General comments:

Overall this is a well-written paper, with an interesting modelling analysis of a last-interglacial reef setting in S-Madagascar. The idea to test the effect of different GIA models on reef architecture is nice, and most parametric choices are well justified. Generally I don’t have many important remarks to make (mostly small ones, see below), but the main thing that felt a bit underwhelming was the last part of the discussion.

Firstly, I understand you cannot make a one on one comparison, but I do think you can get much more out of the comparison with the observations in Figure 10. In the ‘model caveats’ section you emphasize how the burial of MIS 5e deposits by aeolianites would have probably limited subaerial weathering quite a lot. Following that logic, I think it would be very interesting to compare the observed facies elevations with those at the end of MIS 5e (before a lot of material started to be eroded at the surface), not only with the Full scenario, but only at the other two. Another thing to consider is an evaluation of the facies not only at Well C, but perhaps also further land- and seawards. For instance, it would be interesting to see at which (range of) elevations, and in which part of the transects, we would see an MIS 5e transition from reef deposits to beach deposits, similar to the observed reef-flat to intertidal transition in Fig. 10. Again, this could be done both for the end of MIS 5e (before weathering kicks in) and the present-day. I think this way you can have a much better discussion of the different scenarios, and hopefully gain more/better arguments to prefer one GIA scenario over the other.

Secondly, the entire discussion section is centered around the case study in Madagascar, but it is not so clear in the discussion what we really learned from this exercise in terms of broader implications, how it is applicable to other areas, what can be done to improve the analysis etc. I think you can add a bit of depth in the discussion by addressing (some of) those questions. The conclusion section is more convincing in that respect, but it is surprising to read that ‘there is a compelling need for stability within the LIG’ without really discussing this in the discussion section. Again, a more thorough model-observation comparison, as suggested in the paragraph above, could help with this.

Specific comments:

Line 24: I would put a larger range here than 7-8 m, which I think is a little misleading. Actually the Dyer et al. 2021 paper that you cite here makes for a good argument to put the “consensus” on MIS 5e eustatic sea-level peak a little broader.

Line 43: What does high-to-low swing mean here? Shouldn’t it be a high-to-low-to-high swing if there’s a double peak?

Line 57: This sentence is a little confusing, since you start talking about the modern reef system and then about subsequent cycles starting at 400 ka. I think I know what you mean (starting with a 400 ka fringing reef based on the modern reef system), but better clarify.

Line 76: derived? (not derive)

Line 76: please specify how high is high-resolution?

Line 78: has? (not have)

Line 113: is there wave erosion in the model as well? If yes, in what sense? If no, would this affect the outcomes?

Line 140: what are all these numbers based on? If it’s the same studies as for the growth curves it could be specified.

Line 147: You give the Malatesta example of (wave) erosion rate, but I guess that’s not the same as the four weathering processes you mention here? If so it’s a little misleading to cite Malatesta here.

Line 147: The first weathering is sub-aerial weathering? Then maybe specify that

Line 156: Subaerial erosion rate is the same as the (1) maximum weathering rate? Please clarify

Line 156: Not sure if I understand the transformation rate. Does it both transfer (transfer of mass from one part of your model to the other) and transform (change carbonate into carbonate sands and muds)? Which part of the sequence is transformed, and if it’s transported, to where? Please clarify

Line 164: I would change tectonic uplift to tectonics, since tectonics can both result in uplift and subsidence. And what do you mean with subsidence? I would either say “anthropogenic subsidence” (to emphasize this is not tectonic subsidence) or “other vertical land motion processes” (more generic, includes anthropogenic subsidence).

Line 181: Is there a scientific reason why the GrIS would be melting before the AIS, other than that it is a convenient way to create two peaks? I’m no expert, but as I understand from Govin et al. (2011) and Stone et al. (2016) it could be an explanation for different temperatures in Antarctica and Greenland during the early interglacial. Maybe mention this somewhere?

Line 296: I don’t think you can really say “a 295% increase in RSL”, I would rather use the amount in m. For example: when talking melt scenarios the difference between 1 and 5 m RSL is for me about as important as between 2 and 6 m RSL, yet one would be 500% and the other 300%.

Line 299: I think it should be “under the” instead of “under” (twice)

Line 306: Would be interesting to comment why the LIG carbonate package is so much thinner for the G2A5 scenario, for me the reason is not directly obvious

Line 307: Good to mention the amount of carbonate sedimentation as well (in m I suppose?), at least for the Baseline, just to know what kind of amounts we’re talking about.

Line 310: This is a little confusing, the previous sentence suggests you’ll talk about Well C, but Fig. 8a is about Well A. In this sentence and the next, do you mean (Fig. 8c-left), (Fig. 8c-right) and (Fig. 8c-middle) instead of Fig. 8a, Fig. 8c and Fig. 8b? Also, in this sentence you mention the basement, but I don’t see the basement in Fig. 8c? Would be nice to add actually.

