Hi Lauren,

We would like to thank you for reading our manuscript carefully and providing us with comments. We are excited to hear that you enjoyed reading the paper. As a result of the reviews provided by you and others, we realize that the paper will require major revisions before it can be published. Like you, we think that this will be an important contribution to improve our knowledge about inorganic carbon dynamics in the Chukchi Sea and believe that once revised, it will fit well into the journal Biogeosciences. You will find our responses in bold below. Thank you for pointing out the subscript errors and other typos. We will correct all errors in our revised version.

Hi! Great paper; I was really excited to see more information about this topic. Here are some notes.

L69-71,80-83,87-88,629: fix sub- and super-scripts

We will revise accordingly.



L99: italicize species name

We will revise accordingly.



L116: delete "and variability"

We will revise accordingly.



L111,123: can you cite primary literature instead of IPCC?

Agreed. We will add primary literature.



L156: delete parentheses around reference since you're directly talking about it

We will revise accordingly.



L290: silicate can be quite high in this region, affecting your TA calculation. Also, the assumption of zero nutrient concentrations might work for surface water, but not near-bottom, especially as you observed the results of respiration in your pCO2 and pH, that should translate to increased nutrients as well.

We are currently looking into estimating silicate and phosphate to recalculate the inorganic carbon parameters, either from model reanalysis or from nitrate and oxygen-based algorithms. However, we are also revisiting if any analysis with TA and DIC are feasible given the large uncertainty pointed out by you and other reviewers.

L293: I agree. Can you be more specific about the uncertainty that the pH-pCO2 pair can lead to? Especially because you're using pCO2 and pH calculated from it, so really just pCO2. What range or at least order of magnitude of error do you expect this to lead to in your reported TA, DIC, and sat arag? Additionally include error propagation from CO2SYS calculations (should be available in most new versions). To that end I think you need error envelopes around your data lines in Figure 2.

We are adding uncertainty envelopes to all inorganic carbon parameters in our figures. TA and DIC have an uncertainty of about 10 %, hence, we are revisiting our analysis and discussion around these parameters.



L374: correct to "drivers"

We will revise accordingly.

L385: here and all the times after, use Greek letter μ instead of u

We will revise accordingly.



L395: In this whole discussion I think it's important to include your uncertainties in observations of DIC or TA increase (I suspect could be quite high), as that will inform how disparate the Redfield ratio calculations actually are from your measurements.

Agreed - based on the large uncertainty (~10 %) we are rethinking this analysis.



L409: change "we're attributing" to "we attribute"

We will revise accordingly.



L421: delete "to" leaving "temperature increase"

We will revise accordingly.



L437: Can you discuss this in the context of nutrient data as well? Is it possible that some amount of pCO2 decrease is also attributable to mixing stimulating phytoplankton blooms?

Good point, we will bring this point up as well and conclude that with the data we have we can only speculate.



L466: Again I think error discussion is important here (as in write 85.5 +/- X umol/kg), the error could easily be on the order of these discrepancies

We will add uncertainty envelopes. Given the large uncertainty we will likely have to revisit whether the use of TA and DIC in the context of inorganic carbon drivers is feasible.



L554: Correct 'maybe' to 'may'

We will revise accordingly.



L612: I find this paragraph too speculative (off the back of the paragraph discussing the uncertainty about the carbonate chemistry impacts on organisms here) for the serious claims it brings up

We plan to rephrase this paragraph the following:

Indigenous communities are at the forefront of the changing Arctic, including changes in accessibility, availability, and condition of traditional marine foods (Buschman and Sudlovenick, 2022; Hauser et al., 2021).  Several local species are critical to the food and cultural security of coastal Inupiat who have thrived in Arctic Alaska for millenia. While it is not possible to resolve the consequences of the seasonal and interannual variations in carbonate chemistry documented in this manuscript without a proper sensitivity evaluation, the seasonal low pH conditions have the potential to impact organisms like bivalves in a foraging hotspot for walrus (Jay et al., 2012; Kuletz et al., 2015). Walrus, as well as their bivalve stomach contents, are important nutritional, spiritual, and cultural components, raising concerns for food security in the context of ecosystem shifts associated with the variability and multiplicity of climate impacts within the region (ICC, 2015).

L625: change 'was' to were' (subject is 'events')

We will revise accordingly.



L630: italicize species name

We will revise accordingly.



L650-653: again a bold claim for how speculative it is. Also see Mordy et al. 2020 Deep Sea Res II, they have nitrate data from a very similar region (looks like you referenced them but not for your discussion here)

We appreciate your feedback. We are having a challenging time discussing these data with a single point mooring. We will reread Mordy’s paper and reword this paragraph with a broader view, including the possibility of advection and denitrification.



L663: overestimates pH "by" …

We will revise accordingly.



L666: air-sea gas…

We will revise accordingly.



L679: I think this is an important discussion and highlights the necessity for time-series monitoring like this, especially since most of the long-term changes in pH and sat arag are assumed from some discrete datasets that are biased to fall, which you identify as the high-pCO2 time period

Agreed.

Overall, I find this paper interesting and high temporal resolution measurements are super important due to the historical paucity of data, as we try to parse future changes due to climate change. The main thing to change here is the lack of transparency regarding uncertainty in carbonate chemistry parameters, especially since most of them are calculated and not directly measured. I really think that could alter some of your discussion and implications. I also wonder if the discussion couldn't focus more on physical processes than on biological impacts, since by your own admission those are relatively unclear for organisms in the Chukchi Sea?

Thank you again for your thoughtful comments. We will rewrite the discussion given the large uncertainty in TA and DIC. It is difficult to analyze physical drivers with the data we have, but we will do our best.

Thanks very much for an interesting read!

We would like to thank reviewer 1 for their constructive comments and appreciate their time to carefully read the manuscript. We are excited that they found the data and findings interesting and helpful to better understand the changing Chukchi Sea seascape. As a result of their review and others, we realize that the paper will require major revisions before it is ready for publication. Nevertheless, we believe that once it is revised it will make an important contribution to improve our knowledge about inorganic carbon dynamics in the Chukchi Sea and will fit well into the journal Biogeosciences. Our responses are shown in bold below.

Major comments:

Title:

I find that the title of this manuscript is a little bit too ambitious. One mooring location may not represent all Arctic Shelf Seas. Maybe consider changing “Arctic Shelf Sea” to “the Chukchi Sea”.

