This manuscript uses and updated version of the IMAU Firn Densification model to assess 21^st century changes in firn air content (FAC) in Antarctica. New aspects of the work include the inclusion of a dynamic densification law that accounts for changing climate, and the calculation of “accessible FAC” that accounts for possibility that ice slab formation may limit meltwater storage in deeper pore space. The authors conclude that ice shelves in the Antarctic Peninsula Dronning Maud Land are particularly vulnerable to FAC depletion and that for low-accumulation rate shelves, ice slab formation has a particularly deleterious effect on meltwater storage capacity that has not been previously well-captured.

This paper presents important new advances, both in terms of the firn modeling methods and the study of ice slab impacts on ice shelf hydrology. It particularly provides a valuable assessment of how firn meltwater storage will evolve in the 21^st century that comes closer than past studies to being grounded in our current understanding refreezing processes in firn. Overall, the work seems to be technically correct and well-reasoned. However, the paper can be a fit hard to follow at times due to the huge amount of work and information that went into the results. The authors could consider streamlining the main text and moving some of the model development work to the supplementary text.

Major Comments:

[1] I found the organization of Section 2-3 somewhat hard to follow, I think in part because the classic methods-results-discussion format falls short when trying to present what is the equivalent of almost two papers worth of work. You might consider first presenting a streamlined discussion of the model updates, calibration, and performance as one section. The goal of this section would be that the reader finishes it convinced that the FDM v1.2A-C run is a reasonable estimate of the future firn evolution. Right now, this point gets somewhat lost because the text is constantly bouncing back and forth between different experiments. Once this is established, then you can introduce how you will use FDM v1.2AD-C to study FAC across Antarctica. From there, it flows more naturally into the results that really only focus on FDM v1.2AD-C simulations. My suggestion for organization would be something like:

Section 2 – IMAU FDM Model Updates

2.1 – Densification expression

2.2 – Calibration Experiments and Atmospheric Forcing (combined since the experiments are just based on the different forcings)

2.3 – Calibration and Validation Data (aka firn cores)

2.4 – Calibration and Performance of FDM1.2AD-E (demonstrate that dynamic densification is reasonable using historical period)

2.5 – Performance of FDM1.2AD-C (demonstrate that future projections are reasonable)

Section 3 – Experimental setup for the future firn evolution and calculation of accessible firn air content

Section 4 – as it is now

I think that much of the current Sections 2&3 can also be streamlined to focus just on the key information needed to explain to the reader how the dynamical densification is achieved, and that the new model produces believable results for the rest of the 21st century. Maybe some of the detailed discussion of calibration results and extra figures for intermediate steps like FDM1.2AD-E could be moved to the supplement.

Minor Comments:

Title: Perhaps should mention ice shelves, since the paper seems to almost entirely focused on discussing FAC change on ice shelves?

Line 11: Choose to describe the future climate scenarios either in terms of mitigation or in terms of emissions, but not both. It is quite confusing to keep track of whether “strong” means “strong emissions” under SSP8.5 or “strong mitigation” under SSP2.6.

Line 159: Not clear why this is done? Maybe this would be better discussed in the experimental setup section?

Line 173 – 183: This paragraph did not seem particularly enlightening. It just reads like a long list of facts with not much explanation of why they are important, and certainly by the time it becomes relevant in the result, the readers will have forgotten all of this. If any of these statistics are important to the results, I would suggest either bringing them up at that point, or moving to a table and offering a super brief summary of the key points (e.g. accumulation, surface melt, and temperature all increase in the future over all parts of the AIS, with ice shelves seeing the largest increases).

Figure 1: Something to consider that might help readers visual the different experiments would be label on this plot which experiments map to which time periods with background shading or something. The multiple vertical axes are also confusing. Perhaps just break this up into two plots – one with temperature and one with melt and accumulation.

Line 198: Machguth et al. (2016) would be another appropriate citation here.

Section 2.5: A few additional suggestions that can help to bound the range of ice slab permeabilities that you use:

[1] Ashmore, D. W., Mair, D. W. F., & Burgess, D. O. (2019). Meltwater percolation, impermeable layer formation and runoff buffering on Devon Ice Cap, Canada. Journal of Glaciology, 66(255), Article 255. https://doi.org/10.1017/jog.2019.80

Modeling studied which showed that an impermeability threshold of 1 m for ice slabs led to the best fit between modeled and measured SMB on the Devon Ice Cap.

