Evaluation of CryoSat-2 derived sea-ice freeboard over fast ice in McMurdo Sound, Antarctica

摘要

Using in situ data from 2011 and 2013, we evaluate the ability of CryoSat-2 (CS-2) toretrieve sea-ice freeboard over fast ice in McMurdo Sound. This provides the first systematic validationof CS-2 in the coastal Antarctic and offers insight into the assumptions currently used to process CS-2data. European Space Agency Level 2 (ESAL2) data are compared with results of a Waveform Fitting(WfF) procedure and a Threshold-First-Maximum-Retracker-Algorithm employed at 40% (TFMRA40). Asupervised freeboard retrieval procedure is used to reduce errors associated with sea surface heightidentification and radar velocity in snow. We find ESAL2 freeboards located between the ice and snowfreeboard rather than the frequently assumed snow/ice interface. WfF is within 0.04m of the icefreeboard but is influenced by variable snow conditions causing increased radar backscatter from theair/snow interface. Given such snow conditions and additional uncertainties in sea surface heightidentification, a positive bias of 0.14m away from the ice freeboard is observed. TFMRA40 freeboardsare within 0.03m of the snow freeboard. The separation of freeboard estimates is primarily driven bythe different assumptions of each retracker, although waveform alteration by variations in snowproperties and surface roughness is evident. Techniques are amended where necessary, and automaticfreeboard retrieval procedures for ESAL2, WfF and TFMRA40 are presented. CS-2 detects annual fasticefreeboard trends using all three automatic procedures that are in line with known sea-ice growthrates in the region.