Comments on “Surface processes and drivers of the snow water stable isotopic composition at Dome C, East Antarctica – a multi-datasets and modelling analysis” by Ollivier et al.

General comments

The effects of post-depositional processes (water vapor exchanges between surface snow and atmosphere, diffusion of water vapor in the firn, and wind redistribution) on water stable isotopic composition in surface snow in the inland of polar ice sheets strongly limit the quantitative paleo-temperature reconstructions by ice-core stable isotopic records from these regions. In this study, the authors attempted to evaluate the mean effect of post-depositional processes at Dome C using a combination of existing and new dataset of precipitation, surface snow and subsurface isotopic compositions over 2017-2021. It is noted that the sampling schemes are elaborate and the sampling temporal resolution is high, which is not easy to perform at very harsh meteorological conditions such as at Dome C but is very necessary to address the problem of post-depositional processes. I appreciate the authors for their hard work at Dome C and their persistent efforts on this complex issue on the post-depositional processes. Although I agree with the authors for their most interpretations of the observations, I think some explanations should be added and the different post-depositional processes should be clarified.

Specific comments

I note that precipitation, surface snow and subsurface snow show similar δ¹⁸O seasonal cycle, but the maximum (minimum) value of δ¹⁸O occurs in different month, with highest δ¹⁸O value in January and lowest value in May for precipitation, a maximum of δ¹⁸O in February and a minimum in October for surface snow, and a maximum of δ¹⁸O in March and a minimum in November for subsurface snow. I think it must be due to the post-depositional processes, but why the post-depositional processes can lead to the δ¹⁸O shift in time?

In section 3.2.2, the authors find that the surface snow is relatively enriched in δ¹⁸O during warm season (from November to April) due to the sublimation of surface snow. Indeed, the calculated surface moisture fluxes also indicate sublimation occurs mainly in summer (Fig. 2). Theoretically, stronger sublimation should cause more enrichment of δ¹⁸O in surface snow. As a result, I would like to know if there exists a correlation between the surface snow δ¹⁸O and the water vapor flux during the summer. Additionally, the slope between δD and δ¹⁸O in surface snow should be lower than the δD- δ¹⁸O slope of precipitation. I suggest the authors to add above relevant information to further validate the sublimation process because it is a key post-depositional process in inland of Antarctica.

In section 3.3.2, the authors compared precipitation isotopic composition (δ¹⁸O and d-excess) with the simulations of ECHAM6-wiso. Although the daily δ¹⁸O modelled by ECHAM6-wiso shows a good agreement with the observations, the ECHAM6 overestimates (underestimates) the δ¹⁸O (d-excess) likely due to the warm bias of the model. In addition, the daily d-excess is poorly simulated by the ECHAM6-wiso (Fig. 7b). The possible deficiency of the ECHAM6-wiso for the mismatch between the simulations and observations should be brief explained for guiding scientists to better utilize GCMs-wiso.

The authors conclude that the mean effect of post-depositional processes enriches surface snow δ¹⁸O by 3.3‰ to 6.6‰ and lowers the snow surface d-excess by 3.5‰ to 7.6‰. As the authors analyzed, summertime sublimation should contribute significantly to the mean effect. After comparing Fig. 3b and Fig. 6c, I found the δ¹⁸O values of surface snow during wintertime are also higher than the values of precipitation. This suggests that the post-depositional processes rather than sublimation are also significant in winter. So, what kind of post-depositional processes lead to the enrichment of surface snow δ¹⁸O in winter? Is it the diffusion of water vapor? The authors should discuss the post-depositional processes in wintertime.