite recent; the upper and middle parts of the river having evolved separately and then joined together through several stages of river capture and tectonic activities (Shaw and Thomas, 1988, 1992; Goudie, 2005) perhaps in Pliocene or mid-Pleistocene (Thomas and Shaw, 1991). Based on several lines of evidences, Moore and Larkin (2001) suggested that the evolutionary sequence of the Zambezi River started in early Cretaceous when the upper Zambezi was probably linked to the Limpopo via the Shashe River. ...
Fluvial systems in tropical regions are not only the lifeline for modern civilizations there but also form an important continental archive for reconstructing Quaternary palaeoclimates and palaeoenvironments. Tropical rivers drain through a variety of geological and geomorphological settings across the world and have been studied for their hydrology, sediment transport characteristics, facies models, stratigraphic development and flood hazards. This paper presents an overview of the research on tropical rivers globally and identifies major gaps and unanswered research questions. As a part of the IGCP 582 on Tropical Rivers, the participants of this project are continuing to tackle many of these issues. The collection of papers in this special issue is an outcome of this activity and presents new data on processes and alluvial stratigraphy of the tropical rivers from China, India, South America and northern Australia.
... , radiocarbon ages; a-g, possible matching peaks in columns 1 and 4. 1: Cumulative percent by weight of grains <0 , <1 and <1·5 , 2: Drotksy's Cave clastic sediments (this paper), 3: Drotksy's Cave faunal remains indicating past wetter conditions (this paper), 4: Number of 14 C dates indicating past wetter conditions in the summer rainfall zone of southern Africa (Deacon & Lancaster, 1988), 5: Makgadikgadi Basin lake levels (Cooke & Verstappen, 1984;Shaw & Cooke, 1986;Thomas & Shaw, 1991), 6: Ngami, Mababe and Caprivi Basin lake levels (Shaw, 1986;Shaw & Cooke, 1986;Shaw & Thomas, 1988;Thomas & Shaw, 1991), 7: Drotsky's Cave speleothem ages (Cooke & Verhagen, 1977;Shaw & Cooke, 1986;Brook et al. 1990), 8: Gcwihaba Valley calcrete ages (Cooke & Verhagen, 1977; this paper), 9: Dry valleys of the middle and southern Kalahari (Shaw et al., 1992), 10: Tsodilo Hills lake (Brook et al., 1992), 11: Dobe Valley lake at Gi (Helgren & Brooks, 1983). ...
Test excavations conducted at Drotsky’s Cave have provided important new information on the paleoenvironment and archaeology of the western Kalahari desert during the late and terminal Pleistocene. An occupation layer dated to the terminal Pleistocene was rich in Late Stone Age artefacts, pieces of ostrich egg shell, the remains of carnivorous bullfrogs, springhare, and other fauna. A detailed sediment study, along with the evidence of Angoni vlei rat, climbing mouse, an aquaticXenopusfrog, and side neck turtle confirms that conditions were for the most part, substantially more moist than at present between approximately 30,000 and 11,000 years ago. Analysis of a diatom assemblage dated to the terminal Pleistocene implies that the currently dry Gcwihaba Valley was most likely flowing for much of the year. Our evidence supports findings made at other localities in the Kalahari documenting the existence of especially moist conditions during the terminal Pleistocene.
... This lake could only have been maintained by inflow from the Zambezi River, thus the Zambezi was at one stage diverted from its modern east flowing course to the south-west, and the Kafue may also have emptied into this lake (Moore et al., 2007). There is evidence of a complex history with the Upper Zambezi having been diverted at least twice (Shaw and Thomas, 1988;Moore et al., 2007). Uplift of the Chobe and Linyanti faults led to the current complex Chobe/Kwando/Zambezi floodplain systems in Caprivi. ...
The Zambezi River has a catchment area of 1.32 million km2, including parts of eight countries. Three divisions of the river are recognised: the Upper Zambezi separated from the Middle Zambezi by Victoria Falls, and the Lower Zambezi below Cahora Bassa gorge. The Okavango River is also linked to the Upper Zambezi system in wet years in an area of complex geomorphological history. Habitats include forested headwater streams, extensive floodplains, deep gorges, two large man-made lakes and an extensive delta. On floodplains, subsistence fisheries exploit the natural seasonal cycles, while the man-made lakes have commercial-scale fisheries for introduced kapenta, Limnothrissa miodon. Aquaculture is on a small scale, though with larger commercial cage culture enterprises on Lake Kariba. This paper summarises current knowledge on the fish faunas and their origins, the status of the different fisheries and their management, and the conservation status of the river's resources.
