Landscape Metrics as Ecological Indicators for PM₁₀ Prediction in European Cities

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript studies the integration of landscape metrics to enhance PM10 prediction models, considering data from 1,216 European AQ stations and applying a random forest model to calculate complex relationships. The manuscript structure is clear, and the content is sufficient. However, there are still some shortcomings, as outlined below:

Abstract:
The abstract provides a comprehensive introduction to the research objectives, methods, and main results, but some expressions are unclear. It would be helpful to add the importance of PM10 prediction in urban landscapes, especially in European cities. The methods section could include how landscape metrics are integrated into the model and explain their role in improving model performance. The conclusion at the end of the abstract could further elaborate on the specific impact of the research results on urban planning. Currently, the conclusion lacks direct connection with the research results and seems overly general.

Keywords:
The abstract could include "European cities" to clarify the focus of the study.

Introduction:
Although a review of the literature related to particulate matter and landscape metrics is already provided, it is recommended to add relevant European landscape policy and reports in the introduction to further supplement the background.

Study Area and Methods:
The methods are clear.

Results:
The impact of climate factors (temperature, precipitation, wind speed) and landscape metrics on PM10 concentrations during heating and cooling periods is discussed, but the interactions of these variables are not sufficiently explained. For example, is the negative correlation between precipitation and wind speed consistent in both heating and cooling periods? Does the relative importance of different variables change significantly across seasons and buffer zones? These could affect the complexity and interpretability of the model.

The use of Spearman correlation is effective in revealing the relationship between the variables and PM10, but there is a lack of discussion on the potential collinearity or multicollinearity between these variables. High correlations could impact the stability and interpretability of the model.

Discussion:
The discussion is clear and does not require significant changes.

Conclusion:
The conclusion mentions that landscape structure (such as LULC and the presence of green spaces) plays an important role in reducing PM10 pollution, especially in high-emission and barrier zones. However, how these landscape configurations specifically affect the distribution and diffusion of PM10 is not well explained, nor is there any mention of regional or seasonal differences in this process.

The conclusion states that temperature’s effect is more prominent in colder seasons and mentions that climate change may affect PM10 levels. However, the discussion of seasonal variations is brief and lacks an in-depth analysis of the differences in how climate factors affect PM10 concentrations across seasons.

The conclusion suggests addressing PM10 pollution through the integration of urban planning strategies, emission control measures, and public health considerations, but it lacks detailed suggestions for specific policies and planning measures. For example, how can "optimizing urban forms to enhance ventilation" be specifically implemented?

The conclusion also does not adequately discuss the main limitations of the current study, which should be added.

Language:
The English language is correct and readable.

Author Response

Dear Reviewer (1),

Thank you very much for your insightful comments and valuable suggestions. We greatly appreciate the time and effort you have taken to improve the quality of our manuscript. the following changes to our manuscript.

 

• The abstract provides a comprehensive introduction to the research objectives, methods, and main results, but;

 

1. some expressions are unclear.

Since no specific expressions were mentioned, we have clarified some terms we found challenging and revised the abstract accordingly: (L12-13): “particulate matter smaller than 10 µm (PM₁₀)”, (L17): “heating (cold) and non-heating (warm) seasons” and (L23): “air quality (AQ) stations”. Please let us know if there are any specific expressions we did not address or clarify.

Also, we added the definition for “paradigm” (L68-69) by adding “(how landscape structure affects ecological processes)”

The term "aerodynamic" has been removed from line 42, and a simplified definition of particulate matter (PM) and PM₁₀ has been included in the introduction section (L44-50):

“Particle matter is a year-round pollutant, comprised of a mixture of suspended solids and liquids in the air categorized based upon size. The particles are produced from industrial, traffic, and geological sources (dust from roads or via chemical reactions in the atmosphere caused by released chemicals from motor vehicles or industrial sources). PM is not visible. Coarse particles (PM₁₀) are 2.5 to 10 μm in diameter. Their small size allows them to make their way to the air passages deep within the lungs where they may be deposited and result in adverse health effects [1]. PM₁₀ also causes visibility reduction.”

Please let us know if any other terms need clarification.

