Air Quality—Meteorology Correlation Modeling Using Random Forest and Neural Network
Round 1
Reviewer 1 Report
Review results for Manuscript Number: sustainability-2095835
Dear Editor
I read carefully the manuscript titled " Forest Ecosystem Management and Meteorological Correlation Analysis Using Genetic Algorithm Driven by the Artificial Intelligence". The subject of the manuscript is very interesting and relevant. The structure is adequate.
Specific comments to the author(s):
· Quantitative results should be provided in the abstract to make it more comprehensive. The results of the study should be added in the abstract section.
· I advise the authors to add a paragraph concerning the application of the RF approach in various studies. e.g, https://doi.org/10.1007/s12517-022-09531-3; https://doi.org/10.1007/s12145-021-00653-y
· More literature review about the other methods is needed, hence manuscript could be substantially improved.
Re-write the mathematical equation with better resolution.
· Perform a last check in the English language. Some phrases would be improved.
· The discussion section in the present form is relatively weak and should be strengthened with more details and justifications.
· It seems that conclusions are observations only.
Figures resolution needs to be enhanced.
Overall, this work could be accepted for publication after addressing the above issues.
Therefore, I recommend a moderate revision
Author Response
I read carefully the manuscript titled " Forest Ecosystem Management and Meteorological Correlation Analysis Using Genetic Algorithm Driven by the Artificial Intelligence". The subject of the manuscript is very interesting and relevant. The structure is adequate.
Specific comments to the author(s):
- Quantitative results should be provided in the abstract to make it more comprehensive. The results of the study should be added in the abstract section.
Response: Thank you for your comments. The prediction effect of BP-GA model on air quality forecast by meteorological correlation model has been explained in detail in the abstract.
- I advise the authors to add a paragraph concerning the application of the RF approach in various studies. e.g, https://doi.org/10.1007/s12517-022-09531-3; https://doi.org/10.1007/s12145-021-00653-y
Response: Thank you for your comments. The research and application of random forest algorithm in image recognition and classification, target detection and other directions have been added in the third paragraph of the introduction.
- More literature review about the other methods is needed, hence manuscript could be substantially improved.
Response: Thank you for your comments. The research of different machine learning algorithms on air quality early warning models has been added in the fifth paragraph of the introduction.
Re-write the mathematical equation with better resolution.
Response: Thank you for your comments. The resolution of the equations has been enhanced by adjusting the format of equation paragraphs and font format to make the equations meet the resolution requirements.
- Perform a last check in the English language. Some phrases would be improved.
Response: Thank you for your comments. The professionals have checked and polished the sentences and words of the article.
- The discussion section in the present form is relatively weak and should be strengthened with more details and justifications.
Response: Thank you for your comments. The prediction of air quality by fusion model and single model has been added to demonstrate the prediction effect of RF+BP+GA model in this paper. See the discussion section for details.
- It seems that conclusions are observations only.
Response: Thank you for your comments. Firstly, the numerical results of predicting AQI by RF+BP+GA model have been explained in the conclusion part. For details, see the first paragraph of the conclusion part. Secondly, the conclusion part of this paper describes the research process and the research results of this paper, and finally summarizes the research conclusions, reflecting and summarizing the research process and proposing the research development.
Figures resolution needs to be enhanced.
Response: Thank you for your comments. The frame diagram and data charts have been adjusted to improve the image resolution. See Fig.3, Fig.1, Fig.8 ~ Fig.10 for details.
Overall, this work could be accepted for publication after addressing the above issues.
Therefore, I recommend a moderate revision
Response: Thank you for your comments. The article has been modified according to the above opinions.
Reviewer 2 Report
This work focuses on the management strategy of the environmental ecosystem under the Artificial Intelligence (AI) algorithm and explores the correlation between air quality and meteorology. Xi’an city is selected as an example. A fusion model of RF+BP+GA is proposed to predict the air quality and meteorological correlation. The proposed air quality and meteorological correlation model is applied to forest ecosystem management. At the same time, based on the proposed model, six valuable forest ecological management measures are proposed. These suggestions and research findings are expected to provide reference directions for the follow-up forest ecological management, sustainable development, and studying the interaction factors between climate and forest ecosystem.
My advice is to Review Again after Resubmission (Paper is not acceptable in its current form, but has merit. A major rewrite is required. Author should be encouraged to resubmit a rewritten version after the changes suggested.)
Comments:
1) Lack of significant contributions: the technical contributions are incremental, the proposed methodology has limited novelty, and the overall impact is not significant enough. If a new AI method/theory (i.e A fusion model of RF+BP+GA) is being reported in the paper, it should be compared and validated against the other state-of-the-art methods.
2) Weak experiments: experiments are not convincing enough; baselines are not strong (e.g., did not compare with state-of-the-art approaches)
3) The author should revise their paper to answer the four questions below:
First question: What is the main challenge and issues in this study?
