the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 May 2023
Carbon monoxide cycling in the Ria Formosa Lagoon (southern Portugal) during summer 2021
Damian L. Arévalo-Martínez
Riel Carlo O. Ingeniero
Hermann W. Bange
Abstract. Carbon monoxide (CO) is an atmospheric trace gas that plays a crucial role in the oxidizing capacity of the Earth’s atmosphere. Moreover, it functions as an indirect greenhouse gas, influencing the lifetimes of potent greenhouse gases such as methane. Albeit being an overall source of atmospheric CO, the role of coastal regions in the marine cycling of CO and how its budget can be affected by anthropogenic activities, remain uncertain. Here, we present the first measurements of dissolved CO in the Ria Formosa Lagoon, an anthropogenically influenced system in southern Portugal. The dissolved CO concentrations in the surface layer ranged from 0.16 to 3.1 nmol L−1 with an average concentration of 0.75 ± 0.57 nmol L−1. The CO saturation ratio ranged from 1.7 to 32.2, indicating that the lagoon acted as a source of CO to the atmosphere in May 2021. The estimated average sea-to-air flux density was 1.53 μmol m−2 d−1, mainly fueled by CO photochemical production. Microbial consumption accounted for 83 % of the CO production, suggesting that the resulting CO emissions to the atmosphere were modulated by microbial consumption in the surface waters of the Ria Formosa Lagoon. The results from an irradiation experiment with aquaculture effluent water indicated that aquaculture facilities in the Ria Formosa Lagoon seem to be a negligible source of atmospheric CO.
- Preprint
(2741 KB) - Metadata XML
-
Supplement
(305 KB) - BibTeX
- EndNote
Guanlin Li et al.
Status: closed
-
CC1: 'Comment on egusphere-2023-771', Hong-Hai Zhang, 01 Jun 2023
Interactive comment on “Carbon monoxide cycling in the Ria Formosa Lagoon (southern Portugal) during summer 2021” by Guanlin Li et al.
General comments:
The manuscript by Li et al. deals with a measured of the concentrations of carbon monoxide in the water column in the Ria Formosa Lagoon system and determine the potential impact of aquaculture activities on CO cycling in this region. It addresses an under-investigated topic, clearly of interest for the BG community, of general biogeochemical relevance as well as consequences for the role of coastal ecosystems for the (trans)formation of climate-relevant gases. The authors have very carefully exploited their data and provided the appropriate statistical evaluation to support their results, and extracted the main findings of this study. However, from my point of view, I have some suggestions to render the work more attractive to readers. Therefore, I suggest its publication after minor revisions.
Specific comments:
Page 4 Lines 107: The author mentioned in the manuscript that because of a technical problem with the calibration of the CO analyzer, data was corrected by the correction factor. It is not clear for me that the correction factor is for atmospheric samples or for all samples?
Page 9 Lines 263-264: as mentioned in the previous paragraphs ‘with the assumption of a steady state, the sum of the CO sources and sinks is equal to zero’. It should not be mentioned here again in the form of a conclusion. It is recommended to modify or delete.
For the microbial CO consumption experiment, should the influence of dark (thermodynamic) production on CO be considered?
should the sampling density be increased, especially between 0 and 24 hours? And the author mentioned that CO net production rate for Olhão aquaculture effluent is extremely low, microbial CO consumption was counteracting the CO photochemical production almost completely. Would the author consider using some testing methods to analyze the consumption of microorganisms to confirm the experimental results?
For the aquaculture incubation experiments, should the sampling density be increased, especially between 0 and 24 hours? And the author mentioned that CO net production rate for Olhão aquaculture effluent is extremely low, microbial CO consumption was counteracting the CO photochemical production almost completely. Would the author consider using some testing methods to analyze the consumption of microorganisms to confirm the experimental results?
The conclusion should not contain too many references.
Minor comments for the figures:
The quality of the figure should be improved (Fig. 2).
Fig. 5(a), the color for the right vertical axis for CO saturation ratio is oversaturated, I suggest changing to another color.
The legend of Figure 8 is not very clear.