 

2. It would be helpful to add the importance of PM10 prediction in urban landscapes, especially in European cities.

Following your advice, we incorporated a new section at the beginning of the abstract: (L10-14): “Despite significant progress in recent decades, air pollution remains the leading environmental cause of premature death in Europe. Urban populations are particularly vulnerable to pollution from high concentrations of air pollutants, such as particulate matter smaller than 10 µm (PM₁₀). Understanding the spatiotemporal variations of PM₁₀ is essential for developing effective control strategies.”

 

3. The methods section could include how landscape metrics are integrated into the model and explain their role in improving model performance.

Following your advice, we incorporated a new section at (L17-23): “In our previous research, we only calculated the proportion of land uses (PLAND), but according to our current research hypothesis, landscape metrics have a significant impact on PM₁₀ air quality. Therefore, we expanded our independent variables by incorporating landscape metrics that capture compositional heterogeneity, including the Shannon Diversity Index (SHDI), as well as metrics that reflect configurational heterogeneity in urban landscapes, such as Mean Patch Area (MPA) and Shape Index (SHI).”

 

4. The conclusion at the end of the abstract could further elaborate on the specific impact of the research results on urban planning. Currently, the conclusion lacks direct connection with the research results and seems overly general.

Based on your opinion, we removed (L32-33): “This research offers valuable information on urban planning and ecology for sustainable air quality management in European cities” and added a new section at (L33-36): “Our findings suggest that increased urban landscape heterogeneity, smaller patch sizes, and more fragmented land uses associated with PM₁₀ sources such as built-up areas, along with larger and more evenly distributed green spaces, can contribute to the control and reduction of PM₁₀ pollution.”

It is challenging to keep the abstract short and informative while avoiding too many details.

• Keywords: The abstract could include "European cities" to clarify the focus of the study.

We included "European cities" as an additional keyword (L37).

• Introduction:

Although a review of the literature related to particulate matter and landscape metrics is already provided, it is recommended to add relevant European landscape policy and reports in the introduction to further supplement the background.

We added relevant European directive reports to the introduction based on your recommendation: (L39-41): “Air pollution is Europe’s largest environmental health risk, causing cardiovascular and respiratory diseases that impact health, reduce the quality of life, and cause preventable deaths [2].” And (L53-61): “The European Environment Agency (EEA) plays a crucial role in monitoring and assessing air quality, publishing annual reports summarizing pollution levels including PM₁₀ and their effects. EU directives underscore the importance of air quality modeling in managing and mitigating air pollution in European cities. Directive 2008/50/EC, a cornerstone of European air quality legislation, emphasizes the use of modeling techniques to assess and predict pollutant concentrations, particularly in areas where fixed measurements are limited. The directive recognizes the role of modeling in understanding the spatial distribution of pollutants, identifying pollution hotspots, and developing effective mitigation strategies [3]. Modeling plays a crucial role in informing policy decisions, evaluating the effectiveness of emission reduction measures, and predicting future air quality trends in European cities [4].”

• Results:
  The impact of climate factors (temperature, precipitation, wind speed) and landscape metrics on PM10 concentrations during heating and cooling periods is discussed, but

1. the interactions of these variables are not sufficiently explained. For example, is the negative correlation between precipitation and wind speed consistent in both heating and cooling periods?

Thank you for your comment regarding the possible correlation between precipitation and wind speed during the heating and cooling periods. We would like to clarify that our study did not investigate the direct correlation between precipitation and wind speed or any interactions between independent variables. Instead, the Spearman correlation coefficients provided in Figures 2, 3, 4, and 5 strictly assess the relationship between PM₁₀ concentrations and each independent variable included in the model.

The purpose of presenting these correlations is to provide supplementary knowledge on whether these variables influence monthly average PM₁₀ concentrations positively, negatively, or not significantly. This step aligns with our primary objective of improving PM₁₀ concentration modeling by integrating landscape metrics as ecological indicators. The study focuses on identifying significant predictors of PM₁₀ concentrations rather than examining inter-variable relationships.

While interactions between independent variables such as precipitation and wind speed could provide additional insights into the dynamics of PM₁₀ concentrations, they fall outside the scope of our study. Instead, we prioritized the modeling of PM₁₀ levels and the assessment of variable importance to support urban air quality management.

