Effects of Habitat Filtering on Tree Growth and Mortality across Life Stages in an Old-Growth Temperate Forest

Round 1

Reviewer 1 Report

Provide more details, a definition of the "habitat filtering" concept, first outlined in line 43 

Explain the abbreviation "T" in line 95

In the conclusions, the hypothesis needs to be addressed and confirmed or rejected. 

Author Response

Response to the Comments

Dear editors and reviewers,

Re: Manuscript ID: forests-1717513 and Effects of Habitat Filtering on Tree Growth and Mortality Across Life Stages in An Old-Growth Temperate Forest

Thank you for your letter and the reviewers’ comments concerning our manuscript entitled “Effects of habitat filtering on tree growth and mortality across life stages in an old-growth temperate forest”. Those comments are valuable and very helpful. We have read through comments carefully and have made corrections. Based on the instructions provided in your letter, we uploaded the file of the revised manuscript. Revisions in the text are shown using red highlight for additions. The corrections in the paper and our responses to the reviewer's comments are presented following.

Reviewers' Comments to Author:

Reviewer: 1

1. Provide more details, a definition of the "habitat filtering" concept, first outlined in line 43

Response: Thanks for your comment! We have added the concept of “habitat filtering” in lines 43-45.

2. Explain the abbreviation "T" in line 95

Response: Thanks for your comment! We have explained the meaning of the abbreviation “T” in line 95 ; T was time, which represented 5 years.

3. In the conclusions, the hypothesis needs to be addressed and confirmed or rejected.

Response: Thanks for your comment! We have added and acknowledged the hypothesis in our conclusions, and revised our conclusions.

Author Response File: Author Response.doc

Reviewer 2 Report

General comments:

Overview and general recommendation:

It is true that habitat filtering and filtration in ecology plays an important role for the habitat suitability that affects regeneration rates, mortality rates and species richness because a species can arrive and grow at a site but cannot tolerate the site’s abiotic conditions independently from other biotic factors. The aim of this article was to present the effects of habitat filtering on the growth rate and mortality rate across life stages in an old-growth temperate forest.

The article's main contribution is the report of principal component analysis of the many environmental variables that have been collected in large 9 ha plot where practically all trees were measured three times and a dense grid of soil samples were taken for the analysis.

The Authors tried to present their method and the results from their investigation in a short and comprehensive manner which, I think it is a good idea. On one hand I found the manuscript overall interesting, especially for the very large dataset used. However, there are problems mistakes and not clear meanings which I found not so clearly presented in the manuscript. In the results section for example there is no reference for the habitat filtering, the term exists in the title but not in the M & M and results sections. The figures I don’t think that cover clearly and comprehensively the results. Generally, it is difficult for the reader to follow the meaning and understand what is presented in the figures. The way the results are written, one, cannot confirm the findings that discussed in the discussion section. The manuscript cannot be considered for publishing in forests journal as it is written and presented. I suggest the authors to revise the manuscript according to my suggestions. Therefore, I recommend that a major revision is warranted. I explain my concerns in more detail below and I ask the authors address my comments in their response.

Specific comments:

Lines 79-80

The meaning is not clear, please revise

Line 84

… dark brown forest according to the Chinese…

Results section

The authors do not report anything or at least some descriptive statistics about the RGR and MR that have calculated – estimated  by the formulae reported in lines 94-96

Line 148

These variables concern the MR and the RGR, as far as I can read from table1

Lines 145-153

There is a mistake here either all Pearson’s significant correlations marked with asterisk are not shown in tab 1 or there are mistakes in the text in lines 148-151

 Line 151

…VWC had a significantly negative correlation with MR. -> This result concerns the large trees and the RGR not the MR

Table 1.

The significant pearson correlation values marked with asterisk do not correspond with those reported in the text.

Lines 161-173

Why the authors haven’t reported biplot graphs for showing the principal component axes and the related variable graphically. I think that biplots would have been more informative depicting better the relationships and much easier for the reader to follow. So I strongly urge the authors  to include biplots in their revisions

Line 163

What about vars VWC and BD?

Line 166

What about vars Con, Slo and VWC? There are more variables with large negative or positive pca coefficients in tab 3 that are not mentioned in this paragraph.

Line 181

This subsection must be numbered 1.2 instead of 1.1

Line 199

…than for large trees…

Line 201

I’m not convinced for this result

Line 206

…stronger than that of saplings…

Lines 219-238

As I’ve mentioned in my general comments, the way the results are written and presented the reader cannot confirm the findings in the discussion section. MR od saplings, for example, “it is increased with reduced water availability” and the only numbers that are reported in results are those of VWC in tab 3 for the three pca axes 0.213 0.34 0.221 respectively. What are these numbers mean and how they are explaining what is written in the discussion?

Lines 233-234

Does this sentence here results from your analysis in the results section? where?

Lines 239-258

In this paragraph the habitat filtering is discussed but there nothing for this term in the M & M and the Results sections

Author Response

Response to the Comments

Dear editors and reviewers,

Re: Manuscript ID: forests-1717513 and Effects of Habitat Filtering on Tree Growth and Mortality Across Life Stages in An Old-Growth Temperate Forest

Thank you for your letter and the reviewers’ comments concerning our manuscript entitled “Effects of habitat filtering on tree growth and mortality across life stages in an old-growth temperate forest”. Those comments are valuable and very helpful. We have read through comments carefully and have made corrections. Based on the instructions provided in your letter, we uploaded the file of the revised manuscript. Revisions in the text are shown using red highlight for additions. The corrections in the paper and our responses to the reviewer's comments are presented following.

Reviewer: 2

The Authors tried to present their method and the results from their investigation in a short and comprehensive manner which, I think it is a good idea. On one hand I found the manuscript overall interesting, especially for the very large dataset used. However, there are problems mistakes and not clear meanings which I found not so clearly presented in the manuscript. In the results section for example there is no reference for the habitat filtering, the term exists in the title but not in the M & M and results sections. The figures I don’t think that cover clearly and comprehensively the results. Generally, it is difficult for the reader to follow the meaning and understand what is presented in the figures. The way the results are written, one, cannot confirm the findings that discussed in the discussion section. The manuscript cannot be considered for publishing in forests journal as it is written and presented. I suggest the authors to revise the manuscript according to my suggestions. Therefore, I recommend that a major revision is warranted. I explain my concerns in more detail below and I ask the authors address my comments in their response.

Response: Thank you for these constructive comments! Please see our response below and our revised manuscript (important revisions have been marked red).

1. Lines 79-80. The meaning is not clear, please revise

Response: We are grateful for the comment. We revised this sentence. This study was conducted in the old-growth temperate forest, which was a 9-ha forest dynamic plot (FDP) in the Liangshui national reserve (47°10′50″N，128°57′20″E) of northeastern China.

2. Line 84. … dark brown forest according to the Chinese…

Response: We are grateful for the comment. We revised this sentence. The dominant soil is dark brown soil per the Chinese soil classification system.

Results section

3. The authors do not report anything or at least some descriptive statistics about the RGR and MR that have calculated – estimated by the formulae reported in lines 94-96.

Response: Thanks for your comment. We calculated RGR and MR by formula and show in Table 1. Please see Table 1. RGR and MR for saplings (1 cm ≤DBH <10 cm) and large trees (DBH≥ 10 cm) in the 9-ha dynamic plot of typical mixed broadleaved-Korean pine forest.

4. Line 148. These variables concern the MR and the RGR, as far as I can read from table1.

Response: Thanks for your comment. Sorry for the confusions. We revised this section to align the Table 2 with the presentation.

5. There is a mistake here either all Pearson’s significant correlations marked with asterisk are not shown in tab 1 or there are mistakes in the text in lines 148-151.

Response: We apologize for the inconsistent statements in the original manuscript. There were mistakes in the text. We revised this section to align all Pearson’s significant correlations with the presentation.

6. Line 151. …VWC had a significantly negative correlation with MR. -> This result concerns the large trees and the RGR not the MR.

Response: We are grateful for the comment. There were mistakes in the text. We had revised this sentence.

7. Table 1. The significant pearson correlation values marked with asterisk do not correspond with those reported in the text.

Response: We are grateful for the comment. The content of table 2 was correct, the content in the text was wrong, we revised the content of this part of the manuscript, sorry again for our mistakes.

8. Lines 161-173. Why the authors haven’t reported biplot graphs for showing the principal component axes and the related variable graphically. I think that biplots would have been more informative depicting better the relationships and much easier for the reader to follow. So I strongly urge the authors to include biplots in their revisions.

Response: Thanks for your proposal. We made principal component analysis diagrams to give readers a better understanding. Please see Figure 2. PCAs for topographical and soil characteristics of RGR and MR in saplings and large trees.

9. Line 163. What about vars VWC and BD?

Response: Thanks for your comment. We chose the first two factors for each PC axis, e.g. axis 1 (PC1) of sapling-RGR, OC(0.397)>AP(0.395)>VWC(-0.388)>BD(0.357). Therefore, we showed OC and AP in the manuscript.

10. Line 166. What about vars Con, Slo and VWC? There are more variables with large negative or positive pca coefficients in tab 3 that are not mentioned in this paragraph.

