Turbulent Flow over a Rough Surface in a Wind Tunnel

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper successfully summarizes and shares its findings while clearly stating the research statement. The authors have effectively conveyed the objectives and methodology of the study, which provides a comprehensive understanding of the research conducted.

However, a critical aspect that appears to be lacking is the clear articulation of the novel contributions to scientific discovery. While the paper effectively presents the data and analysis, it does not explicitly highlight how this research adds value to the existing body of knowledge.

To enhance the paper's impact, the authors should provide a more detailed discussion on the unique aspects of their findings. This includes identifying the specific gaps in the current literature that their research addresses and explaining how their work advances the field. By doing so, the paper would not only demonstrate its relevance but also establish its significance in contributing to scientific progress.

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper presents a study of estimating the basic log law profile parameters by wind tunnel test. It is of importance to the research community since sometimes it is hard to obtain a good mean wind velocity profile during wind tunnel test. This paper studys the roles of roughness elements play for the mean wind velocity profile.

I have several major concerns:

1. Please clarify the length ratio it can be during a wind tunnel test. If it is a small scale wind tunnel, good mean wind velocity profile can be achieved already decades ago, for example, please see the NIST aerodynamic database.
2.  Figure 17 is a bad estimation of the PSD. According to ESDU85020, which is referred to in this paper, Lu is height-dependent and the higher, the larger the value of Lu. While in Figure 17, Lu is larger at Z=0.15 comapred to that at z=0.4. Also, Lu is z0 dependent. The larger the value of z0, the smaller the value of Lu. According to Table 5, z0 reaches its maximum value at around lambda_p=10% to 12%. Therefore, ideally, Lu should reach to its mimimum value between this range. But in Figure 17, when lambda_p is over 20%, Lu shows a decreasing trend.
3. For Eq.20, if this is derived from Eq.1, it should be log_10((z_2-d_0)/(z_1-d_0)). If ignoring d_0 is accepatable in this estimaiton, please clarify this.

Comments on the Quality of English Language

 

1. The way of referring to equations are varied in the paper. For example, in Line 122 and Line 128, 129, 'equation 2', 'equation 4', 'equation 3' are used. In line 170, 'expression 12' is used. In Line380, 'Eqn.' is used. It is suggested to use 'Eq.1', 'Eq.2' for better identification of the equations.
2.  Line 181, the decimal numbers and percentage number appear in the same calculaiton, which is not professional.
3. Line 199, 'whose' is not a formal writing. 'of which' is a better choice here.
4.  Line 202, it shoule be 'were placed'.
5. Line 286, there should be a comma between 'study' and 'a'.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript explores the simulation of turbulent flow over a rough surface within a wind tunnel. The reviewer has the following concerns:

1. How is the turbulence integral scale estimated in the Karman spectrum? It is related to the selected time interval.
2. The spectrum needs to be compared with the newly established spectrum based on measured data.
3. A literature review on the recent advances in wind characteristics of extreme winds is insufficient. Pls have a brief review on the non-stationary characteristics of wind speeds since 2017.
4. The limitation needs to be clarified in the manuscript.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This paper focuses on the parametrization of zero displacement plane d₀ and aerodynamic roughness length z₀ as a function of roughness and areal density via wind tunnel experiments.

The paper is reasonably structured and the use of English language is good. The paper is generally written well, however there are some details that could be discussed further.

Here are some issues that can be addressed, please note:

• The ABL profiles generated in the wind tunnel are always approximations of the full-scale profiles, and therefore introduce more or less significant errors. Given that the authors derive their methods of estimating d₀ and z₀ from experimental values, are the methods presented in the manuscript also applicable to full-scale ABL profiles?

 

• In the abstract, line 12 the sentence seems grammatically incorrect. It should probably be “… wind tunnel measurements of airflow over rough surface.”

 

• On page 1, line 28 the use of English language is incorrect, as it should be “… flow characteristics still remain a topic of discussion and research”

 

• Page 3, line 96 the sentence is confusing as it refers to eq.1 for 𝛿, when there is no 𝛿 in the eq.1. 𝛿 should also be defined here, as it is not clear what exactly it represents (e.g. which height exactly, it should be drawn in Figure 1).

 

• Page 3, line 99 mentions internal layer, referred to Figure 2. where there is no internal layer shown.

 

• Page 3, line 101, the sentence “After the roughness change and below height 𝛿₁, the wind speed depends only on the new roughness 𝑧₀₂”, is there any reference in literature for this statement?