Second question: 'What is the criticism and gap analysis for academic literature that attempt to provide a solution?'
Third question: 'What are the recommended solution for such challenges and their issues?'
Fourth question: 'What are the implication, contributions, and novelty of the present study?'
Author Response
This work focuses on the management strategy of the environmental ecosystem under the Artificial Intelligence (AI) algorithm and explores the correlation between air quality and meteorology. Xi’an city is selected as an example. A fusion model of RF+BP+GA is proposed to predict the air quality and meteorological correlation. The proposed air quality and meteorological correlation model is applied to forest ecosystem management. At the same time, based on the proposed model, six valuable forest ecological management measures are proposed. These suggestions and research findings are expected to provide reference directions for the follow-up forest ecological management, sustainable development, and studying the interaction factors between climate and forest ecosystem.
My advice is to Review Again after Resubmission (Paper is not acceptable in its current form, but has merit. A major rewrite is required. Author should be encouraged to resubmit a rewritten version after the changes suggested.)
Comments:
1) Lack of significant contributions: the technical contributions are incremental, the proposed methodology has limited novelty, and the overall impact is not significant enough. If a new AI method/theory (i.e A fusion model of RF+BP+GA) is being reported in the paper, it should be compared and validated against the other state-of-the-art methods.
Response: Thank you for your comments. In this paper, contributions to this research are explained from the aspects of research methods and content. For details, see the seventh paragraph of the introduction. In addition, the prediction effect of RF+BP+GA model in this paper is demonstrated by combining the fusion model and a single model. See the discussion section for details.
2) Weak experiments: experiments are not convincing enough; baselines are not strong (e.g., did not compare with state-of-the-art approaches)
Response: Thank you for your comments. In this paper, the prediction results of single model (random forest model) and dual model (BP-GA model) are compared and analyzed in the experimental part, and then the prediction effects of dual model (BP-GA model) and RF+BP+GA model are compared. In addition, the prediction of air quality by fusion model and single model is added to demonstrate the prediction effect of RF+BP+GA model in this paper. See the discussion section for details.
3) The author should revise their paper to answer the four questions below:
First question: What is the main challenge and issues in this study?
Response: Thank you for your comments. The main challenges in the research process of this paper are as follows: the acquisition of meteorological air quality monitoring data of Xi'an city and optimization of BPNN algorithm.
Second question: 'What is the criticism and gap analysis for academic literature that attempt to provide a solution?'
Response: Thank you for your comments. In this paper, it is found that the prediction accuracy of nonlinear air quality data is not high and the generalization ability of the traditional single algorithm is weak by consulting the previous network models for air quality prediction. For high-dimensional data, it is easy to lead to modeling failure, so the air quality meteorological CM cannot be fully analyzed, resulting in low model performance.
Third question: 'What are the recommended solution for such challenges and their issues?'
Response: Thank you for your comments. In view of the problems and defects in previous studies, this paper combines ML algorithm to establish air quality meteorological CM. ML algorithm can process high-dimensional data without feature screening, with high prediction accuracy, strong generalization ability, and strong anti-interference ability. When there are missing values in the data, the prediction accuracy can still be maintained.
Fourth question: 'What are the implication, contributions, and novelty of the present study?'
Response: Thank you for your comments. The significance of this study lies in that, different from the previous air quality prediction models of fusion models, this paper combines RF algorithm, BPNN model, and genetic algorithm to optimize the network weight and threshold prediction in BPNN, reduce the dimension of input factors, reduce the interference of unrelated dimensions, and realize the optimization of urban air quality-meteorological CM. On the one hand, the prediction accuracy is improved, which provides a research reference for the prediction of air quality index (AQI) of air quality meteorological CM. On the other hand, the fusion model prediction is made by combining meteorological and air quality factors, which provides data support for the analysis of air quality problems.
Research contributions and innovations are as follows: the concept of RF+BP+GA fusion model is proposed, and a feature selection method suitable for meteorological CM assessment of air quality is established. The prediction model combining RF, neural network, and GA is used to predict air quality, which has high prediction accuracy. It can not only predict AQI, but also provide implementation strategy basis for forest ecosystem management.
Reviewer 3 Report
The paper explores a possibility of using a fusion model of the artificial intelligence algorithms to predict the air quality. Six forest ecological management measures are proposed. However, a relationship between the proposed model and the proposed management measures is not clear. The paper can be suitable for Sustainability journal after clarifying the relationship between the model and the proposed management measures.
The novelty of the research should be highlighted. Is a fusion model of RF+BP+GA developed by the authors? Or was this model already used in other fields of research by other authors?
The abstract does not reflect the authors' findings. The work does not explain nor discuss RF, BPNN, and GA (L.17-19). The authors provide only a description of the algorithms. There is no description in the main text of how "the proposed air quality and meteorological correlation model is applied to forest ecosystem management" (L.21-22). Six proposed ecological management measures are not "based on the proposed model" (L.25-26).
Introduction, L.160-162. How exactly the fusion model explains the role of forests in climate regulation? There is no description in the manuscript.
In the Methods section the authors describe only the basic principles of RF, BP, and GA. The authors must provide a description of how these algorithms were applied to the experimental data to obtain the results. What was the initial dataset? What were air and meteorological quality features extracted from the initial dataset? What data were used as a training subset? What data were used to predict the AQI value? What software was used? Etc.
The descriptions of RF, BP, GA, and AQI are difficult to understand while there are many clear descriptions in the literature and on websites. The descriptions of RF, BP, GA, and AQI should be improved.
How the information presented in the section 2.4 was used in the study?
The authors should provide details regarding where the experimental data can be found (link or reference).
Section 3 is Results and discussion, and section 4 is Discussion. I would suggest to keep the discussion in one place.
The authors must explain why a time interval of 7 days is sufficient to make statistical inferences. The sample size seems to be extremely small for exploring the performance of the AI algorithms. The authors must explain why the sample size is sufficient to draw any conclusions on the model performance.
The conclusions (section 5) are not fully supported by the results. The interaction between the air guality - climate CM and the forest ecosystem was not explored in the manuscript.
According to the Instructions for Authors, abbreviations should be defined the first time they appear in the abstract; the main text; the first figure or table. The authors should check abbreviations carefully.
No need to provide abbreviations that are not used (QTM, PCC, SCC etc.).
The statements (L.65-84, 125-128, 330-361) should be referenced. Reference 22 is missing in the text.
Author Response
The paper explores a possibility of using a fusion model of the artificial intelligence algorithms to predict the air quality. Six forest ecological management measures are proposed. However, a relationship between the proposed model and the proposed management measures is not clear. The paper can be suitable for Sustainability journal after clarifying the relationship between the model and the proposed management measures.
Response: Thank you for your comments. In this paper, the correlation model of air quality and meteorology is applied to forest ecosystem management to propose management measures. The relationship between the model and proposed management measures is described in the first paragraph of Section 3.3.
The novelty of the research should be highlighted. Is a fusion model of RF+BP+GA developed by the authors? Or was this model already used in other fields of research by other authors?
Response: Thank you for your comments. The research innovation of this paper has been discussed in detail in the seventh paragraph of the introduction. RF+BP+GA fusion model was developed experimentally by the author, and RF+BP+GA fusion model has not been applied to other fields.
The abstract does not reflect the authors' findings. The work does not explain nor discuss RF, BPNN, and GA (L.17-19). The authors provide only a description of the algorithms. There is no description in the main text of how "the proposed air quality and meteorological correlation model is applied to forest ecosystem management" (L.21-22). Six proposed ecological management measures are not "based on the proposed model" (L.25-26).
Response: Thank you for your comments. The prediction effect of BP-GA model in predicting air quality has been described in detail in the abstract. The application of the air quality and meteorological correlation model to forest ecosystem management is described in the first paragraph of Section 3.3, and the management measures are proposed for details.
Introduction, L.160-162. How exactly the fusion model explains the role of forests in climate regulation? There is no description in the manuscript.
Response: Thank you for your comments. In this paper, the correlation model of air quality and meteorology is applied to forest ecosystem management, so as to propose the management measures. The relationship between the model and the proposed management measures is explained in the content and abstract of Section 3.3.
In the Methods section the authors describe only the basic principles of RF, BP, and GA. The authors must provide a description of how these algorithms were applied to the experimental data to obtain the results.
Response: Thank you for your comments. In the seventh paragraph of the introduction, this paper describes the roles of random forest algorithm, BP neural network model, and genetic algorithm in the fusion model.
What was the initial dataset?
Response: Thank you for your comments. The experimental data set of this paper is the real-time air quality test data of Xi'an City from June 24, 2022 to June 30, 2022.
What were air and meteorological quality features extracted from the initial dataset?
Response: Thank you for your comments. AQI data from June 24, 2022 to June 30, 2022 were extracted from the initial data set.
What data were used as a training subset?
Response: Thank you for your comments. AQI data values from June 24, 2022 to June 30, 2022 were used as a training subset.
What data were used to predict the AQI value?
Response: Thank you for your comments. This article uses data from June 24, 2022 to June 18, 2022 to predict the air quality index. See paragraph 1 of Section 2.5 for details.
What software was used? Etc.
Response: Thank you for your comments. In this paper, Spark framework of Hadoop big data platform is used to set three distributed frameworks for experimental operations. See Section 2.6 for details.
The descriptions of RF, BP, GA, and AQI are difficult to understand while there are many clear descriptions in the literature and on websites. The descriptions of RF, BP, GA, and AQI should be improved.
Response: Thank you for your comments. The description of RF, BP, GA and AQI has been improved based on literature. For details, see the first paragraph of Section 2.2, the first and third paragraphs of Section 2.3, and the second paragraph of Section 2.1.
