Round 1

Reviewer 1 Report

1. The paper provides a clear rationale for investigating unmanned aerial systems (UAS) as an alternative to traditional weather balloons. However, it would be beneficial to briefly mention the environmental advantages of UAS over weather balloons in the introduction.

2. The specific sensor used (TMP117) is mentioned, but it would be helpful to provide a brief overview of its capabilities and advantages in comparison to other sensors.

3. The mention of challenges related to inherent delays in temperature readings is intriguing. It would be valuable to elaborate on these challenges and their potential impact on data accuracy.

4. The use of specific data processing techniques such as the Savitzky-Golay filter, iterative smoothing, time shift, and Newton’s law of cooling is mentioned. It would be beneficial to provide a bit more detail on how each technique was applied and why it was chosen.

5. The paper mentions the examination of 28 flights and the assessment of temperature differentials over a range of 100 meters. Providing more information on the methodology for these assessments and the significance of the 100-meter range would enhance the clarity of the research.

6. While the paper mentions an accuracy enhancement of 0.16°C compared to a radiosonde RS-41 data, it would be helpful to include statistical measures such as standard deviation or confidence intervals to assess the significance of this improvement.

7. The conclusion regarding the potential of UAS in capturing accurate high-resolution vertical temperature profiles is promising. It would be beneficial to briefly discuss the practical implications of this potential, such as improved weather forecasting or atmospheric research applications.

8. The paper concludes by suggesting that UAS could revolutionize atmospheric data collection with further refinements. It would be interesting to include a brief statement on what those potential refinements might entail, such as sensor improvements or data processing advancements.

9. The literature review must be improved. Authors should build upon the wider applicability of UAVs owing to their benefits. Can take the help of recently published studies such as following to show the importance of UAVs and UAV-based measurements in current context:

https://link.springer.com/article/10.1007/s11270-020-04973-5

Inoue, J., & Sato, K. (2022). Toward sustainable meteorological profiling in polar regions: Case studies using an inexpensive UAS on measuring lower boundary layers with quality of radiosondes. Environmental Research, 205, 112468.

https://www.mdpi.com/2504-446X/6/12/406

 

 

Overall, the paper provides a solid overview of the research but could benefit from additional details and context to enhance its clarity and impact.

Major revision

Author Response

1. The paper provides a clear rationale for investigating unmanned aerial systems (UAS) as an alternative to traditional weather balloons. However, it would be beneficial to briefly mention the environmental advantages of UAS over weather balloons in the introduction.

The Authors agree with this remark. This aspect is briefly described in lines 44-47. 

2. The specific sensor used (TMP117) is mentioned, but it would be helpful to provide a brief overview of its capabilities and advantages in comparison to other sensors.

The authors concur with this observation, and additional elucidation has been supplied to bolster the rationale for selecting the TMP 117 for subsequent analysis. The decision to employ this sensor was principally influenced by two primary contributing factors: the design characteristics of the thermometer and its low Root Mean Square Error (RMSE) when compared to readings from an established ground station etalon. Lines 301-310.

3. The mention of challenges related to inherent delays in temperature readings is intriguing. It would be valuable to elaborate on these challenges and their potential impact on data accuracy.

In this article, one type of thermometer was used to analyze its inertial properties. It has a specific delay to temperature expressed as “reaction time constant”. It is impossible to change its parameters unless changing the thermometer manufacturer or design of an enclosure. Reaction time constant was evaluated on the ground as to eliminate external factors as much as possible. Another factor influencing the speed and or delay of temperature was climb and or descent rate, which was not analysed as a separate parameter in this study and would be a good addition to our further research. 

4. The use of specific data processing techniques such as the Savitzky-Golay filter, iterative smoothing, time shift, and Newton’s law of cooling is mentioned. It would be beneficial to provide a bit more detail on how each technique was applied and why it was chosen.

The Authors agree, further detail on smoothening methodologies could be expanded to make article clearer. A brief description of how these algorithms were implemented in this specific study and how they work were outlined. Lines 570-583.

5. The paper mentions the examination of 28 flights and the assessment of temperature differentials over a range of 100 meters. Providing more information on the methodology for these assessments and the significance of the 100-meter range would enhance the clarity of the research.

Examining the lower portion of the atmospheric boundary layer may not yield prognostic data as substantially impactful as data encompassing the entire troposphere. Nevertheless, it is important to acknowledge that such information remains valuable, particularly in the context of urban heat island (UHI) assessments, temperature inversion analyses, or other specialized investigations pertaining to specific atmospheric conditions. Lines 39-43.

6. While the paper mentions an accuracy enhancement of 0.16°C compared to a radiosonde RS-41 data, it would be helpful to include statistical measures such as standard deviation or confidence intervals to assess the significance of this improvement.

 
The authors agree with this remark, more details about the accuracy enhancement, such as standard deviation and confidence intervals were mentioned in the discussion part to outline the significance of improvement. The model was found to differ from RS-41 radiosonde data by 0.16±0.014 °C with 95% confidence when applying Newton’s law of cooling. Lines 673-677 and abstract. 

7. The conclusion regarding the potential of UAS in capturing accurate high-resolution vertical temperature profiles is promising. It would be beneficial to briefly discuss the practical implications of this potential, such as improved weather forecasting or atmospheric research applications.

Additional elaboration on the significance of this research in the context of weather forecasting and atmospheric research applications has been incorporated into the Discussion section. The specific focus of this research holds the potential to enhance our capabilities in addressing urban heat island (UHI) phenomena and the formation of inversions, especially in densely populated areas. Furthermore, it contributes to the broader understanding of this field, augmenting our knowledge base in these critical areas of atmospheric science.  Lines 39-43. 

8. The paper concludes by suggesting that UAS could revolutionize atmospheric data collection with further refinements. It would be interesting to include a brief statement on what those potential refinements might entail, such as sensor improvements or data processing advancements.

The authors concur with this remark. The discussion includes several proposals on how UAS could analyze vertical boundary layer structure, including a drone swarm system and real-time processing configuration. Furthermore, potential areas for further research were outlined in the Discussion section. Lines 698-709.

9. The literature review must be improved. Authors should build upon the wider applicability of UAVs owing to their benefits. Can take the help of recently published studies such as following to show the importance of UAVs and UAV-based measurements in current context:

https://link.springer.com/article/10.1007/s11270-020-04973-5 

Inoue, J., & Sato, K. (2022). Toward sustainable meteorological profiling in polar regions: Case studies using an inexpensive UAS on measuring lower boundary layers with quality of radiosondes. Environmental Research, 205, 112468. 

https://www.mdpi.com/2504-446X/6/12/406 

 

Thank you for Your input and UAS research. Overall, the first paper provides a solid overview of the air quality research, our paper could benefit from additional details and context to enhance its clarity and impact. The second article does not have anything in common with our field of study. Analysis of CNN networks while tracking vehicles not related to temperature sensing analysis. 

Overall, the paper provides a solid overview of the research but could benefit from additional details and context to enhance its clarity and impact.

Reviewer 2 Report

The article is devoted to the use of drones to capture meteorological data (temperature, humidity). The article is well structured and the introduction provides a good overview of existing research in this area. This article will be a good addition to existing research. I like that the authors tested several temperature sensors and provided the test results. Overall this is a good study, but I would like to see how the data obtained in this article compares with other studies. Also I have some comments:

1)Please use author names when citing articles.

2)Maybe it will look better:

"After correction, the authors were able to achieve a root-mean-square-error (RMSE) between ascent and descent equal 0.2 °C."

3)At the end of the “introduction” section, point off the problems in your scientific area and which issues have already been solved, which are partially solved, and which are being solved by yours work.

4) Line 297. Change "Te" to "T_e" (add subscript)

Author Response

Comments and Suggestions for Authors 

The article is devoted to the use of drones to capture meteorological data (temperature, humidity). The article is well structured and the introduction provides a good overview of existing research in this area. This article will be a good addition to existing research. I like that the authors tested several temperature sensors and provided the test results. Overall this is a good study, but I would like to see how the data obtained in this article compares with other studies. Also I have some comments: 

1)Please use author names when citing articles. 

Author names have been included to improve readability. 

2)Maybe it will look better: 

"After correction, the authors were able to achieve a root-mean-square-error (RMSE) between ascent and descent equal 0.2 °C." 

Looks good, thank you!

3)At the end of the “introduction” section, point off the problems in your scientific area and which issues have already been solved, which are partially solved, and which are being solved by yours work. 

This has been included in the introduction. Lines 98-113 (with "All Markup"). Thank You.

4) Line 297. Change "Te" to "T_e" (add subscript) 

Thank you for noticing the formatting mistake. We have fixed it in the manuscript.

Reviewer 3 Report

Summary

The relevance of the article's topic is unquestionable. The authors provide a detailed presentation of the most important scientific precedents of the topic, offering an excellent summary of the literature on drone measurements. However, they do not mention any future events or plans that would provide essential information in the discussion of the topic.

The structure of the article is appropriately organized, and the detail of each section is consistent. However, there are some inaccuracies in places, and in some cases, there could be more detailed discussion about the limitations of the given measurement parameters, which is not addressed in the article.

The message of the article is clear, its conclusions are scientifically grounded and meet the desired standard. Based on the above, I recommend accepting the article after minor revisions.

 

Minor issues need to be revised

The authors make no mention of the UAS demonstration campaign by the World Meteorological Organization (https://community.wmo.int/en/uas-demonstration) which is crucial background information for the topic. The comprehensive review of the scientific literature background should refer to this.

The Vertical Take-Off and Landing (VTOL) drone expression is usually used for fixed-wing UAVs. VTOL drones combine the cruising flight efficiency of fixed-wing aircraft with the convenient vertical take-off and landing of multirotors. The DJI drone, mentioned in the article, is a multirotor UAV. That should be corrected.

The authors do not detail in several instances how the parameters used in the experimental configuration might influence the drones' usability in this direction. For example, at a vertical speed of 1 m/s, the limited flight time would not even be sufficient to survey the planetary boundary layer (which is the declared goal of drone measurements by the WMO); the peak altitude of 120 meters is not competitive with radiosondes, and in special categories, preparations are being made in several countries for the WMO campaign. It would be worth dedicating a few sentences to discuss these points.

 

Author Response

Summary 

The relevance of the article's topic is unquestionable. The authors provide a detailed presentation of the most important scientific precedents of the topic, offering an excellent summary of the literature on drone measurements. However, they do not mention any future events or plans that would provide essential information in the discussion of the topic. 

The structure of the article is appropriately organized, and the detail of each section is consistent. However, there are some inaccuracies in places, and in some cases, there could be more detailed discussion about the limitations of the given measurement parameters, which is not addressed in the article. 

The message of the article is clear, its conclusions are scientifically grounded and meet the desired standard. Based on the above, I recommend accepting the article after minor revisions. 

Minor issues need to be revised 

The authors make no mention of the UAS demonstration campaign by the World Meteorological Organization (https://community.wmo.int/en/uas-demonstration) which is crucial background information for the topic. The comprehensive review of the scientific literature background should refer to this. 

The authors are thankful for the valuable suggestion. It is now included in the manuscript and enhances the context of the research greatly. 

The Vertical Take-Off and Landing (VTOL) drone expression is usually used for fixed-wing UAVs. VTOL drones combine the cruising flight efficiency of fixed-wing aircraft with the convenient vertical take-off and landing of multirotors. The DJI drone, mentioned in the article, is a multirotor UAV. That should be corrected. 

Even though rotary wing aircraft do take off and land vertically it is also true that the term Vertical Take-Off and Landing (VTOL) vehicle is usually used to refer to fixed-wing aircraft. The authors replaced the term to avoid confusion. Thank You.

The authors do not detail in several instances how the parameters used in the experimental configuration might influence the drones' usability in this direction. For example, at a vertical speed of 1 m/s, the limited flight time would not even be sufficient to survey the planetary boundary layer (which is the declared goal of drone measurements by the WMO); the peak altitude of 120 meters is not competitive with radiosondes, and in special categories, preparations are being made in several countries for the WMO campaign. It would be worth dedicating a few sentences to discuss these points.

