Dear Dr. Safarov,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Dec 07 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Jianhong Zhou

Staff Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

This research was funded by the Science Committee of the Ministry of Science

and Higher Education of the Republic of Kazakhstan within the framework of the grant AP19677560 “Monitoring and mapping of the ecological state of the Pavlodar air environment using machine learning methods”.

Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

3. Please remove all personal information, ensure that the data shared are in accordance with participant consent, and re-upload a fully anonymized data set.

Note: spreadsheet columns with personal information must be removed and not hidden as all hidden columns will appear in the published file.

Additional guidance on preparing raw data for publication can be found in our Data Policy (https://journals.plos.org/plosone/s/data-availability#loc-human-research-participant-data-and-other-sensitive-data) and in the following article: http://www.bmj.com/content/340/bmj.c181.long.

4. Please ensure that you refer to Figure 2, 4, 5, in your text as, if accepted, production will need this reference to link the reader to the figure.

5. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. 

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

Reviewer #1: The authors of the manuscript entitled “DynamicSeq2SeqXGB for PM2.5 Imputation in Extremely Sparse Environmental Monitoring Networks” validated the DynamicSeq2SeqXGB model for PM2.5 estimation under extreme sparsity conditions. The article addresses the issue of missing data caused by various factors that affect sensor performance.

Avoid using “we” throughout the paper, including in the abstract.

Why was a study conducted in Pavlodar, Kazakhstan validated using data from Beijing? Please justify this choice of dataset.

The article reads more like a chapter (or several chapters) from a thesis rather than a standalone journal paper.

Overall, the authors have conducted good work. However, it would be useful to discuss whether the proposed method can be applied to other geographical locations similar to Kazakhstan.

Can the method used for imputing missing data also be extended to forecast PM2.5 levels for future years?

Reviewer #2: Dear Author(s),

Your manuscript presents an original hybrid methodology that combines temporal context encoding with a tree-based learning framework (DynamicSeq2SeqXGB) to reconstruct missing PM₂.₅ concentrations in an urban air-quality monitoring network with high rates of data loss. The integration of contextual temporal windows, XGBoost regression, and compression strategies (full and selective) is an intelligent and practical contribution to the field of environmental data reconstruction. The external validation with a Beijing dataset adds further credibility to the generalizability of the model.

Overall, the paper is methodologically strong and well-structured, with promising implications for real-time air-quality management. However, before acceptance, several points require clarification and expansion—particularly in describing model architecture, benchmarking against modern baselines, and improving the clarity of figures and text organization.

Best regards,

Comments about the Abstract Section

• The abstract effectively summarizes the motivation and results. However, it should include the specific study period for the Pavlodar dataset (e.g., “2019–2022”) and for the Beijing validation set to give a clear temporal scope.

• Please briefly clarify what “DynamicSeq2SeqXGB” represents.

• Quantitative performance results (e.g., MAE, R² values) should be briefly included in the abstract.

• To improve readability, reduce the number of abbreviations; keep only the essential ones.

Comments about the Introduction Section

• The introduction properly identifies the problem of incomplete PM₂.₅ data and the need for reliable imputation in sparse monitoring networks.

• However, the literature review could be expanded to include recent machine learning–based imputation studies, especially deep learning approaches such as SAITS, BRITS, or GRIN, and a short discussion explaining how the proposed method differs (e.g., less computationally intensive, interpretable, more robust to short sequences).

• The paragraph introducing the “compression strategies” (full vs. selective) should appear earlier in the introduction to help readers understand the motivation for this methodological innovation.

• The introduction would benefit from a clear statement of hypotheses and contributions, e.g.:

o To test whether dynamic contextual features improve PM₂.₅ imputation accuracy.

o To compare the performance of full and selective compression strategies for different stations.

o To validate the model transferability using a geographically independent dataset.

• A short final paragraph summarizing the manuscript structure (e.g., “Section 2 describes the data and methodology…”) would also help guide readers.

Comments about the Methodology Section

• Please specify the exact period and temporal resolution of the datasets used (e.g., hourly PM₂.₅ data collected from five Pavlodar stations between January 2019 and December 2022).

• In the Data Description part, add a short explanation of the environmental characteristics of each station (industrial, residential, urban background, etc.).

• The model architecture section in Supplementary File S2 describes the use of XGBoost within a MultiOutputRegressor. However, it remains unclear how the “Seq2Seq” concept is implemented in this non-neural structure. Please clarify:

o How are sequential dependencies handled—through sliding windows or autoregressive recursion?

o How are masked outputs treated in the loss function?

o How is the model trained to handle varying gap lengths?

• Include a concise algorithmic flowchart or schematic diagram summarizing the DynamicSeq2SeqXGB process (input → context windows → model → imputed output).

