Alert and Flight Initiation Distances of the Coot in Response to Drones

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript describes the results of experiments testing the coot's reaction to the appearance of a drone. The entire manuscript is prepared carelessly and needs to be improved and organized. The order of literature citation is inappropriate, the order of arrangement and description of figures is wrong, the structure of chapter results is inappropriate.  The methodology is inadequately described, and the mixed model used, in which a continuous variable is explained by qualitative variables, is questionable. It may falsify the results obtained. The large disparity between sample sizes is also objectionable. 

 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Quality of the English language needs improvement

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper examines the response of common coots to drones descending at variable altitudes. Coots ignored the presence of drones at an altitude of 50 metres. Lower altitudes triggered alarm and eventually flight. The 50m altitude is a known threshold reported in other studies, some of which are cited in the manuscript. Admittedly, it is still valuable to verify that 50 m is a safe distance in a new bird species (Common Coot). The results for the short distances of alert (7 m) and flight (16 m) are remarkable. These distances are unusually short for wild birds. As this is a striking result, the text should explain the habituation mechanisms that may have reduced the distances. This is an aspect that is not satisfactorily addressed in the manuscript. For example, the possibility that the birds became habituated to the proximity of the drone during the experiment is suggested as an explanation in the methods section (line 127), but was not considered in the previous flights (lines 111-113). These flights may have served to habituate the birds. Another factor in the short alert and flight distance was the slow speed during the vertical descent (1 m/s, line 121). Undoubtedly, the alert and flight distance would be greater if the descent had been made at the speed of an aerial predator. Studies on the flight distance of garden birds show that the variables that determine the flight distance are the direction of the pedestrian and its speed. Here, the speed was not variable but constant: 1 m/s.

 

Alerting and flight distances vary as a function of vertical descent speed. Very slow speeds favour short distances because birds can get used to the object as it comes down. Very fast speeds also favour short distances because birds cannot react quickly and move when the object is close. Therefore, there must be an intermediate speed of descent that triggers maximum distances. This function is unknown and probably non-linear:

 

 

 The present study contributes by experimenting with a descending speed of 1m/s. However, a single data point is not sufficient to understand the function between alert or flight distances and vertical descent speed, which must be the aim of studies such as this.  This issue needs to be addressed in the revised manuscript. Overall, the habituation of the subjects to the harmless approach of the flying object is not convincingly explained in the manuscript.

 

Also, secondary methodological aspects are not satisfactorily explained in the methodology, such as the air speed in each experimental event or the consequences of the drone model on the results. Let us first talk about the air speed: in laboratory conditions, the horizontal air speed should be zero m/s. In the open air, the air speed is never zero, so the flying object must counteract the force of the air to ensure a vertical descent. The effort made by the machine to counteract the force of the wind is reflected in increased noise and zigzag movements, among other stimuli that favour the animal's alertness and flight. Since the environmental conditions are uncontrollable in each experimental event, it is necessary to devote a few lines in the manuscript to explain how the effects of these influences, which are by definition uncontrollable and unpredictable, have biased the results.

 

The next methodological aspect not considered in the experiments is the drone model. Other studies on the effect of drone distance on flight distance have examined the effect of the drone model (size, noise, etc.). The manuscript did not measure the noise generated by the drone. It is obvious to point out that the quieter the model, the shorter the warning and flight distances should be. Although this is obvious, it is still necessary to measure the influence of the drone noise. Furthermore, the effect will be non-linear, requiring several measurements at different noise levels.

 

The species selected for the study is gregarious outside the breeding season. It is therefore plausible that the birds have become habituated to the drone by observing the response of other individuals. Learning, and thus habituation to a stimulus, occurs either directly or by observing other subjects. This is another aspect not discussed in the manuscript, although it is likely to have determined the outcome.

