Effect of gas-transfer velocity parameterization choice on air–sea CO<sub>2</sub>
fluxes in the North Atlantic Ocean and the European Arctic

Wrobel, Iwona; Piskozub, Jacek

doi:https://doi.org/10.5194/os-12-1091-2016

Articles | Volume 12, issue 5

https://doi.org/10.5194/os-12-1091-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/os-12-1091-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 12, issue 5

Research article

|

30 Sep 2016

Research article |

| 30 Sep 2016

Effect of gas-transfer velocity parameterization choice on air–sea CO₂ fluxes in the North Atlantic Ocean and the European Arctic

Iwona Wrobel and Jacek Piskozub

Download

Final revised paper (published on 30 Sep 2016)
Preprint (discussion started on 03 Nov 2015)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC C1256: 'Review of Wróbel and Piskozub: Effect of gas-transfer-velocity parameterization choice on CO2 air-sea fluxes in the North Atlantic and European Arctic”', Anonymous Referee #1, 30 Nov 2015
- AC C1525: 'Respons to referee#1', Iwona Wrobel, 27 Jan 2016
RC C1262: 'Manuscript review', Anonymous Referee #2, 01 Dec 2015
- AC C1535: 'Response to reviewer#2', Iwona Wrobel, 27 Jan 2016

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (10 Feb 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (16 Feb 2016) by Mario Hoppema

RR by Anonymous Referee #3 (23 Feb 2016)

Suggestions for revision or reasons for rejection

The paper “Effect of gas-transfer-velocity parameterization choice on CO2 air-sea fluxes in the North Atlantic and the European Arctic” by Wrobel and Piskozub study the uncertainty of air-sea CO2 flux estimates in the North Atlantic due to the air-sea gas parameterization, particularly with respect to five different commonly used gas transfer rate formulas. They indicate that up to 5% uncertainty can be attributed to this. They suggest that relatively stronger wind speed and lack of seasonal changes in the flux direction are the reason for this uncertainty.

Although I found the study interesting, I feel that the analysis performed is too brief, which resulted in relatively limited new and significant advancement in our current knowledge understanding. There are also a number of English errors, which make the manuscript unpublishable at this stage. Below I have provided the authors with some ideas and suggestions to extent the current work and to further improve the overall presentation of the study as well as some minor corrections.

Major comments.
Although it is useful to quantify uncertainty attributed to the gas transfer formulation, there is very limited new information for the community. Given that the authors have done the computation for the whole global ocean, a more detailed investigation on regional features would be very useful, i.e., in addition to the North Atlantic. The authors state that the uncertainty could be substantial in regions of high wind speed. Is this only valid for the North Atlantic or also for other strong wind regions such as the Equatorial Pacific and the Southern Ocean?

To compute uncertainty highlighted in this study, one would only need to provide surface pCO2, SST, SSS, wind speed, and atmospheric CO2 fields. In addition to the above point, I think the authors should also consider studying how this uncertainty would evolve in the future, regionally and temporally. To do this, one can use the CMIP5 model outputs, which are publicly available. Questions such as will uncertainty due to atmospheric CO2 concentration dominates the uncertainty due to wind speed formulation, etc. has not been studied previously and can be addressed here.
In which regions, and when in the future, the uncertainty become significant thus provides some recommendation for improving the monitoring network to the observational community.

L210-214: I am not entirely convinced by this analysis and I think that this is not entirely true (e.g.based on Fig 4d and 3d). From my visual inspection, the spatial variations in Fig.4 actually resemble very much Fig. 2, and therefore, I would assume the difference shown in Fig. 4 to a large extent stems from the amplitude of delta pCO2 values, rather than the wind speed. Assuming that atmospheric CO2 has very little spatial variability, the delta pCO2 patterns will be very similar with Fig. 2.

Fig. 5 shows that the differences in wind speed parameterization is largest between McGills and Nightingale parameterizations. Based on this, it would be more informative to show the difference between fluxes computed based on these formulations in Fig. 4, as an indicator for upper uncertainty range.

Some of the existing analysis can be illustrated in a simpler and compacted way, such as put values from Figs. 6 and 7 into one single table (Line 234-237, the values from the global estimates, which are missing from Fig. 7 should be added to this table as well). In addition, to better illustrate regions where different wind parameterization could be important for air-sea fluxes estimation, I recommend showing a map of standard deviation/variance of gridded fluxes computed using the different equations.

