Climate-Sensitive Diameter Growth Models for White Spruce and White Pine Plantations

Round 1

Reviewer 1 Report

General comment

This manuscript develops climate-sensitive diameter growth models for two species and simulates trends in diameter increase or decrease under future climate change scenarios. Manuscript is written generally well. Data and methods used in the study are adequate, and both the methods and results presented can be of interest to the audience of the journal. I have a following comments to improve MS quality.

Specific comments

1.      Line 11. You add White pine (Pinus strobus L.).

2.      Please check the location of the dependent variables for eqs 2 and 3.

3.      Line 217: Perhaps the authors need to make a table comparing the fit performance of the various models. A gradual upgrade from a base model to a climate-sensitive model might be an interesting process for readers.

4.      For Figure 2, please use the scattered plot graphs, not the Average inside bark diameter at breast height.

5.      Line 260: Does the simultaneous introduction of the same climate variable and its transformed form into the model cause multicollinearity?

6.      Line 278 and Line 293, eqns. (2) and (3), eqs. (2) and (3) should be consistent.

7.      Lines 289-291: This phrase seems to imply an interaction between the variables; however, the model does not include an interaction term.

8.      Line 299, I am curious this sentence “These plots indicated that autocorrelation…”, how residual plot indicate that autocorrelation? What kind of aucorrelation function would be considered?

9.      Table 4 need a line below the first row.

10.   Line 361: Here or elsewhere, authors are advised to add units when expressing the magnitude of change in diameter.

11.   Line 372: Discussions need to be strengthened. In some places it seems to be a mere restatement of the results, such as in the last paragraph.

12.   Line 421: Perhaps the font is incorrect due to a printing error, please correct it.

13.   Tables, diagrams, typesetting, abbreviations and some annotations are not standardized in this paper, For example, RCP 2.6 and 8.5 or RCP 26 and 85 for Fig 4 and 5; Table 4 and 5 have different format, It seems to be very casual, so it is suggested to modify them through the whole MS.

Author Response

General comment 

This manuscript develops climate-sensitive diameter growth models for two species and simulates trends in diameter increase or decrease under future climate change scenarios. Manuscript is written generally well. Data and methods used in the study are adequate, and both the methods and results presented can be of interest to the audience of the journal. I have a following comments to improve MS quality. 

Answer: Thank you for your encouraging comment. I appreciate it. All changes/additions made to address comments are in RED.

Specific comments

1. Line 11. You add White pine (Pinus strobus L.).

Answer: white pine has been added.

2. Please check the location of the dependent variables for eqs 2 and 3.

Answer: Thank you. Locations of dependent variables have been fixed now.

3. Line 217: Perhaps the authors need to make a table comparing the fit performance of the various models. A gradual upgrade from a base model to a climate-sensitive model might be an interesting process for readers.

Answer: Parameter estimates and fit statistics for the models without and with climate variables are listed in Table 2 and 3, respectively. To avoid changes to be made in the text by combining these tables, both tables have been kept. Readers can compare parameter estimates and fit statistics from these tables.

4. For Figure 2, please use the scattered plot graphs, not the Average inside bark diameter at breast height.

Answer: Scattered plot graphs have been made as suggested and previous graphs have been replaced by these new graphs.

5. Line 260: Does the simultaneous introduction of the same climate variable and its transformed form into the model cause multicollinearity?

Answer: The quadratic transformation of the climate variable made sense in the model as its coefficient had opposite sign to the linear form. That suggests the linear (positive/negative) effect of the climate variable will diminish as the value of the climate variable increases. The multicollinearity was not the issue here.

6. Line 278 and Line 293, eqns. (2) and (3), eqs. (2) and (3) should be consistent.

Answer: These have been fixed now.

7. Lines 289-291: This phrase seems to imply an interaction between the variables; however, the model does not include an interaction term.

Answer: Yes. The interaction was to imply between climate variables but none of interactions were significant.

8. Line 299, I am curious this sentence “These plots indicated that autocorrelation…”, how residual plot indicate that autocorrelation? What kind of aucorrelation function would be considered?

Answer: Thank you! It was meant for heteroscedasticity. I removed the autocorrelation from the text now.

9. Table 4 need a line below the first row.

Answer: The line has been added now.

