Evaluation Strategies on the Thermal Environmental Effectiveness of Street Canyon Clusters: A Case Study of Harbin, China

Round 1

Reviewer 1 Report

Review Sustainability:

Evaluation strategies on the thermal environmental effectiveness of street canyon clusters: A case study of Harbin, China

 

Dear authors,

 

This manuscript presents a method to evaluate and optimize the thermal environment of different types of streetscapes. The problem statement questions whether there is a difference in the thermal environment between the street intersections and the street interior and how to renew the street trees accurately. The manuscript contains a lot of information that contributes to the current state of this research field since field measurements, ENVI-met modeling and statistical techniques are used to derive conclusions. This manuscript fits well in the scope of the journal Sustainability since the authors want to contribute with their research to improve the streetscapes for human life and well-being. I acknowledge that extensive work has been carried out to obtain the results and conclusion. There is however still some work to be done to get the largest value out of the data that was shown in the current version of the manuscript. I believe the authors can publish a very interesting manuscript when the major changes in the result and discussion section are fulfilled.

General concept comments

The authors already published a similar article but with the focus on the direction of a street. The current manuscript builds upon this and contains a bunch of new information. It would however have been better to focus on SVF, TVF and BVF when all streets have a similar direction to exclude differences in shading because of the direction. The manuscript provides an advancement of the current knowledge by building upon earlier scientific knowledge that has been peer reviewed.

Some general comments:

In the first section the topic is nicely introduced and the methods are written in a clear and concise way.

 

Please, refer to code used in the manuscript when this was published in a scientific article or when it is freely available on the internet. When you wrote the code yourself, please add it to an open source platform or mention who can be contacted to obtain the code. In this way your work will have more impact.

 

Make sure that all graphs have a x-axis and y-axis title, so that the reader can interpret the graph without reading the text.

 

Mention out of which data graphs are constructed. This is not clear for the figures in the sections 4.2, 4.3, 4.4, and 5.2. Did you use data of one point in the street canyon/ intersection at several time intervals or an average value over the simulation period for each location in the street canyon/intersection? Please mention this in the text and in the title of the figures.

 

 

Some general comments for particular sections:

 

In the literature review section “2.1 Urban vegetation and thermal environment” the examples seem to be quite randomly chosen to me or do they all apply for Harbin? Please make this clear or include only what is relevant for Harbin.

 

The section “3.2 Research framework” is rather a summation of what can be seen in Figure 3. The manuscript would improve if you elaborate this section by writing the research framework in full sentences instead of a list of things that were done. 

 

Figure 3 is a very nice overview figure. The figure might even improve if you add numbers, as mentioned in the text, in the corner of the boxes.

 

The authors do not give the answer to the general problem statement as mentioned in lines 172-175, while this should be done in the conclusion. A general answer on the question whether there is a difference between the street intersections and the street interior is missing.

 

The measurement campaign as described in the section “3.4.1 Data acquisition” is rather short but the frequency of 30 s makes that enough data was collected. The results might however differ on a different day so it would be good to mention throughout the manuscript that the results are only valid under those weather circumstances. It is further not mentioned in this section and Table 4 which fish-eye lens was used during the campaign. Mention for example the brand. Was the walking speed tracked to keep a constant pace? If so, please mention which device was used and what the walking speed was instead of the vague wording in the current version of the manuscript. Please, add the itinerary or route and the walking direction on one of the maps for example Figure 1c or create a new map with the point measurements. What technology or which device did you use to geolocate the measurement points? Please add this to this section.

 

I would call the “result verification” “model verification” since you test how well the ENVI-met model can reproduce the measurements. I would also add scores such as the bias, mean absolute error (MAE) and root mean square error (RMSE) to this section. Further, the correlation analysis of MRT and wind speed is missing. Please add this and have a look at my specific comment further down to improve this section.

 

In section 4.1. it is not clear at first sight why the streetscape types are created and why the streets are appointed to a streetscape type, so please make this clear in the beginning of this section. I do not understand how streets were appointed to a particular type, what were the thresholds? I think it would be better to appoint the points to a particular streetscape using for example cluster analysis instead of a full street, because one part of a street can differ quite a bit compared to another part as it is for example the case for the NE-SW-1 street. This may cause the large variations in the box-plots (long tails) of the results section. In the current version of the manuscript, it seems that the streetscape type was appointed visually based on the spatial distribution maps, while this is not a scientifically sound method to apply. Please elaborate on which method was used so this is clear. It would be beneficial to add a panoramic or fish-eye photo of each streetscape type. This can for example be presented in a table, accompanied by the name of the streetscape type and some characteristics of this type e.g. TVF > a value.  Further, it should be mentioned which interpolation method was used in ArcGIS to obtain the spatial distribution maps. Additionally, I was wondering whether this is a new classification system that was used or has this method been applied before? If it is the latter, please add some references.

