INTRODUCTION

This paper is written as a part of project on the Pleistocene sediments of the submarine continental margins of the Atlantic Ocean. In the framework of this project, we published papers on the Scotia Sea (Levitan et al., 2020) and submitted papers on the Caribbean Sea and the Gulf of Mexico. This work for the first time considers data on the Pleistocene sediments from the submarine part of the Southwestern African margin ascribed to the Atlantic passive margin.

The studies were conducted using the comparative-lithological method by N.M. Strakhov (1945), volumetric method by A.B. Ronov (1949), and facies analysis of oceanic sediments by I.O. Murdmaa (1987).

In this work, we are based on the “old” stratigraphic scheme of Quaternary sediments (Gradstein et al., 2004), in which Pleistocene is subdivided into two main parts: Early Pleistocene (roughly from 1.8 to 0.8 Ma) and Middle–Late Pleistocene (from 0.8 to 0.01 Ma). They are separated by the Matuyama to Brunhes geomagnetic boundary. The bottom of the Early Pleistocene in the magnetic stratigraphy is represented by the upper boundary of the Olduvai subchrone. The end of the Late Pleistocene coincides with the beginning of the Holocene. According to the Russian version of Quaternary stratigraphy, the lower Pleistocene corresponds to the Eopleistocene, while the middle and upper Pleistocene, to the Neopleistocene (Alekseev et al., 1997).

In the studied area, the stratigraphic subdivision and correlation of sediments (including Quaternary sediments) were made using biostratigraphy (based on planktonic foraminifers, calcareous nanoplankton, and diatoms) and magnetostratigraphy (see below references on the reports of the Deep-Sea Drilling Project).

The names of the lithotypes from these reports were controlled by authors using presented description of smear slides and carbonate content in sediments.

MODERN SEDIMENTATION CONDITIONS

This section presents the results of study of the region extending from equator to 35° S and from the African coastal line to a depth of 3000 m. A relatively narrow zone of submarine continental margin from the north to the south intersects equatorial, tropical, moderate humid, and arid climatic zones.

From the north to the south, the submarine Southwestern African margin borders with such large structures as the deep-water Angola Basin, the Walvis Ridge, and the deep-water Сape Basin (Fig. 1).

Fig. 1.
figure 1

Position of the deep-sea drilling holes. Symbols: (1) deep-sea drilling holes; (2) currents (modified after Berger et al., 1998); (3) isobaths (in m); (4) areas with elevated primary production (modified after Berger et al., 1998), southern area—Benguela upwelling. Dashed line shows the Angola–Benguela hydrological front (ABF). (BC) Benguela current; (SEC) Southern Equatorial Current; (SECC) Southern Equatorial Countercurrent. (I) Angola deep-water basin; (II) Walvis Ridge; (III) Cape deep-water basin.

Full size image

The surface circulation system is mainly controlled by the Benguela Current flowing northward (Fig. 1) and representing the eastern part of the giant South Atlantic circulation system. The southeastern trade winds blowing from the Namib Desert and Ekman mechanism provide the formation of the west-trending discharge current. Therefore, the structure of vertical circulation is mainly determined by the Benguela (Namibian) upwelling, in which jet deviating northward from the Antarctic Circumpolar Current raises nutrient-rich subsurface waters from a depth of 80–50 m to the surface of the southern part of the studied region. This results in the elevated primary production with values over 180 g C cm2/yr along the entire margin and several times higher values in the Benguela upwelling zone (Berger et al., 1998). In the modern epoch, the influence of this zone of high primary production is expanded to a depth approximately 1000 m. The Benguela upwelling is the best expressed south of 21° S. The intensity of the upwelling in the modern epoch is defined by the wind strength (Nelson, 1992). Once in ten years, the El Nino phenomenon in this area brings warm and saline surface waters from the north. This process is accompanied by the termination of upwelling and mass fish mortality (Baturin, 2004).

Different circulation pattern is observed north of the Walvis Ridge–African continental margin junction zone (Fig. 1). In this area, upwelling is replaced by another large circulation cell (Berger et al., 1998).

Continental terrigenous influx was supplied by the Congo, Kunene, and Orange rivers (from the north to the south) (Fig. 1); as well as by aeolian way, mainly from the Namib Desert.

The northern part of the studied margin (from 0 to approximately 15° S) in the modern epoch is dominated by terrigenous sedimentation (with a small spot of diatom clay in the influence zone of the Congo river), while the southern part is dominated by planktonic carbonate sediments, which on the shelf and continental slope in the Benguela upwelling influence zone are supplemented by foraminiferal sands, glauconite sands (on the shelf), diatom ooze, and diatom clays with phosphorite nodules (Emel’yanov et al., 1989–1990; Baturin, 2004).

FACTUAL MATERIAL AND METHODS

The submarine Southwestern African margin was drilled by holes during the following deep-sea drilling legs: DSDP 40 (Bolli et al., 1978) and 75 (Hay et al., 1984); and ODP 175 (Wefer et al., 1998). The position of the holes is shown in Fig. 1.

