Assessing the Effects of Ocean Warming and Acidification on the Seagrass Thalassia hemprichii
Round 1
Reviewer 1 Report
The manuscript provides results of the large mesocosm experiment of ocean acidification and increased water temperature on a wide range of seagrass parameters. The main aim is to understand the extent to which climate change may alter the carbon sink capacity of seagrass beds in the tropical Indo-Pacific. In the time of rapid climate change and shifting attention to the ocean role in the climate warming process, the results of this study should be interesting for wider scientific community. Especially, as they target one of the most important blue carbon sinks – seagrass.
The conducted experiment and performed statistical analysis seem to be satisfactory however, there are few issues that has to be resolved before this paper can be published.
First, I wonder why the highest temperature tested was only 31 °C. In the discussion the authors cite themselves that “T. hemprichii is a tropical seagrass species, and it can thrive at temperatures up to 33ºC”. As massive seagrass die-offs were noted above 31 °C (34°C in Florida) I question the experimental design. If the authors were seeking to answer what is the potential effect of increased temperatures on seagrass, a higher range of elevated temperatures should be included in the experiment.
Moreover, the stated hypothesis is weak. The authors expect that OA and temperature raise will influence seagrass, but they do not state what influence they expect on the seagrass parameters – increase or decrease.
Some of the results are impossible to be checked because the Table 1 is impossible to read - see my comments below. Therefore, this must be corrected before it can be reviewed.
Finally, the discussion is divided into two parts – one treating OA effect on seagrass and the other the temperature effect on seagrass. There are few weak points in both parts (see below) and I feel like the second part shows a mix of results for both OA and elevated temperature. This is a little bit confusing to read. I suggest that maybe there could be a third paragraph that talks about adverse effect of OA and temperature rise. Therefore, I strongly suggest rewriting the discussion because in the current stage, it is confusing for the reader.
There are few grammatical and editing errors therefore, I strongly suggest that the authors take some additional time for grammatical corrections. Some of them are noted in my review below.
Specific comments:
Line 56: “gaining of” – should be “gaining an understanding”
Line 70-75 – Those two sentences that give an example of biodiversity mitigating the effects of environmental change and the giant kelp example seems out of place in this paragraph. Or perhaps I don’t understand them. How does uptake of large quantities of nutrients by kelp forests correspond to the seagrass CO2 uptake? I suggest looking for examples of seagrass communities or rewriting those sentences to make them clearer.
Line 78 – It is hypothesized that the OA and elevated temperatures will influence a range of seagrass parameters. In my opinion, it should be stated in the hypothesis how the OA and temperature will influence seagrass – does the author expect an increase or decrease in tested parameters?
Methods – Line 146 – The authors speak here about the functional groups but what I can see is just taxa listed (fish, bivalves, gastropods, etc.). Those taxa do not represent functional groups because within each taxa group there might be species that play different functions (some bivalves might be deposit feeders while some suspension feeders). Therefore, it is not clear to me why the authors use the term “functional groups”. If this is their aim to test functional groups, it must be described before mentioning them – what does functional group means for the authors. However, as I already mentioned, functional groups are not simply taxa.
Line 152- 58: This sentence is way too long, I get lost. And it is not grammatically correct:
“Some additional functional groups could be found on the live rocks and in the sediments, but their densities were not quantified which the abundance were low and did not influence the dissolved oxygen concentration and nutrients, or these organisms were similar in each mesocosm so it didn’t affect the experiment; these included sea anemones (e.g., Exaiptasia sp.), sponges, hydrozoans, macroalgae, filamentous algae, crustose coralline algae, isopods, amphipods, crabs, gastropods, and a diverse array of polychaete worms.”
I suggest cutting the sentence to two, and correcting: “not quantified which” to “not quantified because”. Also, it should be “the abundance was low”.
Line 202: Add this equation using the equation tool in word. It should be in a separate line.
Line 206: mesh cage with what? Something is missing in the sentence.
Line 238: Above line 146 the shoot density is reported as shoots per m2. Here it is shoot per cm2. I have never seen this unit so I suggest changing it and reporting the data as per square meter.
2.6 Data and statistical analysis – It should be stated which parameters were transformed and which were not. In line 175 there are 4 types of transformation mentioned, why so many were applied? What does it mean for the data? This should be better explained to evaluate whether the statistical tests were applied correctly and to be able to interpret the results correctly.
Paragraph 3.4 – It is hard to ready this paragraph and to follow the results. I suggest rewriting, probably cutting in two would be helpful.
Table 1 – It is impossible to read the first row of the columns that describe what results are presented. According to the caption “A separate nested ANOVA was carried out to assess the effects of pCO2 nested within temp., as well as tank nested within pCO2 for all response variables except for Fv:Fm” however, I can see the results of Temp. x (pCtank) for Fv:Fm, at least I think it’s what I see but as I said, the table has to be correctly presented.
