|My major concerns:|
A- Most of the work presented here is not new.
- All the water isotopic data have been published in the paper by Stenni et al., the cryosphere, 2016. This should be presented as such. As a consequence, there is no reason why the acquisition of the data should be detailed except if additional measurements are presented here. If this is the case, it should be clearly stated.
- The model approach has already been applied at Dome F by Dittmann et al. (2016). So that this paper shares many similarities with this recently published paper. The site is still not the same. Since the two sites are on the East Antarctic plateau but under different influences, a nice added value to the paper would have been to develop a comparison between the two sites both on meteorological and isotopic aspects.
- The weather dynamic at Dome C with warm intrusions of air has already been discussed in the recent paper by Schlosser et al. (2016) concentrating on the year 2009 and 2010 without isotopic data. The only new feature presented here is the evidence of a moisture transport from West Antarctica across the continent and the relation to the Amundsen-Bellingshausen Sea Low. This part is only discussed within 15 lines (section 4e). If this is the major new result, it should be discussed much more.
B- The discussion on hoar isotopic composition (stated as a prominent new result in the conclusion and in contradiction with Stenni et al., 2016) is not clear. Only a few sentences are given in p. 15: « Figure 8a shows observed d18O vs. 2m air temperature for the different types of precipitation: snow, diamond dust, and hoar frost. High precipitation events, for which trajectories were calculated, are marked with circles. The regression lines differ only slightly for the various precipitation types. For all samples, a d18O/T slope of 0.49‰/°C is found (r=0.79, n=498). The slope for the studied high precipitation events only is 0.39‰/°C, lower than for all days (r=0.78, n=21). Also, the relationship between deuterium excess and d18O (Fig 8b) shows no significant differences between the precipitation types.” This statement is not different from the finding by Stenni et al. (2016) who also display the same d18O vs Temperature slope for the different precipitation types (Table 4, hence similar to Figure 8 of this study). In Stenni et al., 2016 as in this manuscript, it is already mentioned that hoar mainly occurs during winter and that this may be the cause of the low d18O and high d-excess. The other possibility proposed in Stenni is the fact that hoar is linked to low level water vapor isotopic composition and that one needs to explore this aspect. The study presented here does not explore this aspect because it better looks at high level (500 hPa, 600 hPa) trajectory. It does not bring any information on low level water vapor isotopic composition so that it cannot conclude anything on the post-deposition influence on hoar isotopic composition. It is not possible to really conclude here on the question raised by Stenni et al., 2016 without knowing what happens in the low level atmospheric layer.
Finally, I find it disturbing that the authors claim that they have different conclusions than in Stenni et al., 2016. Indeed, the author list is quite similar between this study and the Stenni very recent paper.
C- The discussion on inversion vs condensation temperature used for calculating the d18O of precipitation in polar regions has already be explored by Ekaykin and others (see for example: http://eprints2008.lib.hokudai.ac.jp/dspace/bitstream/2115/45456/1/LTS68suppl_022.pdf, bottom of p. 301 and reference to phD thesis with temperature profiles above Vostok). These authors also provide some interesting conclusions that should be confronted to the results given here.
D- Finally, the strong effect of local temperature in d-excess in East Antarctica has already been discussed by others, i.e. Uemura et al., Climate of the Past, 2012. It is a quite well known effect of distillation that decreases the slope (dδD/dδ18O) towards low temperature.
My conclusion is that this study does not present sufficient new results / new approaches for a full publication. If this study can be complemented by a meaningful site comparison (Dome F vs Dome C for example), it may bring an important added value that is worth a publication.