Review Reports - From Prediction to Remediation: Characterization of Tropical Landfill Leachates Using ARIMA and Application of Adsorption and Reverse Osmosis Treatments

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a comprehensive study ARIMA model for tropical landfill leachate management. The research addresses a critical environmental challenge in regions with high seasonal variability, such as Valledupar, Colombia.

1、While the ARIMA (3,0,3) model is justified through stationarity tests (eg, Dickey-Fuller, Ljung-Box), the rationale for selecting this specific model is not discussed. Clarify how competing models were evaluated and why ARIMA (3,0,3) was chosen.

2、The granular activated carbon batch and continuous-flow experiments lack critical operational details. For example: Specify the origin and physicochemical properties of the GAC (eg, surface area, particle size).

3、An indication of whether the leachate has been pre-treated (eg, filtered) prior to adsorption to avoid confounding factors (eg, suspended solids).

4、The membrane specifications (eg, pore size, rejection rates) and pre-treatment steps (eg, microfiltration) are not described. Additionally, the rationale for selecting 18 bar operating pressure and pH 6.0-7.0 should be justified with literature.

5、In Figure 4, the temporal variation of COD/BOD₅ lacks error bars or confidence intervals, making it difficult to assess data variability.

6、Update citations to include recent advancements.

Author Response

Response Letter – Reviewers’ Comments

We sincerely thank the reviewers and the editorial team for their valuable comments and suggestions on our manuscript titled:
“From Prediction to Remediation: Characterization of Tropical Landfill Leachates Using ARIMA and Application of Adsorption and Reverse Osmosis Treatments” (Manuscript ID: 3681334).

Your feedback was instrumental in improving the scientific quality, clarity, and methodological robustness of the study. We carefully revised the manuscript and implemented the suggested changes, including improvements in structure, statistical justification, and discussion.

Below, we provide a detailed response to each comment, indicating the section and line numbers where the corresponding revisions were made.

Reviewer 1.

Reviewer 1, Comment 1: While the ARIMA(3,0,3) model is supported by stationarity tests (e.g., Dickey-Fuller, Ljung-Box), the rationale for selecting this specific model is not fully explained. Please clarify how competing models were evaluated and why ARIMA(3,0,3) was ultimately chosen.

Response:
We sincerely thank the reviewer for this insightful observation. In response, we have clarified and expanded the justification for selecting the ARIMA(3,0,3) model in the revised manuscript. Specifically, a paragraph in Section 2.2.1 – ARIMA Model Selection (lines 93–97) now details the comparison with other candidate models: ARIMA(1,0,1), (2,0,2), (3,0,2), and (2,0,3). Evaluation criteria such as the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Root Mean Square Error (RMSE) were applied, with ARIMA(3,0,3) presenting the lowest scores across all criteria. In addition, Section 2.2.2 – Residual Diagnostics (lines 98–102) now includes the results of the Shapiro–Wilk test (to assess normality) and Breusch–Pagan test (for heteroscedasticity), reinforcing the statistical robustness of the selected model. These additions improve transparency and methodological rigor in the time-series forecasting component of the study.

Reviewer 1, Comment 2: The batch and continuous-flow granular activated carbon (GAC) experiments lack critical operational details. Please specify the origin and physicochemical properties of the GAC used (e.g., surface area, particle size).

Response:
Thank you for this valuable comment. In response, we have incorporated the operational specifications of the granular activated carbon (GAC) used in both the batch and continuous-flow experiments. As now described in Section 2.3.1 – Batch Adsorption System (lines 138–141), the GAC was provided by Sigma-Aldrich, derived from bituminous coal, and has a mineral-based composition. Its specific surface area is 950 m²/g (BET), with a particle size range of 1.5–2.0 mm and a porosity of 0.65, as determined by scanning electron microscopy (SEM). These physicochemical characteristics are also reiterated in Table 3, found in Section 2.3.4 – Continuous Flow Process (lines 180–181), which details the structural and operational parameters of the packed column. These additions enhance methodological clarity and support reproducibility of the adsorption trials.

Reviewer 1, Comment 3: Please indicate whether the leachate was subjected to any pretreatment (e.g., filtration) prior to the adsorption experiments to avoid confounding factors such as suspended solids.