... It is possible that this ponding was caused by wind-deposited material blocking the channel flow into the pans (Cooke and Verstappen, 1984) but it is perhaps more likely that backflooding was caused by a lake high-stand event (Shaw et al., 1997). The deposits have been dated to a period spanning approximately 4000 years by two OSL ages of 31.7 AE 2.2 ka and 27.8 AE 2.4 ka (Shaw et al., 1997)Shaw and Thomas, 1988). The relict Okwa valley penetrates the Gidikwe Ridge in the south west corner of the basin. ...
A distinct series of beach ridges marking the former shorelines of large interconnected lacustrine basins in the Kalahari can be clearly identified from Landsat imagery and Shuttle Radar Topography Mission (SRTM) data. These basins, which form the terminal sump of the Okavango system in northern Botswana, are now almost completely dry. During the Quaternary they were intermittently occupied by large stable lake bodies and are thought to have periodically filled to a point of coalescence inundating an area that, at its largest extent, encompassed 66,000 km 2. Poor chronological control has previously limited the utility of this important palaeo-archive. As part of a region-wide lake palaeo-shoreline research programme , a systematic optically stimulated luminescence (OSL) dating programme has utilised a lightweight hydraulic auger to take samples at depth from relict shoreline features. Twenty drill-sites have generated 140 samples for dating, establishing a firm chronology for multiple lake full phases in all three component basins (Ngami, Mababe and Makgadikgadi) of this mega-lake. This paper presents the final set of ages in the programme, derived from four cores from the western and northeastern shorelines of Makgadikgadi, and uses these ages to establish a chronology of mega-lake high-stands during the last w300 ka providing a rare directly dated, long terrestrial record of positive hydrological excursions within the southern hemisphere.
... The extension of the alluvium into the Machili Basin and to the Kafue fl ats (Figures 15.14 and 15.15) suggests that the Kafue may have also emptied into this major lake. A less extensive lake (Lake Caprivi) is recorded by fossil shorelines at the 936 m level associated with the Mababe and Ngami depressions (Shaw and Thomas, 1988). These authors suggest that this lake post-dated the Lake Palaeo-Makgadikgadi at the 945 m level, although it also required infl ow from the Zambezi. ...
... The deposits have been dated to a period spanning approximately 4000 years by two OSL ages of 31.7 AE 2.2 ka and 27.8 AE 2.4 ka (Shaw et al., 1997). These ages are concurrent with radiocarbon dates from a 934 m asl terrace on the south side of Moremaoto (26,950 AE 200 14 C yrs BP), a 932 m asl terrace upstream of Moremaoto (25,860 AE 200 14 C yrs BP) and mollusc-bearing calcretes on a 925 m asl terrace downstream of Moremaoto (22,540 AE 200 14 C yrs BP) (Shaw and Thomas, 1988). ...
A distinct series of beach ridges marking the former shorelines of large inter-connected lacustrine basins in the Kalahari can be clearly identified from Landsat imagery and Shuttle Radar Topography Mission (SRTM) data. These basins, which form the terminal sump of the Okavango system in northern Botswana, are now almost completely dry. During the Quaternary they were intermittently occupied by large stable lake bodies and are thought to have periodically filled to a point of coalescence inundating an area that, at its largest extent, encompassed 66,000 km2. Poor chronological control has previously limited the utility of this important palaeo-archive. As part of a region-wide lake palaeo-shoreline research programme, a systematic optically stimulated luminescence (OSL) dating programme has utilised a lightweight hydraulic auger to take samples at depth from relict shoreline features. Twenty drill-sites have generated 140 samples for dating, establishing a firm chronology for multiple lake full phases in all three component basins (Ngami, Mababe and Makgadikgadi) of this mega-lake. This paper presents the final set of ages in the programme, derived from four cores from the western and north-eastern shorelines of Makgadikgadi, and uses these ages to establish a chronology of mega-lake high-stands during the last ∼300 ka providing a rare directly dated, long terrestrial record of positive hydrological excursions within the southern hemisphere.