Line 318: Maybe add the duration of the peak for the Full scenario as well.

Line 324: “each scenarios” should be “each scenario”

Line 324: maybe change “shore-ward” to “the area shore-ward” for clarity

Line 327: maybe elaborate what you mean with previous investigations, something like “unlike previous investigations with the same model elsewhere” for clarity

Line 329: I think the word ‘significant’ is very vague here, maybe replace it with something more concrete

Line 332: why only subsidence, and not uplift?

Line 332: I don’t understand this last sentence, why ‘furthermore’?

Line 344: Interesting conclusion, but I’m a bit surprised that this is not mentioned in the discussion section really, where I think you can spend a few sentences that your results (seem to) imply LIG stability?

Fig. 1: It would be nice to have the colour code of the bathymetry included

Fig. 2: I think Figure 1 could be included as a location inset within this figure, as there is less relevant information in the first figure but it’s much bigger than figure 2.

Fig. 6: It is very difficult to see the differences between the three models. Given the huge amount of white space in each figure, I think you can find a way to increase visibility of the reef architecture (De Gelder et al. 2022 Fig. 4 might give some inspiration, no need to cite that of course, and apologies for the self-promotion)

Fig 8: I think it would be nice to have unconformities clearly marked in the lithology as well, both as a visual aid and to help with the results section. For instance in Fig. 8a, it would be nice to see a separation in the lithology for the Baseline model, where you jump from an ~80 ka forereef to a Holocene forereef. Another thing for this figure is that the lithology weirdly stops at the bottom for some cases (white area), I think this should be an easy cosmetic fix. Is it supposed to be basement?

Fig. 9: It is a little difficult to understand the aesthetics of the synthetic wells. Well c in figure c, does it continue below the water surface with a different color? Why is well A shorter in panel c?

Fig. 10: As a colorblind, I struggled a bit linking the legend to the stratigraphy, mostly because the intensity of the colour is very different between the two. Please change the legend to be consistent with the figure.

Another, more important thing about this figure is that I think it would be worthwile to add the stratigraphy at Well C towards the end of MIS 5e, i.e. before a (large?) portion of the stratigraphy was removed by erosion. Was there any reef flat or beach deposits above the reef crest deposits? That would make for an interesting comparison with the data. Also, maybe there were some beach deposits in the Baseline or G2A5 scenarios at the end of MIS 5e, which would be worth comparing with the observed stratigraphy?

References:

de Gelder, G., Husson, L., Pastier, A. M., Fernández-Blanco, D., Pico, T., Chauveau, D., ... & Pedoja, K. (2022). High interstadial sea levels over the past 420ka from the Huon Peninsula, Papua New Guinea. Communications Earth & Environment, 3(1), 256.

Govin, A., Braconnot, P., Capron, É., Cortijo, E., Duplessy, J. C., Jansen, E., ... & Waelbroeck, C. (2012). Persistent influence of ice sheet melting on high northern latitude climate during the early Last Interglacial. Climate of the Past, 8(2), 483-507.

Stone, E. J., Capron, E., Lunt, D. J., Payne, A. J., Singarayer, J. S., Valdes, P. J., & Wolff, E. W. (2016). Impact of meltwater on high-latitude early Last Interglacial climate. Climate of the Past, 12(9), 1919-1932.

This manuscript documents a test case of forward stratigraphic modelling for the purposes of validating paleo sea-level scenarios (Last IntergIacial considered here). This area of research is developing and sharing test cases will help the community to design approaches and best practise. This type of work is incredibly important for the community.

The manuscript was well structured and easy to read - given the dense amount of model output discussed- and the figures are very useful and clearly presented. In particular, I applaud the authors on the level of transparency and detail provided to the reader on model caveats, where the authors have clearly identified what they were unable to constrain and what the potential consequences on the results were – this is further referenced throughout the manuscript.

General comments:

While I found the methods well explained, I think further reasoning of parameter selection could be added to section 2.7 and 2.8 (detailed below).

A more clear introduction to DIONISOS, exactly what it does, and referenced uses would be welcome.

Unit notation: mMyr^-1 should be m.Myr^-1 or m Myr^-1 and consistent units e.g. 10 kyr rather than 10,000 years.

Acronyms: some provided and not used, in particular I think it would be improve readability if FSM is spelled out every time as it’s used sparsely in the manuscript, but is very important and I forgot what it was when I first saw it.  

Data request: a simple table in appendix documenting the parameters used would be preferred for FAIR principles so that the study could be replicated. Acknowledging that the parameters are included through the text.

Technical comments:

Line 5: preference for formal language “sequences, which also provide…”

Line 12/13: reword for clarification “DIONISOS forward stratigraphic model (from Beicip Franlab)…”

Line 23: define eustatic – use eustatic, GMSL and RSL and these terms are often used interchangeably by the community.