How the information presented in the section 2.4 was used in the study?
Response: Thank you for your comments. Section 2.4 of this paper describes the role of forests in climate regulation, and provides a theoretical basis for the application of the correlation model between air quality and meteorology to forest ecosystem management.
The authors should provide details regarding where the experimental data can be found (link or reference).
Response: Thank you for your comments. Experimental data link: Xi 'an Air Quality Index query (AQI) in June 2022 _Xi 'an PM2.5 historical data query in June _ Weather Post Report (tianqihoubao.com)
Section 3 is Results and discussion, and section 4 is Discussion. I would suggest to keep the discussion in one place.
Response: Thank you for your comments. The discussion part has been shown in section "3 Results and discussion", see Section 3.3 for details.
The authors must explain why a time interval of 7 days is sufficient to make statistical inferences. The sample size seems to be extremely small for exploring the performance of the AI algorithms. The authors must explain why the sample size is sufficient to draw any conclusions on the model performance.
Response: Thank you for your comments. This paper uses 7 days of data to test the validity of the model and algorithm. With limited data, better models can be trained to apply air quality and meteorological correlation models to forest ecosystem management.
The conclusions (section 5) are not fully supported by the results. The interaction between the air guality - climate CM and the forest ecosystem was not explored in the manuscript.
Response: Thank you for your comments. In the conclusion part, six valuable forest ecological management measures are proposed after applying the correlation model of air quality and meteorology to forest ecosystem management. See the conclusion for more details.
According to the Instructions for Authors, abbreviations should be defined the first time they appear in the abstract; the main text; the first figure or table. The authors should check abbreviations carefully.
Response: Thank you for your comments. The abbreviations have been defined for the first time in this article. After checking, there is no problem with the abbreviations.
No need to provide abbreviations that are not used (QTM, PCC, SCC etc.).
Response: Thank you for your comments. In this paper, the abbreviations of algorithms and research methods are expressed in full.
The statements (L.65-84, 125-128, 330-361) should be referenced. Reference 22 is missing in the text.
Response: Thank you for your comments. Reference 22 has been searched and supplemented.
Reviewer 4 Report
Section 1 must be improved.
- Authors need to review the format of the paper, there are several formatting errors in text, alignment, formulas such as images, etc.
- Authors should emphasize contribution and novelty, the introduction needs to clarify the motivation, challenges, contribution, objectives, and significance/implication.
- You must properly introduce your work, specify well what were the goals you set yourself and how you approached the problem.
Section 2 must be improved.
- In this section you introduce Data and Methods, maybe you can rename it as Materials and Methods.
- Perhaps it would be appropriate first to introduce the methods of describing data, since the technologies you intend to develop are based on data.
- review the format of the equations, as images do not fit, use the word tool to insert the images
- You must properly introduce the equation, list in detail the variables contained in it with a concise description of the meaning. To make them more readable show them in a bulleted list. In this way the reader will be able to understand the contribution of each variable.
- Add more references to works that have already dealt with the topic (Random Forest applications), for example:” Wind turbine noise prediction using random forest regression”
- Try to enrich the captions of the figures (for example Figure 1), the reader should be able to read the figure without the need to retrieve the information in the paper. Try to summarize the essential parts of the Figure and what you want to explain with it.
- Introduce adequately the GA algorithms
- Move Section 2.4 at the beginning of the section
- Describe with major accuracy the experimental data
- Describe in detail the equipment used to make the measurements (air quality and meteorological parameters). Extract this data from the datasheet of the instrumentation manufacturer. To make reading the specifications of the instruments more immediate, you can insert them in a table, listing the instruments used and the specific characteristics for each.
- add a map that identifies the area affected by the monitoring, and indicates the position of the monitoring stations
- You should better describe the datasets you used, there is no reference to the resource altogether. Where did you download this data? You should specify this so that the reader can reproduce the experiment.
- The section relating to the methodologies based on Machine Learning must be enriched. You must summarize the essential characteristics of the methods you have used and justify your choices. Try to summarize what are the strengths and weaknesses of the methods, in this way you can make the reader understand why you have chosen these methodologies.
- I could not find a detailed description of the evaluation metrics you have adopted. How will you measure your model's performance? This section is essential in order to demonstrate the effectiveness of your methodology. Furthermore, only by adopting adequate metrics will it be possible to compare your results with those obtained by other researchers.
-
Section 3 must be improved.
- In this section you present the results of your work. You should start by summarizing what you have done in your work.
- Next you should summarize the data you collected and the conditions under which you collected it.
- How much data did you collect? How many features and how many records
- In the case of Machine Learning applications, data quality is crucial for the success of the analyses.
- Figure 8 must be improved: The text is too small and appears blurry, making it difficult for the reader to follow the flow of information. I will not repeat this advice again, it also applies to the other occurrences.
-
Section 4 must be improved.
