21.1 Introduction

The Nile Cone is the seaward extension of the Nile Delta and, on present evidence, appears to have been accumulating for at least 30 million years (Macgregor, Reference Macgregor2012; Fielding et al., Reference Fielding, Najman and Millar2016, Reference Fielding, Najman and Butterworth2018). The older portions of the Nile Cone (Oligocene and Miocene in age, or 30–5.3 Ma) extend north across the floor of the Eastern Mediterranean for roughly 400 km and span an E–W distance of ca. 800 km (Fig. 21.1a). The younger portions of the Cone (Pliocene and Pleistocene in age, or 5.3 Ma to 11 ka) are almost as extensive, with an E–W extent of ca. 700 km and a northward extent of ca. 300 km (Fig. 21.1b). Apart from its actual and potential economic value as a source of oil and gas (Makled and Mandur, Reference Makled and Mandur2016), the Nile Cone contains a relatively complete record of major Nile floods. The purpose of this chapter is to review this record and to offer a synthesis of the Quaternary depositional record preserved within the younger and better-dated portion of the Cone.

Figure 21.1 The Nile Cone at different stages in its formation. (a) Oligocene and Miocene (30–5.3 ma). (b) Pliocene and Pleistocene (5.3 Ma to 11 ka). LB denotes Levantine Basin, R the Rosetta Depocentre, and D the Damietta Depocentre. (Simplified from Macgregor, Reference Macgregor2012, Fig. 5.)

21.2 Age and Volume of the Nile Cone

The Nile is one of the most intensively studied rivers in the world. Despite more than a century of detailed investigation throughout the basin, until very recently there was still very little agreement over when the Nile drainage network arose. Estimates varied from Eocene to Holocene. One group of workers argued that the Nile is no older than Pleistocene, when pyroxenes and other minerals diagnostic of an Ethiopian volcanic source first become evident (Shukri, Reference Shukri1949). Another group advocates a much earlier connection between the Ethiopian headwaters of the Nile and the desert Nile in Sudan, with evidence of Oligocene incision by the Blue Nile and Tekezze Rivers (McDougall et al., Reference McDougall, Morton and Williams1975). Fielding et al. (Reference Fielding, Najman and Millar2016, Reference Fielding, Najman and Butterworth2018) have now demonstrated quite conclusively that the modern Nile flow regime was already underway by 31 Ma, confirming an Oligocene connection between the Ethiopian headwaters and the Nile Cone. They analysed sediment samples retrieved from the Nile Cone with biostratigraphic ages of 31, 27.5, 17, 15.5 and 15.2 Ma, 3.25–2.65 Ma and 1.295 Ma. Their analysis of Sr-Nd bulk data and of detrital zircon U-Pb and Hf-isotope data showed a persistent signal in offshore sediments extending back to at least 31 Ma, consistent with a sustained and relatively steady input of sediments derived from the Ethiopian Continental Flood Basalts via the Blue Nile and Atbara tributaries of the desert Nile. Fielding et al. (Reference Fielding, Najman and Butterworth2018) also noted that Eocene sediments onshore in northern Egypt may yet reveal a still earlier Ethiopian connection, and await future study.

There has also been considerable debate about the volume of the Nile Cone and its sources of sediment. Macgregor (Reference Macgregor2012, Table 2) estimated that the volume of compacted sediment deposited onto the Nile Cone between 30 and 10 Ma amounted to 188,097 km³. This was followed by deposition of a further 393,265 km³ during the last 10 Ma. The total compacted volume for the period 30–0 Ma is therefore 581,362 km³, which is substantially higher than earlier estimates. For example, Harrison (Reference Harrison1955) used gravity anomaly data to estimate the areal extent and thickness of the Nile Cone, and came up with a figure of 95,000 km³ assuming no crustal sag and with a figure of 220,000 km³ allowing for the late Cenozoic crustal sag that has indeed taken place. Emery et al. (Reference Emery, Heezen and Allan1966) used Harrison’s submarine contour map and considered that the volume was closer to 140,000 km³ and most likely less. Wong and Zarudski (Reference Wong and Zarudski1969) considered this an underestimate and McDougall et al. (Reference McDougall, Morton and Williams1975) concluded that a volume between 100,000 and 200,000 km³ was plausible, on the grounds that this volume matched that of the gorges eroded in the headwaters of the Blue Nile and Tekezze–Atbara Rivers in Ethiopia. However, as Macgregor has shown, the Ethiopian headwaters of the Nile were not the only contributors of sediment to the Nile Cone. Talbot and Williams (2009, pp. 39–40) have also pointed out that incision by the ancestral Nile during the late Miocene created a canyon between the coast and as far upstream as Aswan, from which roughly 80,000 km³ of rock had been eroded before the Pliocene transgression converted the canyon into a vast estuary that was filled with a similar volume of sediment from the aggrading Nile. Very rapid rates of Pliocene sedimentation are also evident in two marine sediment cores (Baltim-1 and NDOB-1) drilled into the Nile Cone (Makled and Mandur, Reference Makled and Mandur2016, Figures 3 to 5).

Macgregor (Reference Macgregor2012) estimated likely rates of denudation in different sectors of the Nile Basin over the past 30 Ma using a combination of geomorphic evidence (incision into surfaces of known age) and apatite fission track analysis for certain portions of the Red Sea Hills, where he inferred 1.3 km of mean surface lowering in the past 30 Ma. He concluded that the Red Sea Hills region alone had contributed 217,249 km³ or 37% of the total amount of sediment in the Nile Cone (Macgregor, Reference Macgregor2012, Table 1). Earlier workers had neglected the contribution from the Red Sea Hills and had greatly underestimated the quantity of rock eroded from northern Sudan and the now arid Western Desert of Egypt, so that if we exclude the post-depositional carbonate present in the sediments, a Nile Cone volume slightly in excess of 500,000 km³ seems reasonable.

21.3 Analysis of Marine Sediment Cores from the Nile Cone

A variety of analyses (terrestrial and marine microfossils, geochemical and isotopic composition) of marine sediment cores obtained from various depths within the Nile Cone have been used to reconstruct past changes in sediment input and freshwater discharge from the Nile Basin (Rossignol-Strick et al., Reference Rossignol-Strick, Nesterhoff, Olive and Vergnaud-Grazzini1982; Rossignol-Strick, Reference Rossignol-Strick1985, Reference Rossignol-Strick1999; Wehausen and Brumsack, Reference Wehausen, Brumsack, Robertson, Emeis, Richter and Camerlengui1998; Ducassou et al., Reference Ducassou, Mulder and Migeon2008, Reference Ducassou, Migeon and Mulder2009; Rohling et al., Reference Rohling, Abu-Zied, Casford, Hayes, Hoogakker and Woodward2009; Revel et al., Reference Revel, Ducassou and Grousset2010, Reference Revel, Ducassou and Skonieczny2015; Zhao et al., Reference Zhao, Liu and Colin2011, Reference Zhao, Colin, Liu, Paterne, Siani and Xie2012; Blanchet et al., Reference Blanchet, Tjallingii and Frank2013, Reference Blanchet, Contoux and Leduc2015; Hennekam et al., Reference Hennekam, Jilbert, Schnetger and de Lange2014, Reference Hennekam, Donders, Zwiep and de Lange2015; Makled and Mandur, Reference Makled and Mandur2016). Some of these studies have focused on the presence or absence of sapropel layers within the Nile Cone and further afield, which we discuss in greater detail in Section 21.4. In this section, we focus more on geochemical and isotopic studies.

Zhao et al. (Reference Zhao, Liu and Colin2011) investigated the oxygen isotope variations in marine core MD90-964 on the distal Nile fan in the Levantine Basin (Fig. 21.1) of the eastern Mediterranean. The core extended back in time to 1.75 Ma. They identified 21 sapropel layers, which they defined as sediment layers with at least 1% organic carbon. These layers they found to be enriched in barium. They identified another 21 dark layers enriched in barium but from which the carbon had been oxidised and removed. These they termed ‘ghost sapropels’ and ‘hidden sapropels, depending on the strength of the geochemical evidence that they had once been sapropels. Fluctuations in the ratio of titanium to vanadium (Ti/V) at different depths in the core were used as a proxy for fluctuations in the amount of suspended sediment delivered to the Nile from the volcanic highlands of Ethiopia. These fluctuations showed a 23-ka periodicity, indicative of strong precessional influence (see Chapter 4).