[2] Charalampidis, C., Van As, D., Colgan, W. T., Fausto, R. S., Macferrin, M., & Machguth, H. (2016). Thermal tracing of retained meltwater in the lower accumulation area of the Southwestern Greenland ice sheet. Annals of Glaciology, 57(72), Article 72. https://doi.org/10.1017/aog.2016.2

Used thermistor measurements to show that no percolation occurred through a 5.5 m thick ice slab at KAN-U even during the summer of 2012.

Note that using the numbers from Culberg et al. (2021) when looking at individual ice slabs is a bit complicated, because what they really show is that a package of many ice lenses that is 1-2 m thick can inhibit percolation to some degree. I think it’s okay to use since the numbers are consistent with other papers and a 1-2m layer thickness or impermeability is a conservative take-away, but just good to note that their numbers are not totally comparable to some of these other studies on ice slab thickness.

Line 232: It’s not clear where the > 900 kgm^-3 threshold is coming from. Machguth et al. (2016) gives an ice slab density of 873  25 kgm^-3 in their supplement and Rennermalm et al. (2021) gives a density of 862 kgm^-3 (Machguth et al., 2016; Rennermalm et al., 2021).  

Line 273: I am confused about this statement about “higher FAC”. It seems like the preceding sentence says that FAC decreases on average?

Figure 6: I found it very hard to pick out the differences between the top and bottom row. Perhaps a third row with difference plots between rows 1 and 2 would be valuable.

Lines 302 – 305: how do these thresholds compare to what we know about the climatic conditions for firn aquifers and ice slabs from either Greenland (MacFerrin et al., 2019; Munneke et al., 2014) or the Antarctica Peninsula (Van Wessem et al., 2021)?

Figure 7c: Is the “difference between accessible firn air content and total firn air content” just a subtraction or is it a ratio? It is confusing to me why this would be positive for high melt, moderate accumulation ice shelves where my understanding is that accessible FAC would be lower that total FAC.

Lines 328 – 337: How is runoff defined and calculated in IMAU-FDM? I think this needs to be clarified for the reader. I am not fully convinced that the runoff extent or values are particularly meaningful given the caveat that ice slabs remain permeable in the model. The spatial extent of low accessible FAC seems like a far more valuable metric. How consistent is the spatial extent of low accessible FAC with the spatial extent of runoff as calculated by the model?

Line 338: This seems surprising for an area that is currently a firn aquifer and is projected to see increased accumulation. Is this runoff coming from FAC depletion or from the bottom of an aquifer?

Line 353: Please explain how you define an “extreme melt season”. Consider marking these seasons in some way on Figure 9.

Line 414: Seems appropriate to at least mention something about firn aquifers, though I understand that this is not the focus of this study.

References:

MacFerrin, M. J., Machguth, H., van As, D., Charalampidis, C., Stevens, C. M., Heilig, A., Vandecrux, B., Langen, P. L., Mottram, R. H., Fettweis, X., van den Broeke, M. R., Pfeffer, W. T., Moussavi, M., & Abdalati, W. (2019). Rapid expansion of Greenland’s low-permeability ice slabs. Nature, 573(7774), Article 7774. https://doi.org/10.1038/s41586-019-1550-3

Machguth, H., Macferrin, M., Van As, D., Box, J. E., Charalampidis, C., Colgan, W., Fausto, R. S., Meijer, H. A. J., Mosley-Thompson, E., & Van De Wal, R. S. W. (2016). Greenland meltwater storage in firn limited by near-surface ice formation. Nature Climate Change, 6(4), Article 4. https://doi.org/10.1038/nclimate2899

Munneke, P. K., M. Ligtenberg, S. R., van den Broeke, M. R., van Angelen, J. H., & Forster, R. R. (2014). Explaining the presence of perennial liquid water bodies in the firn of the Greenland Ice Sheet. Geophysical Research Letters, 41(2), 476–483. https://doi.org/10.1002/2013GL058389

Rennermalm, Å. K., Hock, R., Covi, F., Xiao, J., Corti, G., Kingslake, J., Leidman, S. Z., Miège, C., Macferrin, M., Machguth, H., Osterberg, E., Kameda, T., & McConnell, J. R. (2021). Shallow firn cores 1989–2019 in southwest Greenland’s percolation zone reveal decreasing density and ice layer thickness after 2012. Journal of Glaciology, 1–12. https://doi.org/10.1017/jog.2021.102

Van Wessem, J. M., Steger, C. R., Wever, N., & Van Den Broeke, M. R. (2021). An exploratory modelling study of perennial firn aquifers in the Antarctic Peninsula for the period 1979-2016. Cryosphere, 15(2), Article 2. https://doi.org/10.5194/tc-15-695-2021