The imprint of neotectonics is frequently obscured in low-relief environments by the sedimentary cover. This is the case in the Chobe Enclave, part of a pristine region of the Middle Kalahari Basin in northern Botswana, where numerous fossil landforms, such as sand dunes, pans, sand ridges, and carbonate islands, can be observed but whose detailed origins are unclear. In this study, a combination of near-surface geophysical surveys, sedimentological analysis, and optically-stimulated luminescence (OSL) dating were used to study the relationship between landscape development and neotectonic activity in one region of the Chobe Enclave. Specifically, electrical resistivity tomography (ERT) and ground penetrating radar (GPR) surveys were used to define the lateral and vertical distribution of sedimentary deposits associated with a prominent sand ridge and its margins. Sedimentological analysis and OSL dating of soil samples from pits taken along the survey lines were then used to calibrate the geophysical results and establish a chronostratigraphical framework. Our results lead to the following hypotheses regarding the evolution of this part of the Chobe landscape: (i) during the Late Pleistocene (>~25 ka), fluvial channels were buried due to epeirogenic movements, resulting in paleo-channels; (ii) between ~25 ka and ~6 ka, a paleo-lake shoreline formed, which resulted in the linear eastern margin observed on the considered sand ridge. Erosional bounding surfaces suggest the syndepositional uplift of the sand ridge morphologies during this time; (iii) a neotectonic influence on the evolution of the landscape is revealed by the incision of the sand ridge forming fluvial watergaps since ~6 ka ago; and (iv) due to continuous aeolian and fluvial reworking, the western-most ridge became gradually isolated from the floodplain resulting in an inherited relief. In conclusion, this investigation allowed the deformation caused by epeirogenic movements acting on a low relief landscape to be characterized in terms of landforms and sedimentary body properties.
The intersection of archaeological material with the landscape is investigated using OSL dating of landforms associated with Middle Stone Age (MSA) archaeology in the Makgadikgadi basin, Botswana. In this study, MSA archaeological sites on the Makgadikgadi pan floor date to two dry periods in the basin during the late Quaternary. Site formation at one site occurred during dry, or seasonally dry conditions that followed a period of high lake levels between 128 ± 18 ka and 81 ± 6 ka. The site was buried by sediments from a subsequent period of high lake levels dating to between 72 ± 5 ka and 57 ± 8 ka. At other investigated sites, the archaeological material was most likely deposited during a second dry period sometime after this. Overlying dunes are much younger (late Holocene) than the late Pleistocene lakebed sediments associated with the archaeological sites. Rapid burial of the archaeological sites by clayey sands has resulted in limited disturbance and weathering of archaeological material which appears to have only been exposed very recently, perhaps in the last 350 years when conditions have been particularly dry and susceptible to deflation. The spatial patterning of both sediment accumulation and deflation strongly influences archaeological visibility both within and around the Makgadikgadi basin.
Visible from space, the Makgadikgadi pans are one of the most distinctive geomorphic features of southern Africa. Now seasonally dry, this collection of pans is the sump of what was once one of Africa’s largest inland lakes. Formed by uplift along the Kalahari-Zimbabwe axis in the early Miocene and re-sculpted during the Pleistocene by both climatic events and neo-tectonism, the geomorphology of this basin tells a story of significant hydrological dynamism within northern Botswana, a story that has arguably influenced the evolution and dispersal of our own species (Chan et al. 2019). The boundary of the 37,000 km2 basin withholds a suite of landforms including saltpans, dunes, relict shorelines, lacustrine spits and fluvial deltas, each with its own inter-linked geomorphic history. Today, its landscape remains geomorphically active, providing one of the largest sources of dust in southern Africa with knock-on effects for global climate.KeywordsSaltpanPalaeolakeShorelineDiatomsSpitIslandsDustQuaternary
The Zambezi River rises at the center of southern Africa, flows across the low-relief Kalahari Plateau, meets Karoo basalt, plunges into Victoria Falls, follows along Karoo rifts, and pierces through Precambrian basement to eventually deliver its load onto the Mozambican passive margin. Reflecting its polyphase evolution, the river is subdivided into segments with different geological and geomorphological character, a subdivision finally fixed by man’s construction of large reservoirs and faithfully testified by sharp changes in sediment composition. Pure quartzose sand recycled from Kalahari desert dunes in the uppermost tract is next progressively enriched in basaltic rock fragments and clinopyroxene. Sediment load is renewed first downstream of Lake Kariba and next downstream of Lake Cahora Bassa, documenting a stepwise decrease in quartz and durable heavy minerals. Composition becomes quartzo-feldspathic in the lower tract, where most sediment is supplied by high-grade basements rejuvenated by the southward propagation of the East African rift. Feldspar abundance in Lower Zambezi sand has no equivalent among big rivers on Earth and far exceeds that in sediments of the northern delta, shelf, and slope, revealing that provenance signals from the upper reaches have ceased to be transmitted across the routing system after closure of the big dams. This high-resolution petrologic study of Zambezi sand allows us to critically reconsider several dogmas, such as the supposed increase of mineralogical “maturity” during long-distance fluvial transport, and forges a key to unlock the rich information stored in sedimentary archives, with the ultimate goal to accurately reconstruct the evolution of this mighty river flowing across changing African landscapes since the late Mesozoic.
