Review of « A model framework on atmosphere-snow water vapor exchange and the associated isotope effects at Dome Argus, Antarctica：part I the diurnal changes» by Ma and others.

This manuscript considers the exchange between water molecules between the firn and the atmosphere, and the impact it can induce on the change of isotopic composition in extremely low accumulation regions of Antarctica. Using the results from Dome C as an analogue for Dome A is a clever strategy that can yield promising results to how to explain the impact of surface processes on the future Dome A ice core. The study takes into account the variations of stability of the atmosphere with systematic calculations of the Richardson number and developed three case studies associated with two sets of summer conditions (clear sky and cloud), and one set of winter conditions.

While the authors used a rather classical set of equations to evaluate the isotopic exchanges during sublimation and condensation, it seems not pertinent here, as it ignores major contributors to the boundary layer processes and only consider the system as a closed box without exchange with the free atmosphere. As a result, the results do not match the observations that were made for the surface snow isotopic composition at Dome C, even though, it is supposed to be the case study used to parametrise the model.

I suggest profound modifications to the model, which take into account exchanges between the atmospheric boundary layer and the free atmosphere on top of the surface processes, and which would match the surface snow changes, at least in order of magnitude, before considering the manuscript for publication.

Major Comments:

1. The box model developed by the authors was parametrised against vapour measurements obtained at Dome C, in order to compensate for lack of measurements at Dome A. The outputs of the model predict changes of vapour isotopic composition that seem realistic, but it is not the case for the changes of snow isotopic composition for which the variations are extremely small (less than 0.02‰) while the observed changes are around 2‰ during a typical night (Casado et al., 2018). The relative changes of snow and vapour isotopic compositions during a typical clear sky night were modelled in this manuscript, and suggested that a closed box model (which is de facto what the authors have implemented since no exchanges between the free atmosphere and the boundary layer are taken into account) is not realistic for this type of event.

1. Another aspect that suggests that exchanges between the boundary layer and the free atmosphere must happen is the Richardson number. Indeed, for negative Richardson numbers, the atmosphere must be quite convective, which suggest that the boundary layer exchanges with both the surface snow and the free atmosphere.

1. The atmosphere is qualified as stable for any positive Richardson number, yet, it seems that some studies suggest that some amount of mixing remains quite strong for 0 < Ri < 0.1 (Zilitinkevich et al., 2008) . This could be discussed.

1. Some limited vapour data exist at Dome A (Liu et al., 2022). While these data might be difficult to compare to your results, in particular consider how high the d-excess is, which could be associated with calibration issues, it should be discussed.

Considering how fundamental these changes are, an updated version of the manuscript could have completely different conclusions.

Bibliography

Casado, M., Landais, A., Picard, G., Münch, T., Laepple, T., Stenni, B., Dreossi, G., Ekaykin, A., Arnaud, L., Genthon, C., Touzeau, A., Masson-Delmotte, V., & Jouzel, J. (2018). Archival processes of the water stable isotope signal in East Antarctic ice cores. The Cryosphere, 12(5), 1745–1766. https://doi.org/10.5194/tc-12-1745-2018

Liu, J., Du, Z., Zhang, D., & Wang, S. (2022). Diagnoses of Antarctic Inland Water Cycle Regime: Perspectives From Atmospheric Water Vapor Isotope Observations Along the Transect From Zhongshan Station to Dome A   . In Frontiers in Earth Science   (Vol. 10). https://www.frontiersin.org/articles/10.3389/feart.2022.823515

Zilitinkevich, S. S., Elperin, T., Kleeorin, N., Rogachevskii, I., Esau, I., Mauritsen, T., & Miles, M. W. (2008). Turbulence energetics in stably stratified geophysical flows: Strong and weak mixing regimes. Quarterly Journal of the Royal Meteorological Society, 134(633), 793–799. https://doi.org/10.1002/qj.264

This manuscript describes a closed box model assuming no atmospheric mixing and simulations of the effect of a mean diurnal cycle at Dome C (using observations) and Dome A (using atmospheric reanalyses and assumptions as inputs). The current title does not reflect the content and the conclusions are not well supported by the analyses and the underlying assumptions in the modelling methodology.

The long introduction gives a good scene setting for the study, which addresses an important topic, but fails to describe the modelling framework in the context of other studies, and fails to provide a clear comparison of the meteorological and snow conditions between Dome C and Dome A (and what are the similarities and differences that need to be accounted for in comparing results for these two sites, for diurnal variations, clear and cloud sky, and winter vs summer conditions).

The description of the model has flaws in the equations for latent heat flux and possibly in the use of relative humidity in the atmosphere and not relative to surface temperature for fractionation coefficients. The information provided in supplementary information is very difficult to understand.

The choice of performing simulations driven by a mean diurnal cycle instead of using the actual wealth of observations is unclear and the implications should be discussed. I am puzzled by how wind effects are accounted for when averaging conditions.

There should be at least a more detailed comparison between the Dome C and Dome A characteristics (including comparison of meteorological conditions and ERA5 results at both sites), instead of current Table 1 (where assumptions versus observational based information should be differenciated).

The assumptions displayed in Figure 1 should be discussed in the context of available information, including the Richardson number, regarding atmospheric exchanges (the closed box assumption validity).

The authors should reflect on what their model explicitely implies in terms of behaviour, and what is effectively "validated" from their approach which does not resolve the diurnal variations in snow measured at Dome C. This physics-based approach is missing.

For these reasons, major revisions are needed, first to ensure accurate equations in the model, and then to reflect on the limitations and suitability of the core assumptions of the closed box model to adress these questions, and third regarding the average diurnal cycle approach, and fourth regarding the detailed comparison between Dome C and Dome A (well beyond "validating" and "applying" this model at the two sites).