Dear authors,

many thanks for responding to the comments raised by the referees. As you can see from their comments, the referees are in general in favour of the pCO2 analysis but they also raise a couple of major issues which need to be addressed. From my reading I believe, that most of the concerns can be fixed by adding more information and the way forward should be clear from the well formulated referee comments. There is however one more critical comment that I would like to address here in my decision letter:

Both referees find the discussion on the role of total alkalinity insufficient. Referee #1 e.g. remake that "Nowhere in the manuscript have the authors presented ALK data that would support the analysis" and referee #2 adds that "My main concern with this paper is with the discussion on total alkalinity, which I dont find satisfactory". While the authors acknowledge this, I am not satisfied with their response, which reads:

"Both Ref#1 and Ref#2 call for information on TCO2 and alkalinity to support the analysis. The discrete sea water sample pCO2 data we present generally include TCO2 results. ALK may thus be calculated. The call for information on TCO2 and ALK brings a complex temporal and biogeochemical variability in the Polar Water into focus. We intend to present another paper on that subject which we find too extensive to be added to the present paper. However, the referee suggestions may perhaps be met by adding summary information in figures of a kind similar to Fig. 6a and expand accordingly the Results and Discussion sections?"

I understand that the research topic calls for a second study, however, the readers of this study cannot be left "waiting" for this follow up study (the situation would be different if the alkalinity study was available and could be referenced). Therefore the authors need to provide sufficient evidence in this study (either in the text or supplement), supporting the observed connection between pCO2 and alkalinity for the readers of this study. I also don´t see that adding this information would compromise a follow up study that has a deeper focus on the mechanisms involved. Expanding the results and discussion section (as suggested by the authors in their response) is certainly a good idea, nevertheless, I wanted to stress here that sufficient evidence needs to be included in this study. Therefore, I have decided that major revisions are required before the study can be considered for publication

Best regards
Peter Landschützer

Dear authors,

many thanks for submitting a revised manuscript. I have asked the same referees again for their view on the revision and I much agree with their assessment that the manuscript has improved and is close to publication. Both referees have suggested a few additional minor comments that I have also copied here at the end of my letter. In this respect, I would like you to consider these last technical/minor comments and resubmit a final manuscript, which I intend to review myself.

Best regards
Peter Landschützer

referee#1

Review of the revised manuscript submitted to Biogeosciences, MS No.: bg-2020-313. “Enhancement of the North Atlantic CO2 sink by Arctic Waters” by Olafsson et al.,

Review invitation: 2/1/21
Review accepted: 2/1/21
Review sent: 3/1/21
General comment:
Authors attempted to reply to almost all questions, suggestions from both reviewers. I find the revised version much easier to read, including the presentation of the data and methods, the results, discussion and interpretations. The conclusion has been also revised and the discussion indicates that these preliminary results certainly call for new analyses as mentioned by the authors in their first replies. Interesting new figures have been added (e.g. TCO2/S, TALK/S, new Figure 6) that should be redrawn (a curve for Atlantic water is apparently missing in these figures).

I think the paper is suitable for publication in its present form (pending few corrections).

I do have few comments (mainly minors) on the revised MS listed below.


;;;;;; Comments on revised manuscript

C-01: Line 25: In the abstract add uncertainty associated to the number listed (flux).

C-02: Line 58: “The North Atlantic is a relatively well observed region of the ocean”. Lauvset et al (2016) reference is for a climatology. For pCO2, TALK and DIC surface observations in the North Atlantic maybe refer to Takahashi et al (2009), Bakker et al (2016) and Reverdin et al (2018) for recent observation synthesis in this region (see comment from reviewer 2).

C-03: Line 205: typo: carrbonic.

C-04: Line 301: typo: and and continued…

C-05: Figure 5a and 6: there is no data in January (correct ?).

C-06: Table 2, Figure 6 and text: for TCO2/S and TALK/S ratios specify units (e.g. µmol.kg-1/psu).

C-07: Figure 6: Red horizontal lines for Atlantic Waters specified in the legend but not seen in figures.

C-08: For a reader not familiar with this topic, would it be nice to show also TCO2 and TALK data (not only the ratio with S; maybe add plots of TCO2 and TALK data in sup mat).

C-09: Line 390: Reference to Figure 6e ?

C-10: Line 395: “Representative annual long term TCO2/S and ALK/S means would be more realistic but are not available”. Would that be possible to get this information from GLODAP in this region or using TCO2 and TALK climatological fields (Takahashi et al 2014; Broullon et al 2019, 2020), although this might not change the results.

C-11: Line 399: reference to Table 1 or 2 ?

C-12: Line 402: Like in the submitted manuscript there is confusion here: “Polar waters having an increasingly higher alkalinity/salinity and alkalinity/TCO2 ratios”. The ratio TCO2/ALK in Polar water is higher than in Atlantic (Table 2). Please rephrase.

C-13: Lines 410 and 448: Typo: Arcipelago

C-14: Line 434: Not clear to see in Figure S3 that pCO2 unchanged for the ratio 0.85. Could this be clarified ?

C-15: Line 445: Not sure how you derive the value of 34.8 10**12 mol CO2/yr and 0.058 PgC/yr. I suspect a typo here: maybe this is 4.8 10**12 mol CO2/yr (as correctly listed in the supplement).

C-16: Line 445: The effect of excess alkalinity of 0.058 PgC/yr is a number that could be recalled in the abstract. Would it be possible to associate an uncertainty for this number ?

C-17: Line 457: Typo: thTCO2/Alk

C-18: Figure S3: legend: typo: contours in in….

;;;;;;; end review

referee#2:

The authors have satisfactorily responded to my queries, and I find the paper much clearer with interesting and challenging results, in particular about the role in the Iceland Sea of the excess alkalinity of Arctic-origin waters on the surface pCO2 and air-sea exchanges.

I have only a few technical comments that can easily be fixed, and that I list below:
l. 49-51: end of sentence to rewrite
l; 142: rewrite ‘to 5 stations’ as ‘with 5 stations’
l. 147: remove ‘, two years’
l. 163: change ‘so were also calibrated...( as ‘the standards... measurements were similarly calibrated’
l. 191: ‘we use discrete samples of pCO2 and TCO2, and calculated...’
l. 226: I would still have liked what is the percentage of missing data that had to be filled with Mauna Loa values.
l. 300: ‘undersaturation’
l. 338 ‘which confirms...’
l. 465: ‘western Arctic Ocean’ (western is a little bit vague ... does it refer to the European Arctic?)
l. 477: carried
l. 498: ‘technical’
Supplementary material:
Figure S3, the caption title should include the mention that it is plotted at t=5°C and S=35 (then, it does not need to be mentioned in the next sentence for the red square)

Dear authors,

many thanks for addressing the remaining referee comments and for submitting your revised manuscript, which I am now happy to accept. Before uploading the final files I would encourage you to increase the resolution of figures 3 and 5b, so they match the resolution of the other figures.

Best regards
Peter Landschützer