Ice shelves are an important component of the Antarctic Ice Sheet as they indirectly control sea level rise by regulating mass flux into the ocean. The coupling of ice shelves with the ocean and the atmosphere makes them vulnerable to climate change. With the oceans absorbing most of the energy from global warming, there is an increased interest in understanding ice-ocean interactions. Basal processes are poorly understood as the base of an ice shelf is difficult to measure due to its inaccessibility.

This thesis explores the effect different internal and basal processes have on ice shelves, and their implications for ice shelf stability, using ground penetrating radar (GPR) and a new phase sensitive radar (ApRES). To achieve this, two study sites were visited. A GPR survey was made on the northern grounding line of the stationary Southern McMurdo Ice Shelf (SMIS) in November 2014 to examine deformation of internal layers, measure ice thickness distribution across a grounding zone, and interpret basal topography. In November 2015, 21 sites in the central Ross Ice Shelf (RIS) were measured with ApRES to estimate the distribution and thickness of marine ice, vertical strain and basal melting/freezing.

Ice thickness of the northern SMIS grounding zone is mapped to high resolution and with an uncertainty of <10 m. Ice is thickest near the grounding line (≈250 m) and thins to 200 m within 3 km seaward of the coast. Basal topography and deformation of internal layers reveal basal processes and interactions of the ice shelf with the ocean. Basal crevasses at the grounding line complicate the radar profile and are created as a result of tidal rather than shear stresses. Downwarping and truncation of internal layers just seaward of the grounding line are caused by basal melting. The generated meltwater directly influences basal topography creating stepped features in the ice shelf base which persist for kilometres from the grounding line.

Widespread marine ice at the base of the RIS is revealed by ApRES point measurements. Marine ice thickness could not be estimated due to possible shortcomings in the hydrostatic equilibrium assumption which produces thickness anomalies in the order of 20 m. This indicates that the marine ice layer has a similar thickness to this uncertainty. In this environment it was found that, calculations of vertical strain and basal melting/freezing made by examining the difference between internal layers with ApRES requires longer than two weeks between repeat visits to measure with sufficient accuracy.

Stationary ice shelves are an ideal location to examine grounding line processes. The ice remains in-situ for sufficient time to deform in response to its extended interaction with these processes. The grounding line of the SMIS has been stable for a long time and is not inherently vulnerable to future warming as it lies on a prograde bed slope. Additionally, the SMIS is buffered from changes in ocean circulation due to its geographical isolation. Despite this, because the SMIS is effectively stationary, any changes at the ice shelf base may lead to thinning at the grounding line or a modification in velocity as ice is limited in its ability to recover.

The relationship between the presence of marine ice and meteoric ice thickness suggests that the distribution of marine ice is primarily controlled by basal topography rather than ocean circulation. Marine ice alters the ice-ocean interaction, and as a result, the RIS will demonstrate a unique response to climate change. The RIS requires further dedicated study in order to examine its stability, and the distribution and thickness of marine ice will play an important role in the initial response it demonstrates to oceanic changes.

This study demonstrates the potential of a phase sensitive and ground penetrating radar to improve knowledge about ice shelf processes. In particular, their ability to reveal complex internal layering and collect high-precision measurements at the ice-ocean interface.