Detecting Growth Phase Shifts Based on Leaf Trait Variation of a Canopy Dipterocarp Tree Species (Parashorea chinensis)

Round 1

Reviewer 1 Report

The article is complex and the content is fresh. It was a real pleasure to read it. Congratulations to the author team!
In essence, your study is an analysis of the vertical structure of forest stand related to leaf morphological and physiological traits. The tested hypotheses are clearly described, and the method, results and discussions are well organized. It's an honest study.
I noticed some minor issues that can be improved, but that do not affect the value of the article.

1. Line 93: you use the symbol DBH (otherwise very well known), but which is explained only in line 131.
2. Line 105: please, delete a space between "onset" and "change".
3. Line 109: the symbol Rd is not explained. The meaning of Rd is presented in lines 153 and 161.
4. Line 130: Census or inventory?
5. Lines 182-184 aren't results? Here is presented the method.
6. Line 204: please, just check the space between Celsius degree and "in".
7. Figure 2: it is not usual to present DBH variation depending on tree height. This is my opinion as an experienced researcher in forest biometry. I recommend here to change the data on the vertical axis with those on the horizontal axis and vice versa. I agree to use in the other graphs the tree height on the horizontal axis and I understand the value of use it in figures 3, 6-9. It was a good idea!

Author Response

Dear Editor,

We have given careful consideration to all the comments made by the two reviewers. Consequently, many changes have been made in the manuscript, which are detailed below in response to each suggestion/question.

We would like to thank the reviewers for their very helpful suggestions.

Thank you very much,

Authors

17 October 2020

**********************************************************************************************************

Reviewer 1

Comments and Suggestions for Authors

The article is complex and the content is fresh. It was a real pleasure to read it. Congratulations to the author team!
In essence, your study is an analysis of the vertical structure of forest stand related to leaf morphological and physiological traits. The tested hypotheses are clearly described, and the method, results and discussions are well organized. It's an honest study.
I noticed some minor issues that can be improved, but that do not affect the value of the article.

Line 93: you use the symbol DBH (otherwise very well known), but which is explained only in line 131.

Authors’ response: We modified as what you suggested in line 94.

Line 105: please, delete a space between "onset" and "change".

Authors’ response: We deleted the space in line 107.

Line 109: the symbol Rd is not explained. The meaning of Rd is presented in lines 153 and 161.

Authors’ response: We modified as what you suggested in line 111.

Line 130: Census or inventory?

Authors’ response: We replaced “census” by “inventory” in line 132.

Lines 182-184 aren't results? Here is presented the method.

Authors’ response: The statement here is to justify our method; we deleted the tedious statement and insert a supplementary figure (Figure S1) in lines 190–192.

Line 204: please, just check the space between Celsius degree and "in".

Authors’ response: We added the space between “℃” and "in" in line 212.

Figure 2: it is not usual to present DBH variation depending on tree height. This is my opinion as an experienced researcher in forest biometry. I recommend here to change the data on the vertical axis with those on the horizontal axis and vice versa. I agree to use in the other graphs the tree height on the horizontal axis and I understand the value of use it in figures 3, 6-9. It was a good idea!

 Authors’ response: We renewed figure 2 and use DBH on the horizontal axis in line 257.

Submission Date

25 September 2020

Date of this review

08 Oct 2020 09:44:19

Reviewer 2 Report

This paper presents an interesting set of data about leaf traits and leaf ecophysiology and their changes throughout ontogeny. However, the authors have extracted several conclusions that I find in part speculative. In addition, although the data seem to be mostly sound, there are some weaknesses in the data set that make the interpretations somehow doubtful. For these reasons, I think that the paper needs some additional work.

My main concern is that the authors interpret different phases during ontogeny as different ecological strategies following the Grime’s scheme. The CSR theory, as I can understand it, describes the plant adaptations to specific environmental conditions. I find difficult understanding why the authors consider that the trees experience these specific environmental conditions as they traverse the canopy throughout ontogeny. For example, why do the authors classify the “canopy phase” as “ruderal”? Ruderal strategy is usually present where disturbances are frequent. The authors did not explain which kind of disturbances are especially frequent during this phase of growth. There is a sentence in the conclusion that is perhaps related to this issue and that need clarification. After summarizing the classification of the different phases, the authors state: “although these strategies were inferred only relative to ontogenetic adaptation in P. chinensis” (l. 544). Does it mean that the classification only reflects the similarity between the leaf traits in the different phases with those of the Grimes’ strategies? If this is the case, you should clarify this at the start of the discussion.

Data collection was made at the end of the rain season. Is there some reason? Perhaps more details about the leaf strategies of your species (leaf life span, phenology, etc.) could be of help to understand your approach.

How many days were used for ecophysiological measurements? I suspect you performed the measurements during just one for light, other one for Rd, etc? Please give more details.

Line 175-176. Why do you mix the midday depression of photosynthesis with the swinging of the crane basket to justify the use of detached shoots? Did the swinging of the basket affect the Rd measurements?