They also found a 78-ka signal, the cause of which remains enigmatic, although it could represent two obliquity cycles. Each obliquity cycle lasts about 41 ka and reflects changes in the tilt of Earth’s axis from a maximum of 24°36′ to a minimum of 21°59′. Times of maximum inclination are times of hotter summers and colder winters in high latitudes; times of minimum tilt are associated with mild winters and mild summers.

The precessional cycle over the past million years has varied between 16.3 ka and 25.8 ka and over the past 10 Ma has varied between 23 ka and 19 ka (Chapter 4). This cycle reflects the changing season of the year during which Earth is closest to the sun and is controlled by the direction in which the spin axis of Earth points in space. At times when the sun is closest to Earth during the northern summer, the tropical monsoons will tend to be stronger and summer precipitation will also be both more intense and greater in amount. Zhao et al. (Reference Zhao, Liu and Colin2011) considered that during episodes of higher monsoonal precipitation in the Ethiopian Highlands, the vegetation cover would be denser and more widespread, leading to a reduction in erosion and hence in the sediment load of the Blue Nile and Atbara. They concluded, somewhat cryptically, that fluctuations in Nile sediment discharge ‘are more the result of river transport capability than of erosion potential in source areas’ (Zhao et al., Reference Zhao, Liu and Colin2011, p. 239).

In a subsequent study of the ratio of iron to aluminium (Fe/Al) within the same core (MD90-964), Zhao et al. (Reference Zhao, Colin, Liu, Paterne, Siani and Xie2012) used variations in this ratio as an index of fluctuations in the contribution of iron-bearing heavy minerals derived from the Ethiopian headwaters of the Nile. They also considered this index to be an indirect proxy for precipitation changes in that region. Their key finding was that times when the Fe/Al ratio was highest often coincided with times of sapropel formation, with a spectral peak indicating a 23 ka precessional signal. They reviewed earlier studies from the Nile Delta and proximal Nile Cone based on Sr and Nd isotopic ratios (Foucault and Stanley, Reference Foucault and Stanley1989) and the clay mineral composition of sapropels (Revel et al., Reference Revel, Ducassou and Grousset2010). These studies indicated reduced inputs of fine-grained Nile sediments during times of sapropel formation in the eastern Mediterranean but higher Nile freshwater discharge during times of stronger summer monsoon. They did not attempt to explain this apparent paradox. The factors determining Nile discharge and Nile sediment load are complex and are reviewed in Section 21.5. We now return to the question of Nile floods and sapropel formation.

21.4 Nile Floods and Sapropel Formation

Marine sediment cores from the floor of the Mediterranean have revealed the presence of multiple sapropel layers in Neogene (Miocene and Pliocene) and Quaternary sediments (Lourens et al., Reference Lourens, Antonarakou and Hilgen1996; Cramp and O’Sullivan, Reference Cramp and O’Sullivan1999; Larrasoaňa et al., Reference Larrasoaña, Roberts, Rohling, Winklhofer and Wehausen2003). The word sapropel comes from the Greek words saprós, meaning putrid or decayed, and pēlós, meaning clay or earth. The Glossary of Geology defines sapropel as ‘an unconsolidated, jellylike ooze or sludge composed of plant remains, most often algae, macerating and putrefying in an anaerobic environment on the shallow bottoms of lakes and seas. It may be a source material for petroleum and natural gas’ (Bates and Jackson, Reference Bates and Jackson1987, p. 588). The Encyclopaedic Dictionary of Physical Geography defines sapropel as ‘a mud or ooze composed predominantly of anaerobically decomposing organic material, usually in aquatic environments’ (Goudie et al., Reference Goudie, Atkinson and Gregory1985, p. 376). The anaerobic conditions that allow sapropels to form on the bed of the eastern Mediterranean are thought to reflect a change in seawater stratification caused by a sustained influx of freshwater from major rivers such as the Nile (Rossignol-Strick et al., Reference Rossignol-Strick, Nesterhoff, Olive and Vergnaud-Grazzini1982; Rossignol-Strick, Reference Rossignol-Strick1985,Reference Rossignol-Strick1999), as well as possible outflows from the Black Sea. Another possibility is increased primary productivity as a result of an influx of nutrients followed by preservation under anoxic conditions (Cramp and O’Sullivan, Reference Cramp and O’Sullivan1999). These two sets of processes are not mutually exclusive.

