Trends and Applications of Principal Component Analysis in Forestry Research: A Literature and Bibliometric Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Principal Component Analysis (PCA) is a commonly used technique for data dimensionality reduction. It has the following obvious advantages: it can reduce the data dimension; eliminate noise and redundancy; discover the hidden structure; reduce the computational complexity: PCA decomposes the eigenvalues of the covariance matrix to transform large-scale data calculations into those of a small number of eigenvectors, thereby improving the computational efficiency. Therefore, PCA has been widely applied in data processing and statistical analysis in scientific research, particularly in forestry research. This study provided the first comprehensive bibliometric and literature review focused exclusively on PCA applications in forestry by using the Web of Science database and visualized through VOSviewer software. It is very necessary and important for reducing data dimensionality and classifying parameters within complex ecological datasets. However, PCA has limitations when dealing with nonlinear data, maintaining data integrity and interpreting results. Therefore, while making use of its advantages, we should try to avoid the occurrence of its disadvantages.

 

The subject completely falls within the general scope of this Journal. The topic is very interesting. Generally, the review is well-structurally organized. This is comparatively a fully new and original contribution in this field.   All stated above, this literature review is complete, accurate and comprehensive.The English level of the authors is very high.

Existing problems as follows:

 

1. Pay attention to distinguishing between the research and researches;
2. P16-17, “A total of 96 articles published between 1993 and 2024 were analyzed using the Web of Science database and visualized through VOSviewer software”; P118-119 “This initial query retrieved 361 articles from Web of Science and 683 from Scopus”; P129-130 “The second component of the study involved a traditional literature review to provide an in-depth synthesis of the 382 selected articles”; authors should clearly tell the readers how to get a total of 96 articles published and analyzed between 1993 and 2024;
3. P143-144 “Figure 2. Distribution of the main publication types related to forest ecosystem services in riparian areas” seems to be unnecessary for this scientific study because it can only provide limited formation, but takes up a considerable amount of space; Pie charts are rarely used in scientific researches;
4. P180-181, should put “Table 2. The most used keywords in articles published about the principal component analysis (PCA) in forestry” in the same pages;
5. P531 after “5. Conclusions” , “1., 2., 3.” Should be replaced by “(1), (2), (3)”.

Comments on the Quality of English Language

The English level of the authors is very high

Author Response

Comments 1

Pay attention to distinguishing between the research and researches

Response 1

Thank you for your observation. We carefully reviewed the entire manuscript and corrected the usage of “research” and “researches” where appropriate, in accordance with standard academic conventions.

Comments 2

P16-17, “A total of 96 articles published between 1993 and 2024 were analyzed using the Web of Science database and visualized through VOSviewer software”; P118-119 “This initial query retrieved 361 articles from Web of Science and 683 from Scopus”; P129-130 “The second component of the study involved a traditional literature review to provide an in-depth synthesis of the 382 selected articles”; authors should clearly tell the readers how to get a total of 96 articles published and analyzed between 1993 and 2024;

Response 2

Thank you for pointing out the need for clarification. We have addressed this issue by adding Figure 1, which illustrates the article selection process in detail. Additionally, we revised the text to explicitly describe how the final set of 96 articles was derived. The updated explanation now reads:

This initial query retrieved 361 articles from Web of Science and 683 from Scopus. After removing duplicates (articles appearing in both databases), 803 unique articles remained. We then excluded records that could not be retrieved, those without abstracts, and articles deemed irrelevant to the topic. This refinement resulted in a total of 382 articles, which formed the basis of the analyses presented in Chapter 1. From a close evaluation of these 382 articles, only 96 were found to contain elements relevant for citation in our bibliometric and conceptual review.

We believe this clarification, along with the accompanying figure, provides a transparent and reproducible overview of the selection and filtering process.

Comments 3

P143-144 “Figure 2. Distribution of the main publication types related to forest ecosystem services in riparian areas” seems to be unnecessary for this scientific study because it can only provide limited formation, but takes up a considerable amount of space; Pie charts are rarely used in scientific researches;

Response 3

We agree with the reviewer’s comment. Figure 2 has been deleted from the revised manuscript.