The branch samples were measured within 8 hours of excision. How did you manage to maintain the samples in a good condition after 8 hours?

With respect to MRT, usually the application of these methods requires splitting the data base in two subsets: one for model fitting and the other one for validation. Did you use this procedure?

In most of your figures, there are several points at the right side corresponding to the tallest individuals, which present strong variability in the response variable with scarce variation in the independent variable. I wonder if this strong individual variability is related to the statement in line 502: “The leaves of P. chinensis in the canopy layer displayed a very large range of variation in Amax_area [29]”. Do you mean that different individuals show different photosynthetic capacities that are no related to differences in light availability or height?

In any case, the weak relationships between leaf traits and the independent variables for the tallest trees pose doubts with respect to your conclusion that during the last phase this species experiences a decline in Amax. Actually, the explanation you give in the discussion for this trend is convincing, but it should be useful if you may provide a better experimental evidence. The number of individuals included in this phase is rather low, and, for this reason, your conclusion may be debatable.

Several sentences in the Discussion are confusing. For example, in line 397 “while Amax_mass and PNUE monotonically decreased with height”. This sounds contradictory with your scheme in Table 1.

Line 447. Leaf lifespan is introduced here as a trait that change with ontogeny. You did not report data on this variable. After that, in line 452, the vulnerable phase is characterized as a period with high mortality due to drought, predation and “missing”? Do you think that a long leaf life span is a favorable trait under conditions of high mortality?

Again, in line 480, you add a variable (RGR) not mentioned before and with no data to confirm this statement.

In sum, although your data are valuable, I would suggest a rewriting of the Discussion to remove these inconsistencies. Especially, I would recommend a better justification for the interpretation of the different phases in the frame of Grime’s strategies.

Author Response

Dear Editor,

We have given careful consideration to all the comments made by the two reviewers. Consequently, many changes have been made in the manuscript, which are detailed below in response to each suggestion/question.

We would like to thank the reviewers for their very helpful suggestions.

Thank you very much,

Authors

17 October 2020

**********************************************************************************************************

Reviewer 2

Comments and Suggestions for Authors

This paper presents an interesting set of data about leaf traits and leaf ecophysiology and their changes throughout ontogeny. However, the authors have extracted several conclusions that I find in part speculative. In addition, although the data seem to be mostly sound, there are some weaknesses in the data set that make the interpretations somehow doubtful. For these reasons, I think that the paper needs some additional work.

My main concern is that the authors interpret different phases during ontogeny as different ecological strategies following the Grime’s scheme. The CSR theory, as I can understand it, describes the plant adaptations to specific environmental conditions. I find difficult understanding why the authors consider that the trees experience these specific environmental conditions as they traverse the canopy throughout ontogeny. For example, why do the authors classify the “canopy phase” as “ruderal”? Ruderal strategy is usually present where disturbances are frequent. The authors did not explain which kind of disturbances are especially frequent during this phase of growth. There is a sentence in the conclusion that is perhaps related to this issue and that need clarification. After summarizing the classification of the different phases, the authors state: “although these strategies were inferred only relative to ontogenetic adaptation in P. chinensis” (l. 544). Does it mean that the classification only reflects the similarity between the leaf traits in the different phases with those of the Grimes’ strategies? If this is the case, you should clarify this at the start of the discussion.

Authors’ response: We removed “ruderal strategy” in discussion, and clarify “all these strategies described here were inferred only relative to ontogenetic adaptation in P. chinensis, which only reflected the similarity between the leaf traits in the different phases with those of the Grimes’ strategies” at the start of the discussion (in lines 403–405).

Data collection was made at the end of the rain season. Is there some reason? Perhaps more details about the leaf strategies of your species (leaf life span, phenology, etc.) could be of help to understand your approach.

Authors’ response: We collected the data at the end of the rainy season, because it is also the main growing season for local plants. The weather in these months has enough sunny days so that we can perform the crane and conduct measurements in the field.

These explanations have been added in lines 140–142.

How many days were used for ecophysiological measurements? I suspect you performed the measurements during just one for light, other one for Rd, etc? Please give more details.

Authors’ response: We used 34 workdays for ecophysiological measurements in field (17 days in Oct 2016, 3 days in Nov 2016, 9 days in Sep 2017 and 5 days in Oct 2017). The weather conditions limit our field work (too much rainfall before October, and too little rainfall after November), and sometimes the equipment failure also limits our efficiency. These explanations have been added in lines 140–142.

We measured R_d, light-response curve, leaf area, thickness, LMA, leaf nitrogen and carbon in same leaf. These explanations have been added in lines 159–160.

Line 175-176. Why do you mix the midday depression of photosynthesis with the swinging of the crane basket to justify the use of detached shoots? Did the swinging of the basket affect the Rd measurements?

Authors’ response: The photosynthesis of P. chinensis decreased at midday, the use of detached shoots could eliminate the immediate effects of path length and gravity on gas exchange. Meanwhile, the swinging of the crane basket also made it difficult to measure a light-response curve in situ, because this measurement requires clamping a leaf for more than one hour. These explanations have been added in lines 181–184.

