Tracking Neogene growth and collapse of the Antarctic Ice Sheet (AIS) by detrital provenance analysis


PhD Candidate: Roland Neofitu

Supervisor: Dr Chris Mark

Funded by: Science Foundation Ireland through grant 18SIRG5559



The Antarctic ice-sheet (AIS) since the time of its major expansion into a permanent ice sheet during the Late Eocene - Early Oligocene (ca. 34 Ma - Zachos et al., 2008), has likely undergone a dynamic history of alternating retreat-expansion cycles in response to global climate variation (and on geological timescales, changes in underlying topography; Gasson et al., 2016; Pierce et al., 2017). This variability was especially pronounced during the Miocene and this has been documented by a wealth of indirect global evidence from primarily marine and lacustrine proxy δ18O fluctuations (Zachos et al., 2008), as well as limited direct evidence from ice-sheet proximal deep-sea marine drill cores (Gasson et al., 2016). These data have been used to argue for a major retreat in the size of the AIS as a response to the Mid-Miocene Climate Optimum (MMCO; marked by a peak in global temperatures and a rise in sea level (e.g., Methner et al., 2020)), followed by a major AIS expansion and stabilisation during the Mid-Miocene Climate Transition (MMCT - 14.2 to 13.8 Ma, saw a rapid cooling of ca. 6-7 °C in the high-latitude Southern Ocean).

The MMCT is recorded by numerous IRD-rich sediment horizons in marine sediment cores around the Antarctic margin, reflecting iceberg calving during times of ice-sheet instability (Kennett and Barker, 1990). A key step in ground-truthing MMCT AIS instability is to determine the location of iceberg calving sites, thus highlighting ice-sheet sectors exhibiting repeated instability. We will use these data to evaluate predictions from the latest generation of paleo-ice sheet models (e.g., Gasson et al., 2016), which will be an important tool for more accurate forecasting of anthropogenic climate change. We will do this by comparing our IRD sources to the modelled zones of ice-sheet retreat during MMCT.

We will use sedimentary provenance techniques to address the above problems, at first focusing on the Pensacola subglacial basin and the adjacent potential sediment sources during the Mid-Miocene. Gaining a better understanding of iceberg calving sites will allow us to comment on the hypothesized model-based Miocene extent of the ice-sheet margin around the Weddell Sea. We will accomplish this by source-terrane fingerprinting the IRD samples in an attempt to match them to key isotopic bedrock data from the continent.

Detrital provenance of IRD heavy mineral phases primarily using the U-Pb system (e.g., apatite and zircon) will be used, to determine where those sediments originated by correlating detrital ages to reported bedrock ages onshore. In IRD where heavy minerals are insufficiently abundant for source-terrane fingerprinting, we are able to use rock-forming phases, such as 87Rb/87Sr dating of K-feldspar, as well as Pb- and Sr-isotope analysis of plagioclase for fingerprinting the IRD.



Gasson, E, et al., 2016, Proceedings of the National Academy of Sciences, v. 113, (13), p. 3459-3464, doi:

Kennett, J.P., and Barker, P.F., 1990, in Proceedings, Ocean Drilling Program, scientific results, v. 113, p. 937-962.

Methner, K., et al., 2020, Scientific Reports, v. 10, (7989), doi:

Pierce, E.L., et al., 2017, Earth and Planetary Science Letters, v. 478, p. 1-13, doi: 10.1016/j.epsl.2017.08.011.

Zachos, J.C., et al., 2008, Nature, v. 451, p. 279-283, doi: