Characteristics of Rain-Induced Attenuation over Signal Links at Frequency Ranges of 25 and 38 GHz Observed in Beijing

Round 1

Reviewer 1 Report

The following are additional corrections that are needed.  There are still English problems in this manuscript.  Remote Sensing must have an editor fluent in written English proofread the manuscript before it goes to print.

l. 90 
Delete extra period after model.

l. 109
Correction: "...by a rain gauge...".

l. 121
rephrase: "Beijing, ...which sits along the northwest edge of the North China Plain."

l. 193
Add comma after "types": "...rain types, and in fact..".

l. 365
"Discussions" should be "Discussion".

l. 372-373
"The convective showers was dominant." should be 
"Convective showers were dominant".  This is an example of the English problems that are still in
the manuscript.

l. 380-381
When the average rainfall rate per 15-min 380
reached above 10 mm/h, the disdrometer-derived A-R model shows noticeable better 381
rainfall rate estimation. 

This has errors. 
"reached above" is incorrect usage; it should be "exceeded" or "went above".
"noticeable" should be "noticeably".

l. 385
"from ITU" should be "from the ITU".

Author Response

Thank you for your review comments and suggestions. Please find our responses in the attached file.

Reviewer 2 Report

Most of my previous comments have been addressed. I only have a few minor comments below. However, I would like to point out that there is still significant language editing required.

1. Please do not use "rains" in the plural form in the context of this study. Common usage is e.g. "Convective and stratiform rain are two types of rain." The same is true for drizzle. Use "types of drizzle" instead of "drizzles".

2. Line 255: The definitions of convective and stratiform rain are not complete: (a) how is rain with a rain rate <5 mm/h classified? (b) how is rain with a rain rate >5 mm/h but std of <1.5 mm/h classified? (c) how is rain with a rain rate of <5 mm/h but a std of >1.5 mm/h classified?

3. Figs. 4 and 5: It would be helpful to add "convective" and "stratiform" in the panel titles of Figs. 4 and 5.

4. Line 394: Please provide actual numbers on how much improvement was achieved by using locally derived coefficients in the text, not just in the table.

Author Response

Thank you for your review comments and suggestions. Please find our responses in the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. Note that a very large number of studies and publications are devoted to the statistics of drops in the clouds, so there are generally accepted concepts, such as the distribution of drops by size (this is first goal in the title of the work). Then your new statement should be proved in detailed theory and compared with the traditional results inside atmospheric physics. Although the authors have added some explanations, the new theory proposed in the paper in equations (1-4), (7) and its result in (8) look meager and not convincing. In particular, the physical meaning and purpose of formula (2) are not clear, the (3) is not related to other formulas, the dimensions of values in both parts of the formula (8) do not coincide, implied dimensional coefficients are not described. The augmented standard formula (4) with N(D) could be incorporated into (6) and also compared with the sum from (1).

Therefore, the resulting formula (8) looks strange compared to the known analogues.

2. My opinion is that the article is not sufficiently developed, it can be published only after additional revision .

At the same time, the extensive data accumulated by the authors on the measurement of losses for radio signals for mobile communications in the second part of the article are of interest and value. It can be published in a new article with a title, for example, " Measurement of signal loss at frequency range of 25 and 38 GHz during rain or drizzle "

3. The most appropriate theory for radio link could be a deepen the formula (5) and Mie scattering.

Modern radars and lidars are more suitable to determine the cloud parameters, especially since near the ground it is easier to measure the precipitation intensity, R in mm/hour, and use the generally accepted Marshall-Palmer algorithm.

4 .It would be good to compare the accuracy of the measurements obtained by the authors with other radar measurements at close frequencies, especially since Figures 4 (a-c) show a poor coincidence of measurements and calculations and the fact is not explained in the text. A bad match is discussed for Figure 5, but probably the Marshall - Palmer formula should be used which is optimized specifically for heavy rains instead of gamma distributions. The redesigned Figure 6 is nice and informative. The words 'more stable’,' increased reliability ' for the method compared to the radar (lines 404-408) are not considered and not proved above in presented text.

