Neogene-Quaternary Palaeoceanography from the Geochemistry

3.3 Subproject

Sub-Project 3.3

Neogene-Quaternary palaeoceanography from the geochemistry

of successions on the South African margin

Participants* Coordinator

Institution / Names / Email addresses
University of Cape Town
(UCT) / John Compton* /
Alfred Wegener Institute
(AWI) / Gabriele Uenzelmann-Neben /

Requested Funding:

Total for the 5-year duration project beginning in 2004:Euros 366000

2004

2005

2006

2007

2008

UCT

127000

58000

80000

61000

40000

Summary

Multiple geochemical proxies will be used to infer the evolution of the continental margin of South Africa from Neogene and Quaternary successions. Strontium isotope stratigraphy will be combined with biostratigraphy to develop an improved age model of deposition and diagenesis on the margin. Nd isotopic changes will be related to changes in oceanic circulation and upwelled waters along the margin. Seismic stratigraphy will be integrated with geochemistry, litho- and biostratigraphy in order to determine the relation among climate, tectonic uplift, upwelling intensity, ocean currents (erosional events) and sea-level fluctuations. The improved age model will allow us to correlate events on the western margin with global tectonism (e.g., opening of oceanic gateways), climate (glaciation of Antarctica and the Northern Hemisphere), eustacy and ocean circulation (e.g., intensity of North Atlantic Deep Water formation). The age of Quaternary successions will be determined using oxygen isotope stratigraphy and AMS radiocarbon ages of the Holocene mud belt. High-resolution Quaternary records will be compared to continental climate proxy records and Northern Hemisphere (including ice core) records.

Scientific Motivation and State of the art

The southern tip of Africa occupies a unique and critical position relative to the global circulation of the oceans as it interfaces with a major confluence of ocean currents. The warm, Agulhas Current flows along the eastern continental margin leaking flow (“rings”) into the South Atlantic as it retroflects back into the southern Indian Ocean. The cold, Benguela Current flows along the western continental margin as part of the South Atlantic gyre. These surface, as well as associated intermediate and deep water, currents form an important part of the overall thermohaline circulation of the ocean in which cold salty waters that sink and spread throughout the deep ocean are returned to the surface in a large scale, idealised ‘conveyor belt’ loop (Broecker and Denton, 1989). Of particular interest is the amount of warm water that leaks into the South Atlantic via the Agulhas retroflection because it is believed to control much of the heat transfer important to global climate. The present-day thermohaline circulation is believed to have been established around 25 million years ago (25 Ma) when Antarctica became fully isolated by the opening of Drake Passage (Shackleton and Kennett, 1975). The tectonic isolation of Antarctica allowed establishment of the Circum Antarctic Current, build up of ice on the continent of Antarctica and a significant shift towards colder climates and the current Ice Age (Kennett, 1982).

Palaeoceanographers are keen to determine past changes in oceanic circulation because of the strong linkage to climate change, both on long, million-year and short, thousand-year time scales. The ratio of the neodymium (Nd) isotopes 143 to 144 (expressed in shorthand relative to chondritic meteorites as εNd) has proven to be a useful geochemical tracer of changes in ocean circulation because of the short residence time of Nd in seawater. As a consequence of the short residence time of Nd, the εNd values of seawater are distinct in each of the major ocean basins and εNd values can be used to trace the mixing of these different ocean water masses (e.g., Piepgras and Wasserburg, 1982; Jeandel, 1993). Changes in εNd of marine precipitates have been used to document changes in Pacific - Atlantic exchange through the Panama gateway (Frank et al., 1999) and through the Drake Passage (Piepgras and Wasserburg, 1982; Jeandel, 1993) as well as to infer long-term (Stille, 1992; Stille et al., 1996) and short-term (Rutberg et al., 2000) changes in ocean circulation. Most previous work has focused on ferromanganese deposits that form at very slow rates of several mm per Myr in the deep ocean. However, there have been several studies that have shown the usefulness of phosphorite analyses (Shaw and Wasserburg, 1985; Stille, 1992). The advantages of phosphorite samples are that they precipitate at much faster rates than Fe-Mn oxides, they are generally widespread on continental margins and they take up abundant Sr from which the age of the sample can be derived.

