Elsevier

Earth and Planetary Science Letters

Volume 375, 1 August 2013, Pages 408-417
Earth and Planetary Science Letters

Miocene to Pliocene changes in South African hydrology and vegetation in relation to the expansion of C4 plants

https://doi.org/10.1016/j.epsl.2013.06.005Get rights and content

Highlights

  • Western southern African climate became continuously drier since 8 Ma.

  • Maximum C4 savannah expansion in western South Africa occurred between 5 and 4.5 Ma.

  • Main water vapour source of rainfall shifted to the Indian Ocean between 7 and 4.5 Ma.

  • Miocene C4 plant expansion resulted from increased equator-pole temperature gradient.

Abstract

The Miocene expansion of C4 plants (mainly tropical grasses) between 8 and 4 million years (Ma) remains an enigma since regional differences in the timing of the expansion rules out decreased CO2 (pCO2) as a dominant forcing [e.g. Tipple and Pagani, 2007. The early origins of terrestrial C4 photosynthesis. Annu. Rev. Earth Planet. Sci. 35, 435–461]. Other environmental factors, such as low-latitude aridity and seasonality have been proposed to explain the low tree versus grass ratio found in savannahs and tropical grasslands of the world, but conclusive evidence is missing. Here we use pollen and stable carbon (δ13C) and hydrogen (δD) isotope ratios of terrestrial plant wax from a South Atlantic sediment core (ODP Site 1085) to reconstruct Miocene to Pliocene changes of vegetation and rainfall regime of western southern Africa. Our results reveal changes in the relative amount of precipitation and indicate a shift of the main moisture source from the Atlantic to the Indian Ocean during the onset of a major aridification 8 Ma ago. We emphasize the importance of declining precipitation during the expansion of C4 and CAM (mainly succulent) vegetation in South Africa. We suggest that the C4 plant expansion resulted from an increased equator-pole temperature gradient caused by the initiation of strong Atlantic Meridional Overturning Circulation following the shoaling of the Central American Seaway during the Late Miocene.

Introduction

Regarding their carbon fixation pathways, plants can be categorised into 3 types: C3 plants, plants using the Crassulacean Acid Metabolism (CAM), and C4 plants. C3 plants are trees, bushes, shrubs, cold-season grasses and sedges. Most CAM plants are succulents and are relatively abundant in dry regions of SW Africa (Winter and Smith, 1996, Mucina and Rutherford, 2006). C4 plants are mainly warm-season grasses and sedges and are found predominantly in tropical and subtropical grasslands and semi-deserts (Collatz et al., 1998). Globally, C4 plants account for about 20% of terrestrial carbon fixation (Lloyd and Farquhar, 1994), making up 2/3 of all grasses in tropical and subtropical regions and more than 90% of tropical savannah ecosystems (Sage, 2001).

C4 grasses might have evolved as early as the Early Eocene (Strömberg, 2011), but Vincinetti et al. (2008) infer that changes of photosynthetic pathway in grass clades occurred at least 15 times after the initial appearance of C4 grass in the Oligocene. Finally, C4 grass dominated ecosystems expanded during the Miocene to Pliocene (Strömberg, 2011), when the world's climate changed from warm greenhouse to colder ice-house regimes with the establishment and expansion of permanent Antarctic ice sheets (Zachos et al., 2001) and the steepening of the climatic gradient from the equator to the poles (Gussone et al., 2004). The large time lag between the evolution of C4 grasses and the expansion of C4 dominated savannahs suggests different triggers for evolution on the one hand and ecological expansion on the other (Osborne, 2008, C Grasses consortium, 2010, Strömberg, 2011).

The expansion of C4 grasslands neither occurred synchronously, nor continuously (Edwards et al., 2010). On most continents, C4 plants expanded in grasslands and savannahs of tropical and subtropical climate regimes previously dominated by C3 grass species (Tipple and Pagani, 2007, C Grasses consortium, 2010, Strömberg, 2011). For North-East Africa, Feakins et al. (2013) inferred two distinct steps in the establishment of C4 vegetation. The first step is an increase in C4 biomass during the late Middle Miocene and the second one a re-expansion of C4 vegetation from C3 grasslands and woodlands in the Late Miocene to a mix of C4 grassland, C3/C4 shrubland, and C3 woodland in the Pliocene. Stable isotope composition from paleosols from the Awash Valley in Ethiopia showed that open C4 rich vegetation of the latest Miocene returned towards more woody C3 vegetation during the Early Pliocene, and again changed to open C4 vegetation during the Pleistocene (Cerling et al., 2011).

C4 plants have a higher water-use efficiency and are increasingly more competitive than C3 plants in tropical dry forests, savannahs, grasslands and deserts, respectively (Bourlière, 1983, Beerling and Osborne, 2006, Tipple and Pagani, 2007). In tropical savannahs and grasslands, the length and intensity of the rainy season (typical major growing season) are important factors governing the competition of woody plants (trees and shrubs which are C3) versus tropical grasses (mainly C4) (Beerling and Osborne, 2006, Still and Powell, 2010). Not only the total amount of precipitation is important but also its seasonality as intensified fire regime is a corollary of strong seasonality (Archibald et al., 2009). Frequent fires increase woody plant mortality preventing savannahs to be converted into woodlands and causing a shift of the woodland/grassland ecotone (Keeley and Rundel, 2005). Therefore, changes in hydrology may have been an important factor for the Miocene to Pliocene C4 plant expansion (Tipple and Pagani, 2010).

The use of the C3 and C4 carbon fixation pathways results in distinct stable carbon isotopic signatures of the respective plants’ tissues and leaf waxes, where the C3 plants are typically most depleted in 13C with values ranging between −30‰ and −42‰ for long-chain n-alkanes (typical components of plant waxes), and C4 plants are less 13C depleted with values ranging between −18‰ and −26‰ (Rommerskirchen et al., 2006, Vogts et al., 2009). By investigating stable carbon isotope compositions of plant-wax derived n-alkanes in marine sediments, it has been shown that inferences can be made on changes in mean annual precipitation and length of the rainy season in the source area of the leaf wax based on the climate preference of C3 and C4 plants (Collins et al., 2011, Vogts et al., 2012). Plants using the CAM carbon fixation pathway are also adapted to arid conditions and have intermediate bulk and lipid stable carbon isotopic compositions (Winter and Smith, 1996, Feakins and Sessions, 2010).

Pollen in marine sediments are well suited to qualitatively record vegetation changes on the adjacent continent (e.g. Hooghiemstra et al., 2006). Asteraceae and tree pollen can be used to infer development of woodlands and semi-desert, respectively. The relative amounts of grass pollen along the African coast are indicative of the latitudinal distribution of savannah and grasslands (e.g. Dupont, 2011). However, pollen of C4 and C3 grasses cannot be distinguished. On the other hand, isotope measurements cannot distinguish between grassy and non-grassy C4 plants. By combining pollen and isotope analysis we can disentangle the C4 grass development in western South Africa.

In contrast to stable carbon isotopes, hydrogen isotope compositions of plant lipids are directly related to hydrological changes. Principally, plant waxes record isotopic changes in rainfall isotopic composition (e.g., Sachse et al., 2012). The isotopic composition of precipitation is affected by global ice volume (by changing the mean ocean surface water isotopic composition, ice effect), condensation temperature (temperature effect), increased depletion by rainout at longer distance from the source (continental effect) and at higher altitudes (altitude effect) as well as rainfall intensity (amount effect) (Gat, 1996, Dansgaard, 1964). Second order effects are from carbon fixation pathways (CAM, C3, C4), plant life forms (trees versus grasses), biosynthesis, and fractionation during evaporation from soil as well as during transpiration from leaves (see review by Sachse et al., 2012).

Aridification accompanied by an intensification of the dry season in connection with enhanced fire activity since the Miocene has been proposed to have triggered the expansion of C4 grass dominated vegetation (Linder, 2003, Keeley and Rundel, 2005, Ségalen et al., 2006, Osborne, 2008, Dupont et al., 2011). To test this hypothesis we reconstruct hydrological changes in connection to vegetation development in western South Africa during the Miocene to Pliocene (13.8–2.8 Ma) using the stable hydrogen and carbon isotope compositions of sedimentary n-alkanes derived from plant leaf wax as well as pollen records from deep-sea sediments of ODP Leg 175 Site 1085 (Fig. 1).

Western southern Africa is one of today's important semi-arid regions with abundant C4 and CAM plants in the vegetation (Fig. 1). The gradient from desert, grassland, savannah to finally dry-forest vegetation from the west coast to the hinterland is expected to be highly sensitive to climatic changes.

At present southern African rainfall depends on interactions between the seasonal migration of the Intertropical Convergence Zone (ITCZ) and its southern branch, the Congolian Air Boundary (CAB), over land (Gasse et al., 2008), on sea surface temperatures (SSTs) of the adjacent oceans (Nicholson and Entekhabi, 1987, Gasse et al., 2008, Rouault et al., 2003), and on insolation (Suzuki, 2011). Today, the major source of oceanic moisture for precipitation in southern Africa is the warm surface water of the Indian Ocean (Gimeno et al., 2010). Upwelling induced cold South-East Atlantic surface water inhibits strong moisture-laden sea breezes except in the southernmost winter-rainfall area (Fig. 1) (Tyson and Preston-Whyte, 2000). South of the ITCZ the CAB separates air masses of the Atlantic and Indian Oceans and hinders Atlantic moisture to penetrate far into the hinterland. The South Atlantic high pressure cell blocks the westward flow of moist air from the Indian Ocean by enhancing persistent trade winds along the western coastline. Except its southernmost tip, most areas of southern Africa are characterized by seasonal rainfall with predominance in austral summer, when the ITCZ is at its southernmost position.

Modern vegetation in western southern Africa between 25 and 33°S consists of Nama Karoo semi-desert and grassy scrubland, which is rich in C4 plants. The Namib Desert lies west of the Nama Karoo stretching along the Namibian coast from 15 to 26°S (Giess, 1998). The Succulent Karoo, which is rich in CAM plants, lies south of the Namib Desert and southwest of the Nama Karoo, where summer rainfall ceases and only very little rainfall takes place during austral winter. CAM plants are particularly frequent in the desert strip around the Orange River mouth (Cowling et al., 1997). In the far south, the rich Cape flora grows in the winter rainfall region (Mucina and Rutherford, 2006), which is dominated by Fynbos, an exclusively C3 vegetation (Werger and Ellis, 1981).

The strong seasonality of the southern African climate with a hot wet and a cooler dry season probably developed around the Miocene/Pliocene transition (Linder, 2003, Osborne, 2008). Only at the southernmost tip of Africa the modern winter rainfall regime with summer drought conditions became established somewhat earlier, approximately 10–8 Ma ago, and is associated with decreasing SSTs during the onset and intensification of Benguela upwelling (Linder, 2003, Bytebier et al., 2011, Dupont et al., 2011, Rommerskirchen et al., 2011).

Section snippets

Material

ODP Site 1085 (29°22.5′S, 13°59.4′E, 1713 m water depth) is located on the southern rim of the Orange River deep-sea fan (Wefer et al., 1998). The site is situated in the mid‐Cape Basin south of the modern upwelling centre. The upwelling offshore from the Orange River is subdued and less productive than the Namibian Upwelling between Walvis Ridge and Lüderitz (Berger et al., 2002). ODP Site 1085 is located approximately 300 km away from the mouth of the perennial Orange River at the margin of

Results

Pollen concentrations per ml sediment are extremely low before 11 Ma, and then rise to values between 40 and 200 pollen and spores (with one exception) until 7 Ma. Afterwards pollen concentrations strongly fluctuate between values of 100 and 600. The increase in pollen concentrations is associated with lower SSTs and increased total organic carbon (TOC) in sediments of ODP Site 1085 (Fig. 2) (Rommerskirchen et al., 2011).

Between 9 and 6 Ma grass pollen percentages (Poaceae%) rise from a minimum of

Terrestrial input

Terrestrial plant matter, e.g. leaf wax components and pollen and spores, are delivered to deep-sea sediments by fluvial (Orange River discharge, Fig. 1) and/or aeolian transport (SE trade winds), and comprise representative signatures of the nearby adjacent continental vegetation (Schefuß et al., 2003, Rommerskirchen et al., 2003, Dupont, 2011).

The organic matter contribution by the Orange River as indicated by the BIT index decreases steadily until 11 Ma and is very low since (Fig. 2) (

Conclusions

  • Stable carbon isotopes of terrestrial plant wax combined with pollen data retrieved from ODP Site 1085 offshore western southern Africa indicate an increase of grasslands since 8 Ma and a maximum expansion of C4 grass savannahs between 5 and 4.5 Ma. Aridification, however, continued leading to a decrease of the grasslands and an expansion of desert and semi-desert vegetation.

  • The compound-specific hydrogen isotope record from the same sediments indicates a change in the main source of

Authors contributions

Experimental geochemical work and data analysis was carried out by F.R. and E.S.; palynological analysis was conducted by L.M.D.; F.R., L.M.D., G.M. and E.S. contributed to the interpretation and writing.

Additional information

Data are stored in the Pangaea database (www.pangaea.de).

Acknowledgements

We thank Sarah Feakins and an anonymous reviewer for their constructive comments. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, Bonn, Germany) within the research unit “Understanding Cenozoic Climate Cooling: The Role of the Hydrology Cycle, the Carbon Cycle, and Vegetation Changes” (FOR 1070), Grant no. SCHE 903/6. Laboratory assistance was provided by Sabrina Renken, Elke Joost, Katharina Siedenberg, Abhinav Gogoi, Monika Segl and Wolfgang Bevern.

References (93)

  • E.C. Hopmans et al.

    A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids

    Earth Planet. Sci. Lett.

    (2004)
  • Y. Huang et al.

    Mapping of C4 plant input from North West Africa into North East Atlantic sediments

    Geochim. Ccosmochim. Acta

    (2000)
  • M.H. Iversen et al.

    High resolution profiles of vertical particulate organic matter export off Cape Blanc, Mauritania: degradation processes and ballasting effects

    Deep Sea Res. I

    (2010)
  • L.J. Maher

    Nomograms for computing 0.95 confidence limits of pollen data

    Rev. Palaeobot. Palynol.

    (1972)
  • L.J. Maher

    Statistics for microfossil concentration measurements employing samples spiked with marker grains

    Rev. Palaeobot. Palynol.

    (1981)
  • E.M. Niedermeyer et al.

    Orbital- and millennial-scale changes in the hydrologic cycle and vegetation in the western African Sahel: insights from individual plant wax delta D and delta C-13

    Quat. Sci. Rev.

    (2010)
  • C. Robert et al.

    Clay mineral assemblages, siliciclastic input and paleoproductivity at ODP Site 1085 off Southwest Africa: a late Miocene-early Pliocene history of Orange river discharges and Benguela current activity, and their relation to global sea level change

    Mar. Geol.

    (2005)
  • F. Rommerskirchen et al.

    Chemotaxonomic significance of distribution and stable carbon isotopic composition of long-chain alkanes and alkan-1-ols in C4 grass waxes

    Org. Geochem.

    (2006)
  • S. Schouten et al.

    Distributional variations in marine crenarcheotal membrane lipids: a new tool for reconstructing ancient sea water temperatures?

    Earth Planet. Sci. Lett.

    (2002)
  • D.P. Schrag et al.

    The oxygen isotopic composition of seawater during the Last Glacial Maximum

    Quat. Sci. Rev.

    (2002)
  • L. Ségalen et al.

    Neogene climate change and emergence of C4 grasses in the Namib, southwestern Africa, as reflected in ratite 13C and 18O

    Earth Planet. Sci. Lett.

    (2006)
  • R.W. Smith et al.

    A re-evaluation of the use of branched GDGTs as terrestrial biomarkers: implications for the BIT Index

    Geochim. Cosmochim. Acta

    (2012)
  • C.P. Summerhayes et al.

    Variability in the Benguela current upwelling system over the past 70,000 years

    Prog. Oceanogr.

    (1995)
  • B.J. Tipple et al.

    A 35 Myr North American leaf-wax compound-specific carbon and hydrogen isotope record: implications for C4 grasslands and hydrologic cycle dynamics

    Earth Planet. Sci. Lett.

    (2010)
  • L. Vidal et al.

    Late Miocene stable isotope stratigraphy of SE Atlantic ODP Site 1085: relation to Messinian events

    Mar. Geol.

    (2002)
  • A. Vogts et al.

    Distribution patterns and stable carbon isotopic composition of alkanes and alkan-1-ols from plant waxes of African rain forest and savanna C3 species

    Org. Geochem.

    (2009)
  • A. Vogts et al.

    n-Alkane parameters from a deep sea sediment transect off southwest Africa reflect continental vegetation and climate conditions

    Org. Geochem.

    (2012)
  • M.A.J. Werger et al.

    Photosynthetic pathways in the arid regions of South Africa

    Flora

    (1981)
  • T. Westerhold et al.

    Middle to late Miocene oxygen stratigraphy of ODP site 1085 (SE Atlantic): new constrains on Miocene climate variability and sea-level fluctuations

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2005)
  • K.A.F. Zonneveld et al.

    Preservation and organic chemistry of Late Cenozoic organic-walled dinoflagellate cysts: a review

    Mar. Micropaleontol.

    (2008)
  • S. Archibald et al.

    What limits fire? An examination of drivers of burnt area in Southern Africa

    Global Change Biol.

    (2009)
  • D.J. Beerling et al.

    The origin of the savanna biome

    Global Change Biol.

    (2006)
  • W.H. Berger et al.

    Upwelling history of the Benguela–Namibia system: a synthesis of Leg 175 results

  • H.J.B. Birks et al.

    Numerical Methods in Quaternary Pollen Analysis

    (1985)
  • G.J. Bowen et al.

    Interpolating the isotopic composition of modern meteoric precipitation

    Water. Resour. Res.

    (2003)
  • J.M. Bremner et al.

    Sedimentological aspects of the 1988 Orange River floods

    Trans. R. Soc. S. Afr.

    (1990)
  • M. Butzin et al.

    Miocene ocean circulation inferred from marine carbon cycle modeling combined with benthic isotope records

    Paleoceanography

    (2011)
  • B. Bytebier et al.

    Estimating the age of fire in the Cape flora of South Africa from an orchid phylogeny

    Proc. R. Soc. B: Biol. Sci.

    (2011)
  • I.S. Castañeda et al.

    Wet phases in the Sahara/Sahel region and human migration patterns in North Africa

    Proc. Natl. Acad. Sci.

    (2009)
  • T.E. Cerling et al.

    Woody cover and hominin environments in the past 6million years

    Nature

    (2011)
  • J.A. Coetzee

    Pollen analytical studies in east and southern Africa

    Palaeoecol. Afr.

    (1967)
  • G.J. Collatz et al.

    Effects of climate and atmospheric CO2 partial pressure on the global distribution of C4 grasses: present, past and future

    Oecologia

    (1998)
  • J.A. Collins et al.

    Interhemispheric symmetry of the tropical African rainbelt over the past 23,000 years

    Nat. Geosci.

    (2011)
  • R.M. Cowling et al.

    Vegetation of Southern Africa

    (1997)
  • W. Dansgaard

    Stable isotopes in precipitation

    Tellus

    (1964)
  • Cited by (0)

    1

    Now at: Intertek Food Services GmbH, Bremen, Germany.

    View full text