Response:
We appreciate this pertinent observation. In the revised manuscript, we have clarified that no pretreatment procedures (such as filtration or sedimentation) were applied to the leachate prior to the adsorption experiments. This clarification is now explicitly included in Section 2.3 – Removal of Organic Matter in Leachate with Activated Carbon (lines 133–137). The choice to use untreated raw leachate was intentional, aiming to evaluate the performance of the adsorption process under real field conditions, including the natural presence of suspended solids and colloids. This approach provides a more realistic assessment of treatment performance and aligns with the study’s objective to reflect the operational complexities encountered in actual landfill scenarios.

Reviewer 1, Comment 4: Membrane specifications (e.g., pore size, rejection rates) and pretreatment steps (e.g., microfiltration) are not described. Furthermore, the rationale for selecting an operating pressure of 18 bar and a pH of 6.0–7.0 should be supported by literature.

Response:
Thank you for your detailed comment. In response, we have revised the manuscript to include the requested membrane specifications and process rationale. As now described in Section 2.4.2 – Membrane Characteristics and Operating Conditions (lines 192–203), the reverse osmosis membrane used was a thin-film composite (TFC) polyamide membrane consisting of a selective polyamide layer (200 nm), a microporous support (20–50 μm), and a nonwoven backing (120–150 μm). The membrane has a nominal pore size of ~0.0001 μm and achieves contaminant rejection rates exceeding 98% for organic and ionic pollutants. We also clarified that no pretreatment steps (e.g., microfiltration) were applied before membrane filtration to simulate field-representative conditions. Furthermore, the selected operating pressure of 18 bar and pH range of 6.0–7.0 were supported with recent literature (Tejera et al., 2020; Chen et al., 2021; Ahmed et al., 2024) that report these conditions as optimal for leachate treatment using RO membranes. These additions improve the technical rigor and provide clear justification for the experimental configuration.

Reviewer 1, Comment 5: In Figure 4, the temporal variation of COD/BOD₅ lacks error bars or confidence intervals, making it difficult to assess data variability.

Response:
We thank the reviewer for this important remark. In the revised manuscript, Figure 4 has been updated and renumbered as Figure 5, and now includes error bars representing the 95% confidence interval (CI95%) for each COD and BOD₅ measurement. These intervals were calculated based on triplicate analyses conducted during each semiannual sampling period. Additionally, a secondary curve displaying the BOD₅/COD biodegradability ratio has been added to illustrate its temporal evolution. These updates are described in Section 3.1.3 – Physicochemical Composition of Urban Waste Landfill Leachates (lines 361–387), improving the interpretability of the results and providing greater clarity on the variability and biodegradability patterns observed in the leachate.

Reviewer 1, Comment 6: Please update the references to reflect recent advancements.

Response:
We greatly appreciate this recommendation. In response, the manuscript has been updated to incorporate recent and relevant literature published between 2021 and 2024, particularly in the areas of ARIMA modeling, hybrid leachate treatment systems, and AI-enhanced environmental monitoring. These updates are reflected throughout the revised manuscript, particularly in the Introduction and Discussion sections. New references include studies such as Zafra-Mejía et al. (2021) on leachate forecasting using ARIMA, Çavuş & Ertekin (2022) on integrated treatment systems, and López et al. (2024) on the application of machine learning for leachate risk assessment. These additions improve the scientific rigor and demonstrate alignment with cutting-edge developments in sustainable leachate management.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper investigates leachate characterization and treatment at a tropical landfill in Colombia. It combines time-series forecasting (ARIMA model) with two remediation techniques: granular activated carbon (GAC) adsorption and reverse osmosis (RO). However, before publication I would like to point out some week points that can be improved.

My first concern in on the novelty of the manuscript. Since, both ARIMA forecasting and the treatment methods (GAC/RO) are established techniques. Their combination is useful but not particularly novel. Moreover, The Langmuir fit is expected under acidic conditions and is well-documented in literature.
While ANOVA and Tukey are applied, no confidence intervals or error bars are reported. Moreover, ARIMA model diagnostics are correct, but model selection justification (e.g., AIC/BIC) is missing.
The paper omits a life-cycle perspective or any cost-benefit/environmental burden analysis (especially for RO, which is energy-intensive). It would be nice if author can add this in manuscript.

Author Response

Response Letter – Reviewers’ Comments

Below, we provide a detailed response to each comment, indicating the section and line numbers where the corresponding revisions were made.

Reviewer 2.

Reviewer 2, Comment 1: My primary concern relates to the novelty of the manuscript. While both ARIMA forecasting and treatment methods (GAC/RO) are well-established techniques, their combination is useful but not particularly novel. Additionally, the Langmuir fit under acidic conditions is already well-documented in the literature.

Response:
We sincerely thank the reviewer for this thoughtful and constructive comment. While we acknowledge that ARIMA modeling, granular activated carbon (GAC) adsorption, and reverse osmosis (RO) are individually well-established techniques, the novelty of our study lies in the integration of these components into a unified framework, specifically applied to the operational realities of tropical landfills. The manuscript presents a comprehensive methodology that combines time-series forecasting, physicochemical characterization, and treatment validation (adsorption and membrane-based), enabling both predictive and responsive management strategies in highly variable field conditions.

Additionally, the use of raw, untreated leachate—without pre-stabilization or filtration—adds significant practical relevance, as it reflects the true complexity of leachates found in operational settings, particularly in regions with limited infrastructure.

In response to the reviewer’s specific observation on the Langmuir isotherm, we have expanded the discussion in Section 3.2.2 (lines 491–508). Our results show that the Langmuir model provided the best fit at pH 4.0 (R² = 0.9685), indicating dominant monolayer chemisorption. This is consistent with recent studies on leachates with high organic and metal content (Jeppu & Clement, 2021; Duwiejuah et al., 2023; Abdel-Shafy et al., 2024). This contextualization enhances the scientific contribution and positions our findings within the broader international literature.

Reviewer 2, Comment 2: Although ANOVA and Tukey tests are applied, confidence intervals and error bars are not reported.

Response:
We appreciate your valuable suggestion. In the revised version of the manuscript, Figure 4 has been updated and renumbered as Figure 5, and now includes error bars representing the 95% confidence intervals (CI95%) for COD and BOD₅ values, calculated from triplicate experimental measurements. This graphical update ensures better representation of data variability. Additionally, a secondary curve representing the BOD₅/COD biodegradability ratio has been added to illustrate the temporal behavior of the leachate's biodegradability. These modifications are now reflected in Section 3.1.3 (lines 361–387), and the figure caption has also been updated. These improvements enhance the statistical robustness and clarity of the results presented.

Reviewer 2, Comment 3: While the ARIMA model diagnostics are correct, the rationale for model selection (e.g., AIC/BIC) is missing.

Response:
Thank you for pointing out this important aspect. In response, we have incorporated a comparative analysis of multiple ARIMA model configurations based on Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Root Mean Square Error (RMSE). These criteria were applied to evaluate competing models (ARIMA(1,0,1), (2,0,2), (3,0,2), (2,0,3)), and ARIMA(3,0,3) was selected as the best-fitting model due to its lowest AIC, BIC, and RMSE values. This justification has been added to Section 2.2.1 – ARIMA Model Selection (lines 93–97). Additionally, residual diagnostics were conducted (Section 2.2.2) to validate model assumptions and reinforce statistical reliability. These revisions improve the methodological transparency and rigor of the modeling approach.

Reviewer 2, Comment 4: The manuscript omits a life-cycle perspective or any cost-benefit/environmental load analysis, particularly for RO, which is energy-intensive. It would be beneficial to add this component.

Response:
We appreciate this critical observation. In response, we have added a new subsection titled “3.3.4 Energy Consumption and Environmental Perspective of Reverse Osmosis”, located at the end of Section 3.3 and spanning lines 607–623 of the revised manuscript. This section discusses the energy consumption and environmental footprint associated with reverse osmosis (RO) systems, drawing on recent literature (Chen et al., 2021; Tejera et al., 2020; Ahmed et al., 2024; Abdel-Shafy et al., 2024; Xiang et al., 2025). The discussion emphasizes the importance of adopting a life-cycle perspective, especially considering the energy-intensive nature of RO. It also presents hybrid systems and renewable energy alternatives as strategies to mitigate environmental impacts. This addition enhances the manuscript by framing the technical findings within a broader sustainability and policy-relevant context.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors

After reading and critically evaluating the manuscript entitled "From Prediction to Remediation: Characterization of Tropical Landfill Leachates Using ARIMA and Application of Adsorption and Reverse Osmosis Treatments", the following are the main points regarding the weaknesses that need to be corrected or adjusted so that the article is in line with the standards of the Sustainability journal:

Initially, the authors must justify the choice of the ARIMA (3,0,3) model. There is not enough description of the selection process (e.g., use of AIC, BIC or comparison between different orders). Additionally, the discussion of the model results is very focused on stationarity, but there is a lack of robust analysis of the residuals (despite the Ljung-Box), such as tests for normality and heteroscedasticity of the residuals.