We will change the title to “Sea ice loss translates into major shifts in the carbonate environmental conditions in the Chukchi Sea”

Salinity-normalized data

Although the authors pointed out that freshwater can affect the inorganic carbon system and they attempt to normalize DIC and TA to a reference salinity, they didn’t clarify the freshwater end-member they used for salinity normalization. Since normalized data are critical for further quantitative analysis, they should be clearly presented.

The freshwater end-members were discussed in the methods section “2.8 Influence of freshwater and temperature on inorganic carbon system”

In addition, it is not clear to me if the authors used normalized data in Results and Discussion sections because they presented changes in DIC, TA, and NO3, not nDIC, nTA, nNO3, thus it is really hard to evaluate their data interpretation.

We agree that this was not clear. On line 335 in the preprint, we wrote: “Comparison of normalized (n) nDIC, nTA, npCO₂, and nNO₃ with the non-normalized variables suggests that the large pCO₂ and pH changes in spring, summer, and fall are not driven by freshwater (S3).”

We will clarify: “Comparison of normalized (n) nDIC, nTA, npCO₂, and nNO₃ with the non-normalized variables suggests that the large pCO₂ and pH changes in spring, summer, and fall are not driven by freshwater (Figure S3). We will therefore use the non-normalized variables throughout the paper.“

Minor comments:

Line 25 Need to define “pCO₂” the first time it appears.

We will no longer abbreviate it in the abstract and then introduce the abbreviation on line 80.

Line 37 I feel that “CO2-depleted surface water” is not a very accurate description. The authors need to explain how they define “depleted”. Maybe “low pCO2 surface water” is good enough.

Agreed. We will reword to “have a low pCO₂” here and throughout the manuscript.

Line 37 I feel that ice formation-induced water convection in late fall may also play a role.

Good point, we will investigate this further and discuss accordingly.

Line 56 The citation is not complete.

It seems that unlinking Zotero from the word document right before submission caused all sorts of issues with the references. We will fix this reference and all others.  Lesson learned for the future.

Line 67-71 Please carefully check throughout the entire text and modify the terms as “CaCO3”, “CO2”, CO32−, “pCO2”, and “Ωarag”. I cannot list all the wrong spell here.

We apologize - not sure how we did not notice during the final read. All terms will be fixed.

Line 95 Need to define CEO at the first time it appears

We will revise accordingly.

Line 190-192 Should Move to 2.5 CTD and Oxygen section. How about DO data collected from Aug 2017 to Aug 2018?

We will revise accordingly.

Line 220 What is CEO-2? Need to explain.

Not relevant here. We will delete “2”.

Line 220 …mooring near the HydroC…

We will revise accordingly

Line 264 I don’t understand why the r2 for the validation part is even higher than that of the training part. Why select data from 15/9/2017 to 14/3/2018 as the training dataset?

Dates for the training dataset were chosen arbitrarily over a 6-month timespan so as not to influence the algorithm performance. We will investigate why r² for the validation is higher than for the training dataset and provide an explanation in the paper.

Line 253-255 The parameter symbol of 𝛼 (not a) should be consistent throughout the text and equation.

We will revise accordingly.

Line 288-289 TA, DIC, and Ωarag (Figure 2 i-k) were calculated based on data from the HydroC pCO2, pHest and S, T, and pressure (P) from the SBE16.

We will rephrase to: “TA, DIC, and Ω_arag (Figure 2 i-k) were calculated based on measured pCO₂, S, T, and pressure (P) and algorithm-based pH (pH^est).”

Line 306 (Sulpis et al. (2020) found that…

We will revise accordingly.

Line 323 Please cite the literature for supporting this statement “also at times by the Mackenzie River outflow from the eastern Beaufort Sea and the large Russian Arctic rivers located to the west of the Chukchi Sea.”

We will add the following citations:

Asahara, Y., Takeuchi, F., Nagashima, K., Harada, N., Yamamoto, K., Oguri, K., and Tadai, O.: Provenance of terrigenous detritus of the surface sediments in the Bering and Chukchi Seas as derived from Sr and Nd isotopes: Implications for recent climate change in the Arctic regions, Deep Sea Res. Part II Top. Stud. Oceanogr., 61–64, 155–171, https://doi.org/10.1016/j.dsr2.2011.12.004, 2012.

Jung, J., Son, J. E., Lee, Y. K., Cho, K.-H., Lee, Y., Yang, E. J., Kang, S.-H., and Hur, J.: Tracing riverine dissolved organic carbon and its transport to the halocline layer in the Chukchi Sea (western Arctic Ocean) using humic-like fluorescence fingerprinting, Sci. Total Environ., 772, 145542, https://doi.org/10.1016/j.scitotenv.2021.145542, 2021.

Line 331 nP = (P -P(s=0))/S* Sref +P(s=0), (2)

We will revise accordingly.

Line 336 Not clear how the authors determine the freshwater endmember for DIC, TA, pCO2, and NO3. I suggest listing all estimated freshwater end-member in a table in the supporting information.

The freshwater end-members were discussed in the methods section “2.8 Influence of freshwater and temperature on inorganic carbon system”. We can add a table with all calculated end-members.

Line 337 I don’t think pCO2 can be directly normalized by salinity because it doesn’t change linearly with salinity.

We will investigate this in more depth and will address this comment accordingly.

Line 336-338 How does the brine rejection process during ice formation affect seasonal variations in these variables? Especially during the winter.

Good point. We will discuss this in the context of a Beaufort Sea study by deGrandpre et al., (2019), however our inorganic carbon data does not show evidence of brine rejection processes.

DeGrandpre, M. D., Lai, C. Z., Timmermans, M. L., Krishfield, R. A., Proshutinsky, A., and Torres, D.: Inorganic Carbon and pCO2 Variability During Ice Formation in the Beaufort Gyre of the Canada Basin, J. Geophys. Res. Oceans, 124, 4017–4028, https://doi.org/10.1029/2019JC015109, 2019.

Line 338 and 345 Should be (Figure S3) and (Figure S4)

We will revise accordingly.

Line 344 pCO2 ,NT = pCO2 * exp(0.0423(Tref - T)) (3)

Should list the equations in a number order

We will revise accordingly.

Line 346 …surface waters were entrained to…

We will revise accordingly.

Line 348 “pCO2”

We will revise accordingly.

Line 357 Please explain why selected 51% of sea ice concentration as a threshold. People normally use <15% as a threshold for open water.

We adjusted the figure using <15 % as a threshold. It does not affect the broader results.

Line 375 Needs to define “springtime” in this study.

We will define spring, summer and fall at the beginning of the study.