 

In sections 4.2. and 4.3. a precision of 0.01 °C is used while the measurement equipment to validate the model has a precision of 0.3 °C. Therefore it doesn’t make sense to report values with such high precision. Please round all values up to a precision of 0.1 °C and do not conclude that there is a difference when values are in a range of 0.3 °C because anything less than this cannot be measured and is thus too small to detect.

 

In the “Simulation result” section it is not clear whether you subtract the value of the control group from the experiment group value or vice versa. Please mention this in the beginning of section 4.2.1. and mention in the following sections on RH and MRT that you used the same method. Further, it is confusing that a reduction in temperature/RH/MRT is positive and an increase is negative, so please modify this.

It is difficult to understand what the difference stands for. Maybe it is better to show the CG next to the EG box-plots so it becomes clear for the reader what is going on. Is it for example because there is already a big difference in the CG scenario that larger differences are obtained for particular streets? This is not clear in the current version of the manuscript because only differences are presented.

Which intersections are studied exactly and where are the simulation points located? Please indicate which intersections are studied by labeling them on one of the maps in the earlier figures and show a figure or the simulation grid.

 

Sections 5.1 and 5.2 still belong to the result section and not to the discussion section.

 

I think it would help the reader in section 5.2 if you show graphs of TVFcg and TVFeg combined with Ta, RH and MRT instead of or additionally to showing the difference because it is quite complex to understand what is going on when you just show the difference.

 

In the discussion section I miss a discussion where these results are compared to previous literature and I miss concrete guidelines for adjusting streetscapes to a thermal enjoyable environment, so please add guidelines for every streetscape type you studied in this manuscript. Further, you could also emphasize on the difference between the locations with a high and a low SVF, representing the intersections and narrow urban street canyons respectively.

 

The conclusion is rather a short and concise description of the results. This is good but the conclusion should also contain the answer to the research question and now this is not clearly answered. Further, some general advice for future research and for the design of streetscapes should be included in the conclusion.

 

 

 

Specific comments

 

Please find the remarks for the indicated line number, table or figure below: 

 

22: Write simulations instead of simulation since multiple simulations were done.

 

24-26: You can only say this when you mention as well that this was the scenario with the least spacing between the trees and that smaller spacing has not been tested and might even lead to a more comfortable thermal environment.

 

27: Please express changes in the unit of the variable, i.e. °C for temperature and mean radiant temperature, since a relative difference can be large when the values of the variable are low.

 

29: Do you mean “cost” instead of “economy”?

 

31: In which way is this a critical point for further investigation? I did not find this in the text.

 

32-34: These aspects are not described in the manuscript.

 

39-41: Since when is this trend of overheating visible? Please add a source.

 

42: You do not mention that this is about Harbin. Suggestion: “The temperature in the main area of the city Harbin is significantly higher than in the suburbs…”

 

45: This is vage, give numbers.

 

46-47: Try to use more relevant sources for this paragraph.

 

50-51: Please add a source that contains this information.

 

58: Another good source is:

Milošević, D. D., Bajšanski, I. V., & Savić, S. M. (2017). Influence of changing trees locations on thermal comfort on street parking lot and footways. Urban forestry & urban greening, 23, 113-124.

 

59: It is a feasible technical approach in Harbin, but not for all cities. Please add therefore “Harbin” in this line or the name of the broader region for which this appies.

 

68: Source 8 Shashua-Bar et al. (2000) encloses the study of urban forests, rather than forest areas as it is described in the current version of the manuscript.

 

72: SVF has not been mentioned before, so write first “sky view factor (SVF)”.

 

78-79: Mention that these numbers are for a study in Montreal, these numbers will differ depending on the climatic zone and longitude.

 

98-100: This sentence is not clear to me. What is exactly the best? Now it seems like two methods are the best, but only one thing can be the best.

 

103: Do not mention the nationality of researchers. Write rather: “In Cesena, Italy, researchers studied the microclimate of Bufalini square and found that a green-colored surface …”

 

112-113: I do not understand what you mean with “and the perimeter configuration of the residences provided”.

 

121: Sky viewing factor can be removed.

 

126-129: Do you refer to a particular map here? If so please refer to the source and rephrase the sentence otherwise to make clear you speak about maps of indicators in general.

 

134: Mention wherefore TVF and BVF stand for.

 

135: Mention Boston and I think you mean: “…examined the spatial distribution and amount of street trees their shade”.

 

136: “... another study”, study was missing in the sentence. I was also wondering if they used big data to obtain GSV or is it rather using GSV to obtain a big amount of data so the SVF can be calculated? If it is the latter, please adjust the sentence.

 

140: “by the same research team” is redundant and can be removed.

