Multi-Scale Analysis of Green Space Patterns in Thermal Regulation Using Boosted Regression Tree Model: A Case Study in Central Urban Area of Shijiazhuang, China
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors- Please rearrange the keywords alphabetically.
- Abbreviations should be introduced only once in parentheses, after which the abbreviation alone should be used consistently. However, this guideline is not followed in the article. For example, the term "urban heat island (UHI)" is defined both on line 32 and again on line 130.
- It is recommended that the authors provide appropriate references for the content presented in lines 78 to 88. If this material is based on their own analysis or data, they should clearly explain the methodology used to extract or generate this information.
- The text in Figure 1 is difficult to read due to its small size and low resolution. It is recommended that the authors enhance the clarity of the figure by increasing the font size and ensuring high image quality for better readability.
- Figure 1 is not referenced or explained in the main text of the article. To ensure coherence and clarity, the authors should explicitly mention and describe Figure 1 within the relevant section, explaining its relevance to the study.
- Equations 1 to 7 are not accompanied by any references. Given that constant values are used in solving these equations, providing appropriate citations would enhance the credibility and transparency of the methodology. It is recommended that the authors reference the original sources from which the equations and constants were derived.
- The content presented in Figure 10 is not clearly readable due to its low resolution. It is recommended that the authors provide a higher-quality version of the figure to ensure that all details are legible and interpretable.
What is the main question addressed by the research?
- The research's central question is how urban green space patterns, specifically at various scales, influence thermoregulation in urban environments, with a case study in Shijiazhuang, China. The study uses a boosted regression tree model to explore the non-linear relationships between specific landscape metrics (e.g., patch density, edge density, and shape index) and land surface temperature (LST) across different spatial scales.
• Do you consider the topic original or relevant to the field? Does it
address a specific gap in the field? Please also explain why this is/ is not
the case.
2)The topic is not only highly relevant to urban climate studies and urban planning, but it also introduces a novel approach. The study's original and captivating use of a multi-scale analysis combined with machine learning (boosted regression tree model) to explore the cooling effects of urban green spaces. This research addresses a gap by analyzing the synergistic effects of green space patterns at various spatial scales, which has not been extensively explored in previous studies, particularly in high-density cities like Shijiazhuang.
• What does it add to the subject area compared with other published
material?
- This study contributes a novel methodological framework that integrates multi-scale landscape metrics with machine learning models, enhancing our understanding of how urban green space configurations can be optimized for heat mitigation. It focuses explicitly on non-linear relationships and scale-dependent cooling mechanisms, offering actionable insights for urban planners in high-density cities. These insights empower planners with the knowledge to make informed decisions about green space planning for heat mitigation.
• What specific improvements should the authors consider regarding the
methodology?
- The methodology appears solid overall; however, further elaborating on the BRT model's validation process and considering additional model comparisons (e.g., decision trees vs. other regression models) could strengthen the study. It would also be beneficial to include a more detailed discussion on how the spatial resolution of the datasets (such as Landsat 8 and Sentinel-2) might influence the results, particularly in urban areas with fragmented green space.
• Are the conclusions consistent with the evidence and arguments presented and do they address the main question posed? Please also explain why this is/is not the case.
- Yes, the conclusions are consistent with the evidence and clearly summarise how different green space metrics contribute to thermal regulation across different scales. The study successfully addresses the main question, linking landscape patterns with cooling effects and offering a systematic framework for urban planning. This successful addressing of the main question reassures the audience of the study's thoroughness and relevance.
• Are the references appropriate?
- The references appear comprehensive and relevant to the subject area. However, it would be helpful to include more recent studies focusing on machine learning applications in urban heat island mitigation and additional comparisons of landscape metrics.
• Any additional comments on the tables and figures.
- As noted in the previous review, Figures 1 and 10 need higher resolution for clarity. Additionally, Figure 1 should be referenced within the main text for better integration with the discussion. For Figure 10, a more detailed description of the optimization matrix and its implications for green space strategies could further enhance its utility for planners.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsThe article is well structured and has a comprehensible central theme.