“higher pH and Ωarag and lower pCO2” compared to what? The overall means?

We will add: “compared to the overall mean”

Line 377 I don’t see the spikes in June of 2020

We address this in section 3.3.

Line 384 If the oxygen data is already known as bad data as shown in Fig. S2, how can the author trust the oxygen changes can be used for quantitative analysis?

We trust the relative oxygen values from the SBE63 can be used for quantitative analysis because the sensor deployed in 2017-2018 was pre- and post-calibrated, and the sensor deployed in 2018-2019 was pre-calibrated with processed data showing good agreement (relative difference of 4%) from 2018 recovery and 2018 deployment (smaller than the relative difference of 6% between the last two measurements prior to the 2018 recovery). However, due to the large uncertainties in TA and DIC we will likely not conduct this analysis in the revised paper.

Line 388 Not clear how the authors estimated TA increase with the assumption of NO3 consumption. No observation of NO3 in the 2018 spring. Need more explanation.

Due to the large uncertainty in TA we will put less weight on this analysis in the revised version of the paper. However, in this example, we used oxygen to derive DIC, and then estimated TA with + 0.15 µmol TA per μmol DIC consumed.

Line 404 If the water column is well-stratified, does the observation at 33 m reflect the surface mixed layer or bottom layer at this period? If it is reflecting the variation of the surface mixed layer, the CO2 air-sea exchange should be considered. If it is reflecting the variations of bottom layer, how CaCO3 mineral dissolution in the surface layer increases TA in the bottom layer?

The water column is stratified upon sea ice break-up as a result of sea ice melt, so it is unlikely that there is influence of CO₂ air-sea gas exchange at 33 m. However, ikaite particles could penetrate the mix layer depth and affect TA at the bottom. Since TA uncertainty is large (~10 %) we will put less weight on this analysis and dissolution of ikaite will only be discussed as a possibility rather than a fact.

Line 407 Needs to define “fall” in this study

We will define all seasons at the beginning of the manuscript.

 …drops in pCO2 ?

We will revise accordingly.

Line 431 There is no figure (Figure 7) showing changes in Ωarag

We will add (Figures 3 and 7).

Line 458 DIC changes or nDIC change?

We addressed this comment above. We use non-normalized variables in the manuscript.

Line 491 NO3 decreased below 10 μmol kg-1 shown in Fig. 2f

We disagree. The figure shows that it is above 10 μmol kg^-1 in July 2020.

Line 514 the authors need to clarify that they are comparing the acidification rate at the surface vs. in the water columns

Thank you for pointing this out. We will rewrite L513-519.

Line 515-516 I didn’t see the result of the statement “The weaker trend was calculated with data starting in 1994, whereas the stronger trend used data starting eight years later.” in Qi et al. (2022). Need to explain how the author concluded that.

We will rewrite L513-519.

Line 520 Please clarify if the model results were derived from the depth of the CEO.

This was stated in the methodology section: “Each simulation was used to calculate the annual trend of aragonite saturation state and pH at the closest depth and grid cell to the CEO mooring.”

Line 604 …pH (e.g. Gianguzza et al., (2014).

We will revise accordingly.

Line 625… and lower pCO2, were driven by a combination

We will revise accordingly.

Line 627-629 It is not necessarily and exclusively to be sea ice algae, because the water column is not well-stratified in the spring, the phytoplankton is also able to grow at the subsurface or to be entrained into the subsurface.

Please check Arrigo et al., 2017 and Ouyang et al., 2022

Arrigo, K. R., Mills, M. M., van Dijken, G. L., Lowry, K. E., Pickart, R. S., & Schlitzer, R. (2017). Late spring nitrate distributions beneath the ice‐covered northeastern Chukchi Shelf. Journal of Geophysical Research: Biogeosciences, 122(9), 2409-2417.

Ouyang, Z., Collins, A., Li, Y., Qi, D., Arrigo, K. R., Zhuang, Y., ... & Cai, W. J. (2022). Seasonal Water Mass Evolution and Non‐Redfield Dynamics Enhance CO2 Uptake in the Chukchi Sea. Journal of Geophysical Research: Oceans, 127(8), e2021JC018326.

We will review these publications and rephrase accordingly.

Line 693 “dataset”

We will revise accordingly.

Line1273 I suggest enlarging Figure 2 in the y-axis scale. Since this is the most important and valuable figure in the manuscript, it should be made more readable.

Also, it is really hard to see the measured pH (grey line) in Fig. 2c. Suggest changing it to another color.

Thank you for these helpful comments. We will redo these figures accordingly.

Line 1286-1287 Please enlarge the green diamond and its error bars in the figure to make it more readable.

We will follow these suggestions.

Supplement

Line 7-8 the format of the date should be consistent with the label in the figure.

Will do.

Figure S3 Please enlarge the labels to make them more readable. Also, suggest changing the normalized data to a more readable color.

Will do.

Figure S4 Suggests changing the temperature-normalized data to a more readable color.

Will do.

We would like to thank reviewer 2 for their constructive comments and appreciate their time to carefully read the manuscript. We agree that it is important to present new data from an Arctic coastal area and are excited that the reviewer found that this dataset presentation is within the scope of Biogeosciences. We agree with this reviewer’s comments that we need to more carefully address advection and diffusion. As a result of their review and others, we realize that the paper will require major revisions before it is ready for publication. Nevertheless, we believe that once it is revised it will make an important contribution to improve our knowledge about inorganic carbon dynamics in the Chukchi Sea and will fit well into the journal Biogeosciences. Our responses are shown in bold below.

Review comments for manuscript: “Sea ice loss translates into major shifts in the carbonate environmental conditions in Arctic Shelf Sea” submitted to Biogeosciences by Hauri et al.

General comments

Time-series observations of marine biogeochemical properties in coastal regions are important for assessing environmental changes that are directly linked to peoples’ livelihoods. In addition, there are few time-series stations, especially high-latitude oceans, where comprehensive biogeochemical observations have been made. Therefore, it is invaluable to present new data in the Arctic coastal area, although there is a large amount of missing data. At this point, I believe that this manuscript can contribute to scientific progress within the scope of Biogeosciences.

However, caution should be exercised when interpreting the observed results. This is because coastal areas usually show high spatial and temporal variations in biogeochemical properties, and data from vertical one-dimensional observations cannot remove the influences of lateral advection and diffusion. The latter is particularly important in coastal areas. To interpret the observed results, it does not seem that the discussion is conducted appropriately. Overall, there was a sense of over-discussion. For these reasons, I have to judge whether the scientific quality is inadequate.