Changes in lake status are often interpreted as palaeoclimate indicators due to their dependence on precipitation and evaporation. The Global Lake Status Database (GLSDB) has since long provided a standardised synopsis of qualitative lake status over the last 30,000 ¹⁴C years. Potential sources of uncertainty however are not recorded in the GLSDB. Here we present an updated and improved relational-database framework that incorporates uncertainty in both chronology and the interpretation of palaeoenvironmental data. The database uses peer-reviewed palaeolimnological studies to produce a consensus on qualitative lake-status histories, whose chronologies are revised and standardized through the recalibration of radiocarbon dates and the application of Bayesian age-depth modelling for stratigraphic archives. Quantitative information on absolute water-level elevation is preserved if available from geomorphological sources. We also propose a new probabilistic analytical framework that accounts for these uncertainties to reconstruct synoptic, integrated environmental signals. The process is based on a Monte Carlo algorithm that iteratively samples individual lake-status histories within the limits of their uncertainties to produce many possible scenarios. We then use Recursively-Subtracted Empirical Orthogonal Function analysis to extract dominant patterns of lake-status variability from these scenarios.
As a proof of concept, we apply this framework to 67 sites in eastern and southern Africa whose lake-status histories cover part of the late Pleistocene and/or Holocene. We show that, despite the sometimes large temporal and interpretation uncertainties, and the inclusion of highly discontinuous lake-status time series, identifying the major known millennial-scale climatic phases during the last 20,000 years is possible. Our framework was also able to identify an antiphased response between the lake basins in eastern and interior southern Africa to these changes. We propose that our new database and methodology framework serves as a template for efficient lake-status data synthesis, encourages the incorporation of lake-status data in more palaeoclimate syntheses, and expands the possibilities for the use of such data in the evaluation of climate models.
Namibia has a wide range of landscapes and the classification provided in the Atlas of Namibia is adopted by way of introduction to their nature and diversity. This is followed by a description of three main landscape regions: the Namib Desert, the Kalahari Desert, and the Great Escarpment. Finally, the chapter includes a regional analysis of the river systems and the coastline of Namibia, with feature boxes on two iconic landscapes, the Fish River Canyon and Sandwich Harbour.