Different granularities of the statistical areas to be analysed were used to consider the different requirements for green spaces. I consider this to be a good method for a rough assessment of the decisive parameters that influence the thermal regulation of a green space in conjunction with the adjacent surface temperatures.
A further positive point is linked to the large number of parameters that make it possible to analyse green spaces both qualitatively and quantitatively. It is thus possible to formulate appropriate recommendations for action depending on the level of detail of the statistical unit with the help of the indicators used and the various granularities.
As already critically described in the paper, the analysis of the green spaces is only concerned with ‘conventional’ green spaces. Areas that trigger synergy effects with the green space and may also have a thermally regulating effect as a result are neglected (e.g. water areas). The differentiated (thermal) effects of green spaces, which vary depending on the time of year and day as well as the vitality of the green spaces (e.g. drought stress), are also neglected. When considering the temperature, reference is also made here to the surface temperature. It would also be interesting to analyse what effects the consideration of the air temperature instead of the surface temperature has on the result.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors- reference for climate zones in Fig 1?
- Section 2.2.2. More information on choice of the model (did authors try any other models, why was this one chosen, any literature for similar applications) and choice of model parameters during implementation
- Figure 3 - an inset with the map of the lcoation (like a cartographic map) will be helpful to correlate the lcoation and areas of the city)
- "(1) ED’s marginal benefits increased exponentially with scale, contrasting with linear patterns in tropical climates like Bangkok" - where does Bangkok suddenly turn up from?
- The work would benefit from a section on sensitivity and error analysis since you are combining two sources of data.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThis article investigates the multi-scale thermal regulation mechanisms of urban green spaces in Shijiazhuang, China, employing a Boosted Regression Tree (BRT) model to analyze nonlinear relationships between landscape metrics and land surface temperature (LST). By integrating remote sensing data, multi-granularity grid systems, and machine learning, the study proposes a three-tier optimization framework to enhance green space planning in high-density urban areas. The comments below may help the authors improve the article:
- Clarify the validation methodology for the SVM-classified green spaces against Gaode satellite imagery, including quantitative accuracy metrics like kappa coefficients.
- Address potential biases in LST retrieval from Landsat 8 due to temporal mismatches between image acquisition and extreme heat events.
- Elaborate on the handling of null-value grids during fishnet segmentation and its impact on statistical robustness.
- The discussion of multi-scale green space optimization strategies would benefit from engaging with recent work on "Envisioning the Invisible: Unleashing the Interplay Between Green Supply Chain Management and Green Human Resource Management: An Ability-Motivation-Opportunity Theory Perspective Towards Environmental Sustainability," which provides a robust framework for integrating ecological and managerial dimensions of sustainability, particularly in high-density urban contexts.
- Discuss the absence of socioeconomic drivers in green space distribution analysis and its implications for planning relevance.
- Explain why the coefficient of variation in patch size was retained despite its low cooling contribution.
- Quantify the cooling efficiency disparity between Shijiazhuang’s semi-arid climate and tropical regions in edge density responses.
- To strengthen the policy implications of threshold-driven green space planning, consider citing "Study on the Dynamic Analysis of the Evolutionary Game and Influence Effect of Green Taxation in Promoting the Development of New Energy Industry," which offers empirical insights into how fiscal instruments can incentivize sustainable infrastructure development and align with climate-adaptive governance.
- Strengthen the novelty claim of the three-tier framework by contrasting it with existing multi-scale green space strategies in the literature.
- Justify the exclusion of blue infrastructure (water bodies) in the multi-scale cooling analysis despite their acknowledged role in humidity regulation.
- Discuss the implications of resampling LST data from 30m to 900m resolution on thermal pattern accuracy.
- The manuscript’s emphasis on climate-resilient urban systems could be enriched by referencing "The Need for the Green Economy Factors in Assessing the Development and Growth of Raw Materials Companies," which critically evaluates the role of ecological metrics in balancing resource efficiency with socio-environmental outcomes, a key consideration for Shijiazhuang’s green space strategies.
- Expand the limitations section to address the lack of diurnal LST variations and their impact on threshold generalizability.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Comments and Suggestions for Authorsaccept
Reviewer 4 Report
Comments and Suggestions for AuthorsAccept as is.