For the presentation, the resolutions of the figures are poor for the discussion of monthly variations. I recommend focusing on the relevant time series in addition to the entire time series, as shown in Fig. 2. In the introduction, I could not find the purpose of this work. It just refers presenting data. For the discussion, I had the impression that the authors made a forced interpretation of the results. One of the reasons is probably that the authors do not have a clear purpose. Based on these points, the presentation of the manuscript as a research paper is inappropriate. And, about one-third of the manuscript is used for the description of materials and methods. I recommend submission to a journal with a purpose of data presentation.

As a whole, with the current version of the manuscript, I cannot recommend publication in Biogeosciences. Judging from a lot of missing references, I think the manuscript is not completed.

Specific comments

(1) The CEO is located at the Chukchi Sea site, where water masses from the Pacific flow to the Arctic basin. Therefore, it is expected that the physical and biogeochemical properties are influenced by lateral advection and mixing. However, there were almost no words related to these influences. The authors discussed variations and changes in biogeochemical properties, mostly from the viewpoint of vertical mixing. Therefore, it is necessary to confirm that the lateral advection and mixing are negligible at the site.

Thank you for bringing this up. We have ADCP data from some years and will use it to show that this region is known for having relatively low current speeds (it is a depositional area and the settling organic material feeds the local benthic hotspot).

(2) DIC, TA and Ω_arag are calculated from pCO₂ and pH. As pointed out by the authors, this combination leads to large error, although the error of Ω_arag is not so large. Therefore, it is necessary to show ranges of uncertainty as a result of calculation (propagation of errors), if authors are purposed to discuss carbon budget-like calculation. Or, it seems to be a good way to compare the calculated and observed values. But this is made only for DIC but not for TA. The authors conduct carbon budget-like discussion in some sections of the manuscript. For this, uncertainty of DIC and TA should be clarified.

 We agree. We will add uncertainty envelopes to all figures and discuss our results based on the uncertainty.

(3) The pH algorithm (eq.1) is evaluated with ship-based data. But I wonder the evaluation can be related to the evaluation of pH and pCO₂ data obtained by sensors attached to the mooring. I think that the evaluation by ship-based data is independent to the mooring data.

We disagree with this assessment. The fact that the algorithm works so well for ship-based measurements shows that the algorithm is robust across time and space. The algorithm was trained with mooring based data from the time period between mid-September to mid-March and evaluated with ship-based data collected across the eastern Chukchi Sea from August and September. Figure 5 shows that the algorithm-based data is to the most part within 0.02 pH units of the ship-based pH, with exception of some surface data.

(4) It is pointed (lines 681-684) out that a large gradient of pCO₂ exists. This implies that a slight tilt of mooring system causes an artificial variation of properties. With this condition, is it possible to detect fine carbon budget calculation? In this point also, uncertainty of the data should be clarified.

We will address the uncertainty more clearly and will highlight changes in pressure to indicate potential tilt in the mooring.

(5) At line 632, it is described that CaCO₃ mineral dissolution was observed. But it is very hard to find the evidence of CaCO₃ mineral dissolution through the manuscript. Discussion at lines 632-642 is too speculative. TA increases under the supersaturation of aragonite and calcite can be accounted for by e.g., water mixing.

We are rethinking the interpretation of TA based on its uncertainties and will also include water mixing in the discussion.

(6) From the title of manuscript, I expected examination of climate change, which leads to “shift” of carbonate system from one condition (sea ice existence) to another (sea ice loss). However, the content of manuscript does not show shift of climatological time scale. It is however possible to see a shift of seasonal time scale. Even in this case, differences of carbonate system between conditions in the sea ice existence and those in sea ice loss should be clarified.

We will address this more clearly in the discussion.

(7) Salinity influence is examined in the manuscript (section 2.8) but from the viewpoint of river water alone. Influence of sea-ice melting water should be also examined.

Our analysis did already include the influence of sea ice meltwaters: we state in section 2.8 that “Inorganic carbon chemistry at the CEO site can be influenced by freshwater from sea ice melt and meteoric sources (precipitation and rivers).“

To find the most appropriate representative freshwater end-member for the inorganic carbon variables we determined the intercept of a regression line across the full timeseries of each variable. This approach therefore includes both riverine and sea ice end members.

(8) In marine CO₂ community, it is required to indicate that pH is measured at which water temperature. Please describe that the pH is at in situ temperature.

We will make this clearer in the methods.

As a whole, discussion on processes based on water mixing are acceptable, although examination of horizontal mixing should be added.

We will use our ADCP data to show relatively low current speeds.

But discussion on processes based on biogeochemical changes are too speculative.

We will revisit and rethink our analysis and discussion.

Technical corrections:

In this manuscript, the observed results and the interpretation/speculation based on the result are written together in the section of results (section 3). This may lead readers and authors themselves to misunderstanding. For example, bloom and CaCO₃ dissolution are described in the abstract. But these are not observations but speculation. I think that there are no problems to describe them in the main text, but description in the abstract makes readers misunderstand as if they were observed. I recommend the authors rewriting section 3, separating observed results and speculation.

Thank you for this suggestion. We will rework it accordingly.

Through the manuscript, “CO₂-deplete” or “CO₂ depleted” is repeatedly used. It is very rare that CO₂ is depleted, being different from nutrients. “CO₂ reduced” is better.

Replaced “depleted” with “low” throughout the manuscript.

Lines 56-57: If “National Snow and Ice Data Center” is a reference, add published year. Not found in the reference.

Will add.

Line 69: “CaCO3”, “CaCO₃”

The sub and superscript errors will be fixed throughout the manuscript.

Line 70: Bednarsek et al. 2021 is cited here. But it is difficult to find the reference because of incomplete information of the reference at lines 762-764.

We had issues with unlinking the word doc from Zotero after our final read. We apologize for these inaccuracies. All references will be fixed. Lesson learned for the future.

“CO32-“, “CO₃^2-”

We will revise accordingly.

Line 79: Bates, 2015; Bates et al., 2009; Pipko et al., 2002; Mathis and Questel, 2013 are not found in the reference.

We will revise accordingly.

Line 80: “CO2”, “CO₂”

We will revise accordingly.

Line 81: Bates, 2015; Bates et al., 2009 are not found in the reference.

We will revise accordingly.

Line 83: Bates et al., 2009 is not found in the reference.

We will revise accordingly.

 “pCO2”, “pCO₂”

We will revise accordingly.