10. Line 361: Here or elsewhere, authors are advised to add units when expressing the magnitude of change in diameter.

Answer: units (cm) have been added now where they were missing before.

11. Line 372: Discussions need to be strengthened. In some places it seems to be a mere restatement of the results, such as in the last paragraph.

Answer: The discussion has been revised.

12. Line 421: Perhaps the font is incorrect due to a printing error, please correct it.

Answer: This has been fixed now.

13. Tables, diagrams, typesetting, abbreviations and some annotations are not standardized in this paper, For example, RCP 2.6 and 8.5 or RCP 26 and 85 for Fig 4 and 5; Table 4 and 5 have different format, It seems to be very casual, so it is suggested to modify them through the whole MS.

Answer: RCP values have been fixed in figures 4 and 5. The format in the tables have also been fixed now.

Reviewer 2 Report

From my point of view, the paper needs major revision. Also I would like to see answer to some of my comments. If possible, I would like to see some data in the supplementary material.  

First, these models are perfectly suitable for the even aged forests. What about uneven aged forests? Also, couldn't you use artificial intelligence methods to present the models?

Second, why are other factors not included in the presented models, such as the factors of the stand competition factors, slope, direction, and elevation above sea level in the models?

The introduction section is short. There is no discussion about how to choose climate scenarios. The purpose and necessity of the research is not well defined. Also, the text of the article needs more references that are not used at all. For example, lines 36 to 41

  In relation to climate scenarios, no explanation is given in the introduction and method. Why these scenarios were chosen and what are their characteristics.

Please check these articles,

In the results section, it is better to add a table or a figure about the effect of climate on climatic factors, to what extent the temperature or humidity changes under the influence of climate change scenarios?

For example, refer to the following research.

"Modeling Tree Growth Responses to Climate Change: A Case Study in Natural Deciduous Mountain Forests." Forests 13, no. 11 (2022): 1816.

Projected biodiversity in the Hyrcanian Mountain Forest of Iran: an investigation based on two climate scenarios. Biodivers Conserv 32, 3791–3808 (2023). https://doi.org/10.1007/s10531-022-02470-1

In the discussion, climate change should be discussed more and the reason why other factors were not included in the models should be discussed. Also, under what conditions can these models be used for mixed forests?

Author Response

From my point of view, the paper needs major revision. Also I would like to see answer to some of my comments. If possible, I would like to see some data in the supplementary material. First, these models are perfectly suitable for the even aged forests. What about uneven aged forests? Also, couldn't you use artificial intelligence methods to present the models? Second, why are other factors not included in the presented models, such as the factors of the stand competition factors, slope, direction, and elevation above sea level in the models?

The introduction section is short. There is no discussion about how to choose climate scenarios. The purpose and necessity of the research is not well defined. Also, the text of the article needs more references that are not used at all. For example, lines 36 to 41 

 In relation to climate scenarios, no explanation is given in the introduction and method. Why these scenarios were chosen and what are their characteristics.

Please check these articles,  

In the results section, it is better to add a table or a figure about the effect of climate on climatic factors, to what extent the temperature or humidity changes under the influence of climate change scenarios?

For example, refer to the following research.

"Modeling Tree Growth Responses to Climate Change: A Case Study in Natural Deciduous Mountain Forests." Forests 13, no. 11 (2022): 1816. 

Projected biodiversity in the Hyrcanian Mountain Forest of Iran: an investigation based on two climate scenarios. Biodivers Conserv 32, 3791–3808 (2023). https://doi.org/10.1007/s10531-022-02470-1 

In the discussion, climate change should be discussed more and the reason why other factors were not included in the models should be discussed. Also, under what conditions can these models be used for mixed forests?

Answer:

Thank you for your constructive comments. All changes/additions made to address reviewers’ comments are in RED.

I have presented the summary of the data used in this study in Table 1 and added a table (Table 6) that displays the summary of projected climate variables for sample sites used in this study. I have not used the data from uneven aged stands in analyzing climate effects of diameter growth in this study. Therefore, I have added some text to caution the users of these models at the end of “Discussion section” lines 450-456. Using artificial intelligence (AI) methods was beyond the scope of this study. Therefore, AI methods were not used to present the models. Similarly, the objective of this study was to develop diameter growth models by incorporating climate variables. As a result, competition factors were not used here. However, other site related variables (elevation, latitude, longitude, etc.) were used which is described in methods (lines 173-174 and 261-262).