Minor comments
L30-32: I suggest revising this sentence into something like: “The seasonal flux in the Arctic computed from the two climatology data sets are opposite to one another, possibly due to insufficient spatial and temporal data coverage, especially in the winter.”

L55: Studying the rate of the ocean CO2 sink…

L55: … especially its long-term trends, one needs …

L60: parameterization

L83 … European Space Agency funded OceanFlux …

L85: Maybe better to replace “opening sourced” with “publicly available”

L98: calculations.

L98 users can choose

L99: how is ice age affect air-sea flux? clarify. Do you mean sea-ice area?

L100: clarify what is meant by “whitecapping”?

L101: configure them in ..

L111: Bakker et al.

L117. clarify what is meant here by “preprocessed”?

L133: What is meant by “sea state”?

L137: the solubility unit should be g m-3 microatm-1

L175: in water])

L184: do you mean “Ocean Flux GHG project”?

L195: NIghtingale et al. (2000)

L195: Takahashi et al., (2009)

L197: replace “all-season” with “annual”

L205: are shown

L219 “… by up to 30% …”

L269:271, again not clear to me where are 33% and 50% quantities derived from?

L257-258, where does the 9% value come from? Fig. 7 only shows values for North Atlantic and Arctic.

Double check the units throughout the text. I believe most of the [Tg] should be [Pg].

Figs caption: would be useful to state that positive values represent outgassing and negative otherwise, when showing the air-sea flux of CO2 values.

Fig1. the color bar can be improved, e.g., use ranges from -20 to 20 since there seems to be no grid points showing values of -40.

Fig. 1. Add zero contour lines to distinguish uptake and outgassing areas.

Fig. 6a and 6b are identical, and the units in x-axes are wrong as well.

Fig. 7. Clarify what is meant by normalized here. Shouldn’t it be unit-less instead of Tg/year?

Fig. 8: units should be in Pg instead of Tg.

Last paragraph of introduction: In addition to formula in the gas transfer velocity, the selection of wind product also contribute to the uncertainty, therefore would be good to mention this, citing the study by Gregg et al., 2014 (Ocean Modelling: Sensitivity of simulated global ocean carbon flux estimates to forcing by reanalysis products.)

There is also a study by Landshutzer et al. (2011, in Biogeosciences Discussion, A model study on the sensitivity of surface ocean CO2 pressure with respect to the CO2 gas exchange rate), which can be cited as well.

I would also recommend a native speaker to review the English prior to resubmission.

Hide

RR by Anonymous Referee #2 (02 Mar 2016)

Suggestions for revision or reasons for rejection

I think that many of the comments have been addressed. However, the paper still has many grammar and spelling mistakes and needs a thorough editing. I think there should be some minor revisions before acceptance:

1)SOCAT is not a climatology and should not be referred to as such (see the Abstract). Whatever was done with the FluxEngine is fine, but when talking about SOCAT it should not be called a climatology.
2) First review and Abstract - The authors misunderstood my criticism of the word uncertainty. I do not mind the concept of comparing different k parameterizations. I mind use of the word uncertainty. The fact that parameterizations result in fluxes that are 10% or 50% different does not mean that is uncertainty in the k value or the resulting flux. The uncertainty in the k param. has to do with the methods used to make the measurement. I would like it better if the difference in fluxes related to choice of k was called something else, rather than uncertainty (for example, range or difference).
3) In my initial review, I made a comment about Figure 7 as well as Figure 8. The authors misunderstood. Point about fig 7 - This figure does not add anything over Figure 6. Why do you need both figures? Point about fig 8 - This is an interesting result and should be described in more detail. It is fine that the authors think this is outside of the scope of the current manuscript. I just want to make it clear that there were 2 seaparate comments. Here I would like the authors to address only the comment about Figure 7.
4) Again, I think it is strange to cite workshops here. It doesn't mean the point of the authors is not valid. I leave it to the editor to decide, but I think the paper reads less professionally when workshop proceedings are cited. There should be substantial support in the literature for the points made by the authors other than 'these folks said it is important at a workshop'.

