Correlation between Non-Structural Carbohydrates and C:N:P Stoichiometric Ratio of Haloxylon ammodendron under Different Water–Salt Gradients

Round 1

Reviewer 1 Report (Previous Reviewer 1)

I remember I have reviewed an earlier submitted version of this manuscript. Although several improvement is shown in the current version, I consider that several flaws still exist and much improvement is needed. First, I appreciate the English writting of this manuscript. The readability is higher than average. Second, I have several doubts about the presentation of the results and more statistical analyses are needed. I have several specific recommendations as follows:

1. The presentation of experimental design is problematic. I do not consider the presentation of water-salt gradient is correct. The environmental gradient in the experimental design is much more complexed than just variations in the percentages of "water" and "salt" in the soil. The Fig. 2 presents the correlation between salt and water content is not meaningful for the present research. I suggest to move this figure to the supplementary. Table 2 showes the gradient of pH, TN and TP in the soil of the three sites. Therefore, I think the water and salt gradient are not the only playmaker in the shape of plant stoichiometry. Hence, I think the presentation of experimental design is problematic. The authors need to explain the gradient in another form and it is not rigorous to just say the variations in soil water and salt content altered the plant performance.

2. Following the above question of the design of environmental gradient, the secondary problem exists in the analyses of the data. Since the environmental gardient is complexed with a constitution of factors of water, salt, pH, TN and TP, the authors must analyze the contribution of each factor in shaping the plant stoichiometry using methods such as ordination, multivariate regression model or structural equation model. Based on this analysis, we can see which factor is more deterministic in the shape of plant stoichiometry and verify if the soil water and salt content "indeed" induces the plant stoichiometry. The current analyses are not sufficient to tell a whole story. If the percentage of contribution of each factor to the plant stoichiometry can be determined (e.g. soil water >soil salt > TN>TP or reverse or other hierachies), a novel story about the tradeoff between water, soil and nutrient can be told and a comprehensive ecological perspective can be proposed.

3. I have doubts about Fig.3 and Fig.5. Fig. 3 displays a decrease of NSC along the Site I, II and III. Based on the gradient, the site III is drier than the other two sites. Generally, plants have higher NSC concentrations when confronting drier condition (see Würth, M.K.R., Peláez-Riedl, S., Wright, S.J. et al. Non-structural carbohydrate pools in a tropical forest. Oecologia 143, 11–24 (2005). https://doi.org/10.1007/s00442-004-1773-2). Therefore, I think the water gradient may be not a deterministic factor comparing to other factors in affecting the plant performance.

Fig. 5 shows negative correlations between plant NSC and carbon concentration, C/N and C/P, which are counter-intuitive. NSC is constituted by carbon, hydrogen and oxygen. If the NSC concentration increases, the content of carbon should have a greater probability to increase. Maybe more carbon was used to construct the structural carbohydrate and induces this consequence. I suggest the authors to analyze the structural carbohydrate concentration if you aim to tell a whole story.

 

The manuscript is well-written and the language is not problematic.

Author Response

I would like to thank you for your careful and insightful comments on the manuscript. We hope that editors and reviewers can see our efforts.

Point 1: The presentation of experimental design is problematic. I do not consider the presentation of water-salt gradient is correct. The environmental gradient in the experimental design is much more complexed than just variations in the percentages of "water" and "salt" in the soil. The Fig. 2 presents the correlation between salt and water content is not meaningful for the present research. I suggest to move this figure to the supplementary. Table 2 showes the gradient of pH, TN and TP in the soil of the three sites. Therefore, I think the water and salt gradient are not the only playmaker in the shape of plant stoichiometry. Hence, I think the presentation of experimental design is problematic. The authors need to explain the gradient in another form and it is not rigorous to just say the variations in soil water and salt content altered the plant performance.

Response 1:

We are grateful for reviewer’s guidance.

According to suggestions of reviewer, we have moved Figure 2 to the supplementary file.

Regarding the issue of experimental design, I would like to provide the following explanation:

For arid areas, the important environmental factor affecting plant survival is water content. Our experimental research area is located in a typical arid area, with severe soil salinization and low soil nutrient content. We have conducted continuous monitoring in the region for several years and accumulated a large amount of data. Moreover, the research transect set up in our research area has obvious water and salt gradients. Based on the water and salt gradients, we measured the background values of soil nutrients under this gradient, and only then did we obtain the data in Table 2.

Point 2: Following the above question of the design of environmental gradient, the secondary problem exists in the analyses of the data. Since the environmental gardient is complexed with a constitution of factors of water, salt, pH, TN and TP, the authors must analyze the contribution of each factor in shaping the plant stoichiometry using methods such as ordination, multivariate regression model or structural equation model. Based on this analysis, we can see which factor is more deterministic in the shape of plant stoichiometry and verify if the soil water and salt content "indeed" induces the plant stoichiometry. The current analyses are not sufficient to tell a whole story. If the percentage of contribution of each factor to the plant stoichiometry can be determined (e.g. soil water >soil salt > TN>TP or reverse or other hierachies), a novel story about the tradeoff between water, soil and nutrient can be told and a comprehensive ecological perspective can be proposed.

Response 2:

We are grateful for reviewer’s guidance.

According to suggestions of reviewer, I tried to use the random forest model to calculate the contribution percentage of water, salt, soil organic carbon, soil total nitrogen and soil total phosphorus. The results are shown in the table below:

Feature Name	Feature importance
Water content	31.30%
Salt content	27.60%
Soil organic carbon (SOC)	18.90%
Soil total nitrogen (TN)	14.40%
Soil total phosphorus (TP)	7.80%

This also confirms that our experimental design is to some extent reasonable.

Point 3: I have doubts about Fig.3 and Fig.5. Fig. 3 displays a decrease of NSC along the Site I, II and III. Based on the gradient, the site III is drier than the other two sites. Generally, plants have higher NSC concentrations when confronting drier condition (see Würth, M.K.R., Peláez-Riedl, S., Wright, S.J. et al. Non-structural carbohydrate pools in a tropical forest. Oecologia 143, 11–24 (2005). https://doi.org/10.1007/s00442-004-1773-2). Therefore, I think the water gradient may be not a deterministic factor comparing to other factors in affecting the plant performance.

 

Fig. 5 shows negative correlations between plant NSC and carbon concentration, C/N and C/P, which are counter-intuitive. NSC is constituted by carbon, hydrogen and oxygen. If the NSC concentration increases, the content of carbon should have a greater probability to increase. Maybe more carbon was used to construct the structural carbohydrate and induces this consequence. I suggest the authors to analyze the structural carbohydrate concentration if you aim to tell a whole story.

Response 3:

We are grateful for reviewer’s guidance.

According to suggestions of reviewer, I would like to explain as follows:

In our study, the content of non-structural carbohydrates (NSC) decreased with the decrease of water and salt content, which may be contrary to our previous understanding, but the reasons for this are multifaceted. Firstly, different research regions and species result in different experimental results; Secondly, it is permissible for the results of our field research to differ from those of indoor control experiments; Thirdly, plants in arid areas may invest more NSC to resist drought stress in order to survive, resulting in a decrease in their own carbon composition, leading to a decrease in the content of NSC.

Regarding the issue of Figure 5, I would like to explain as follows:

The reviewers are quite right that more carbon may be used to construct structural carbohydrates. This is also one of the experiments we will do in the future, and it is also an important content that we will analyze in the future.

The revised draft will be uploaded as an attachment for your review. Finally, thank you again for your valuable comments on the content of this article during your busy schedule, which have greatly benefited me and have greatly inspired my future scientific research.

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

I consider that this study increase our understanding of the survival and growth strategies of H. ammodendron in arid environments by analyzing the characteristics of the C:N:P ratio and NSC accumulation and distribution under different water–salt gradients. Besides, I reckon that the provided results will be relevant to the restoration of degraded ecosystems and desertification control.

The text is clear and the performed analysis are correct. In addition, the discussion is coherent and discuss successfully the obtained results.

Nonetheless, a few changes should be done to improve the manuscript:

Line 86, page 2. Please add the authors of Haloxylon ammodendron at the first time its name appears in the text.