In addition, I found the absence of experimental replicates for the laboratory tests. The number of replicates per treatment (in the tests with activated carbon and reverse osmosis) is not explicit. This compromises the statistical robustness of the results.

Additionally, regarding the scarcity of hydrometeorological data to validate correlations between precipitation and leachate generation, there are no time series or statistical analyses that support this association quantitatively (e.g., correlation or regression analysis with precipitation data).

Regarding the pollutant removal results, it was found that the data do not detail the uncertainties. Removal efficiencies (e.g., 97% COD removal) are presented without intervals of confidence, standard deviations or associated experimental. In addition, the use of the ANOVA test followed by Tukey is correct, but the experimental design (especially in the part with activated carbon) suggests the need for a more robust multivariate analysis (e.g. MANOVA or mixed models), given the multifactorial nature (pH, time, dosage)

In this sense, it is recommended that the authors:

1) strengthen the statistical and experimental basis;

2) make the writing of the manuscript more objective and the text should follow the IMRAD structure, with shorter sections;

3) expand the discussion with recent international literature (2020-2024), especially studies that address tropical climates and integration of predictive models with hybrid treatment technologies;

4) Review the academic English to increase the final quality of the text;

5) include implications and limitations at the end of the conclusion and avoid repeating information from the results in the conclusion section, indicating how the results can be replicated or expanded in other tropical or urban contexts.

Comments on the Quality of English Language

Dear Authors,

Review the academic English to increase the final quality of the text.

Author Response

Response Letter – Reviewers’ Comments

Below, we provide a detailed response to each comment, indicating the section and line numbers where the corresponding revisions were made.

Reviewer 3.

Reviewer 3, Comment 1: Although the ARIMA(3,0,3) model is supported by stationarity tests, the manuscript lacks a detailed justification for its selection. There is no clear comparison with alternative models using AIC, BIC, or RMSE. Furthermore, while the Ljung-Box test is mentioned, residual analysis omits critical tests for normality and heteroscedasticity.

Response:
We sincerely thank you for your valuable comment regarding the need for a more detailed justification of the ARIMA model selection and residual analysis. In response, we have expanded Section 2.2.1 – ARIMA Model Selection (lines 93–97) to include a comparative evaluation of candidate models—ARIMA(1,0,1), (2,0,2), (3,0,2), and (2,0,3)—based on Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Root Mean Square Error (RMSE). The selected model, ARIMA(3,0,3), demonstrated the lowest values for all three criteria.

Additionally, we have strengthened Section 2.2.2 – Residual Diagnostics (lines 98–102) by incorporating the results of the Shapiro–Wilk test for normality and the Breusch–Pagan test for heteroscedasticity. These additions confirm the statistical adequacy of the model and ensure that our time-series approach adheres to rigorous validation standards. This comprehensive adjustment enhances both the transparency and reliability of the modeling framework.

Reviewer 3, Comment 2: The manuscript lacks explicit information on the number of experimental replicates for laboratory tests. This omission compromises the statistical robustness of the results.

Response:
We appreciate this critical observation. In response, we have clarified in the revised manuscript that all laboratory experiments—both adsorption and reverse osmosis—were conducted in triplicate (n = 3). This information is now explicitly stated in Section 2.3.1 – Batch Adsorption System (line 143) and Section 2.4.3 – Analytical Methods and Data Collection (line 209). The reported values across the manuscript now reflect mean ± standard deviation, and the application of Shapiro–Wilk and Levene’s tests ensured the validation of normality and homoscedasticity assumptions prior to performing ANOVA and Tukey’s post hoc tests. These updates improve the transparency and statistical integrity of the experimental design.