Lines 85-86: Mathis and Questel, 2013; Pipko et al., 2002; Bates, 2015 are not found the reference.

We will revise accordingly.

Line 87: “pCO2”, “pCO₂”

We will revise accordingly.

Line 88: “pCO2”, “pCO₂”

We will revise accordingly.

Lines 88-89: Hauri et al., 2013 is not found in the reference.

We will revise accordingly.

Line 89: Yamamoto-Kawai et al. (2016) is not found in the reference.

We will revise accordingly.

Line 95: “CEO”, “The Chukchi Ecosystem Observatory (CEO)”

We will revise accordingly.

Line 96: Grebmeier et al., 2015; Moore et al., 2000 are not found in the reference.

We will revise accordingly.

Line 98: Grebmeier et al., 2015; Blanchard et al., 2013 are not found in the reference.

We will revise accordingly.

Line 101: Moore et al. 2022 is not found the reference.

We will revise accordingly.

Line 111: IPCC 2022 is not found in the reference.

We will revise accordingly.

Line 123: IPCC 2022 is not found in the reference.

We will revise accordingly.

Line 141: “The Chukchi Ecosystem Observatory (CEO)”, “The CEO”

We will revise accordingly.

Line 172: “quadrature”, “quadruple”?

“To add in quadrature” is a mathematical description for equation 7 in Orr et al., (2018)

Orr, J. C., Epitalon, J. M., Dickson, A. G., and Gattuso, J. P.: Routine uncertainty propagation for the marine carbon dioxide system, Mar. Chem., 207, 84–107, https://doi.org/10.1016/j.marchem.2018.10.006, 2018.

 Lines 196 and 199: “pH_ext”, “pH_SeaFET”, define terms at the first appearance.

 We will revise accordingly.

Line 204: “quadrature”, “quadruple”?

Quadrature is correct. See above.

Line 219: “2.5 CTD and Oxygen”, it is better to remove the description written at lines 190-193 and summarize it here.

We will delete lines 190-193 and replace Lines 222 - 225 with:

“The other CTD was a pumped Sea-Bird MicroCAT (SBE37-SMP-ODO), with an integrated optical dissolved oxygen sensor (SBE63; Figure S2), integrated within the SeapHOx instrument that was deployed in fall 2016, 2017, and 2018. The SBE37-SMP-ODO did not record any CTD or oxygen data during the 2016 deployment and only recorded CTD and oxygen data between August and November 3 in 2018 due to battery failure.”

Line 220: What is the CEO-2? Different from CEO?

Not relevant here, will remove.

“morning”, “mooring”

We will revise accordingly.

Line 235: What do you mean “relative oxygen values from the pumped SBE63”? In the figures, absolute values seem to be used.

We agree with this criticism and will redo the figure to show oxygen values relative to the mean.

Line 257: In the caption of Fig. 3, explanation of dots is not found. Indicate which displays pH_SeaFET.

Thank you for pointing this out. We will address this in the caption and revision.

Line 258: No explanation of black dots is found in the caption of Fig. 3. Instead, black circles are explained.

Will do - see above.

Line 261: Define “pH_SeaFET” clearly.

Thank you, we will address this throughout the paper.

Line 283: “quadrature”, “quadruple”?

Quadrature is correct.

Line 290: “to be zero”, “to be negligible” is better because there observed ~10-20 umol/kg NO₃ is observed.

Will do.

Line 302: Show which temperature is used for pH, at in-situ temperature or a constant temperature (e.g., 25°C).

Line 302: pH is reported in total scale and at in-situ temperature for the entirety of this paper.

Line 306: “(Sulpis et al., 2020)”, ”Sulpis et al. (2020)”

We will take care of all reference issues and would like to apologize again.

Lines 306-307: “(Lueker et al., 2000), “Lueker et al. (2000)”

We will revise accordingly.

Line 308: “(Lueker et al., 2000), “Lueker et al. (2000)”

We will revise accordingly.

Line 311: “DIC(pH^est,pCO₂)”, “DIC calculated from pH^est and pCO₂”

We will address this in the revised version.

Lines 336-338: The comparison does not deny influences of freshwater. It just shows that the large changes cannot be accounted for by freshwater alone.

We will rephrase.

Line 338: “(S3)”, “(Figure S3)”?

We will address this in the revised version.

Line 345: “(S4)”, “(Figure S4)”? Figure S4 does not display the average temperature.

We will rewrite to:  “The impact of these large temperature swings was analyzed by temperature normalizing pCO2 (pCO2,NT; Figure S4) following Takahashi et al., (2002): pCO2,NT = pCO2 * exp(0.0423(Tref - T)); where Tref is the average temperature across the full timeseries.

Line 354: DiGirolamo et al. (2022) is not found in the reference. 2013?

We will address this in the revised version.

Line 361: Sivers et al. 2018 and Horowitz et al. (2018) are not found in the reference.

We will address this in the revised version.

Line 362: Boucher et al. (2020) and Seferian (2019) are not found in the reference.

We will address this in the revised version.

Lines 381-383: If sea-ice melting are associated with the TA and DIC variations, both should be decreased. As NO₃ also seems to be reduced, partially effects of biological activity?

Thank you for pointing this out. We will rewrite this to state that the water column was stratified at this time. 

Line 384: Increased from what?

We are using relative O2 changes in this context.

 Line 387: Again, decreased from what?

 Absolute values do not seem relevant in this paragraph and would make the readability more difficult.

Line 391: From Fig. 2, Ω_arag shows > 1.0 in the corresponding period. Can we expect dissolution of CaCO₃? It may be possible that it occurred as a result of dissolution. But for the explanation, time-lag is necessary between the phenomena.

We will more carefully look at the timing of Ω_arag increase. Although it is likely that we will not discuss dissolution at all as a result of the uncertainties in TA.

Lines 415-418: This part seems to insist that less dense water goes down to the deeper layer. I recommend discussion from the viewpoint of horizontal water transport.

Correct, the entrainment of less dense water to deeper depths is shown in Figure 6.

Line 448: “Bottom waters”, Not waters at 33m? 

We will clarify.

Lines 457-459: Is TA value of 4 umol kg^-1 significant?

We will revisit the whole discussion around TA.

Lines 481-484. Figure 8 displays changes of TA/DIC ratio. What does it mean and how can we realize relationship with organic matter remineralization?

Thank you for pointing this out, we will revisit and think about whether centering the discussion around the TA/DIC is feasible.

Lines 510-528: The discussion made here is related to long-term changes of acidification. It is not appropriate to short-term changes of carbonate properties obtained in the present study.

