The Effects of Microclimate Parameters on Outdoor Thermal Sensation in Severe Cold Cities

Round  1

Reviewer 1 Report

The paper deals with the thermal sensations for outdoor in severe cold cities in China. In the paper, the authors present the results of the questionnaire where several people gave a vote to their thermal sensations about temperature, solar irradiance, relative humidity, wind, and so on. At the same time, the authors recorded the climate of the day, aiming at finding a correlation between the results of the questionnaire and the data acquired.

In my opinion, it is not clear what the aim of the paper is. There is also confusion, among the authors, about the word “microclimate”. Microclimate is a word that means climate referred to a close space, but in the outdoor, the use of this word is misleading. Thermal sensations referred to microclimate have, in fact, already been studied since Fanger studies as the authors have written in the introduction. Fanger and the following researchers after him have studied the indoor microclimate and the thermal sensation to seek the best values for temperature, relative humidity, air speed, mean radiant temperature, to ensure users’ comfort. In this case, what is the novelty of the paper? What is the aim of the paper? Is it just a correlation between the results of a questionnaire and external data acquisitions? How the research can be useful for the scientific community? In a sense, the correlation between comfort sensation and indoor microclimate have a sense as HVAC system can be controlled to maintain specific indoor condition. In outdoor case, instead, what is the relation among thermal sensation, climate and engineering/science?

Point 1: In my opinion, it is not clear what the aim of the paper is. There is also confusion, among the authors, about the word “microclimate”. Microclimate is a word that means climate referred to a close space, but in the outdoor, the use of this word is misleading. Thermal sensations referred to microclimate have, in fact, already been studied since Fanger studies as the authors have written in the introduction. Fanger and the following researchers after him have studied the indoor microclimate and the thermal sensation to seek the best values for temperature, relative humidity, air speed, mean radiant temperature, to ensure users’ comfort. In this case, what is the novelty of the paper? What is the aim of the paper? Is it just a correlation between the results of a questionnaire and external data acquisitions? How the research can be useful for the scientific community? In a sense, the correlation between comfort sensation and indoor microclimate have a sense as HVAC system can be controlled to maintain specific indoor condition. In outdoor case, instead, what is the relation among thermal sensation, climate and engineering/science?

Response 1: The authors are extremely grateful to receive such positive feedback and have made serious revisions to resolve the problems raised by the reviewer. The study is very closely related to the subject of urban and rural planning. Because that a more comfortable microclimate is often obtained by adjusting the site scale in urban planning and design. We added an introduction to the practical significance of the study. And the whole manuscript has been heavily revised. Thank you again for your valuable advice.

The authors conducted a survey in different seasons in Harbin and tried to do a regression between the meteorological parameters and the thermal sensation vote. It is valuable to do this kind of research in the area with serve cold climate. However, many problems are found in this manuscript. The academic writing skill are poor. Major technical problems are found in the Section of Results. Here are my comments as follows:

1). Many typos are found in the manuscript. Please check the manuscript carefully before submitting it. Several expressions were found to be Chinglish and it should be avoided in the academic writing.

2). Method Section, past tense is suggested when describing the research method part.

3). Section 2.2. average air temperature should be a value, not a range.

3). Section 3.1. Mentioned in the previous part that the serve cold season was defined as those having average air temperature below -10 °C. Why the minimum air temperature was 16.5 °C in serve cold season in Part 3.1? 

4). How to determine the difference between sitting while rest or chatting and chess or card entertainment regarding the period of one activity? For example, chatting can also last long time. Also, if the authors are intended to analyse what kind of activity is suitable for the outdoor environment in Harbin, the observation data should be analysed regarding to different seasons.

5). Line 163. Using the word “activist” may be not appropriate. How can we know the difference between activist and normal user from the current observation?

6). Fig. 1(b). The purpose of this figure is unclear. The number of survey samples obtained in different seasons were not the same. In this case, comparing percentage of ASV in different season can lead to wrong conclusion.

7). It is not appropriate that the author lead to a conclusion of “people are only psychologically comfortable but not physically comfortable”? Many researchers have shown TSV=0 might not be the most comfortable situation, please do more literature review before drafting a paper.

