The Effect of Urban Greenspace on Land Surface Temperatures: A Spatial Analysis in Sheffield, UK

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article covers the topic: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK.

This article addresses a current issue. Such analyses are important and can add value to science due to their utilitarian nature. The content of the paper is acceptable. The research approach and the way the study was conducted are of appropriate quality. The authors demonstrate a high level of research expertise.

However, the article requires some refinement before publication. Issues for improvement:

1. State the research objective more clearly in the Abstract.
2. Add the research objective and research problems, as well as the structure of the article, in the Introduction.
3. Expand the description of the methodological assumptions and indicate the added value of the chosen research approach (why this particular approach is best for solving the identified research problems). Add research objectives and research problems. It would be beneficial to also include a study outline.
4. Separate the Results and Discussion sections.
5. Expand the Conclusions section.

Author Response

Response to Reviewers’ Comments

Manuscript Number: Land-3772248
Title: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK

 

Dear Editor and Reviewers:

Thank you for your comments concerning our manuscript entitled “The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK”. The valuable and helpful comments have been taken on board in improving our manuscript. We have read the comments carefully and made revisions that we hope can meet your approval. 

Additionally, we have taken the time to repeat the analysis to adjust and amend the results from the previous measure of surface radiant temperature to land surface temperature. We believe that this makes it both more straightforward to interpret our findings and also easier to compare our results with that from other literature (where LST measurements are far more common than surface radiant temperature). We hope that you share this view. We have updated the paper throughout - including new figures and tables.

The specific corrections in the manuscript and the responses to the reviewer’s comments are as follows.

 

Reviewer1:

R1 Comment 1 (R1-1): This article addresses a current issue. Such analyses are important and can add value to science due to their utilitarian nature. The content of the paper is acceptable. The research approach and the way the study was conducted are of appropriate quality. The authors demonstrate a high level of research expertise. However, the article requires some refinement before publication. Issues for improvement: 

R1 Comment 1 (R1-1) response: Thank-you for your comment and support. We are glad that you find the paper acceptable and of an appropriate quality, highlighting the high level of research expertise. We also thank-you for raising the suggestions for improvements below.

 

 

R1-2: State the research objective more clearly in the Abstract.

R1-2 response: Thank you for your suggestion. We have revised the abstract accordingly to clarify the research aim and objectives:

“The aim of this study is to explore how various characteristics of green spaces—including type, configuration (size and shape), location, and distance from the urban centre—affect their cooling effect. Landsat remote sensing land surface temperature data were analyzed through Geographic Information Systems, using Sheffield as a case study.”

 

We believe that this makes it clearer and easier for readers to understand and we hope that you agree.

 

 

R1-3: Add the research objective and research problems, as well as the structure of the article, in the Introduction.

R1-3 response: Thank-you for highlighting this. In the revised manuscript, we have now stated the research aim more explicitly (see line 132): “Given the knowledge gap mentioned above this study aims to quantitatively investigate how various characteristics of green spaces, including type, configuration (size and shape), and location, (distance to urban centre - reflecting an urban rural gradient), affect their cooling effect.” 

 

At the end of the Introduction, we have also added a paragraph outlining the structure of the manuscript (see line 142): “The structure of this manuscript is as follows. In the Materials and Methods section, we introduced the study area (i.e., Sheffield) and its climatic context, the methods for acquiring and processing remote sensing data, the approach to land cover identification, and the method for analyzing LST data. In the Results and Discussion section, we primarily analyzed and discussed the impact and extent of different green space types on their thermal effects, as well as the relationship between landscape metrics and the intensity of these thermal effects. Based on the findings of this study, practical suggestions for urban planners and landscape practitioners were proposed. The conclusion of this study is presented in Section 4.”

 

 

R1-4: Expand the description of the methodological assumptions and indicate the added value of the chosen research approach (why this particular approach is best for solving the identified research problems). Add research objectives and research problems. It would be beneficial to also include a study outline.

R1-4 response: In Section 2.3.2 Calculation of Cooling Effect, we have added an explanation of the rationale for selecting these three quantitative indicators (see line 267): “Three quantitative indicators, including cooling intensity; cooling extent and comparison with baseline temperature were employed in this study with the following two considerations: Firstly, these indicators can help quantitatively describe the cooling effect of green space on the surrounding environment, aligning with our research aim; Secondly, these are commonly used in relevant studies [26, 27, 29, 45] and can facilitate comparison and analysis with the results of similar research.”

We have expanded on many of the assumptions of the method within the limitations section of the work - in line with other reviewer comments (see lines 575-601).

 

 

R1-5: Separate the Results and Discussion sections.

R1-5 response: We understand the concerns here but all authors agree that due to the complexity of the different results that it is preferable for a combined results and discussion section - where results are individually discussed next to where they appear in the text. We believe that this makes the content much easier to understand and reduces confusion and ambiguity. A combined Results and Discussion sections are permitted within the journal of Land. We hope that you understand our position.

 

 

R1-6:Expand the Conclusions section.

R1-6 response: 

We have made some changes to the conclusion section - including a sentence on the key impact of the work for policy makers (see line 628).