A more detailed explanation was provided in section 2.2.4. While it is true, that with the theoretical flight time of 50 minutes only the lowest 1500 m would be covered, which is a significant part of the boundary layer. On the other hand, the vertical rate might be increased if the temperature gradient decreases. This velocity was chosen to minimize the inertial error. With a lower gradient, this error is not as pronounced, therefore the velocity can be increased to reach a higher altitude. We are working on a technical solution as well as regulatory approval to make higher altitude measurements possible. On the other hand, altitudes up to 120 meters above ground can give valuable insight into the hydrothermal structure in urban environments, such as temperature inversions, urban heat islands, and other special conditions, which would be difficult or impossible to measure in situ by other means. These points have been included in the manuscript.

Reviewer 4 Report

 

This paper investigates using drones to measure the key parameters of urban air environment. The strengths of the drone-based method are declared by comparison with the weather balloon-based counterpart. The proposed method can be superior compared to the existing method, if necessary quantitative comparisons are made. Therefore, the reviewer would like to give the following comments for reference.

1) General concept comments

1.         Literature review. Up to 17 references are analyzed in this part. This part is now more like a technical report with statements. The reviewer would like to see more summaries of the existing methods including what has been done well by previous works, and what are the drawbacks therein.

2.         Additionally, where possible (if allowed by the journal), the authors might as well compare the existing methods in a table. This can be done in the introduction or after statement of the proposed method. The terms to be compared may be related to the content of this paper, for instance, the hardware and algorithms used.

3.         To follow such a summary in the first comment, please clearly state the contributions or novelties of this work in the introduction. Also, please strengthen this again in the conclusion.

4.         The reviewer recommends that the authors add a figure of the drone system. This can be an illustrative diagram with drone, its hardware, and ground station, since this task is unique compared to the existing drone applications.

5.         It is declared in the abstract that “By comparing the data … we spotlight the UAS's superior cost-effectiveness, mobility, and reusability in comparison to weather balloons”. But in the quantitative analysis, such comparisons with weather balloon-based method are not made. Only different filter algorithms are compared.

6.         To follow the previous comment, what is the cost of the proposed drone-based method? This should be compared with existing drone-based method.

7.         What is the applicable range of altitude and temperature? Is there a limit on these conditions? The reviewer guesses the upper limit of altitude when using drones should be lower than that of using balloons, but this should be clearly reported in the paper.

8.         Regarding safety – will the drone fly out of the operator view? In other words, is the mission to measure the urban environment a visual flight or an instrumental flight? In both cases, one should guarantee there is no loss of control. Please discuss.

9.         What is the core novelty of this paper? It seems that the used hardware and algorithm are not new. Please discuss.

2) Specific comments referring to line numbers, tables, or figures

1.         In Lines 64, 71, and 79, for instance, if the number of the cited reference is at the beginning of one sentence, it will look better to add “Ref.” in front of the number.

2.         Figures with presented data are in low quality, see for instance Figure 2(a) and (c). Please guarantee they are in high resolution.

3.         The shaded bar of “20s after transition H0 -> H1” cannot be found in Figure 2(a) and (c), but only appears in (b) and (d). If so, The reviewer thinks the legends should be different of these two groups of figures.

4.         In Line 298, the double quotation marks for t look strange - „t”. Please use the style in Line 297, i.e., “”.

5.         Please correct unnecessary italic fonts in equations, e.g., “d” in Eq. (1) that represents differential operation should be normal instead of italic since it is not a variable. In Line 300, the expression “T(t)”, both T and t should be italic. These are just two examples. Please carefully check the paper.

6.         In Line 297, the subscript e of “Te” should be written just as it is in Eq. (1) – in terms of its size.

7.         Title of Subsection 3.2, there is an unnecessary dot in the end.

8.         Figure 4, the authors use colorbar for lines. Although it is explained below “colors represent different time intervals under examination”, There should be a colorbar legend on the side of these two subfigures.