Comments 4

P180-181, should put “Table 2. The most used keywords in articles published about the principal component analysis (PCA) in forestry” in the same pages;

Response 4

We have adjusted the layout of the manuscript to ensure that Table 2 and its corresponding title and comments appear on the same page.

Comments 5

P531 after “5. Conclusions” , “1., 2., 3.” Should be replaced by “(1), (2), (3)”.

Response 5

Thank you for your observation. We have modified the numbering format in the conclusion section as suggested, replacing “1., 2., 3.” with “(1), (2), (3)”.

Supplementary comment

However, PCA has limitations when dealing with nonlinear data, maintaining data integrity and interpreting results. Therefore, while making use of its advantages, we should try to avoid the occurrence of its disadvantages.

Response

Even though this is not an observation requiring modifications, we considered these explanations particularly valuable, which is why we included them in the newly added chapter 6. Future directions and research gaps and in Table 3. Benefits and limitations of PCA in forestry research.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript shows that Principal Component Analysis (PCA) has become a key tool in forestry for reducing data dimensionality and enhancing ecological analysis, with increasing integration into advanced technologies and machine learning. This comprehensive review of 96 articles (1993–2024) highlights PCA’s expanding applications in forest management, species classification, and environmental monitoring, positioning it as a foundational method for future research and innovation in the field. The manuscript is an interesting mix of review and data compilation. However, as a review article, it would benefit from providing a clearer overview of current knowledge, future perspectives, and areas that require further development. While most references are current, many are not from the last five years—a notable issue for a state-of-the-art review. I found the text to lack clear direction and structure, making it difficult to follow. All figure legends should be updated to provide detailed and adequate descriptions of all elements, colours, and figures included. As presented, the manuscript does not give the reader a clear understanding of the subject matter. In my opinion, the review still requires substantial improvement, particularly regarding current approaches and how they might be enhanced. Please include a discussion of how sample size influences results, which statistical tools are most commonly used, and whether or why these should continue to be employed. A more critical review of the topic is needed, especially if that is the authors’ intention. Additionally, please ensure the text is written in clear and concise paragraphs, as excessively long sentences and paragraphs detract from readability and make the manuscript less engaging.

Comments on the Quality of English Language

CHeck grammar, spelling and verbosity.

Author Response

Comments 1

However, as a review article, it would benefit from providing a clearer overview of current knowledge, future perspectives, and areas that require further development.

Response 1
Thank you for your valuable suggestion. In response, we have substantially expanded the manuscript by adding a new Section 6 – “Future Directions and Research Gaps”. This section outlines key areas for further exploration, including the integration of PCA with non-linear dimensionality reduction techniques (e.g., t-SNE, UMAP), its use in real-time forest monitoring, its potential for analyzing socio-ecological systems, and the development of standardized protocols. These additions provide a clearer perspective on current knowledge and help map out meaningful directions for future forestry research using PCA.

Comments 2

While most references are current, many are not from the last five years—a notable issue for a state-of-the-art review.

Response 2

We have re-analyzed the specialized literature and added 2 additional articles published in 2023, 7 articles in 2024, and 1 article in 2025 to Table 3. Some of the areas in Forestry where PCA is used (extract from literature) and to the reference list. These additions strengthen the recency and relevance of the review.

Comments 3

I found the text to lack clear direction and structure, making it difficult to follow. Additionally, please ensure the text is written in clear and concise paragraphs, as excessively long sentences and paragraphs detract from readability and make the manuscript less engaging.

Response 3

We appreciate the reviewer’s helpful observation. In response, we have revised the manuscript to improve clarity, structure, and paragraph coherence. Specifically:

We restructured Section 3.2.3 and Section 4.2, breaking long paragraphs into shorter, more focused units with clearer thematic transitions.

We simplified complex or lengthy sentences to enhance readability and reader engagement.

We also ensured a more consistent logical flow across subsections, including clearer topic sentences and closing summaries.

We believe these changes significantly improve the manuscript’s readability and thank the reviewer for highlighting this issue.

Comments 4

All figure legends should be updated to provide detailed and adequate descriptions of all elements, colours, and figures included.

Response 4

Thank you for your observations. The figures were updated.

Comments 5

As presented, the manuscript does not give the reader a clear understanding of the subject matter.

Response 5

We appreciate your feedback. To improve clarity and aid comprehension, we have inserted a new summary visual (Figure 4) and accompanying table within Section 3.2.1, which synthesize the diverse applications of PCA in forestry. This includes a domain-level overview and comparative table highlighting the benefits and limitations of PCA across major research areas such as biomass estimation, disease monitoring, soil quality analysis, and remote sensing. These additions are intended to provide the reader with a clear and immediate understanding of PCA’s role and significance in forestry.

Comments 6

In my opinion, the review still requires substantial improvement, particularly regarding current approaches and how they might be enhanced.

Response 6
Thank you for this constructive critique. We have addressed this point by extending Section 4.2 with a new paragraph that specifically discusses how current PCA-based approaches can be enhanced. We emphasize the value of domain-specific preprocessing, the benefits of hybrid PCA–machine learning models, and the importance of explainable AI in forestry analytics. These refinements aim to improve the reader’s understanding of how PCA methodologies can evolve to meet the demands of increasingly complex forestry datasets and real-world applications.

Comments 7

Please include a discussion of how sample size influences results, which statistical tools are most commonly used, and whether or why these should continue to be employed.

Response 7

Thank you for this insightful comment. We have expanded the Discussion section to include a dedicated paragraph addressing the influence of sample size on statistical outcomes. Specifically, we highlight that larger sample sizes improve the reliability and precision of multivariate analyses such as PCA, ANOVA, and Pearson/Spearman correlations. For PCA in particular, larger datasets allow for more stable and accurate identification of principal components, reducing the risk of overfitting or misrepresenting ecological variance. We also note that PCA remains one of the most widely employed statistical tools in forestry research due to its strengths in dimensionality reduction, pattern recognition, and data interpretation—especially when dealing with large, complex datasets. These strengths justify its continued use, particularly when combined with complementary techniques such as regression models or machine learning algorithms.

Comments 8

A more critical review of the topic is needed, especially if that is the authors’ intention.

Response 8

Thank you for the valuable suggestion. We agree that a more critical perspective was necessary. Accordingly, we have made the following changes:

Section 3.2.3 has been revised to include a deeper critical discussion of PCA’s limitations, such as its linear nature and sensitivity to data preprocessing. We now compare PCA to alternative methods like kernel PCA, t-SNE, and UMAP, and we highlight scenarios where PCA may not be suitable.

Section 4.2 has been rewritten to more critically evaluate recent PCA innovations. We now discuss the assumptions and trade-offs of PCA-based models, and pose open research questions related to forest heterogeneity, nonlinear ecological relationships, and model scalability.

Throughout these revisions, we have also emphasized challenges and potential pitfalls in PCA applications, to provide a more balanced and analytical perspective. We hope these improvements strengthen the manuscript’s critical depth and better reflect the complexity of the subject matter.

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for addressing my comments. The manuscript has improved considerably and is likely to be accepted.

Author Response

Dear Editor,

We have made efforts to address the issue regarding the quality of the figures as appropriately as possible:

• Figure 5 has been removed, as it was not particularly representative, and the information it conveyed is also discussed in the text.
• Figure 6 has been improved by using more vivid colors (the figure is generated using Excel, and the maximum pixel resolution cannot be adjusted further).
• Figures 7 and 8, which are generated with the VOSviewer software, have been enhanced to the extent possible by maximizing the size and clarity of the details, although we were not able to modify the pixel resolution.
• Figures 3 and 4 have been entirely redone to improve their quality.

We hope these revisions meet the requirements for final acceptance.