 

170-172: The following reads better: “As a medium-scale low-density city, Harbin contains buildings of similar height as the trees along the roads, and therefore the thermal environment of the street canyon is more affected by street trees compared to high-density cities.”

 

172-175: To highlight that this is the problem statement you can start this sentence better with: “There is however still a lack…” or “A remaining question is whether there is a difference…”

 

184: There is only one study area, so change the subtitle in “3.1 Study area”.

 

190: Add the source [33] at the end of the sentence.

 

194: Remove “clusters” because this wasn’t explained yet and write “canyons” instead of “canyon”.

 

196: Use “m” instead of “meters”.

 

197: Here you use “height-street with ratio” and in Table 1 you use “street aspect ratio”. Use instead one of both terms throughout the manuscript.

 

Fig. 1: Please add an additional map to show where the Beixu District is located within Harbin. Reduce the whitespace between the map and the name of each map and add some more whitespace between the title of the above map and the map under it. Enlarge the text labels on the maps (1c and 1d) because they are too small to read. Add scale bars to all maps and add the year or date of the Baidu satellite map. Improve the title of Figure 1, mention for example that this figure describes the location of Harbin and the structure of the study area.

 

Fig. 2: This information is not essential to me to understand the manuscript, so this can maybe be moved to supplementary material in order to make the manuscript more concise. It would be an added value if the location where the panoramic photos are taken is indicated on one of the maps in Figure 1.

 

Table 1: It would be good to add 1 and 2 to NE-SW since you use these abbreviations further in the manuscript. It would be an added value if the average building height is added to the table.

 

Fig. 3: What does “selection of measured range” stand for? This is currently not explained in the main text and is not clear from the figure.

 

207: Mention that you established three scenarios.

 

214 and Fig. 3: Replace “park and green open space” by “park with green open space” since the green open space is located within the park.

 

215: Add “as a final result” between “and” and “proposals” to indicate this is the aim of the paper.

 

217: “The” instead of “he”.

 

219: Add reference [10] additionally to [35].

 

226: Be more specific on which plant canopy image analyzer that was used so your work is reproducible. Does the software have a name?

 

228: Can you refer to a publication that used the same MATLAB code? Or can you include the MATLAB code that you created yourself in the supplementary material? Or can you add it into a repository and share the link in the manuscript? This would be very beneficial for researchers that want to apply the same method.

 

229: These formulas can be placed in the supplementary material since they are commonly used and you refer to another source where they can be found and where they are explained.

 

231: Mention that you used the Albero plant library in ENVI-met.

 

232: Suggestion to use “The background meteorological …” instead of “The meteorological …”

 

233: For which location did you extract the data? The airport? Which variables did you download (temperature, relative humidity and wind speed?)? Please add this fundamental information to the text.

 

234: Mention why both the background meteorological data and the additional on site measurements are needed. 

 

235: Use the primary source that introduced this formula and the primary source that used this formula in this context instead of a source that applies the formula based on previous work.

 

235-237: Is the wind speed data downloaded from weatherspark.com not at 10 m height already? So then no transformation from 2 m to 10 m is needed.

 

Fig. 4: Add which side is 660 m and which side is 750 m, this will also give an indication of the scale. Mention the source and the date of the satellite map.

 

Fig. 5: Mention the name of the software.

 

Fig. 6:  I suggest moving this figure to the supplementary material.

 

Table 3: Please add the temporal frequency e.g. hourly. It should be “wind speed at 10 m height” instead of “wind speed in 10m height”. Is the roughness length 0.01 because you took data from the weather station at the airport?

 

248: Mention the name of the measuring device: Testo-430.

 

Fig. 9: This figure is very similar to Figure 1c, so this figure can be removed to make the text more concise.

 

259-262: This explanation comes a bit too late in the text, please move it to the research framework section.

 

263: How did you obtain the coordinates of the 79 points? Which device did you use?

 

268: Remove “pre-stored”. Please refer to a scientific article where this code was used before, add the Python code in the supplementary material or add a link where the code can be found.

 

271: Please add a reference.

 

272: Do you mean the “Cityscapes Dataset” instead of the “Cityspace Dataset”? Please add a reference e.g. Cordts et al., 2016.

Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., ... & Schiele, B. (2016). The cityscapes dataset for semantic urban scene understanding. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3213-3223).

 

Fig. 10: This figure is taken from reference [39] and does not add any additional value since you already refer to this source in the text, so please remove this figure to make the text more concise.

 

275-280: This paragraph can be shortened by just mentioning the used techniques. You do not have to explain the techniques in detail with all the formulas since you mention the sources where the used method is described. You can move the formulas to the supplementary material to shorten the manuscript.

 

281: Please refer to a scientific article where this code was used before, add the Python code in the supplementary material or add a link where the code can be found.

 

282-283: Table 5 shows some examples but now it reads like you show everything. Modify the sentence therefore as follows: “Some examples of the generated recognition results…”

 

Table 5: This is a very nice and clear example.

 

287: Remove “simulation”.

 

291-293: The figure to prove this statement is missing. Add therefore also the correlation analysis of MRT and mention in the text that you use MRT as the thermal environment index because there are multiple ones that can be used. Can you proceed if this result verification of MRT is really bad? Can we trust the model? I suggest adding also the correlation analysis for wind speed since this might explain why the MRT of the measurement deviates from the modeled MRT. If the correlation analysis of wind speed shows good agreement then the differences are likely to originate from a difference in the radiation, this could be due to the simplifications that are made in the model.

 

295-296: This sentence is difficult to understand, therefore this suggestion: “In the model, the streetscape can be adjusted by adjusting the spacing and number of street trees.”

 

303-304: Please visualize the measuring points on figure 1c or figure 12. Do you mean the same thing with the first and last part of this sentence? I do not understand the difference between those two locations, can you clarify this?

 

Table 6: It would be beneficial to add a column in Table 6 with the number of trees in the study area for each scenario.

 

Fig. 12: The legend is too small to read and please add a scale bar to make the extent of the study area clear.

 

310: Suggestion to change the subtitle into: “4.1. Spatial distribution and types of streetscape interfaces”.

 

313: Mention how you distinguished three types of streetscapes before you start introducing them.

 

314-315: Suggestion: “The BVF and SVF value of such streets is relatively low. This TVF streetscape type is present in the NW-SE, N-N-S and NE-SW-1 streets.”

 

Fig. 13: The text of the legends and scale bars is too small to read, please enlarge this.

 

334: A cooling of 0.64% is not relevant, so please remove this.

 

338: You mention that the ventilation has an effect, which is plausible, but you do not prove this. Can you add a figure or data to prove this?

 

341: Here you can maybe also mention (after checking the data) that there is still enough ventilation due to the wider street.

 

342-346: Write this paragraph in a more concise way.

 

Fig. 14 & 15: These figures can be merged into one figure, by indicating at the top of the figure which streets belong to which streetscape type. Both figures are missing a title on the y-axis and it is confusing that a reduction in temperature is positive and an increase in temperature is negative. Which data is shown on these plots? Data of one point in the street canyon at several time intervals or an average value over the simulation period for each location in the street canyon of that street? Please mention this in the text and in the title of the figure.

 

352: Remove “significant”.

 

354: Temperature data is always continuous, except for when you specify temperature classes, so you cannot speak about a discrete data distribution here. The data is stretched out, you have long tails indicating a large spread in the data of the point you took into account.

 

369: Remove “significant”.

 

375-376: Note that the differences are in general small and most of them are not noticeable (<0.1 °C).

 

386 & 388: Remove the percentages that give the relative change since this is misleading. 57.1% and 66.7% seems to be a large decrease but this is due to the very small values.

 

389-390: This is a nice result, but it can be more framed in practical guidelines e.g. when is a TVF of 0.16 commonly reached and what should be done for streets where this is not reached e.g. how many trees should be added of the type of trees you had in your study area.

 

Fig. 17: Here it would also be easier to interpret when you plot SVF and TVF against Ta of each scenario, the CG and the three EGs (same remark for  Figure 21 and Figure 25). Why is BVF not included? How did you retrieve the inflection point? How did you define there is a significant inflection point? Why did you not use a quadratic fit?

 

396: Mention under which scenario the maximum is reached.

 

397: Replace “cooling” by “humidification”.

 

398: Please modify “best effect” into “largest effect” since higher RH does not mean that the thermal comfort will improve, it is rather the opposite. The higher the RH, the more difficult it is to relieve the human body from heat since it is more difficult to sweat. You did note once throughout the manuscript that RH strongly depends on the prevailing temperature, it would be good to mention it here as well since it is difficult to say something about the thermal environment, solely based on RH.

 

400: Modify this sentence as follows: “The NE-SW-1 street had the largest spread and in RH change during the day due to large fluctuations in RH.”

 

406: In which way does the setback distance of the street canyon buildings affect the humidication effect of vegetation? Please explain this.

 

Fig. 18 & 19: Same remarks as for Figures 14 and 15.

 

412: There seems to be a higher value than 2.51% in Figure 19.

 

414-415: Very good remark you made here. Maybe it is interesting as well to present the results based on the change in vegetation for each scenario since in some streets more vegetation will be added compared to other streets because some streets already contain some vegetation in the current situation.

 

423: Remove “can improve the” and replace “of” by “the”.

 

425-428: I do not understand this sentence, there is also a typo “benign” and I think “decreases” should be “increases”?

 

432: Replace “discrete” by “spread out”.

435-441: Mention here as well there is a relationship between temperature and relative humidity.

