Radar-derived bed reflectivity across Pine Island Glacier

Over the last two decades remote sensing observations of accelerating and thinning ice streams draining the West Antarctic Ice Sheet (WAIS) have reignited concerns over the long-term future of the ice sheet and its contribution to global sea-levels. Pine Island Glacier (PIG) is one of the key ice streams contributing to mass loss from the WAIS, and a number of numerical-modelling initiatives have been set up to improve our understanding of the main processes involved, ultimately to aid forecasting of its future contribution to sea level rise in the 21st and subsequent centuries. Critical to such modelling is a good representation of the basal boundary conditions, with radar systems offering the greatest opportunities for the acquisition of pertinent data. In the austral season 2007-08 we towed a 1 MHz monopulse radar, configured to stack groups of traces at 5-10 m intervals, along and across the main trunk and several tributaries of PIG. The data were used to determine bed reflectivity and englacial attenuation across the PIG catchment. Overall, bed reflectivity is heterogeneous and generally bright beneath PIG, albeit with two trends: (i) the basal interface becomes brighter downstream, with a positive step in brightness coinciding with flow out of a deep upstream basin; and (ii) brighter basal echoes are observed on the southern flanks of the basin, towards the neighbouring Thwaites Glacier catchment. The correspondence of variations in bed reflectivity with subglacial topographic features suggests a subglacial geological influence, and subglacial volcanism, known to exist in Pine Island Glacier and thought to exist beneath Thwaites Glacier, may be responsible for some particularly bright spots. We find little correspondence between bed reflectivity and the spatial distribution of ice-flow velocities, suggesting that the links between high bed reflectivity, high basal lubrication, and thus high ice-flow rates, are not straightforward.