8). Section 3.4. Linear regression might not be the best way to do the regression. The authors have figured out that the distribution of the TSV follow normal distribution, so the author should have noticed the limited samples in the two ends of the vote. In other words, it is very likely that this regression can only present the relation between the meteorological parameters and the vote between “-1” and “+1”.

9). Section 3.4. The independent variables used in the regression have not been standardized, the units for the variables are different. It is not suitable to compare the relative impact of the independent variables have to the ASV, because they are incomparable under different units.

10). Minor problems, the authors are suggested to review more recently published high-quality journal papers in the past 5 years, especially for the northern part of China that have the similar climate.

Point 1: Many typos are found in the manuscript. Please check the manuscript carefully before submitting it. Several expressions were found to be Chinglish and it should be avoided in the academic writing. 


Response 1: The authors are extremely grateful to receive such positive feedback and have made serious revisions to resolve the problems raised by the reviewer. To resolve language issues, the whole manuscript has been heavily revised.

Point 2: Method Section, past tense is suggested when describing the research method part.

Response 2: We have modified the research method part, including to use the past tense.

Please see the section 2.

Point 3: Section 2.2 average air temperature should be a value, not a range. Section 3.1. Mentioned in the previous part that the serve cold season was defined as those having average air temperature below -10 °C. Why the minimum air temperature was 16.5 °C in serve cold season in Part 3.1?

Response 3: We change the “average temperature” to “temperature”, and the minimum air temperature was -16.5 °C.

The revised parts are highlighted in yellow in: section 3.1.

Point 4: How to determine the difference between sitting while rest or chatting and chess or card entertainment regarding the period of one activity? For example, chatting can also last long time. Also, if the authors are intended to analyse what kind of activity is suitable for the outdoor environment in Harbin, the observation data should be analysed regarding to different seasons.

Response 4: In the current document, the observation data of the activity type is not analysed. So we have made appropriate deletions to the activity type.

Please see the section 3.2

Point 5: Line 163. Using the word “activist” may be not appropriate. How can we know the difference between activist and normal user from the current observation?

Response 5: It is not appropriate to use the word “Activists”. In fact, "activist" is normal user from the current observation. To resolve language issues, the whole manuscript has been heavily revised.

Point 6: Fig. 1(b). The purpose of this figure is unclear. The number of survey samples obtained in different seasons were not the same. In this case, comparing percentage of ASV in different season can lead to wrong conclusion.

Response 6: Fig. 1 (b) shows the approximate distribution of the votes. We have modified the picture so that it does not show seasons. We have rechecked the other figures in the article and modified the pictures that do not meet requirements, such as Fig. 2(a) and (b).

Point 7: It is not appropriate that the author lead to a conclusion of “people are only psychologically comfortable but not physically comfortable”? Many researchers have shown TSV=0 might not be the most comfortable situation, please do more literature review before drafting a paper.

Response 7: Thank you for your suggestion. We deleted this inappropriate conclusion and modified it.

The revised parts are highlighted in yellow in: section 3.2 and section 5.

Point 8: Section 3.4. Linear regression might not be the best way to do the regression. The authors have figured out that the distribution of the TSV follow normal distribution, so the author should have noticed the limited samples in the two ends of the vote. In other words, it is very likely that this regression can only present the relation between the meteorological parameters and the vote between “-1” and “+1”

Response 8: I delete the description of normal distribution. In the study, we found that ATSV and OCV are normally distributed. Due to limited space, it is not introduced in detail in the article.

The revised parts are highlighted in yellow in: section 5.

Point 9: Section 3.4. The independent variables used in the regression have not been standardized, the units for the variables are different. It is not suitable to compare the relative impact of the independent variables have to the ASV, because they are incomparable under different units.

Response 9: To compare the relative impact of independent variables on ATSV, we standardize the coefficients of the equations (1-4), as shown in Table 6.

Point 10: Minor problems, the authors are suggested to review more recently published high-quality journal papers in the past 5 years, especially for the northern part of China that have the similar climate.

Response 10: We have reviewed the relevant papers in recent years, and modified the introduction. Thank you again for your valuable advice.

The revised parts are highlighted in yellow in: section 1.

Reviewer 3 Report

Dear Authors

I read with great interest your research.

The manuscript, unfortunately, is characterized by several lacks which reduces its efficacy. This is the reason why I would suggest major revisions.