 

442: Mention mean radiant temperature (MRT) earlier in the text. Explain in the introduction what it stands for.

 

446: What do you mean with “equivalent”? It is not clear to me what you are describing in this sentence.

 

449-451: The characteristics of the buildings are not changed in the different experiments, so how can they have an impact? There is also a word missing after “less”. Do you mean “less effective”?

 

451: Replace “average radiation temperature” by “MRT”.

 

452-454: The fact that EG-15 is the better scenario is probably due to the ventilation or radiation, since for both T and RH the EG-12 scenario is the scenario which changes the variables the most. It would therefore be interesting to have a look at the changes in wind speed as well.

 

Fig. 22 & 23: Same remarks as for Figures 14 and 15.

 

460: Express the maximum decrease in °C instead of %.

 

462: Formulate this sentence in a more scientifically sound way.

 

471: Express the maximum decrease in °C instead of %.

 

474: What do you mean with “not apparent” is there another word that is the opposite so you do not have to use “not”?

 

478: Why is this the case for these interactions? Can you please elaborate on this?

 

485: Modify as follows: “To reduce the MRT in an cost-effective way, the deltaSVF and deltaTVF values should be increased to at least …”

 

487: Change into: “5. Discussion”.

 

498: Do you mean “most cost-effective way” instead of “and be economical”?

 

496-512: How come that 15 m and 18 m are the most suited solutions here and not 12 m? What is the mechanism behind this? Please include also the data of these points to prove this.

 

Table 8: This table gives a very nice overview. Change the precision however to 0.1 °C and 0.1% as mentioned before. Some text disappeared at the bottom of the third column of the table.

 

520: Is there a reason why you took the N-N-S street? Please mention why you studied this particular street. Which scenario did you take to calculate the difference? Please add this to the text.

 

525-528: How can you achieve this in practice? Please add a sentence to the text on how this could be achieved in practice.

 

533: Can you explain way deltaTVF should be lower than -0.07 and deltaSVF larger than 0.04?

 

536-537: This sentence is redundant to me, so it can be removed.

 

550: -0.16 should be -0.17 or it should be modified in Figure 27a.

 

560: Mention in a new sentence immediately after the numbers 0.33 and 0.35. that these correlations are too weak to make strong conclusions.

 

560-561: This indicates that you should find a relationship with the BVF as well, so please plot this graph as well to see whether this is correct.

 

562-563: This is an important conclusion, nice to read.

 

593: The decrease was limited.

 

595: Replace “growth level of streets” by “increase was highest for” and remove “was the highest” in line 596.

 

595-596: Use °C for MRT instead of % since relative changes can be tricky.

 

600: Replace “economy” by “cost”.

 

607-609: You cannot conclude this based on one street/graph. If you want to include this then you should prove this is the case as well for other streets with the same streetscape type.

 

612: A general conclusion is missing. What does all of this mean? What would your general advice be?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work aims to investigate the impact of streetscape interface measurement on the thermal environment via using ENVI-met simulation. Three design variables, including the sky view factor, building view factor, and tree view factor, were studied. The model was validated and provided acceptable results compared with the field measurement. The paper is relevant to the journal themes and well organized; however, more detail in methodology requires explanation and clarification. Therefore, I suggest a better paper based on the following comments:

 Abstract:

The definition of streetscape interface measurement should provide. The studied variables (SVF, TVF, and BVF) should be mentioned in the abstract.

Introduction:

·       The introduction and literature reviews provide sufficient background. However, it should provide more critical reasons why the study focuses on the SVF, TVF, and BVF.

·       Abbreviations/ acronyms should be defined the first time.

Methodology:

·       In Table 1, what is the meaning of Layer 7 and Layer 6-7? Please clarify.

·       In Table 1, where are the locations of the space type A, B, and C in the studied area? And please provide the reason why these three locations were selected instead of other locations.

·       Figure 3, What is D=12m? EG-12? And CG? These terms need an explanation.

·       What is the condition of tree plantings and planting distance for the control group?

·       The authors should do more literature reviews about the grid setting and grid sizes for ENVI-met modeling. Why was the simulation performed with the grid number 220x220x25 and the grid resolution of 3mx3mx3m? Is this setting acceptable for the simulation for the neighborhood scale?

·       Besides the walking routes, the measurement points at 30 m intervals should be presented on the map.

·       Line226, ‘A plant canopy image analyzer photographed the vegetation in the model’ what did the model the study use? Please clarify.

·       Table 2, please provide more detail; what are the building and paving materials?

·       Equations (4), (5), and (5), please provide more information about Cy, Cx, and n used in the equations.

Results:

·       Table 6, what is the current spacing of vegetation in the control group?

·       Figure 13, the text and color shade in the figure is difficult to read.