Citation: https://doi.org/10.5194/egusphere-2023-771-CC1 - AC2: 'Reply on CC1', Guanlin Li, 25 Oct 2023
-
RC1: 'Comment on egusphere-2023-771', Anonymous Referee #1, 07 Jun 2023
General comments:
This manuscript reports a dataset comprising CO’s surface water concentrations, air-sea exchange fluxes, bacterial consumption rates and net production rates in the Ria Formosa Lagoon. It also presents a CO mass-balance budget. The authors conclude that the Ria Formosa Lagoon is a source of atmospheric CO and that photoproduction and microbial consumption are the main source and sink of CO in this system, respectively. This study has the potential to make a cumulative contribution to the understanding of air-sea fluxes and biogeochemical cycling of CO in the severely under-sampled coastal oceans.
However, I have a major concern on the quality of the data and thus the resulting conclusions. My concern primarily arises from the extremely sparse sampling resolutions: only one time point (late afternoon and high tide) on two days for CO concentration sampling and only one station (Sta. 7) for CO consumption sampling. It is well known that surface water CO concentration changes substantially over diel time scales mainly imposed by daytime photoproduction and bacterial consumption throughout day and night. This diel cycle can be further complicated by tidal variation in coastal areas. It is also well recognized that microbial CO consumption presents large spatiotemporal variations in coastal regions. The data obtained from such poor sampling resolutions can only capture a slim part of the diel cycle and thus cannot support the conclusions reported. For example, if we use the Kbio determined at Sta. 7 (0.40 h-1) to estimate the surface water CO concentration in early morning (before dawn, assuming 7 hours of nighttime) at this same locality, we get 0.04 nM, which is below the CO concentration at equilibrium with air (~0.1 nM). Then, the late-afternoon source turns into a nighttime sink. This situation may apply to many other stations, because the reported surface water CO concentrations were mostly quite low.
The second important factor that led to me casting doubt on the quality of the data is the huge correction factor (3.12) for CO concentrations. There is no explanation for what caused the technical problem leading to this correction. Any time- and concentration-dependence of the correction factor? The uncertainty of your atmospheric CO measurement was 14.2%, which alone leads to significant variations in the correction factor. This variability can be larger if the uncertainties of the atmospheric CO measurements at Mace Head and Terceira are taken into account. Please present an uncertainty analysis of this correction.
In addition, there are errors in the calculation of CO concentrations (eq. 1, see specific comments on this).
The CO budget (eq. 11) is unreliable because the poor sampling resolutions invalidate the stead-state assumption. Moreover, the budgeting misses a potentially important source (terrestrial input including the WWTP effluent) and a potentially important sink (output to the outside sea).
The interpretation of the data from the incubations of the aquaculture is superficial and often inaccurate or incorrect (see specific comments). In fact, the data allows you to derive the gross production in the “light” treatment.
Some suggestions:
- Make sure that the calculations of and corrections on CO concentrations are correct.
- Base your conclusions on the data. Do not over-extrapolate temporally and/or spatially.
Specific comments:
Introduction: Put your study in a broad context: why do you choose a human-impacted lagoon system?
L32: Ossola et al. (2022) used model compounds not natural CDOM and their experiments were conducted at pH 5.8-6.5, conditions atypical of coastal and open oceans. Replace it with a more appropriate reference.
Integrate “Study site description” into the last paragraph of the “Introduction”. Elaborate more about the human activities (including aquaculture) in the region.
L63-64: Give more details about how seawater was transferred from the Niskin bottles to the quartz glass bottles (e.g., the material of the tube used for sample transfer; if the bottles were overflowed; if sunlight was avoided during the transfer; what caps were used to close the bottles, etc.).
L72: Samples for dark incubation were collected only on one day? Then was it May 25 or 26?
L73: How many replicate incubations?
L74: “cooling box”—what was the temperature inside the box, well below the in situ temperature? If so, could significantly affect bacterial activity.
L75-78: Was temperature stable during the incubation? At what temperature? At or close to in situ temperature?
CO measurements: What was the precision and detection limit of the method.
L81: Did the 113.9 ppb standard cover the range of CO concentrations in the headspace of your samples? If not, what was the linear response range of the CO analyzer and was the range large enough to cover your samples’ headspace concentrations?
L85-89: Provide more details about how the headspace was created and how the headspace gas was subsampled (i.e., creation of headspace: how water was removed from and how gas was injected into the bottle? Subsampling the headspace gas: how the headspace gas was sampled? During both the creation and subsampling of the headspace gas, how the headspace was maintained at the atmospheric pressure?)
Equations 1-5: List units for all parameters. Is Vw the volume before making the headspace or after making the headspace? Beta in Wiesenburg and Guinasso (1979) is in mL gas (mL H2O)-1 atm-1. Moreover, the units of R are atm L mol-1 K-1. How did you get the units of nmol L-1 for COsurf (L94)?
Equation 1: The equation seems incorrect. The first term should be multiplied “P” and “Vw (volume of remaining water after the creation of headspace) then divided by the volume of the water before the creation of the headspace.
L104: “in equilibrium with the headspace”. Note COeq is the dissolved CO concentration in the water remaining after the creation of the headspace.
Equation 4 is the first term of equation 1. If Ceq here refers to the CO concentration in surface water in equilibrium with the atmosphere (see equation 6), then x’ here refers to the dry mole fraction of CO in the air samples. Do not mix this x’ with that in equation 1.
L107: Please explain what was the technical problem? The correction fact f = 3.12. This was huge!
L111-113: “nearly uniform atmospheric background CO mole fractions”. This is usually true over open oceans but not near the coast. The relative standard deviation of your measurements (5/35.1=14.2%) indicates this.
Equation 6. See comment on equation 4.
L124-129: Did you measure u2?
Equation 7: Flux densities were calculated using spot windspeeds? If not, how windspeeds were processed? (if averaged, give the relevant spatial and temporal scales).
Figure 3: Temperature, pH, and nutrients are shown in Figure 3 but not cited in the text.
Figure 2: What is “Stations 114” in the figure caption?
L180: Justify using one station (Sta. 18) to represent a region (Praia de Faro zone).
L188: CO concentration at Sta. 13 was among the lowest (Figures 2 and 4).
L188-189: “biological CO production”. This is purely speculative. FDOM at Sta. was high as well.
L189-191: “WWTP thermal effluent plume”. What was your purpose of mentioning this plume?
L196-197: n = 14 for the relationship between CO and phosphate but n = 28 for the relationship between CO and FDOM. Why? Any implication for the significant correlation between CO and phosphate?
L201-202: “The lagoon was a source of atmospheric CO in May 2021”. This is a premature conclusion since you only sampled on two days, each at a specific time point and tidal level (see general comment). Better to say “The lagoon was source of atmospheric CO at the time of sampling”.
Figures 3, 4, and 5(b) and Table 1: Samples were collected on two days. Are the data shown in these figures the average of the two samples? If averaged, what are the variabilities of the two sampling times? Justify the use of average values.
Figure 8: What do you mean by “CDOM absorption factor”?
L266-267: Why CO concentration did not decrease during night? Hard to understand. Bacteria were killed by solar UV during daytime?
L267-269: In fact, the dark incubation roughly followed an exponential decay trend. No evidence suggests that a steady state reached at the end of the incubation.
L284-289: Much of the discussion is purely speculative.
L291-293: Such a comparison does not mean much because you measured net production while the others reported photoproduction.
Technical corrections:
L16: remove “of the Ria Formosa Lagoon”.
L21: replace “it” with “this reaction”.
L27: remove “may”.
L28: “early” à “earlier”; “indicate”à”reported”.
L125: at 10 m.
Equation 9: define u2.
L161: in the Ria…
L175: Show the Ramalhete and Faro–Olhão inlets in the maps of Figure 3.
L176: replace first “.” with “,”.
Figure 4: Caption is incorrect (now the same as that of Figure 3). Chl-a is missing.
L197: add “,” after “r = 0.860”.
Figure 8: Absence of symbol legends. What do you mean by “CDOM absorption factor”?
Citation: https://doi.org/10.5194/egusphere-2023-771-RC1 - AC1: 'Reply on RC1', Guanlin Li, 25 Oct 2023
-
RC2: 'Comment on egusphere-2023-771', Anonymous Referee #2, 06 Oct 2023
** Major comments **
The authors report a data-set of dissolved CO concentration at 14 stations in Ria Formosa Lagoon collected on 25 and 26 May 2021 (n=28). The data are discussed in relation to basic environmental variables (temperature, salinity, FDOM) measured with a EXO-2 YSI probe. They made one incubation of CO photo-production.
