the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
The impact of coral reef ecosystems and upwelling events on the marine carbon dynamics of Southern Taiwan
Pei-Jie Meng
Chia-Ming Chang
Hung-Yen Hsieh
Anderson B. Mayfield
Abstract. The ocean is the largest carbon reservoir and plays a crucial role in regulating atmospheric CO2 levels, especially in the face of climate change. In coral reef ecosystems, the complexity and importance of the carbonate system must be better appreciated as atmospheric CO2 concentrations continue to rise. This study measured pCO2 over time and space in Nanwan Bay, a coral reef ecosystem in Southern Taiwan, to identify factors that influence its variation. The results showed that mean pCO2 values varied seasonally, with values of 394 µatm in spring, 406 µatm in summer, 399 µatm in fall, and 367 µatm in winter. These seasonal differences (ΔpCO2) were -2, 14, 7, and -29 µatm, respectively. These findings suggest that the Nanwan Bay coral reef ecosystem acts as a sink for atmospheric CO2 during the spring and winter, with an average sea-air gas flux of -1 gC m-2 year-1 and a net annual uptake of -29 t. The carbonaceous parameters of the surface water in this high-biodiversity sub-tropical marine ecosystem were influenced not only by seasonal temperature variation but also by vertical mixing, intermittent upwelling, and biological effects.
- Preprint
(2099 KB) - Metadata XML
- BibTeX
- EndNote
Pei-Jie Meng et al.
Status: open (extended)
-
RC1: 'Comment on egusphere-2023-1097', Anonymous Referee #1, 02 Aug 2023
reply
The manuscript discusses the carbonate system measurements in Nanwan Bay, Taiwan, and in particular focuses on the processes that influence pCO2 in Nanwan Bay and seasonal variability in whether the bay is a source or a sink.
I have concerns about the methods used in this study. The study uses the gas exchange relationship FGAS = k × KH × (pCO2,seawater−pCO2,air), and so conclusions about the magnitude of the air-sea gas exchange are highly dependent on pCO2,seawater, pCO2,air, and the gas exchange rate k which is in turn highly dependent on wind speed. I believe that there are serious issues that need to be addressed with each of these parameters.
pCO2,seawater
The pCO2,seawater values used in this study were calculated from pH and TA measurements, however these data are not presented. pH measurements were recorded on two different scales, and it was not explained why, which measurements were made on which scale, or how this may have influenced any of the results.
It is not clear how the values obtained during the study were averaged to obtain the mean pCO2,seawater values for each site. This is problematic because the sites are not evenly spaced, and so a simple arithmetic mean would result in certain areas (ie., around the nuclear power plant outflow) being overweighted, while others would be underweighted.
It is also stated that measurements were taken on particular days, but it does not say how many times these measurements were taken. It appears that there were at least three measurements at S10 during many of the cruises, with these three measurements sometimes showing distinct variations in surface pCO2 of up to ~ 50 µatm during the same day, and greater than this at depth. These large variations call into question how representative the values presented are of the system as a whole, and some discussion is needed to justify why the numbers here would be representative.
pCO2,air
This study uses pCO2,air values measured at Dongsha Island, a remote island ~500km away from Taiwan and 250 km away from the nearest landmass. As such, I think there needs to be some justification as to why these values are representative of Nanwan Bay. Could there be anthropogenic influences to the pCO2 near Nanwan Bay that do not influence Dongsha Island? Could the seasonal terrestrial signal have a greater impact on Nanwan Bay than Dongsha Island? A sensitivity analysis to determine how much influence variations in pCO2,air values would have on the calculated fluxes would be beneficial.
Wind speed
This study uses the seasonal average wind speed to calculate the seasonal air-sea gas exchange for Nanwan Bay, however I think there needs to be more justification about why this is reasonable. The authors formulation of k is quadratically dependent on wind speed, and so a few relatively brief high wind speed events could result in the gas exchange being underestimated if only the average wind speed is used. Data showing that the wind speed in the bay is highly consistent, or some work to characterize how such occurrences would alter the estimated air-sea fluxes would be beneficial.
These are my biggest concerns about the paper, however there were several other issues that I think need to be addressed. I have included more specific comments below.
Title – The paper doesn’t discuss the impact of coral reef ecosystems on the marine carbon dynamics of Nanwan Bay.
Line 51 – I think this statement could benefit from a reference.
Line 57 – I think being more specific about exactly what is meant by hydrological characteristics would be useful.
Line 70 – I would use conversely rather than similarly as the data show the opposite trend to the previously mentioned
Figure 1 – The writing on the inset is too small and too similar in color to the rest of the inset, making it illegible. As the goal of this inset seems to be indicating where Nanwan Bay is with respect to Taiwan, I would recommend using a figure that just showed land and sea, rather than depth, as it’s hard to tell where Taiwan is. If the goal is to also show bathymetry over a larger area, I think the inset needs to be bigger. The color bars for both the figure and the inset need labels, and I would encourage the authors to use a more colorblind friendly color map. The ‘x’ marks denoting the sampling sites should be a different color to the underlying bathymetry.
Line 85 – I’d consider removing ‘may’ – carbonate dynamics on coral reefs typically vary substantially.
Lines 90-93 – This sentence is unclear. I think the authors should make it more explicit how physiological changes in resident organisms in response to environmental change can in turn affect seawater carbon levels.
Lines 94-108 – This reads as a site description and could potentially be moved to methods.
Line 101 – I would like more detail here regarding what these habitats are and what their relative proportions are.
Line 103 – What specific impacts do these upwelling events have on temperature and nutrients?
Lines 111-113 – It’s not clear if this sentence is referring to Nanwan Bay or upwelling regions in general. What causes some upwelling regions to be sources and some to be sinks?
Line 113 – I would advise the authors to reorder this section here and start by talking about the mechanism by which productivity alters carbon cycling, and then focus on how an increase in nutrients can enhance these effects. I’d also consider replacing basic productivity with primary productivity throughout the paper, as I believe it’s a more widely used term.
Lines 117-121 – As far as I can tell, the authors have not characterized the P/R ratio of the bay but instead use calculated pCO2 values to determine if it’s a sink or a flux. As such, I’d suggest deleting this sentence, or changing it to discuss how pCO2 gradients determine if the ocean is a sink or a source. I’m also not sure what is meant by a range of biogeochemical processes – to me this paper does not constrain biogeochemical processes.
Lines 125-127 – More information needs to be provided about the sampling. How many times was each station sampled each day? What time of day was each sample taken at? Which depths were the samples taken at each station? I think a table reporting this would be greatly beneficial.
Lines 130 & 127 – How were these accuracies determined? Are these the factory standard accuracies?
Lines 141-149 – It is not clear to me why pH values are being reported on two different scales.
Line 157 – Very. Minor, but I believe CO2SYS is spelt without a subscript. I’d also encourage the authors to indicate which version of CO2SYS they have used.
Line 160 – which measured pH scale was used, NBS or total? Both?
Line 162 – I believe Dickson and Millero (1987) refit the values from Mehrbach et al. (1973)
Line 170 – I would keep this section focused on the equation and say where the data came from later on.
Line 173 – I would include a citation on this equation.
Lines 177-180 – I believe more justification is needed here as to why samples measured at Dongsha Island are applicable to Nanwan Bay, given that these values directly influence the direction and magnitude of the air-sea CO2 flux. Nanwan Bay’s higher proximity to land may mean that anthropogenic and terrestrial effects (i.e., effects from terrestrial growing seasons) alter its pCO2 dynamics in comparison to Dongsha. pCO2 data from somewhere on Taiwan that demonstrated a similar trend to that observed on Dongsha would strengthen this argument.
COME BACK TO LINES 189 TO 207
Table 1 – It would be good to see timeseries of these data either in the paper or in supplementary materials, rather than just this correlation matrix.
Line 218-220 – I’m afraid I don’t totally understand why these findings indicate vertical mixing in spring and winter and upwelling in spring and would ask the authors to explain this more.
Lines 222-224 – I think the connection between TA and these factors needs to be drawn out more, particularly since riverine outflow following rainfall can have high TA values and so increase TA even as it freshens water. I also think the paper would benefit by discussing specifics related to this site – are there any big rivers that flow into it? What’s the seasonal rainfall cycle like?
Figure 3 – All the font sizes in this figure need to be increased. The color bar should be labeled.
Lines 231-233 – I’m assuming the means calculated here are spatial, in which case I think care needs to be taken to account for the different distances between sampling points. The closeness of S31 and S33 mean that area will be overweighted if a mean is calculated without any weighting, while the lack of points near S15 would mean that this area will be underweighted. Given that these mean values are what are used to calculate the air-sea flux, I think the authors have to address this, and at the very least need to be clearer about how mean values are calculated. I would also stress that only surface water values were used, and say at what depth surface water values were obtained.
Figure 4 – The font size needs to increased throughout these figures. This figure also implies that each site was sampled multiple times per day – this needs to be mentioned and described in the methods.
Lines 240-244 – I believe the authors have the reasoning here backwards. Vertical variation is an indication of stratification, or a poorly mixed water column, while well mixed water masses tend to have constant properties throughout the water column.
Figures 5-8 – These figures could potentially be collapsed into one by calculating an average profile for each parameter each season, then plotting those average profiles on a single figure.
Figures 5-8 – These figures show pretty high surface pCO2 variability over the course of a day. How is that being accounted for when calculating average CO2 values? Are similar levels of CO2 variability present at other sites?
Lines 272-276, and Figure 10 – I think what pCO2 are at Tobs and Tmean could be made clearer. As far as I understand, pCO2 at Tobs is the pCO2 value you would measure at a given temperature if pCO2 if temperature was the only factor affecting it, while pCO2 at Tmean would be the pCO2 you would get by normalizing observations to the annual mean temperature. To me, these alone wouldn’t be the temperature and non-temperature effects. Instead, the temperature effects would be pCO2 at Tobs – pCO2 at Tmean, while the non-temperature effects would be the difference between the measured pCO2 and the pCO2 at Tobs. If this is the case, then I think nT – T just gives you the range between these values – it doesn’t tell you about the relative influence of temperature or non-temperature effects. You’d get the same nT – T if the nT value was 5 µatm higher than the mean and the T value was 35 µatm lower than the mean as you would if the nT value was 20 µatm higher than the mean and the T value was 20 µatm lower, but the influence of each effect would not be the same. This would affect a large amount of the analysis that follows (e.g. lines 297-299).
Line 277 – Is this a spatial mean rather than the mean in the equation above? The annual mean doesn’t seem like it should be changing over time. This is covered in the caption to Figure 10, but I think it should be included in the main text.
Lines 281-283 – Why do you believe that? What are possible non-temperature effects? Why would they act in the other direction to temperature?
Lines 283-286 – Why would consistently higher water temperatures mean that pCO2 is more variable? Seems like more variable temperatures would lead to greater pCO2 variability.
Figure 10, and other figures with multiple axes – I think the figures could be made a bit clearer if the axes were colored to match their data. You could change the color of the nT-T and right axis to match each other.
Figure 11 – It’s not super clear what’s meant here by control factors, or where the standard deviations have come from.
Lines 299-304 – I think there needs to be some discussion of why Chl a influences pCO2 here if the authors are going to claim that Chl a is one of the main factors affecting pCO2 in autumn. Why would an increase in Chl a increase pCO2? It seems plausible that if there was an increase in Chl a, then you’d expect a lower pCO2 due to an increase in photosynthesis. Conversely, could it be that an increase in pCO2 increases Chl a?
Line 300 – This is the first time Chl a has been mentioned. How was it measured? Where was it measured?
Figure 12 a & b – Both of the significant relationships seem to be heavily influenced by single points at higher temperatures and Chl a concentration than the remaining points. Are the fits still significant without those points?
Lines 312-314 – Doesn’t Fig. 12c show pCO2 at Tmean rather than at the actual temperature? If it is showing it at actual temperature, then what is 12a showing? I’d also recommend replacing actual temperature with measured temperature, which is how it’s been used previously.
Lines 313-316 – I don’t think Chl a itself is an effect, and so I think the authors need to explain the connection further.
Lines 319-323 – It’s not clear how this relates to the results of this paper. Is DNC calculated somewhere? How do measured changes in the mentioned parameters compare to what would be expected from DNC? How would these changes alter the chemical and biological processes of the surface water?
Line 324 – I think this should be more specific about what exactly the benthic environment of Nanwan Bay is, what is being regenerated, and what processes are leading to it being regenerated.
Lines 324-335 – This paragraph seems a little out of place as nutrients have not really been discussed to this point. I think it’s a worthwhile addition, but I think it needs some information about how nutrients would influence pCO2.
Line 338 - ∆pCO1 > 0 doesn’t necessarily mean the seawater is supersaturated with CO2, just that CO2 will move from the seawater to the atmosphere.
Lines 353-356 – I think there needs to be some justification as to why it’s reasonable to use the mean values. The gas exchange rate has a quadratic dependence on wind speed, so if there are a few wind events where the maximum wind speed is much greater than the mean wind speed then using the mean could lead to a large underestimation of the gas exchange. As noted in lines 362-363, wind speed is a crucial factor in determining air-sea gas exchange, so I think more justification is required as to why the cited speeds are reasonable.
I also again think that there needs to be more information about how the mean surface pCO2 values are calculated – how are they averaged both temporally and spatially? There also appears to be something of a diel cycle in the surface pCO2 values (e.g., Fig 6d) – how would this influence the variability in the gas exchange rates?
Lines 367-373 – What is different about this reef that might make it a sink rather than a source?
Lines 373-374 – What are the land-based inputs to Nanwan Bay? What effect might they have?
Lines 375-379 – As far as I understand, the wind speed measurements in this paper come from buoys. If this is the case, why is the difference between ship-based and satellite-based estimates of gas exchange relevant to this study?
Citation: https://doi.org/10.5194/egusphere-2023-1097-RC1
Pei-Jie Meng et al.
Pei-Jie Meng et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 173 | 64 | 13 | 250 | 7 | 5 |
- HTML: 173
- PDF: 64
- XML: 13
- Total: 250
- BibTeX: 7
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1