Response: Thanks for your comment. We compared the absolute value of each factor in each PC axis, and selected the first two as our reference. PC=PC1×Variance contribution(%)+PC2×Variance contribution(%)+PC3×Variance contribution(%), we used this formula to calculate the highest value(Table 2 and 4).

11. Line 181. This subsection must be numbered 1.2 instead of 1.1.

Response: We are grateful for the comment. We made changes here, and revised content related to this.

12. Line 199. …than for large trees…

Response: Thanks for your comment. Done. please see line 199.

13. Line 201. I’m not convinced for this result

Response: We are grateful for the comment and sorry for the unclear statement. This result has been modified. This result is similar to the effect of density dependence for sapling to large tree transitions[22]. Although trees in both life stages are affected by this effect, the effect is more pronounced in the early stages showing higher growth rates and mortality in an old-growth temperate forest.

14. Line 206. …stronger than that of saplings…

Response: Thanks for your comment. Done. please see line 206.

15. Lines 219-238. As I’ve mentioned in my general comments, the way the results are written and presented the reader cannot confirm the findings in the discussion section. MR od saplings, for example, “it is increased with reduced water availability” and the only numbers that are reported in results are those of VWC in tab 3 for the three pca axes 0.213 0.34 0.221 respectively. What are these numbers mean and how they are explaining what is written in the discussion?

Response: We are grateful for the comment and sorry for the unclear statement. This section has been modified. We compared the absolute value of each factor, and selected the first two as our reference. Phosphorus plays a special role in energy metabolism, is a component of many coenzymes, and is directly involved in oxidative phosphorylation and photosynthetic phosphorylation. The phosphorus in plants is mainly absorbed from soil nutrients. Thus, enhanced supplies of TP and AP have the potential to promote growth in saplings and large trees, respectively, because of increased light availability. Interestingly, the MR of large trees showed similar responses to the RGR of saplings on elemental availability, possibly because the production of fine roots was higher in poor soils than in fertile soils[35]. The lifespan of large trees can be shortened in high TP soils[36]. In PCA of Saplings and MR, VWC of the three pca axes were 0.213, 0.34, 0.221, respectively. These three values were not the highest in each PCA axis, so the discussion was wrong and would be deleted.

16. Lines 233-234. Does this sentence here results from your analysis in the results section? where?

Response: Thanks for your comment. Sorry for the confusions. PCA showed the highest values related to increasing TN in RGR of large trees(Table 2 and 4). We added a location to the manuscript that reflected the result. PC=PC1×Variance contribution(%)+PC2×Variance contribution(%)+PC3×Variance contribution(%), we used this formula to calculate the highest value(Table 2 and 4).

17. Lines 239-258. In this paragraph the habitat filtering is discussed but there nothing for this term in the M & M and the Results sections

Response: We added the concept of “habitat filtering”, and added “Habitat filtering is evidenced if RGR or MR of small trees is not related to topographic and soil factors, but that of large trees is.” in materials and methods.

Author Response File: Author Response.doc

Round 2

Reviewer 2 Report

The authors have revised the manuscript a lot. My comments concerns tmainly he interpretation of the biplots in figure 2. See my comments below:

Line 102
dominant soil is dark brown soil according to the Chinese soil classification system

Line 157
Table 1 What the values 0.049, 0.015 etc represent? Percentages? clear decimal numbers? Please give units of the values presented in the table

Figure 2
The authors must give a more detailed explenation in the results section in the text regarding the biplots. The loading for example that represented by the arrows give information that has to be shown.
When two arroes are close, forming a small angle, the two variables they represent are positively correlated. Example: TN and VWC in Sap-MR, OC,AN in Lar-RGR, BD and Con, Slo and Ele in Lar-MR. Some other arrows that connected by a 90° angle are not corre;ated at al example BD and Slo , VWC and TP in Sap-RGR or Ap and OC in Sap-MR. Some others are connected with a very lagre angle ~180° showing strong negative correlation such as An, OC and Bd in Lar-RGR Cos(Asp) and Con and Slo in Sap-RGR. They also must discuss these findings in the discussion section arguing how the topographic and soil factors affect the RGR and MR of saplings and large trees comparing their findings with other similar finding from the literature.

Author Response

Response to the Comments

Dear editors and reviewers,

Re: Manuscript ID: forests-1717513 and Effects of Habitat Filtering on Tree Growth and Mortality Across Life Stages in An Old-Growth Temperate Forest

Thank you for your letter and the reviewers’ comments concerning our manuscript entitled “Effects of habitat filtering on tree growth and mortality across life stages in an old-growth temperate forest”. Those comments are valuable and very helpful. We have read through comments carefully and have made corrections. Based on the instructions provided in your letter, we uploaded the file of the revised manuscript. Revisions in the text are shown using red highlight for additions. The corrections in the paper and our responses to the reviewer's comments are presented following.

