Dear Mal et al.,

Thanks for your detailed response to the points raised by the two referees.

Please submit your manuscript with the suggested changes. This will go in review again with one of the initial referees.

Best regards,
Louise

Dear authors,

Your paper has now been assessed again by one of the original referees, and I am happy to inform you that your manuscript is acceptable for publication once you have satisfactorily implemented the last few edits suggested by the referee (see below). These final edits will be assessed by me, before final publication.

Best,
Louise

----------
referee comments:

I am pleased to see the much more convincing results in Sect.3 and acknowledge the additional work the authors have put into this analysis and several other parts of the manuscript. I have only a few remaining minor comments on the revised version.

One main point is that there are several places where the reader is referred to McMillan et al., 2019. This is okay when referring to details about the approach (as e.g. in l.111). However, Sect.5 is impossible to interpret at all without switching between those two papers several times. I think, also a “Brief Communication” should be understandable on its own (this refers mainly to my comments on l.104 and l.117).

After the following minor comments are addressed, I suggest the manuscript for publication:

l.53 Why is this region called “Spirit site”? Isn’t it Adélie Land / George V Land? How are OIB flights distributed there? You say, you want to make the validation comparable to McMillan et al., 2019, but why don’t you use the regions “Dronning Maud Land” and “Wilkes Land” then?

Fig. 1a) I suggest to use a more gradual color scale like “Viridis” or “Plasma”. Furthermore, it is difficult to decide if black color is the shade of the relief or means R<=0.7 (both would be expected in rugged terrain).

l.104 I guess that “dispersion” means MAD. Please define (also for Sect. 5).

l.111 Briefly mention what “reference datasets” are (OIB/airborne profiles, refer to Sect.2).

l .117 In order to give the reader the ability to interpret the significance by her/his own, please also mention the median bias and the MAD between S3 and the reference dataset (or at least the order of magnitude). I know that this is given in Tab.3 of McMillan et al., 2019 but mentioning at least the key results (or order of magnitude) here would make this section understandable on its own.

Dear Mal et al,

thanks for revising your manuscript according to the referees suggestions. I think that the color scale in your Fig 1a is fine.

Best,
Louise