Our western scientific understanding of climate relies on modern observations and, for areas and timescales for which those are unavailable, inferences from paleoclimate records. Much of this dissertation aims to develop tools to understand Quaternary climate of the past 2.6 million years, which is characterized by cycles of glaciation. Understanding the increase in magnitude and duration of these glacial-interglacial cycles has motivated international efforts to retrieve an Oldest Ice core in Antarctica containing samples of the ancient atmosphere. To aid in careful site selection, we explore the use of marine dust records as dating templates for Oldest Ice dust, which can be optically logged in a borehole without retrieval of an intact ice core. A new marine dust record close to South American dust sources offers the most promising template for this purpose. While ice cores provide unique and crucial records of global and polar climate, records of regional paleoclimate are also important. Within the timescale of the late Pleistocene to Holocene, noble gases dissolved in groundwater and their isotopic composition serve as valuable tracers of terrestrial climate, but the interpretation of excess dissolved gas (excess air) remains ambiguous. We develop a new sampling and analytical method for dissolved noble gas isotopes and demonstrate that this method is precise and unbiased through air-equilibrated water standards, inter-laboratory comparisons, and comparisons to previous data. We use this new method along with prior data to show that more excess air is generally associated with shallower water tables, though one arid site complicates this finding. Finally, we consider climate futures and the role of geoscience in upholding or subverting colonial relationships with land and people. We move the discussion beyond metaphorical invocations of decolonization, which obscure the full meaning and power of returning land and life to Indigenous peoples. Even motivated by climate change, colluding with state and capitalist forces for a techno-centric energy transition reproduces colonial resource extraction. Aligning geoscience with Indigenous land and water struggles and challenging assumptions of energy and material use offer ways to affirm life rather than death.

Gases preserved in ice cores provide a potential direct archive for atmospheric oxygen. Yet, oxygen-to-nitrogen ratios in ice cores (expressed as δO₂/N₂) are modified by a number of processes related to gas trapping and gas losses in the ice. Such complications have long hindered the use of ice core δO₂/N₂ to derive true atmospheric oxygen concentrations. Recently, a persistent decline in δO₂/N₂, observed in four different ice cores (GISP2, Vostok, Dome F, and EDC), is interpreted to reflect decreasing atmospheric O₂ concentrations over the late Pleistocene (Stolper et al., 2016). The rate of δO₂/N₂ change is -8.4±0.2 ‰/Myr (1σ). Using new measurements made on EDC samples stored at -50 °C and therefore free from gas loss, Extier et al (2018) confirms the decrease in δO₂/N₂ with a slope of -7.0±0.6‰/Myr (1σ).

Here, we present new δO₂/N₂ measurements made on 1.5-million-year-old blue ice cores from Allan Hills Blue Ice Areas, East Antarctica. We use argon-to-nitrogen ratios (δAr/N₂) in the ice to correct for the fractionations during bubble close-off and gas losses. In those processes, δAr/N₂ is fractionated in a fashion similar to δO₂/N₂ (Huber et al., 2006; Severinghaus and Battle, 2006). Paired δO₂/N₂-δAr/N₂ values measured from the same sample were classified into three different time slices: 1.5 Ma (million years old), 950 ka, and 490 ka. Between 950 ka and 490 ka, we observe a decline in δO₂/N₂ similar to that observed in the aforementioned deep ice cores. This observation gives us confidence in the validity of the Allan Hills blue ice δO₂/N₂ records. Between 1.5 Ma and 950 ka, however, there is no statistically significant trend in ice core δO₂/N₂. Our results show a surprising lack of variability from 1.5 to 0.95 Ma; even during the past ~0.9 Ma, the rate of decline was very slow.