Lourens et al. (Reference Lourens, Antonarakou and Hilgen1996) provided a chronology of sapropel formation in the Mediterranean calibrated against the astronomical time scale. Their ages are midpoint ages for sapropel formation and there appears to be a 3,000-year time lag between sapropel formation and the correlative precessional minimum (see Chapter 4) shown in their insolation index. The sapropels are numbered consecutively from the youngest (S1) back in time. Cramp and O’Sullivan (Reference Cramp and O’Sullivan1999) noted the strong lack of any spatial and temporal continuity of the sapropels, which they considered was a result of redeposition and/or post-depositional geochemical alteration of the sapropels, the former process shown by the fact that some sapropel layers are more than 4 m thick, which would be unusually thick for an undisturbed sapropel layer. They further noted that although most sapropels coincided with warmer climatic phases on land (i.e., interglacial and interstadial phases), some also formed during times of colder climate, such as sapropel S8 and S6. The conspicuous sapropel S5 was synchronous with marine isotope stage 5e (125 ka), or the peak of the last interglacial.

In the case of the Nile, the floodwater hypothesis is, in principle, eminently testable by comparing well-dated phases of very high Nile discharge with dated phases of sapropel formation. While such hypothesis testing is certainly possible for about the last 125,000 years (Williams et al., Reference Williams, Duller and Williams2015a), it becomes more difficult with earlier Nile flood episodes simply because the ages obtained for these earlier events have much larger error terms, making it hard to establish whether the flood phases were indeed synchronous with the sapropels.

The most recent sapropel S1 in the eastern Mediterranean is a composite unit, with cited ages showing a high degree of variability (Thomson et al., Reference Thomson, Mercone, De Lange and Van Santvoort1999; Mercone et al., Reference Mercone, Thomson, Abu-Zied, Croudace and Rohling2001). For example, Cramp and O’Sullivan (Reference Cramp and O’Sullivan1999) suggest an age range of 12–6 ka. Other age estimates suggest 13.7–12.4 ka near the base and 9.9–8.9 ka near the top (Ducassou et al., Reference Ducassou, Migeon and Mulder2009), 9.5–6.6 ka (Revel et al., Reference Revel, Ducassou and Grousset2010) and 10.1–6.5 ka with a gap at 8.2–7.9 ka (Hennekam et al., Reference Hennekam, Jilbert, Schnetger and de Lange2014). In fact, formation of sapropel S1 may have ended as recently as 5 ka (Higgs et al., Reference Higgs, Thomson, Wilson and Croudace1994), which is also when the Nile deep-sea turbidite system became inactive as a result of reduced sediment discharge from that river associated with reduced rainfall in the Ethiopian headwaters of the Nile (Ducassou et al., Reference Ducassou, Mulder and Migeon2008, Reference Ducassou, Migeon and Mulder2009). There are several reasons for these age discrepancies, of which the most obvious is differential rates of post-depositional oxidation of the sapropel. Another is that the sediment cores come from different sectors of the Nile Cone, some of which might have experienced low rates of sedimentation whereas others might have experienced more rapid sedimentation at the same time, which would have been more conducive to sapropel formation. Since the Nile Delta distributary channels were active at different times, we would expect to find varying rates of sediment deposition across the proximal sector of the Nile Cone (Hennekam et al., Reference Hennekam, Donders, Zwiep and de Lange2015).

Three deep-sea sediment cores were retrieved from the western sector of the Nile Cone at depths of 1389 m (MS27PT), 2823 m (FKS05) and 2221 m (FKS04), at up to 200 km NW of the present coast (Ducassou et al., Reference Ducassou, Mulder and Migeon2008). The clastic mud beds rich in terrestrial organic matter have been reworked by turbidity currents and are characterised by an upward coarsening particle size composition followed by an upward fining sequence, indicative of a waxing flow succeeded by a waning flow regime. Ducassou and her co-authors concluded that the mud beds rich in organic matter indicated higher Nile discharge and wetter intervals in the Nile headwaters, with the most recent wet phase lasting from 12–10 ka to 6–4 ka (Ducassou et al., Reference Ducassou, Mulder and Migeon2008).

A comprehensive analysis of more than forty deep-sea sediment cores collected across the Nile Cone and spanning the last 200 ka enabled Ducassou et al. (Reference Ducassou, Migeon and Mulder2009) to derive a detailed chronology for changes in sediment input and for times of sapropel formation associated with episodes of very high Nile discharge (Ducassou et al., Reference Ducassou, Migeon and Mulder2009, Figs. 4 and 5). Table 21.1 summarises this work. Turbidity flows in deep-sea fans on the Nile Cone were most active during times of rising and high sea level associated with wetter climates, and least active during times of low sea level or high sea level stands coupled to arid periods, as in the last 5,000 years. Capozzi and Negri (Reference Capozzi and Negri2009) independently confirmed the important role played by sea level fluctuations upon deep water circulation in the Mediterranean and indirectly on the timing of sapropel formation.

Revel et al. (Reference Revel, Ducassou and Grousset2010) carried out a detailed study of fluctuations in the strontium and neodymium isotopic composition of the terrigenous component of sediment in marine core MS27PT located 100 km WNW of the Rosetta distributary outlet on the Nile Delta (see Chapter 20, Fig. 20.1). They also analysed changes in major elements, with a 10-year resolution for iron fluctuations during wetter periods across North Africa in the past 100,000 years (98–69 ka, 60–50 ka, 38–30 ka and 14–5 ka). They found high amounts of desert dust influx during the arid phases coinciding with MIS 4 and MIS 2 (Last Glacial Maximum [LGM]), which were also times of eolian dust deposition in East Antarctica. The 98–69 ka wet phase coincides with the formation of sapropels S4 and S3, the 60–50 ka wet phase with sapropel S2, and the 14–5 ka wet phase with sapropel S1. Their final conclusion was that the Ethiopian headwaters of the Blue Nile and Tekezze–Atbara Rivers were significantly wetter during 8–5 ka, with a longer rainy season and/or highly variable rainfall intensity. However, a 12-m sediment core obtained from the Dendi eastern crater lake near the Blue Nile headwaters in the central Ethiopian uplands indicates peak wetness (and, presumably, high Blue Nile flow) between 10.0 and 8.7 ka and peak aridity between 4.0 and 2.6 ka (Wagner et al., Reference Wagner, Wennrich and Viehberg2018).

The response of the Nile to millennial-scale climatic change since the LGM was the focus of a study by Box et al. (Reference Box, Krom and Cliff2011) in which they analysed the ⁸⁷Sr/⁸⁶Sr ratios and major element geochemistry from two sediment cores. Core 9501 was retrieved on the distal edge of the Nile Cone, south of Cyprus; core 9509 came from the eastern margin of the Nile Cone off the coast of southern Israel. The authors noted that the sediment in these cores represented suspended sediment discharged from the Nile into the Levantine Basin, and so provided a record of erosion and sediment transport in the Nile headwaters. They found that during the dry interval of the last 5,000 years as well as before 11 ka, in core 9509, the influx of sediment deduced to be from the Blue Nile/Atbara (10–12 g/cm² per yr and ca. 6 g/cm² per yr, respectively) was greater than during the 11–5 ka period of peak humidity, when it averaged only 2 g/cm² per yr. The sediment flux from the White Nile showed a very different response, increasing from 5 g/cm² per yr at ca. 13 ka to >15 g/cm² per yr by 5 ka. The influx of Saharan dust decreased during this interval, but was high both before and after the 13–5 ka humid episode. The authors postulated that during wetter intervals in the seasonally wet Ethiopian Highlands, expansion of the plant cover caused a reduction in erosion and sediment yield in the Blue Nile/Atbara headwaters, but in the perennially wet Ugandan headwaters of the White Nile, an increase in precipitation would generate an increase in erosion and in the transport of sediment through the Sudd marshes.

This interpretation raises several questions. First, until about 5 ka, there was overflow from Lake Turkana via the Pibor into the Sobat and so into the White Nile (see Chapter 9). White Nile sediment would therefore show an enhanced volcanic signature during wetter climatic intervals. How is this sediment input into the lower White Nile to be distinguished from the contributions emanating from the Blue Nile and Atbara? Second, wadis were actively contributing sediment to the Desert Nile during wetter climatic periods (Woodward et al., 2015 and Chapter 14). Some of this sediment came from the Red Sea Hills, where the bedrock geology is not too dissimilar to that in the White Nile headwaters. Some came from erosion of the Mesozoic Nubian Sandstone Formation, and some represented recycled desert dust. It would be easy to mistake the sediment input from these wadi sediments for a higher White Nile sediment input. Third, and equally important, it is by no means certain that an increase in precipitation in the equatorial headwaters of the White Nile would necessarily lead to an increase in sediment yield. The opposite is more likely. Peak sediment yields in present-day rivers occur in the monsoonal tropics and in semiarid regions, but not in rivers flowing through densely vegetated catchments in equatorial climatic regions (Douglas, Reference Douglas1967, Reference Douglas1969; Milliman and Meade, Reference Milliman and Meade1983; Milliman, Reference Milliman and Ruddiman1997; Williams, Reference Williams and Vita-Finzi2012a).

These caveats are not intended as a criticism of the quality of the analytical work conducted on the marine sediment cores. Rather, they illustrate the complexity of the Nile’s response to climatic change and the difficulties involved in construing subtle environmental changes in the Nile headwaters from the geochemical, isotopic and microfossil evidence preserved in marine sediment cores retrieved from the Nile Cone.

A further example of complex response to climatic change in the Nile Basin is shown by the work of Cécile Blanchet and her co-workers (Blanchet et al., Reference Blanchet, Tjallingii and Frank2013) on Holocene marine sediment core P362/2–33 collected from the SW margin of the Nile Cone offshore from the Rosetta distributary outlet in the Nile Delta. Using a combination of grain size analysis, bulk element composition and variations in the strontium and neodymium isotope values, they inferred high Blue Nile sediment input during 9.5–7.3 ka, which they attributed to high spring insolation, and low Blue Nile sediment influx during 7.3–4 ka, which they related to high autumn insolation in the Ethiopian Highlands. Two arid phases were identified between ca. 8.5 and 7.3 ka and between 4.5 and 3.7 ka. The White Nile was thought to have had a high sediment input to the Nile during 8–4 ka, when the inferred Blue Nile sediment influx was relatively low. Once again, disentangling the White Nile signal is far from simple, so that inferences about the past role of the White Nile should be viewed with considerable caution.

Hennekam et al. (Reference Hennekam, Jilbert, Schnetger and de Lange2014) analysed the stable oxygen isotopic composition of the planktonic foraminifer Globigerinoides ruber (∂¹⁸O_ruber) and the bulk sediment inorganic geochemistry in a Holocene marine sediment core (PS009PC) from the southeast Levantine basin of the Nile Cone (Fig. 21.1). The high rates of sediment deposition at this site enabled them to compile a high-resolution record of changes in Nile discharge, as well as to identify times when the main source of precipitation in the Nile headwaters was via the Indian Ocean or the Atlantic. They found that early Holocene Nile discharge came primarily from the Indian Ocean, with a peak at 9.5 ka. Oscillations in the summer monsoon at this time appeared to reflect fluctuations in solar activity. They identified five periods of enhanced Nile flow, with a recurrence interval of 500 to 1,000 years, associated with evidence of increased anoxia in the marine sediment core. The calibrated ages they obtained for these periods of very high Nile flow were ca. 9.7 ka, ca. 9.1 ka, ca. 8.6 ka, 7.7 ka and 6.6 ka. These ages are in remarkably good agreement with independent evidence of high White Nile flow at 9.7–9.0 ka, 7.9–7.6 ka and 6.3 ka, and for high Blue Nile flow at 8.6 ka, 7.7 ka and 6.3 ka (Williams, Reference Williams2009a, and Chapters 8 and 11). In this core, sapropel S1 is a composite unit, with ages of 10.1–8.2 ka for the lower portion S1a, and 7.9–6.5 ka for the upper portion S1b (Hennekam et al., Reference Hennekam, Jilbert, Schnetger and de Lange2014). Whether the 8.2–7.9 ka hiatus relates to the 8.2-ka cold event in the North Atlantic and Greenland remains uncertain but it is coeval with a phase of much reduced Nile flow.

