Estimating Position, Diameter at Breast Height, and Total Height of Eucalyptus Trees Using Portable Laser Scanning

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. Please avoid using abbreviations in titles, especially if they are not widely known by the general public.
2. The results of this study should not only serve forestry monitoring in Brazil, so the significance and necessity of this study should be described in a broader context in the background of the study; Please rewrite the first paragraph of Introduction;
3. Please avoid citing like line 59, as each study has its own unique results. If all of these results are useful for this study, please list them all. Full text check and fix this error;
4. Figure 1 needs to be reset, the current picture is difficult to identify the location of the study area, and please change the annotation method of latitude and longitude to use N, E, S, W;
5. The annotations in Figure 5 are too small to be illegible, and Figure 6 has the same problem;
6. Table 2 is proposed to be placed in appendix;
7. The conclusion needs to be carefully considered, and even if this experiment is not done, it can be concluded that the decrease in point cloud density will affect the accuracy of the measurement. In addition, it is necessary to reorganize the structure of the conclusions and point out the superiority of the method proposed in this study, because the semi-automatic processing method will increase the time and labor costs.

Author Response

1. Please avoid using abbreviations in titles, especially if they are not widely known by the general public.

         Thanks. Portable laser scanning was inserted, and SLAM was removed. In red in the title.

2. The results of this study should not only serve forestry monitoring in Brazil, so the significance and necessity of this study should be described in a broader context in the background of the study; Please rewrite the first paragraph of the introduction.

        Thanks. We rewrote the first paragraph. Lines 49-53.

3. Please avoid citing like line 59, as each study has its own unique results. If all of these results are useful for this study, please list them all. Full text check and fix this error.

        Thank you. We removed references that are not important to our study. References # 3-7, 9-11, and 13. Those that are important are listed in red in the text. Lines 83-91 and 105-107.

4. Figure 1 needs to be reset; the current picture is difficult to identify the location of the study area, and please change the annotation method of latitude and longitude to use N, E, S, and W.

         We have changed the figure, thank you.

5. The annotations in Figure 5 are too small to be illegible, and Figure 6 has the same problem;

         We have changed the figures, thank you.

6. Table 2 is proposed to be placed in appendix;

         Thank you; we have inserted the table as an appendix.

7. The conclusion needs to be carefully considered, and even if this experiment is not done, it can be concluded that the decrease in point cloud density will affect the accuracy of the measurement. In addition, it is necessary to reorganize the structure of the conclusions and point out the superiority of the method proposed in this study, because the semi-automatic processing method will increase the time and labor costs.

         Thanks. We changed the conclusions, lines 544-558.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a study of PLS-SLAM for estimating tree structural attributes (DBH and height) in Eucalyptus plantations. The work explores the influence of point cloud density on estimation accuracy, which is of practical significance for operational forest inventory. However, the current version of the manuscript suffers from issues related to scientific writing quality, logic flow, interpretation of results, and formatting consistency. Below are specific comments:

1. Line 17–20, Page 1: The phrase “Based on the factors mentioned above” is vague, and the logical link to the study objective (the objective of this study was to evaluate the accuracy of different point densities) is weak. Please clearly state the research goal and contextualize it in terms of the challenges or gaps in applying PLS-SLAM for forest inventories.
2. Line 23, Page 1: “> 100 points.m-2” should be modified.
3. Line 24, Page 1: “Successfully segmenting” is vague. Do you mean individual tree segmentation? Please specify what is “successfully”.
4. Line 25–26, Page 1: Clarify what (5.9%) and (4.2%) refer to.
5. Line 26, Page 1: “36,000 returns.m-2” is hard to interpret.
6. Line 29–34, Page 1: Statements such as “reduce the accuracy”, “PLS-SLAM has potential…” are vague without data to support them. Consider adding quantitative conclusions here.
7. Line 39, Page 1: “75.8% are a fast-growing species.” → should be corrected to “75.8% consist of fast-growing species.” or “are fast-growing species.”
8. Line 84–86, Page 2: The structure “such as: Does this new paradigm...” is both grammatically incorrect and stylistically too informal for a scientific article. Replace this with a list of concerns expressed as noun phrases (e.g., “concerns about accuracy….”).
9. Line 95, Page 3 and Line 162, Page 4: DBH and H were already defined earlier — no need to redefine. Just use the abbreviations. Please remove the repeated definitions.
10. Line 109–110, Page 3: The mention of Köppen-Geiger classification should be supported with a reference.
11. Line 117, Page 3: “1.30 m high” → suggest rephrasing to “1.30 m above ground level (AGL)” for clarity.
12. Line 119, Page 3: “the average if” → should be “the average of”.
13. Equation (1), Page 4: tan should be in upright Roman font. Use a multiplication dot (·) between tan and d. Also, the text defines d = distance to the tree, but Fig. 2 uses variable I. Please standardize the notation across text and figures.
14. Line 151, Page 4: “.. (x/y/z)”. Suggest removing.
15. Line 187, Page 4: “sloop smooth” → should be “slope smooth”.
16. Line 188, Page 4: “weighing” → should be “inverse distance weighting”.
17. Line 245–246, Page 6: You state that only one individual was not identified at 500 and 100 pts·m². However, from Fig. 7c, it appears two trees were missing. Please double-check and clarify.
18. Line 263, Page 6: “Accuracy was slightly higher for larger trees” contradicts Table 1 on Page 9. RMSE for DBH > 27.3 cm is higher, not lower. This interpretation needs correction.
19. Line 350–356, Page 8: The word “All” should be lowercase.
20. Line 376, Page 9: “Ransac” → should be capitalized as “RANSAC”.
21. Line 398, Page 9: You mention “efficiency” without conducting a time or labor analysis. This statement should be either removed or supported by comparative experiments.
22. The manuscript evaluates the influence of different point cloud densities (e.g., 36,000, 500, 100 points·m⁻²) on tree attribute estimation. However, the method used to reduce point cloud density is not clearly described. Was the downsampling performed via random sampling, voxel grid filtering, or another spatial method? Please add it.
23. The number of trees used in the evaluation (n = 71) appears to be quite limited given the stratification across point densities and size classes (e.g., small vs. large DBH). With such a small sample, it is challenging to draw statistically robust conclusions, especially regarding trends like “accuracy decreases more for smaller trees” or “height estimation is more affected in taller trees.”