Yes, we are only presenting a few years of data. However, we think that it is important to talk about seasonal to interannual variability as found in our study in the context of long-term change.

Line 611: Boyd et al. (2018) is not found in the reference.

We will address this in the revised version.

Lines 628-629: Stabeno et al. (2020) and Koch et al. (2020) are not found in the reference.

 We will address this in the revised version.

Line 629: “pCO2”, pCO₂”

We will fix this everywhere.

Line 632: CaCO₃ mineral dissolution is not observed. That is just a speculation from TA changes.

We will revise the text around dissolution.

Lines 632-633: Are there any evidences that show or suggest ikaite crystal dissolution? At least, citation of previous studies is necessary.

As mentioned and cited in the text, ikaite crystal dissolution has been found in other regions.

Lines 645-646: In section 3.3, no evidence is presented, only possibility.

We will reword and add more citations, as pointed out by a different reviewer.

Lines 646-647: Increase in density does not always leads to resuspension. Data from e.g., turbidity meter are necessary.

We will investigate our turbidity and sediment trap data.

Line 662: The weather quality goal of is not +/-0.003 but 0.02.

We will correct this in the revised version.

Line 663: From Figs. 3 and 4c, it is hard to see overestimation of 0.0008. It is too good.

 Thank you, we agree and have replaced “(Figures 3 and 4c)” with “(not-shown)”.

Line 682: Hauri et al. (2013) is not found in the reference.

 We will address this in the revised version.

Lines 762-764: There are no pieces of information of journal.

We will correct this in the revised version.

Lines 893-895: There are no pieces of information of journal.

 We will correct this in the revised version.

Line 852: Insert one line before this line.

 We will correct this in the revised version.

Line 1065: “pCO2”, “pCO₂”

We will correct this in the revised version.

Lines 1085-1087: There are no pieces of information of journal.

We will correct this in the revised version.

Line 1272: DiGirolamo et al., 2022 is not found in the reference.

We will correct this in the revised version.

Line 1273: It is very hard to distinguish between black and gray on the figure.

We will redo this figure.

Lines 1334-1336: The top figure is a figure for O₂, but no explanation is given. What dashed lines indicate?

This will be clarified in the caption.

Table 1: Table 1 needs to be reworked. It is an unfriendly table for readers. Parameters such as NO₃, pCO₂, CTD, are placed in rows of the first column.

We will adjust this table.

Table 2: This table also needs to be reworked.

 We will adjust this table.

We would like to thank reviewer 2 for their constructive comments and appreciate their time to carefully read the manuscript. We agree that it is important to present new data from an Arctic coastal area and are excited that the reviewer found that this dataset presentation is within the scope of Biogeosciences. We agree with this reviewer’s comments that we need to more carefully address advection and diffusion. As a result of their review and others, we realize that the paper will require major revisions before it is ready for publication. Nevertheless, we believe that once it is revised it will make an important contribution to improve our knowledge about inorganic carbon dynamics in the Chukchi Sea and will fit well into the journal Biogeosciences. Our responses are shown in bold below.

Review comments for manuscript: “Sea ice loss translates into major shifts in the carbonate environmental conditions in Arctic Shelf Sea” submitted to Biogeosciences by Hauri et al.

General comments

Time-series observations of marine biogeochemical properties in coastal regions are important for assessing environmental changes that are directly linked to peoples’ livelihoods. In addition, there are few time-series stations, especially high-latitude oceans, where comprehensive biogeochemical observations have been made. Therefore, it is invaluable to present new data in the Arctic coastal area, although there is a large amount of missing data. At this point, I believe that this manuscript can contribute to scientific progress within the scope of Biogeosciences.

However, caution should be exercised when interpreting the observed results. This is because coastal areas usually show high spatial and temporal variations in biogeochemical properties, and data from vertical one-dimensional observations cannot remove the influences of lateral advection and diffusion. The latter is particularly important in coastal areas. To interpret the observed results, it does not seem that the discussion is conducted appropriately. Overall, there was a sense of over-discussion. For these reasons, I have to judge whether the scientific quality is inadequate.

For the presentation, the resolutions of the figures are poor for the discussion of monthly variations. I recommend focusing on the relevant time series in addition to the entire time series, as shown in Fig. 2. In the introduction, I could not find the purpose of this work. It just refers presenting data. For the discussion, I had the impression that the authors made a forced interpretation of the results. One of the reasons is probably that the authors do not have a clear purpose. Based on these points, the presentation of the manuscript as a research paper is inappropriate. And, about one-third of the manuscript is used for the description of materials and methods. I recommend submission to a journal with a purpose of data presentation.

As a whole, with the current version of the manuscript, I cannot recommend publication in Biogeosciences. Judging from a lot of missing references, I think the manuscript is not completed.

Specific comments

(1) The CEO is located at the Chukchi Sea site, where water masses from the Pacific flow to the Arctic basin. Therefore, it is expected that the physical and biogeochemical properties are influenced by lateral advection and mixing. However, there were almost no words related to these influences. The authors discussed variations and changes in biogeochemical properties, mostly from the viewpoint of vertical mixing. Therefore, it is necessary to confirm that the lateral advection and mixing are negligible at the site.

Thank you for bringing this up. We have ADCP data from some years and will use it to show that this region is known for having relatively low current speeds (it is a depositional area and the settling organic material feeds the local benthic hotspot).

(2) DIC, TA and Ω_arag are calculated from pCO₂ and pH. As pointed out by the authors, this combination leads to large error, although the error of Ω_arag is not so large. Therefore, it is necessary to show ranges of uncertainty as a result of calculation (propagation of errors), if authors are purposed to discuss carbon budget-like calculation. Or, it seems to be a good way to compare the calculated and observed values. But this is made only for DIC but not for TA. The authors conduct carbon budget-like discussion in some sections of the manuscript. For this, uncertainty of DIC and TA should be clarified.

 We agree. We will add uncertainty envelopes to all figures and discuss our results based on the uncertainty.

(3) The pH algorithm (eq.1) is evaluated with ship-based data. But I wonder the evaluation can be related to the evaluation of pH and pCO₂ data obtained by sensors attached to the mooring. I think that the evaluation by ship-based data is independent to the mooring data.

We disagree with this assessment. The fact that the algorithm works so well for ship-based measurements shows that the algorithm is robust across time and space. The algorithm was trained with mooring based data from the time period between mid-September to mid-March and evaluated with ship-based data collected across the eastern Chukchi Sea from August and September. Figure 5 shows that the algorithm-based data is to the most part within 0.02 pH units of the ship-based pH, with exception of some surface data.