The Okavango Delta is southern Africa's largest wetland ecosystem and probably the most pristine large wetland ecosystem in the world. Alex du Toit was the first to recognize the role of faulting in the origin of the Delta, proposing that the Delta lies within a graben structure related to the East African Rift Valley system. The history of its rivers is more ancient, extending back to the breakup of Gondwana. Two mantle plumes initiated the breakup of Gondwana: the 140 Ma Karoo plume and the 130 Ma Parana-Etendeka plume. Domical uplift and rifting associated with these plumes created two major river systems: the Okavango-Zambezi-Limpopo system and the Vaal-Orange system. The climate at the time of breakup was hot and humid and the interior experienced extensive erosion, so much so that by the Oligocene (ca. 30 Ma), the sub-continent had been planed to base level, rising only a few hundred metres above sea level and mantled by thick, leached soils (now known as the African Erosion Surface). Warping in the continent interior created uplifted arches and depressions, most notably the Kalahari basin. Arching severed the link between the lower Limpopo and its central African headwaters (Zambezi-Okavango), and a large lake formed in the Kalahari depression (Lake Palaeo-Makgadikgadi). This lake gradually disappeared, partly due to sedimentation but mainly due to the increasing dry climate. The East Africa Rift system commenced in the Afar about 30 Ma ago and began to propagate southwards. In southern Malawi and especially in Zambia, the path of rifting has not yet been clearly established and the region is characterized by numerous horsts and grabens. One of these grabens passes through the Okavango Delta. The formation of these grabens has profoundly affected the courses of the rivers in the region. The Okavango River debouches into the graben forming a large alluvial fan. Lakes have periodically existed at the toe of the Okavango fan where it abuts the bounding fault scarps, but these are not permanent. Some Okavango water discharges across the bounding fault scarps and flows into the Makgadikgadi depression to the southeast. The Okavango River catchment is largely underlain by Kalahari sand, which forms the major particulate sediment carried by the river. Consequently, sediment carried by the river is mainly fine sand, with little silt and mud. The dissolved solid concentration in the river water is low (ca. 40 mg/L) and consists mainly of silica and calcium and magnesium bicarbonates. However, the volume of water entering the Delta each year is large and hence the solutes constitute the largest component of the sediment carried into the Delta. The Okavango River discharges onto the alluvial fan where water is carried in channels that form the major primary distributaries. Channel margins are formed by vegetal material and are permeable, leaking water which sustains permanent swamps in the upper portion of the alluvial fan. The arrival of the seasonal flood increases the rate of channel leakage, forming the seasonal swamps on the lower fan. The advance of the flood water across the seasonal swamps is slow, as much of it infiltrates the ground, taking four to five months to traverse the 250 km length of the fan. Bedload is confined to channels and as water leaks through the channel margins, channel beds aggrade, increasing leakage, which further promotes bed aggradation. Channels eventually fail and water diverts elsewhere. Channel formation and failure results in a constant shift in the distribution of water across the Delta surface. The demise of a channel system results in desiccation of the surrounding peat, which is then destroyed by slow-burning peat fires. Nutrients and fine particulate material accumulated in the peat is released, enriching the soil. Most of the water delivered to the Delta annually is lost to the atmosphere by evapotranspiration because of the semi-arid climate. Transpiration of groundwater by terrestrial plants is the dominant means of water loss. The high transpiration rate of trees is particularly important: trees grow on islands and their transpiration lowers the water table beneath islands so there is a net flow of water towards islands. Solutes, especially silica and carbonates of calcium and magnesium, precipitate, leaving only the very soluble sodium carbonate in solution. Its concentration rises and impacts on the vegetation, resulting in a zonation in the distribution of plant species on islands. The salinity of the groundwater ultimately rises to the point where gravity induced advection occurs, thus transferring the sodium carbonate to the deep groundwater. This process prevents the formation of surface saline brine in the Delta and surface water remains fresh. The accumulation of precipitated solutes results in expansion and this form of chemical precipitation is the major mechanism of sedimentation in the distal regions of the Delta. Islands are mainly initiated as a consequence of termite activity during dry periods. Sand ridges which form by channel bed sedimentation may also result in islands. Constant changes in the distribution of water across the fan due to channel failure have profound effects on the ecology of the Delta: regions of swamp may revert to dry land, when rain flushes accumulated salts from the island soils; and formerly dry areas become seasonally or even permanently flooded. Such constant changes, operating on time-scales of decades to centuries, underpin the immense habitat diversity of the Okavango Delta.
The Kalahari features a long-lived lacustrine system which may exist since the Early Pleistocene. The emergence of an extant cichlid fish radiation from this (palaeo-) lake during the Middle Pleistocene indicates an ancient lake character. The early history of the system remains speculative, but it is established that lake extensions matching modern Lake Victoria in size have occurred during the Late Pleistocene. It has been assumed that the hydrographical dynamics chiefly depended on the inflow from the Okavango River and thus on ITCZ-controlled precipitation. Our studies, which focused the hydromorphological and palaeolimnological development of the Makgadikgadi Basin during the last 50 ka, suggest that from c. 46¿16 ka it did not receive water from the Okavango River but from palaeo-rivers located in the northern and south-western catchment. A northward shift of the winter rainfall zone during the Last Glacial Maximum sustained a high lake level for a period of c. 6 ka. During Heinrich Event 1 (17¿16 ka) the lake probably desiccated abruptly and completely. Higher lake levels, controlled by water from the Okavango river system, were reached again during the Holocene before the lake dried up in the middle of the last millennium.