 

We wish to thank the reviewer for their time and helpful suggestions which we believe have continued to strengthen our manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear author,

This manuscript explored the impact of different types of urban green spaces in Sheffield on the surface temperature, which has certain theoretical value and practical significance, but there are several key issues in the method design and data processing, which affect the scientificity and promotion of the results:

1. The classification standards for woodlands, grasslands and parks are unclear, the boundaries overlap, the gray space inside the park (roads, squares, buildings, etc.) is not distinguished, and some "park" areas have no vegetation coverage, reflecting the rough classification or even misjudgment. The use of 30-meter resolution images is difficult to reflect the complex internal structure of urban green spaces, lacks classification accuracy verification, and is prone to systematic errors. It is recommended to clarify the boundaries of green space classification, refine the green and non-green areas in the park, and combine high-resolution data or field surveys to improve data quality
2. The study only focuses on green space, ignoring the structure, type and area of the gray space in which it is located. The urban heat island effect is the result of the energy interaction between green space and the surrounding gray space. The layout and nature of the gray space will significantly affect the cooling effect of the green space. Failure to include gray space in the analysis framework leads to an incomplete explanation of the thermal environment regulation mechanism. It is recommended to increase the gray space related variables and combine the green space joint modeling analysis to improve the overall understanding of the urban thermal environment.
3. The fixed width annular buffer zone is delineated with the city hall as the center, without considering the terrain and urban functional heterogeneity, which may cover up the real spatial mechanism. It is recommended to supplement the theoretical basis and optimize the zoning scheme in combination with the terrain, land use and thermal environment characteristics.
4. The conclusion is limited in generalization due to the lack of multi-temporal and cross-regional verification. It is based only on single season and urban data, lacking time and space comparison. It is recommended to supplement multi-temporal and multi-regional data to improve universality.
5. The policy recommendations lack connection with the current planning and social equity perspective, and lack the support of the socio-economic background. It is recommended to deepen the policy-level discussion and pay attention to the issue of green space equity.

Author Response

Response to Reviewers’ Comments

Manuscript Number: Land-3772248
Title: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK

 

Dear Editor and Reviewers:

Thank you for your comments concerning our manuscript entitled “The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK”. The valuable and helpful comments have been taken on board in improving our manuscript. We have read the comments carefully and made revisions that we hope can meet your approval. 

Additionally, we have taken the time to repeat the analysis to adjust and amend the results from the previous measure of surface radiant temperature to land surface temperature. We believe that this makes it both more straightforward to interpret our findings and also easier to compare our results with that from other literature (where LST measurements are far more common than surface radiant temperature). We hope that you share this view. We have updated the paper throughout - including new figures and tables.

The specific corrections in the manuscript and the responses to the reviewer’s comments are as follows.

 

Reviewer 2:

R2 Comment 1 (R2-1): This manuscript explored the impact of different types of urban green spaces in Sheffield on the surface temperature, which has certain theoretical value and practical significance, but there are several key issues in the method design and data processing, which affect the scientificity and promotion of the results:

R2 Comment 1 (R2-1) response: Thank you for highlighting the theoretical value and practical significance of the paper and for raising the suggestions for improvements below.

 

 

R2-2: The classification standards for woodlands, grasslands and parks are unclear, the boundaries overlap, the gray space inside the park (roads, squares, buildings, etc.) is not distinguished, and some "park" areas have no vegetation coverage, reflecting the rough classification or even misjudgment. The use of 30-meter resolution images is difficult to reflect the complex internal structure of urban green spaces, lacks classification accuracy verification, and is prone to systematic errors. It is recommended to clarify the boundaries of green space classification, refine the green and non-green areas in the park, and combine high-resolution data or field surveys to improve data quality

R2-2 response: Our work purposely seeks to draw out the issues of classification of the units of analysis used. This is because currently we identified work that focused explicitly on the role of parks [8,20] - which as you rightly state is no reflection of the vegetation coverage. We do make this point in the paper (see line 87): “However, one problem with analysing parks as the units of analysis is that their land cover (and therefore heat reduction capacity) can be highly varied - including areas of trees, lawn, rocks, soil, water, build environments and so forth [21]” We have edited the text in the methodology to make this point clearer when introducing Figure 3 (see line 248):

“The boundaries of parks purposely overlap with those of woodland and grassland due to definitions, as we sought to investigate the differences between analysis that focussed specifically on parks [8, 20], compared to that which focused on vegetation [17,22].”

We have complemented this with an additional sentence in the discussion to reinforce the point, which we appreciate was not made clearly enough. Thank-you for highlighting this (see line 368):

“This demonstrates the importance of utilising the most appropriate units of analysis whereby for land surface temperature analysis, vegetation coverage rather than park boundaries should be sought.”

 

Despite its limitations, the use of 30m resolution satellite data analysis reflects the current highest resolution of comprehensive freely available data (as stated in the methodology - Section ‘2.2.1. Land surface temperature retrieval’). Our work follows similar work that utilised 30m resolution data [17,26]. Furthermore, a Systematic Review [36] looked into the various imagery used for data analysis and found that 55% of papers cited LANDSAT as a methodology used to study heating. 

Field survey recordings alone could not achieve results at a city scale - required to explore the relationships of the 3 different green space types with size and shape as well as different locations within an urban-rural gradient. We do raise the issue of resolution again in the limitations and future research section. We have included additional text here to highlight the issues raised (see line 579):

“The use of 30-meter resolution images makes it difficult to reflect the complex internal structure of urban green spaces, lacks classification accuracy verification, and is prone to systematic errors.”