 

Author Response

This paper investigates using drones to measure the key parameters of urban air environment. The strengths of the drone-based method are declared by comparison with the weather balloon-based counterpart. The proposed method can be superior compared to the existing method, if necessary quantitative comparisons are made. Therefore, the reviewer would like to give the following comments for reference. 

 

1) General concept comments 

1. Literature review. Up to 17 references are analyzed in this part. This part is now more like a technical report with statements. The reviewer would like to see more summaries of the existing methods including what has been done well by previous works, and what are the drawbacks therein.

Authors are thankful for your insights. Some of the notable methods were summarized in lines 247-259. 

2. Additionally, where possible (if allowed by the journal), the authors might as well compare the existing methods in a table. This can be done in the introduction or after statement of the proposed method. The terms to be compared may be related to the content of this paper, for instance, the hardware and algorithms used.

The authors concur with this observation that the format of literature representation in this article might be perceived as somewhat perplexing. In response, we have provided a more concise summarization of the work conducted by other researchers. The inclusion of a table in this context may not offer significant clarity, as various authors have employed a range of techniques, some of which may bear similarities or even be identical, yet discerning this with certainty can prove challenging. Some of the methodologies or certain details in other researchers' work were omitted, making structured comparisons in table difficult, therefore authors believe a summary in this instance of other’s people work suits best. 

3. To follow such a summary in the first comment, please clearly state the contributions or novelties of this work in the introduction. Also, please strengthen this again in the conclusion. (METHODS)

A novel method was introduced including hardware, measurement and data processing techniques. The accuracy of the measurements improved compared to some other authors’ work and is in line with the World Meteorological Organization’s (WMO) requirements. Such an evaluation has not been found to include the surface atmospheric layer (~100 m) in other authors’ work. 

Introduction: lines 106-110.

Conclusion: 682-692.

4. The reviewer recommends that the authors add a figure of the drone system. This can be an illustrative diagram with drone, its hardware, and ground station, since this task is unique compared to the existing drone applications.

Authors thank the reviewer for the observation. We agree with the comment and have added a photograph of the vehicle with the measurement system (figure 2). 

5. It is declared in the abstract that “By comparing the data … we spotlight the UAS's superior cost-effectiveness, mobility, and reusability in comparison to weather balloons”. But in the quantitative analysis, such comparisons with weather balloon-based method are not made. Only different filter algorithms are compared.

The sentence is indeed misleading. It will be rewritten, so that the actual work is described. Lines 15-19. 

6. To follow the previous comment, what is the cost of the proposed drone-based method? This should be compared with existing drone-based method.

In compliance with Your observation, we have calculated the cost of the system and compared it with the approximate cost of radiosonde launch. The price of the sensing unit was included, however this is difficult to compare against prices of other systems with vehicles included. Lines 54-76.

7. What is the applicable range of altitude and temperature? Is there a limit on these conditions? The reviewer guesses the upper limit of altitude when using drones should be lower than that of using balloons, but this should be clearly reported in the paper.

The maximum reachable altitude with a vertical speed of 1 m/s and the maximum flight time of 50 minutes is 1500 m. The temperature range is limited from -20°C to 50°C as stated by the UAV manufacturer. It is suitable for all, but several of the coldest days in the region. All hardware limitations were outlined in separate section 2.2.4. 

8. Regarding safety – will the drone fly out of the operator view? In other words, is the mission to measure the urban environment a visual flight or an instrumental flight? In both cases, one should guarantee there is no loss of control. Please discuss.

According to the requirements of the European Union Aviation Safety Agency all flights in the “open” category, in which we operate, must maintain a visual line-of-sight between the vehicle and the pilot. It is possible for organizations to obtain a certificate allowing autonomous flight in which case remote control is not required. To obtain this certificate, organizations have to provide operations risk assessment. In the context of this study, it is worth noting that the maximum altitude above ground level was limited to 120 meters. Consequently, a continuous visual line-of-sight was maintained with the multirotor throughout the operations, obviating the need for any specialized requirements. However, it is important to mention that in the event of a cloud layer obstructing the line of sight or if the need arose to extend the flight to higher altitudes, specific certifications would be required to ensure compliance with relevant regulations and safety standards. 