The lithological and stratigraphic data on the Quaternary deposits obtained during these cruises were used to compile the Neo- and Eopleistocene lithological-facies maps with isopachs on a scale 1 : 10 000 000 in the transverse azimuthal equal-area projection. GEBCO map published in 2004 was used as the bathymetric base (www.gebco.org).

Then, the maps were processed with A.B. Ronov’s volumetric method to calculate areas and volumes of mapped lithological gradations. Data on the water content and density of natural sediments presented in the reports on the aforementioned legs allowed us to recalculate the volumes into masses of dry sediment and, correspondingly, to determine their accumulation rates.

RESULTS

Neopleistocene. The Neopleistocene lithological-facies map (Fig. 2) demonstrates well expressed longitudinal and transverse facies zoning. The northern part up to approximately 14° S is dominated by terrigenous hemipelagic clays, which are replaced by the diatom clays on the lower continental slope in the Congo transect. Further southward up to approximately 22° S, these sediments grade into carbonate zone with fields of diatom nanoclays (slightly northward of the Kunene mouth), foraminiferal–coccolithic clays on the shelf and continental slope, as well as foraminiferal–coccolithic ooze on the northwestern termination of the Walvis Ridge. Southward (approximately up to the Orange transect), there is the influence zone of the Benguela upwelling with alternating foraminiferal–coccolithic clays, hemipelagic clays, coccolithic ooze, and diatom–coccolithic clays. These sediments are characterized by the high glauconite admixture and ubiquitously contain phosphorite nodules. The southernmost part of the studied margin in the Neopleistocene was covered by alternating units of coccolithic ooze and carbonate turbidites.

Fig. 2.
figure 2

Neopleistocene lithological-facies map, Symbols: (1) hemipelagic clays; (2) carbonate turbidites; (3) nano-ooze; (4) foraminiferal nano-ooze; (5) nanoclays; (6) foraminiferal nanoclays; (7) nanodiatom clays; (8) diatom clays; (9) deep-sea drilling holes; (10) isopachs.

Full size image

It should be noted that practically all studied sediments contain a few percents of Сorg. Thereby, the largest contents were found in two sites: in the Congo transect and in the Benguela upwelling influence zone.

Sediments over 100 m thick are confined to the mixing zone of the Congo River and seawaters, as well as to the Benguela upwelling zone. The most part of the submarine margin is covered by sediment sequence from 100 to 50 m thick, while sediments < 50 m thick in the Neopleistocene were developed over almost half of the margin in two areas: between 7° and 22° S and south of the Orange mouth. The sediment thickness systematically decreases from the coast to the pelagic zone over the entire extent of the studied region.

Processing of the described map using A.B. Ronov’s volumetric method showed that the mapped area is 226.9 thou. km2 and the volume of accumulated sediments is 14.7 thou. km3 (Table 1). By volume, the Neopleistocene sediments are arranged in the following series (in decreasing order, vol %): hemipelagic clays 37.4, nano-ooze 18.4, carbonate turbidites 9.5, nanoclays 8.8, coccolithic–foraminiferal clays 8.2, diatom–coccolithic ooze 6.8, diatom clays 4.8, and diatom–coccolithic clays 4.1. Recalculation of the sediment volumes for dry sediment mass (Table 2) showed that the total mass of the Neopleistocene sediments in the studied area is 11.4 × 1018 g and includes 5.4 × 1018 g terrigenous sediments, 4.8 × 1018 g carbonate sediments, and 1.2 × 1018 siliceous sediments.

Table 1. Areas (S, thou. km2) and volumes (V, thou. km3) of Pleistocene sediments on the submarine margin of Southwestern Africa
Full size table
Table 2. Mass of dry sediment (M, 1018 g) and accumulation rate (I, 1018 g/Ma) of Pleistocene sediments of the Southwestern Africa submarine margin
Full size table

Eopleistocene. The facies structure of the Eopleistocene lithological-facies map (Fig. 3) resembles that of the Neopleistocene. However, the area occupied by weakly siliceous and siliceous sediments (including those on the surface of the northeastern termination of the Walvis Ridge) becomes wider, the thickness of terrigenous sediments decreases, while that of siliceous sediments, increases.

Fig. 3.
figure 3

Eopleistocene lithological-facies maps. Symbols: (1) hemipelagic clays; (2) carbonate turbidites; (3) nano-ooze; (4) foraminiferal nano-ooze; (5) nanoclays; (6) foraminiferal nanoclays; (7) nanodiatom ooze; (8) nanodiatom clays; (9) diatom clays; (10) isopachs (in m); (11) deep-sea drilling holes.

