Development and Validation of a Custom Stochastic Microscale Wind Model for Urban Air Mobility Applications

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The realistic, computationally afordable models of the atmospheric turbulence around buildings are very important for enabling credible simulations of UAV usage in urban environments. In this sence the paper is of high interest for the public.

The reviewer has the following remarcs:

line 253: dimension error "AOI of 400 × 400 × 300, cm, discretized using a uniform Cartesian grid with a spatial resolution of 1m."

The origin of fig. 10a and 10c is not clearly declared in the text of the figure;

Table 1 could be separated in two for clarity;

Fig.6 could be split in 3 and figure parts (for TS1, 2 and 3) be located respectively in the beginning of sub-subsections 3.2.1, 3.2.2, 3.2.3 to ease the readers interpretation of the results;

subpart 4.3: TI (probably Turbulence Index) NOT DEFINED in the paper;

Actually in section 5 the authors have admitted that their model has certain weaknesses and  limitations. This shell be adequately reflected also in the abstract of the paper.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper describes an efficient approach to modelling wind variability in urban domains using SWM, a synthesis of QUIC-URB, a steady flow solver, and TurbSim, a stochastic turbulence solver. The authors examine the performance of SWM in three cases where CFD and experimental data are available for comparison, and apply it to study a scenario with practical value by embedding their solver inside a mesoscale atmospheric model.

The paper has been written with great care and leaves little to criticise in terms of expression, description, illustration or analysis. The only potential shortcomings relate to the applicability of SWM to address the motivation for this work, which is the difficulty of regulating the operation of uncrewed and passenger-carrying uncrewed aerial vehicles in urban environments where practical experimentation is unsafe.

The authors argue that a sufficiently accurate simulation system could be employed to assess such operations in lieu of physical experiments, and show that such a system must reproduce microscale spatial and temporal wind variability. At the conclusion of the paper, it is clear that SWM has strengths and limitations; it can simulate complex environments with considerably less computational resources than CFD, but the results are of variable quality and may be least reliable where they would be of most value - scenarios with strong winds and high wind variability.

The most important question that remains unanswered at the conclusion of the paper is whether or not the proposed system is sufficiently accurate to validate the safety of uncrewed air mobility vehicles in simulation. This deficiency could be addressed by assessing how the stability of representative urban air mobility vehicles might vary given the difference between simulated and experimental velocity and turbulent intensity values. Without a metric of this kind, the value of the suggested approach is difficult to assess.

A second question that should be addressed in the paper is whether models based on deep learning and machine learning, which are becoming operationally used in numerical weather prediction, could provide an alternative to the conventional approximations used to reduce computational loads in SWM. The authors suggest that enhancements of public-domain variants of QUIC may be worthy of attention. It is surprising that they omit newer approaches.

The attached marked up version of the manuscript has a few very minor comments. This is a paper to be extended rather amended.

 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Observations:
1. Be careful when using abbreviations. They must be explained in advance before the first use;
2. In the text of the work, there is an inconsistency in the way figures are referred to (when Figure is used, when Fig.). I recommend using the same way of referring throughout the text of the work.
3. The conclusions chapter of the work is missing. The fact that the authors did not centralize the ideas and results they reached in such a final chapter of conclusions creates the feeling in the reader that the work is not completed.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Authors present a stochastic microscope wind model to validate and support operation of UAM vehicles. Authors claim the model can rapidly generate high-resolution, quasi-non-stationary urban wind fields. The model is validated using wind tunnel data and later refined by use of urban wake simulation. The model is also coupled with a weather forecasting model.

My first comment is that, in general, the references lack of information. The format should be reviewed for completeness. Dates for conferences, issues for Journals and DOI should be provided when applicable. For government reports, links should be provided. 

The manuscript would benefit from a clearer and more coherent organizational structure. Several sections introduce related ideas without a well-defined hierarchy, which affects the overall readability. In addition, the repeated use of the term “QUIC”, appearing more than ninety times in twenty-two pages, suggests that some explanations may be unnecessarily reiterated. Consolidating background material, reducing redundancy, and improving the flow between sections would substantially strengthen the presentation.

Acronyms should be specified only when at first appearance. However, some of them are presented multiple times in the manuscript. That creates unnecessary repetition and sometimes are confusing.

Figure 4 states a "for each" which can be understood as an iteration loop. However, there is no iteration in the arrow. Suggest updating the chart or the text to reflect the idea with more clarity.

Unless there is a specific reason to mention of QES system between lines 193-205, it should be removed since it does not add meaningful contributions to this work once the authors did not use that moving forward in this manuscript.

Figure 5, number "1" in the TS2 case is missing. 

Lines 231-263 could be replaced by a simplified paragraph, since the information is on the table 1.

Figure 6 is confusing for TS1 and TS2. If there is not enough space for improvement, recommended splitting into multiple figures. 

Figure 7 is not clear. Recommending add legend to specify which curve represents what. Also, having the curves with similar style is not helpful. The X/W_ is also not clear to what each of them represents. The block on the left has y-axis as Z/W while the two other blocks have Y/W labels. Since all rows seem to have the same X/W for three cases, you could remove some of the axis labels and make them a single figure to enlarge each plot.

Figures should follow Figure 13 standard (in terms of legend)

The word "etc" should be avoided.

The symbol "~" should be replaced by "approximately" or similar.

Although the abstract and stated contributions emphasize the role of SWM, the manuscript gradually shifts its focus and later sections discuss only the QUIC-related results. This discontinuity weakens the coherence of the narrative and obscures the contribution associated with SWM. The QUIC system is not mentioned in the abstract. A clear distinction or emphasis where SWM fits in the overall architecture would be appreciated.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I congratulate the authors for the work presented.