Dear Dr. Maselli,

I think you have addressed most of the referee comments. But some of them have not been fully addressed. I have a few comments for further revisions.

1. As Referee #2 (Ref2 hereafter) commented, using the MSA data to determine the sea ice component of the Br signal is tricky. The substantial negative values of the non-sea-ice Br (Fig. 8) worries me. Can you give any reasonable explanation for the negative values? The large scatter of the data shown in Fig. R6a,b does not look very convincing, either, even though the correlation may be statistically significant. When you did a linear fitting between the MSA and Br data, you used smoothed data only for MSA. Wouldn’t it be better if you use smoothed data also for Br?

2. Both Ref1 and Ref2 commented on biogenic sources of Br. I was also puzzled by the argument on biogenic sources of Br. To me, it is not very clear if the contribution of biogenic sources is important or not. Does the manuscript mean that the sea ice source is dominant in spring and biogenic sources are dominant in summer? Not having read the revised manuscript, I could not evaluate if the manuscript has been revised adequately. Please revise the manuscript carefully to make the logic clear.

3. The “ 3 step linear regression” is explained in author’s response to Ref1. Please explain it also in the revised manuscript.

4. Ref1 suggests that a more detailed consideration of potential impacts of acidity on the stability of Br and MSA in ice cores is needed, this has not been fully addressed. Though the authors argue that they analyzed the discrete samples obtained from the melt stream, this does not address the Ref1’s concern about a possibility that MSA might get destabilized at the melt head. Potential impacts of acidity variability within each ice core should be discussed, too, to fully address the Ref1 comment.

I have the following editorial comments.

1. In Section 2.1, Fig. 1 needs to be referred to.

2. The order of the figures or figure numbers should be re-organized. For example, Fig. 3 is referred to before Fig.2, and Fig.S3 before S1 and S2 (line 173) . Please check the figure numbers and the order that they are referred to.

3. The legend for the blue color in Fig.8a reads exPb (x2 after 1925). Does this mean “after 1940?”

4. Line 923. Delete one of the two “significant”s.

5. Figure 5. Pink circles and the fonts beside the color bar are too small to see. Same applies to some of the other figures and supplementary figures. In some of the figures, legends, axis labels etc. are too small to see. Please make them larger.

Dear Dr. Maselli,

Thank you for your revised manuscript. Please address the following Editor comments and revise the manuscript.

1. The statements on the MSA and Br temporal trends do not seem to agree with Table S1, which is confusing. For example, though Lines 185-186 reads “Ice core Br levels at each site were stable until ~1830”, Table S1 shows different inflection points for Summit and Tunu, neither of which is 1830. Another example is Lines 283-284, reading “ the decrease in both MSA and bromine at both sites in the early 1800s”. But Table 1 shows different inflection points for MSA and Br, and for Summit and Tunu. Inflection 1 for Summit MSA is 1854 and that for Tunu Br is 1842. I don’t think 1854 and 1842 can be called “early 1800s”. Please revise the sentences describing the trends of MSA and Br to avoid these confusions. Also, please display the inflection points and straight regression lines in Fig. 2.

2. The manuscript argues somewhat convincingly about MSA as a sea ice proxy, with the main argument based on a correlation between MSA and sea ice. But the correlation maps (Figures 6 and 7) are not very convincing. In black bordered regions both positive and negative correlations seem to be observed. But partly because the maps are too small, and colors for negative and positive correlations are hard to distinguish, it is hard to see in some regions if the correlation is positive or negative. I’m also confused by Fig. 6 because different black bordered regions are chosen for 1900-2010 and 1979-2010 without enough explanation. Please revise the correlation maps in Figs. 6 and 7 so that the correlations that the authors argue can be clearly seen, or maybe correlation maps can be removed.

3. Please add enrBr(Na) and nsiBr to Figs. 6 and 7 together with OWIP and MSA. This would strengthen the arguments of the manuscript, if enrBr(Na) and nsiBr have very different trend from that of MSA in the recent period. Currently they are in the supplementary, but without OWIP trend in the same figure.

4. Please show enrBr(Na) and nsiBr in the main text, not in the supplementary, because enrBr is a sea ice proxy published by previous studies, and to strengthen the conclusion
on the findings between enrBr/nsiBr and sea ice trends.

5. It would be better to add a few words to the final line of the manuscript "leaving room for the possibility that bromine may still be an effective proxy for local Antarctic sea ice conditions” “*and for preindustrial sea ice reconstructions*."

1. Section 2.1 and Fig. 3.
22 cm water eq and 12 months resolution, makes each month less than 2 cm. Is this really resolved all the way in the record at Summit (Fig 3)? How come there is a shift between the two time periods in Na, if Na is the main parameter used to make the annual layer counting?

2. To me, lines 532-535 (P. 22) seem to contradict with lines 540-542.
Please clarify the logic.

3. The caption of Fig. 6 (line 1073).
3-1. There seems to be no red curve in Fig.6. There are thin and thick
orange curves instead.
3-2. The legend for the thin orange curve reads "MSA no volc". Isn't
this "MSA, outliers removed"?

4. Same comments for Fig. 7 caption.

5. The last line of the Fig. S9 caption.
The longitude seems to be wrong. It should be same as that for (a).

I could not understand how the authors defined annual layer boundaries in the Summit core. If they defined the minima of S/Na as annual layer boundaries, it would have been affected by anthropogenic sulphate, which has changed the seasonal cycle of sulphate in Greenland. The manuscript did not convince me if the seasons were determined consistently across the pre-anthropogenic and anthropogenic eras. Though the authors seem to have used other parameters as well, it is not clearly written. Please revise the statement on ice core dating.