Line 88: At the L0 to L1B processing stage, Hamming-weighting and zero-padding were applied, both of which have been shown to be essential for sea ice retrieval and which are not included inthe ESA standard processing at this time (Lawrence et al., 2019a).

This statement is not true for CryoSat (which does have Hamming/ZP in the ESA processing). I suggest that the paragraph is changed to state that Hamming/ZP are only currently missing from the ESA processing for S3 and not CS2.

The arcticle could be clearer on where CryoSat and Sentinel-3 operate in different modes (SAR for S3, SARin for CS2) over certain areas of the Arctic such as the Canadian Archipelago,  what differences in freeboard might be expected in these areas, and the subsequent effect on regional and pan-Arctic statistics. The authors might consider restricting the analysis or including an analysis of  areas of the Arctic where CS2 and S3 operate in the same mode (SAR). For example, on line 292, 'significant discrepancies between CS2 and S3 freeboards' are noted but not explained. This could be partially explained by the different modes in operation.

On line 292, 'Candian Archipelago', should be 'Canadian Archipelago'