Theses Doctoral

Late Cenozoic Evolution of Aridity and C4 Vegetation in North Africa

Rose, Cassaundra Ashley

Northern Africa has experienced major shifts towards aridity and extensive C4 vegetation over the late Cenozoic, but due to a scarcity of spatially and temporally extensive paleoenvironmental records, the timing, patterns, and causes of these shifts are still under debate. Both long-term aridification and large amplitude orbital-scale climate variability have been recognized, with little understanding of how these two patterns relate to each other over time. African’s climate and environmental history of the last 7 Myr is of particular interest because hydrological and vegetation variability is considered the driving selection mechanism for human evolution. In addition, the age of the initiation of desert conditions in the modern Sahara desert, Earth’s largest warm desert and the largest source of dust to the modern atmosphere, is unknown.
The stable isotope ratios of carbon and hydrogen in sedimentary plant leaf wax biomarker compounds have recently been shown to quantitatively track source vegetation photosynthetic pathways and the hydrogen isotope composition of plant source water, which is dominantly controlled by the amount of precipitation in Africa. These proxies have been applied to reconstruct long-term vegetation changes in East Africa and SW Africa over the last 14 Ma, as well as orbital-scale variability from various locations around the African continent, but they have not been extended further back in time or combined in tandem to robustly assess both long-term and orbital-scale climate and vegetation variability and how they relate to each other.
In this thesis, I have utilized quantitative plant leaf wax stable isotope proxies to examine both orbital-scale and long-term changes in North African aridity and vegetation from a variety of regions over the last 25 Ma, with particular emphasis on the last 4.5 Ma. In Chapter 2, I investigated the evolution of hydrological and vegetation gradients from the equator to the sub-Sahara in NW Africa over the last 25 Myr using leaf wax stable isotopes at two marine sediment core locations, producing the longest existing leaf wax stable isotope record in Africa to my knowledge, and one of the longest such records globally. In this study I found that NW African environments were remarkably similar at both latitudes from 25 – 10 Ma, but at 10 Ma C4 vegetation abruptly expanded in the north, indicating sudden aridification in the Sahara region at that time. The hydrogen isotope record was stable long-term, with variability similar to that of known orbital-scale cyclicity in the Pliocene and Pleistocene, possibly suggesting that orbital-scale cyclicity or other factors obscured or were larger than any long-term changes in the hydrogen isotope ratio of precipitation. Saharan aridification at 10 Ma is consistent with climate model predictions of aridity due to the closure of the Tethys Seaway connection between the Indian Ocean and Mediterranean Sea near that time. The 10 Ma expansion in C4 vegetation is earlier than most other regions globally.
To examine long-term changes in orbital-scale variability in the Eastern Sahara and Mediterranean Sea, I constructed a record of eastern Mediterranean sedimentary leaf wax carbon and hydrogen isotopes, leaf wax abundance, lignin biomarkers, and oxygen isotope ratios of planktonic foraminifera G. ruber during two 100-kyr periods of equal eccentricity near 3.0 and 1.7 Ma (Chapter 3). I found that precession-scale variability dominates the record during both periods, and Eastern Saharan precipitation and the vegetation assemblage, which was C4-dominated, do not change on average between the two periods.
Chapter 4 extended the eastern Mediterranean record of Chapter 3 by sampling leaf wax stable isotopes in sapropel sediments (deposited during North African humid periods) at ~0.25 Myr resolution back to 4.5 Ma, placing the orbital-scale Chapter 3 results in long term context. I found that Eastern Saharan environments were persistently C4-dominated (>68%) throughout the entire interval, and that long-term hydrogen and carbon variability were similar in magnitude to orbital-scale cycles back to 4.5 Ma, strongly indicating that orbital-scale variability has been the dominant environmental control in NE Africa since the early Pliocene. This record contrasts sharply with observations of a transition from C3-C4 mixed vegetation to abundant C4 grasslands in East Africa over the same period of time. The results may suggest that long-term precipitation shifts did not occur in NE Africa since the Pliocene, or that the resolution of this approach is not sufficient to detect long-term shifts. It is likely that NW Africa also experienced similarly large hydrological variability over the same period of time, which may explain the unclear long-term hydrological signal in Chapter 2. The results emphasize that East Africa has not been representative of northern Africa as a whole since the Pliocene.

Geographic Areas


  • thumnail for Rose_columbia_0054D_13083.pdf Rose_columbia_0054D_13083.pdf application/pdf 8.32 MB Download File

More About This Work

Academic Units
Earth and Environmental Sciences
Thesis Advisors
deMenocal, Peter B.
Ph.D., Columbia University
Published Here
December 29, 2015