Reviewer 3, Comment 3: The manuscript lacks time-series data or statistical analysis to substantiate the correlation between rainfall and leachate generation.

Response:
Thank you for highlighting this gap. In response, we have included a statistical analysis using monthly time-series data from 2016 to 2023 to substantiate the relationship between precipitation and leachate generation. The results, based on Pearson correlation and linear regression, revealed a strong and statistically significant association (r = 0.981; R² = 0.962). These findings are now presented in the newly added Figure 2, and the analysis is explained in Section 3.1.1 – Seasonal Leachate Generation Trends (lines 248–270). This addition is supported by recent literature (2020–2025) and reinforces the conclusion that leachate production in tropical environments is closely linked to rainfall variability.

Reviewer 3, Comment 4: The reported removal efficiencies (e.g., 97% COD removal) lack associated uncertainties such as confidence intervals or standard deviations.

Response:
Thank you for your valuable observation. In the revised manuscript, we have updated Section 3.3.1 – Organic Matter Removal Efficiency (lines 533–549) to report removal efficiencies as mean ± standard deviation, calculated from triplicate experimental replicates. Additionally, the 95% confidence intervals (CI95%) are described within the text to reflect the uncertainty and variability inherent to the experimental conditions. These improvements enhance the transparency, reproducibility, and statistical robustness of the treatment performance analysis.

Reviewer 3, Comment 5: The use of ANOVA and Tukey tests is correct, but given the multifactorial design (pH, time, dosage), a more robust multivariate analysis such as MANOVA or mixed models is recommended.

Response:
We thank the reviewer for this valuable suggestion. In response, we conducted a multivariate analysis of variance (MANOVA) to rigorously evaluate the combined effects of pH and activated carbon dosage on the removal efficiencies of COD and BOD₅. The MANOVA results indicated statistically significant effects for both individual factors and their interaction:

Wilks’ Lambda for pH = 0.312, F = 6.78, p = 0.001
Dosage = 0.284, F = 7.12, p = 0.001
Interaction (pH × dosage) = 0.197, F = 4.59, p = 0.005

These methodological improvements have been integrated into Section 2.3.2 – Statistical Analysis (lines 153–157), and the findings are now discussed in Section 3.2.1 – Batch Adsorption Performance (lines 406–413). These additions enhance the statistical rigor of the analysis and provide more robust support for interpreting the effects of experimental conditions on treatment efficiency.

Reviewer 3, Comment 6: 1) Strengthen the statistical and experimental basis

Response:
We have comprehensively strengthened the manuscript’s statistical and experimental components through the following targeted revisions, each of which has been integrated into the relevant sections of the manuscript:

Justification of ARIMA model selection using AIC, BIC, and RMSE, along with residual diagnostics (Shapiro–Wilk and Breusch–Pagan tests) has been added in Sections 2.2.1–2.2.2 (lines 93–97 and 98–102).
An explicit statement of experimental replication (n = 3), with data reported as mean ± standard deviation, was included in Section 2.3.1 (line 143) and Section 2.4.3 (line 209).
A Pearson correlation and linear regression analysis between precipitation and leachate generation (r = 0.981; R² = 0.962) was incorporated in Section 3.1.1 (lines 248–270) and presented in Figure 2.
Confidence intervals (CI95%) and standard deviation are now reported for removal efficiency data in Section 3.3.1 (lines 533–549).
A multivariate analysis of variance (MANOVA) was implemented to assess the combined effects of pH and dosage, and these results are described in Section 2.3.2 (lines 153–157) and discussed in Section 3.2.1 (lines 406–413).

Together, these revisions significantly enhance the methodological rigor and analytical robustness of the study, ensuring that the results are statistically sound and experimentally reproducible.

Reviewer 3, Comment 7: 2) Improve objectivity and adhere to the IMRAD structure

Response:
We appreciate your suggestion to enhance manuscript structure and objectivity. In response, we have revised and reorganized the manuscript to strictly follow the IMRAD structure. All sections—Introduction, Methods, Results, and Discussion—were evaluated and refined to ensure logical progression, clarity, and academic coherence.

The analytical content was reformulated with a more objective tone, redundant descriptions were removed, and concise, evidence-based language was implemented. Subsections were reorganized using a hierarchical structure of headings and subheadings, especially within the Methods and Results. These changes were applied across the entire manuscript to strengthen its readability and scientific rigor.