To summarise thus far: during phases of very high Nile flow, clastic muds rich in continental organic matter and highly organic sapropels accumulated on the floor of the eastern Mediterranean (Fig. 21.2) (Krom et al., Reference Krom, Stanley, Cliff and Woodward2002; Ducassou et al., Reference Ducassou, Mulder and Migeon2008, Reference Ducassou, Migeon and Mulder2009; Rohling et al., Reference Rohling, Abu-Zied, Casford, Hayes, Hoogakker and Woodward2009; Revel et al., Reference Revel, Ducassou and Grousset2010, Reference Revel, Ducassou and Skonieczny2015; Zhao et al., Reference Zhao, Liu and Colin2011, Reference Zhao, Colin, Liu, Paterne, Siani and Xie2012; Blanchet et al., Reference Blanchet, Tjallingii and Frank2013, Reference Blanchet, Contoux and Leduc2015; Hennekam et al., Reference Hennekam, Jilbert, Schnetger and de Lange2014, Reference Hennekam, Donders, Zwiep and de Lange2015; Makled and Mandur, Reference Makled and Mandur2016). The sapropel record reveals that episodes of middle to late Pleistocene high flow in the Blue and White Nile coincide very broadly with sapropel units S8 (217 ka), S7 (195 ka), S6 (172 ka), S5 (124 ka), S3 (81 ka), S2 (50 ka) and S1 (Williams et al., Reference Williams, Adamson, Prescott and Williams2003, Reference Williams, Williams and Duller2010, Reference Williams, Duller and Williams2015a; McDougall et al., Reference McDougall, Brown and Fleagle2008). Sapropel 5 (124 ka) was synchronous with major flooding and the formation of the 386-m lake in the lower White Nile Valley (Barrows et al., Reference Barrows, Williams and Mills2014) and with a prolonged wet phase at ca. 125 ka at Kharga Oasis in the Western Desert of Egypt (Kieniewicz and Smith, Reference Kieniewicz and Smith2007).

Figure 21.2 Phases of sapropel formation in the Nile Cone in relation to episodes of high Nile flow during the last 250 ka. (After Williams et al., Reference Williams, Duller and Williams2015a, Fig. 8.)

Although the sapropel record in the eastern Mediterranean is incomplete, with some evidence of complete removal of sapropels by post-depositional oxidation (Higgs et al., Reference Higgs, Thomson, Wilson and Croudace1994), it is nonetheless a longer and more complete record than that presently available on land, and so can serve as a useful surrogate record for Nile floods and phases of enhanced summer monsoon precipitation.

21.5 Conclusion

The Nile Cone extends north from the present coastline for 300–400 km and spans an E–W distance of 700–800 km (Fig. 21.1). Its total sediment volume amounts to about 580,000 km³, of which 180,000 km³ were deposited between ca. 30 Ma and 10 Ma (Oligocene and Miocene), and the remaining 400,000 km³ during the last 10 million years (Pliocene and Quaternary). About 37% of the total sediment in the Nile Cone (220,000 km³) came from erosion of the Red Sea Hills and roughly 100,000 ± 50,000 km³ of rock came from erosion in the headwaters of the Blue Nile and Tekezze/Atbara Rivers.

Deep-sea sediment cores contain a useful, if sometimes enigmatic, record of past changes in Nile sediment influx and an indirect record of possible changes in precipitation, erosion and sediment yield in the Ethiopian and Ugandan headwaters of the Nile. Clastic muds rich in continental organic matter and highly organic sapropels accumulated on the floor of the eastern Mediterranean during phases of very high Nile flow. Episodes of middle to late Pleistocene high flow in the Blue and White Nile coincide very broadly with sapropel units S8 (217 ka), S7 (195 ka), S6 (172 ka), S5 (124 ka), S3 (81 ka), S2 (50 ka) and S1. Sapropel S1 is a composite unit, with ages of 10.1–8.2 ka for the lower portion and 7.9–6.5 ka for the upper portion in cores studied by Hennekam et al. (Reference Hennekam, Jilbert, Schnetger and de Lange2014). The 8.2–7.9 ka hiatus coincides with a phase of much reduced Nile flow.