Line 41: change sea-Level to sea-level

Line 41: confusing wording, suggest “sea-level patterns, for example Kopp et al…”

Line 43: please clarify further what ‘ice proxies’ are

Line 59: A brief explanation why Madagascar was chosen would be useful here, particularly as there is very little data available for the site. Is it selected primarily because it is considered stable?

Line 63: Add ref for DIONISOS and spell out acronym here.

Line 76: spell out DEM

Line 77: remove acronym SFM (not used again)

Line 94: reword for clarification “seven sediment contributors, that includes five coral…”

Line 121: ref for CAWRC and/or Wave Watch

Line 121: Appendix table including the exact sites (lon/lats) used

Line 131: repeats from Line 92 – although written more clearly here. Remove from one location.

Line 133: remove a “…assignment of a a ...”

Line 133: I’m unclear what production versus depth curves are, sediment deposition/erosion? Please clarify.

Line 134: clarify “provided production curve…”

Line 135: ref for model creators of production curves

Line 141-142: how do these parameters compare to Montaggioni? Are they taken from Montaggioni or different and if different, what was the reasoning. I’m unclear on why these values were used.

Line 142: clarify “throughout the duration of the time period”

Line 154: repeats “Therefore, we utilize a maximum weathering of 100 m.Myr-1.” On line 155: “we use a maximum 100 m Ma-1 weathering rate” and contains different units.

Line 154: Reasoning for weathering rates. While I feel the maximum weathering rate is acceptable  given the comparison and evidence Montaggioni and Trudgill, I think this could be explained more clearly. Was there a ref for mechanical and bio erosion under marine conditions?

Line 157: Restate to make clear exactly what is referred to, there are a lot of rates and curves, these need to be explicitly stated. Here “the transformation rate is treated as a constant…” And Line 141: “Our simulations are run with a maximum x for coral facies” – add maximum growth rates. Check elsewhere.

Line 157: Is the 50 m Myr-1 rate a standard parameter or does it need to be supplied, if so why was it selected?

Line 162: for additional understanding by the reader, I suggest you add something akin to “down slope, determined by the slope steepness (ref)”.

Line 162: add reference to Section 2.8 with “user defined curve”. While it’s the next paragraph, the sentence as it stands makes you expect a description of the RSL curve, which is not forthcoming, and that makes you question whether in fact it is provided by the user or not.

Section 2.8 – add references to support the sea-level scenarios. It is mentioned later in discussion, that these scenarios are preferred by the community but not here when you first introduce them.

Line 174: add to clarify that the model runs over the last 410 kyr, with a (global?) sea-level curve used for baseline, and the two altered scenarios only change sea-level from (130-115ka?). That you run to present to represent all post depositional processes so that the synthetic wells would be comparable to modern observations is very understated here! And a key requirement for this approach.

Section 2.9 – great!

Section 3.0 – incredibly well described, a pleasure to read. 

Line 227: strange wording “upper 7.5 m of LIG limestone to have eroded away” reword.

Line 288: Here you state commonly cited scenarios of ice sheet melt – but they are not cited.

Line 301: space between the & LIG

Line 306: clarification “…two melting scenarios (Full and G2A5)…”

Line 321: I do not understand what this referring to “actual exposed MIS 5e facies are lacking” – do you mean in the synthetic logs? If so, reword. The terms ‘actual’ and ‘exposed’ made me initially think of real exposed outcrop. If you are referring to real outcrop then reference.

Line 332-333 : is a general simplification of facies formed and classified by the model an issue here too?

Line 344: I think this is somewhat overstated for the conclusions given the short paragraph discussing this at the end of the discussion. It should either be heavily edited (as below) or removed from the conclusions to discussion, or requires more development of this concept in the discussion to support this conclusion. From my understanding “We further conclude that a stable LIG sea-level signal is most consistent with the synthetic stratigraphy presented here, specifically because no reef facies are preserved under the double-peak sea-level signal.”  As the statement stands, you are taking one sites, for which you have limited observations and interpreting this as representative for global sea-level (which you have not previously stated). In general, I think more discussion is required regarding Fig. 10 and the facies difference and elevation differences, in fact the caption contains more pertinent information (Boydens facies interpretation of peak sea-level) that is not contained in the main text. When you call it roughly anagolous to the observation, this is not explained what the similarities ad differences are.  

Lastly, I would consider including some broader implications for this type of study. Do you consider that this model is easily applied to other areas, is the baseline study and lack of GIA at this site critical? Are the differences in the resulting synthetic stratigraphy significant enough to be able to conclusively determine the LIG melt history (double peak or not). Are the different locations of LIG sea-level expected to be so different this will need to be applied to each? What are the intended next steps/ areas that need further development?