(4) It is pointed (lines 681-684) out that a large gradient of pCO₂ exists. This implies that a slight tilt of mooring system causes an artificial variation of properties. With this condition, is it possible to detect fine carbon budget calculation? In this point also, uncertainty of the data should be clarified.

We will address the uncertainty more clearly and will highlight changes in pressure to indicate potential tilt in the mooring.

(5) At line 632, it is described that CaCO₃ mineral dissolution was observed. But it is very hard to find the evidence of CaCO₃ mineral dissolution through the manuscript. Discussion at lines 632-642 is too speculative. TA increases under the supersaturation of aragonite and calcite can be accounted for by e.g., water mixing.

We are rethinking the interpretation of TA based on its uncertainties and will also include water mixing in the discussion.

(6) From the title of manuscript, I expected examination of climate change, which leads to “shift” of carbonate system from one condition (sea ice existence) to another (sea ice loss). However, the content of manuscript does not show shift of climatological time scale. It is however possible to see a shift of seasonal time scale. Even in this case, differences of carbonate system between conditions in the sea ice existence and those in sea ice loss should be clarified.

We will address this more clearly in the discussion.

(7) Salinity influence is examined in the manuscript (section 2.8) but from the viewpoint of river water alone. Influence of sea-ice melting water should be also examined.

Our analysis did already include the influence of sea ice meltwaters: we state in section 2.8 that “Inorganic carbon chemistry at the CEO site can be influenced by freshwater from sea ice melt and meteoric sources (precipitation and rivers).“

To find the most appropriate representative freshwater end-member for the inorganic carbon variables we determined the intercept of a regression line across the full timeseries of each variable. This approach therefore includes both riverine and sea ice end members.

(8) In marine CO₂ community, it is required to indicate that pH is measured at which water temperature. Please describe that the pH is at in situ temperature.

We will make this clearer in the methods.

As a whole, discussion on processes based on water mixing are acceptable, although examination of horizontal mixing should be added.

We will use our ADCP data to show relatively low current speeds.

But discussion on processes based on biogeochemical changes are too speculative.

We will revisit and rethink our analysis and discussion.

Technical corrections:

In this manuscript, the observed results and the interpretation/speculation based on the result are written together in the section of results (section 3). This may lead readers and authors themselves to misunderstanding. For example, bloom and CaCO₃ dissolution are described in the abstract. But these are not observations but speculation. I think that there are no problems to describe them in the main text, but description in the abstract makes readers misunderstand as if they were observed. I recommend the authors rewriting section 3, separating observed results and speculation.

Thank you for this suggestion. We will rework it accordingly.

Through the manuscript, “CO₂-deplete” or “CO₂ depleted” is repeatedly used. It is very rare that CO₂ is depleted, being different from nutrients. “CO₂ reduced” is better.

Replaced “depleted” with “low” throughout the manuscript.

Lines 56-57: If “National Snow and Ice Data Center” is a reference, add published year. Not found in the reference.

Will add.

Line 69: “CaCO3”, “CaCO₃”

The sub and superscript errors will be fixed throughout the manuscript.

Line 70: Bednarsek et al. 2021 is cited here. But it is difficult to find the reference because of incomplete information of the reference at lines 762-764.

We had issues with unlinking the word doc from Zotero after our final read. We apologize for these inaccuracies. All references will be fixed. Lesson learned for the future.

“CO32-“, “CO₃^2-”

We will revise accordingly.

Line 79: Bates, 2015; Bates et al., 2009; Pipko et al., 2002; Mathis and Questel, 2013 are not found in the reference.

We will revise accordingly.

Line 80: “CO2”, “CO₂”

We will revise accordingly.

Line 81: Bates, 2015; Bates et al., 2009 are not found in the reference.

We will revise accordingly.

Line 83: Bates et al., 2009 is not found in the reference.

We will revise accordingly.

 “pCO2”, “pCO₂”

We will revise accordingly.

Lines 85-86: Mathis and Questel, 2013; Pipko et al., 2002; Bates, 2015 are not found the reference.

We will revise accordingly.

Line 87: “pCO2”, “pCO₂”

We will revise accordingly.

Line 88: “pCO2”, “pCO₂”

We will revise accordingly.

Lines 88-89: Hauri et al., 2013 is not found in the reference.

We will revise accordingly.

Line 89: Yamamoto-Kawai et al. (2016) is not found in the reference.

We will revise accordingly.

Line 95: “CEO”, “The Chukchi Ecosystem Observatory (CEO)”

We will revise accordingly.

Line 96: Grebmeier et al., 2015; Moore et al., 2000 are not found in the reference.

We will revise accordingly.

Line 98: Grebmeier et al., 2015; Blanchard et al., 2013 are not found in the reference.

We will revise accordingly.

Line 101: Moore et al. 2022 is not found the reference.

We will revise accordingly.

Line 111: IPCC 2022 is not found in the reference.

We will revise accordingly.

Line 123: IPCC 2022 is not found in the reference.

We will revise accordingly.

Line 141: “The Chukchi Ecosystem Observatory (CEO)”, “The CEO”

We will revise accordingly.

Line 172: “quadrature”, “quadruple”?

“To add in quadrature” is a mathematical description for equation 7 in Orr et al., (2018)

Orr, J. C., Epitalon, J. M., Dickson, A. G., and Gattuso, J. P.: Routine uncertainty propagation for the marine carbon dioxide system, Mar. Chem., 207, 84–107, https://doi.org/10.1016/j.marchem.2018.10.006, 2018.

 Lines 196 and 199: “pH_ext”, “pH_SeaFET”, define terms at the first appearance.

 We will revise accordingly.

Line 204: “quadrature”, “quadruple”?

Quadrature is correct. See above.

Line 219: “2.5 CTD and Oxygen”, it is better to remove the description written at lines 190-193 and summarize it here.

We will delete lines 190-193 and replace Lines 222 - 225 with:

“The other CTD was a pumped Sea-Bird MicroCAT (SBE37-SMP-ODO), with an integrated optical dissolved oxygen sensor (SBE63; Figure S2), integrated within the SeapHOx instrument that was deployed in fall 2016, 2017, and 2018. The SBE37-SMP-ODO did not record any CTD or oxygen data during the 2016 deployment and only recorded CTD and oxygen data between August and November 3 in 2018 due to battery failure.”

Line 220: What is the CEO-2? Different from CEO?

Not relevant here, will remove.

“morning”, “mooring”

We will revise accordingly.

Line 235: What do you mean “relative oxygen values from the pumped SBE63”? In the figures, absolute values seem to be used.

We agree with this criticism and will redo the figure to show oxygen values relative to the mean.

Line 257: In the caption of Fig. 3, explanation of dots is not found. Indicate which displays pH_SeaFET.

Thank you for pointing this out. We will address this in the caption and revision.

Line 258: No explanation of black dots is found in the caption of Fig. 3. Instead, black circles are explained.

Will do - see above.

Line 261: Define “pH_SeaFET” clearly.

Thank you, we will address this throughout the paper.

Line 283: “quadrature”, “quadruple”?

Quadrature is correct.

Line 290: “to be zero”, “to be negligible” is better because there observed ~10-20 umol/kg NO₃ is observed.

Will do.

Line 302: Show which temperature is used for pH, at in-situ temperature or a constant temperature (e.g., 25°C).

Line 302: pH is reported in total scale and at in-situ temperature for the entirety of this paper.

Line 306: “(Sulpis et al., 2020)”, ”Sulpis et al. (2020)”

We will take care of all reference issues and would like to apologize again.

Lines 306-307: “(Lueker et al., 2000), “Lueker et al. (2000)”

We will revise accordingly.

Line 308: “(Lueker et al., 2000), “Lueker et al. (2000)”

We will revise accordingly.

Line 311: “DIC(pH^est,pCO₂)”, “DIC calculated from pH^est and pCO₂”

We will address this in the revised version.

Lines 336-338: The comparison does not deny influences of freshwater. It just shows that the large changes cannot be accounted for by freshwater alone.

We will rephrase.

Line 338: “(S3)”, “(Figure S3)”?

We will address this in the revised version.

Line 345: “(S4)”, “(Figure S4)”? Figure S4 does not display the average temperature.

We will rewrite to:  “The impact of these large temperature swings was analyzed by temperature normalizing pCO2 (pCO2,NT; Figure S4) following Takahashi et al., (2002): pCO2,NT = pCO2 * exp(0.0423(Tref - T)); where Tref is the average temperature across the full timeseries.

Line 354: DiGirolamo et al. (2022) is not found in the reference. 2013?

We will address this in the revised version.

Line 361: Sivers et al. 2018 and Horowitz et al. (2018) are not found in the reference.

We will address this in the revised version.

Line 362: Boucher et al. (2020) and Seferian (2019) are not found in the reference.

We will address this in the revised version.

Lines 381-383: If sea-ice melting are associated with the TA and DIC variations, both should be decreased. As NO₃ also seems to be reduced, partially effects of biological activity?

Thank you for pointing this out. We will rewrite this to state that the water column was stratified at this time. 

Line 384: Increased from what?

We are using relative O2 changes in this context.

 Line 387: Again, decreased from what?

 Absolute values do not seem relevant in this paragraph and would make the readability more difficult.

Line 391: From Fig. 2, Ω_arag shows > 1.0 in the corresponding period. Can we expect dissolution of CaCO₃? It may be possible that it occurred as a result of dissolution. But for the explanation, time-lag is necessary between the phenomena.

We will more carefully look at the timing of Ω_arag increase. Although it is likely that we will not discuss dissolution at all as a result of the uncertainties in TA.

Lines 415-418: This part seems to insist that less dense water goes down to the deeper layer. I recommend discussion from the viewpoint of horizontal water transport.

Correct, the entrainment of less dense water to deeper depths is shown in Figure 6.

Line 448: “Bottom waters”, Not waters at 33m? 

We will clarify.

Lines 457-459: Is TA value of 4 umol kg^-1 significant?

We will revisit the whole discussion around TA.

Lines 481-484. Figure 8 displays changes of TA/DIC ratio. What does it mean and how can we realize relationship with organic matter remineralization?

Thank you for pointing this out, we will revisit and think about whether centering the discussion around the TA/DIC is feasible.

Lines 510-528: The discussion made here is related to long-term changes of acidification. It is not appropriate to short-term changes of carbonate properties obtained in the present study.

Yes, we are only presenting a few years of data. However, we think that it is important to talk about seasonal to interannual variability as found in our study in the context of long-term change.

Line 611: Boyd et al. (2018) is not found in the reference.

We will address this in the revised version.

Lines 628-629: Stabeno et al. (2020) and Koch et al. (2020) are not found in the reference.

 We will address this in the revised version.

Line 629: “pCO2”, pCO₂”

We will fix this everywhere.

Line 632: CaCO₃ mineral dissolution is not observed. That is just a speculation from TA changes.

We will revise the text around dissolution.

Lines 632-633: Are there any evidences that show or suggest ikaite crystal dissolution? At least, citation of previous studies is necessary.

As mentioned and cited in the text, ikaite crystal dissolution has been found in other regions.

Lines 645-646: In section 3.3, no evidence is presented, only possibility.

We will reword and add more citations, as pointed out by a different reviewer.

Lines 646-647: Increase in density does not always leads to resuspension. Data from e.g., turbidity meter are necessary.

We will investigate our turbidity and sediment trap data.

Line 662: The weather quality goal of is not +/-0.003 but 0.02.

We will correct this in the revised version.

Line 663: From Figs. 3 and 4c, it is hard to see overestimation of 0.0008. It is too good.

 Thank you, we agree and have replaced “(Figures 3 and 4c)” with “(not-shown)”.

Line 682: Hauri et al. (2013) is not found in the reference.

 We will address this in the revised version.

Lines 762-764: There are no pieces of information of journal.

We will correct this in the revised version.

Lines 893-895: There are no pieces of information of journal.

 We will correct this in the revised version.

Line 852: Insert one line before this line.

 We will correct this in the revised version.

Line 1065: “pCO2”, “pCO₂”

We will correct this in the revised version.

Lines 1085-1087: There are no pieces of information of journal.

We will correct this in the revised version.

Line 1272: DiGirolamo et al., 2022 is not found in the reference.

We will correct this in the revised version.

Line 1273: It is very hard to distinguish between black and gray on the figure.

We will redo this figure.

Lines 1334-1336: The top figure is a figure for O₂, but no explanation is given. What dashed lines indicate?

This will be clarified in the caption.

Table 1: Table 1 needs to be reworked. It is an unfriendly table for readers. Parameters such as NO₃, pCO₂, CTD, are placed in rows of the first column.

We will adjust this table.

Table 2: This table also needs to be reworked.

 We will adjust this table.