My main concerns are:

1. You should specify the practical implication of their findings. Otherwise most of all conclusions risk to seem “naif”.

2. Subjective investigations have not been carried out consistently with International Standards in the field (ISO 10551).

3. You measured only solar radiation and did not account for the IR contribution related to the presence of surfaces (do they know the mean radiant temperature?). In addition, they did not measure clothing insulation and they made use of an anemometer with a very poor accuracy (±1m/s). So, it is very difficult trusting in their multivariable regressions (especially in the presence of low air speed values).

4. You did not correlate their ATSV votes with most common metrics used for assessing outdoor climates (eg. UTCI, PET) or cold heat stress (e.g. WBGT, IREQ).

5.In cold conditions maybe a further question on local thermal sensation (local cooling) is necessary

I am confident that you will consider my observations merely as suggestions addressed to the best result. You will find other details in the marked pdf file.

Best regards

Comments for author File: https://susy.mdpi.com/user/review/displayFile/6612272/NBbFijJ3?file=review&report=3422967

Point 1: You should specify the practical implication of their findings. Otherwise most of all conclusions risk to seem “naif”. 


Response 1: The authors are extremely grateful to receive such positive feedback and have made serious revisions to resolve the problems raised by the reviewer. We added an introduction to the practical significance of the study.

The revised parts are highlighted in yellow in: section 1. Introduction.

Point 2: Subjective investigations have not been carried out consistently with International Standards in the field (ISO 10551)

Response 2: We have added a reference to ISO 10551 to the manuscript, and other similar issues have also been revised. The revised parts are shown in section 2.4 and References [22].

Point 3: You measured only solar radiation and did not account for the IR contribution related to the presence of surfaces (do they know the mean radiant temperature?). In addition, they did not measure clothing insulation and they made use of an anemometer with a very poor accuracy (±1m/s). So, it is very difficult trusting in their multivariable regressions (especially in the presence of low air speed values).

Response 3: In the process of data processing, we compared thermal radiation with solar radiation in outdoor. In the open outdoor environment, the solar radiation has the largest proportion of the total thermal radiation during the day, which is supported by References [14]. At the same time, we need interviewees to describe how they feel about solar radiation. Therefore, solar radiation is chosen as one of the parameters.

In this paper, we supplement the introduction of these situations. The revised parts are shown in: section 2.5.

We rechecked the data of the thermal probe and corrected the accuracy of the tool.

The revised parts are shown in Table 3.

Point 4: You did not correlate their ATSV votes with most common metrics used for assessing outdoor climates (eg. UTCI, PET) or cold heat stress (e.g. WBGT, IREQ).

Response 4: To remedy this deficiency, we cite the results of a paper. The gap between NAT and NPET is compared. In the past, we didn't investigate and calculate PET, so we couldn't correlate ATSV and PET in the paper.

The revised parts are highlighted in yellow in: section 4.

Point 5: In cold conditions maybe a further question on local thermal sensation (local cooling) is necessary.

Response 5: Local thermal sensation has a great influence on people's thermal sensation, especially in the cold area. But we haven't investigated it yet. Thank you again for your valuable advice.

Round  2

Reviewer 1 Report

The paper deals with the thermal sensations for outdoor in severe cold cities in China, aiming at finding a correlation between external climate parameters and thermal sensations, which were evaluated through questionnaire.

I really appreciate the efforts of the authors in changing the manuscript and in conducting the 886 questionnaires and analyzing all the data, but I still do not understand the aim of the work. How can these results can be used by other researchers? I think that this work is an end in itself.

The work is really similar to Fanger work, but the aim of Fanger was to find the setpoint of temperature to have optimal human comfort in internal microclimates, which can be controlled though HVAC system. In this work, the authors find some parameters that have some influence in human comfort for people that stay in open spaces. However, here there is no possibility to control “something” to increase human comfort. So, what is the practical use of the Equations 1-4?

Author Response

    These suggestions led us to discover the inadequacies in the article, and we are very grateful to the reviewer. The whole manuscript has been heavily revised.

(1) We supplement the purpose of the research.

      The revised parts are highlighted in yellow in: section 1. Introduction.