Hide

ED: Reconsider after major revisions (04 Mar 2016) by Mario Hoppema

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (14 Apr 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (15 Apr 2016) by Mario Hoppema

RR by Anonymous Referee #3 (23 Apr 2016)

Suggestions for revision or reasons for rejection

The authors have addressed some (but not all) of my major comments in the previous round, albeit not satisfactorily. E.g., in addressing my suggestion to include CMIP5 models, the authors provided rebuttal, that the analysis would be “burdensome and actually pointless”.

I have suggested to explore how the uncertainty of CO2 fluxes attributed by the wind speed parameterization may evolve in the future. They responded that “it has been done and results are published” by showing Fig. 12.19 of the IPCC-AR5. I don’t understand this argument. The figure is completely different than what I suggested. Further, based on this figure, they claimed that “This is not anyone would call robust prediction”. In fact, the figure indicates nothing about the robustness of the model predictions. It indicates that the inter-models mean projects a small future changes, which in most cases are still within the range of internal variability.

Nevertheless, I understand that for those not familiar with processing CMIP5 models, this task could be overwhelming and time consuming. I do think that my suggestion on analyzing the SOC or other key ocean regions is feasible, and their argument for focusing on the North Atlantic and Arctic regions is not very convincing.

The authors also did not respond to my comments on replacing information in Figs. 6 and 7 into a Table. This has led me to identify further errors in the manuscript:
Line 237: according to values provided earlier on Line 231, 42% should be 47%
Line 332: 42% should be 47%
Line 272: “50%” should be “more than 50%”
Fig 7b, purple bar does not look correct (i.e. it shows here smaller than the light blue bar, but it is larger in Fig. 6b)

The suggestion to have an English native speaker reviewing the manuscript prior to resubmission appears to be ignored as well. Overall, I have the feeling that the authors did not spend sufficient time to grasp the reviewer’s suggestions, which in most cases are constructive. This is also evident from the authors’ response to Reviewer#2 comments, e.g., his/her request to add more discussion to Fig. 8, and NOT to remove it.

More minor errors identified:
L115-118: I suggest to clarify the sentence, e.g., by splitting into two sentences.
Line 129: 130, lower case north
Line 137: OF should be F?
L140: unit of alpha should be: g ‘C’ ..
L147: So equation (1) now becomes …
L152: Equation(2) can also be represented as:
L156: referred to as the ..
L275-276: I believe the Southern Ocean, a region with stronger surface wind, could contribute to the global differences and should be discussed here.
L388: OceanFlux
L300: gas flux
L552 and 557: I am not sure what is Ocean Science policy on citing reference materials that are in submission stage.
Units in Figs. 1 and 4 color panels: superscript “-2“ and “-1”
Units in Fig. 3 color panel: superscript “-1”
Fig. 3 caption: unit is wrong. (also in L572)
Fig. 5 caption and y-axes label: units should be Pg C/month, (also L583)

Hide

ED: Reconsider after major revisions (05 May 2016) by Mario Hoppema

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (13 Jun 2016) Author's response Manuscript

ED: Reconsider after major revisions (01 Jul 2016) by Mario Hoppema

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (12 Aug 2016) Author's response Manuscript

ED: Publish subject to minor revisions (Editor review) (23 Aug 2016) by Mario Hoppema

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (02 Sep 2016) Author's response Manuscript

ED: Publish subject to technical corrections (14 Sep 2016) by Mario Hoppema

AR by Iwona Wrobel-Niedzwiecka on behalf of the Authors (19 Sep 2016) Author's response Manuscript

Short summary

We used a recently developed tool – FluxEngine, to calculate monthly net carbon air–sea CO₂ fluxes for the extratropical North Atlantic Ocean, European Arctic, and global values, using several available parameterizations of gas transfer velocity on different dependence of wind speed. The aim of the study is to constrain the uncertainty caused by the choice of parameterization in the North Atlantic, a large sink of CO₂ and a region with good measurement coverage, characterized by strong winds.

Effect of gas-transfer velocity parameterization choice on air–sea CO2 fluxes in the North Atlantic Ocean and the European Arctic

Download

Interactive discussion

Peer-review completion

Effect of gas-transfer velocity parameterization choice on air–sea CO₂ fluxes in the North Atlantic Ocean and the European Arctic