Measurement and Analysis of Radar Signals Modulated by the Respiration Movement of Birds
, Jun Yan 1, Deren Li 1, Huiping Hu 2, Deyong Kong 3,*, Wenjing Bao 4 and Shangde Wu 4
Round 1
Reviewer 1 Report
The manuscript describes a topic of some interest to the radar community.
Start with the positives. The context is adequately described and the presentation of real experimental data is always valuable. The English is fine and the manuscript is easy to follow.
Unfortunately, there is major doubt over the technical correctness of the results. The descriptions and explanations are quite simple, which do not allay the serious concerns over the technical accuracy. To begin with, what is the purpose of using low resolution range profiles when the resolution is 150cm? The entire target (duck, dead or alive) is much smaller than this resolution, which means the radar is operating in a point target regime. It is therefore alarming to see fluctuations reported in Figure 4: the plotted range is only 40cm, so the LRRP should only be a single range bin. A similar concern exists for the other set of measurements, the high resolution range profiles (HRRP). For the S-band data, the range resolution is 7.5cm, which is commensurate with the target size. In other words, the target would only span 2-3 range bins, yet the plots in Figure 5 are overly smoothed. The source of this smoothing is not described anywhere; it appears the graphs are taken straight from the network analyser. Upon closer inspection, the peak responses of the HRRP are offset from 11.8m, which is perfectly possible with a live animal’s movements. However, the authors took the backscatter at precisely 11.8m across multiple pulses (assumed, as the authors mention “cross range”, which makes no sense in this context) to analyse the animal’s breathing rate. It is a strange choice, especially since there is a null very near to the 11.8m range in the X-band HRRPs.
There are some minor issues, such as figure labelling, missing details in the text, but they are insignificant compared to the technical results. The most unfortunate part is, this paper’s broad conclusion that a duck’s radar cross section fluctuates much less than 10dB, as reported in the literature, may hold. That would be a valuable contribution. However, unless the major flaws are addressed, it is impossible to recommend acceptance of this manuscript.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The presented article is fascinating.
I have only a few comments, which I hope could improve the article.
-- Measurements involving other ducks/ birds could be useful to exclude individual characteristics of birds.
-- Is the respiration of birds influencing non-laboratory radar signals and how its contribution can be removed?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The reviewer does not recommend beginning the abstract with a negative reference to a previous publication. Rather it is better to focus on the contribution by introducing the conflicting alternatives and then proposing the choice identified in this manuscript.
The subject of live and dead duck respiration would not be culturally appealing to many English-speaking readers. It probably is better to refer to the ducks as specimens.
Line 67 “where m is the mass and ? is the mean density of the bird.
The equation (3) is probably ?(?)=??2(?), where ?(?) is the time varying radius of the equivalent water sphere.
Line 78 “fall” and line 213
Table 1 Please define HRRP, LRRP, S, X, and Ku more precisely for the uninitiated reader.
Line 87 dBsm should be defined as decibels received from a 1 square meter target.
In Figs. 4 and 5, the numbers on the horizontal axis need to be separated.
Line 237 “fewer” should be “few”
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
‘Measurement and analysis of radar signals modulated by the respiration movement of birds’
The paper nicely demonstrate the RCS fluctuations produced by respiration of a duck and show that the strong fluctuating RCS values detected by radar are not caused by respiration, as earlier mentioned in the literature. The authors suggest that this signal is probably caused by wing-beat frequencies as used by many researchers already. This study demonstrates the factor of respiration adds a very limited amount to the strong RCS fluctuations.
Detailed comments:
Ln 41: although all references are fine, it feels a bit odd not to include one of the papers by Bruderer and/or Pennycuick who has worked their life on flight characteristics.
Ln 81: description of the table is missing; table should be self-explanatory
Ln 82/83: this result just falls from the sky, without any reference to table, formula or whatever
Ln 89: SSAT should be SAT ?