A 61 cm core through a speleothem column in Drotsky's Cave, Botswana, has yielded a U-series dated pollen record of Holocene vegetation changes in the Kalahari Desert. Between c.10000-7000BP, the site was surrounded by an arid grassland with dry-adapted trees and shrubs such as Acacia and Commiphora. An increase in pollen of Combretaceae and Cyperaceae, and the appearance of pollen of such mesic savanna plants as Grewia between c. 7000—6000 BP, may indicate the onset of slightly wetter conditions. Combretaceae and other mesic pollen types increase between c. 6000-3000 BP, but the late Holocene record appears variable, with the period c. 5000-4000 BP being somewhat dryer than the millennia before and after. It would appear that Kalahari vegetation has changed less during the Holocene than many other African environments previously investigated. Despite the limitations imposed by pollen analysis of cave sites, the record from Drotsky's Cave shows good agreement with regional trends derived from other studies. The use of U-series dated pollen spectra derived from speleothems holds potential for expanding the geographic scope of palynological studies to areas lacking suitable lakes and bogs, and for extending the dated pollen record back an order of magnitude beyond the 14C timescale.
This article presents an overview of palaeofluvial geomorphology research in southern Africa. For the purposes of this article this includes South Africa, Zimbabwe, Namibia, Lesotho, Swaziland and Botswana. Although interest in fluvial systems has a long history in southern Africa, the scientific study of rivers was initiated by the discovery of the first alluvial diamond along the banks of the Orange River in 1867. Since then, significant progress has been made in unravelling the fluvial history of southern Africa from the early Archaean Ventersdorp Contact Reef River to modern channel process studies. The development of an understanding of palaeofluvial systems has occurred along two main lines. The first was alluvial diamond exploration work undertaken by the large mining houses. The second line was of a more 'academic' interest and included determining the impact of superimposition. tectonics, base level anti climate changes. The review suggests that southern Africa fluvial systems have shown large-scale changes in drainage pattern, discharge and sediment yield and that these can be related to a complex set of causative factors including the geological template, the Jurassic rifting of Gondwana, tectonic episodes and climate change.
In the absence of closed sites, fossiliferous strata and regional or even local stratigraphic frameworks, Quaternary scenarios for the Kalahari have been developed only in the past 20 years based on deductive geomorphological studies. Landforms studied include palaeo-lakes, caves, pans, dry valleys and spring deposits, all indicative of wetter episodes. Attempts to establish the presence of episodes of aridity from the vast linear dune fields have not been successful, primarily because the function of linear dunes has been misunderstood. A chronological framework for the past 40,000 years has been established, based on a corpus of 239 radiocarbon dates, of which nearly half, using calcrete as the dating medium, require careful interpretation. One of the most important findings is that the entire Kalahari experienced a wetter episode from 16,000–13,000 B.P., followed by a period of lowering water tables and return to the present climatic mean. The Kalahari is thus out of phase with Africa north of the Equator, and with other major sub-tropical deserts. This has implications for global palaeo-climatic modelling, although the phenomenon could be explained by the orbital forcing hypothesis. The recent use of U/Th dating on cave sinters suggests this pattern extends back into earlier glacial/interglacial cycles.
Linear dune features in south-central Angola have been mapped using recently released Soviet General Staff topographical maps, satellite imagery, aerial photography and limited ground truthing along the Angolan/Northern Namibian border. Analysis indicates that there are a range of relict aeolian features extending as far as 14°E 15°S. In the wetter eastern sector, extensive relict linear dunes (in the order of 200 km in length) are highlighted by vegetation contrasts and a trellis drainage pattern, which feeds into the main southward flowing drainage. In the drier western sector, the linear features are less extensive and appear to have a more direct association with reworked fluvial deposits of the Cuvelai and Cunene drainage systems. These features provide evidence that linear dunes were present over a greater area of the western Mega Kalahari than previously thought (Thomas, 1984). Recent studies in the Caprivi Strip (Thomas et al., 2000) and western Zambia (O'Connor and Thomas, 1999) can be linked to the identified linear features which form a continuation of this northern dune sector, interrupted only by the main drainage channels of the Zambezi, Cuando, Cuito and Cubango Rivers.