 

• Page 3, line 107, 𝛿₂ is called the thickness of the inner layer, in line 96 𝛿 is called boundary layer thickness, without the inner layer. This needs to be more specific as the terminology is being mixed up. In line 113 the “inner layer height” is mentioned again, what is the inner layer in regards to the boundary layer? Also, it is either “thickness” or “height”, and should be uniform throughout the manuscript.

 

• Just a curiosity of the reviewer, how are the cubes fixed to the floor of the wind tunnel?

 

• Please remove Figure 5, it is unnecessary and does not add value to the manuscript.

 

• In the Figure 9, the legends can be enlarged.

 

• Page 11, line 263, please describe the generalized reduced gradient nonlinear optimization method in a few sentences so that the reader who is not familiar with the method can at least get a general idea of it.

 

• Page 11, line 264 and 265, it is completely unclear to this reviewer how u* was calculated. What methods by other authors are used? References to some equations or literature would be useful. Given the importance of this estimations, it seems required to spend at least a paragraph explaining in detail how this was done for different formulations.

 

• Figure 14 shows turbulence intensities measured in the wind tunnel, how do they relate to full-scale values? It is notoriously difficult to accurately reproduce both mean velocity profiles and turbulence intensity profiles in the wind tunnel in relation to full-scale profiles. In the Figure 14 there should be some comparison to full-scale because these values simply show arbitrary profiles generated in the wind tunnel without any relation to full-scale, i.e. these profiles might not even remotely resemble scaled versions of ABL if not shown otherwise.

 

• Figure 17, the percentages can be placed in the legend of every figure, as they are difficult to notice on the far left of the page.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

1. For the first comment, the reviewer meant to say for a 3mx2mx17m wind tunnel, what is usually the scale down ratio for a model being tested in this wind tunnel. For small scale test, the reviewer has seen good wind profile generated, especially for thoes will small z0 value.
2. Speaking scale down length ratio, the authors did not mention any length scale down ratio of the wind tunnel test in this manuscript. Moreover, the wind velocity profile, such as Figure 8,9,15, the height is from 0 to 1.1m. Therefore, the reviewer can assume that all the results presented in this manuscript is the raw data being measured during the test. In other words, this is a 'full-scale' test. However, considering the dimensions of the wind tunnel, it is impossible to do a full-length-scale test. Therefore, the results presented here can not be directly used in the future experiment, for example, simulating wind pressures on the bridge section or a residential building.
3. For comment 2, the reviewer did not question anything about the value of z0 and the roughness density. The reviewer questioned about the integral turbulence length scale. For rougness density at 20%, 30%, 40%, and 50%, the value of z0 are 0.0222, 0.0153, 0.0107 and 0.0072 according to Table 5. According to ESDU 85020, when the height is less than 15m, the larger the value of z0, the smaller the value of streamwise turbulence length scale. But in this paper, Lu,x are 0.23, 0.19, 0.18, and 0.19. Ideally, it should be showing an increase trend since z0 is decreasing with the roughness density increase from 20% to 50%. The reveiwer hope the authors can understand what the reviewer mean to say here.
4. It is not only the PSD of z=0.4m being changed in Figure 16. The PSD plot of z=0.15m has also been changed by the author. This should be clarified by the author.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for the performed corrections. All questions were answered to a satisfactory degree, and I would therefore recommend the publication of this manuscript.

Author Response

Dear Reviewer,

I hope this message finds you well. I would like to express my deepest gratitude for the time and effort you dedicated to reviewing our article titled "Turbulent flow over a rough surface in wind tunnel". Your meticulous evaluation and valuable observations have been instrumental in improving the quality and accuracy of the content presented.

Sincerely,
Raúl Sánchez

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

1. For comment 2, it is very important to set the length scale down ratio because with different length scale down ratio, the wind field characteristics can be very different. The authors claimed that they normalized the aerodnamic characteristics by the geometry of the cubes but the reviewer did not observe any differences in Figure 8 and 9. It is very important to determine the similarity ratios before the wind tunnel tests. The reviewer does not agree with the author's statement but the reviewer believe their experimental exploration can provide some valuable information to the research community.
2.  For comment 4, the reviewer doubt the method they used to get the value of Lu. By autocorrelation function, Lu can be obtained by measuring Vx at the same time at many different spots along x direction, which seems impossible in the exmperiment. The authors must clarify how they get the autocorrelation function.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf