Early Pliocene Ross Sea paleoclimate and a new application of the diatom oxygen isotope proxy

Oxygen isotope measurements ( 18 O) provide valuable insight into climate conditions.These data are advantageous as paleoclimate proxies, but records are sparse in high-latitude waters, where calcareous microfossils are poorly preserved in marine sediments.Over glacial time scales, high-latitude regions likely have the widest variability in terms of the  18 O values of marine waters; therefore, the paucity of data in these regions greatly hinders our ability to reconstruct global climatic and environmental change.This research is motivated to investigate Southern Ocean diatoms as a way to begin filling the gap in our knowledge.Diatoms are a cosmopolitan group of siliceous phytoplankton with characteristic ecological affinities.Diatom silica is increasingly being analyzed as a  18 O archive, and has the potential to provide comparable data to  18 O variations recorded by benthic foraminifera, but to date, studies have focused on Holocene sediments.This thesis presents oxygen isotope values from Early Pliocene Ross Sea diatoms, assessment of purification procedures, and implications for paleoclimate and the history of the West Antarctic Ice Sheet.ANDRILL AND-1B is a 13 Ma record from the Antarctic Geologic Drilling Project in the Ross Sea that provides Lower to Mid-Pliocene diatomite units.The context provided by age models and paleoenvironmental studies, along with more geologically recent Southern Ocean diatom  18 O measurements, make AND-1B an ideal dataset for further exploring the use of diatom  18 O values as a record of changes in salinity and temperature in high-latitude seawater.Samples were purified, geochemically and mineralogically assessed using X-ray diffraction (e.g., identification of amorphous opal-A vs. mineral contamination) and X-ray fluorescence (e.g., elemental quantification of clay contamination), then analyzed for  18 O values using a step-wise fluorination (SWF) -CO 2 laser heating technique and a Thermo MAT 253 isotope ratio mass spectrometer.Correction methods based on different assumptions place constraints on true diatom  18 O values, with margins of error dependent on final sample purity.The results are intermediate between Holocene Antarctic diatom  18 O studies, indicating that diatom  18 O values can remain largely or entirely unaltered for at least 4.5 Ma.Interpreting these values through the current paleotemperature model, with the contingent paleoenvironmental constraints afforded by AND-1B, suggests water  18 O values offset about -10‰ relative to today's ocean.Stacked benthic foraminifera  18 O values for the Pliocene suggest an ocean average less than 1‰ different than the present; therefore, this points to a strong component of glacial water in a warmer, fresher Pliocene Ross Sea.