In the text the authors indicate P < 0.05 for signifficant results and P > 0.05 for non-signifficant results. I suggest the use of P ≤ 0.05 for signifficant results (besause 0.05 is not consider in the both options exposed in the study).

Please, unify the style header of Table 2, 3 and 4 (adjust "Water salt-gradient" term in one line in header of Table 4 like in Table 2 and 3).

 

Author Response

I would like to thank you for your careful and insightful comments on the manuscript. Based on your suggestion, we have made modifications to this. We hope that editors and reviewers can see our efforts.

Point 1: Line 86, page 2. Please add the authors of Haloxylon ammodendron at the first time its name appears in the text.

Response 1:

I would like to thank you for your careful and insightful comments on the manuscript.

According to suggestions of reviewer, we have made modifications to this. (Line 86-87)

Point 2: In the text the authors indicate P < 0.05 for signifficant results and P > 0.05 for non-signifficant results. I suggest the use of P ≤ 0.05 for signifficant results (besause 0.05 is not consider in the both options exposed in the study).

Response 2:

I would like to thank you for your careful and insightful comments on the manuscript.

According to suggestions of reviewer, I reanalyzed the data and set the significance to P ≤ 0.05. The research results are the same, and our significance is all less than 0.05. But for the sake of rigor, we still set the significance in the text to P ≤ 0.05.

Point 3: Please, unify the style header of Table 2, 3 and 4 (adjust "Water salt-gradient" term in one line in header of Table 4 like in Table 2 and 3).

Response 3:

I would like to thank you for your careful and insightful comments on the manuscript.

According to suggestions of reviewer, we have standardized the tables. The details are as follows:

Table 4. Changes in water potential of assimilating branches and secondary branches under different water–salt gradients

Water–salt gradient	Assimilating branches		Secondary branches
	Predawn	Midday	Predawn	Midday
I	–4.62±0.29 ab	–5.92±0.94 b	–4.94±0.82 c	–5.91±1.09 c
II	–4.24±0.79 a	–4.77±0.95 a	–3.55±0.75 a	–3.78±1.02 a
III	–4.93±0.76 b	–5.29±0.85 ab	–4.23±0.85 b	–4.53±1.12 b

The revised draft will be uploaded as an attachment for your review.

Finally, thank you again for your valuable comments on the content of this article during your busy schedule, which have greatly benefited me and have greatly inspired my future scientific research.

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The study collected numerous data including soil CNP, plant CNP and plant NSC, the latter two indicators usually indicate the plant life-history strategy and are commonly correlated with environmental variables such as soil CNP. I personally admire the data collection in the present study and the data size is sufficient to probe several ecological theories. However, two major deficts downgrade the quality of the manuscript.

First, I personally recommend the authors to reevaluate the theme of this study. The title shows that the authors aimed to study the effects of soil salinity and soil water content on plant stoichiometry and NSC. However, after I read the manuscript, I found that the gradient of soil salinity and soil water content is likely not crucial in determining the plant CNP and NSC and the water-salt stress cannot be a theme for this manuscript. Table 2 showed that the soil TN and TP or soil nutrient availablity decreases along the water-salt stress gradient of HD, MD and LD. Thus, soil salinity and soil water content are not the only involving factors for the environmental gradient design. The environmental gradient is complex for the present study. If the authors consider the soil salinity and soil water content as controling factor, the soil nutrient availability should have been consistent across the sites of HD, MD and LD and thus water-salinity stress can be the only driving factor. In fact,  soil nutrient availability rather than the soil salinity and water content is likely the deterministic factor for plant CNP after I checked the experimental results. For instance, Fig. 3 showed that the N concentration and P concentration of both assimilative and secondary branches decreased along the gradient of HD, MD and LD. This variation in plant nutrient concentration actually corresponds to the variation in soil nutrient concentrations in the three sites. Thus, what is the significance of soil salinity and water content? Soil salinity and water seem to have no effects on plant nutrient concentration. Therefore, I suggest the authors to replace the prior theme of water-salinity stress with soil nutrient availability. And a huge revision and rewriting may be needed.

Second, the structure of the manuscript is scattered despite that the English writing is acceptable. After I read the manuscript, I can not grasp the key information of the study and the text is a bit long. 

