EVIDENCE FOR EAST ANTARCTIC OLIGOCENE GLACIATION FROM ODP SITE 693, DRONNING MAUD LAND (DML)

The abrupt, >1 per mil, increase in delta¹⁸O at the Eocene-Oligocene Transition (EOT) is attributed to ice growth on Antarctica. Dronning Maud Land (DML) adjacent to the Weddell Sea, East Antarctica, is identified in numerical models to be one of the three sites of glacial nucleation, based, in part, on their elevation (DeConto and Pollard, 2003). Ice growth progressed more rapidly at the other sites, the Transantarctic and the Gamburtsev Mountains, eventually coalescing into a continental-scale ice sheet. However, the dynamics of glacial advances and retreat are poorly known. ODP Site 693, in 2359 m of water, at the margin of DML provides an opportunity to decipher some of these dynamics. We are examining a 127 m thick early Oligocene to Miocene sequence deposited above a Cretaceous unconformity.

Site 693 is a prime location to examine bedrock erosion and its marine deposition through measurement of the weight % of the sand-size fraction and identifying the provenance of the sand grains. The sand-size fraction, which, in this location, can be attributed to iceberg melting (ice-rafted detritus, IRD), provides a first-order measurement of the amount and scale of continental glaciers and iceberg activity. These icebergs were most likely were derived from the adjacent continent. Our analysis of weight % coarse fraction for samples across this period indicates a contrast between middle/early Miocene and early Oligocene. In the early Oligocene, IRD comprises between 5% to 20% of the sample. IRD is generally much less abundant in the early and middle Miocene. We interpret this to mean that initial cooling and ice formation led to ice-rafting off of Dronning Maud Land in the early Oligocene. During the late Oligocene and early/middle Miocene, either warming led to a decrease in iceberg activity or the icebergs carried significantly less sand. We are in the process of dating detrital sand grains (zircons, hornblendes and biotites) to see if they came from the Grunehogna Craton. This part of DML contains billion year old rocks. The presence of billion year old sand grains in the Oligocene and Miocene sediment would support DML glaciation at this time.