     In severe cold cities, researchers are trying to establish a set of criteria to evaluate the quality of the outdoor public space microclimate, and to help designers create a comfortable environment. The purpose of our study is to obtain the relationship between microclimate parameters and thermal sensation, which will be used to guide the evaluation and design of microclimate in urban public space in the future.

     In the outdoor, we can adjust the microclimate by adjusting the aspect ratio of the street, the street orientation, the open degree of the field, the high-rise building and the multi-storey, the greening, the setting of sunshade shed and so on.

(2) We've added some literature.

The revised parts are highlighted in yellow in: section 1. Introduction.

(3) We supplement the purpose of the equation 1-4.

The revised parts are highlighted in yellow in: 3.3.

     In equation 1-4, ATSV can be used as a reference index for the design and evaluation of public space. In the actual design, we can obtain the microclimatic parameters (temperature, wind speed, humidity and solar radiation) by actual measurement or computer simulation, and calculate the value of ATSV. Select appropriate design methods and techniques based on the results for construction or alteration. At present, we are establishing and perfecting this evaluation method as a new subject.  

     In addition, the study considered the impact of behavior on evaluation results. This effect is different from the indoor environment, not just the movement or rest, or what kind of clothing people wear. The subjects were not individuals, but a group of people who were purposefully active. Their state is likely to change and they can make themselves more comfortable by taking the initiative, such as opening the neckline, taking off their coats or gloves, or choosing a more comfortable environment. These factors are closely related to actual engineering projects, and that is valuable to urban planners.

     Thank you again for your help and advice！

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript looks much clear than the first submission. I believe the authors make their efforts on revising it based on our reviewers’ comments last time. Only a few minor problems should be carefully revised in the academic English writing.

1. The word “respondent” may not be professional in thermal comfort survey. The authors are suggested to use “subject” or “volunteer” instead in the whole manuscript.

2. Abstract, the word “poll” should be “survey/questionnaire/vote”.

3. Introduction, more literature can be reviewed when the authors wish to support their points on the design/planning stage, such as some simulated studies on microclimate and outdoor thermal comfort using CFD technique.

Author Response

Response 1: These suggestions led us to discover the inadequacies in the article, and we are very grateful to the reviewer. To resolve language issues, the whole manuscript has been heavily revised.

The revised parts are highlighted in yellow in: section Abstract.

Response 2: The revised parts are highlighted in yellow in: section Abstract.

Response 3: We have supplemented the literature on the study of severe cold regions, and supplement the purpose of the research.

The revised parts are highlighted in yellow in: section 1. Introduction.

Thank you again for your help and advice！

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors

I read with great interest this manuscript devoted to the assessment of the thermal sensation outdoor in severe cold cities. The paper reveals significant modifications with respect to the earlier version but, unfortunately, it is still characterized by several lacks which reduces its efficacy.

I am confident that authors will consider my observations merely as suggestions addressed to the best result.

Best regards.

My main concerns are:

1. Introduction is well structured but there are not enough bibliographic references useful to support the reasoning. In particular(L65-L67) stated that, at the present, there are few studies devoted to cold cities. Well, few is not zero. So,  you are invited to specifies the few available studies.

2. In the methods section there are some issues not well addressed.

·       Above all, why did you measure microclimatic parameters only at one height form the ground)? Why did you  use 1.5 m for a standing person? (usually microclimatic measurements are carried out at 3 heights. 0.10; 1.10, 1.70 m).

·       You measured globe temperature, but they do not assess its correlation with ATSV. This is singular from a physical point of view (see anger’s, Gagge’s and Parsons textbooks). In the presence of hot surfaces IR radiation affects the thermal sensation even under a tree or in shadowed spaces. Given explanations from L173 to L179 are unclear and not robust.

·       You should specify the reason why did not correlate ATSV with main heat and cold stress indices (IREQ/WBGT).

3. Results and discussion. L266-L271. The procedure of “standardization” (maybe normalization) is unclear and not well discussed. The seasonal trends of coefficients in table 6 (as in equations 1-4) exhibit a strange behavior. For instance: coefficient for solar radiation changes from 0.618 (transition season) to 0.528 in hot season (it should increase). The coefficient for wind speed is always negative (how justify the effect of hot air on the thermal sensation in cases such as air temperature is greater than skin temperature?). This is also for humidity: apart from the sign, under severe cold conditions humidity weights on the thermal sensation as in summer. This is quite no sense being the effect of latent heat flows in winter quite negligible.