Author Response

Thank you for your review comments and suggestions. Please find our responses in the attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

1.Note that a very large number of studies and publications are devoted to the statistics of drops in the clouds, so there are generally accepted concepts, such as the distribution of drops by size (this is first goal in the title of the work). Then your new statement should be proved in detailed theory and compared with the traditional results inside atmospheric physics. Although the authors have added some explanations, the new theory proposed in the paper in equations (1-4), (7) and its result in (8) look meager and not convincing. In particular, the physical meaning and purpose of formula (2) are not clear, the (3) is not related to other formulas, the dimensions of values in both parts of the formula (8) do not coincide, implied dimensional coefficients are not described. The augmented standard formula (4) with N(D) could be incorporated into (6) and also compared with the sum from (1).

Therefore, the resulting formula (8) looks strange compared to the known analogues.

2. My opinion is that the article is not sufficiently developed, it can be published only after additional revision .

At the same time, the extensive data accumulated by the authors on the measurement of losses for radio signals for mobile communications in the second part of the article are of interest and value. It can be published in a new article with a title, for example, " Measurement of signal loss at frequency range of 25 and 38 GHz during rain or drizzle "

3. The most appropriate theory for radio link could be a deepen the formula (5) and Mie scattering.

Modern radars and lidars are more suitable to determine the cloud parameters, especially since near the ground it is easier to measure the precipitation intensity, R in mm/hour, and use the generally accepted Marshall-Palmer algorithm.

4. It would be good to compare the accuracy of the measurements obtained by the authors with other radar measurements at close frequencies, especially since Figures 4 (a-c) show a poor coincidence of measurements and calculations and the fact is not explained in the text. A bad match is discussed for Figure 5, but probably the Marshall - Palmer formula should be used which is optimized specifically for heavy rains instead of gamma distributions. The redesigned Figure 6 is nice and informative. The words 'more stable’,' increased reliability ' for the method compared to the radar (lines 404-408) are not considered and not proved above in presented text.

Author Response

Thanks

Round 3

Reviewer 3 Report

This article is an important study for wireless communication, and in particular the attenuation of the 25-GHz and 38-GHz signals during rains having different intensities. The results are well presented and interpreted. The article will be interesting and informative for readers. 

Author Response

Thanks for your comments

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

Review provided in the attached file.

Comments for author File: Comments.pdf

Author Response

We would like to thank the reviewer for very valuable comments and feedback. Our replies are attached in the document.

Author Response File: Author Response.pdf

Reviewer 2 Report

Review of remotesensing-1123396 entitled "Characteristics of the Raindrop Size Distribution and Rain-induced Attenuation Over Millimeter-wave Links Observed in Beijing" by Han et al. 2021.

The authors use 25 and 38 GHz field data from a millimeter wave link and distrometer and rain gauge data to study attenuation caused by precipitation and it's relation to physical parameters of the precipitation. It is always interesting to see real world data used for these types of studies as opposed to just models and the results are interesting and worth publishing. There are a few challenges with the current manuscript (which are listed in more detail in the specific comments below). The first one is that there is not enough of a connection made to the existing literature on the topic. Also the organization of the study is rather unconventional in that there is no "Data and methodology" section but the content that is supposed to go there is scattered in other sections. That is not per se a problem but unfortunately some of the data and methodology information is either missing or not presented with enough detail. Therefore I recommend to possibly accept the manuscript for publication after major revisions.

Specific comments:

1. Line 59: Incomplete citation.

2. Paragraph starting line 58: The authors only cite the ITU A-R relationship but there is a lot more research and literature available on this topic in the radar context. It is important that the authors link their research with the existing literature and make clear which unsolved (!) questions they are trying to answer.

3. The organization of the paper is a bit unconventional. Generally, in a research article the reader expects a section on the data and the methodology and general set up (maps, etc.) after the introduction. This information is spread out over several sections which in principle works fine. However, to readers who don't read the whole paper from the beginning to the end (i.e. most readers!) it will be easier to find what they are looking for in a more traditional organization. This is not a requirement but rather a suggestion on how to make the content more accessible.

4. Section 2.3: The section is entitled "Measurement Location" but also gives information about the time of measurement and the stratiform/convective partitioning. This information will be very hard to find under the current title. Please consider re-organizing the sections as mentioned in the previous comment. Also, what are IAP and CAS?

5. More on Section 2.3: Please give a more detailed description of the time frame. Or was it really January 1st 2017 to December 31st 2018? How were the 192 hours of precipitation distributed over the time frame?

6. Even more on Section 2.3: Please give more detail on the stratiform/convective partitioning. How exactly was that derived from the radar data? What algorithm did the radar use? What was the location of the radar compared to the location of the distrometer? Did you just use the classification of the radar data point closest to the distrometer?