·       The distribution range of ΔT in Figure 14 and Figure 15 should be the same for easier comparison. As similar to Figures 18 and 19

Discussion:

·       The quadratic correlation characteristic of ‘first fall and then rise’ needs more discussion. What causes that effect? How do the findings show similarities or differences to other works?

 

 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear authors, The manuscript underwent a significant improvement compared to the previous version and most of the remarks were addressed or explained. There are however still some small improvements of the current manuscript version possible. The nuances that lacked in the previous version have been addressed and make that this version of the manuscript can be published after modifying the following items: 21: SVF, TVF, and BVF, write the full meaning of these abbreviations since it is the first time you mention them in the abstract and one might not be familiar with it. 49: Add reference [2] 53: “The urban center area of Harbin is the historical foundation of the city.” Add “the”. 59: Replace “MRT” by “mean radiant temperature (MRT)”. 197-198: “In the study area, the underlying surface of the street canyons are of the same material, with asphalt roadway and red brick pavement for pedestrians.” This should be: “In the study area, the underlying surface of the street canyons is of the same material, with asphalt roadway and red brick pavement for pedestrians.” or: “In the study area, the underlying surfaces of the street canyons are of the same material, with asphalt roadway and red brick pavement for pedestrians.” Figure 1 was nicely improved, but please keep the reference year 1927 of the old map 1b in its description. Please add a legend that explains the newly added elements in figure 1c and describe in the figure’s title that the walking route and measurement points are indicated. Further, the street names are still too small to read in Figure 1c and d. Maybe just enlarge all the rectangles of figure 1 a bit? 210: Remove “three” 213: Replace “;” by “.” 216: Replace “...were simulated and…” by “...were simulated, using MRT and…” 217: Replace “Establish” by “Therefore” 222-225: Change “Optimization strategies of specific street canyon types will be discussed. Established spatial optimization strategy models for three specific street canyon types (the inner space of street canyon street intersection space, park and green open space), a comparison of the results of strategy simulation.” into: “Optimization strategies of specific street canyon types will be discussed by establishing spatial optimization strategy models for three specific street canyon types (the inner space of street canyon street intersection space, park and 224 green open space) and comparing the results of strategy simulation.” 226-227: This sentence could benefit from a reference. 228: Add “are provided” at the end of this sentence since the verb is currently missing. 238: Modify into: Most studies adopt a grid resolution of 1.5m-5m. 249: Refer at the end of the sentence that the reader can find the MATLAB code in the supplementary material 1, e.g. “(see supplementary material 1 for MATLAB code) [18].” 252: Add at the end of the sentence: “(see supplementary material 4)” 258-260: Please check once more whether the wind speed data is downloaded for 2 m height and not 10 m height as it is the case for me when I download data from weatherspark.com. If you download data at 10 m height then the conversion is not needed. 266: Replace “planting” by “plant” 273: Replace “instrument” by “instruments” 275-278: Replace: “…instruments at a consistent height of 1.5m, and walking at a constant speed at an average pace Here the uniform speed refers to the perceived uniform speed, rather than the absolute uniform speed in the physical sense (Figure 6).” by: “…instruments at a height of 1.5 m (Figure 6), and walking at a more or less constant speed with an average pace.” 290-291: Replace: “We obtained the coordinates of 79 measurement points through ArcGIS according to the measured positions, and the use of the Baidu Street View Image Open Platform to obtain the panoramic image ID” by “The coordinates of the 79 measurement points were used to obtain the panoramic image ID from the Baidu Street View Image Open Platform” 297: Add after “location”: “(see Python code in supplementary material 2)” 299: Replace “Cityspaces” by “Cityscapes” 301-302: Replace : “Project a hemispherical environment (cylindrical projection) onto a circular plane (azimuthal projection), producing a fisheye image.” by “To produce a fisheye image, a hemispherical environment (cylindrical projection) is projected onto a circular plane (azimuthal projection).” 310: Add after “BVF”: “(see Python code in supplementary material 2)” 319: Suggestion to remove “specific” 321-323: Replace: “The correlation between the measured and the simulated wind speed is not strong, which is because the measurement method leads to a large instantaneous change in wind speed.” by: “The limited correlation between the measured and the simulated wind speed, indicates a large deviation between them, which is due to the large instantaneous change in wind speeds of the measurement method.” Figure 8: The legends are still quite small. Try to maximize the figures over the full page in order to improve the size of the legends. 341: Replace: “and used the ArcToolBox (Inverse Distance Weighted)” by “and an inverse distance weighted interpolation was used ” 342: Remove: “-10” 345: Refer to a source that describes the typology classification method. This method must be scientifically described somewhere before you apply it. 346: Add “(figure 10a)” at the end of the sentence. 353: Add “(figure 10b)” at the end of the sentence. 358: Add “(figure 10c)” at the end of the sentence. Figure 9: The text of the legends and scale bars is too small to read. 374-375: Please add a plot (in the supplementary material) that shows that the ventilation pattern changes between the CG and optimized scene to prove that this is the cause for the slight temperature increase. It does not matter whether the wind environment is improved or not, the figure should just show the difference in air flow to prove that your statement is correct. 383: Replace: “than NW-SE street canyon” by: “than in the NW-SE street canyon” 391-392: Replace: “...and the data distribution was discrete, which showed that the air temperature decreased at different times and the whole day was very different.” by: “...and the large range in temperature differences, indicates that there is a large variation during the day in the change of air temperature when trees are added in the NE-SW-2 street.” 