We addressed the importance and complexity of these interactions in the discussion section (L493-508): “The cooling period is generally characterized by transitional weather patterns, which can significantly influence atmospheric dynamics. For example, Wind Patterns Mixing and Precipitation Effects. The cooling period often experiences changes in wind patterns, which can enhance the vertical mixing and dispersion of pollutants, even under high pressure conditions. The increase in wind speed during this period could counteract the trapping effect of high pressure, leading to lower concentrations of PM₁₀. For example, [79] highlights the importance of wind patterns and their influence on the transport and dispersion of pollutants, particularly during seasonal transitions. Furthermore, the cooling period, depending on the specific geographic location, could also be associated with increased precipitation, which can further contribute to the removal of PM₁₀ from the atmosphere through wet deposition. Analyzing long-term meteorological data and PM₁₀ measurements for the study area can provide a comprehensive understanding of the seasonal and temporal variations in this relationship. Understanding these relationships is crucial to developing effective strategies to manage and mitigate PM₁₀ pollution, ultimately improving air quality and protecting public health.”

Also, we mentioned it in the conclusion chapter (L722-723): “Given the complex interplay of these factors, a comprehensive approach to mitigation is essential.”

We appreciate your attention to this aspect and agree that investigating interactions among independent variables, such as those mentioned, could offer valuable insights. We consider this a promising avenue for future research that builds on our current findings. We hope this explanation provides the clarity needed and assures the alignment of our approach with the study’s objectives. Please feel free to share any further suggestions or points requiring elaboration.

 

2. Does the relative importance of different variables change significantly across seasons and buffer zones? These could affect the complexity and interpretability of the model.

Thank you for your comment. We appreciate your suggestion and would like to assure you that we have already addressed the influence of seasonality and buffer zones on variable importance in our analysis.

In Section 3.2, we examined how different variables affect PM₁₀ concentrations during the heating and cooling periods, as well as within the 1000 m and 3000 m buffer zones. Below are some key points from our results:

• Heating Period (1000 m buffer zone):

Temperature, total precipitation, and wind speed were the most influential variables, all showing negative correlations with PM₁₀ concentrations. We also analyzed the impact of landscape metrics and soil texture, which exhibited varying correlations.

• Heating Period (3000 m buffer zone):

Temperature remained dominant, with wind speed and total precipitation showing consistent negative correlations with PM₁₀ concentrations.

• Cooling Period (1000 m buffer zone):

Total precipitation was the leading factor influencing PM₁₀ concentrations, while landscape metrics (e.g., MPS and SHI) and soil texture (e.g., silty clay and loam) showed positive and negative correlations, respectively.

• Cooling Period (3000 m buffer zone):

Air pressure was identified as the most significant variable, showing a negative correlation with PM₁₀ concentrations, suggesting a dispersal effect. Landscape metrics and soil texture also contributed to explaining PM₁₀ variability.

We believe that the analyses presented in Figures 2, 3, 4, and 5 effectively illustrate how variable importance varies across different seasons and buffer zones, enhancing the interpretability of our model.

Thank you again for your valuable feedback. Please let us know if further clarification is needed.

3. The use of Spearman correlation is effective in revealing the relationship between the variables and PM10, but there is a lack of discussion on the potential collinearity or multicollinearity between these variables. High correlations could impact the stability and interpretability of the model.

 

We acknowledge the critical importance of addressing potential collinearity or multicollinearity among the independent variables. “Random Forest is a machine learning technique that is particularly well-suited to handling multicollinearity, a condition where independent variables in a dataset are highly correlated. Unlike linear regression models, which can suffer from inflated standard errors and misleading statistical inferences due to multicollinearity, Random Forests are less affected because of their tree-based structure and random feature selection process during training. Random Forests build multiple decision trees, each using a random subset of features, which reduces the influence of correlated predictors on the model's performance [5,6].” We added this explanation to the methodology chapter (L239-246).