Reviewer 3, Comment 8: 3) Expand the discussion with recent international literature (2020–2024)

Response:
In response, we have expanded the discussion section to incorporate recent international literature published between 2020 and 2024, which reinforces the relevance and contextual depth of our findings. The following references were added to reflect advancements in tropical leachate management, ARIMA-based forecasting, and hybrid modeling-treatment strategies:

Zafra-Mejía et al. (2021) – Application of ARIMA models in tropical mountain landfills.
Çavuş & Ertekin (2022) – Evaluation of integrated treatment approaches for leachate remediation.
López et al. (2024) – Use of machine learning for environmental risk assessment in landfill systems.
Xiang et al. (2025) – Comparative analysis of advanced leachate treatment technologies.
Cherni et al. (2021) – Review of recent developments in leachate treatment under varying climatic and operational conditions.

Reviewer 3, Comment 9: 4) Review academic English to enhance overall quality

Response:

We acknowledge the importance of clear and precise academic language in communicating scientific findings. Accordingly, the entire manuscript has been thoroughly revised to enhance the quality of written English. Particular attention was given to improving grammatical accuracy, eliminating redundancy, refining technical terminology, and ensuring coherence and logical flow across sections.

These language adjustments improve readability, consistency, and contribute to a more effective presentation of the research. The revised text now ensures better clarity and accessibility for an international academic audience, thereby enhancing the overall scientific impact of the manuscript.

Reviewer 3, Comment 10: 5) Include implications and limitations at the end of the conclusion

Response:
We thank the reviewer for this important recommendation. In response, the Conclusion section (lines 625–647) has been revised to avoid redundancy and to emphasize the broader implications of the study. The final paragraph now clearly outlines how the proposed integrated approach—which combines forecasting, physicochemical characterization, and treatment validation—can support the development of modular and adaptable leachate management strategies in tropical and urban environments.

Additionally, we acknowledge key limitations, including the laboratory-scale nature of the validation, the need for pilot-scale implementation, and the importance of coupling the system with renewable energy sources. These updates strengthen the manuscript by aligning the conclusions with the study’s applied relevance and highlighting directions for future research and innovation.

Reviewer 4 Report

Comments and Suggestions for Authors

The submitted article presents an integrated and technically sound approach for the treatment of landfill leachate in tropical areas, a topic of great relevance for environmental sustainability and urban waste management. The combination of predictive statistical modeling (ARIMA) with advanced treatments such as adsorption with activated carbon and reverse osmosis is innovative and applicable to real contexts.

Strengths
- The study addresses a reality that is underexplored in the international literature, by focusing on landfills located in tropical regions with high rainfall variability.
- The article presents a solid experimental design, with appropriate use of statistical tests (ANOVA, Tukey, Shapiro-Wilk, Dickey-Fuller, Ljung-Box) and basis in international technical standards (APHA).
- The adsorption and reverse osmosis tests showed COD and BOD removal rates above 94%, highlighting the effectiveness of the technologies tested.
- The work has direct applicability in tropical urban contexts, where leachate treatment infrastructure is often precarious.

Points for Improvement
Despite the overall quality of the manuscript, there are some gaps and adjustments to be considered:

- The Materials and Methods section is excessively dense; it should be restructured with clearer subtitles to facilitate reading.
- The discussion lacks a broader comparative analysis with recent literature. It is recommended to reinforce this aspect, highlighting how the results compare with similar studies.
- The temporal variability of COD and BOD data is commented on, but not explored in statistical depth.

Specific Suggestions
- Expand the discussion with references to comparable international studies, especially on adsorption at acidic pH.
- Present a more critical conclusion, considering limitations and possibilities for technological scaling.

Conclusion
I recommend accepting the manuscript after minor revisions aimed at improving clarity, expanding the scientific contextualization, and providing additional elements of practical applicability. The relevance of the topic, combined with its methodological solidity, makes this study a pertinent contribution of interest to the scientific community in the area of waste management and environmental sustainability.

Author Response

Response Letter – Reviewers’ Comments

Below, we provide a detailed response to each comment, indicating the section and line numbers where the corresponding revisions were made.

Reviewer 4

Reviewer 4, Comment 1: The Materials and Methods section is overly dense and should be restructured with clearer subheadings to improve readability.