7. Section 2.1 mentions only one distrometer but two distrometer locations are shown in Fig. 3. Were there two or one? Also, please consider showing Figure 3 earlier in the paper.

8. Figure 2: How much data was in each of the categories? Also, I think it would be easier to compare the different curves if they were plotted in just one panel but with different colors.

9. Line 148: Figure 3 shows something different.

10. Line 152: I don't believe it is possible to know whether the precipitation is convective or stratiform from the DSDs. This statement should be removed.

11. Starting line 236: Why are the authors talking about light, moderate, and heavy rain here? They only show stratiform and convective rain. While it is generally true that convective rain is heavier than stratiform rain, light/heavy vs stratiform/convective rain are still two very different concepts and should not be mixed.

12. Figure 6 is very confusing and it is not clear to me which equations and coefficients were used to calculate which curves. E.g. in the figure caption it says that panel (a) shows attenuation but on the y axis it says that it is the received signal. Which equation was used to calculate the lines in (b)? Which one for the curves in (c)?

13. It would be interesting to see more cases similar to that presented in Figure 6. It is difficult to draw conclusions from just one case.

Author Response

We would like to thank the reviewer for your valuable comments and feedback. Our responses are attached in the document.

Author Response File: Author Response.pdf

Reviewer 3 Report

Review

The authors aim to increase the accuracy in assessing the effect of raindrops on the loss of acoustic signal propagation at frequencies of 25 and 38 GHz. The specific attenuation is usually calculated from the rain rate using a power law that was developed from fitting the curve to power coefficients. The reason for the loss of attenuation is many-times repeated reflection from the air-water interface and also the general scattering of the sound wave power on curves surfaces of raindrops.

Therefore, the analysis of the size, shape of the drops, as well as the speed of rain, is an important subject of research to find out the sound loss most accurately. Analysis and modeling of the distribution of raindrops or drizzles is carried out in numerous scientific literature, for example, see the links at the end of the review. Are generally accepted facts:

(1) the size distribution of droplets in rain of varying intensity follows the Marshall-Palmer distribution:

(2) the number droplet’s distribution in the initial stage of rain (drizzle) is described by the gamma distribution function.

These concepts are not only not compared with the experimental results obtained, but are not even mentioned by the authors. At the same time, the size distributions obtained by the authors are shown in Figure 2. The first graph of this figure in its initial part vaguely resembles the gamma distribution characteristic of drizzle, the right part of this graph is questionable due to too large drops for drizzle-type precipitation. The last 2 graphs of this group, 2(e) and 2(f), seem to be wrong, since there can be no drops with diameters of 8 mm in the rain due to their splitting into two. The reason for the impossibility of too large a drop is that the mass of a huge drop exceeds the surface tension forces , this is a well-known fact established earlier. Probably section 2 should be heavily redone.

The authors carried out direct measurements of losses in the sound transmission line under conditions of various rains. This is a positive contribution to the study of the question raised, but it is not sufficiently detailed. Also, the interpretation is not sufficient. This work does not solve the goal and does not even create an algorithm for improving the estimation of losses in variable weather conditions. Therefore, the conclusions do not correspond to paper’s title and theme.

In general, this article should undergo a major revision before being published in journal.

• Fielding M.D., J.C. Chiu, R.J. Hogan, G. Feingold, E. Eloranta, E.J. O’Connor, and M.P. Cadeddu. Joint retrievals of cloud and drizzle in marine boundary layer clouds using ground-based radar, lidar and zenith radiances . Meas. Tech. Discuss., 8, 1833–1889, 2015. doi:10.5194/amtd-8-1833-2015
• MORRISON HUGH AND GRABOWSKI WOJCIECH W. 2007. Comparison of Bulk and Bin Warm-Rain Microphysics Models Using a Kinematic Framework. AMS j. of Atmospheric Science, 64, 2839- 2861. DOI: 10.1175/JAS398
• Miles N.L., Verlinde J., Clothiaux E. 2000. Cloud Droplet Size Distributions in Low-Level Stratiform Clouds/ J Atmosp. Science AMS, v.57, 295- 311.
• Marshall J.S., Palmer W. M. K. The distribution of rain drops with size. J Meteor, 1948, 5: 165–166.

Author Response

We would like to thank the reviewer's very valuable suggestions and corrections. Our replies are summarized in the attached document.

Author Response File: Author Response.pdf