Author Response

Dear reviewer,

Thank you for your valuable comments on the content of this article during your busy schedule, which have greatly benefited me and have greatly inspired my future scientific research.
I will upload the content of my feedback as an attachment.

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a field study about the effects of salt and soil water availability (SWA) on twig carbon (C), nitrogen (N) and phosphorus (P) in a woody desert species. It is important that authors highlight in the Abstract that this is a field study, as otherwise it is hard to understand why the comparison between sites with high salinity and high SWA vs sites with low salinity and low SWA. The natural occurrence of both factors gives a realistic information of plant behavior, but complicates understanding the mechanisms through which each environmental factor affects twig stoichiometry. In fact, the Discussion is very speculative. One crucial limitation to this study is missing a proxy of water stress. Authors assumes that higher SWA implies lower drought, but higher salinity (together with higher SWA) can implies higher osmotic stress and drought stress, as some results suggest (e.g. the ratio SS: ST).

It would be recommended to provide plant size (e.g. height and/or crown or stem diameter) to know if decreases in nutrient concentrations are due to dilution effects within larger plants.

Thirty one 30x30m quadrats separated 90m and 121 30x30m quadrats cannot fit exactly in the same length as it is drawn in Figure 1.

The objective is too general and so it is the Introduction, which does not address the mechanisms affecting stoichiometry in young and older twigs.

The decision of naming the microsites of high SWA and high salinity as HD (high drought?) is highly confusing, as you later say these are the microsites of lower drought. Rename. Anyway, if you do not have any proxy of plant water status (e.g. plant water potential, relative water content, osmotic potential…) you should not speak of drought as straightforward as you do. You ignore drought stress; you should measure it for discussing how it might affect twig stoichiometry.

The three clusters indicate that this species cannot tolerate microsites of low SWA and high salinity (it probably would induce too much osmotic stress), as soil analyses are made from soil collected under the canopy of plants.

The reader has no information of what is the assimilation branch and the secondary branch. Rename and define. This is missing in section 2.3.2 or elsewhere. Consider including a picture to clarify this.

Why authors do not make a two-way ANOVA for analyzing main and interaction effects of microsite and branch (twig) type? Anyway, I am not sure the ANOVA is the best tool to use in your study. Multivariate analyses can be more appropriate, as that in Figure 4.

The Results section is very hard to follow. You should shorten the sub-sections to focus only on those results more relevant for the Discussion. Else, the reader will be tired and will not be able to take home any message. The reader will always have the results in the figures; there is no need to detail every result.

Legends are not enough self-explanatory, especially those of those of tables 3 to 5.

Explain to readers why the need of using a DCA and a RDA. Why not just a RDA?

The Discussion is unclear and speculative at many passages, with strong statements poorly or not at all supported by literature or your experimental design. Just as an example of the above criticism, check the first sentence of the Discussion linking NSC and water potential and transport. There are many passages having the same problem. Try to be more moderate in your statements.

The idea that C:N or C:P are proxies of nutrient use efficiency needs further development. Otherwise the reader might not understand this point.

Another example of strong statement: “leaves are key organs for carbohydrate accumulation”. It depends on the leaf habit, species, etc.

In lines 394-399, you must note that you have not measured biomass partitioning; just C concentrations.

Lines 419-422 (and others) need a reference supporting the statement.

Line 424. You should not speak of carbon starvation, as you have not measured C gain, C mobilization or C use in respiration, for example.

The role of P in NSC mobilization and storage is interesting but also speculative; so please moderate the tone.

Line 428. You should bring the explanation of N:P higher or lower than 16 to Results, for readers to understand why you focus on this specific value (and not other).

The conclusion is speculative for the reasons described above. Note that you miss any proxy of water stress. The question of why starch does not degrade under higher water stress is questionable, in view of abundant literature.

Please mend the typo for reference in lines 533-535.

Author Response

尊敬的审稿人，

感谢您在百忙之中对本文内容的宝贵意见，使我受益匪浅，对我今后的科学研究也有很大的启发。
我会将我的反馈内容作为附件上传。

请参阅附件。

Author Response File: Author Response.docx