- this section is very short it is better to add to the end of the previous section and rename Results and Discussion
69) “2021 Climate Blue Book” Add references to allow the reader to learn more about the topic
85-86) Add references to support these statements.
90) “Lolli et al. (2020)” Use a coherent format for citation, replace with Lolli et al. [12]. I will not repeat this advice again, it also applies to the other occurrences.
239-254) Check the text format, it seems that the line spacing is different from the rest of the paper
258-262) Introduce adequately the topic (GA)
413) rearrange the subplots in Figure 8, so they don't fit, either you order them in three rows and one column or two rows and two columns
431) Make table fit on the same page.
483-508) Check the text format, the alignment of the lines is not consistent with the rest of the paper
Author Response
Section 1 must be improved.
- Authors need to review the format of the paper, there are several formatting errors in text, alignment, formulas such as images, etc.
Response: Thank you for your comments. The paper format has been revised.
- Authors should emphasize contribution and novelty, the introduction needs to clarify the motivation, challenges, contribution, objectives, and significance/implication.
Response: Thank you for your comments. The research contribution and innovation of this paper have been explained in the seventh paragraph of the introduction. See the beginning of the abstract for details on the motivation of this study. As for the research challenges in this paper, please refer to the second paragraph of the conclusion.
- You must properly introduce your work, specify well what were the goals you set yourself and how you approached the problem.
Response: Thank you for your comments. The research background of this paper has been introduced in detail, and the research status of scholars is discussed. The research process of this paper is introduced in detail in the introduction section.
Section 2 must be improved.
- In this section you introduce Data and Methods, maybe you can rename it as Materials and Methods.
Response: Thank you for your comments. The title of Section 2 has been amended to read "2. Materials and Methods".
- Perhaps it would be appropriate first to introduce the methods of describing data, since the technologies you intend to develop are based on data.
Response: Thank you for your comments. The research data of this paper has been shown in Section 2.1.
- review the format of the equations, as images do not fit, use the word tool to insert the images
Response: Thank you for your comments. Word has been used to insert the equations and images for this article.
- You must properly introduce the equation, list in detail the variables contained in it with a concise description of the meaning. To make them more readable show them in a bulleted list. In this way the reader will be able to understand the contribution of each variable.
Response: Thank you for your comments. The characters of all equations have been explained one by one.
- Add more references to works that have already dealt with the topic (Random Forest applications), for example:” Wind turbine noise prediction using random forest regression”
Response: Thank you for your comments. The research and application of random forest in image recognition and classification, target detection and other directions have been added in the third paragraph of the introduction.
- Try to enrich the captions of the figures (for example Figure 1), the reader should be able to read the figure without the need to retrieve the information in the paper. Try to summarize the essential parts of the Figure and what you want to explain with it.
Response: Thank you for your comments. The title of the diagram has been enriched.
- Introduce adequately the GA algorithms
Response: Thank you for your comments. The definition of genetic algorithm has been described in detail, as shown in paragraph 3 of Section 2.3.
- Move Section 2.4 at the beginning of the section
Response: Thank you for your comments. The beginning of Section 2.4 has been moved to the second paragraph according to the writing logic.
- Describe with major accuracy the experimental data
Response: Thank you for your comments. A supplementary description of the experiment in this paper has been provided. For details, see the first paragraph of Section 2.1.
- Describe in detail the equipment used to make the measurements (air quality and meteorological parameters). Extract this data from the datasheet of the instrumentation manufacturer. To make reading the specifications of the instruments more immediate, you can insert them in a table, listing the instruments used and the specific characteristics for each.
Response: Thank you for your comments. The experimental software environment and platform in this paper have been introduced in detail, see Section 2.6 for details.
- add a map that identifies the area affected by the monitoring, and indicates the position of the monitoring stations
Response: Thank you for your comments. This paper mainly uses fusion model to detect meteorological and air quality indexes, and the data come from Shaanxi Province Air Quality implementation and release system.
- You should better describe the datasets you used, there is no reference to the resource altogether. Where did you download this data? You should specify this so that the reader can reproduce the experiment.
Response: Thank you for your comments. The experimental data in this paper have been explained in detail, and the source of the data set has been explained. See Section 2.1 for details.
- The section relating to the methodologies based on Machine Learning must be enriched. You must summarize the essential characteristics of the methods you have used and justify your choices. Try to summarize what are the strengths and weaknesses of the methods, in this way you can make the reader understand why you have chosen these methodologies.
Response: Thank you for your comments. The advantages, disadvantages, and definitions of using RF, BP, GA, and AQI in this paper have been described in detail. For details, see Paragraph 1 of Section 2.2, Paragraphs 1 and 3 of Section 2.3, and Paragraph 2 of Section 2.1.
- I could not find a detailed description of the evaluation metrics you have adopted. How will you measure your model's performance? This section is essential in order to demonstrate the effectiveness of your methodology. Furthermore, only by adopting adequate metrics will it be possible to compare your results with those obtained by other researchers.
Response: Thank you for your comments. The effectiveness of the research method in this paper has been demonstrated in Section 3.3 by discussing the research results of this paper combined with current domestic scholars' research.
-
Section 3 must be improved.
- In this section you present the results of your work. You should start by summarizing what you have done in your work.
Response: Thank you for your comments. The research process of this paper has been explained in the beginning of Section 3.1.
- Next you should summarize the data you collected and the conditions under which you collected it.
Response: Thank you for your comments. The data and sources collected in this paper have been described in Section 2.1.
- How much data did you collect? How many features and how many records
Response: Thank you for your comments. This article collected data from June 24, 2022 to June 18, 2022 to predict the air quality index. And real-time air quality test data from June 24, 2022 to June 30, 2022, a total of 14 pieces, are detailed in the first paragraph of Section 2.1.
- In the case of Machine Learning applications, data quality is crucial for the success of the analyses.
Response: Thank you for your comments. The data chart of this paper has been optimized, see Figure 8 to Figure 10 for details.
- Figure 8 must be improved: The text is too small and appears blurry, making it difficult for the reader to follow the flow of information. I will not repeat this advice again, it also applies to the other occurrences.
Response: Thank you for your comments. Figure 8 has been optimized and adjusted, as shown in Figure 8. Font and image display have been adjusted for other charts in the paper, as shown in Figure 3, Figure 1, Figure 9 and Figure 10.
-
Section 4 must be improved.
- this section is very short it is better to add to the end of the previous section and rename Results and Discussion
Response: Thank you for your comments. The discussion part has been explained as the third section of "3. Results and discussion", and the title of this part has been adjusted to "Results and Discussion".
69) “2021 Climate Blue Book” Add references to allow the reader to learn more about the topic
85-86) Add references to support these statements.
Response: Thank you for your comments. Related documents of "2021 Climate Blue Book" have been added, see Reference 30 for details.
90) “Lolli et al. (2020)” Use a coherent format for citation, replace with Lolli et al. [12]. I will not repeat this advice again, it also applies to the other occurrences.
Response: Thank you for your comments. The citation format of the article "Lolli et al. (2020)" has been modified to "Lolli et al. [12]". See article reference format for details.
239-254) Check the text format, it seems that the line spacing is different from the rest of the paper
Response: Thank you for your comments. The text format of the article has been checked and corrected.
258-262) Introduce adequately the topic (GA)
Response: Thank you for your comments. GA has been introduced in detail. For details, see paragraph 3 of Section 2.3.
413) rearrange the subplots in Figure 8, so they don't fit, either you order them in three rows and one column or two rows and two columns
Response: Thank you for your comments. Figure 8 has been arranged in three rows and one column, as shown in Figure 8.
431) Make table fit on the same page.
Response: Thank you for your comments. The table and text have been placed on the same page.
483-508) Check the text format, the alignment of the lines is not consistent with the rest of the paper
Response: Thank you for your comments. The text format of the article has been checked and corrected.
Round 2
Reviewer 1 Report
Dears
please cite the related subject in introduction
https://doi.org/10.1007/s13762-022-04367-6
Author Response
Dears
please cite the related subject in introduction
https://doi.org/10.1007/s13762-022-04367-6
Reply: Thank you for your suggestion. The related subjects in https://doi.org/10.1007/s13762-022-04367-6 has been cited, see paragraph 6 of the introduction: RF and neural network models have successfully lent to air quality predictions [30].
Reviewer 2 Report
Accept in present form
Author Response
Accept in present form
Reply: Thanks for the careful reading.
Reviewer 3 Report
In the Abstract the authors claim that "The proposed air quality and meteorological correlation model is applied to forest ecosystem management". However, the authors fail to explain how exactly they apply their model to forest ecosystem management.
In the Abstract the authors claim that "based on the proposed model, six valuable forest ecological management measures are proposed". However, six forest ecological management measures proposed in Section 3.4 are not based on the proposed model.
In the response to report 1 the authors claim that the relationship between the model and proposed management measures is described in Section 3.3. However, there is nothing about the relationship between the model and proposed management measures in Section 3.3.
The authors should find a relationship between "six valuable forest ecological management measures" and the proposed air quality and meteorological correlation model, or give up the attempts to claim that the proposed forest ecological management measures are based on the proposed air quality and meteorological correlation model.
Author Response
In the Abstract the authors claim that "The proposed air quality and meteorological correlation model is applied to forest ecosystem management". However, the authors fail to explain how exactly they apply their model to forest ecosystem management.
Reply: Thank you for your suggestion. It has been added in the article that the relevant models of air quality and meteorology can implement monitoring in forest ecosystem management, help forest managers timely obtain meteorological data, conduct online statistical analysis, and take effective early warning and preventive measures for forest ecosystem management. See the third and fourth paragraphs of Section 2.5 for details.
In the Abstract the authors claim that "based on the proposed model, six valuable forest ecological management measures are proposed". However, six forest ecological management measures proposed in Section 3.4 are not based on the proposed model.
Reply: Thank you for your suggestion. The correlation model of air quality and meteorology proposed in this paper provides data support for forest ecosystem management and forest ecosystem monitoring. In order to facilitate forest managers to carry out effective management, six measures are proposed. The relationship between the two is application. For details, see paragraph 1 of Section 3.4.
In the response to report 1 the authors claim that the relationship between the model and proposed management measures is described in Section 3.3. However, there is nothing about the relationship between the model and proposed management measures in Section 3.3.
Reply: Thank you for your suggestion. It has been added in the article that the relevant models of air quality and meteorology can implement monitoring in forest ecosystem management, help forest managers timely obtain meteorological data, conduct online statistical analysis, and take effective early warning and preventive measures for forest ecosystem management. See the third and fourth paragraphs of Section 2.5 for details.
The authors should find a relationship between "six valuable forest ecological management measures" and the proposed air quality and meteorological correlation model, or give up the attempts to claim that the proposed forest ecological management measures are based on the proposed air quality and meteorological correlation model.
Reply: Thank you for your suggestion. It has been added in the article that the relevant models of air quality and meteorology can implement monitoring in forest ecosystem management, help forest managers timely obtain meteorological data, conduct online statistical analysis, and take effective early warning and preventive measures for forest ecosystem management. See the third and fourth paragraphs of Section 2.5 for details. The role of air quality and meteorology-related models in the implementation and monitoring of forest ecosystem management and the application relationship of air quality and meteorology-related models in the implementation and monitoring of forest ecosystem management are detailed in the third and fourth paragraphs of Section 2.5 and the first paragraph of Section 3.4.
Reviewer 4 Report
The authors addressed the reviewer's comments with attention and modified the paper with the suggestions provided. The new version of the paper has improved both in the presentation and in the contents.
Minor revision:
- Try to enrich the captions of the figures, the reader should be able to read the figure without the need to retrieve the information in the paper. Try to summarize the essential parts of the Figure and what you want to explain with it. For example Figure 5 caption
- Check the format of the table 2 is not coerent with the journal specification
- check the equation alignment
- Figure 8: Make Figure and Caption fit on the same page.
- table 5: Make Table and header fit on the same page.
Author Response
The authors addressed the reviewer's comments with attention and modified the paper with the suggestions provided. The new version of the paper has improved both in the presentation and in the contents.
Reply: Thanks for the careful reading.
Minor revision:
- Try to enrich the captions of the figures, the reader should be able to read the figure without the need to retrieve the information in the paper. Try to summarize the essential parts of the Figure and what you want to explain with it. For example Figure 5 caption
Reply: Thank you for your suggestion. The title of the picture has been explained and supplemented in combination with the content of the picture in the article, and the reference of the corresponding chart in the article has been modified. The title of Figure 1 has been changed to "the change of AQI index value in Xi'an from June 24 to June 30, 2022", the title of Figure 3 has been changed to "the importance of sample characteristics in the process of random forest training", and the title of Figure 4 has been changed to "the connection structure between BP neural network neurons", The title of Figure 5 is changed to "initialization operation, exchange operation and mutation operation flow of genetic algorithm". See the title of Figure 1, Figure 3, Figure 4 and Figure 5 for details.
- Check the format of the table 2 is not coerent with the journal specification
Reply: Thank you for your suggestion. The format of Table 2 has been changed according to the journal form format requirements.
- check the equation alignment
Reply: Thank you for your suggestion. The equation format in the paper has been changed according to the format requirements, and the equation is aligned in the center.
- Figure 8: Make Figure and Caption fit on the same page.
Reply: Thank you for your suggestion. Figure 8 and its caption have been placed on the same page.
- table 5: Make Table and header fit on the same page.
Reply: Thank you for your suggestion. Table 5 and headers have been placed on the same page.
Round 3
Reviewer 3 Report
It is obvious that quality-meteorology correlation models "can be used for forest meteorological and ecological environment monitoring, meteorological monitoring of planting plants, conventional meteorological monitoring of field ecological stations, conventional monioring of water cycle, heat balance, carbon cycle and other research topics, regional water monitoring, provincial, municipal and county meteorological monitoring networks, meteorological monitoring of tourism suitability index of parks and scenic spots and forest fire prevention" (L. 469-475). The third and fourth paragraphs of Section 2.5 are a good explanation of the importance of the model. These paragraphs are appropriate for Introduction.
However, "six valuable forest ecological managed measures" are not based on the proposed model. The finding of the research is as follows: (1) RF+BP+GA model is proposed to predict the air quality and meteorological correlation; (2) for a very small dataset from 24 June 2022 to 30 June 2022, the RF+BP+GA model predictes the AQI better than RF and BP-GA models.
The proposed measure 1 (L. 597-600) does not follow the authors' findings and is not based on the RF+BP+GA model.