The age of marine minerals can be determined by their strontium (Sr) isotope composition. In contrast to Nd, Sr has a long residence time in the oceans and is uniformly well mixed. In addition, the ocean Sr isotope composition has steadily increased at variable rates from the Oligocene to the present-day and allows mineral precipitates to be dated (DePaolo and Ingram, 1985). One of the principal marine minerals on the South African margin is phosphorite, a phosphorus-rich rock type that forms in highly productive upwelling areas (Birch, 1990; Baturin, 2000). The utility of Sr isotope analysis is that it provides a means to date phosphorite formation directly (Compton et al., 1993) and, depending on the steepness of the marine Sr isotope record, can provide ages with uncertainties of less than 1 Myr (Hodell et al., 1991; Farrell et al., 1995). The age of sediment deposition will be determined by integrating Sr isotope analysis of biogenic grains and biostratigraphy (foraminifera and pollen). In addition to biostratigraphy, the pollen record will provide important insights into the evolution of the Cape Floral Kingdom. Improved age resolution is critical to our understanding of the complex depositional and diagenetic history of these shelf successions and their relation to sea-level fluctuations and climate change. Recent results from the Orange River Prodelta (Compton et al., 2002; in press) indicate that upwelling was initiated in the earliest Miocene, significantly older than the late Miocene age indicated at the Walvis Ridge (Siesser, 1980). Episodes of phosphogenesis on the margin can be linked to sea-level fluctuations and periods of increased organic matter accumulation related to greater productivity (upwelling) and preservation of organic matter (Mallinson and Compton, 1997). Climate, upwelling and formation of phosphorite are connected and, therefore, it is important to understand the age of phosphorite formation in order to infer past changes in climate.

Our understanding of present-day ocean circulation is far from complete. For example, it is debated to what extent the return flow to the South Atlantic is from leakage of warm Agulhas Current vs. cold waters passing through Drake Passage (Gordon, 1986; Rintoul, 1991; Heywood et al., 2002). Intermediate water masses remain relatively poorly understood in comparison to surface and deep water masses. What does appear clear is that significant ocean mixing occurs in the southeast Atlantic at the confluence of the Benguela Current with the Agulhas retroflection (J Lutjeharms, pers comm). Previous studies have looked at the deep and intermediate water εNd signatures around southern Africa (Albarede and Goldstein, 1992; Jeandel, 1993; Rutberg et al., 2000). These studies show the importance of Atlantic deep and intermediate-depth water flow into the Indian Ocean near the Agulhas Plateau. To date, there has been no attempt to look at the εNd values of intermediate and surface waters from outer shelf and upper slope deposits of the western and southern margins.

Scientific goals

Neogene

The geological deposits on the continental margin of South Africa contain a number of marine minerals whose εNd and Sr isotope values have the potential to document changes in ocean circulation and source of upwelled waters at the critical and complex juncture of the Agulhas and Benguela currents. The focus of this proposal is to determine the age of phosphogenic episodes and their corresponding variations in εNd from the paired analysis of Sr and Nd isotope compositions of specific mineral grain types from both the western margin (Benguela) deposits and southern margin (Agulhas) deposits. It is believed that these analyses will provide new insights into the sources and dynamics of South African margin water masses and their relation to major tectonic and climatic events over the last 25 Ma. These geochemical indicators of phosphogenesis and ocean circulation will be linked to the litho-, bio- and seismic stratigraphy of the margin. The architecture of sediment deposits and major erosional unconformities revealed by the seismic profiles proposed by G Uenzelmann-Neben will allow us to develop an integrated model of margin evolution.

Critical questions are:

1. What is the age of phosphogenic episodes on the western and southern South African margin and how do they relate to major palaeoceanographic events, such as upwelling, changes in ocean currents, the opening of Drake Passage, sea-level fluctuations, global cooling, aridity and the evolution of the Cape Floral Kingdom?

Preliminary results (Compton et al., 2002; in press) indicate that much of the phosphorite formed during periods of rising or highstands of sea level, but in some areas phosphorites formed during overall marine regressions. Major erosional unconformities on the margin suggest that tectonic uplift of the margin has also played a significant role along with eustatic sea-level fluctuations. A major phosphogenic episode represented by a 23.5-25.4 Ma phosphate-rich gravel lag deposit on the western margin corresponds to the Oligocene to Miocene marine transgression and highstand and to the final, deep-water opening of Drake Passage, a major oceanic gateway that initiated modern, thermohaline oceanic circulation. The apparent paucity of phosphorites on the margin older than this initial major phosphogenic episode suggests a major change in palaeoceanography on the margin across the Oligocene/Miocene boundary. Factors other than rising sea level may have included a significant increase in upwelling related to establishment of thermohaline circulation and the Benguela Current. These phosphorites may mark the transition from the greenhouse to the icehouse Earth where increased oceanic circulation resulted in a removal of deep-water phosphorus.

2. How does the Nd isotope composition of the phosphorite and other marine minerals vary over the last 25 Ma and what can we infer from these data on changes in the exchange of Pacific, Indian and Atlantic water masses?

Phosphorite εNd values are expected to reflect the εNd values of upwelled waters with much of the Nd scavenged from the water column and remobilised during early diagenesis to be taken up by phosphate minerals. On the western margin upwelling waters are primarily coming from water depths of less than 300 m (V Shannon, pers com.) and are associated with nutrient-rich Central South Atlantic Water. Upwelled waters on the Agulhas Bank are most likely Antarctic Intermediate Water which is nutrient-rich and believed to flow equatorward and counter to the overlying Agulhas Current (A Meyer, pers. com.). The results of this study should allow us to determine how similar Central South Atlantic Water was to Antarctic Intermediate Water and how these water masses changed over the last 25 Myr. The Agulhas Current is poor in nutrients and is not expected to be associated with increased productivity and phosphorite formation; however, ferromanganese deposits on the Agulhas Bank/Plateau may provide εNd signatures of the Agulhas Current that could be compared to the εNd of upwelled phosphorite samples. In addition, we plan to look at Mn nodules from the western continental rise to document changes in the εNd signatures of North Atlantic Deep Water. The age of the Mn nodules probably only extends back to the latest Miocene, but will provide data on oceanic circulation changes over the last 5 million years, a critical time period that includes onset of Northern Hemisphere glaciation.

Quaternary

Quaternary successions will be studied in order to understand short-term (<1 Myr) changes on the margin and will provide a useful comparison to the long-term (> 1 Myr) timescales of the Neogene successions. The age model of the Quaternary will be based on oxygen isotope stratigraphy and AMS radiocarbon ages of Holocene sediments. The Quaternary is characterised by high-frequency and high-amplitude variations in climate and sea level that make it significantly different from the Neogene.

Key scientific objectives for the Quaternary include:

Are phosphorites forming today on the western shelf mud belt deposits? Modern phosphorites have been documented to occur in diatomaceous ooze off Namibia (Veeh et al. 1973; Baturin, 2000) as well as perhaps in the Holocene mud belt of the Orange River Prodelta (Compton et al., 2002).

How are terrigenous sediments delivered by rivers and wind distributed on the western margin and how does this sediment distribution relate to changes in ocean currents and sea level?

The high sedimentation rates that result from sediment focussing allow for high organic carbon preservation and promote pyrite and phosphorite formation in the Holocene mud belt. Variations in the terrigenous flux from the continent should reflect changes in terrestrial erosion rates that are linked to tectonism (relief) and climate (rainfall). Marine records of continental erosion rates can be compared to the climate proxy records of the last 200,000 yr at the Tswaing impact crater site.

How do high-resolution (100-500 yr) climate proxy records over glacial/interglacial cycles and the Holocene relate to terrestrial climate change records, Northern Hemisphere climate records and ice core records?

Moderate to high sedimentation rates in the Quaternary successions will allow us to look at a number of different proxies of climate, tectonic and ocean circulation changes. Similar records are well known from the polar ice caps and from marine and terrestrial Northern Hemisphere sites, but there are relatively few from the Southern Hemisphere, particularly on the southern African margin whose ocean circulation may play a critical role in controlling climates of the North Atlantic. These records will provide increased coverage of climate change records and allow the timing of events to be compared and to evaluate lead or lag times between hemispheres. The Holocene ice records indicate that, although far more subtle than during the glacial periods, there were significant variations in Holocene climate. Holocene climate variations are of particular interest in their application to our understanding of possible future climate change.

Work Schedule

TASK / YEAR 1 / YEARS 2-3 / YEAR 4 / YEAR 5
Co-ordinate
Projects /
Collect and analyse samples / / /
Write-up theses; papers / 2 MSc
1 PhD
1 post-doc / 1 MSc
/ 1MSc
1PhD
1post-doc

Sample material

The UCT core repository houses a large collection of offshore material recovered over the years by various ships dredging the Agulhas Bank and there are a large number of vibracores from the western margin donated by diamond exploration companies. In addition, there have been a number of recent scientific expeditions along the southern African margin that have recovered a large number of piston, gravity and surface cores. Some of this material is housed in the UCT core repository, but the majority of it is available from core repositories in Europe (Germany and France). Future research expeditions are planned for over the next 5 years (IMAGES) that will augment these collections, including cores from the Agulhas Plateau. The location of existing samples is well documented and their coverage is sufficient to address the objectives of this proposal.

Analytical methods

The mineralogy of the samples will be determined by petrography (including scanning electron microscopy, SEM) and x-ray diffraction (XRD). Grains will be selected for Sr and Nd analyses on the basis of texture. Previous results have shown how critical it is to analyse different grain types from the same sample to reinforce interpretations. The different grain types will include phosphorite (CFA-cemented) peloidal grains with no internal structure, CFA replaced limestone, fossil fish bones (skeletal phosphorite grains), fish otoliths, benthic foraminifera, echinoid spines, mollusc shell and any ferromanganese deposits. Samples will be dissolved at room temperature in twice-distilled 5M glacial acetic acid for Sr and Nd analyses to minimise dissolution of other minerals. The amount of Sr and Nd contributed by minerals other than CFA in the phosphorite grain types is considered insignificant because these samples are predominantly composed of CFA that generally has a high Sr and Nd content (Shaw and Wasserburg, 1985). Major and trace element composition will be determined using ICP-MS and XRF. Stable isotopes (oxygen and carbon) will be measured on the mass spectrometer in Archaeology at UCT. AMS radiocarbon analyses will be measured in an overseas lab.

Funds Requested

Personnel

This project is anticipated to involve at least two post-doc, two PhD and four MSc post graduate students. An inflation/cost of living increase of approximately 10% is included. The proposed study will result in the training of new postgraduate students at the MSc and PhD level in the area of marine geochemistry and palaeoceanography. These fields are generally under represented in South Africa and there is a need for well qualified specialists.

Year 1 / 1 PhD @ R60,000 / 1 MSc @ R40,000 / 1 postdoc @ R100,000 / = R200,000
Year 2 / 1 PhD @ R65,000 / 2 MSc @ R45,000 / 1 postdoc @ R110,000 / = R265,000
Year 3 / 2 postdoc2 @ R120,000 / 2 PhD @ R70,000) / 2 MSc @ R50,000 / = R480,000
Year 4 / 1 postdoc @ R135,000 / 1 PhD @ R80,000 / 2 MSc @ R55,000) / = R325,000
Year 5 / 1 postdoc @ R150,000 / 1 PhD @ R90,000); / 1 MSc @ R60,000 / = R300,000
5 year total = R1,570,000 (2 postdocs, 2 PhD, 4 MSc) (Euros 187,000)

Analytical costs (years 1-4)