 

To address your remarks regarding possible multicollinearity between independent variables of Spearman's rank correlation coefficient we calculated the Variance Inflation Factor (VIF). Our Variance Inflation Factor (VIF) values are all less than 9, indicating that multicollinearity among the predictor variables is not a concern in our modeling process. These procedures are detailed in the Methods section (L281-288): “To address potential multicollinearity among the independent variables identified through Spearman correlation, Variance Inflation Factor (VIF) values were calculated. This step ensures that the predictor variables used in the modeling process are independent and do not exhibit problematic levels of multicollinearity. The VIF analysis was conducted in Python, using the processed Pandas DataFrame. The results, presented in Appendix B, show that all VIF values are below 9, confirming that multicollinearity is not a concern in our analysis. This additional step strengthens the reliability of our modeling approach.”

The corresponding results are provided in Appendix B (L744). We hope these additions clarify and support our approach.

 

Conclusion:

The conclusion mentions that landscape structure (such as LULC and the presence of green spaces) plays an important role in reducing PM10 pollution, especially in high-emission and barrier zones.

1. However, how these landscape configurations specifically affect the distribution and diffusion of PM10 is not well explained, nor is there any mention of regional or seasonal differences in this process.

The conclusion mentions that landscape structure (such as LULC and the presence of green spaces) plays an important role in reducing PM₁₀ pollution, especially in high-emission and barrier land use land covers. This is influenced by regional and seasonal variations, as heating and non-heating periods lead to differences in pollutant dispersion and concentration patterns. Specific landscape configurations, such as the size, shape, and connectivity of green spaces, significantly impact the diffusion and distribution of PM₁₀. For instance, larger, contiguous patches of forests and urban parks are more effective in mitigating PM₁₀, while fragmented and complex built-up areas tend to exacerbate pollution levels. Please check (L701-705): “Landscape composition, especially the presence of green spaces, is vital to reduce monthly average PM₁₀ levels. Larger patches in high-emission areas (LULC groups 1 and 3) tend to increase PM₁₀, particularly during heating periods due to concentrated emissions. In contrast, larger patches in barrier zones (LULC group 2) help mitigate PM₁₀ levels by acting as buffers or facilitating pollutant dispersion.”

A new section has been incorporated into the conclusion, as suggested, to further enhance it based on your valuable feedback. (L706-710): “This study highlights that the specific landscape configurations, such as the size, shape, and connectivity of green spaces, significantly impact the diffusion and distribution of PM₁₀. For instance, larger, contiguous patches of forests and urban parks are more effective in mitigating PM₁₀, while fragmented and complex built-up areas tend to exacerbate pollution levels.”

2. The conclusion states that temperature’s effect is more prominent in colder seasons and mentions that climate change may affect PM10 levels. However, the discussion of seasonal variations is brief and lacks an in-depth analysis of the differences in how climate factors affect PM10 concentrations across seasons.

Thank you for your valuable comment highlighting the need for a deeper analysis of seasonal variations in how climate factors affect PM₁₀ concentrations. We appreciate your input and would like to clarify that our study already provides a detailed examination of these seasonal differences in section 4.2. Effects of Climatological Variables, as summarized below:

(L451-453): Temperature significantly influences PM₁₀ levels, with its effect being more pronounced during the heating period. During the heating season, its predictive influence is nearly quadrupled compared to the cooling period. This substantial increase in impact is likely due to limited atmospheric mixing and reduced pollutant dispersal during colder months.

(L454-458, 461-475): The study also found distinct seasonal correlation patterns between temperature and PM₁₀ concentration. A negative correlation was observed during the heating season, likely due to increased emissions from heating sources in colder conditions. Conversely, a positive correlation was found during the cooling season, possibly due to enhanced evaporation and the formation of secondary particles.

(L457-460, 476-485): Total precipitation emerges as the dominant predictor during the cooling period, suggesting that it plays a more active role in mitigating PM₁₀ concentrations through washout effects when heating sources are minimal. The study reveals a consistent negative correlation between monthly average total precipitation and monthly average wind speed with monthly average PM₁₀ concentrations in both seasons. This suggests that higher wind speeds and precipitation rates are effective in reducing PM₁₀ levels throughout the year by dispersing particulate matter or facilitating its removal from the atmosphere.

(L486-504): The research found a significant negative correlation between monthly average air pressure at sea level and monthly average PM₁₀ concentration, especially during the cooling period. This contradicts several studies that have indicated a positive correlation between air pressure and PM₁₀ concentrations. This discrepancy might be attributed to changes in wind patterns during the cooling period, which can enhance vertical mixing and pollutant dispersion, even under high-pressure conditions.

We believe these insights underscore the seasonal variations in the influence of climatological factors on PM₁₀ concentrations and offer valuable contributions to developing targeted mitigation strategies. We sincerely appreciate your feedback and remain open to any further suggestions that could enhance the depth and clarity of our analysis.

3. The conclusion suggests addressing PM10 pollution through the integration of urban planning strategies, emission control measures, and public health considerations, but it lacks detailed suggestions for specific policies and planning measures. For example, how can "optimizing urban forms to enhance ventilation" be specifically implemented?

Thank you for your valuable feedback. Numerous policies and planning measures can be implemented to optimize urban forms for enhanced ventilation and reduced PM₁₀ concentrations. Cities can establish design guidelines and regulations that mandate or incentivize developers to incorporate features that promote air circulation within their projects. These could include stipulations for lower aspect ratios in street canyons (a place where the street is flanked by buildings on both sides creating a canyon-like environment), the strategic orientation of buildings to align with prevailing wind patterns, and the incorporation of ground-level porosity through features like arcades or elevated structures [7–9]. Furthermore, zoning regulations can be adjusted to limit building heights and densities in areas designated as ventilation corridors or those adjacent to sensitive land uses like schools and hospitals [9–11]. Policies promoting mixed-use development and reduced reliance on private vehicles can also contribute to improved air quality [9]. Incentivizing green infrastructure, such as green roofs and strategically planted trees, can further enhance air quality through pollutant removal and improved ventilation, provided their placement and species selection are carefully considered [8,12,13]. Top of Form

We have added a new part to the conclusion to address your points and provide more specific suggestions for urban planning strategies. (L727-733): “Cities can adopt design guidelines to enhance air circulation, such as lower aspect ratios in street canyons, strategic building orientations aligned with wind patterns, and ground-level porosity through features like arcades or elevated structures [7–9]. Zoning regulations can restrict building heights and densities in ventilation corridors or near sensitive areas like schools and hospitals [9,11]. Mixed-use development, reduced vehicle reliance, and green infrastructure, such as green roofs and strategically placed trees, further improve air quality when carefully planned [8,12,13].”

We hope these additions will provide a clearer understanding of how these strategies can be implemented to address PM₁₀ pollution.

4. The conclusion also does not adequately discuss the main limitations of the current study, which should be added.

We have added the limitations of our study, followed by suggestions for future studies, at the end of the conclusion. (L833-840): “This study is limited by the thematic accuracy (≥80%) of the Urban Atlas LULC dataset, which may introduce uncertainties into spatial modeling. While landscape metrics such as PLAND, SHDI, SHI, and MPA were effectively used, future research could improve predictions by integrating urban form indexes (e.g., connectivity, compactness, and 3D structural metrics) to better capture spatial configurations. Additionally, using daily PM₁₀ concentrations instead of monthly averages could enhance temporal precision and provide deeper insights into short-term variations in air quality.”

We sincerely hope that the revisions we have made address your concerns and improve the manuscript to your satisfaction. Thank you once again for your valuable feedback and suggestions.

 

Sincerely,

Seyedehmehrmanzar Sohrab

PhD Student, Department of Physical and Environmental Geography
University of Szeged

 