I have added more citations in the introduction (lines 36-41). I also added the citation listed in the comments that was relevant. The text (in the introduction and discussion) has been extended to incorporate these citations.

The reason why these 3 climate change scenarios are chosen are mentioned in the methods (lines 216-218). I think describing climate change in the introduction is not necessary. I have conducted several studies on climate effects on tree growth and published them in various journals (as shown below for you information) and hence referred to one of the studies about the description of climate variables here.

As mentioned earlier, Table 6 has been added to the end of Results section to present the projected values of climate variables for the 80 year growth period (2021-2100).

The Discussion section has been modified to incorporate newly cited studies and limitations of models (lines 436-440and 450-456). Thank you!

1. Sharma, M. 2022. Modeling climate effects on site productivity of plantation grown jack pine, black spruce, red pine, and white spruce using annual/seasonal climate values. Forests 2022, 13, 1600. https://doi.org/10.3390/f13101600.
2. Sharma, M. 2022. Climate effects on black spruce and trembling aspen productivity in natural origin mixed stands. Forests2022, 13(3), 430, https://doi.org/10.3390/f13030430
3. Sharma, M. 2021. Climate effects on jack pine and black spruce productivity in natural origin mixed stands and site index conversion equations. Trees, Forests and People, https://doi.org/10.1016/j.tfp.2021.100089.
4. Sharma, M. 2021. Modelling climate effects on diameter growth of red pine trees in boreal Ontario, Canada. Trees, Forests and People, https://doi.org/10.1016/j.tfp.2021.100064.
5. Sharma, M. and J. Parton. 2019. Modelling the effects of climate on site productivity of white pine plantations. Canadian journal of forest research 49: 1289-1297.

6.      Sharma, M. and J. Parton. 2018. Climatic effects on site productivity of red pine plantations. Forest Science 64:544-554.

7.      Sharma, M. and J. Parton. 2018. Analyzing and modelling effects of climate on site productivity of white spruce plantations. Forestry Chronicle 94: 173-182.

8. Sharma, M., Subedi, M. Ter-Mikaelian, and J. Parton. 2015. Modeling climatic effects on stand height/site index of plantation-grown jack pine and black spruce trees. Forest Science 61: 25-34.
9. Subedi, N. and M. Sharma. 2013. Climate-diameter growth relationships of black spruce and jack pine trees in boreal Ontario, Canada. Global Change Biology, 19: 505–516.

Reviewer 3 Report

Journal: Forests (ISSN 1999-4907)

Manuscript ID: forests-2762906

Title: Climate Sensitive Diameter Growth Models for White Spruce and White Pine Plantations

Overall  Comments and Suggestions for Authors

To authors,

Regarding the climate-sensitive diameter growth models for Pinus strobus L. and Picea glauca (Moench) Voss plantations in Ontario, Canada, this manuscript may be interesting to the relevant researchers who deals with similar issues such as climate-related forest biometrics and tree growth models especially in the region. Overall, the manuscript contents were clearly described, and structured well enough. I do not find any severe flaws or drawbacks for major revision or rejection. Instead, I pointed out several issues that the author may want to consider. I wish the model can help to understand the diameter growth of the species under the climate change.

I hope that this manuscript can be improved based on peer-review’s comments. My specific comments were provided in detail as follows.

Kind regards,

Reviewer

Point 1.

In Materials section, if data were from monoculture plantations, wasn’t it possible to check the stand biological age? Was it or not? Also, why did the author use the breast height age? Instead of stump age or biological age? Wouldn’t it be more applicable for practical forestry?

Point 2.

In Methods sections, all methodological processes were clearly explained.

By the way, why did the author not use the stand density or basal area as fixed effect? I assumed that the stand density may influence the diameter growth. Wasn’t it significant? Or did the author consider the stand characteristics just as random-effects? From my perspective, there must be some parts that can be explained as fixed-effects instead of random-effects.

 Point 3.