Author Response

Response 1: These suggestions led us to discover the inadequacies in the article, and we are very grateful to the reviewer. The whole manuscript has been heavily revised. We have supplemented the literature on the study of severe cold regions, and supplement the purpose of the research.

The revised parts are highlighted in yellow in: section 1. Introduction.

Response 2: When I set the height of measurement, I referred to the relevant literature and found that the height measured in outdoor environment is generally in the range of 1.10m~1.70m, such as [15] (1.10m) [16,25] (1.50m) [17] (1.70m). Therefore, this study referred to the literature of the same region [16], and set the height of measurement to 1.5 m.

The revised parts are highlighted in yellow in: L137.

As you are concerned, the selection of microclimatic parameters is indeed one of the key aspects, especially with regard to thermal radiation. Thermal radiation contains many items, including solar radiation and IR radiation. The tool is JTR05, which has a spectral range of 0.3-3.2μm, solar radiation (0.15-4 μ m),) near infrared radiation (0.782.5 μ m) and part of the mid-infrared radiation line.

At the beginning of the experiment, the first consideration is to collect the radiation data of various bands. So we use the black sphere thermometer JTR04A, and calculate it according to the method of L164-165. The results obtained include most of the thermal radiation in the wave band.

This study is based on space design, so we do not calculate the thermal radiation from electrical appliances, motor vehicles and human bodies. The research site is open, away from the engine rooms and cars, the investigation time is during the day. Under these conditions, solar radiation accounts for a large proportion of thermal radiation, which is supported by literature and our previous tests.

The microclimate parameters in the study must be distinguishable (for example, when a person comes outside from home, he can feel the air is cold. But if we ask him "do you feel the change in infrared radiation", most respondents can't answer it), and then we can use existing design methods to adjust it. Therefore, we finally adopt the radiation values measured by JTR05 (including direct and diffuse, near infrared and partial mid-infrared, etc.).

The revised parts are highlighted in yellow in: 2.5.

In thermal perception studies, WBGT and IREQ were considered valuable indicators, and we also looked at a number of factors in the survey: clothing level, outdoor stay time, thermal history (order of exposure to different thermal environments and outdoor conditions), Personal background and weather conditions, etc. It was later found that WBGT is more suitable for indoor and industrial environments, evaluating the thermal intensity of the human body throughout the working cycle. This method has been used by the United States and some European countries to evaluate the thermal environmental meteorological conditions of high temperature workshops for many years, and the ISO International Organization for Standardization has formally adopted this method as the standard (ISO7243) since 1982. WBGT index method (the latest standard GB/T4200-2008) is also used in the revised high temperature work classification standard (GB/T4200-1997) in China in 1997. The actual investigation found that the thermal intensity of public activity space in cold cities is at an acceptable range for most of the time, and the duration of outdoor public activities is relatively short, and the state of public activities may change at any time. Therefore, WBGT does not apply. On the other hand, our study is not about individuals, but about a group of people who are purposefully active, whose state is likely to change and who can make themselves more comfortable by taking the initiative, for example, by opening their necks. Take off your coat or gloves, so there is no further analysis of IREQ in this study.

These two indicators are not included in this revision.

Response 3: The source of the standardization coefficient has been supplemented.

The revised parts are highlighted in yellow in: 3.3.

The standardized coefficient reflects the relative influence of microclimatic parameters on the evaluation results. For example, the coefficient of radiation in the hot season is 0.528 and the temperature is 0.272. In the transition season, the temperature effect value becomes larger (0.36), and the coefficient of the other three parameters is bound to decrease, but it does not mean that the actual value of the three parameters will decrease. There are many interesting phenomena that need further study.

Therefore, we do not compare these coefficients more.

The effect of hot wind on thermal sensation is beyond the scope of this study.

Because the air in the outdoor environment is always flowing and people are less sensitive to the air flow than the indoor, the "hot" wind, which can be perceived as "hot" wind, is not only hotter than the human body, but also higher than the initial temperature of the environment. Usually, the hot air rises and the cold air comes in through the lower layer. Therefore, without considering the external heat source, the equipment and the car exhaust heat, the wind in the cold urban area is the cold wind blowing from the periphery. Therefore, the effect of hot air on thermal sensation has not been demonstrated in this study. Climatic conditions also have an effect on humidity, resulting in a very small coefficient of humidity.

