Dear Berhanu et al.


Thank you for your detailed answer to the referee comments and for the updated manuscript. After long considerations, I came to the conclusion that the manuscript does not yet fully solve the main points that have been raised during the referee process. May I therefore kindly ask you to rewrite your manuscript once again?

Overall, I think that the manuscript can be more quantitative and that important data and details from the supplement can be included in the text to tackle the concerns raised by the referee. Let me try to summarize the main issues that I still see with the revised manuscript:
• A more careful and quantitative analysis of the fraction of snow at the surface that is impacted by mixing with the surrounding snow. I think it was not so much questioned that you remove data points from the snow surface, but it was questioned whether this should be done at 2, 7 or 12 cm. A quantitative argumentation for example based on Fig. 5 looks like a good idea to me. Throughout the manuscript, it is still difficult to understand which threshold you used and why.
o Maybe use dashes lines in Figure 3 and 5 to indicate the sections of each data set individually, as opposed to the shaded area.
o Clarify the importance or use of the 7 cm threshold that is mentioned in the text. Or, omit stating or using it, when you decide that the individual thresholds are better.
o Was the snow that fluffy that it could mix with surrounding snow at the top few cm?
• Based on the referee comments, the choice of C0 is not straightforward. I have to admit that I also do not fully understand that choice. Could you give more details about the problem with the calibration? Still, all measurements of the same time (so, each number in the plots) were done with the same calibration? So, discussing UV#0-Control#0 vs. UV#1-Control#1 etc. should still be possible? Results and conclusions from that traditional analysis might add important information to the manuscript. I feel it is not enough to assume that there was no change in the amount of nitrate in that depth [of C0] (lines 255 ff). Can you justify this based on your data and some statistics?
• Figure 3: Could you indicate the photolytic zone for your snow? Could you indicate the times of the samples 0-6 in the caption (as noted in table 1 of the supplement). I further suggest showing all data in these figures and not just a selection.


Minor Things:
• Experimental Section: Could you add a description about the snow. Was it fluffy? Did it settle or harden during your experimental time; did it else change its optical appearance? Do you have any means to give estimates of the density (like from volume of molten sample vs. snow sample).
Can you estimate the photolytic zone for your snow? Or in other words, is it valid to compare to compare to light transmission measurements at Dome C (line 179 ff).
• The definition of isotope ratio line 68 and 72 do not match
• Line 296: “could still have been”, you observed such drifts, no need for conjungtive.
• Figure 10 and text 459 ff: I had difficulties relating the text and figure 10. Why, for example, is 0-7 cm excluded for all samples, when in the response to the referee you mention that for some samples 0-2 cm is excluded (page 7)? Also, I don not see “progressively more negative (data) over time” in Figure 10. Does this not look like a stable plateau?
• Could you comment in the conclusions, why you see a change in isotope ratio driven by photolysis, but no change in concentration.


I hope you find these comments helpful, please don't hesitate to contact me if any questions should arise.
best regards, Thorsten bartels-rausch