In Results section, one may wonder about the model structure. Wasn’t there any chance to make it parsimonious? Wouldn’t it be better to make it simple if available? From my point of view, many of climatic variables in the shape parameter cannot indicate a various model pattern. It eventually expressed as a shape parameter after the computation. Also, I don’t think 40 sites were enough to consider as many variables as the author used. Taking into account the number of observations and variables, the selected climatic variables were relatively a lot.

Point 4.

In Result section, I appreciate the residual plots over predicted diameter that the author provided. Although it was a little bit biased, the model can be available within the range as the author mentioned. However, I also wondered about the other type of residuals, e.g., the residual plots over age and other climatic predictors the author applied? Were they not biased, either? I consider it be important to make it convincing more.

Point 5.

In Discussion, can the author suggest any practical range of models such as age and/or climatic predictors? It will be useful to simulate the model in practice. Some cautions and limitations as well as recommendation can be supplemented in the Discussion.

Point 6.

Minor point.

Line 11: better add common name both species “for (Pinus strobus L.) and white spruce 11 (Picea glauca (Moench) Voss)”

Author Response

Overall comments and Suggestions for Authors

Regarding the climate-sensitive diameter growth models for Pinus strobus L. and Picea glauca (Moench) Voss plantations in Ontario, Canada, this manuscript may be interesting to the relevant researchers who deals with similar issues such as climate-related forest biometrics and tree growth models especially in the region. Overall, the manuscript contents were clearly described, and structured well enough. I do not find any severe flaws or drawbacks for major revision or rejection. Instead, I pointed out several issues that the author may want to consider. I wish the model can help to understand the diameter growth of the species under the climate change.

I hope that this manuscript can be improved based on peer-review’s comments. My specific comments were provided in detail as follows.

Answer:

Thank you for your positive and constructive comments. I appreciate that. All changes/additions made to address reviewers’ comments are in RED.

Point 1.

In Materials section, if data were from monoculture plantations, wasn’t it possible to check the stand biological age? Was it or not? Also, why did the author use the breast height age? Instead of stump age or biological age? Wouldn’t it be more applicable for practical forestry? 

Answer:

As mentioned in the manuscript, tree growth (height and diameter) before the breast height was erratic. That erratic growth was not necessarily due to climate conditions. There could have been other biotic and abiotic factors affecting seedling growth. To avoid the erratic growth, breast height age was used in this study.  

Point 2.

In Methods sections, all methodological processes were clearly explained.

By the way, why did the author not use the stand density or basal area as fixed effect? I assumed that the stand density may influence the diameter growth. Wasn’t it significant? Or did the author consider the stand characteristics just as random-effects? From my perspective, there must be some parts that can be explained as fixed-effects instead of random-effects. 

Answer:

This is an excellent comment. Initially, the objective was to incorporate climate effects on diameter growth but not the stand density. However, after receiving these comments, I got curious about the stand density effects on diameter growth. Therefore, I expressed the asymptote and the shape parameter in terms of stem density (stems/ha and BA/ha) and fit the models to the data. However, both stand density terms (stem density and BA) were not significant in the regression. This could be because of the lack of tracking stand density information from the beginning. Had we tracked the stand density from the beginning and used the dynamic values in regression, at least one of them could have been significant.

 Point 3.

In Results section, one may wonder about the model structure. Wasn’t there any chance to make it parsimonious? Wouldn’t it be better to make it simple if available? From my point of view, many of climatic variables in the shape parameter cannot indicate a various model pattern. It eventually expressed as a shape parameter after the computation. Also, I don’t think 40 sites were enough to consider as many variables as the author used. Taking into account the number of observations and variables, the selected climatic variables were relatively a lot. 

Answer:

I agree that there were a lot of climate variables (68). However, there were 5 trees from each site. Since all 5 trees did not reach the breast height during the same calendar year due to erratic early growth, average diameter could not be calculated for each site. As a result, there were 40*5=200 trees for each species in the data set. In addition, I did not used all 68 climate variables at the same time to fit the models. I used one climate variable at a time to fit the model for each species. As far as the parsimonious is concerned, the model looks complicated, but it is very easy to use as there are only 4 climate variables and age as dependent variables for each species. In the absence of climate variables, the model could not be simpler.

Point 4.