Kalahari dry ‘fossil’ valleys (mekgacha) have been little used in palaeoenvironmental reconstruction, partly because the origins and functions of the valleys are a subject of debate, and partly because they contain few datable sediments. In the Middle Kalahari the valleys are endoreic, and, following a decline in ground-water tables owing to human interference over the past 150 years, rarely contain water. Evidence from three sites suggests that the valleys contained standing water between 15000 and 12000 yr BP, and that the main Okwa-Mmone system flowed into palaeolake Makgadikgadi at 920 m a.s.l. during this period.
The Southern Kalahari contains the exoreic Molopo network which is hydrologically more active, being prone to surface flows in response to extreme rainfall events. The spring—fed Kuruman valley has two terraces, the lower composed of a series of flood deposits of late Holocene age. The flood of February 1988 provides a useful analogue for the conditions under which these deposits were formed, and of the temporal regime of the Kuruman River. Comparison with adjacent sites suggests that evidence put forward to indicate cycles of climatic amelioration in the last 4000 yr may, in part, represents the effect of extreme precipitation events. Historical floods show a strong correlation to Southern Oscillation high-phase (cold event) episodes, suggesting possibilities for extension of the Southern Oscillation record through palaeoflood studies.
A shell of the gastropod Potadoma was discovered in fluvio-lacustrine sediments of the Kalahari (Botswana). A late Pleistocene age of the sediments of ca.
46,000cal. years BP was established by radiocarbon dating. It is the first record of the freshwater genus Potadoma from southern Africa. Modern counterparts as well as other Pleistocene species of Potadoma had been reported elsewhere from (palaeo-) habitats located at least 2,000km further in the north, e.g., in the Congo Basin.
So far it can be only speculated about possible causes for such a disjunct distribution. The discovery of Potadoma and five more fossil gastropods from the same sediments was used to outline palaeolimnological features. Considering the
geomorphological setting, the assemblage of fossil gastropods indicates that the nowadays mainly dry Boteti River was permanently
flowing through the western Makgadikgadi Pans ca. 46,000cal. years BP. The existence of such a riverine palaeoenvironment
in the pans demonstrates that the so-called Lake Palaeo-Makgadikgadi was comparatively small during that time, not spanning
the area of investigation.
The Kalahari basin formed as a response to down-warp of the interior of the southern Africa, probably in the Late Cretaceous. The down-warp, along with possible uplift along epeirogenic axes, back-tilted rivers into the newly formed Kalahari basin and deposition of the Kalahari Group sediments began. Initial deposition of basal gravels occurred in the channels of the Cretaceous rivers, with other unsorted gravel beds deposited at the base of scree slopes along the edges of valleys and fault-bounded structures. The accumulation of gravels continued as the down-warp of the basin progressed with interbedding of the gravel layers with sand and finer sediment carried by the rivers. Thick clay beds accumulated in the lakes that formed as a result of the back-tilting of rivers, with sandstone being deposited in braided streams interfingering with the clays and covering them in some areas as the shallow lakes filled up with sediment.
The longitudinal profile of the Zambezi River forms two concave-upwards sections, with their boundary at the Victoria Falls. This form has been ascribed to the process of pediplantation and the Victoria Falls identified as one of several knick points that have traversed the river since the breakup of Gondwanaland. An alternative model explains the river's long profile by suggesting that the Upper and Middle Zambezi evolved as entirely separate river systems, which only joined together in comparatively recent times.The alluvial sequence of the Middle Zambezi is described and interpreted in terms of the latter hypothesis. The river capture event caused a change in grade and is marked by a deposit believed to record a cataclysmic flood. Capture is inferred to have resulted from overtopping of the lake that formed the end sink of the proto-Upper Zambezi, rather than from headward erosion of the proto-Middle Zambezi. This event is dated, from archaeological evidence, to the peak of the last interglacial, at the end of the Middle Pleistocene. Subsequent drainage diversions between the Middle Zambezi and the Kalahari are interpreted as the product of rifting of the Chobe Graben and aggradation of the Chobe Swamps.River capture by overtopping implies high rainfall over Central Southern Africa at the peak of the last interglacial, which is contrary to predictions that Africa's rain belts then lay north of their modern mean positions. This anomaly is resolved by postulating a southern polar warming episode at that time and agrees with suggestions that the last interglacial was marked by large scale ablation of the West Antarctic Ice Sheet.