 

 

R2-3:The study only focuses on green space, ignoring the structure, type and area of the gray space in which it is located. The urban heat island effect is the result of the energy interaction between green space and the surrounding gray space. The layout and nature of the gray space will significantly affect the cooling effect of the green space. Failure to include gray space in the analysis framework leads to an incomplete explanation of the thermal environment regulation mechanism. It is recommended to increase the gray space related variables and combine the green space joint modeling analysis to improve the overall understanding of the urban thermal environment.

R2-3 response: Our title along with the paper’s aims reflects our focus on green and not grey spaces. This said, we acknowledge the interconnected effects. We do include the grey space in analysis - as land surface temperatures, for example, will reflect the composition of grey spaces. Additionally, the baseline temperature comparison attempts to directly compare the influence of green space compared to areas without green space. We do, however, agree that more comprehensive grey space related variables combined with our green space modelling could help to improve the overall understanding of the urban thermal environment but that this would be outside the current scope of the work (and the timescales available for revisions). We have therefore included the following statement in section ‘3.4. Future work & Limitations’ to address this (see line 598):

“We acknowledge that in future work, the inclusion of more comprehensive grey space related variables combined with green space modelling could help to improve the overall understanding of the urban thermal environment.” 

Additionally, please see our response to the comment below - where in response we have included more detail (maps and tables) regarding the urban characteristics of the surrounding grey spaces.

 

 

 

R2-4:The fixed width annular buffer zone is delineated with the city hall as the center, without considering the terrain and urban functional heterogeneity, which may cover up the real spatial mechanism. It is recommended to supplement the theoretical basis and optimize the zoning scheme in combination with the terrain, land use and thermal environment characteristics.

R2-4 response: One of the reasons for the choice of city (Sheffield) is that like many UK cities, it has a single core centre and broadly follows a concentric ring model - as opposed to more complex multicentre cities. We have added some text concerning this in the Materials and methods section (see line 213):

“Sheffield as a city works as a single core centre that grows radially from the centre with the main urban density reducing as you go further away towards the suburb.”

 

We have now also included additional map content in a new appendix (topography, land use and NDVI maps) (as also recommended by Reviewer 3 - comment 3) and have additionally provided some descriptive statistics for the 6 different zones reflecting the urban characteristics  (including building height and land use for example) - within this new appendix. We believe that these collectively now help the reader understand the case study site in more detail. Thank-you for raising this point that will aid user readability.

 

 

 

R2-5:The conclusion is limited in generalization due to the lack of multi-temporal and cross-regional verification. It is based only on single season and urban data, lacking time and space comparison. It is recommended to supplement multi-temporal and multi-regional data to improve universality.

R2-5 response: Whilst we acknowledge the point, we feel that this would be beyond the scope of the paper which explores the relationships of the 3 different green space types with size and shape as well as different locations within an urban-rural gradient. We have validated the spatial distribution of the single time period (6th September 2023) with that from the period over the entire summer (1st June to 31st August 2023) using simple linear regression. We found a very strong relationship (r-squared of 90%) - justifying our approach. Full results:

y = a + b*x

a (Offset): -8.868264

b (Gain): 1.278550

R (sumXY - sumX*sumY/N): 0.946723

N (Number of elements): 237653

F (F-test significance): 2053712.985029

meanX (Mean of map1): 29.644812

sdX (Standard deviation of map1): 1.978788

meanY (Mean of map2): 29.034099

sdY (Standard deviation of map2): 2.672355

 

Note that we continued to prefer the single time period as this date was specifically chosen as the hottest day of the year and therefore demonstrated a worst case scenario as opposed to the summer average which would be a more conservative measurement. Given, climate change, we feel that this is the most helpful and insightful approach.

Furthermore, we acknowledge the limitation and ensure that within the discussion we compare our findings with those of other papers to extend generalization. 

We have therefore added the following additional text within section ‘3.4. Future work & Limitations’ (see line 584):

“Whilst we acknowledge that our paper focuses on a single city and one time point and therefore lacks multi-temporal and cross-regional exploration, future research should seek to explore temporal patterns. We did, however, compare the spatial pattern of LST between our single day (6th September 2023) and that for the entire summer period (1st June-31st August 2023) and found a strong association (R-squared of 90%) - justifying our data. This single day was specifically selected as the hottest day of the year which therefore in the light of climate change is likely to reflect important human health implications.”

 

 

R2-6:The policy recommendations lack connection with the current planning and social equity perspective, and lack the support of the socio-economic background. It is recommended to deepen the policy-level discussion and pay attention to the issue of green space equity.

R2-6 response: Thank-you for raising this important issue - we agree and have happily included the following text in the policy section of the manuscript (see line 546):

“Firstly, whilst not the focus of our work, findings highlight potential social justice issues in relation to green space cooling effects. Previous work has identified that more deprived areas in Sheffield tend to lack larger high quality green spaces [66] and more widely within England - deprived areas have lower tree canopy cover [67]. Given that findings of this work place a particular importance on the size of green space and on wooded areas for surface temperature reductions, it is likely that deprived areas will be disproportionately disadvantaged. Future work should seek to explicitly investigate these relationships.”