The CO data were corrected by a factor of 3 due to analytical problem with the CO Gas Analyzer (that was not specified). This correction factor was computed by comparing atmospheric CO with data from atmospheric monitoring stations Mace Head (Ireland) and Terceira (Azores). While it is laudable that the authors are transparent about this correction, I think this is a problem because CO concentrations are quite close to equilibrium (saturation ratio on average of <8) so uncertainty on the dissolved concentration propagates to a large error on the saturation ratio.
The authors derived the FDOM data from the YSI EXO-2 when they had access to a UV-Vis Spectrophotometer (as stated) so they could have made spectrophotometric measurements of CDOM, with little extra consumables and workload. This would have been much more precise because my experience with the YSI EXO-2 is that the FDOM sensor tends to provide erratic measurements due to the interaction of scattering light on the sensors and presence of suspended particles in water. I would expect in a lagoon environment quite a lot of these interferences, I would not trust the FDOM data from the sensor.
While the Chl-a sensor of the YSI EXO-2 performs better than FDOM sensor, it is still not optimal (compares poorly with discrete Chla measurements), and again the authors could have gone into the trouble of measuring a relatively basic variable such as Chla concentration on 28 samples. The Chl-a is not just a descriptive variable but is used to calculate the CO production by phytoplankton in the mass budget.
The production of CO in the dark was only estimated on one station at the effluent of the aquaculture facility. While I see from the M&M that these incubations require some work, I think it could have been feasible to make more of these incubations in different water masses of the Ria Formosa Lagoon, since the authors had all of this apparatus and equipment on site.
In conclusion, I find that the sampling effort of CO concentrations was “light” (only 14 stations during 2 consecutive days). It could have been useful to make surveys during low tide, or during other seasons. If the production of CO is mostly due to photo-production, it would have made sense to make night-day cycles. Indeed, CO concentrations change by a factor of 5 during night-day oscillations (Ohta 1997), so potentially a much stronger signal that the variability across the data-set.
** Minor comments **
In L200: Please note that the saturation ratio is proportional to the CO concentration because the saturation ratio is computed from the CO concentration, so this finding is meaningless; refer to Berges (1997).
The authors mention they took samples in the late afternoon (~17:00h local time), so well after the maximum of daily irradiance (when CO would have been expected to be max. I imagine that the 14 stations spread over a distance of about 4km were not sampled instantaneously. Can you specify the time of sampling ?
The sample abbreviation (T) is used for temperature either in °C or K
The authors should explain why they sampled in quartz glass bottles ?
The authors should explain how the quartz glass bottles were closed gas tight. With stoppers ?
** References **
Berges JA (1997) Ratios, regression statistics, and “spurious” correlations, Limnol. Ocemrqr., 42(5). 1006-1007
Ohta K (1997), Diurnal Variations of Carbon Monoxide Concentration in the Equatorial Pacific Upwelling Region. Journal of Oceanography, 53, 173-178.
Citation: https://doi.org/10.5194/egusphere-2023-771-RC2 - AC3: 'Reply on RC2', Guanlin Li, 25 Oct 2023
Status: closed
-
CC1: 'Comment on egusphere-2023-771', Hong-Hai Zhang, 01 Jun 2023
Interactive comment on “Carbon monoxide cycling in the Ria Formosa Lagoon (southern Portugal) during summer 2021” by Guanlin Li et al.
General comments:
The manuscript by Li et al. deals with a measured of the concentrations of carbon monoxide in the water column in the Ria Formosa Lagoon system and determine the potential impact of aquaculture activities on CO cycling in this region. It addresses an under-investigated topic, clearly of interest for the BG community, of general biogeochemical relevance as well as consequences for the role of coastal ecosystems for the (trans)formation of climate-relevant gases. The authors have very carefully exploited their data and provided the appropriate statistical evaluation to support their results, and extracted the main findings of this study. However, from my point of view, I have some suggestions to render the work more attractive to readers. Therefore, I suggest its publication after minor revisions.
Specific comments:
Page 4 Lines 107: The author mentioned in the manuscript that because of a technical problem with the calibration of the CO analyzer, data was corrected by the correction factor. It is not clear for me that the correction factor is for atmospheric samples or for all samples?
Page 9 Lines 263-264: as mentioned in the previous paragraphs ‘with the assumption of a steady state, the sum of the CO sources and sinks is equal to zero’. It should not be mentioned here again in the form of a conclusion. It is recommended to modify or delete.
