Decoupling of ΔO<sub>2</sub>∕Ar and particulate organic carbon dynamics in nearshore surface ocean waters

Rosengard, Sarah Z.; Izett, Robert W.; Burt, William J.; Schuback, Nina; Tortell, Philippe D.

doi:https://doi.org/10.5194/bg-17-3277-2020

Articles | Volume 17, issue 12

https://doi.org/10.5194/bg-17-3277-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/bg-17-3277-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 17, issue 12

Research article

|

29 Jun 2020

Research article |

| 29 Jun 2020

Decoupling of ΔO₂∕Ar and particulate organic carbon dynamics in nearshore surface ocean waters

Sarah Z. Rosengard, Robert W. Izett, William J. Burt, Nina Schuback, and Philippe D. Tortell

Download

Final revised paper (published on 29 Jun 2020)
Supplement to the final revised paper
Preprint (discussion started on 08 Jul 2019)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of “Decoupling of net community production and particulate organic carbon dynamics in near shore surface ocean waters” by Rosengard et al.', Anonymous Referee #1, 01 Aug 2019
- AC1: 'Author response to Referee #1 - October 21, 2019', Sarah Rosengard, 22 Oct 2019
RC2: 'Review of "Decoupling of net community production and particulate organic carbon dynamics in near shore surface ocean waters" by Rosengard et al.', Anonymous Referee #2, 30 Aug 2019
- AC2: 'Author response to Referee #2 - October 21, 2019', Sarah Rosengard, 22 Oct 2019

Peer-review completion

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

ED: Reconsider after major revisions (24 Oct 2019) by Emilio Marañón

AR by Sarah Rosengard on behalf of the Authors (25 Oct 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (25 Oct 2019) by Emilio Marañón

RR by Anonymous Referee #2 (17 Nov 2019)

Suggestions for revision or reasons for rejection

The authors did a good job incorporating some of the reviewers comments but there are still a number of things that in my opinion need to be addressed before publication.

The main problem that I see with the interpretation of the data is that the authors are attributing a large fraction of the difference between O2/Ar- and POC-derived rates to C export. However, if this was the case one would expect the POC-derived respiration rates to be overestimated (and larger than those derived by O2/Ar), which is not the case here. Also, if we can assume that C export is relatively constant throughout the day, estimated GPP rates would not be affected, so one would expect better agreement between rates of GPP than R, which again is not the case. So either I am missing something obvious or there must be a different process accounting for the differences observed. One possibility would be DOC dynamics (DOC produced during the day and consumed during the night). I am not an expert in optics, but could the beam attenuation be missing the larger phytoplankton, and could that account for the differences? In any case, the authors need to justify the observed differences in all rates, not only NCP.

The authors have included estimated rates of GPP and R in the revised manuscript as I had suggested, which is nice. However, I feel that they haven’t really incorporated these rates fully in the manuscript. There is no mention of these rates until the discussion section (they are not mentioned in the abstract, methods or results, and they are not presented in Figure 4), despite the fact that looking at these rates is probably critical for understanding the source of the discrepancy between methods.

In the response to one of my comments about the N2O correction in the nearshore upwelling location the authors mention: “We excluded two additional CTD casts during which N2O and O2 concentrations were measured on the reasoning that they were likely influenced by an external water mass”, and then go on to say “we neglect lateral advection of N2O into the drifter site because there is little change in [O2]bio/[N2O]bio between the two CTD casts used to apply a vertical mixing correction during drifter deployment, suggesting a water masses with similar N2O signatures” Wouldn’t the 2 abnormal CTD casts be an indicator of potential lateral advection? Would lateral advection be a potential explanation for the NCP discrepancies?

Regarding the validity of the N2O correction at drifter site 2 the authors say:” According to Izett et al. (2018), the sub-saturation of [N2O]bio in the mixed layer at drifter site 2 invalidates the vertical mixing approach in this environment”. Also, there seem to be a subsurface O2 maximum at that site with a very limited increase in N2O below the mixed layer, again invalidating the N2O method. However, in the revised manuscript the authors still use the N2O method to justify the lack of vertical mixing which I think is misleading.

Specific comments

L285-287 Potentially you could use all values (ignoring the diel cycles) to obtain a linear fit and use the slope to estimate mean NCP over the time of the deployment, and then check how does that estimate compare with the mean of your daily estimates. One problem that I see with the current approach and that is obvious when looking at Fig 3 is that occasionally the mismatch between where one of the linear fits end to where the next one begin is large, which might introduce large uncertainty. For example, in Fig3, drifter 2, first day of DO2/Ar, your fitted curve is missing the initial drop in O2 at the beginning of the night. Actually you can see that at the end of that night the DO2/Ar is lower than at the end of the previous night, indicating net heterotrophy. In contrast, using the approach proposed in this manuscript the resulting NCP is instead 26 mmol C m-2 d-1.

L296-298 It is unclear to me how the authors estimate NCP over 3-hour intervals

L356 Under the conditions that there are other unaccounted POC losses the diel POC method is not suitable to estimate NCP. It might still be suitable to estimate GPP, if the POC loss is constant throughout the day.

L450-451 Could the increase in nutrients during the experiment indicate entrainment of nutrient-rich water?

L540-541 How do you explain the net decrease in POC if there was a developing diatom bloom?

L570-572 I am not sure what this means

L581-582 Could the decrease in Si indicate phytoplankton uptake rather than export?