 

References:

[66] Mears, M., Brindley, P., Maheswaran, R., & Jorgensen, A. (2019). Understanding the socioeconomic equity of publicly accessible greenspace distribution: The example of Sheffield, UK. Geoforum, 103, 126-137. doi: https://doi.org/10.1016/j.geoforum.2019.04.016

 

[67] Sales, K., Walker, H., Sparrow, K., Handley, P., Vaz Monteiro, M., Hand, K. L., Buckland, A., Chambers-Ostler, A., and Doick, K. J. (2023). The canopy cover Webmap of the United Kingdom's towns and cities. Arboricultural Journal. doi: https://doi.org/10.1080/03071375.2023.2233864.

We wish to thank the reviewer for their time and helpful suggestions which we believe have continued to strengthen our manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

This study conducts a detailed and comprehensive quantitative investigation of how green space characteristics—including type, configuration (size and shape), and location (particularly distance from the urban center)—influence urban climate improvement. However, several methodological and presentational issues require attention prior to publication consideration.

Sentences require revision for improved clarity and precision. For instance, in the abstract “This study analysed Landsat remote sensing land surface temperature data within the city of Sheffield, UK via a Geographic Information System.” This sentence could be strengthened to better convey the study's scientific contribution.

The absence of urban topographic maps or remote sensing imagery makes it impossible to visually assess the critical distribution of green spaces and impervious surfaces.

The rationality of concentric circular buffers requires validation – it’s unclear whether this design accurately captures the spatial relationship between urban green spaces and impervious surfaces.

Relying on a single satellite image introduces significant uncertainty and randomness. A time-series analysis using multiple images is strongly recommended to verify findings.

The 50m interval LST extraction from 100m-resolution Landsat thermal infrared data is methodologically questionable as interpolation cannot create meaningful higher-resolution data.

The study's small spatial scale (single site) and temporal limitation (one image) raise concerns about result generalizability. Expanding study areas/timeframes would enhance applicability to other regions/periods.

The combined effects of LST inversion uncertainty and resolution constraints necessitate validation. A comprehensive uncertainty assessment should be conducted, encompassing not only data sources but also the modeling process. For methodological guidance, please refer to:

Examining the reliable trend of global urban land use efficiency from 1985 to 2020 using robust indicators and analysis tools

How well do we really know the world? Uncertainty in GIScience

Figures 1-3 require clarification: Ambiguous definition of "heating area" and Unclear individual messages of Figures 1 vs. 2 vs. 3. Recommend adding detailed captions explaining each figure's purpose.

Figure 5 appears redundant and could be merged with other figures to streamline presentation.

Potential confusion exists between: Zone1-6 in Figure 1 and  Zones in Tables 1-2 and Figures 6-7.

Adding panel labels (a, b, c...) would significantly enhance figure readability.

Author Response

Response to Reviewers’ Comments

Manuscript Number: Land-3772248
Title: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK

 

Dear Editor and Reviewers:

Thank you for your comments concerning our manuscript entitled “The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK”. The valuable and helpful comments have been taken on board in improving our manuscript. We have read the comments carefully and made revisions that we hope can meet your approval. 

Additionally, we have taken the time to repeat the analysis to adjust and amend the results from the previous measure of surface radiant temperature to land surface temperature. We believe that this makes it both more straightforward to interpret our findings and also easier to compare our results with that from other literature (where LST measurements are far more common than surface radiant temperature). We hope that you share this view. We have updated the paper throughout - including new figures and tables.

The specific corrections in the manuscript and the responses to the reviewer’s comments are as follows.

 

Reviewer 3:

R3 Comment 1 (R3-1): This study conducts a detailed and comprehensive quantitative investigation of how green space characteristics—including type, configuration (size and shape), and location (particularly distance from the urban center)—influence urban climate improvement. However, several methodological and presentational issues require attention prior to publication consideration.

R3 Comment 1 (R3-1) response: Thank-you for highlighting the detailed and comprehensive investigation of our paper and for raising the suggestions for improvements below.

 

 

R3-2:Sentences require revision for improved clarity and precision. For instance, in the abstract “This study analysed Landsat remote sensing land surface temperature data within the city of Sheffield, UK via a Geographic Information System.” This sentence could be strengthened to better convey the study's scientific contribution.

R3-2 response: Thank you for your comment. We have revised the abstract accordingly to clarify the research aim (see line 9): “The aim of this study is to explore how various characteristics of green spaces—including type, configuration (size and shape), location, and distance from the urban centre—affect their cooling effect. Landsat remote sensing land surface temperature data were analysed through Geographic Information Systems, using Sheffield as a case study.”

 

We hope this makes it clearer and easier for readers to understand.

 

 

R3-3:The absence of urban topographic maps or remote sensing imagery makes it impossible to visually assess the critical distribution of green spaces and impervious surfaces.

R3-3 response: Thank-you for raising this point. This comment aligns with the views of another reviewer too. We agree that adding additional content demonstrating the conditions of the study area would help readers who are likely to be unfamiliar with the study site. As such we have added additional map content in a new appendix (topography, land use and NDVI maps) (as also recommended by Reviewer 3 - comment 3) and have additionally provided some descriptive statistics for the 6 different zones reflecting the urban characteristics  (including building height and land use for example) - within the new appendix. 

 

We believe that these collectively now help the reader understand the case study site in more detail. Thank-you for raising this point that will aid user readability.

 

 

 

R3-4:The rationality of concentric circular buffers requires validation – it’s unclear whether this design accurately captures the spatial relationship between urban green spaces and impervious surfaces.