Full size image

Calculations showed that the volume of the Eopleistocene sediments is 14.8 thou. km3 (Table 1). The sediment types occupy the following volumes (in vol %): diatom-coccolithic clays 31.8, hemipelagic clays 29.1, nano-ooze 10.8, carbonate turbidites 9.5, diatom clays 4.8, coccolithic–foraminiferal (nano) clays 4.7, diatom–coccolithic ooze 2.0%, and nanoclays 1.4. Recalculation of volumes for dry sediment matter (Table 2) showed that the total dry mass of the Eopleistocene sediments in the studied area is 9.4 × 1018 g, including 4.2 × 1018 g terrigenous sediments, 2.1 × 1018 g carbonates, and 3.1 × 1018 g siliceous sediments.

The calculation of accumulation rates of the main sediment groups (Table 2) showed that during Pleistocene this parameter increased by 1.6 times for terrigenous sediments and by 3 times for carbonate sediments, but decreased by 2.2 times for siliceous sediments.

DISCUSSION

An increase of terrigenous influx from Africa during Pleistocene was likely related to increasing uplifting of the African rift zone, which is ascribed to the most neotectonically active mountainous belts (Trifonov and Sokolov, 2015). It should be reminded that the upper reaches of the northeastern tributaries of the Congo River are located in the southern part of this rift zone. Thereby, a definite role of glacioeustatic fluctuations of the World Ocean levels also cannot be ruled out.

It is noteworthy that the growth of carbonate accumulation in the Neopleistocene compared to the Eopleistocene is inconsistent with previously revealed tendency to decreasing carbonate accumulation rate in the pelagic part of the Atlantic Ocean (Levitan and Gelvi, 2016). Thus, the shallow-depth floor areas (submarine part of the Southwestern Africa continental margin) accumulated more carbonate sediments in the Neopleistocene compared to the Eopleistocene than at greater depths. As mentioned previously, the situation at large depths is explained by the increasing flow of deep and bottom water flows dissolving carbonates from the high latitudes in the Neopleistocene (Levitan and Gelvi, 2016). Since this influence practically does not reach the upper part of intermediate and surface waters, the trend of increasing productivity of carbonate-concentrating organisms, which was determined for the Pacific and Indian oceans (Levitan et al., 2013, 2014), was preserved during Pleistocene in this zone

The history of silica accumulation on the Southwestern African submarine margin is sharply different for two main areas: on the Congo transect and in the Benguela upwelling influence zone. Deep-sea drilling data convincingly indicate that the accumulation of diatom clays related to the nutrient influx with the Congo River was intensified in the Neopleistocene compared to the Eopleistocene (similar to the accumulation of hemipelagic clays) (Wefer et al., 1998). At the same time, the silica accumulation in the influence zone of the Benguela upwelling was much stronger and wider in area in the Eopleistocene than in the Neopleistocene. In general, the influence of the Benguela upwelling on the silica accumulation on the studied continental margin in the Pleistocene was much greater than that of dissolved nutrient influx of the Congo River.

Of great interest is the question what caused the intensification of silica accumulation in the Benguela upwelling zone in the Eopleistocene? Two alternative approaches are possible: (1) intensification of aeolian activity above the Namib Desert; (2) increasing influx of the Antarctic subsurface waters.

Only a few reliable materials are available for the first mechanism. Judging from literature data, the Pleistocene (especially, Late Pleistocene) was characterize by the alternation of wet and arid episodes with the predominance of the latter. For instance, the episodes of elevated humidity during last 420 ka were recorded within 420–385, 230–207, and 120–117 ka and coincided with interglacials (Geyh and Heine, 2014). Arid episodes were likely correlated with the intensification of southeastern trade winds and upwelling (Shi et al., 2001). They likely correspond to the episodes of increasing dust concentration in atmosphere during glaciation, which were revealed in the ice drilling cores (Brook et al., 2006). Based on these data, the intensification of upwelling should be expected in the Neopleistocene rather than in the Eopleistocene.

The second point of view is developed in detail in (Berger et al., 1998). Based on numerous paleoceanographic proxies, these authors showed that the powerful flow of the subsurface Antarctic waters, which were enriched in dissolved silica and entered the upwelling zone, beneath relatively warm surface waters with low contents of dissolved silica already has existed since Pliocene. Its maximum occurred 2 Ma, and then the flow gradually decreased, also during the period from the Eopleistocene to the Neopleistocene.

We support the second point of view. It is additionally confirmed by the almost complete coincidence of the variation pattern of biogenic silica mass accumulation rates in the Benguela upwelling with that of the Weddel Sea, showing the same maximum at 2 Ma (Cortese et al., 2005).

CONCLUSIONS

Our study presents the lithological-facies maps (with isopachs) on a scale of 1 : 10 000 000 compiled by us for the Neo- and Eopleistocene of the submarine continental margin of Southwestern Africa. Processing of these maps using A.B. Ronov’s volumetric method allowed us to quantify the sedimentation parameters: areas and volumes of mapped sediments of the corresponding stratigraphic units, mass of dry sediment, and its accumulation rates.

Based on analysis of obtained data, it was concluded that the terrigenous and carbonate sedimentation was intensified while siliceous sedimentation became weaker during Pleistocene. The last fact is likely related to the decreasing intensity of the Benguela upwelling.