9. What is the core novelty of this paper? It seems that the used hardware and algorithm are not new. Please discuss.

This paper highlights the temperature measurement accuracy associated with the measurement method proposed. The measurements adhere to the accuracy and inertial requirements of WMO. Moreover, the accuracy greatly exceeds the ‘breakthrough’ requirement, which would result in a significant improvement for the numerical weather prediction application. Furthermore, it is noteworthy that the phenomenon of hysteresis is seldom discussed as a potential issue in other articles. This article aims to shed light on and thoroughly analyze this aspect, emphasizing its significance in the context of the research. 

2) Specific comments referring to line numbers, tables, or figures 

1. In Lines 64, 71, and 79, for instance, if the number of the cited reference is at the beginning of one sentence, it will look better to add “Ref.” in front of the number.

In response to another reviewer, names of the authors of the references cited were included. 

2. Figures with presented data are in low quality, see for instance Figure 2(a) and (c). Please guarantee they are in high resolution.

Figure 2 was updated and all the subfigures are now in high quality. Note that Figure 2 is now referenced as Figure 3. 

3. The shaded bar of “20s after transition H0 -> H1” cannot be found in Figure 2(a) and (c), but only appears in (b) and (d). If so, The reviewer thinks the legends should be different of these two groups of figures.

The bars can now be visible in all the figures. Now they share a single legend. 

4. In Line 298, the double quotation marks for t look strange - „t”. Please use the style in Line 297, i.e., “”.

Thank You. The quotation marks are also inconsistent. The authors have replaced them with “”. 

5. Please correct unnecessary italic fonts in equations, e.g., “d” in Eq. (1) that represents differential operation should be normal instead of italic since it is not a variable. In Line 300, the expression “T(t)”, both T and t should be italic. These are just two examples. Please carefully check the paper.

The equations and variables across the whole manuscript were corrected according to Your observation. 

6. In Line 297, the subscript e of “Te” should be written just as it is in Eq. (1) – in terms of its size.

The authors are thankful for pointing out the inconsistency. It is corrected. 

7. Title of Subsection 3.2, there is an unnecessary dot in the end.

The dot is deleted. 

8. Figure 4, the authors use colorbar for lines. Although it is explained below “colors represent different time intervals under examination”, There should be a colorbar legend on the side of these two subfigures.

The legend was indeed missing. It is added to the figure. 

Round 2

Reviewer 4 Report

The authors have made significant efforts to improve the paper quality. Most of the issues are addressed accordingly. The reviewer still has the following comments for them before accepted for publication.

1.      For the newly added Figure 2, please denote the components such as sensors on the figure.

2.      Lines 324-325, it is a little bit weird to see in the end of this sentence “section (a)”. Should this be Figure 2(a)?

3.      Line 326, the word “motors” is in the same line with the figure, this is also weird.

4.      In Figure 5, the supplemented colorbar should be accompanied by numeric scales that illustrate the flight time, for instance, 1, 2, 3, … (e.g., in hours)

Author Response

The authors have made significant efforts to improve the paper quality. Most of the issues are addressed accordingly. The reviewer still has the following comments for them before accepted for publication.

Thank You for Your feedback.

For the newly added Figure 2, please denote the components such as sensors on the figure.

Thank You for the suggestion. We have revised the figure with borders for components and named them. Exact sensors and placement in housing can be seen in Figure 1.

Lines 324-325, it is a little bit weird to see in the end of this sentence “section (a)”. Should this be Figure 2(a)?

Thank You for the suggestion. We have renamed the reference.

Line 326, the word “motors” is in the same line with the figure, this is also weird.

Thank You. Word is not in the manuscript anymore.

In Figure 5, the supplemented colorbar should be accompanied by numeric scales that illustrate the flight time, for instance, 1, 2, 3, … (e.g., in hours)

Thank You for the suggestion. We have revised Figure 5. A numeric scale in minutes has been added.