The swinging of the basket was acceptable for the R_d measurements, because this measurement just needs clamp the leaf for minutes.

The branch samples were measured within 8 hours of excision. How did you manage to maintain the samples in a good condition after 8 hours?

Authors’ response: After the gas exchange measurements within 8 hours of excision, the leaf lamina thickness of these fresh leaves was determined using a digital caliper (accuracy: 0.01 mm, No. 111-101-40, Guanglu Co., Guilin, China) based on the mean value of three repetitive measurements across the lamina while avoiding major veins. The areas of the leaves were measured with a leaf area meter (LI-3000C, LI-COR Inc., Lincoln, NE, USA). After then, all leaves were dried at 60 ℃ in an oven for 3 days, and the leaf dry mass was measured. These explanations have been added in lines 211–212.

With respect to MRT, usually the application of these methods requires splitting the data base in two subsets: one for model fitting and the other one for validation. Did you use this procedure?

Authors’ response: MRT is a form of multivariate regression in which the response is explained by the explanatory variables. However, it is also a method of constrained clustering, because it determines clusters that are similar in a chosen measure of traits dissimilarity, with each cluster defined by a set of environmental values (De'ath 2002). As a clustering method, MRT needs to input enough samples to generate a tree structure that partitions the data set into mutually exclusive groups, but this process doesn’t need validation.

The fitness of MRT is defined by relative error (RE) and cross-validated relative error (CVRE). RE indicated the total impurity of the leaves divided by the impurity of the root node (the undivided data), and gives an over-optimistic estimate of how accurately a tree will predict for new data. Predictive accuracy is better estimated from the CVRE, which varies from zero for a perfect predictor to close to one for a poor predictor (De'ath 2002).

These explanations were in lines 229–233.

Reference: De'ath, G. Multivariate regression trees: A new technique for modeling species–environment relationships. Ecology 2002, 83, 1105–1117; DOI:10.1890/0012-9658(2002)083[1105:mrtant]2.0.co;2.

In most of your figures, there are several points at the right side corresponding to the tallest individuals, which present strong variability in the response variable with scarce variation in the independent variable. I wonder if this strong individual variability is related to the statement in line 502: “The leaves of P. chinensis in the canopy layer displayed a very large range of variation in Amax_area [29]”. Do you mean that different individuals show different photosynthetic capacities that are no related to differences in light availability or height?

Authors’ response: This description was not very suitable to justify our origin meaning that “the highest photosynthetic efficiency appeared in this phase”, sorry. We modified the statement as “The leaves of P. chinensis in the canopy layer displayed higher A_{max_mass} than other local Dipterocarp species” in lines 516–518.

In any case, the weak relationships between leaf traits and the independent variables for the tallest trees pose doubts with respect to your conclusion that during the last phase this species experiences a decline in Amax. Actually, the explanation you give in the discussion for this trend is convincing, but it should be useful if you may provide a better experimental evidence. The number of individuals included in this phase is rather low, and, for this reason, your conclusion may be debatable.

Authors’ response: Because the canopy crane jib only covered a few emergent trees, very few individuals were included in emergent phase. We expect more experiments for emergent trees in the future.

These explanations have been added in lines544–546.

Several sentences in the Discussion are confusing. For example, in line 397 “while Amax_mass and PNUE monotonically decreased with height”. This sounds contradictory with your scheme in Table 1.

Authors’ response: The statement here was incomplete, sorry. It should be “A_{max_mass} and PNUE peaked at low %TRANS (3.50% for A_{max_mass} and 2.60% for PNUE) and then monotonically decreased with height”.

These modifications have been added in lines 412.

Line 447. Leaf lifespan is introduced here as a trait that change with ontogeny. You did not report data on this variable. After that, in line 452, the vulnerable phase is characterized as a period with high mortality due to drought, predation and “missing”? Do you think that a long leaf life span is a favorable trait under conditions of high mortality?

Authors’ response: The leaf lifespan is unimportant in vulnerable phase, sorry. We modified the statement as “Low PNUE in the vulnerable phase might indicate that P. chinensis seedlings invested much less photosynthetic product into growth.” (in lines 459–463)

Again, in line 480, you add a variable (RGR) not mentioned before and with no data to confirm this statement.

Authors’ response: This statement sounds unnecessary, sorry. We deleted it in line 494–496.

In sum, although your data are valuable, I would suggest a rewriting of the Discussion to remove these inconsistencies. Especially, I would recommend a better justification for the interpretation of the different phases in the frame of Grime’s strategies.

Authors’ response: We deleted “ruderal” strategy and modified the canopy phase as “competitive” strategy, and addressed that variance in emergent phase still deserves more experiments in the future.

More modifications are marked using Microsoft Word Track Changes in the present version.

Submission Date

25 September 2020

Date of this review

13 Oct 2020 20:28:29

Round 2

Reviewer 2 Report

I think the authors have responded appropriately to my suggestions. The paper has been improved. I think this is an interesting piece of work.