R3-4 response: The buffer approach is commonly used in the wider literature in similar analyses that operate at city wide spatial scales. We have followed and replicated  such work where various studies that look at land surface temperature using this method to study single core centre cities. For example:

• Chen, W., Zeng, J., Chu, Y., & Liang, J. (2021). Impacts of landscape patterns on ecosystem services value: A multiscale buffer gradient analysis approach. Remote Sensing, 13(13), 2551. https://doi.org/10.3390/rs13132551
• Lei, H., Koch, J., & Shi, H. (2020). An analysis of spatio-temporal urbanization patterns in Northwest China. Land, 9(11), 411. https://doi.org/10.3390/land9110411

We have added this into the method section with the following text (see line 215):

“This buffer approach is commonly used in the wider literature in similar GIS based analyses that operate at city wide spatial scales.”

 

 

R3-5:Relying on a single satellite image introduces significant uncertainty and randomness. A time-series analysis using multiple images is strongly recommended to verify findings.

R3-5 response: This point is responded to within the related comment R3-7 below.

 

 

R3-6:The 50m interval LST extraction from 100m-resolution Landsat thermal infrared data is methodologically questionable as interpolation cannot create meaningful higher-resolution data.

R3-6 response: We acknowledge the point. The approach is common and widely applied - as stated in the methodology. 

 

Despite its limitations, the use of resampled 30m resolution Landsat satellite data analysis is common and widely applied - as stated in the methodology: Our work follows similar work that utilised 30m resolution data [17,26]. Furthermore, a Systematic Review [36] looked into the various imagery used for data analysis and found that 55% of papers cited LANDSAT as a methodology used to study heating. Examples of papers that used this method includes:

• F. Yuan and M. E. Bauer, “Comparison of impervious surface area and normalized difference vegetation index as indicators of surface urban heat island effects in Landsat imagery,” Remote Sens. Environ., vol. 106, no. 3, pp. 375–386, Feb. 2007
• X.-L. Chen, H.-M. Zhao, P.-X. Li, and Z.-Y. Yin, “Remote sensing image-based analysis of the relationship between urban heat island and land use/cover changes,” Remote Sense Environ., vol. 104, no. 2, pp. 133–146, Sep. 2006,
• Gaoyuan Yang, Zhaowu Yua, Gertrud Jørgensen, Henrik Vejre How can urban blue-green space be planned for climate adaptation in high-latitude cities? A seasonal perspective.
• Huiying Fan, Zhaowu Yua, Gaoyuan Yang, Tsz Yiu Liu, Tsz Ying Liu, Carmen Huang Hung, Henrik Vejre, How to cool hot-humid (Asian) cities with urban trees? An optimal landscape size perspective

 

We have added a sentence in the methodology to directly raise this issue (see line 174):

“Whilst we recognise that this may potentially introduce inaccuracies - it is a widely used practice which brings significant benefit of increased spatial resolution” 

 

 

R3-7:The study's small spatial scale (single site) and temporal limitation (one image) raise concerns about result generalizability. Expanding study areas/timeframes would enhance applicability to other regions/periods.

Whilst we acknowledge the point, we feel that this would be beyond the scope of the paper which explores the relationships of the 3 different green space types with size and shape as well as different locations within an urban-rural gradient. We have validated the spatial distribution of the single time period (6th September 2023) with that from the period over the entire summer (1st June to 31st August 2023) using simple linear regression. We found a very strong relationship (r-squared of 90%) - justifying our approach. Full results:

y = a + b*x

a (Offset): -8.868264

b (Gain): 1.278550

R (sumXY - sumX*sumY/N): 0.946723

N (Number of elements): 237653

F (F-test significance): 2053712.985029

meanX (Mean of map1): 29.644812

sdX (Standard deviation of map1): 1.978788

meanY (Mean of map2): 29.034099

sdY (Standard deviation of map2): 2.672355

 

Note that we continued to prefer the single time period as this date was specifically chosen as the hottest day of the year and therefore demonstrated a worst case scenario as opposed to the summer average which would be a more conservative measurement. Given, climate change, we feel that this is the most helpful and insightful approach.

Furthermore, we acknowledge the limitation and ensure that within the discussion we compare our findings with those of other papers to extend generalization. 

We have therefore added the following additional text within section ‘3.4. Future work & Limitations’ (see line 584):

“Whilst we acknowledge that our paper focuses on a single city and one time point and therefore lacks multi-temporal and cross-regional exploration, future research should seek to explore temporal patterns. We did, however, compare the spatial pattern of LST between our single day (6th September 2023) and that for the entire summer period (1st June-31st August 2023) and found a strong association (R-squared of 90%) - justifying our data. This single day was specifically selected as the hottest day of the year which therefore in the light of climate change is likely to reflect important human health implications.”

 

 

 

R3-8:The combined effects of LST inversion uncertainty and resolution constraints necessitate validation. A comprehensive uncertainty assessment should be conducted, encompassing not only data sources but also the modeling process. For methodological guidance, please refer to:

Examining the reliable trend of global urban land use efficiency from 1985 to 2020 using robust indicators and analysis tools

How well do we really know the world? Uncertainty in GIScience

R3-8 response: Thank-you for raising this point. The first paper provided is very interesting but it is noted that the entire purpose of the paper is explicitly to test and validate the data. In contrast, in our paper, we seek to quantitatively investigate how various characteristics of green spaces, including type, configuration (size and shape), and location, (distance to urban centre - reflecting an urban rural gradient), affect their cooling effect. Therefore, it would be impossible to provide the same level of detail to validate findings. 