The proposed measure 2 (L. 601-609) does not follow the authors' findings and is not based on the RF+BP+GA model.
The proposed measure 3 (L. 610-612) does not follow the authors' findings and is not based on the RF+BP+GA model.
The proposed measure 4 (L. 613-615) does not follow the authors' findings and is not based on the RF+BP+GA model.
The proposed measure 5 (L. 616-619) does not follow the authors' findings and is not based on the RF+BP+GA model.
The proposed measure 6 (L. 620-629) does not follow the authors' findings and is not based on the RF+BP+GA model.
Thus, there is a logical gap between the authors' findings and proposed forest management measures.
I would recommend the authors to exclude "six valuable forest ecological managed measures" from the manuscript. However, in this case the paper would be more suitable for Atmosphere than for Sustainability.
Unfortunately, I cannot recommend the manuscript for publication in present form.
Author Response
It is obvious that quality-meteorology correlation models "can be used for forest meteorological and ecological environment monitoring, meteorological monitoring of planting plants, conventional meteorological monitoring of field ecological stations, conventional monioring of water cycle, heat balance, carbon cycle and other research topics, regional water monitoring, provincial, municipal and county meteorological monitoring networks, meteorological monitoring of tourism suitability index of parks and scenic spots and forest fire prevention" (L. 469-475). The third and fourth paragraphs of Section 2.5 are a good explanation of the importance of the model. These paragraphs are appropriate for Introduction.
Reply: Thank you for your comments. The third and fourth paragraphs of Section 2.5 have been moved to the introduction, providing the research background for the air quality monitoring model in the forest ecological environment monitoring, and complementing the important role of climate change and forest sustainable development. See the second paragraph of the introduction for details.
However, "six valuable forest ecological managed measures" are not based on the proposed model. The finding of the research is as follows: (1) RF+BP+GA model is proposed to predict the air quality and meteorological correlation; (2) for a very small dataset from 24 June 2022 to 30 June 2022, the RF+BP+GA model predictes the AQI better than RF and BP-GA models.
Reply: Thank you for your comments. The purpose of the six forest ecological management measures proposed in this paper is to make suggestions on the sustainable development management strategies of forest ecosystems under climate change. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, this revision will delete the "six valid forest ecological managed measures". We also applied the air quality meteorological related model in the first paragraph of Section 3.4 to the forest ecosystem monitoring, which can provide data support for the forest ecosystem management. The relevant content of promoting the effective management of forest managers is moved to the last paragraph of Section 2.5 for supplementary explanation.
The proposed measure 1 (L. 597-600) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 1 will be deleted in this revision. See Section 3.4 for details.
The proposed measure 2 (L. 601-609) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 2 will be deleted in this revision. See Section 3.4 for details.
The proposed measure 3 (L. 610-612) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 3 will be deleted in this revision. See Section 3.4 for details.
The proposed measure 4 (L. 613-615) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 4 will be deleted in this revision. See Section 3.4 for details.
The proposed measure 5 (L. 616-619) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 5 will be deleted in this revision. See Section 3.4 for details.
The proposed measure 6 (L. 620-629) does not follow the authors' findings and is not based on the RF+BP+GA model.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, measure 6 will be deleted in this revision. See Section 3.4 for details.
Thus, there is a logical gap between the authors' findings and proposed forest management measures.
Reply: Thank you for your comments. The purpose of the six forest ecological management measures proposed in this paper is to make suggestions on the sustainable development management strategies of forest ecosystems under climate change. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring application of the air quality meteorological model, this revision will delete the "six valid forest ecological managed measures". We also applied the air quality meteorological related model in the first paragraph of Section 3.4 to the forest ecosystem monitoring, which can provide data support for the forest ecosystem management. The relevant content of promoting the effective management of forest managers is moved to the last paragraph of Section 2.5 for supplementary explanation.
I would recommend the authors to exclude "six valuable forest ecological managed measures" from the manuscript. However, in this case the paper would be more suitable for Atmosphere than for Sustainability.
Reply: Thank you for your comments. Based on the fact that the RF+BP+GA model in this paper is applicable to the monitoring and application of the air quality meteorological model, this revision will delete the "six valid forest ecological managed measures", and delete the relevant description of "six valid forest ecological management measures" in the abstract and conclusion. This supplements the research in this paper to provide research reference for the AQI value of the correlation model for predicting air quality and meteorology and data support for the analysis of air quality problems. See the contents at the end of the abstract for details.
Unfortunately, I cannot recommend the manuscript for publication in present form.
Reply: Thank you for your comments. By revising the above opinions in detail, and deleting the "six valid forest ecological managed measures" and establishing the logical relationship based on the RF+BP+GA model, this article is revised in order to provide research reference for the AQI value of the correlation model of air quality and meteorology. See the revised part of the abstract, the second and third paragraphs of the introduction, the third paragraph of Section 2.5, and the conclusion for details.