References

1. Subramanian, A.; Khatri, S.B. The Exposome and Asthma. Clin. Chest Med. 2019, 40, 107–123, doi:10.1016/j.ccm.2018.10.017.
2. Beloconi, A.; Vounatsou, P. Revised EU and WHO Air Quality Thresholds : Where Does Europe Stand ? Atmos. Environ. 2023, 314, 120110, doi:10.1016/j.atmosenv.2023.120110.
3. UNION, T.E.P.A.T.C.O.T.E.-P. DIRECTIVE 2008/50/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL. Off. J. Eur. Union 2008, 133.
4. European Parliament; Council of the European Union DIRECTIVE (EU) 2024/2881 on Ambient Air Quality and Cleaner Air for Europe. Off. J. Eur. Union 2024, 2881, 1–70.
5. Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32, doi:10.1023/A:1010933404324.
6. Evans, J.S.; Murphy, M.A.; Holden, Z.A. Modeling Species Distribution and Change Using Random Forest Chapter 8 Modeling Species Distribution and Change Using Random Forest. 2011, doi:10.1007/978-1-4419-7390-0.
7. Choi, W.; Ranasinghe, D.; Bunavage, K.; DeShazo, J.R.; Wu, L.; Seguel, R.; Winer, A.M.; Paulson, S.E. The Effects of the Built Environment, Traffic Patterns, and Micrometeorology on Street Level Ultrafine Particle Concentrations at a Block Scale: Results from Multiple Urban Sites. Sci. Total Environ. 2016, 553, 474–485, doi:10.1016/j.scitotenv.2016.02.083.
8. Huang, Y.; Lei, C.; Liu, C.H.; Perez, P.; Forehead, H.; Kong, S.; Zhou, J.L. A Review of Strategies for Mitigating Roadside Air Pollution in Urban Street Canyons. Environ. Pollut. 2021, 280, 116971, doi:10.1016/j.envpol.2021.116971.
9. Li, S.; Zou, B.; Ma, X.; Liu, N.; Zhang, Z.; Xie, M.; Zhi, L. Improving Air Quality through Urban Form Optimization: A Review Study. Build. Environ. 2023, 243, 110685, doi:10.1016/j.buildenv.2023.110685.
10. Hadžiabdić, M.; Midžić-Kurtagić, S.; Arnaut, S.; Begić, T.; Ćorović, F. Green Cantonal Action Plan for Sarajevo, Bosnia & Herzegovina, Study of Urban Ventilation Corridors and Impact of High-Rise Buildings; 2019;
11. ClimateADAPT Stuttgart: Combating the Heat Island Effect and Poor Air Quality with Ventilation Corridors and Green-Blue Infrastructure; Stuttgart, 2023;
12. Vitaliano, S.; Cascone, S.; D’Urso, P.R. Mitigating Built Environment Air Pollution by Green Systems: An In-Depth Review. Appl. Sci. 2024, 14, doi:10.3390/app14156487.
13. Hassan, A.M.; ELMokadem, A.A.; Megahed, N.A.; Abo Eleinen, O.M. Urban Morphology as a Passive Strategy in Promoting Outdoor Air Quality. J. Build. Eng. 2020, 29, 101204, doi:10.1016/j.jobe.2020.101204.

 

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Although this is not my area of research, I find the article to be well-written and interesting. The only suggestion I have is that some key terms might be clarified to make them more understandable to a wider audience.
Also, I think it would be useful to include additional model validation data to confirm the accuracy of the predictions.
Additionally, I recommend including suggestions for future research directions.

Author Response

Dear Reviewer (2),

Thank you very much for your thoughtful feedback. We have carefully revised our manuscript based on your suggestions to improve its quality and clarity.

 

Comment 1: Although this is not my area of research, I find the article to be well-written and interesting.

1. The only suggestion I have is that some key terms might be clarified to make them more understandable to a wider audience.

Response 1: We appreciate your point about clarifying key terms to make the content more accessible to a wider audience. Ensuring our findings are understandable and impactful for readers beyond our specific field is a priority, and we will revise the manuscript accordingly. Since no specific terms were mentioned, we clarified a few terms we found unclear and revised the abstract: (L12-13): “particulate matter smaller than 10 µm (PM₁₀)”, (L17): “heating (cold) and non-heating (warm) seasons” and (L23): “air quality (AQ) stations”.

Also, we added the definition for “paradigm” (L68-69) by adding “(how landscape structure affects ecological processes)”

The term "aerodynamic" has been removed from line 42, and a simplified definition of particulate matter (PM) and PM₁₀ has been included in the introduction section (L44-50):

“Particle matter is a year-round pollutant, comprised of a mixture of suspended solids and liquids in the air categorized based upon size. The particles are produced from industrial, traffic, and geological sources (dust from roads or via chemical reactions in the atmosphere caused by released chemicals from motor vehicles or industrial sources). PM is not visible. Coarse particles (PM₁₀) are 2.5 to 10 μm in diameter. Their small size allows them to make their way to the air passages deep within the lungs where they may be deposited and result in adverse health effects [1]. PM₁₀ also causes visibility reduction.”