Overall, the manuscript contains many informal phrases and sentences. A thorough revision by a native English speaker or professional editing service is highly recommended to improve clarity and readability.

Author Response

This manuscript presents a study of PLS-SLAM for estimating tree structural attributes (DBH and height) in Eucalyptus plantations. The work explores the influence of point cloud density on estimation accuracy, which is of practical significance for operational forest inventory. However, the current version of the manuscript suffers from issues related to scientific writing quality, logic flow, interpretation of results, and formatting consistency. Below are specific comments:

1. Line 17–20, Page 1: The phrase “Based on the factors mentioned above” is vague, and the logical link to the study objective (the objective of this study was to evaluate the accuracy of different point densities) is weak. Please clearly state the research goal and contextualize it in terms of the challenges or gaps in applying PLS-SLAM for forest inventories.

         Thanks, modified in lines 19-22.

2. Line 23, Page 1: “> 100 points. m-2” should be modified.

         Thanks, modified in lines 26-28.

3. Line 24, Page 1: “Successfully segmenting” is vague. Do you mean individual tree segmentation? Please specify what “successfully” is.

         Thanks, modified in lines 26-31.

4. Line 25–26, Page 1: Clarify what (5.9%) and (4.2%) refer to.

         Thanks, modified in lines 30-31.

5. Line 26, Page 1: “36,000 returns.m-2” is hard to interpret.

         Thanks, modified in line 31.

6. Line 29–34, Page 1: Statements such as “reduce the accuracy” and “PLS-SLAM has potential…” are vague without data to support them. Consider adding quantitative conclusions here.

         Thanks, modified in lines 34-44.

7. Line 39, Page 1: “75.8% are a fast-growing species” → should be corrected to “75.8% consist of fast-growing species.” or “are fast-growing species.”

         Thanks, modified in line 55.

8. Line 84–86, Page 2: The structure “such as: Does this new paradigm...” is both grammatically incorrect and stylistically too informal for a scientific article. Replace this with a list of concerns expressed as noun phrases (e.g., “concerns about accuracy…”).

         Thanks, modified in line 115.

9. Line 95, Page 3 and Line 162, Page 4: DBH and H were already defined earlier — no need to redefine. Just use the abbreviations. Please remove the repeated definitions.

         Modified in line 119 and 146.

10. Line 109–110, Page 3: The mention of Köppen-Geiger classification should be supported with a reference.

         Reference inserted, line 139.

11. Line 117, Page 3: “1.30 m high” → suggest rephrasing to “1.30 m above ground level (AGL)” for clarity.

         Modified in line 147-148.

12. Line 119, Page 3: “the average if” → should be “the average of”.

         Modified in line 150.

13. Equation (1), Page 4: tan should be in upright Roman font. Use a multiplication dot (·) between tan and d. Also, the text defines d as the distance to the tree, but Fig. 2 uses the variable I. Please standardize the notation across text and figures.

         Figure and equation changed, thanks.

14. Line 151, Page 4: “.. (x/y/z).” Suggest removing.

         Removed. Line 185.

15. Line 187, Page 4: “sloop smooth” → should be “slope smooth.”

         Changed, line 221.

16. Line 188, Page 4: “weighing” → should be “inverse distance weighting.”

         Changed, line 223.

17. Line 245–246, Page 6: You state that only one individual was not identified at 500 and 100 pts·m². However, from Fig. 7c, it appears two trees were missing. Please double-check and clarify.

         Thank you for checking.

