Glacial Antarctic warm events as captured by the RICE ice core
Millennial-scale climatic oscillations have been documented in numerous ice cores in Antarctica and Greenland for the last glacial period. These glacial events, termed Antarctic Isotope Maxima (AIM) and Dansgaard – Oeschger (DO) events in Antarctica and Greenland, respectively, are understood as ocean-mediated mechanisms, with asynchronous progression into warming and cooling. This bipolar seesaw originates from the out-of-phase strengthening and weakening of Antarctic Bottom Water (AABW) and North Atlantic Deep Water (NADW). While proxy records provide strong evidence for enhanced AABW formation during AIM events, the mechanism of its formation during these events remains indiscernible. The process of AABW formation, however, is likely to be different from today. In modern times, AABW formation occurs in front of ice-shelves, whereas during glacial periods, the ice sheets extended to the continental shelf, and the ice-shelves were absent. Ice cores allow the reconstruction of an array of environmental conditions. Here, we present new data from a high-resolution coastal ice core record that sensitively captures changes in temperature, sea ice conditions, primary productivity, and atmospheric circulation pattern. The new data provide important insights into the drivers of AIM events, AABW formation, and the evolution of the bipolar seesaw.
This thesis focuses on the high resolution record of major ions bracketing 26-40 kilo year (ka) Before Present (BP), from Roosevelt Island, West Antarctica. Roosevelt Island is a grounded ice rise, located in the north-east of Ross Ice Shelf, in the Southern Ross Sea. The ice core was drilled to bedrock at 764m, as a part of the Roosevelt Island Climate Evolution (RICE) project. The record has been dated to 753m which captures the past 83 ka BP. The major ion record comprises of concentrations of sodium, magnesium, calcium, potassium, chloride, sulphate, nitrate, and Methane Sulphonic Acid (MSA), with an average resolution of ~25 years.
We demonstrate that the RICE MSA, sea salt aerosols, and non-sea salt calcium records are sensitive recorders of primary productivity, open-ocean area, and latitudinal position of Southern Hemisphere westerly winds, respectively. Our results suggest that AIM events in the Ross Sea region are driven by open-ocean convection. We test this hypothesis using the outputs from LOVECLIM, an intermediate complexity earth system model, based on contrasting scenarios, with and without open-ocean convection in the Ross Sea region. Our results confirm that open-ocean convection as a prominent mechanism for the formation of AABW in the Ross Sea region during AIM events.
Availability of high resolution, continuous flow methane data for the AIM 4 event provides an opportunity to examine the relative phasing of the events in RICE and North Greenland Ice Core Project (NGRIP) ice cores. We demonstrate that the RICE AIM 4 event leads by over 200 years the Greenland DO 4 event. Our data suggest that the Antarctic lead is caused by the coupled interaction between open ocean convection and local insolation changes, which also highlights the significant role of the Southern Ocean in the evolution of bipolar seesaw. Using non-sea salt calcium data in RICE and other Antarctic ice cores to reconstruct atmospheric circulation pattern, we find an equatorward migration of Southern Hemisphere westerly winds between 31-32 ka BP. This shift evokes large scale consequences that include a reduction in atmospheric CO2 and strengthening of tropical rainfall. This atmospheric rearrangement might be linked to reduced summer insolation, leading to Southern Ocean sea surface cooling and subsequent sea ice expansion.