For the microbial CO consumption experiment, should the influence of dark (thermodynamic) production on CO be considered?
should the sampling density be increased, especially between 0 and 24 hours? And the author mentioned that CO net production rate for Olhão aquaculture effluent is extremely low, microbial CO consumption was counteracting the CO photochemical production almost completely. Would the author consider using some testing methods to analyze the consumption of microorganisms to confirm the experimental results?
For the aquaculture incubation experiments, should the sampling density be increased, especially between 0 and 24 hours? And the author mentioned that CO net production rate for Olhão aquaculture effluent is extremely low, microbial CO consumption was counteracting the CO photochemical production almost completely. Would the author consider using some testing methods to analyze the consumption of microorganisms to confirm the experimental results?
The conclusion should not contain too many references.
Minor comments for the figures:
The quality of the figure should be improved (Fig. 2).
Fig. 5(a), the color for the right vertical axis for CO saturation ratio is oversaturated, I suggest changing to another color.
The legend of Figure 8 is not very clear.
Citation: https://doi.org/10.5194/egusphere-2023-771-CC1 - AC2: 'Reply on CC1', Guanlin Li, 25 Oct 2023
-
RC1: 'Comment on egusphere-2023-771', Anonymous Referee #1, 07 Jun 2023
General comments:
This manuscript reports a dataset comprising CO’s surface water concentrations, air-sea exchange fluxes, bacterial consumption rates and net production rates in the Ria Formosa Lagoon. It also presents a CO mass-balance budget. The authors conclude that the Ria Formosa Lagoon is a source of atmospheric CO and that photoproduction and microbial consumption are the main source and sink of CO in this system, respectively. This study has the potential to make a cumulative contribution to the understanding of air-sea fluxes and biogeochemical cycling of CO in the severely under-sampled coastal oceans.
However, I have a major concern on the quality of the data and thus the resulting conclusions. My concern primarily arises from the extremely sparse sampling resolutions: only one time point (late afternoon and high tide) on two days for CO concentration sampling and only one station (Sta. 7) for CO consumption sampling. It is well known that surface water CO concentration changes substantially over diel time scales mainly imposed by daytime photoproduction and bacterial consumption throughout day and night. This diel cycle can be further complicated by tidal variation in coastal areas. It is also well recognized that microbial CO consumption presents large spatiotemporal variations in coastal regions. The data obtained from such poor sampling resolutions can only capture a slim part of the diel cycle and thus cannot support the conclusions reported. For example, if we use the Kbio determined at Sta. 7 (0.40 h-1) to estimate the surface water CO concentration in early morning (before dawn, assuming 7 hours of nighttime) at this same locality, we get 0.04 nM, which is below the CO concentration at equilibrium with air (~0.1 nM). Then, the late-afternoon source turns into a nighttime sink. This situation may apply to many other stations, because the reported surface water CO concentrations were mostly quite low.
The second important factor that led to me casting doubt on the quality of the data is the huge correction factor (3.12) for CO concentrations. There is no explanation for what caused the technical problem leading to this correction. Any time- and concentration-dependence of the correction factor? The uncertainty of your atmospheric CO measurement was 14.2%, which alone leads to significant variations in the correction factor. This variability can be larger if the uncertainties of the atmospheric CO measurements at Mace Head and Terceira are taken into account. Please present an uncertainty analysis of this correction.
In addition, there are errors in the calculation of CO concentrations (eq. 1, see specific comments on this).
The CO budget (eq. 11) is unreliable because the poor sampling resolutions invalidate the stead-state assumption. Moreover, the budgeting misses a potentially important source (terrestrial input including the WWTP effluent) and a potentially important sink (output to the outside sea).
The interpretation of the data from the incubations of the aquaculture is superficial and often inaccurate or incorrect (see specific comments). In fact, the data allows you to derive the gross production in the “light” treatment.
Some suggestions:
- Make sure that the calculations of and corrections on CO concentrations are correct.
- Base your conclusions on the data. Do not over-extrapolate temporally and/or spatially.
Specific comments:
Introduction: Put your study in a broad context: why do you choose a human-impacted lagoon system?
L32: Ossola et al. (2022) used model compounds not natural CDOM and their experiments were conducted at pH 5.8-6.5, conditions atypical of coastal and open oceans. Replace it with a more appropriate reference.
Integrate “Study site description” into the last paragraph of the “Introduction”. Elaborate more about the human activities (including aquaculture) in the region.
L63-64: Give more details about how seawater was transferred from the Niskin bottles to the quartz glass bottles (e.g., the material of the tube used for sample transfer; if the bottles were overflowed; if sunlight was avoided during the transfer; what caps were used to close the bottles, etc.).
L72: Samples for dark incubation were collected only on one day? Then was it May 25 or 26?
L73: How many replicate incubations?
L74: “cooling box”—what was the temperature inside the box, well below the in situ temperature? If so, could significantly affect bacterial activity.
L75-78: Was temperature stable during the incubation? At what temperature? At or close to in situ temperature?
CO measurements: What was the precision and detection limit of the method.
L81: Did the 113.9 ppb standard cover the range of CO concentrations in the headspace of your samples? If not, what was the linear response range of the CO analyzer and was the range large enough to cover your samples’ headspace concentrations?
L85-89: Provide more details about how the headspace was created and how the headspace gas was subsampled (i.e., creation of headspace: how water was removed from and how gas was injected into the bottle? Subsampling the headspace gas: how the headspace gas was sampled? During both the creation and subsampling of the headspace gas, how the headspace was maintained at the atmospheric pressure?)
Equations 1-5: List units for all parameters. Is Vw the volume before making the headspace or after making the headspace? Beta in Wiesenburg and Guinasso (1979) is in mL gas (mL H2O)-1 atm-1. Moreover, the units of R are atm L mol-1 K-1. How did you get the units of nmol L-1 for COsurf (L94)?
Equation 1: The equation seems incorrect. The first term should be multiplied “P” and “Vw (volume of remaining water after the creation of headspace) then divided by the volume of the water before the creation of the headspace.
L104: “in equilibrium with the headspace”. Note COeq is the dissolved CO concentration in the water remaining after the creation of the headspace.
Equation 4 is the first term of equation 1. If Ceq here refers to the CO concentration in surface water in equilibrium with the atmosphere (see equation 6), then x’ here refers to the dry mole fraction of CO in the air samples. Do not mix this x’ with that in equation 1.
L107: Please explain what was the technical problem? The correction fact f = 3.12. This was huge!
L111-113: “nearly uniform atmospheric background CO mole fractions”. This is usually true over open oceans but not near the coast. The relative standard deviation of your measurements (5/35.1=14.2%) indicates this.
Equation 6. See comment on equation 4.
L124-129: Did you measure u2?
Equation 7: Flux densities were calculated using spot windspeeds? If not, how windspeeds were processed? (if averaged, give the relevant spatial and temporal scales).
Figure 3: Temperature, pH, and nutrients are shown in Figure 3 but not cited in the text.
Figure 2: What is “Stations 114” in the figure caption?
L180: Justify using one station (Sta. 18) to represent a region (Praia de Faro zone).
L188: CO concentration at Sta. 13 was among the lowest (Figures 2 and 4).
L188-189: “biological CO production”. This is purely speculative. FDOM at Sta. was high as well.
L189-191: “WWTP thermal effluent plume”. What was your purpose of mentioning this plume?
L196-197: n = 14 for the relationship between CO and phosphate but n = 28 for the relationship between CO and FDOM. Why? Any implication for the significant correlation between CO and phosphate?
L201-202: “The lagoon was a source of atmospheric CO in May 2021”. This is a premature conclusion since you only sampled on two days, each at a specific time point and tidal level (see general comment). Better to say “The lagoon was source of atmospheric CO at the time of sampling”.
Figures 3, 4, and 5(b) and Table 1: Samples were collected on two days. Are the data shown in these figures the average of the two samples? If averaged, what are the variabilities of the two sampling times? Justify the use of average values.
Figure 8: What do you mean by “CDOM absorption factor”?
L266-267: Why CO concentration did not decrease during night? Hard to understand. Bacteria were killed by solar UV during daytime?
L267-269: In fact, the dark incubation roughly followed an exponential decay trend. No evidence suggests that a steady state reached at the end of the incubation.
L284-289: Much of the discussion is purely speculative.
L291-293: Such a comparison does not mean much because you measured net production while the others reported photoproduction.
Technical corrections:
L16: remove “of the Ria Formosa Lagoon”.
L21: replace “it” with “this reaction”.
L27: remove “may”.
L28: “early” à “earlier”; “indicate”à”reported”.
L125: at 10 m.
Equation 9: define u2.
L161: in the Ria…
L175: Show the Ramalhete and Faro–Olhão inlets in the maps of Figure 3.
L176: replace first “.” with “,”.
Figure 4: Caption is incorrect (now the same as that of Figure 3). Chl-a is missing.
L197: add “,” after “r = 0.860”.
Figure 8: Absence of symbol legends. What do you mean by “CDOM absorption factor”?
Citation: https://doi.org/10.5194/egusphere-2023-771-RC1 - AC1: 'Reply on RC1', Guanlin Li, 25 Oct 2023
-
RC2: 'Comment on egusphere-2023-771', Anonymous Referee #2, 06 Oct 2023
** Major comments **
The authors report a data-set of dissolved CO concentration at 14 stations in Ria Formosa Lagoon collected on 25 and 26 May 2021 (n=28). The data are discussed in relation to basic environmental variables (temperature, salinity, FDOM) measured with a EXO-2 YSI probe. They made one incubation of CO photo-production.
The CO data were corrected by a factor of 3 due to analytical problem with the CO Gas Analyzer (that was not specified). This correction factor was computed by comparing atmospheric CO with data from atmospheric monitoring stations Mace Head (Ireland) and Terceira (Azores). While it is laudable that the authors are transparent about this correction, I think this is a problem because CO concentrations are quite close to equilibrium (saturation ratio on average of <8) so uncertainty on the dissolved concentration propagates to a large error on the saturation ratio.
The authors derived the FDOM data from the YSI EXO-2 when they had access to a UV-Vis Spectrophotometer (as stated) so they could have made spectrophotometric measurements of CDOM, with little extra consumables and workload. This would have been much more precise because my experience with the YSI EXO-2 is that the FDOM sensor tends to provide erratic measurements due to the interaction of scattering light on the sensors and presence of suspended particles in water. I would expect in a lagoon environment quite a lot of these interferences, I would not trust the FDOM data from the sensor.
While the Chl-a sensor of the YSI EXO-2 performs better than FDOM sensor, it is still not optimal (compares poorly with discrete Chla measurements), and again the authors could have gone into the trouble of measuring a relatively basic variable such as Chla concentration on 28 samples. The Chl-a is not just a descriptive variable but is used to calculate the CO production by phytoplankton in the mass budget.
The production of CO in the dark was only estimated on one station at the effluent of the aquaculture facility. While I see from the M&M that these incubations require some work, I think it could have been feasible to make more of these incubations in different water masses of the Ria Formosa Lagoon, since the authors had all of this apparatus and equipment on site.
In conclusion, I find that the sampling effort of CO concentrations was “light” (only 14 stations during 2 consecutive days). It could have been useful to make surveys during low tide, or during other seasons. If the production of CO is mostly due to photo-production, it would have made sense to make night-day cycles. Indeed, CO concentrations change by a factor of 5 during night-day oscillations (Ohta 1997), so potentially a much stronger signal that the variability across the data-set.
** Minor comments **
In L200: Please note that the saturation ratio is proportional to the CO concentration because the saturation ratio is computed from the CO concentration, so this finding is meaningless; refer to Berges (1997).
The authors mention they took samples in the late afternoon (~17:00h local time), so well after the maximum of daily irradiance (when CO would have been expected to be max. I imagine that the 14 stations spread over a distance of about 4km were not sampled instantaneously. Can you specify the time of sampling ?
The sample abbreviation (T) is used for temperature either in °C or K
The authors should explain why they sampled in quartz glass bottles ?
The authors should explain how the quartz glass bottles were closed gas tight. With stoppers ?
** References **
Berges JA (1997) Ratios, regression statistics, and “spurious” correlations, Limnol. Ocemrqr., 42(5). 1006-1007
Ohta K (1997), Diurnal Variations of Carbon Monoxide Concentration in the Equatorial Pacific Upwelling Region. Journal of Oceanography, 53, 173-178.
Citation: https://doi.org/10.5194/egusphere-2023-771-RC2 - AC3: 'Reply on RC2', Guanlin Li, 25 Oct 2023
Guanlin Li et al.
Guanlin Li et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 288 | 97 | 33 | 418 | 28 | 9 | 10 |
- HTML: 288
- PDF: 97
- XML: 33
- Total: 418
- Supplement: 28
- BibTeX: 9
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1