398-402: Remove because the behavior of NE-SW-2 and NE-SW-1 is very similar to me. If you want to prove that their behavior isn’t similar then you should do this with a statistical test. You do not include this so I would omit this paragraph. 414: Please add here shortly a sentence on the following comment you made in your reply: “According to the urban road design code, the safe visual distance must be considered when planting trees at street intersections, that is, enough space should be reserved at road intersections to ensure driving sight. Therefore, as a street intersection space, the number of trees is less than that of general streets, which makes the thermal environment change limited.” 417-418: Replace “...NE-SW-2 street is the most discrete, indicating…” by: “...NE-SW-2 street has the largest range, indicating…” 421-428: This paragraph is still confusing to me. How can the change in both SVF and TVF increase at the same time? Is this because you let BVF decrease? Or is this because you defined delta SVF as SVFcg - SVFeg and TVF as the opposite TVFeg-TVFcg? If so please use the same definition. I would expect an inverse curve around the y-axis since SVF and TVF are related to each other (SVF+TVF+BVF=1). I suppose the points represent the value of each intersection? But how come that there are more points than intersections? Please explain how you constructed these graphs. The paragraph and figure 14 is about the difference in SVF and TVF, so why do you speak about values for SVF and TVF instead of changes in SVF and TVF in the last sentence? Please make the necessary changes to clarify this paragraph and figure 14. Figure 14: Not including BVF because no relationship was found, is hiding some of the results. Please describe that no correlation was found for BVF, with the hypothesis of what might be causing this. 429-430: Please mention in the beginning of this chapter that RH is strongly correlated with temperature and depends thus on the prevailing temperature. 432: Remove: “was obtained” 437: Replace: “improve” by: “affect”. As mentioned in my previous report, higher humidity is not necessarily an improvement. 441: Replace: “The relative humidity distribution…” by: “The change in relative humidity distribution…” or: “The distribution of the change in relative humidity…” 443-444: Replace: “which promote the evaporation of humidity and make the increase of humidity small.” with: “which promotes the evaporation of humidity and reduces the increase in humidity.” 447: Remove: “was obtained” Figure 18: Same remarks as for figure 14. 488-490: The properties of the buildings were not changed between the different experiments so they cannot play a role in the change of MRT. Replace therefore “This is because the building interface of N-N-S street is discontinuous. Besides, the building is far away from the street trees, so the double-row street trees cannot provide shadow together with the building. Hence, the reduced MRT value is less effective.” by “This is because the added street trees of N-N-S street are placed further from each other because of the width of the street, so they do not provide as much shade as the street trees of the narrower streets NE-SW-1 and NW-SE, which results in a less effective reduction of the MRT.” 494-495: Replace: “The EG-12 scene of NW-SE and N-N-S streets reduced the maximum MRT, similar data dispersion.” by: “For the NW-SE and N-N-S streets, the EG-12 scenario reduced the MRT the most.” 500- 501: Replace: “The data distribution of the whole day is also more discrete, indicating that the variation of the MRT throughout the day is relatively large.” by “The wide spread data distribution indicates that the variation of the MRT throughout the day is relatively large.” 513-514: Replace: “..., the variation of average radiation temperature of NE-SW-1 is unapparent (ΔMRT=4.32℃, 5.15℃, 2.51℃),...” by: “..., followed by a change in average MRT of NE-SW-1 (ΔMRT=4.32℃, 5.15℃, 2.51℃),...” 519: Replace: “...is the most discrete,...” by: “...has the largest range,...”. 520-522: Replace: “The difference EG-12 at the intersection of NW-SE, NE-SW-2 and N-E-W streets provides more shade than other streets.” by: “This is probably due to the fact that the intersections of NW-SE, NE-SW-2 and N-E-W streets provide more shade than the other intersections under the EG-12 scenario.” 526: Replace: “;” by: “.” Figure 22: Same remarks as for figures 14 and 18. Here I would expect that a change towards a smaller SVF (more trees) leads towards a lower MRT or an improvement in the MRT and thus higher delta MRT, as described in the text 502-504 but this is not the case on the figure so I think deltaSVF is defined here as SVFeg-SVFcg, while cg-eg was used anywhere else. 530: Please refer in this section to the supplementary material 5 when needed. 534-537: Please add one or multiple sources that describe the same mechanisms. 571: Replace: “Because adding street trees on N-N-S streets can significantly increase TVF and significantly decrease SVF, taking N-N-S streets as an example.” by: “The N-N-S street is taken as an example because adding street trees in the N-N-S street can significantly increase TVF and significantly decrease SVF compared to the other TVF-led streetscape streets.” 589-590: Replace “radiation temperature” with “MRT” Figures 11, 13, 15, 17, 19 and 21: Please enlarge the notes that were added to these figures because now they are rather small to read. In the reply on the previous changes the following was mentioned: “Because our SVF, TVF and BVF are acquired from street view images, it is difficult to plant street trees and calibrate VFs with 0.01 accuracy in reality. Therefore, the significance of such accurate critical values in this paper is to accurately discuss the rules. There is a big difference between design practice and theory. In design practice, street trees are generally equidistant, and simulation results with different spacing have more reference value and practical guidance for design practice.” I agree that there is a large difference between design practice and theory, this is however not stressed in the current version of the manuscript, so please add this critical note at least in the discussion and conclusion. Tell the readers that there are e.g. theoretically reductions in temperature but that one cannot measure them since the changes are of such an order of magnitude that it cannot be measured.