Reviewer:

The authors have revised the manuscript a lot. My comments concerns tmainly he interpretation of the biplots in figure 2.

Response: Thank you for these constructive comments! Please see our response below and our revised manuscript (important revisions have been marked red).

1. Lines 102. dominant soil is dark brown soil according to the Chinese soil classification system

Response: We are grateful for the comment. We revised this sentence. The dominant soil is dark brown soil according to the Chinese soil classification system.

2. Line 157. Table 1 What the values 0.049, 0.015 etc represent? Percentages? clear decimal numbers? Please give units of the values presented in the table

Response: Thanks for your comment. We gave units of the values presented in the table 1, and the units of the values were also given in materials and methods 2.1. Table 1. RGR(cm cm^-1 year^-1) and MR(year^-1) for saplings (1 cm ≤DBH <10 cm) and large trees (DBH≥ 10 cm) in the 9-ha dynamic plot of typical mixed broadleaved-Korean pine forest.

Results section

3. Figure 2
   The authors must give a more detailed explenation in the results section in the text regarding the biplots. The loading for example that represented by the arrows give information that has to be shown.
   When two arroes are close, forming a small angle, the two variables they represent are positively correlated. Example: TN and VWC in Sap-MR, OC,AN in Lar-RGR, BD and Con, Slo and Ele in Lar-MR. Some other arrows that connected by a 90° angle are not corre;ated at al example BD and Slo , VWC and TP in Sap-RGR or Ap and OC in Sap-MR. Some others are connected with a very lagre angle ~180° showing strong negative correlation such as An, OC and Bd in Lar-RGR Cos(Asp) and Con and Slo in Sap-RGR. They also must discuss these findings in the discussion section arguing how the topographic and soil factors affect the RGR and MR of saplings and large trees comparing their findings with other similar finding from the literature.

Response: Thanks for your detailed comment. We used Figure 2 to analyze the effects of topography and soil factors on the RGR and MR of saplings and large trees, as well as the relationship between the factors.

Results

in Sap-RGR, Asp (Sin (Asp) and Cos (Asp) ) was strongly negatively correlated with Con and Slo. In Sap-MR, OC and VWC showed a strong positive correlation, whereas OC was strongly negatively correlated with Slo. In Lar-RGR, OC and Slo showed a strong positive correlation. In Lar-MR, OC was strongly negatively correlated with Slo.

Discussion

Sin (Asp) and cos (Asp) were used to represent north-south and east-west, respectively [37]. The results showed that the more sunny the slope was, the higher RGR of saplings was. Con and Slo were negatively correlated with the RGR of saplings (Figure 2 Sap-RGR). The results of this study were similar to those of Shen et al. in subtropical forest [38].The negative correlation between convexity and slope and RGR of saplings is likely associated with greater moisture and nutrient accumulation in local lows. High convexity and slope may indicate a hilltop, low RGR of saplings is probably associated with greater run-off of water at high convexity and slope [39]. In contrast, high RGR of large trees was associated with high convexity and slope. This may be caused by the strong competitiveness and wind resistance of the large trees. The negative correlation between OC and Slo of Sap-MR is stronger than that of Lar-MR because of the larger angle between OC and Slo, this indicates the variation in the MR in saplings in habitats with high soil organic carbon was significantly higher than that in large trees. In Sap-MR, OC and Slo showed a strong negative correlation, whereas in Lar-RGR OC and Slo showed a strong positive correlation. This was consistent with Dina's findings that species-habitat association status significantly affected demographic parameters [40].

References

37. Jones, M.M.; Tuomisto, H.; Borcard, D.; Legendre, P.; Clark, D.B.; Olivas, P.C. Explaining variation in tropical plant community composition: influence of environmental and spatial data quality. Oecologia 2008, 155, 593-604.
38. Shen, Y.; Santiago, L.S.; Shen, H.; Ma, L.; Lian, J.; Cao, H.; Lu, H.; Ye, W. Determinants of change in subtropical tree diameter growth with ontogenetic stage. Oecologia 2014, 175, 1315-1324.
39. Farciaoliva, F.; Maass, J.M.; Galicia, L. Rainstorm analysis and rainfall erosivity of a seasonal tropical region with a strong cyclonic influence on the Pacific Coast of mexico. Journal of Applied Meteorology 1995, 34, 2491–2498.
40. Dina, O.; Jin, G. Species-habitat association affects demographic variation across different life stages in an old-growth temperate forest. Perspectives in Pant Ecology, Evolution and Systematics 2019, 125482.

Author Response File: Author Response.doc