Assessment of Antarctic sea ice by surface validated satellite measurements

Satellite investigations have documented Antarctic sea ice area, but are restricted in their ability to provide volume, as the procedure to derive thickness is still under development. This procedure requires the measurement of sea ice freeboard, the segment of ice held above the ocean surface by buoyancy. This measurement can be made by satellite altimeters and in conjunction with density and snow depth information; sea ice thickness can be estimated via the hydrostatic equilibrium assumption. The ability to monitor the spatial and temporal characteristics of the thickness distribution must be improved as we strive to understand the linkages between the glaciological, atmospheric and oceanic components of the Antarctic climate system. A key sector in which these components interact is the Antarctic coast. There, offshore winds drive coastal polynyas creating vast amounts of sea ice, and ice shelf interaction modifies ocean properties. Together they condition the ocean for downwelling, driving the global oceanic circulation. In light of this, the coastal Antarctic is a fundamental region in regard to Antarctic sea ice processes and the Earth climate system. McMurdo Sound occupies a coastal area in proximity to an ice shelf in the south-western corner of the Ross Sea. The sound has witnessed scientific investigation for over a century with a fully established research programme since the 1960s. However, the sea ice research in this region is spatially restricted. This thesis aims to expand the knowledge of sea ice in McMurdo Sound to a larger area using space-borne remote sensing instrumentation and design of in situ measurement campaigns. In doing so, this work evaluates the capabilities of satellite platforms to record sea ice freeboard in the coastal Antarctic, whilst developing knowledge of ice shelf-sea ice interaction. This work provides the first satellite altimeter based investigation of sea ice freeboard in McMurdo Sound using ICESat over the period 2003-2009. No observable trend was observed for first-year sea ice freeboard in the region in line with larger scale assessments in the Ross Sea. However, there was significant increase in the freeboard of a temporary multiyear sea ice regime, the segment of the largest increase linked to the outflow of supercooled Ice Shelf Water (ISW) from the McMurdo and Ross Ice Shelf cavities. This remote sensing assessment supports the in situ and modelling work of many others who have identified the influence of ISW on sea ice processes in this region, in particular, that it is thicker than it would otherwise be. The influence of ISW on altimetric sea ice thickness retrievals was also quantified using a Global Navigation Satellite System (GNSS) evaluation of freeboard to thickness conversion. This revealed that a sub-ice platelet layer, created by supercooled ISW and with an estimated solid fraction of 0.16, accumulates beneath the sea ice cover and influences the thickness estimates from the GNSS-derived surface elevation. A cautionary conclusion is reached that within 100 km of ice shelves this buoyant influence should be considered, and in close proximity (< 50 km) can result in overestimations of sea ice thickness of ~ 12 %. It is also suggested that the sea ice freeboard anomalies that result from enhanced growth, driven by supercooled water advection could be used to map the presence of ISW in the coastal Antarctic. Looking to future ability to monitor Southern Ocean sea ice thickness from space, the first comprehensive evaluation of CryoSat-2 (CS-2) over Antarctic sea ice is provided. Using three separate retracking procedures, CS-2 is shown to be capable of detecting the development of a fast ice cover in McMurdo Sound. The role played by a snow cover with layering typical of the Antarctic appears to cause a positive bias in the ice freeboard for a waveform fitting procedure currently used over Arctic sea ice. The identification of open water and the establishment of accurate sea surface heights are also indicated as causing errors (in the order of cms) in the study region. CS-2 is shown to be capable of recording sea ice growth over two growth cycles in McMurdo Sound. This work has advanced the application of satellite investigative techniques to Antarctic sea ice, providing hope that such techniques may be capable of revealing larger scale connections between sea ice and ice shelves