The Segmented Zambezi Sedimentary System from Source to Sink: 2. Geochemistry, Clay Minerals, and Detrital Geochronology

Abstract

Elemental geochemistry, Nd isotopes, clay minerals, and U-Pb zircon ages integrated by petrographic and heavy-mineral data offer a multi-proxy panorama of mud and sand composition across the Zambezi sediment-routing system. Detrital-zircon geochronology highlights the four major episodes of crustal growth in southern Africa: Irumide ages predominate over Pan-African, Eburnean, and Neoarchean ages. Smectite, dominant in mud generated from Karoo basalts or in the equatorial/winter-dry climate of Mozambican lowlands, prevails over illite and kaolinite. Elemental geochemistry reflects quartz addition by recycling (Uppermost Zambezi), supply from Karoo basalts (Upper Zambezi), and first-cycle provenance from Precambrian basements (Lower Zambezi). Mildly negative for sediments derived from mafic granulites, gabbros, and basalts, åNd values are most negative for sand derived from cratonic gneisses. Intrasample variability among cohesive mud, very coarse silt, and sand is principally caused by the concentration of Nd-rich monazite in the fine tail of the size distribution. The settling-equivalence effect also explains deviations from the theoretical relationship between åNd and TNd,DM model ages, suggesting that monazite carries a more negative åNd signal than less dense and less durable heavy minerals. Elemental geochemistry and Nd isotopes reveal that the Mazowe-Luenha river system contributes most of the sediment reaching the Zambezi Delta today, with minor supply by the Shire River. Sediment yields and erosion rates are lower by an order of magnitude on the low-relief Kalahari Plateau than in rugged Precambrian terranes. On the Plateau, mineralogical and geochemical indices testify to extensive breakdown of feldspars and garnet unjustified by the presently dry climate. Detrital kaolinite is recycled by incision of Cretaceous-Cenozoic paleosols even in the wetter lower catchment, where inefficient hydrolysis is testified by abundant fresh feldspars and undepleted Ca and Na. Mud geochemistry and surficial corrosion of ferromagnesian minerals indicate that, at present, weathering increases only slightly downstream the Zambezi River.Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive publisher-authenticated version is available on the publisher Web site.

The Zambezi deep-sea fan: mineralogical, REE, Zr/Hf, Nd-isotope, and zircon-age variability in feldspar-rich passive-margin turbidites

“…At present, however, all sediment delivered to the Indian Ocean is generated in the Lower Zambezi catchment downstream of Lake Cahora Bassa, where major left (northern) tributaries are the Luia and the Morrunguze draining high-grade rocks of the Southern Irumide Province, and the Shire, the outlet of Lake Malawi, which drains largely garnet-free mafic granulites of the Blantyre domain (Goscombe et al 2020). Among right (western) tributaries, the Mazowe and Luenha rivers sourced in the Archean Zimbabwe Craton and cutting across polymetamorphic gneisses remobilized during the Neoproterozoic Pan-African orogeny are estimated to provide between half and two-thirds of the sediment reaching the Zambezi Delta today (Garzanti et al 2022a). Additional detritus is derived from Permian -Triassic Karoo clastic rocks (Fernandes et al 2015).…”

“…7) of the Zambezi sedimentary system from coastal Mozambique to the deep-sea fan, including detritus generated in SW Madagascar. The composition of sediment carried by the Zambezi River and its tributaries, together with the geological and geomorphological characteristics of the vast Zambezi catchment, are illustrated and discussed in full detail in two companion papers (Garzanti et al 2021a and2022a).…”

“…In many segmented river systems where all detritus generated upstream is impounded in large dams, the mineralogical signal can be nevertheless transmitted downstream by reworking of previously deposited channel, floodplain, and terrace deposits, thus damping the dam effect (e.g., Garzanti et al 2000Garzanti et al , 2015Vezzoli et al 2016;Malkovski et al 2019;Thomson et al 2022). This may well occur where the dominant erosional foci are located upstream of the dams and the river flows across open lowland landscape downstream, which is hardly the case for the Zambezi River, where all compositional signatures change radically and irreversibly downstream of Lake Kariba first, and of Lake Cahora Bassa next (Garzanti et al 2021a(Garzanti et al , 2022a. The narrow and steep river valley lacks floodplains downstream of the dams, where the youthful river course is largely carved in bedrock, forming particularly impressive rapids along the Cahora Bassa Gorge (Davies et al 2000).…”

“…Cosmogenic nuclides in the uppermost Zambezi catchment (Wittmann et al 2020), sediment concentration in the middle catchment (Bolton 1984), and numerical models in the middle and lower catchment (Ronco et al 2010) suggest that annual sediment yields vary from as low as 2 ± 2 t/km 2 for the low-relief Kalahari Basin to 200 t/km 2 for Middle and Lower Zambezi tributaries flowing steeply across basement rocks exposed in the Archean Zimbabwe Craton or in the Proterozoic lrumide, Umkondo, and Zambezi Belts in southern Zambia, northern Zimbabwe, and western Mozambique. By integrating all available constraints (see Garzanti et al 2022a), our best guess is that 5 -10 million tons (≤ 10%) are generated in the upper catchment and trapped in the Kariba reservoir, 50 -60 million tons (60 -65%) in the middle catchment and trapped in the Cahora Bassa reservoir, and 20 -25 million tons (25 -30%) in the lower catchment and carried to the Zambezi Delta annually. These estimates imply that less than a third of the original sediment flux is making its way to the ocean today.…”

“…The present article, intended as a complement of previous studies on land, focuses on the Zambezi deep-sea fan deposited in the Mozambique Channel between the African landmass and Madagascar Island (Fig. 1), thus completing the source-to-sink study of the entire Zambezi sedimentary system (Garzanti et al 2014a(Garzanti et al , 2014b(Garzanti et al , 2021a(Garzanti et al , 2022a(Garzanti et al , 2022b. Our primary aims are to illustrate and discuss the variability of petrographic, heavy-mineral, elemental-geochemistry, Ndisotope, and U-Pb detrital-zircon geochronological signatures of Middle Pleistocene to Holocene turbidite deposits, highlight provenance changes in space and time, reconstruct sedimentary and geochemical budgets, and assess the relative amounts of detritus supplied from Africa versus Madagascar as well as the changing contributions from different rivers of SW Madagascar.…”