         There was an error in the figure: n = 69.

         As shown in the figure below, only one tree was not detected.

         The figure has been updated accordingly.

18. Line 263, Page 6: “Accuracy was slightly higher for larger trees” contradicts Table 1 on Page 9. RMSE for DBH > 27.3 cm is higher, not lower. This interpretation needs correction.

         As shown in the figure below, the accuracy of the DBH estimates for the largest trees was greater, i.e., lower RMSE values (Please, visualize the figure in digital archive).

19. Line 350–356, Page 8: The word “All” should be lowercase.

         Modified, thanks.

20. Line 376, Page 9: “Ransac” → should be capitalized as “RANSAC”.

         Modified, thanks.

21. Line 398, Page 9: You mention “efficiency” without conducting a time or labor analysis. This statement should be either removed or supported by comparative experiments.

         Removed, line 442, thanks.

22. The manuscript evaluates the influence of different point cloud densities (e.g., 36,000, 500, 100 points·m⁻²) on tree attribute estimation. However, the method used to reduce point cloud density is not clearly described. Was the downsampling performed via random sampling, voxel grid filtering, or another spatial method? Please add it.

         Inserted, line 190.

23. The number of trees used in the evaluation (n = 71) appears to be quite limited given the stratification across point densities and size classes (e.g., small vs. large DBH). With such a small sample, it is challenging to draw statistically robust conclusions, especially regarding trends like “accuracy decreases more for smaller trees” or “height estimation is more affected in taller trees.” Overall, the manuscript contains many informal phrases and sentences. A thorough revision by a native English speaker or professional editing service is highly recommended to improve clarity and readability.

         Defining sample size is always a difficult topic. From a scientific perspective, the larger the sample, the more reliable the conclusions should be. On the other hand, the main reason for working with sampling is that measuring the entire population is expensive, onerous, and often unfeasible. Because of this, choosing the ideal sample size will always be a dilemma, since although the largest possible sample is desired, this desire is limited by available resources. Science, especially statistics, has adapted to these limitations. The t-statistic, for example, was designed precisely to draw conclusions in cases of small samples. According to this statistic, after 30 degrees of freedom, t-values ​​begin to vary very little. Based on this, it could be said that, in the case of the t-statistic, 30 sample units would be considered a fairly representative size for drawing conclusions. Comparing the t and normal distributions, as n increases above 30, the t distribution approaches the normal distribution more closely, and in this case, the z-test could be used instead of the t-test without major drawbacks. Given this context, the reviewer expresses concern about the stratifications, which could significantly reduce the value of n. In fact, only one stratification was performed; the 71 trees were divided into two height groups: 35 trees less than or equal to 31.4 meters and 36 trees greater than 31.4 meters. In the case of point cloud density, no stratification was performed; that is, at least 35 trees were always evaluated. Thus, although it is not possible to say with 100% certainty what the ideal sample size is, based on the presented context, the sample size used in this work appears acceptable in light of the best statistical references. However, it's important to emphasize that regardless of the sample size, interpretations will always be probabilistic; that is, nothing can be stated with 100% certainty. Therefore, statistics should be viewed solely as a tool that helps produce evidence that, combined with other evidence, will help form a value judgment. In this sense, the authors believe that the statistical evidence gathered, along with references from other scientific studies conducted under similar conditions and the theoretical foundation on the topic, allowed them to draw the conclusions they reached. Of course, no scientific study is definitive and absolute, and therefore, the conclusions reached in this study based on the evidence gathered may be confirmed or refuted in the future based on new evidence, thus fostering scientific advancement in a natural way.

         We appreciate the extensive review work that enabled the improvement of the manuscript.

Respectfully,

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

1.Please avoid citations like line 444. If the results of these studies are necessary, list them separately. Please check this problem in the whole manuscript；

2. Please check the expression in the upper right corner table throughout the manuscript, as the expression on line 441 is wrong；

3. Please increase the resolution of the images, all images in the manuscript are currently blurry.

Author Response

1.Please avoid citations like line 444. If the results of these studies are necessary, list them separately. Please check this problem in the whole manuscript

Response: We carefully reviewed all references cited in the manuscript and corrected those that did not conform to the journal’s formatting guidelines. The reference list has been updated accordingly in the revised version.

 

2. Please check the expression in the upper right corner table throughout the manuscript, as the expression on line 441 is wrong:

 

Response: Thank you for your observation. However, we did not identify a table or expression in the upper right corner on line 441 of the manuscript. At this specific location, no table is present. We kindly ask for clarification in case we have misunderstood the comment. Nevertheless, we have thoroughly reviewed all tables and expressions in the manuscript to ensure their correctness.

 

3. Please increase the resolution of the images, all images in the manuscript are currently blurry.

Response: We have addressed this issue by improving the resolution of all images in the manuscript. Due to file size limitations, the high-resolution versions of the figures have been provided in a separate file, as recommended in the journal’s submission guidelines