Author Response

Responses to REVIEWER #1 (Round 2)

 

Manuscript title:Evaluation strategies on the thermal environmental effectiveness of street canyon clusters: A case study of Harbin, China

 

Manuscript number: sustainability-1866592

 

Author name: Guanghao Li, Qingqing Cheng, Changhong Zhan, Ken P Yocom

 

 

 

The authors would like to thank the reviewer for the explicit revision comments on our manuscript. The revised paragraphs based on REVIEWER #1 are labeled in BLUE.

All the comments have been revised according to the reviewer's opinion. The following are comments that need particular clarification:

 

1.	216: Replace “...were simulated and…” by “...were simulated, using MRT and…”	Thank you for your comments. The method is described here, and such a change would break the logical integrity. So we think it's better to keep the original sentence.
2.	226-227: This sentence could benefit from a reference.	Thank you for your comments. This is the logic derived from my research conclusions, not from references.
3.	421-428: This paragraph is still confusing to me. How can the change in both SVF and TVF increase at the same time? Is this because you let BVF decrease? Or is this because you defined delta SVF as SVFcg - SVFeg and TVF as the opposite TVFeg-TVFcg? If so please use the same definition. I would expect an inverse curve around the y-axis since SVF and TVF are related to each other (SVF+TVF+BVF=1). I suppose the points represent the value of each intersection? But how come that there are more points than intersections? Please explain how you constructed these graphs. The paragraph and figure 14 is about the difference in SVF and TVF, so why do you speak about values for SVF and TVF instead of changes in SVF and TVF in the last sentence? Please make the necessary changes to clarify this paragraph and figure 14.	Thank you for your comments. First of all, this does not mean change in both SVF and TVF increase at the same time, but fits all data points and finds that the current situation presents such a rule, which is not my artificial regulation rule. Because δSVF and δTVF will not increase at the same time, there will be a critical point of tree planting in theory, so that the changes of SVF and TVF can meet the temperature optimal result at the same time. This is a promising finding, but more research is needed to determine how to apply it. Secondly, regarding BVF, although SVF and TVF are related to each other (SVF+TVF+BVF=1), ΔSVF+ΔTVF + ΔBVF ≠1, so there is no curve you mentioned. Finally, an intersection has multiple value points instead of one. Please refer to Figure 1c.
4.	Figure 14: Not including BVF because no relationship was found, is hiding some of the results. Please describe that no correlation was found for BVF, with the hypothesis of what might be causing this.	Thank you for your comments. No correlation of BVF was found, which may be because street trees are generally planted in front of buildings, and vegetation changes shade buildings, making it difficult to determine the change value of BVF. Moreover, because the cost of urban building renewal is much higher than that of vegetation renewal, the proposed BVF correlation does not have a direct guiding significance for landscape vegetation renewal.
5.	Figure 18: Same remarks as for figure 14. Figure 22: Same remarks as for figures 14 and 18. Here I would expect that a change towards a smaller SVF (more trees) leads towards a lower MRT or an improvement in the MRT and thus higher delta MRT, as described in the text 502-504 but this is not the case on the figure so I think deltaSVF is defined here as SVFeg-SVFcg, while cg-eg was used anywhere else.	Thank you for your comments. Figure 18, Figure 22: Same response as for figures14.
6.	534-537: Please add one or multiple sources that describe the same mechanisms.	Thank you for your comments. This statement is a model description of two types of scenarios, not from references.