• The hyperparameter selection procedure needs more transparency. Indicate whether parameters such as n_estimators, max_depth, learning_rate, and subsample were optimized via grid search, random search, or expert tuning.

• Confirm that data normalization was based solely on the training data to avoid information leakage from the test set.

• The section describing the compression strategies should explicitly state the purpose of each strategy and provide pseudocode or decision rules for their implementation.

• Define all abbreviations once at first use (e.g., DSXGB, WQI, MAE) and avoid redefining them later.

• The performance metrics (MAE, RMSE, R², MAPE) are appropriate; however, consider adding a brief justification for their selection and limitations (e.g., MAPE sensitivity to low values).

Comments about the Results and Discussion Section

• The results are comprehensive, and the comparative analysis between the “full” and “selective” compression strategies is a highlight of this paper.

• In the first results paragraph, explicitly mention that meteorological features were also modeled, as this strengthens the environmental interpretability of the imputation.

• In Table 5 and Figure 6, emphasize which stations benefited most from each strategy and suggest possible environmental or operational reasons (e.g., sensor stability, industrial influence, or background variability).

• In the discussion of station-specific patterns, consider adding correlation statistics between the “optimal strategy” and site-level indicators such as standard deviation, homogeneity index, or mean PM₂.₅ level. This will make the “station preference” findings more quantitative.

• Add one figure (or supplementary figure) summarizing which strategy performed best for each station using a simple visual indicator (e.g., color-coded bar or pie chart).

• The Beijing validation experiment is a major strength. Please explain in more detail how the artificial missingness pattern was generated (“based on app_pspu gap profile”) and whether alternative missingness types (random, seasonal, clustered) were tested.

• The reported performance (MAE = 8.50 μg/m³, R² = 0.944) is excellent, but the baseline comparisons are limited to simple interpolation and averaging methods. For high-impact publication, please consider adding at least one modern ML baseline (e.g., Random Forest, missForest, or a simple neural network imputer) for stronger benchmarking.

• Include a short analysis of error behavior across PM₂.₅ concentration levels (e.g., whether errors increase during pollution peaks).

• Clarify any potential data leakage risk between training and validation sets due to overlapping temporal windows.

• The discussion could benefit from a short paragraph on the computational efficiency of the proposed method (training time, computational cost), since this is important for real-time deployment.

Comments about the Conclusion Section

• The conclusion is concise and well organized.

• Please consider:

o Adding 1–2 sentences quantifying the performance improvements (e.g., “up to 79% improvement over baseline methods”).

o Highlighting the practical guidance your findings offer (for example, “stations with stable temporal variance benefit from selective compression, while those with high variability perform better with full compression”).

o Mentioning future directions, such as integrating an automatic strategy selector based on station features or exploring real-time adaptive imputation for online monitoring systems.

Comments about the References Section

Some methodological discussions could be further supported by recent literature. For instance, when describing machine-learning-based modeling of particulate matter concentrations, a relevant reference (e.g., Int. J. Environ. Sci. Technol., 2023, 20, 5349–5358) could be considered, as it demonstrates the performance of XGBoost for PM₁₀ prediction in a regional case study.

Additionally, when presenting comparative analyses of ensemble and boosting algorithms applied to long-term air-quality forecasting, another pertinent study (PLOS ONE, 2025, 20(10), e0334252) may be cited to enrich the methodological context.

Reviewer #3: This publication is a complete scientific work that addresses the pressing issue of correctly filling in lost monitoring data. The results of the analysis are presented in the form of tables, graphs, and diagrams, the input data array is available, and the author's DynamicSeq2SeqXGBoost model is available on the git-hub repository. The main content of the article is presented in a scientific style, with material presented in an easy-to-understand manner, relevant graphics, and accurate references to the sources used. The work contains all the necessary components of a scientific publication. The work is research-oriented and scientifically novel, consisting of methodological approaches to the data recovery procedure based on an improved model. The practical application of the research results concerns the implementation of the approach proposed by the authors in monitoring atmospheric air, including other priority pollutants , which will contribute to raising public awareness of the dangers of atmospheric air pollution.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

Attachments

Attachment

Submitted filename: REVIEW.docx

DynamicSeq2SeqXGB for PM2.5 Imputation in Extremely Sparse Environmental Monitoring Networks

PONE-D-25-46692R1

Dear Dr. Safarov,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager®  and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support .

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jie Zhang

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

Reviewer #1: I would like to thank the authors for addressing my points. The paper presented a good method and worth to investigate.

Reviewer #2: The authors have addressed all reviewer comments clearly and comprehensively. The revisions have strengthened the manuscript by improving its clarity, methodological transparency, and practical applicability. No ethical or publication concerns were identified. The revised version is suitable for publication.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: No

**********