Review of Wind Flow Modelling in Urban Environments to Support the Development of Urban Air Mobility

Round 1

Reviewer 1 Report

The paper review dozens of papers regarding wind flow modelling in urban environments that support UAM, especially regarding vertical take-off and lending. The paper is very well written, but it seems that the writers come from the UAM community rather than the boundary layer community, thus they have some need for more information that I will try to fill using the following remarks.

Lines 51-63. It would help if you mentioned that there is well-established literature about wind flow in urban environment that is related to UAM although the articles don’t use these keywords. these papers can help the UAM community, thus you should write that the papers covered by this review are a set of selected papers that are representative statistics.

Line 113. It will be helpful to explain about the turbulence length and temporal scale that should be understood, does millimeter scale also important? Does a few meters scale also important?

Figure 7. Long term prediction are up to two weeks and not between day-week. There are alsp seasonal prediction that can predict the wind field up to few months, although their accuracy is doubltable. E.g. Harris, L., Zhou, L., Lin, S. J., Chen, J. H., Chen, X., Gao, K., ... & Stern, W. (2020). GFDL SHiELD: A unified system for weather‐to‐seasonal prediction. Journal of Advances in Modeling Earth Systems, 12(10), e2020MS002223. And many more.

Section 3.2 - RANS. There are many papers using k-omega closure that is better near walls or k-omegaSST closure which is a fusion between k-e and k-omega.  E.g. 

An, K., & Fung, J. C. H. (2018). An improved SST k− ω model for pollutant dispersion simulations within an isothermal boundary layer. Journal of Wind Engineering and Industrial Aerodynamics, 179, 369-384.

Kim, D., Park, J., Park, J. S., & Lee, S. (2023). Optimized Ristorcelli’s Compressibility Correction to the k-ω SST Turbulence Model for Base Flow Analysis. International Journal of Aeronautical and Space Sciences, 1-13.

Line 329. Discussion of a hybrid model of CFD + machine learning of real urban flow is needed. E.g. BenMoshe, N., Fattal, E., Leitl, B., & Arav, Y. (2023). Using Machine Learning to Predict Wind Flow in Urban Areas. Atmosphere, 14(6), 990.

Line 494. CFD can take days, but ML can take minutes. E.g. BenMoshe, N., Fattal, E., Leitl, B., & Arav, Y. (2023). Using Machine Learning to Predict Wind Flow in Urban Areas. Atmosphere, 14(6), 990.

Figure 9. Is not clear, why there is an arrow in the left box between the model and the historical data, and why there is an arrow between WRF to mesoscale model (WRF is a mesoscale model). Please add an explanation for this figure in the text.

Author Response

The authors wish to thank the reviewer for the helpful comments on the
paper. All comments/suggestions have been considered and duly implemented
to enhance the manuscript. Please see the attachment for a point by point response. Also, please note that a pdf file of a version of the revised
paper with all modifications highlighted in blue is provided along with the final paper. Thank you! 

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper reviews methods for modelling air flow in urban environments for the application area of eVTOLS. The scope includes identifying methods that can improve upon the existing definitions of atmospheric disturbance levels in MOC SC-VTOL, including the explicit quantification of the different disturbance levels and the probability of their occurrence.

Overall I think the theme of reviewing modelling methods for this purpose is interesting, and timely for this particular field. I don’t find that the review itself draws clear conclusions about viable next steps for this line of research. I think the summary of deductions should be revisited: several of these are reframing what was said in the introduction, and so are not outcomes from the review. Others are not well evidenced. See below:

-“There is ambiguity in determining what wind parameters…” – where is the evidence for this ambiguity?

-“The flow patterns around buildings must be carefully considered” – this was known before conducting the review

-“This, it is important to strategically position vertiports” – this seems outside of the scope of the review, as it is not related to modelling wind flow

-“There are many models to simulate low altitude wind conditions” – general, and arguably known before doing the review(?)

Compare and contrast: the simulation setup column seem arbitrarily chosen and inconsistent. I think this section really needs to have a set of metrics that are used to contrast each of the methods. These metrics should be used throughout the study

The summaries of the methods worth including, although I think these could be much shorter and just signpost towards relevant literature where they must have been summarised previously.

The argument on p2 that the publication is sporadic from 2011-2018, and then an almost steady increase is not representative of the data being shown in figure 1.

The breadth of the search, which is driven by the search keywords (p2), brings up 42 papers since 2011 What was the reason for the 2011 cut-off? Going to older dates may have brought up a body fo work on rotary and fixed-wing operations in urban environments. And given that the review doesn’t go into the specific of the vehicle, I think the findings in those fields would still be relevant to more modern UAM systems.

It isn’t really made clear what the exact use case is here. For example, P16 – line 493 mentions routine FAM use, but what does this actually mean? This is really needed to frame the entire discussion.

