Interactive comment on “ A detailed radiostratigraphic data set for the central East Antarctic Plateau spanning the last half million years ”

The paper by Cavitte et al. (2021) presents an extensive ice-penetrating radar dataset of internal reflecting horizons (IRHs) traced on both airborne and ground radar data over the Dome C area of East Antarctica. The authors go in length to describe the processing of the data, the uncertainties associated to the radar systems, and their processing steps; as well as the methods used to date (directly at the ice core; indirectly using a 1-D ice flow model) their IRHs. This dataset will be very useful to future modelling studies interested in calculating past accumulation rates, ice-flow speeds, and assess past ice-sheet stability over this sector of East Antarctica; and indeed repC1


Answers to RC3
Overall questions I am not very familiar with the standards for data release that may have been put out by the AntArchitecture program in order to facilitate bringing together different data sets in the future, but is the main piece that is new that these data are now available, or that they have been processed in a way that allows them to be used all together? Or, it seems likely both. Does the way that these data are archived follow a standard that was either established by previous data contributions, or does this set one that future archival efforts should follow? What are the ways that ensure that these data can be used together with other data sets in order to advance the goal of assembling a continent-scale radiostratigraphy? Perhaps this is not the paper to describe the protocols and how fields are archived, but given the focus on contributing to AntArchitecture, it would be interesting to mention. Answer: No protocol was defined for archiving IRH data in a consistent way under the AntArchitecture program so far. It is an important task of the program, but until then, contributions to the AntArchitecture continent-wide data set require the IRH data to be open-access. However, we have followed the structure of other published IRH data sets so that there is some uniformity in the data structures across data releases. We have added the following sentence at the end of the introduction: "We archive this data set based on the standards established by previous IRH data contributions (e.g Winter et al., 2019;Bodart et al., 2021;Beem et al., 2021), adapted to fit the information available for this specific data set." The point in the discussion about tracing the same IRH in radar data from different systems may be worth mentioning earlier. It could even be that this  4) has also been moved up). We find that the discussion section now reads more sequentially, after removing this point about tracing the same IRH in radar data from different systems. It now focuses on the geometry of the IRHs, how easy/complicated they are to trace in the region, discusses new model developments for which these IRHs would be interesting and ends with considerations for the AntArchitecture project.

Given it's prominence in the figures, how accurate are the Zwally et al. (2012)
drainage divides? I assume these are from ICESat data, but one version used ERS, and regardless there would be higher resolution data products from which to designate the divide. Would this matter? Does the divide appear where it is expected from the LDC survey? Answer: The Zwally et al (2012) divides used were indeed from ICESat data, from https://earth.gsfc.nasa.gov/cryo/data/polar-altimetry/ antarctic-and-greenland-drainage-systems. The divide position is not discussed in this data release paper: it is only shown to impress upon the readers that the BE-OI search focused on the divide area. But we have now replaced the drainage divides with the Rignot et al (2019) more recent positions, which differ a little, but not significantly across the LDC region. Given that the overall slope is of the order of a meter or two across the DELORES LDC survey region, the divide line does not move in any discernible way in the small area of the DELORES survey region using the higher resolution DELORES radar data.

C4
The title mentions "last half million years" but the deepest layer is ∼700 kyr. Is it preferred to exclude credit for these deepest layers in the compilation because they are dated only with the age model? In that case the title should reflect the past 350 ka -I didn't follow why half a million was referred to in the title. Answer: Although the ages defined of the oldest IRHs are estimates, they are relatively robust. E.g. the 1D age model was updated recently to include a stagnant ice layer (unpublished) and ages predicted are still on the order of half a million years. Lilien et al (2021) used another 1D model over the LDC region and they agree that isochrones can be dated to at least half a million, (they have isochrones between the age of 71 and 565 ka), and their deepest continuous isochrone is dated at 465 ka. We can therefore reasonably assume that our IRH data set goes back half a million years. The choice of title is supported by the modelled deep isochrone ages. But since the deepest isochrone age is an estimate, we have chosen to modify the title to reflect this to: "A detailed radiostratigraphic data set for the central East Antarctic Plateau spanning from the Holocene to the mid-Pleistocene".

Minor Points
Line 40: Perhaps also indicate that accessibility is an issue to radar-data collection in Antarctica (not just size -though of course that is the major challenge) Answer: We have now added this information at this line, which now begins with: "However, due to its sheer size and much lesser accessibility,...". Further changes have been made based on Reviewer 2's comments.
Line 43: It would be worth being consistent with how AntArchitecture is referred to, as the manuscript uses "AntArchitecture project", "program", "community", and "action group", yet I think nearly all of these uses refer to the same. Perhaps C5 "action group" is best since that is how it is referred to on the SCAR page. Would it be appropriate to cite the white paper or link to AntArchitecture page? Answer: We agree and "action group" has now been adopted throughout. The White paper is under way but not published yet, so instead, we include a link to the AntArchitecture report (AntArchitecture, 2017) Line 45-48: "potentially reducing the lack of unique solutions, a persistent problem until now . . ., as well as solving the problem of modeling 3D data in simple 1-D or 2-D models. . ." -I would suggest revising this sentence as it isn't completely clear what is being stated. And, my read of this is to say that a continent-scale radar data set would in itself help to constrain parameter values and would facilitate solving 3-D problems that are better than 1-D or 2-D problems. I may not be understanding the point but I would caution that radar data are one part of the problem, and the ability to use them to solve an inverse problem or constrain a 3-D model and get a robust solution depends on the problem as it is set up (not only including the data). Unfortunately, a 3-D radiostratigraphy won't on it's own make all ice-sheet problems well posed. Answer: We agree. Our wording was awkward and we only meant that a more extensive data coverage (spatially and temporally) can help reduce the uncertainties of modeling experiments that forward or inverse model IRH data by providing more tuning targets. We have reformulated this section as follows: "The construction of a comprehensive Antarctic-wide IRH data set will both play a key role for projects such as the Beyond EPICA -Oldest Ice European search for million-year-old ice (Van Liefferinge and Pattyn, 2013;Parrenin et al., 2017) and provide valuable additional constraints for inverse models, potentially helping in cases where a unique solution could not be found due to a lack of data constraints (a persistent problem until now, e.g. Morse et al., 1998;Eisen et al., 2008;Koutnik et al., 2016;Parrenin et al., 2017;Muldoon, 2018), or providing large-scale constraints 1D, 2D and 3D ice flow models (e.g. Leysinger Vieli et al., 2007Vieli et al., , 2011Passalacqua et al., C6 2018;Muldoon, 2018;Sutter et al., 2020)."  We agree, and based on reviewer 2's suggestions, we now show the figure without the Landmark panel view. Furthermore, we now display an OIA transect that cuts across the Concordia Subglacial Trench and therefore displays what is discussed later in the manuscript: that the deepest IRHs drop off due to basal process influence, in particular as we move away from LDC.