Frémand and her co-authors in their paper entitled "Antarctic BedMAP data: FAIR sharing of 60 years of ice bed, surface and thickness data" make available all the geophysical data (mainly by radio echo-sounding) acquired since the 1960s and crucial in our knowledge of the topography of the bedrock elevation under the Antarctic ice sheet. The opening of these data is an important advance for the community and was widely expected. It will allow various teams to take advantage of the initial dataset to allow the emergence of new methodologies and propose more accurate Digital Elevation Models (DEMs) that are compatible with the physics of ice sheet models in particular (e.g. see the BedMachine initiative). These DEMs are an essential boundary condition for modeling ice dynamics and projecting the future of Antarctica and its contribution to sea level rise. I therefore consider this paper important, well structured and recommend it for publication after minor adjustments.

• Web links appears with a date of access, I presume this is an editorial issue which I guess will be solved before the final publication?
• On first reading, I had trouble understanding whether the shapefile data was the same as the CSV data. I admit that this is fairly obvious when you look at Tables 2 and 3, but the doubt lingered with me for a while. I think it comes from the introduction to section 3.3, the first sentence starting with "the CSV data were converted to shapefile" (L. 260). I think that presenting the need to offer summarised data might come before using the word "convert" which seems to me to be a bit of a misnomer, it is more than a format conversion.
• L 262-264. "For Bedmpa1 (there is a typo), due to the difficulty... not possible to convert BedMap1... : only the bedmap1 shapefile points are provided". I don't understand this sentence, I see a contradiction, the conversion is not possible but the shapefile is provided. This needs to be rephrased. I presume the idea is that BEDMAP1 is provided as only one shapefile and not split by campaign but this is not clear.
• Figure 2 panel d. Maybe the color scale could be adjusted, any red dot can be seen and it is hard to distinguish the two shade of orange used.
• L 309. A space is missing after the dot.
• L 318. The word pixel is introduced when cell was used before. I believe that only one word should be used.
• Regarding the publishing of the data sources. Quantarcica (https://www.npolar.no/quantarctica) is widely used by the community, I would encourage the BEDMAP team to use this platform to further promote their dataset.
• Regarding the access of the data. The front page is very clear (https://www.bas.ac.uk/project/bedmap/#data), one could ask why the Bedmap1 gridding product is not provided. It becomes a bit more fuzzy when we visit the sub pages.
• When we clic on the links provided under the statistically-summarized data points Shapefiles, the title of the following pages  have dropped the word « statistically summarized » to « standardized shapefiles ». This extends the confusion I mentioned earlier.
• When on one of the metadata or data webpages (e.g. https://ramadda.data.bas.ac.uk/repository/entry/show?entryid=a72a50c6-a829-4e12-9f9a-5a683a1acc4a) there is a list of associated datasets. This list of associated datasets is not consistent from one page to another and do not understand why. If the aim is to promote the other datasets proposed by BEDMAP, I do not understand why these paragraphs are not simply entitled « Other BEDMAP products » with only one link to the front page.
• I would find it useful if the data could be downloaded in a few clicks and not mission by mission (e.g. I would have liked to be able to download all the BedMap3 shapefiles in one click for example)

   

General Comments:

This is a significant paper, and one that will be of great interest to researchers.  Not only does it mark the release of a significant data set, but it also does an excellent job of addressing data management considerations.  The information is presented clearly, and there are many aspects which are highly citable.  I have only minor suggestions and corrections, listed below.

Specific Comments:

Line 115 (suggestion): It may worth referencing Section III 1 c of the Antarctic Treaty here - identifying the need beyond a moral imperative to make the data available

Line 246 (suggestion): Consider providing an abbreviated example of what the header and lines might look like.  Information can be condensed using '...', but I think an example would make it easier to visualise the content of this paragraph.

Line 294 (question): -9999 values often present problems for interpolation, or situations where data ranges can encompass -9999.  ALthough that is not the case here - are there any alternatives for better representing NULL/NaN/Uknown values?  I recognise that this is an issue with CSV, but it may be worth thinking about better ways of encoding this information.

Lines 309 & 326 (question) - You discuss how uncertainty in elevation values was handled, but was there also consideration for varying spatial accuracies?

Technical comments:

Figs 1&2 - Polar stereographic projection?

There are many instances where intensifiers(e.g. see https://advice.writing.utoronto.ca/revising/wordiness/) are unnecessarily used.

97: "user-friendly" subjective assertion - consider omitting.

148: "rapidly" is there evidence or a citation?  Otherwise omit.

153: "considerable" - consider omitting

154:  "Indeed," - unnecessary

170: "in order" - unnecessary

265: "extremely" - unnecessary

267: "Such an" - unnecessary

356: "In order" - unnecessary

365: "easily" - subjective assertion - consider omitting

377: "full" - unnecessary

386: "particularly" unnecessary

Paragraph at 414 (several): Change to "This data release will benefit the glaciology and broader Earth scienc community, particularly in emerging fields such as machine learning and geostatistics which can now make use of this standardised data, and reproduce and create new compilation grids at different scales independently from the Bedmap grids.  These standardised, freely available, and previously-unpublished datasets will lead to improved assessments of fundamental properties of the Antarcitc Ice Sheet and predictions of its future contributions to sea level rise, increasing the (life cycle) of these important data."  -- ('life cycle' is an odd choice of words - maybe just 'value'?)

Line 455 - Normally I leave Acknowledgements alone - but "beyond measure" seems a bit hyperbolic.  It's up to you.

Typos:

Line 111: Theses - These

Line 361: Althought - Although

394: Bedamp3 -> Bedmap3

415: geostatiscs -> geostatistics

416: indenpendently -> independently

Line 422: Direct link -> Direct links

Line 423: Capitalise RAMADDA (or standardise usage)