But these coefficients can still show the trend of microclimate parameters in cold cities in different seasons, helping designers to find the design priorities.

Thank you again for your help and advice！

Author Response File: Author Response.docx

Round  3

Reviewer 1 Report

After the revision, the quality of the manuscript has improved, in terms of both novelty and clarity. I have now understood the objective of the paper.

I think that, before the publication, some other revisions are necessary.

-          The authors claim that residents of severe cold city are more adaptable to seasonal climate change. Is this a physiological adaptation? Have the authors proven that – for example – tourists in these cities feel more uncomfortable?  Is there an influence of specific clothes that residents commonly use, which can be an alternative reason of the higher comfort sensation?

-          The value of R2 for equations 1-4 is good but not excellent. As people do not often feel variations in thermal comfort if relative humidity vary around the middle range, have the authors verify if an improvement of the R2 can be obtained excluding Rhum as a discrete value from equations 1-4 and applying instead a penalization of ATSV if Rhum overcome threshold values (e.g. 40-60%, or 30-70%)?

-          In the questionnaire, the authors have collected data about the gender and the age of the volunteers. Have the authors some statistical information about the comfort results in terms of these data?

-          In the conclusions, the authors should state the limitations of their study and possible future works.

Author Response

Point 1: The authors claim that residents of severe cold city are more adaptable to seasonal climate change. Is this a physiological adaptation? Have the authors proven that – for example – tourists in these cities feel more uncomfortable?  Is there an influence of specific clothes that residents commonly use, which can be an alternative reason of the higher comfort sensation? 

Response 1: Thank you very much for your advice. In response to this problem, I modified the presentation in the article. The revised parts are highlighted in yellow in: section 4(L284-L289). In my article, I think this adaptation is a physiological adaptation achieved by behavioral regulation. My subjects come from all kinds of public activity spaces, people generally choose the right clothes according to the season, rarely appear extremely abnormal dress. There is another reason that when extreme situations occur, people may leave the outdoor space, assuming that a tourist does not wear cotton clothes during the cold season, then he has two options. One is that he can wear a glove hat, pull a high collar, and do a good job of local insulation. When the first option does not work, he can only choose to leave the outside space, such as staying in the car or indoors. These causes have made the survey of comfort higher.

Point 2: The value of R2 for equations 1-4 is good but not excellent. As people do not often feel variations in thermal comfort if relative humidity vary around the middle range, have the authors verify if an improvement of the R2 can be obtained excluding Rhum as a discrete value from equations 1-4 and applying instead a penalization of ATSV if Rhum overcome threshold values (e.g. 40-60%, or 30-70%)?

Response 2: I explained it in my paper. The revised parts are highlighted in yellow in: section 4(L328-L329). The question you have pointed out is of great value. At present, my findings contain too few samples of humidity beyond the comfort range to support further analysis and calculation. My partners and I plan to carry out a new survey to supplement the deficiencies in this article. About R2: the subjects surveyed in this paper are random objects in urban space, in which R2 is usually lower, for example, [15,16], with R2 values of 0.521 × 0.886, but it is sufficient to explain the problem from the perspective of urban planning.

Point 3: In the questionnaire, the authors have collected data about the gender and the age of the volunteers. Have the authors some statistical information about the comfort results in terms of these data?

Response 3:  I explained it in my paper. The revised parts are highlighted in yellow in: section 4(L326-L328). We have measured the results of comfort, but these results do not show the difference in the perception of microclimate between "different sexes" or "different ages". The reason is that we didn't choose the same number of respondents for age and sex. So, we didn't put the data in this article.

Point 4: In the conclusions, the authors should state the limitations of their study and possible future works

Response 4:On these limitations and prospects, I have added in the article. The revised parts are highlighted in yellow in: section 4(L323-L332). Your advice has helped me so much that I have more ideas for the study.In addition to illustrating these shortcomings, we are preparing to carry out the next study. The first is to recruit regular volunteers (the same number of people in each age group, the same proportion of men and women). In different seasons, volunteers will wear the same clothes for outdoor activities and then interview. In the new experiment, continue to study the questions that this article does not go deep into. Another research is to establish the microclimate evaluation standard and design method for outdoor public space in severe cold cities, which is our main research direction in the future. Thank you again for your help and understanding during this period.