Fig. 2 – Figure caption - The latin name should be in italics. Also, the first two sentences repeat each other. I suggest rewriting.
Fig. 3 – Figure caption – The latin name should be in italics, but the beginning of the sentence should not (The chemical content of Thalassia hemprichii”). Also, the caption is quite confusing with so many brackets. I suggest rewriting.
Line 354 – It should be “Neither of the decomposition rates…”. Also in this sentence, there is a missing closing bracket.
Fig. 4 Again, the Figure caption – The latin name should be in italics, but the beginning of the sentence should not (Decomposition and carbon sequestration rates in Thalassia hemprichii).
In all of the Figures’ captions – it is not clear what the letters above bars represent – it is written that it is a Turkey test difference but what do letters from a to d mean?
Line 380: Having this elevated level of underground biomass could more generally aid in seagrass survival when other environmental conditions become unfavorable.
Paragraph 383-390 – Nothing in this paragraph seems to explain the lack of OA effect on seagrass productivity. The given examples are not related to the current experiment – how does the elevated current velocity and light could explain the current results? And was there any carbon limitation or was there carbon excess during the experiment? I suggest finding better sources or explain what does the author mean and discussing this result a bit more.
Line 391 – And where is the second reason? If it relates to explaining the lack of OA effect on seagrass productivity, why this is in a new paragraph?
Line 396-397 – This sentence is a bit odd – “influences their work with C.nodosa” - maybe specify what was influenced, saying “their work” seems not to be very correct and informing since the reader doesn’t have to know their work.
Line 417-428 – Is it possible that the highest temperature used in this experiment was not high enough to note the negative consequences of ocean warming? The authors cited above that in Florida, a temperature of 34 degrees was recorded for two months which in effect caused a die-off of seagrass. The highest temperature in the experiment is 31, therefore I am questioning the experimental design. I am not convinced the tested temperature range was high enough to observe the pattern of change.
Line 425-428 – I am not convinced this is true given the previously cited example of Florida temperatures raise to 34 degrees and it’s effect on seagrass massive die-off.
Line 439-440. What negative effects of temperature?
Line 449 – “macrofauna fauna” – delete fauna
Supplement:
Line 7: Lack of closing bracket.
Line 13: I suggest replacing “over the course” with “throughout”
Line 15-16: Should be “throughout the study” (delete “duration of”).
Line 29: Should be “the experimental stage”
Table S2 – Replace “over the duration” with “throughout”
Fig. S3 – This sounds a little bit odd: “C) Plastic bins that were later transported to the experimental mesocosms.”. I suggest rewriting “C) Plastic bins later transported to the experimental mesocosms.
Fig. S4 – I suggest decreasing the size of points in the graphs so it is easier to read e.g. Fig. S4 A and B don’t look good visually.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This study is an experimental analysis of the effects of increased temperature and pCO2 on several growth and physiology response variables in the Indo-Pacific seagrass Thalassia hemprichii. The experiment was performed using six mesocosms, in which three were held at ambient pCO2, while three were held at elevated pCO2. All six mesocosms received four temperature treatments (25, 28, 31, plus an additional 28 “recovery” treatment).
This is an impressive and difficult experimental study on a question of fundamental importance in seagrass biology and marine ecology. The experimental treatments appear to be well-controlled using established methods, and the results and implications are well described and discussed in the context of the current field, in perfect native English.
My only concern is regarding the experimental design and the statistical analysis, and might well be my own misunderstanding. I think it would be useful to readers for the authors to revise for clarity on the following questions.
1. Why were the mesocosms divided into two groups? The authors state that one group was used as a “control” and the others as “treatments;” however, the three “controls” were only controlled for CO2, whereas all mesocosms were used as both “controls” and “treatments” for temperature, as all temperature treatments were applied to all mesocosms. This is a split-plot design (one treatment is within blocks, the other is between blocks). A simpler design would be to use each mesocosm as a separate experimental block, so that we would have a randomized block design, which I think would be superior in allowing use of only one error term, rather than two terms in the present design (one for pCO2 and a different one for temperature). In the randomized block design, all pCO2 and all temperature treatments would be applied at the same time to all six mesocosms, so that the entire experiment would be performed within each tank, and replicated six times. Wouldn’t this be a preferable design with higher statistical power if there is a tank effect?
2. There is no mention of error terms; what term was used for assessment of the pCO2 and what for the temperature effects? The temperature error should have been a within-term, while the pCO2 error should have been a between-term.
3. How were the error bars calculated in Figures 2 and 3? They should have been calculated using the appropriate error term in the relevant ANOVA, but the authors do not provide this information, as far as I understand.
4. Table 1 is unreadable to me in the PDF I was provided – there is a line wrap problem that makes it impossible to know which column of numbers is under which column heading.
