Assessing the Spatial Mapping of Heat Vulnerability under Urban Heat Island (UHI) Effect in the Dhaka Metropolitan Area

Round 1

Reviewer 1 Report

General comment

This review is on the article “Assessing the spatial mapping of heat vulnerability to urban heat island (UHI) effect in the Dhaka Metropolitan Area”

In general, I found the article a very informative.

I will explain this in more detail by going through the article chapter by chapter:

Abstract

Abstract needs to be improved.

The results of the study are provided only in lines no 24-26. Authors need to provide enough results of the study in the abstract.

Introduction

The introduction is well written and few minor comments on the introduction

Lines 74 -76 and 78-81 citations are missing. 

Lines 132 -142 objectives and research approach are not easy to identify, so I recommend arranging the writing section (Lines 132-142) with a smooth flow.   

Materials and Methods

Figure 1 – Improve the visibility of Thana boundary by reducing the visibility of the road, as your study is much more focused on that.  Also, you do not need to provide the projection parameters on the map (This is valid for the other figures as well). As it is very difficult to read the location names in the Bangladesh Union map, it needs to be improved.

Authors need to justify why only a single Landsat image was used to generate the LST to calculate the Heat vulnerability.

Authors should not include the results in the materials and methods section (Especially the land use classification). As land use classification is one of the key approaches of this study, authors need to provide the full details of the accuracy assessment of land use classification in section 3.1. It is very important to provide the sample locations, used for the accuracy assessment in Figure 2.

Table 1 – Should include the data source for each vulnerability indicator.  

Section 2.5.1, 2.5.2, and 2.5.3 are not required as all the indicators explain in table 1 with details

Overall the methodology section is quite confusing as some of the most important information are not available, for example, what are the data formats for indicators (raster data or vector data) if it is vector what are the technique and methods used for the spatial aggregation of data. I highly recommended addressing all those points in the proper way (suggestion – use table 1 to indicate the data format as well as the spatial resolution).  

Results,

The Tejgaon Industrial Area has been identified as the area having the highest temperature, it is difficult to trace it on the map, as the location information is not represented on the map. So, authors need to provide important locations on the map to improve the readability of the paper.  

Still, we can find the methodology parts in the results (3.3.1, 3.3.2, and 3.3.3). Therefore, I highly recommended moving the methodology section in materials and methods and keeping only the results and their explanation in the results section.

Figure 7 – Need to improve the figure and is it possible to say that there is a correlation between NDVI, NDBI, NDWI with VHI as R² values represent less than 1%. As this is not reflected anywhere in the abstract, discussion, and conclusion it is better to remove it.

Discussion,

Discussion is weak, it is not properly aligned with the research finding of the study, and it is discussing much more general thinks. The discussion should be improved.   

Conclusion

The conclusion is OK, if possible providing the limitation of the study would be good.    

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Extreme heat challenges have been increasingly severe, due to climate change and urbanisation, in many cities all over the world. Beating the urban heat is extremely important and mapping the heat vulnerability is an important approach to presenting the heat challenges. Overall, this paper is good and the analysis is good. However, authors have to make revisions.

[Q1] The introduction should be rewritten to indicate more recent studies on mapping heat vulnerabilities. Namely, authors should change the angle to present the introduction. Authors should also shorten the introduction to remove some abundant analysis on heat impacts.

[Q2] Line 33, please remove the. Please improve the English throughout the paper.

[Q3] Line 37-40, authors should also the increasing urban height leading to ventilation reduction, the reduction of urban greening and water bodies. Please refer He, B. J., Ding, L., & Prasad, D. (2020). Relationships among local-scale urban morphology, urban ventilation, urban heat island and outdoor thermal comfort under sea breeze influence. Sustainable Cities and Society, 60, 102289.

[Q4] Authors must consider heatwave as an important background for analysing urban heat challenges. Please consider the paper: He, B. J., Wang, J., Zhu, J., & Qi, J. (2022). Beating the urban heat: Situation, background, impacts and the way forward in China. Renewable and Sustainable Energy Reviews, 161, 112350.

[Q5] Authors must add the most recent data on extreme heat challenges. Around line 50-63.

[Q6] Information in line 70-94 can be moved to the study area description. Please add more information about the heat vulnerability and the use of heat vulnerability index, please refer:  Xiang, Z., Qin, H., He, B. J., Han, G., & Chen, M. (2022). Heat vulnerability caused by physical and social conditions in a mountainous megacity of Chongqing, China. Sustainable Cities and Society, 103792.

[Q7] Line 142, please more information about the research significance.

[Q8] In Section 2, authors should add more information about the drivers that relevant to the urban heat island formation, please refer to Qi, J., Ding, L., & Lim, S. (2021). Toward cool cities and communities: A sensitivity analysis method to identify the key planning and design variables for urban heat mitigation techniques. Sustainable Cities and Society. Meanwhile, the temperature increase relevant to heatwave should be presented.

[Q9] Section 2.5, please read through the paper Xiang, Z., Qin, H., He, B. J., Han, G., & Chen, M. (2022). Heat vulnerability caused by physical and social conditions in a mountainous megacity of Chongqing, China. Sustainable Cities and Society, 103792. Some new method is integrated via the social media.

[Q10] Please improve the resolution of Figure 2.

[Q11] Section 3.2, please change the title of this section. It should be the Urban heat island map. Not urban heat island. Moreover, please add the time of Figure 2.

[Q12] Figure 7, why are the R square so low in this figure?

[Q13] Please add a section of implications in the discussion section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript entitled “Assessing the spatial mapping of heat vulnerability to urban heat island (UHI) effect in the Dhaka Metropolitan Area” presents a remote sensing study to obtain the HVI for DMA in Bangladesh. The study separates the main characteristics of UHI formation and provides the background of each variable.

In general, the study is promising. However, before considering it suitable for publication in Sustainability Journal, the authors must clarify some points about the study. My main concern is the material used: only one Landsat image, which is unclear why the authors use this specific date (April 18^th, 2021). Second, the methodological sequence is also not precise. Thirdly, the correlation index is too low, with no discussion why do the authors obtain this value. Finally, the study requires an extensive English Grammar Review and manuscript restructuration.

Please, consider the following comments, questions, and minor review separated by each section:

Abstract:

It is recommended not to use so many abbreviations in the abstract. Also, be careful with the confusing sentence on lines 26 and 27, mixing results with the conclusion.

Introduction:

The introduction section presents some works on UHI in a very general way, including for Bangladesh. However, there is no work on heat vulnerability in other parts of the world. The content seems disorganized, not bringing a clear gap to be studied in the manuscript. There is also concern about the cited papers, which may not be related to the content presented through indirect citation.

Minor comments:

- Line 33, page 1 – “The In the”….

- Line 70, Page 2 – Not found this information in any of the references presented.

- Line 77, Page 2 – Unclearly sentence.

- Line 78, Page 2 – reference is missing.

- Line 100, page 3 – reference is not in the journal’s guideline bibliography format.

Materials and Methods

This section describes the study area and the dataset and index calculation. To make it more apparent to the readers, the authors must explain the steps and the procedure to obtain the HVI. Also, it is not clear which values of HVI can be considered stressful or not stressful. The equation (1) is a function, but Table 1 provided does not explain the relation between the variables a,s, and e. The authors must clarify how the Table 1 variables are related and the values to be considered for DMA. Also, it is unclear why only one Landsat image is used and why this date is chosen. Please, make a deeper description of how LST was calculated (and the time of LST obtained). Is NTL calculated or obtained by another source? Is the satellite image period adequate for the study?

Minor comments:

- Line 147, page 3 – keep standard units.

- Line 176, page 5 – Do not start a subsection with “however.”

Results

The results section presents the results of the study. There are some missing points that the authors must clear. The PCA calculation must be explained in the Methods section. Also, the correlation indexes are too low, with no explanation. Also, I recommend that Results and Discussion should be in one section.

Minor comments:

- The authors made figures 2 and 3?

- Figure 5c: Moderate.

- Line 395, page 17: “The In this”….

Discussion

This section is not a discussion about the results. It looks like more of a conclusion section.

Conclusion

This section looks abstract.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The work attempted to assess the spatial profile of heat vulnerability to UHI in the Dhaka metropolitan area. While I do see the value of the work, some intepretations of the data have to be improved. I'd suggest the publication of the work with Sustainability after my comments given below being addressing.

1. While the propossed approve could identify hotspots in the area of your interest, how did you validate the thermal perception of the hotspots?
2.  The linear fittings in Fig.7 do not tell too much as the R^2 is very low in each dataset fitting. I'd suggest the authors to limit their intepretation of the datasets in discussions without the linear fittings.
3. Some recent studies around the topic were missing, as examples, a paper titled 'Urban heat island and its interaction with heatwaves: A review of studies on mesoscale' and the other one titled 'Advancement in urban climate modelling at local scale: Urban heat island mitigation and building cooling demand'.
4. Abstract, the core for the assessment of the HVI, that is, the three components, should be highlighted.
5. The limiations of the work should be discussed in the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I can recommend that this paper is good for publication now.

Author Response

Authors are thankful to the learned reviewer for the positive insight

Reviewer 2 Report

The revision has addressed my concerns, and I recommend publishing it.

Author Response

We are thankful to the learned reviewer for his positive insight. 

Reviewer 3 Report

The authors had improved the manuscript with the reviewer's comments. I still have concerns about the statistics presented and its discussion, which looks like it was removed and changed for a correlogram.

Nevertheless, the authors did not explain the selected indexes and why HVI does not correlate to any selected index. The discussion does not provide any relationship between solar radiation or thermal properties to HVI, which are important factors to discuss UHI.

Author Response

The authors had improved the manuscript with the reviewer's comments. I still have concerns about the statistics presented and its discussion, which looks like it was removed and changed for a correlogram.

Nevertheless, the authors did not explain the selected indexes and why HVI does not correlate to any selected index. The discussion does not provide any relationship between solar radiation or thermal properties to HVI, which are important factors to discuss UHI.

Response: Thank you very much for constructive comments. We incorporated all questions raised by learned reviewer. We did not change other statistics in the manuscript, only changed the statistics of correlation. Because we performed correlation analysis again using more number of points. Previously, we considered less number of points from parameters and HVI, so we got insignificant results. Therefore, in revised version, we did with more points.

We are very sorry that we did not discuss the relationship between HVI and parameters. In the revised version, we did correlation analysis between HVI and parameters using band collection statistics tool in ArcGIS 10.5 software. We discussed the relationship in discussion. We also discussed the relationship between solar radiation and HVI. In the present study, we directly did not consider the solar radiation for HVI calculation. But, it can be stated that the surface and air temperature can be increased because of higher solar radiation and higher rate of absorption of solar radiation by built-up areas. Therefore, it can act as exposure for HVI. As a proxy, We used LST and NDVI (inverse relation) for HVI calculation. We also calculate the correlation for HVI and LST-NDVI and found that high positive relation for LST and adverse relation for NDVI. We again added limitation in conclusion section about the absence of solar radiation for HVI calculation.

“After generation of HVI model with exposure, sensitivity and adaptive capacity, it needs to be identifying the most sensitive parameters for HVI occurrences. Therefore, the correlation has been performed between HVI and HVI conditioning factors. We found that Elderly population (r: 0.61), very young or children (r: 0.607), illiterate people (r: 0.585), female (r: 0.57), LST (0.564), working age people (r: 0.539), poverty (r: 0.51) are very sensitive to heat vulnerability. It is evident that LST can create urban heat island effect, which in turn, causes higher vulnerability to heat. On the other hand, because of chronic medical conditions and health issues, elderly people over the age of 65 are particularly vulnerable to urban heat sensitivity. Young individuals under the age of nine are more sensitive to heat waves than older adults, owing to their lower sweating rate and body mass ratio. Women are more sensitive to heat waves than males owing to their greater body fat percentage (BF%) and physical strength. Illiterate people are particularly vulnerable to heat wave vulnerability because they are unaware of the potential danger of a heat wave. Employed people make up a large portion of the population in a city, and their responses to heat wave sensitivity are varied, and they are mostly exposed due to their work requirements. People living in poverty have poor male nutrition, poor housing conditions, and an obstructive socioeconomic situation, all of which contribute to heat wave sensitivity. In addition, kaccha structure (r: 0.217), disable population (r: 0.34), Floating population (r: 0.37), population density (r: 0.44) have been considered as moderate sensitive to heat vulnerability. Huong et al. [71] reported that disabled persons are more vulnerable to heat exposure owing to their reliance on others, as well as their physical health and fragility. Furthermore, floating people are susceptible to heat wave sensitivity because to their homelessness, lack of access to water, severe sickness, and lack of access to power, as well as their terrible living conditions [71]. It can be said that the sensitivity of a household's Kacha structure may be determined by assessing its susceptibility to heat waves. The Kacha constructions are not fully sturdy, and the heat from the tin roof has a negative impact on the residents. In contrast, we found that road length (0.11), NDVI (0.06), health institution (-0.046), household with electricity (-0.012), packa structure (0.015) have inverse or very low relationship with HVI. Because, larger and broader roads can reduce traffic and land congestion. So, this feature is adaptable. On the other hand, increasing the number of health institutions helps treat and prevent heat wave disasters. Pucca building includes adequate flooring, ceiling, and structural ornamentation, as well as socio-economic variables that adapt to heat waves. Electricity facilitates the use of fans, air conditioners, refrigerators and other cooling devices during a heat wave. Increasing vegetation minimises urban heat island and preserves the environment eco-friendly. When compared to the average temperature of green space and residential areas, differences in solar radiation heating in urban areas are a primary factor contributing to the increase in air temperature and surface temperature in urban areas, particularly around roads, commercial and industrial areas, and especially around residential areas. Therefore, presence of vegetation can reduce the heat vulnerability effect. “

Round 3

Reviewer 3 Report

The authors had improved the discussion about the HVI correlation with other indexes.