The conclusion in the final paragraph (p19) mentions that a database approach should be used. Firstly, I’m not clear on where the database approach came from (as opposed to bespoke simulations for each forecast). Secondly, if it is a database of precalculated data, why does the model need to be quick?

Specific points

P3 line 73 “closer to ground” than what?

Figure 2: I don’t understand the definitions of the different layers. Is the roughness sublayer supposed to be overlapping with the UCL? If there is no quantitative definition for the boundaries, then this should be stated clearly.

P6 – “is complex to” – I don’t understand this part.

P7 “chaotic” – is this an accurate use of the word here?

P10 “may reduce the computation time by several orders of magnitude” – in comparison to what?

P10 “The prime reason…” This sentence would have been much more useful if it were given in the introduction, to set the scope.

P12 – the paragraph “Figure 8…” doesn’t add any value for me, it just seems to restate what is clear from the figure.

P15 – Windshape doesn’t seem to tackle the problem of quantifying flow topology for urban environments, so it seems out of scope here.

P16 – line 496 – the wind database is something that could have been produced by the other methods too, so why is it only mentioned here? It could be produced from experiments with similar fidelity to CFD (?)

P16 – Turbsim – I am not clear on whether including this method actually provides details of the flow topology due to the individual buildings, which is something that the paper states earlier as being important. Also, please describe what method Turbsim is using.

P19 – Point 6. What is the ‘expense’, is this financial expense? If so, where is the evidence given for the comparison of financial cost of wind tunnel tests against CFD simulations?

Author Response

The authors wish to thank the reviewer for the helpful comments on the
paper. All comments/suggestions have been considered and duly implemented
to enhance the manuscript. Please see the attachment for a point by point response. Also, please note that a pdf file of a version of the revised
paper with all modifications highlighted in blue is provided along with the final paper. Thank you! 

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper conducts a comprehensive survey of research on wind flow estimation and forecasting techniques that contribute to the advancement of Urban Air Mobility (UAM). 

Important considerations:

1. In the introduction section, it is advisable to include additional citations from existing research on Urban Air Mobility (UAM), emphasizing the insufficient attention given to wind flow modeling. Moreover, it is recommended to delve into the designs of Urban Traffic Management (UTM) architectures, with a specific focus on the role of Unmanned Aircraft Systems Service Suppliers (USS). These entities provide a range of operational services, encompassing meteorology and the assessment of wind conditions. Incorporating these details into the introduction would enhance the comprehensiveness of the discussion.
2. The definition and introduction of Urban Air Mobility in the paper appear scattered (line 1-5, line 80-85). As a comprehensive review article, it is suggested to provide a targeted definition and introduction in a specific subsection.
3. The author should further reflect on the differences between existing urban wind field models (Chapter 3) and the models applied in the UAM. It would be beneficial to discuss which models are suitable for UAM, their advantages, and disadvantages. This comparative analysis can be included in the summary section of Chapter 6.
4. In the concluding sections of Chapter 6, the introduction to the next steps of research is somewhat limited. It is suggested to supplement the discussion with challenges, trends, and recommendations for future research in UAM. Beyond contributing to the development of wind models for UAM, consider addressing challenges, exploring emerging trends, and providing recommendations for future UAM research based on the identified literature gaps.

Consideration for improving the work:

1. In line 22-23, please provide a more precise definition of the low-altitude range for UAM operations. It is suggested to refer to the airspace classification systems of the ICAO and various countries, as 2km does not seem to be the standard flight altitude for UAM. The same concern applies to line 180.
2. In line 23-25, when discussing the complexity of wind conditions in urban low-altitude environments, please include relevant literature citations to demonstrate the risks for unmanned aerial vehicles.
3. In lines 603-606, it is suggested to supplement examples illustrating the applicability of various models under different research backgrounds and requirements. For instance, consider cases involving drones with different payloads, diverse operational categories, and varying flight stages, including takeoff, landing, and cruising.

Minor corrections:

1. In line 562, the phrase "be operated closer to ground and obstacles" should be further clarified to avoid ambiguity, for example, "urban buildings, structures, and natural obstacles."
2. In line 517, when first introducing the abbreviation "AAM," please provide its full name.
3. It is suggested to explicitly mention in the caption of Figure 8 that the wind modeling technique is used in the UAM domain.

The manuscript is well-written and needs minor adjustments (corrections) in terms of grammar and typing.

Author Response

The authors wish to thank the reviewer for the helpful comments on the
paper. All comments/suggestions have been considered and duly implemented
to enhance the manuscript. Please see the attachment for a point by point response. Also, please note that a pdf file of a version of the revised
paper with all modifications highlighted in blue is provided along with the final paper. Thank you! 

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

All relevant comments from the review have been effectively addressed in the revisions.

The manuscript is well-written and needs minor adjustments (corrections) in terms of grammar and typing.