Ocean Temperature Changes Across the Eocene-Oligocene Boundary

One of the major challenges in palaeoclimate research is understanding the transition from the ice-free, high carbon dioxide 'greenhouse world' of the Cretaceous and Paleogene to the colder, low carbon dioxide 'icehouse world' in which we live today. During this long-term climate evolution, one of the most dramatic events was the relatively rapid growth of permanent ice sheets on the Antarctic continent and co-occurring drop in sea level about 34 million years ago (the Eocene-Oligocene, or E-O, boundary). It is thought that this event was caused by a decrease in atmospheric carbon dioxide levels and/or a change in ocean circulation. Obviously, a crucial piece of data is a high-resolution record of ocean temperatures during this transition; to achieve this, we will develop geochemical records in sediments from ODP Site 1218 in the equatorial Pacific. However, our ability to interpret such records from Site 1218 / and comparable sites / is limited by a number of factors. The classical approach for obtaining sea surface temperature records is by analysing the oxygen isotopic composition in the calcium carbonate shells of foraminifera living in the surface waters of the ocean. Unfortunately, it is becoming clear that such records are typically compromised, because the foraminifera shells dissolve and recrystallise in the deep part of the ocean, effectively erasing the original temperature signal. Bottom water temperature records, based on oxygen isotopes in benthic foraminifera, are more reliable; these records, however, are hard to interpret because the oxygen isotopes are affected by both temperature and global ice volume / which obviously changed dramatically across the E-O boundary. Thus, to obtain high resolution ocean temperature records requires the application of new approaches. One of these tools is based on the composition of cellular fats in marine microorganisms / the TEX86 proxy. The rings in these compounds are related to the structure of the organism's cell membrane, such that the number of rings increases with the temperature of the growth environment. Because this proxy is based on organic compounds, it is not susceptible to the same dissolution concerns that affect proxies based on calcium carbonate shells. To obtain bottom water temperatures we will measure the amount of magnesium that has partitioned into the benthic foraminiferal calcium carbonate shells (Mg/Ca ratios). Mg/Ca ratios are not affected by ice volume and have been used to develop long-term records of deep ocean cooling; but they are also affected by the distribution of carbonate species in the ocean and that appeared to change dramatically at the E-O boundary as well. Therefore, we will measure both the Mg/Ca ratios and B/Ca ratios, which reflect carbonate ion distributions; together they can be used to decipher both the temperature and alkalinity change in the deep ocean. These two approaches will be applied in parallel for twenty-five samples spanning the E-O boundary. This will allow us to determine exactly how much the deep and shallow Equatorial Pacific cooled across this major climate event. It will also allow us to determine if there were any negative feedbacks in the system; for example, did the build up of ice sheets slow down erosion and allow carbon dioxide concentrations to build up again? Finally, by comparing deep and shallow ocean cooling, we will be able to evaluate models of ocean circulation change.