5. I’m not sure I understand the language used in the paragraph on statistical analysis – pCO2 is not a random nesting factor within temperature, rather it’s a fixed factor crossed with temperature. The fact that it varies between tanks rather than within tanks does not make it nested within temperature. “For the remaining response variables”: This preposition I don’t understand, as pCO2 is not a response variable, it’s a treatment. Also, how is random variation among mesocosms incorporated into the ANOVA? This is tank variation, but tank variation should be a random factor used for the error in the pCO2 effect, not a fixed factor.
6. There are no degrees of freedom reported for the F values in Table 1. This is basically the same question as I have above: what were the error terms used? In a split-plot design there are two error terms, different for the two (within versus between) treatments, and they give different denominator degrees of freedom in the observed ANOVA F.
7. The “recovery” temperature treatment of 28C seems to be displayed in Figure 2A for Fv:Fm, but not for any of the other response variables. Why not? It does not seem to be discussed, or referred to in the Figure captions. in Figure 2A the recovery response seems to be significantly lower than the sequential response at 28C; is this true, and what is the cause if known?
I thank the editors for giving me the opportunity to review this very interesting and important paper, and I hope my comments are useful to the authors and journal.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The study assesses the effects of ocean warming and acidification on a tropical seagrass species. While such studies are currently limited, especially in tropical systems, the study’s finding would contribute in filling gap of knowledge of how seagrass species responds to environmental change, especially climate change in the tropics. However, I found that this manuscript is not suitable for publication yet in its current form. It needs several improvements or clarifications, so that it becomes publishable.
Below are my comments:
- L11-L15. I think, the introduction parts in the Abstract needs to be improved. It is too broad. Mentioning first about the important contribution of seagrass bed in climate change mitigation is a good opening. It’d be better to proceed with a statement about how climate change (ocean acidification/OA and ocean warming/OW) might affect seagrasses, and thus this is what the study is aimed.
- L63-L74. I agree, assessing the impact of OA (and OW) at the scale of ecosystem or community might better reflect nature’s actual condition, rather than focusing only the effect on the species level. However, more detailed rationales in the context of seagrass ecosystem needs to presented. The introduction needs to present specific rationales of each functional groups included in the mesocosm. For example, how different component of seagrass bed ecosystem might respond to OA and OW, or why is the sea urchin species included (or why not other species). So that, this study is worth doing.
- L58-L75 and L146-L158. I understand, that the reason to include many other species in the mesocosm is to generally represent functional groups in seagrass beds ecosystems. But just because few studies have explored the effect of climate change within the level of ecosystem does not justify such mesocosm design employed in this study over simpler species-based experiment. Do the authors expect Thalassia hemprichii would respond differently among the mesocosm experiment and simpler species-based experiments? If so, how would those different functional groups influence the way Thalassia hemprichii responds to environmental change (OA and OW)?
Furthermore, why are those species (or why not other species) selected to represent functional groups in the mesocosm? Is there any specific reason? Would the presence of those species influence how Thalassia hemprichii responds to OA and OW? If there is no clear link between the presence of those species with Thalassia hemprichii, then the mesocosm design is not justified.
I don’t see any clear role of those ecological functional groups in the hypothesis In L78-81. I think, this experiment needs another control group, in which Thalassia hemprichii is situated without any presence of those functional groups.
Indeed, throughout the Discussion, there is no clear and relevant statement about the presence of ecological functional groups used in the mesocosm.
Furthermore, did all units of mesocosm have similar species abundance? How were the community structures in each mesocosm controlled?
This study needs to provide clarification or address all of these concerns.
L15. “…ocean acidification (OA)”. I think it this should be “ocean acidification (OA) and ocean warming (OW)”, as this study is assessing the impact of ocean acidification in waters with elevated temperatures.
L35. 400ppm is not partial pressure of CO2, but is CO2 concentration. ppm is a unit for concentration, while unit for partial pressure is mmHg. Sentences about this should be revised throughout the manuscript.
L37. What is GCC? I know, this means global climate change. The full term must be stated first before the acronym.
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
All of my concerns and suggestions raised in the first round of review were addressed by the authors. I appreciate the point-by-point answers to my concerns. The main concern about experimental design (the highest temperature used in the experiment) has been explained and is satisfactory. I appreciate that this has been also underlined in the abstract and the text. The presentation of results and methods has improved (The table that was impossible to read before, now is clear). The discussion has improved as the parts that were speculatory were removed and too-long sentences have been rewritten. It seems that the text has been gramatically corrected however, as I am not a native speaker, I cannot make the final judgment about the quality of the language.
Given the above comments, I recommend that this manuscript can be published.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors have provided acceptable clarification to the previous reviewers' comments. There is one point, but very important, that needs to be followed up. The authors still wrongly used the unit ppm for pCO2. This mistake can be found in many others throughout the manuscript (for example L153, L183, and others). It'd be better to consistently use CO2 concentration (with ppm), rather than pCO2. But, automatically replacing "pCO2" with "CO2 concentration" (using the find & replace feature of a word editor) would likely change the structure or context of the sentence, thus I suggest all the "problematic" sentences need to be revised accordingly.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