Response:

We thank the reviewer for this insightful recommendation. In response, the Materials and Methods section (lines 76–222) has been comprehensively revised and restructured to improve clarity, organization, and overall readability. The following specific improvements were implemented:

Clearly defined subheadings were introduced to segment the methodology into logical and coherent components, including:

▪︎ Study area and site description

▪︎ Leachate flow monitoring

▪︎ ARIMA time-series modeling

▪︎ Physicochemical characterization of leachate

▪︎ Batch and continuous adsorption systems

▪︎ Reverse osmosis configuration and operation

▪︎ Statistical analysis

Paragraphs were shortened and repetitive or redundant descriptions were removed to enhance conciseness and improve focus on core procedures.

The structure was reorganized to follow a logical, hierarchical sequence, grouping related methods and analytical techniques together to facilitate ease of navigation for the reader.

Reviewer 4, Comment 2: The discussion lacks a broader comparative analysis with recent literature. It is recommended to reinforce this aspect by comparing the results with similar studies.

Response:
We thank the reviewer for this important observation. In response, the Discussion section has been expanded to incorporate a broader comparative framework supported by recent international literature published between 2021 and 2024. These references were integrated into the discussion of leachate behavior, treatment performance, and predictive modeling, allowing for improved contextualization of our findings. Specifically, we added:

Zafra-Mejía et al. (2021), which validates the use of ARIMA models for leachate forecasting under tropical landfill conditions.
Çavuş & Ertekin (2022), which proposes a hybrid ARIMA–ANN model to enhance environmental forecasting in waste systems.
López et al. (2024), which demonstrates the application of machine learning techniques to improve prediction of leachate parameters and inform real-time operational decisions.

These comparative references were incorporated throughout the Discussion section, particularly in relation to time-series forecasting, hybrid treatment frameworks, and the applicability of predictive tools in leachate management systems. This enriched discussion improves the depth, relevance, and scientific positioning of the manuscript within the global research context.

Reviewer 4, Comment 3: Temporal variability of COD and BOD data is mentioned, but not explored in sufficient statistical depth.

Response:
We appreciate this valuable comment. In response, we have enhanced the statistical treatment of COD and BOD₅ variability by incorporating the following changes in Section 3.1.3 – Temporal Analysis of Organic Load (lines 361–387):

Error bars based on 95% confidence intervals (CI95%) were added to the updated Figure 5, derived from semiannual triplicate measurements (2019–2023), allowing precise quantification of temporal dispersion in COD and BOD₅ concentrations.
The BOD₅/COD ratio was introduced as a secondary variable to evaluate changes in biodegradability trends over time.
A more detailed interpretation was included, emphasizing variations in organic load and biodegradability, and linking these patterns to environmental drivers such as rainfall and landfill operation.

Reviewer 4, Comment 4: Expand the discussion with references to comparable international studies, particularly those addressing adsorption at acidic pH.

Response:
Thank you for your valuable recommendation. In response, we expanded the discussion in Section 3.2.1 – Batch Adsorption Performance (lines 491–508) to incorporate recent international studies (2020–2025) that support improved adsorption performance and Langmuir model fitting under acidic conditions. Specifically, the following references were included:

Jeppu & Clement (2021) – highlighting the effect of pH on monolayer chemisorption behavior;
Duwiejuah et al. (2023) – reporting enhanced adsorption capacity in acidic environments due to electrostatic mechanisms;
Abdel-Shafy et al. (2024) – confirming that low pH stabilizes active adsorption sites and minimizes competition from multivalent ions.

These studies demonstrate that acidic pH conditions improve surface protonation and promote stronger electrostatic attraction, resulting in greater adsorption capacity and better model fit. This extended discussion strengthens the scientific foundation of our findings and aligns the study with current international trends in leachate treatment research.

Reviewer 4, Comment 5: Present a more critical conclusion, addressing limitations and scalability opportunities.

Response:
We greatly appreciate the reviewer’s insightful suggestion. In response, the Conclusion section (lines 625–647) has been revised to adopt a more critical and analytical tone. Rather than merely summarizing results, the revised text now highlights key limitations, including:

The need for long-term performance evaluation of treatment systems;
The importance of addressing membrane fouling dynamics;
The requirement to assess energy efficiency optimization for reverse osmosis treatment in practical settings.

Furthermore, the conclusion explores technological scalability opportunities, such as the design of modular treatment units and the implementation of hybrid system configurations suited to tropical environments. These additions improve the relevance and applicability of our findings, reinforcing their significance for both scientific advancement and the pursuit of sustainable environmental management solutions.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

No more comments.

Author Response

We would like to extend our sincere thanks to Reviewer 1 for their time, insightful suggestions, and constructive comments, which were essential in improving the quality and clarity of our manuscript. We are pleased to provide the following response:

Reviewer 1 – Comment 1:

No more comments.

Response:
We thank Reviewer 1 for their time and thoughtful feedback throughout the review process. We appreciate the opportunity to improve our manuscript based on your valuable suggestions during the first round and are pleased that the revised version has addressed your concerns.

Sincerely,

The authors

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for the improvements. I am satisfied now.

Author Response

We thank Reviewer 2 for the time dedicated to reviewing our manuscript and for the positive assessment provided in this second round. We are pleased to present our formal response below.

Reviewer 2 – Comment 1:

Thanks for the improvements. I am satisfied now.

Response:
We sincerely thank Reviewer 2 for the positive evaluation and final remarks. We greatly appreciate your constructive feedback during the review process, which contributed to enhancing the clarity, structure, and scientific rigor of our manuscript. We are pleased that the revised version has met your expectations.

Sincerely,
The authors

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

After analyzing the “second version of the manuscript”, it is clear that the authors have satisfactorily met the main recommendations made in the previous review. Below, I hhighligt te main points that were successfully revised:

1) Statistical justification for the ARIMA (3,03) model: The new version includes a comparison between candidate models based on criteria such as AIC, BIC and RMSE, technically justifying the choice of the model;

2) Analysis of the model’s residuals: Additional statistical tests (Sapiro-Wilk and Breusch-Pagan) were incorporated, strengthening the robustness of the temporal modeling;

3) Replication of experiments: The lack of replicates, pointed out in the first version, was corrected. The authors report that all tests were performed in triplicate, and the results are now presented wit mean and standard deviation;

4) Correlation between precipitation and leachate generation: The second version includes Pearson’s correlation analysis (r=0.981; R² = 0.962), complementing the association previously treated only descriptively with a statistical basis;

5) Expression of uncertainties in removal efficiencies: Pollutant removal efficiencies (such as COD) now feature dispersion measures and appropriate statistical treatment;

6) Review of statistical analysis: MANOVA analysis was introduced for multifactorial adsorption data (pH x dosage), which meets the experimental design of the study;

7) Improvement in structure and scientific writing: The structure of the articles is more objective and consistent with the IMRAD format. The scientific language has been improved;

8) Update of the literature review: The literature was expanded with current references (2020-2024), including in relation to the integration of predictive models with hybrid technologies.

Point that could still be reinforced (without preventing approval): The conclusion section could more clearly explain the “limitation of the study” and possible “practical implication and replicability” of the results in other contexts. I recommend that this aspect be evaluated during the final editing stage.

In view of the consistent fulfillment of the requests made in the previous round of evaluation, I consider the manuscript APPROVED for publication, recommending only minor editorial adjustments, if necessary, during the final production.

Best Regards

Ad Hoc Reviewer

Author Response

We appreciate Reviewer 3’s attention to the broader scope of the study and the importance of presenting a well-structured and conclusive summary. The following response outlines the specific adjustments made to address this important observation.

Reviewer 3 – Comment 1:

The conclusion section could more clearly explain the “limitation of the study” and possible “practical implication and replicability” of the results in other contexts.

Response:
We thank Reviewer 3 for highlighting the need to clarify the study’s limitations, the relevance of the proposed approach, and its potential for replication in other settings. In response, we have revised the Conclusion section (lines 637–651) to:

Explicitly acknowledge the main limitation of the study, namely the lack of validation under real operational conditions;
Highlight the replicability of the proposed methodological framework, particularly in tropical regions with similar climatic variability and leachate generation patterns;
Recommend future research focused on field-scale implementation, the integration of renewable energy sources, and the inclusion of life cycle assessment (LCA) to evaluate the environmental footprint of both adsorption and membrane-based treatment technologies.

Sincerely,

The authors