In Result section, I appreciate the residual plots over predicted diameter that the author provided. Although it was a little bit biased, the model can be available within the range as the author mentioned. However, I also wondered about the other type of residuals, e.g., the residual plots over age and other climatic predictors the author applied? Were they not biased, either? I consider it be important to make it convincing more.  

Answer:

I did not think of residual plots over age and other climate variables before. I just plotted residuals against breast height age and the plots for both species looked even better than the ones for predicted diameters. All residuals are clustered around the zero line (as shown below for white pine) and there was no bias for higher age for both species. This has been mentioned in the text (lines 306-308). The plots were similar over the climate variables also.

Point 5.

In Discussion, can the author suggest any practical range of models such as age and/or climatic predictors? It will be useful to simulate the model in practice. Some cautions and limitations as well as recommendation can be supplemented in the Discussion. 

Answer:

I have added some cautions and limitations at the end of Discussion section now (lines 450-456)

Point 6.

Minor point.

Line 11: better add common name both species “for (Pinus strobus L.) and white spruce 11 (Picea glauca (Moench) Voss)”

Answer:

Common name (white pine) has been added for Pinus strobus L. now. Thank you!

Author Response File: Author Response.pdf

Reviewer 4 Report

Dear Authors,

Please find my comments in the attached file.

Also, overcitation of yourself is too evident in this paper.

Regards

Comments for author File: Comments.pdf

Author Response

All changes/additions made to address reviewers’ comments are in RED.

 How the possition of the sample plot was selected?

Answer: The position of the plot was selected to represent tree population on the site. It is clearly stated in the manuscript now.

Validation of your models on independent data is missing.

Answer: Unfortunately, independent data were not available for validation.

The statements in this paragraph are proved only by using visualization of data. I think proper statistical evaluation is missing, for example scatter plot of modelled and measured values would be usefull.

Answer: Scattered plots of measured values of diameters have been displayed in Figure 2 now.

The comparison of observed and modeled values was done only for climate independent model.

Please add additional graph or draw on the same figure 2 how the introduction of climatic variables improved your predictions.

Answer: These comparisons were to check whether the model applied was appropriate to model diameter growth or not. Comparison with and without climate variables are made in Figures 4 and 5.

it is very difficult to see the trends if they are here. Thus you have to add appropriate trend line or loess regression line to see whre model works and where it does errors

Answer: Trend lines have been added in the graph now.

It would be more clear to see ploted observed values against modeled in the scatter plot. Also inclussion of precision and accuracy of the model would be useful.

Answer: The residual plots for diameter class are shown in Figure 3. These plots by age class were similar to the ones by diameter class. It has been stated in the text now. For example, the plot by age class was even better than the one by diameter class as shown below.

It would be more clear to see ploted observed values against modeled in the scatter plot.

Answer: Once again, these plots have been made in Figure 3.

I think there is no need to have such large figures. To save the space they could be reduced and located side by side.

Answer: These plots were made large to clearly see the effects of climate on diameter growth. I would leave it to manuscript editors. They can reduce the size of the plots.

I think overcitation of yourself it is not a bonus for your paper. There are more authors that worked on the moddelling as well.

Answer: I wanted to emphasize that the climate effects depend not only on tree species but also on location and the tree species it is growing with if it is grown in mixed stands. I did not have any intention to overcite a paper. I have published 9 papers on climate effects but only have cited a couple of them in this manuscript.

Different formating. Be constant in it.

Answer: The format has been fixed now.

This is not the conclusion but rather summary of your work.

Answer: As I understand, the conclusion should be stand alone. That might be the reason it may look like as summary. Since other reviewers have not made any comments on this, I have not changed the conclusion.

In table 1 is presented only description of the data but not the data itself.

Answer: I have restated this as “Summary statistics of the data used in this study have been presented in Table 1” now. Thank you so much for your constructive comments!

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I appreciate the authors’ efforts in addressing my comments. The revised manuscript has been improved significantly. I am now satisfied with authors’ revision and have no further comments. Therefore, I would recommend accepting the paper for publication.

Reviewer 4 Report

Dear Authors,

Please take a look to the figure 3 once again. it is very difficult to see the trend line in the presented file forests-2762906-v2 clean.

Also please check your reference style.

Regards