We hope that our soft validation through comparison of the spatial pattern of LST between the single time period and that from the entire summer period (as previously discussed in the previous response) partially offsets these concerns.

 

R3-9: Figures 1-3 require clarification: Ambiguous definition of "heating area" and Unclear individual messages of Figures 1 vs. 2 vs. 3. Recommend adding detailed captions explaining each figure's purpose.

R3-9 response: Thank-you for raising this issue. The captions have been refined for Figure 1, 2 and 3 - explaining the maps further. The new captions include:

Figure 1. Map showing temperature variation within Sheffield along with the different 2 km buffer zones. Temperature variation here is brightness calculated through remote sensing imagery and is used to understand areas that experience more heat.

Figure 2. Map showing overlap of woodland obtained from Land Cover Map (LCM) (shown in dark green hatch) overlaid on base map indicating green spaces. This provides an idea of how many of the green spaces are classified as woodland as per LCM.

Figure 3. Map locating areas of parks, woodland and grassland within Sheffield.

 

 

R3-10: Figure 5 appears redundant and could be merged with other figures to streamline presentation.

Potential confusion exists between: Zone1-6 in Figure 1 and  Zones in Tables 1-2 and Figures 6-7.
R3-10 response: The scale of the Figures would make it difficult to reflect the material of Figure 5 within wider views of Figures 1-3. We would be happy to move this an appendix if the reviewer felt this more appropriate but we feel that readers might misinterpret the temperature baseline comparison without being in the main text.

Responding to the potential confusion between zones and distance from green spaces, we have tried to be consistent with the terminology as both measures are distance related. We have amended the first column in Table 1 from ‘Distance’ to ‘Distance from green space’ for clarity and consistency. Thank-you for raising this.

 

R3-11:Adding panel labels (a, b, c...) would significantly enhance figure readability.

R3-11 response: In our other publications within Land - we have always followed this same format for Figures. We would be happy to alter the Figure formatting/labelling in the editing stage post review, as required.

 

 

We wish to thank the reviewer for their time and helpful suggestions which we believe have continued to strengthen our manuscript.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

Thank you for your thoughtful response and revisions. The newly added site context enhances the presentation of the study. Your work is particularly commendable in the comparative analysis of the proportions of woodland, grassland, and parkland in the zonal structure and their impact on land surface temperature. I have two further comments:

1. In Table A2, you present land use types in different areas of Sheffield. You should further clarify whether green space (such as parks, grasslands, and woodlands) exists as a standalone use or is integrated into specific land use types. It is recommended to add the following research directions to the discussion: how the combination of green space and land use affects the regulatory effect of surface temperature; the joint effects between different green space types, especially how their spatial distribution and interactions affect surface temperature regulation.

2. Sheffield's administrative boundaries are currently uneven, and the proportion of zonal structures 5 and 6 is particularly low, which may affect the reliability of the conclusions. It is recommended to point out this limitation and improvement directionin the discussion.

Author Response

Response to Reviewers’ Comments

Manuscript Number: Land-3772248 Title: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK

Thank-you kindly for your comments. We have responded in turn under each individual comment below.

 

R2 Comment 1 (R2.1): Thank you for your thoughtful response and revisions. The newly added site context enhances the presentation of the study. Your work is particularly commendable in the comparative analysis of the proportions of woodland, grassland, and parkland in the zonal structure and their impact on land surface temperature. I have two further comments:

R2 Comment 1 (R2.1) Response: We welcome your positive comments regarding our revision and would like to thank the reviewer for their time and continued suggestions. 

 

R2.2: 1. In Table A2, you present land use types in different areas of Sheffield. You should further clarify whether green space (such as parks, grasslands, and woodlands) exists as a standalone use or is integrated into specific land use types. It is recommended to add the following research directions to the discussion: how the combination of green space and land use affects the regulatory effect of surface temperature; the joint effects between different green space types, especially how their spatial distribution and interactions affect surface temperature regulation.

R2.2 Response: Thank-you for raising this important point. To clarify, Table A2 does not contain land use types but building use. As such, green spaces are separate to this and not included in these data. Please can we point you towards Table 2 in the main paper - as the most relevant and appropriate data for green space by Zone interpretation. We have amended the text for Table A2 to make this clearer (changing the existing text of “Building Land Use” to “Building Use”). The contents of Table A2 is separate to land use data used in Figure A2. We believe that these changes make these important distinctions clearer.

The purpose of Table A2 is to show how the zones vary and not to explicitly explore the relationships between green space and land/building use. As such, we do not feel comfortable discussing how the combination of green space and land use affects LST but do acknowledge that this is important future work. We have therefore added a sentence in the Future work and Limitations section [line 607]:

“In particular, future work should seek to explore how the combination of green space and land use affects the regulatory effect of surface temperature; especially how the spatial distribution and interactions affect surface temperature regulation.”

Thank-you for raising this point and bringing it to our attention. 

 

R2.3: 2. Sheffield's administrative boundaries are currently uneven, and the proportion of zonal structures 5 and 6 is particularly low, which may affect the reliability of the conclusions. It is recommended to point out this limitation and improvement direction in the discussion.

R2.3 Response: Thank-you for raising this point. Our analysis is limited to only urban areas (see Figure 1). We have added the following text into the Materials and Method section (2.2.1. Land surface temperature retrieval subsection) [line 223]: “Zones 5 and 6 in particular contain large amounts of undeveloped rural land, however, it should be noted that our analysis excludes such areas, as only the urban land areas are included, as shown within Figure 1.” 

As suggested we raise the point again in the limitations as surrounding rural areas will of course affect the temperature within the urban areas of these zones [line 601]: “It should be noted that in particular Zones 5 and 6 contain large amounts of surrounding rural areas which are likely to influence the temperatures of nearby adjacent urban areas in the study area within our findings. Future work should explore and incorporate the temperature effects of these surrounding areas.” 

Additionally, we also raise the issue within the Discussion when talking about the findings related to Zone 5 and 6, with the following new text at line 395 (new text underlined; exiting text without underlining):

“At the same time, we also found that regardless of the type of green space, the cooling intensity in Zones 5 and 6 is relatively weak. This may in part be influenced by the surrounding rural areas affecting temperatures within the urban study area - which is predominately an effect only in Zones 5 and 6 (as evident in Figure 1).”

We would like to take the opportunity to thank-you for your comments and suggestions that have continued to improve our manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

Although this study has undertaken substantial work in an attempt to precisely differentiate temperature variations among distinct urban zones to demonstrate the cooling effect of UGS, it suffers from several fundamental issues that significantly undermine the reliability of its analysis, thereby rendering the research largely inconclusive.

First, the spatial resolution of Landsat8 Band 10 is 100m, whereas the spacing between zones in this study is set at 50m. The image resolution is insufficient to support the analytical scale employed here (resampling to 30m does not enhance the inherent resolution).

Second, the study relies on temperature data from only a single time point. It is well-established that temperature is influenced to a greater extent by meteorological factors (such as wind) than by surface characteristics. A snapshot of temperature is susceptible to transient weather conditions and cannot adequately represent long-term patterns. Consequently, this study fails to establish a robust correlation between LST and surface features.

Lastly, Landsat temperature retrieval carries considerable uncertainty, typically exceeding 1°C. The temperature differences observed among zones in this study fall within this margin of error. The distinctions in temperature between zones are likely heavily influenced by retrieval inaccuracies. The study neither check this error margin nor accounts for its impact. Thus, the zonal temperature differentiation analysis lacks validity.

To address these limitations, it is recommended to consult the previously suggested literature for guidance in refining the analysis and methodology.

Author Response

Response to Reviewers’ Comments

Manuscript Number: Land-3772248 Title: The Effect of Urban Greenspace on Land Surface Temperatures: a spatial analysis in Sheffield, UK

Thank-you for your comments. We have responded in turn under each individual comment below.

R3 Comment 1 (R3.1): Although this study has undertaken substantial work in an attempt to precisely differentiate temperature variations among distinct urban zones to demonstrate the cooling effect of UGS, it suffers from several fundamental issues that significantly undermine the reliability of its analysis, thereby rendering the research largely inconclusive.

R3 Comment 1 (R3.1) Response: We thank the reviewer for acknowledging the efforts of our work. We disagree, however, that the interpretation of the analysis is “inconclusive”. Our work identifies that the cooling effect of woodland was significantly stronger than that of grassland and urban parks, with a cooling intensity reaching up to 2.93°C, and a cooling extent that can reach up to 500 meters beyond its boundary. The size of woodland had a greater reduction in land surface temperature than the shape of the woodland. Our findings are frequently supported by the work of others, for example (see line 312):

“From Figure 6, it can be seen that the LST of woodland and its surrounding environment is consistently the lowest among the three types of green spaces, particularly within a relatively close distance to green spaces. This is supported by the findings of Schwaab et al. (2021): by comparing LST and land-cover data across 293 European cities, which found that tree-covered urban green spaces have a cooling effect that is 2-4 times higher than that of treeless urban green spaces [47]. Findings from many similar studies around the world also support this [17, 29, 48, 49].”

 

R3.2: First, the spatial resolution of Landsat8 Band 10 is 100m, whereas the spacing between zones in this study is set at 50m. The image resolution is insufficient to support the analytical scale employed here (resampling to 30m does not enhance the inherent resolution).

R3.2 Response: The spacing between the six Zones is 2km (except Zone 1 which was 1km to capture the city core). The distance from green space was measured at intervals of 50m, up to 500m. Effects were sought across multiple distance intervals to identify trends. As such, we believe that the resolution is acceptable. The data represents the highest resolution currently available and is widely accepted in the literature.

Despite its limitations, the use of resampled 30m resolution Landsat satellite data analysis is common and widely applied - as stated in the methodology: Our work follows similar work that utilised 30m resolution data [17,26]. Furthermore, a Systematic Review [36] looked into the various imagery used for data analysis and found that 55% of papers cited LANDSAT as a methodology used to study heating. Examples of papers that used this method includes:

• F. Yuan and M. E. Bauer, “Comparison of impervious surface area and normalized difference vegetation index as indicators of surface urban heat island effects in Landsat imagery,” Remote Sens. Environ., vol. 106, no. 3, pp. 375–386, Feb. 2007
• X.-L. Chen, H.-M. Zhao, P.-X. Li, and Z.-Y. Yin, “Remote sensing image-based analysis of the relationship between urban heat island and land use/cover changes,” Remote Sense Environ., vol. 104, no. 2, pp. 133–146, Sep. 2006,
• Gaoyuan Yang, Zhaowu Yua, Gertrud Jørgensen, Henrik Vejre How can urban blue-green space be planned for climate adaptation in high-latitude cities? A seasonal perspective.
• Huiying Fan, Zhaowu Yua, Gaoyuan Yang, Tsz Yiu Liu, Tsz Ying Liu, Carmen Huang Hung, Henrik Vejre, How to cool hot-humid (Asian) cities with urban trees? An optimal landscape size perspective

 

We acknowledge the resampling of 30m data issue in the methodology (see line 174):

“Whilst we recognise that this may potentially introduce inaccuracies - it is a widely used practice which brings significant benefit of increased spatial resolution” 

 

Furthermore, in the Section 3.4 Future work and limitations, we discuss the potential of high resolution (3.5m-sq) LST satellite data in the future.