Ln 89: reference is for Ku band only
Ln 92: remove ‘which is impossible’;
Ln 135-137: move the statement to the end of the paper (epilogue)
Ln 139: why is the respiration frequency taken from literature? Since the authors focus on detailed measurements of respiration size of the body, the actual respiration could be easily taken from their own data of the living bird. Furthermore, the range is huge 30-50 breaths/minute
Ln 144: remove ‘the number of # x is tracking id’ or move this to the figure caption
Ln 149: How is this value calculated and where is this presented, table, figure? Take the reader by hand
Ln 152: ‘we recorded 30 frames’ is method not result
Ln 163-165: Please give the exact results as mean and standard deviations
Ln 164-165: last sentence is not a result, please remove
Ln 166-176: Please rewrite to this section as results, thus facts and move the rest of the text to the discussion. Also the statement that HRRP and LRRP could measure two different things should be into methods (why you measure two times) and what these results mean (discussion)
Ln 181: Would recommend to talk about the ‘living duck’ and ‘dead duck’ instead of ‘alive duck’ and ‘after death’. Also replace this throughout the text. Furthermore, explain the reader how to read the figure, thus also the repeated cases on the right (?), are these important to show them as legend, instead of writing down in the caption that 30 repeated measures are shown as different colors; what do the three different graphs mean? Different headings or what? Not clear. From the text I assume S- X and Ku-band? Modify the figure and add A, B and C into the figure and explain in the caption. Remove the text and arrows mentioning ‘duck position’, but explain in figure caption. Finally the text along the x-axis means distance? And in what values?
Ln 183: remove ‘of the signals of the duck’
Ln 183-191: rewrite as a result section. The rest of the text move to method section. Please rewrite the method section clearly into why you provided LRRP and HRRP and to measure what exactly.
Ln 188-190: Rewrite as facts: The differences between two peaks (one breath) in S-band (xx ms), X-banx (1400 ms) and Ku-band (yy ms) corresponds to a reparation rate of 42 breaths/minute (i.e. 0.7 Hz). In other words for all three bands and not leaving one out
Ln 192-201: repeating the result from the line above; furthermore move this paragraph to the discussion section.
Ln 206-207: make the figure caption clear as suggested for figure 4.
Ln 210: many weather radars also work in the C-band;
Ln 216-217: figure should be moved to the results section and also referred to in the result section
Ln 220-222: labels along x-axis and y-axis should be clear. Reference to living and dead duck or view should be removed (also in figure 4 and 5) and explained in the figure caption.
Ln 223-235: suggestion to replace this paragraph by the conclusion section ?
Ln 226: add reference
Ln 239: remove ‘famous’
Ln 241: remove ‘SATs’
Ln 247: ‘which is much less than the 10 dB-level’?? of what? Be exact and use the right reference
Ln 247-249: This is already in use for ~50 years in radar aeroecology, see for example Bruderer 2010 Wing-beat characteristics of birds recorded with tracking radar and cine camera. Ibis 272-291 and references therein as:
Bruderer, B. 1969. Zur Registrierung und Interpretation von Echosignaturen an einem 3-cm-Zielverfolgungsradar. Ornithol.Beob. 66: 70–88.
Bruderer, B. 1971. Radarbeobachtungen u¨ber den Fru¨hlingszug im Schweizerischen Mittelland (Ein Beitrag zum Problem der Witterungsabha¨ngigkeit des Vogelzugs). Ornithol. Beob. 68: 89–158.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have clarified many of the doubts raised in the first review. The updated manuscript is much improved. One more minor clarification requiring attention is the last paragraph of Section 3. The sentence “In this experiment, we recorded the data 235 after the duck flaps its wings for a while and we could even hear the loud sound of the duck's breathing, quick and loud; hence, the respiration rate of the duck was much higher than the normal state.” Is not followed by any results or measurements of the elevated respiration rate. Can the authors add more information on this aspect? Apart from this, the manuscript can be accepted.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