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors,

I read carefully this revised version of your manuscript focused on the assessment of the thermal sensation outdoor by means of subjective surveys.

I have some minor concerns (on the pdf marked file) and some words to spend about your replies not completely satisfactory.

1. About the response #2 I need to stress that IR radiation does not come only from the sun (direct/reflected/diffused) but it is emitted from all surfaces (not necessarily form cars or electrical devices). Have you experience of the asphalt in summer? It is hot and it emits in the range of 9-10 micrometres according to Wien’s law. This the reason why outdoor indices (e.g. PET, UTCI, WBGT) account for both IR sources (sun and surfaces) by means of the mean radiant temperature. In addition, concerning the relationship between thermal sensation (ATSV and OCV) and main heat/cold stress indices (e.g. WBGT or IREQ), it would be preferable that you explicitly state that in this investigation you have preferred to stress the relationship with main microclimatic parameters (in the pdf the support references). Please observe that the first ISO 7243 Standard version has issued in 1989 (not in 1982) and it has been recently updated (2017).

2. About the response #3 you should stress the anomalous trends of normalized coefficients just with one sentence (e.g. this phenomenon requires further investigation). In addition (L168-L170) interviewed vote on two scales (e.g. ISO 10551) not on the perception related to a specific microclimatic quantity.

Cross fingers.

Best regards.

Comments for author File: Comments.pdf

Author Response

Point 1: About the response #2 I need to stress that IR radiation does not come only from the sun (direct/reflected/diffused) but it is emitted from all surfaces (not necessarily form cars or electrical devices). Have you experience of the asphalt in summer? It is hot and it emits in the range of 9-10 micrometres according to Wien’s law. This the reason why outdoor indices (e.g. PET, UTCI, WBGT) account for both IR sources (sun and surfaces) by means of the mean radiant temperature. In addition, concerning the relationship between thermal sensation (ATSV and OCV) and main heat/cold stress indices (e.g. WBGT or IREQ), it would be preferable that you explicitly state that in this investigation you have preferred to stress the relationship with main microclimatic parameters (in the pdf the support references). Please observe that the first ISO 7243 Standard version has issued in 1989 (not in 1982) and it has been recently updated (2017). Response 1: The revised parts are highlighted in yellow in: section 4 (L293-307), and section 3.3. With your advice, I realize the importance of infrared radiation to thermal sensing research. Although the radiometer we use cannot measure infrared radiation directly, we use a black ball thermometer to measure the globe temperature. The difference between the globe temperature and the air temperature represents the average radiation intensity, and the average radiation temperature is compared with that of the solar radiation. Unfortunately, the depth of the authors' research on the average radiation temperature is insufficient, and the analysis may not be deep enough. At your suggestion, I have carefully revised the presentation of the study, emphasized its relationship with major microclimate parameters, and added references. Point 2: About the response #3 you should stress the anomalous trends of normalized coefficients just with one sentence (e.g. this phenomenon requires further investigation). In addition (L168-L170) interviewed vote on two scales (e.g. ISO 10551) not on the perception related to a specific microclimatic quantity. Response 2: The revised parts are highlighted in yellow in: section 2.5, and section 3.3(L252-254), References (L444-450), 4(322-331). In addition to illustrating these shortcomings, we are preparing to carry out the next study. I have added some limitations and future work to the paper. (322-331) Thank you again for your help and understanding during this period.

Author Response File: Author Response.docx

Round  4

Reviewer 1 Report

I think that now the paper is suitable for publication. I only suggest to further discuss the limitation and future works (as explained in the response to the reviewer) in the conclusions and not in section 4. 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round  1

Reviewer 1 Report

The paper deals with the thermal sensations for outdoor in severe cold cities in China. In the paper, the authors present the results of the questionnaire where several people gave a vote to their thermal sensations about temperature, solar irradiance, relative humidity, wind, and so on. At the same time, the authors recorded the climate of the day, aiming at finding a correlation between the results of the questionnaire and the data acquired.