Many of the features of interest can be quite large - for example the average urban woodland in Sheffield according to the Parks and Countryside data is just over 5.25 hectares in size - meaning that even a resolution of 100m does not appear insufficient and supports the work undertaken.

Finally, please note that cooling effects were identified as far away as 500m - demonstrating that the resolution is able to identify such effects without issue and effects are felt far outside the resolution of the input LST data. Such findings support the approach employed.

 

R3.3: Second, the study relies on temperature data from only a single time point. It is well-established that temperature is influenced to a greater extent by meteorological factors (such as wind) than by surface characteristics. A snapshot of temperature is susceptible to transient weather conditions and cannot adequately represent long-term patterns. Consequently, this study fails to establish a robust correlation between LST and surface features.

R3.3 Response: We purposely selected a single time point as this represented the hottest day of the day in order to demonstrate effects on a worst case scenario. This is already stated in the paper (line 589):

“Whilst we acknowledge that our paper focuses on a single city and one time point and therefore lacks multi-temporal and cross-regional exploration, future research should seek to explore temporal patterns. We did, however, compare the spatial pattern of LST between our single day (6th September 2023) and that for the entire summer period (1st June-31st August 2023) and found a strong association (R-squared of 90%) - justifying our data. This single day was specifically selected as the hottest day of the year which therefore in the light of climate change is likely to reflect important human health implications.”

Given climate change, we believe that it will be more representative of future predictions to use a single hottest day approach than a longer term period - which will average out important variability. We note that another reviewer raised this same point in our earlier submission but is now happy with the revisions that include more explicit text covering this point in the limitations section, alongside the association of the spatial distribution of LST for our single day with that from the longer summer period of the same year.

 

R3.4: Lastly, Landsat temperature retrieval carries considerable uncertainty, typically exceeding 1°C. The temperature differences observed among zones in this study fall within this margin of error. The distinctions in temperature between zones are likely heavily influenced by retrieval inaccuracies. The study neither check this error margin nor accounts for its impact. Thus, the zonal temperature differentiation analysis lacks validity.

R3.4 Response: As with all data of this type, there will be unavoidable uncertainty and inaccuracies (see How to Lie to Maps, ISBN-13: 978-0226435923). However, we strongly disagree that the work lacks validity. We have followed accepted approaches to measure LST at the most appropriate current spatial resolutions at a city wide spatial scale. We used both cooling intensity and baseline comparisons to draw our conclusions. A Systematic Review [36] looked into the various imagery used for data analysis and found that 55% of papers cited Landsat as a methodology used to study heating. 

 

Please note that one might expect relatively modest differences when comparing zone by zone as differences are likely to be gradual through distance, and it should be the trends that are key.

However, critically and contra to the reviewer’s statement, Tables 4, 5 and all six Figures in Appendix B all demonstrate total zonal effects exceeding the cited typical uncertainty levels of 1°C. Given that these were generated using two separate approaches (cooling intensity and comparison to a baseline temperature), we are confident in the validity. 

Our findings of distance effects through the urban gradient (shown through our Zones), are also supported by other literature, such as Jonsson (2004) who identified increased woodland cooling effects in more urban areas [57] - further demonstrating the validity of our work (see line 442).

 

 

R3.5: To address these limitations, it is recommended to consult the previously suggested literature for guidance in refining the analysis and methodology.

R3.5 Response: We thank the author for the previously suggested literature. The Goodchild reference raises the issue of uncertainty and its implications but does not provide any guidance on how to overcome these important issues. 

The Zhong et al reference is interesting but its application is very different (global urban land use efficiency rather than surface temperature measurements). It is noted that the entire purpose of the paper is explicitly to test and validate the data. In contrast, in our paper, we seek to quantitatively investigate how various characteristics of green spaces, including type, configuration (size and shape), and location, (distance to urban centre - reflecting an urban rural gradient), affect their cooling effect. Therefore, it would be impossible to provide the same level of detail to validate findings. Furthermore, replication of global scale work of this type is impractical for our analysis as off the shelf products for global LST do not exist (unlike for population data such as Worldpop and GHS-Pop). We also note that the use of such satellite derived population data (Worldpop and GHS-Pop for example) have been shown to have just as many limitations and uncertainties as the use of LST and our approach and have frequently been shown to be a poor reflection of actual population distributions from census derived material (https://doi.org/10.5194/essd-11-1385-2019; https://doi.org/10.3390/ijgi10100681; https://doi.org/10.1371/journal.pone.0271504; https://doi.org/10.3390/su10051363). 

 

We would like to take the opportunity to thank-you for your comments and suggestions through the review process.