L623-625 Again, if C export explains the difference in NCP estimates, why is POC-R not larger than O2/Ar-R?

Hide

ED: Reconsider after major revisions (20 Nov 2019) by Emilio Marañón

AR by Sarah Rosengard on behalf of the Authors (26 Mar 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (27 Mar 2020) by Emilio Marañón

RR by Anonymous Referee #2 (16 Apr 2020)

Suggestions for revision or reasons for rejection

The authors have made important changes in the manuscript and included a comparison of O2-derived and cp-derived GPP and CR rates to understand what is driving the decoupling of O2 and POC. I think this has greatly improved the manuscript, which tackles an important question that will be of interest to the scientific community. However, I do think that the discussion and conclusions could be better organized and more concise. I find the discussion hard to follow at times, and I think the authors could narrow down a bit better what the most plausible processes causing the observed discrepancies are (and this should be reflected in the conclusions).

As I view it, in drifter 1, the similarities in CR but discrepancies in GPP indicate that C export alone cannot explain GPP discrepancies. Granted that there could be enhanced export during daytime, but given how similar CR rates are it would mean that the export is pretty much halted at night (actually sometimes CR-O2/Ar is larger than CR-POC). A more plausible explanation for these observations is a combination of C export and DOC dynamics. The authors mention DOC production but mostly just as a “POC loss” term, whereas the POC method is really missing a combination of DOC production and DOC respiration. If there is net DOC production during the day from newly fixed C (see Karl et al., 1998) you would expect GPP-POC to be underestimated, and respiration of fresh DOC at night (missed by the POC method) would result in an underestimation of CR-POC. If C export (that would result in an overestimation of CR-POC) is of similar magnitude than DOC respiration, CR-POC would end up being of similar magnitude than CR-O2, but differences in GPP would be greater due to DOC production. The alternative explanation would be the proliferation of large diatoms not captured by the beam attenuation measurement, but the decrease in chla does not really support this idea.

I do not totally understand how the correlation of 3h-NCP from both methods (Figure 5b) indicates that there are sub-daily variations in POC losses, even though I agree that potential daily changes in export could contribute to the differences. A graph showing the difference between 3-hour DeltaO2/Ar and DeltaPOC as a function of time of the day might be more useful to prove this.

In drifter 2, I think that DOC dynamics alone could very well explain the differences observed. I found the discussion of the daily changes in heterotrophic biomass as a potential cause for the observed differences interesting, and I wonder whether this needs to be brought up before, with drifter 1, as it could also be affecting the differences in that case, combined with C export and DOC dynamics.

Because the authors did not measure net DOC accumulation, and the range in the reported fraction of NCP that goes into the DOC pool is quite wide, I do not think it is justifiable to estimate POC export based a chosen DOC/NCP value.

In summary I think this manuscript should be published, but I do recommend some work in fixing typos and mistakes, better organizing the text, and making it a bit more concise (particularly the discussion and conclusions).

Specific comments (line numbers refer to the document that includes the answer to reviewers and the tracked changes).

L479 Is a PQ of 1.1 is justifiable for NCP? If we assume that NCP approximates new production it is probably closer to 1.4.

Equations: It would be helpful to specify the units of each term as well as the direction of the fluxes. I understand that tD and tN sum 1 (day), in which case (tD + tN) does not need to be added to the denominator in equations 2c and 9c. Or for consistency should be added to the denominator in equations 2a and 9a

Eq1: shouldn’t Fmix be subtracted? (positive dO2/dz means an oxygen flux into the mixed layer).

Eq 2b why is the PQ used to convert CR? A RQ should be used instead.

Eq 3: The percent symbol (%) is not needed. O2/Ar should be defined somewhere

Eq 9a, 9b: why are these divided by the PQ? they are already in C units.

L780 This is a weird sentence for the results section, as the conclusions have not been presented yet.

L931-936 I find this paragraph confusing. Without looking at the discrepancies or similarity of GPP and CR you would be unable to know what is causing the discrepancies in NCP.

Table 1. I would delete POC export estimates as the authors did not measure DOC production. Also the numbers do not seem to match my calculations.

Table 2. Why is the Export+DOC column only added to drifter 1?

References
Karl,D.M.,D.V.Hebel,K.Bjorkman, and R. M. Letelier. 1998. The role of dissolved organic matter release in the productivity of the oligotrophic North Pacific Ocean. Limnol. Oceanogr. 43: 1270–1286

Hide

ED: Publish subject to minor revisions (review by editor) (19 Apr 2020) by Emilio Marañón

AR by Sarah Rosengard on behalf of the Authors (13 May 2020) Author's response Manuscript

ED: Publish as is (20 May 2020) by Emilio Marañón

AR by Sarah Rosengard on behalf of the Authors (22 May 2020) Manuscript

Download

Article (518 KB)
Full-text XML

Short summary

Net community production sets the maximum quantity of phytoplankton carbon available for the marine food web and longer-term storage in the deep ocean. We compared two approaches to estimate this critical variable from autonomous measurements of mixed-layer dissolved oxygen and particulate organic carbon, observing a significant discrepancy between estimates in an upwelling zone near the Oregon coast. We use this discrepancy to assess the fate of organic carbon produced in the mixed layer.

Decoupling of ΔO2∕Ar and particulate organic carbon dynamics in nearshore surface ocean waters

Download

Interactive discussion

Peer-review completion

Decoupling of ΔO₂∕Ar and particulate organic carbon dynamics in nearshore surface ocean waters