Please let us know if any other terms need clarification.

 

Comment 2: Also, I think it would be useful to include additional model validation data to confirm the accuracy of the predictions.

Response 2: Regarding your suggestion to include additional model validation data, we recognize the importance of confirming the accuracy of predictions for the robustness of our study, We have calculated two additional validation metrics such as RMSE (Root Mean Square Error) and MAE (Mean Absolute Error). These metrics were applied to our test data (prediction set) to validate the model's performance, ensuring both accuracy and comparability with our previous modeling [2]. The results of these validation metrics are presented in Table (Appendix A) to cover your recommendation (L742). Furthermore, we have included a brief explanation of this process in the methodology chapter to provide clarity on the validation approach (L257-258): “RMSE (Root Mean Square Error) and MAE (Mean Absolute Error)” and (L260-264): “Additionally, the inclusion of RMSE and MAE provided a more comprehensive assessment of model performance, quantifying the average magnitude of prediction errors and ensuring robustness [3–5]. These metrics not only confirmed the model's predictive accuracy but also facilitated meaningful comparisons with our previous modeling approaches [2], further strengthening the reliability of our findings.”

We also added a brief explanation of the validation metrics results to the Results chapter and referred to Appendix A in the text. (L306-312): “To enhance model validation, RMSE and MAE metrics were calculated, showing clear improvements compared to the previous study. Incorporating landscape metrics significantly reduced prediction errors. For example, in the 1000 m buffer zone during the cooling period, RMSE decreased from 4.84 to 3.72, and MAE from 3.58 to 2.74. Similar improvements were observed across all buffer zones and periods. These findings confirm the enhanced predictive accuracy of the RF model. Detailed metrics are presented in Appendix A.”

Comment 3: Additionally, I recommend including suggestions for future research directions.

Response 3: We have added the limitations of our study, followed by suggestions for future studies, at the end of the conclusion. (L733-740): “This study is limited by the thematic accuracy (≥80%) of the Urban Atlas LULC dataset, which may introduce uncertainties into spatial modeling. While landscape metrics such as PLAND, SHDI, SHI, and MPA were effectively used, future research could improve predictions by integrating urban form indexes (e.g., connectivity, compactness, and 3D structural metrics) to better capture spatial configurations. Additionally, using daily PM₁₀ concentrations instead of monthly averages could enhance temporal precision and provide deeper insights into short-term variations in air quality.”

We sincerely hope the revisions we have made fulfill your suggestions and address your concerns. Thank you once again for your constructive feedback and kind guidance.

Best Regards,

Seyedehmehrmanzar Sohrab

PhD Student, Department of Physical and Environmental Geography
University of Szeged

 

 

References

1. Subramanian, A.; Khatri, S.B. The Exposome and Asthma. Clin. Chest Med. 2019, 40, 107–123, doi:10.1016/j.ccm.2018.10.017.
2. Sohrab, S.; Csikós, N.; Szilassi, P. Effect of Geographical Parameters on PM10 Pollution in European Landscapes: A Machine Learning Algorithm-Based Analysis. Environ. Sci. Eur. 2024, 36, doi:10.1186/s12302-024-00972-z.
3. Res, C.; Willmott, C.J.; Matsuura, K. Advantages of the Mean Absolute Error ( MAE ) over the Root Mean Square Error ( RMSE ) in Assessing Average Model Performance. 2005, 30, 79–82.
4. Chai, T.; Draxler, R.R. Root Mean Square Error (RMSE) or Mean Absolute Error (MAE)? -Arguments against Avoiding RMSE in the Literature. Geosci. Model Dev. 2014, 7, 1247–1250, doi:10.5194/gmd-7-1247-2014.
5. Karunasingha, D.S.K. Root Mean Square Error or Mean Absolute Error? Use Their Ratio as Well. Inf. Sci. (Ny). 2022, 585, 609–629, doi:10.1016/j.ins.2021.11.036.

 

Author Response File: Author Response.docx