The Middle Kalahari of northern Botswana contains three palaeolake basins, with shorelands enclosing an area of 60,000 km2, linked to the Okavango Delta and the adjacent Chobe and Zambezi rivers. Two major stages of the palaeolake system at 940-945 m (Lake Palaeo-Makgadikgadi) and 936 m (Lake Thamalakane) are identified, prior to the final desiccation of the lakes over the last 2,000 years. The former is tectonic in origin, and last existed c. 35,000 B.P., whilst the Lake Thamalakane stage has existed on a number of occasions, particularly at 17,000–12,000 B.P., and can be explained by increased precipitation and inflow feeding a complex system of fault-controlled lakes, ultimately controlled by the Mambova falls at the Chobe-Zambezi confluence. A number of factors influencing the behaviour of the Okavango Delta and its adjoining basins can be identified on different time scales, and are discussed in relation to planning for future water development.
This work considers new evidence for palaeo-environmental change taking place during the Pleistocene in northern Botswana. Duricrusted strandlines along the northeastern margin of Sua Pan provide palaeo-environmental data pertaining to the Makgadikgadi subbasin (MSB) with inferences regarding the larger Makgadikgadi–Okavango–Zambezi (MOZ) rift depression. Field, XRD and geochemical data show that MSB strandlines comprise calcretes (LU1 type), MgO-rich calcretes with silica (LU2 type), sil-calcrete (LU3 type) and silcrete (LU4 type). Early freshwater episodes appear to have been followed by calcrete-dominated drying phases interspersed with repeated silcretisation. Calcretisation through pan littoral sediments may have been both biogenically and environmentally induced. Calcite precipitation was in part controlled by the Mg/Ca ratio of pore water in the pan littoral zone suggesting closed basin type evaporative conditions, which were followed by a major desiccation interval. Phases of silcrete precipitation appear to be related to periods when the geochemistry of the lake littoral more closely resembled present-day Na–CO3–SO4–Cl-type brines. Silica saturated acidic, moderately saline groundwater preceded Si precipitation which took place as the pH reduced. Si mobilisation occurred (inter alia) as a result of quartz grain dissolution enhanced by diatoms, bacteria and algal growth in the moist pan littoral. SiO2-rich pore waters migrated through cracked and desiccated calcrete into areas of lower salinity and lower pH resulting in preferential calcite removal and silcrete precipitation. Approximate TL dates imply that exposed littoral sand underwent calcretisation during the drying phases of extensive palaeo-lakes which occurred prior to 110 ka, 80–90 ka and 41–43 ka. These wet periods compare fairly well with Vostok core chronologies for southern Africa.
A systematic drilling and optical dating programme on Middle Kalahari beach ridge (relict shoreline) sediments has enabled the identification of multiple episodes of lake high stands of an extensive palaeolake system at the terminus of the Okavango Delta, northern Botswana. This paper presents 23 ages from the Mababe Depression and establishes four shoreline construction phases in the late Quaternary coeval with other sub-basin lake high stands (Lake Ngami). These synchronous lake phases result from a coalescence of the sub-basins into a unified palaeolake, Lake Thamalakane, covering an area of ∼ 32,000 km2. Six additional ages are also presented from the Chobe enclave to the north of the basin where shoreline ridges were emplaced at the same time as Lake Thamalakane phases. This suggests that increased flow in the Chobe and Zambezi system significantly contributed to the Middle Kalahari lake phases in both the post-glacial and Holocene periods. The integration of these new data and their compatibility with other regional and tropical palaeo-archives is discussed in the light of understanding Quaternary climate drivers within the Kalahari.
ResearchGate has not been able to resolve any references for this publication.