In my opinion, it is not clear what the aim of the paper is. There is also confusion, among the authors, about the word “microclimate”. Microclimate is a word that means climate referred to a close space, but in the outdoor, the use of this word is misleading. Thermal sensations referred to microclimate have, in fact, already been studied since Fanger studies as the authors have written in the introduction. Fanger and the following researchers after him have studied the indoor microclimate and the thermal sensation to seek the best values for temperature, relative humidity, air speed, mean radiant temperature, to ensure users’ comfort. In this case, what is the novelty of the paper? What is the aim of the paper? Is it just a correlation between the results of a questionnaire and external data acquisitions? How the research can be useful for the scientific community? In a sense, the correlation between comfort sensation and indoor microclimate have a sense as HVAC system can be controlled to maintain specific indoor condition. In outdoor case, instead, what is the relation among thermal sensation, climate and engineering/science?

Reviewer 2 Report

The authors conducted a survey in different seasons in Harbin and tried to do a regression between the meteorological parameters and the thermal sensation vote. It is valuable to do this kind of research in the area with serve cold climate. However, many problems are found in this manuscript. The academic writing skill are poor. Major technical problems are found in the Section of Results. Here are my comments as follows:

1). Many typos are found in the manuscript. Please check the manuscript carefully before submitting it. Several expressions were found to be Chinglish and it should be avoided in the academic writing.

2). Method Section, past tense is suggested when describing the research method part.

3). Section 2.2. average air temperature should be a value, not a range.

3). Section 3.1. Mentioned in the previous part that the serve cold season was defined as those having average air temperature below -10 °C. Why the minimum air temperature was 16.5 °C in serve cold season in Part 3.1? 

4). How to determine the difference between sitting while rest or chatting and chess or card entertainment regarding the period of one activity? For example, chatting can also last long time. Also, if the authors are intended to analyse what kind of activity is suitable for the outdoor environment in Harbin, the observation data should be analysed regarding to different seasons.

5). Line 163. Using the word “activist” may be not appropriate. How can we know the difference between activist and normal user from the current observation?

6). Fig. 1(b). The purpose of this figure is unclear. The number of survey samples obtained in different seasons were not the same. In this case, comparing percentage of ASV in different season can lead to wrong conclusion.

7). It is not appropriate that the author lead to a conclusion of “people are only psychologically comfortable but not physically comfortable”? Many researchers have shown TSV=0 might not be the most comfortable situation, please do more literature review before drafting a paper.

8). Section 3.4. Linear regression might not be the best way to do the regression. The authors have figured out that the distribution of the TSV follow normal distribution, so the author should have noticed the limited samples in the two ends of the vote. In other words, it is very likely that this regression can only present the relation between the meteorological parameters and the vote between “-1” and “+1”.

9). Section 3.4. The independent variables used in the regression have not been standardized, the units for the variables are different. It is not suitable to compare the relative impact of the independent variables have to the ASV, because they are incomparable under different units.

10). Minor problems, the authors are suggested to review more recently published high-quality journal papers in the past 5 years, especially for the northern part of China that have the similar climate.

Reviewer 3 Report

Dear Authors

I read with great interest your research.

The manuscript, unfortunately, is characterized by several lacks which reduces its efficacy. This is the reason why I would suggest major revisions.

My main concerns are:

1. You should specify the practical implication of their findings. Otherwise most of all conclusions risk to seem “naif”.

2. Subjective investigations have not been carried out consistently with International Standards in the field (ISO 10551).

3. You measured only solar radiation and did not account for the IR contribution related to the presence of surfaces (do they know the mean radiant temperature?). In addition, they did not measure clothing insulation and they made use of an anemometer with a very poor accuracy (±1m/s). So, it is very difficult trusting in their multivariable regressions (especially in the presence of low air speed values).

4. You did not correlate their ATSV votes with most common metrics used for assessing outdoor climates (eg. UTCI, PET) or cold heat stress (e.g. WBGT, IREQ).

5.In cold conditions maybe a further question on local thermal sensation (local cooling) is necessary

I am confident that you will consider my observations merely as suggestions addressed to the best result. You will find other details in the marked pdf file.

Best regards

Comments for author File: Comments.pdf