Ice thickness, internal structure and subglacial topography of Bowles Plateau ice cap and the main ice divides of Livingston Island, Antarctica, by ground-based radio-echo sounding

Ground-based radio-echo sounding studies of Livingston Island ice cap, Antarctica, were started in 1999 at Johnsons and Hurd glaciers, in Hurd Peninsula, close to the Spanish Antarctic Station Juan Carlos I. Radar profiling continued in 2000 and 2003 along the main ice divides of the ice cap and in the upper parts of Huron and another unnamed glacier, both draining Bowles ice plateau towards Moon Bay. More detailed radar studies on Bowles ice plateau were performed in December 2006 using an icepenetrating radar VIRL-6, with central frequency of 20 MHz, and a Ramac/GPR radar with 200 MHz antennae, intended for deep penetration to bedrock and shallow penetration to determine the firn layer thickness, respectively. The radar equipment was installed on two Nansen sledges and two pulkas towed by snowmobiles. Georeferencing of radar data was accomplished by using a GPS receiver working in stand-alone mode. The endpoints of the radar profiles were more accurately georeferenced using differential GPS. The data collected were used to construct the ice thickness, glacier surface and bedrock elevation maps and to estimate the average ice thickness and total ice volume of the plateau, as well as the spatial distribution of snow accumulation. For time-to-depth conversion, an average radio-wave velocity determined from common midpoint measurements at several locations in Hurd Peninsula was used. The thickest ice (550 m) was found in the upper part of Huron glacier. This thickness is almost twice the maximum thickness found along the main ice divides (330 m) and almost three times that found in Hurd Peninsula (200 m). The bedrock in two large areas in the northern and southern parts of the plateau, towards Moon Bay and Huntress glacier, lies below sea level, at depths down to -200 m, indicating that, should the ice cap fully disappear, several separated islands would likely appear, even if isostacic rebound is taken into account.