 

Finally, the manuscript should include an explanation of the relationship between the selected bird species and the anthropic environments. What was the human presence in the study areas? Were you the first to fly drones over the coot sites? Can you be sure that the animals you observed did not come into contact with recreational drones at any of the eight sites you visited? Perhaps drone flying is banned in some of the selected sites, but not in others. Perhaps it can be shown that coots are familiar with drones, even noisier ones than those used in your study.

 

 

Minor comments

 

Page 2, line 90. Typo: delete ':'

 

Page 3, lines 102-107. Drone choice. Please add the noise (dB) produced by the drone.

 

Page 4, lines 146-147. Location was defined as a random factor because FIDs were repeatedly recorded in only eight locations. Please include a table, either in the main text or in the supplementary material, with the basic statistics of the variables (see Table 1) per location. Report these statistics: N, mean, SE or SD and median. This request is based on the need to assess whether the results of this paper are driven by observations from a fraction of the sites. For example, most of the observations could have been made at half of the sites, and therefore the fieldwork would have been based on visits to four sites, even though there were eight sites in total. A table showing the number of observations (N) at each of the eight sites is important. Similarly, the means and medians of the variables could be the result of particular behaviour at a few sites rather than a stable pattern across all sites. In addition, it would be useful to add the variable 'site' to Tables A1 and A2 so that this random factor has its statistical value.

 

Page 4, lines 150-151. Please indicate the dates of the year when the experimental flights were carried out. The behaviour of coots varies according to the season of the breeding cycle (territorial birds) vs. wintering (gregarious birds). Indicate if there were any cases of coots with chicks.

 

Page 6, line 181. Delete 'different': The influences of factors on flight initiation distance (FID). If more than one factor exists, they will necessarily differ.

 

Page 7, line 197. A gradual and continuous vertical approach could promote habituation and thus reduce distances. Please include an explanation of this point in the manuscript.

 

Page 8, lines 262 - 275. Conclusions. The length of the conclusions should be reduced. For example:

1.     The first sentence describes the work done. It does not describe the conclusion drawn from that work. (lines 262-263).

2.     The third sentence describes a fact that was known before the work was done: drones are more easily detected by waterfowl than by land birds (lines 265-268). This is a sentence that should be in the methods but not in the conclusions. Or it should be moved to the introduction, if one wants to turn this third sentence into a prediction of the work, the results of which will confirm or refute it. This would be a trivial prediction, because a negative result is very unlikely. But quantifying the distance is no longer a trivial task. It is also useful to compare it with the distance of the same bird species when it is on land. This second distance should be shorter, shouldn't it?

3.     The fourth sentence does not seem to be a conclusion of this work, but a general guideline independent of the results (lines 267-268). It should accompany the third sentence as it moves to the Introduction, or perhaps to another more appropriate place in the Discussion. For example, on line 255, before the fourth sentence of that paragraph.

4.     The sixth sentence describes the decisions made in planning the fieldwork (lines 270-273). It is therefore a sentence that belongs in the methods section. If it is a decision made after the fieldwork has been completed (e.g. flight trials in mixed species flocks should have been excluded from the analyses), then the appropriate place for this sentence is in the subsection "2.5 Analysis" (l. 142).

5.     The seventh sentence (lines 273-275) states the obvious: new studies will provide more information about the impact of drones on birds. And with more information, the use of drones can be better managed. This sentence is not so much a conclusion of the paper as a generalisation independent of the present paper. Therefore, it should not be included in this concluding paragraph. Replace it or simply remove it from the manuscript.

The conclusion paragraph, after incorporating the suggested changes, would read as follows: "In conclusion, common coots showed an earlier disturbance response to drone flight experiments when they had previously detected the drone. In contrast to previous studies, the present study placed more emphasis on bird AD than FID, potentially providing more reliable evidence for the management of drone use in ornithology". If you wish to enrich the conclusions, add further conclusions as long as they are closely derived from the results presented in the Tables and Figures of the manuscript. However, make sure that the new sentences are not the results themselves.

 

Page 9, lines 279 -284. Table A. Ensure that there are no typographical errors in the format of Tables A1 and A2. Some examples:

1.     None of the p-values are shown in bold. The custom of highlighting p-values in bold is applied when the numerical value below 0.05 is not considered statistically significant in multivariate models or multiple correlations between associated or covarying variables. For example, the three behaviours included in the GLMM model are mutually exclusive, so their frequencies would covary, perhaps inversely or directly. It would be understandable if you had taken this into account and applied the Bonferroni criterion to the p-value. Consider this possibility, and if you apply the Bonferroni criterion to the p-values, the bold emphasis would be useful.

2.     Delete the numbers '1' to '82'. Alternatively, explain what these numbers stand for

3.     Explain what "See Yes" means. Perhaps it should be written as "See (Yes)"

 

 

Page 9, Line 279. The legend for Table A1 should be more specific. For example, it could read: "Coefficients of the variables included in the prediction model of the Common Coot alarm distance (AD)". Since the factors are by definition different, it is recommended to improve the expression "under different factors". Perhaps "selected factors", "relevant factors", etc.

 

Page 9, line 289. The use of artificial intelligence (AI) in manuscript generation. It is noticeable that no AI programme was used. I myself have to translate the revision into English using AI programmes (DeepL translator, free version) and improve the style by rewriting the translation using other AI (DeepL Write, free version, and Grammarly, paid subscription). I would agree that I did not use AIs to 'create' the review either, but this is only true in a loose sense. This is not a major issue, as AI applied to scientific texts is a "hot topic" in fierce evolution, so we are in a learning phase about the acceptable uses of AI vs. other uses that we should avoid.  As a sign of transparency in the statement, it may be more appropriate to add a second sentence to recognise the use of IAs in post-translation editing and sentence improvement. Perhaps in the future this use of IAs will be included in the editorial standards as acceptable. Perhaps it is unnecessary to go into detail, as it is obvious that it is necessary to use AIs for translators and proofreaders. In the past, these aids were printed in dictionaries and manuals; today they are available in IA programs. To summarise: consider adding to this sentence the name of the AIs used to improve the manuscript, even if they were only used to improve the style and the English translation. This comment is a suggestion, not an imperative.

 

Page 11, line 304. Perhaps the names of "Da" and "Huo" belongs to the same human being. If this is the case, it should be written as "Da Huo, Qi Sun and Taijun Zuo".

 

Page 12, line 373. Typo: Please add the letter 'r' to the name Mølle. His name is Møller.

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript has been corrected. 

Please remove first line in Results. In the Disscussion, replace the word population with group or flock.  

Comments for author File: Comments.pdf

Author Response

请看附件

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The new version of the manuscript has improved the shortcomings in basic information and other minor problems. A good example of this is the determination of the authors to measure the drone noise, and I commend them for their diligence in offering a solution.

It is odd that there has been a change in the sample size. It is unusual for data to be subject to change in the course of peer review. The analysis and the figures are updated, but not the data itself. Please take this seriously and provide a satisfactory response to the reasons for the change in the cover letter. What are the reasons for adding more cases? Would the authors add more cases again if they were forced to do a third review? Maybe the other reviewers pointed out the need to increase the sample. Perhaps the authors needed to increase the sample size between the first and second review, but why?

Other problems with the study are not easy to resolve. The authors acknowledge these limitations, so the manuscript remains in the best possible condition. Apart from a few typographical errors, which are detailed below, I have no other important comments or improvements to suggest.


Minor comments

Page 1, line 8. Typo: delete 'Correspondence'. The word is written twice.

Page 1, line 8. Typo: Check the size of the letters in brackets.

Page 3, Figure 1. Authors, reviewers and editors have a need for guidance on the assessment of study area maps when boundaries are included. For example, the geopolitical classification of a large number of small islands (Fig. 1) is the subject of enthusiastic debate in international forums.   If a reviewer is not to examine a study area map, this should be stated in the reviewer's guide. Note: If this is already the case, please accept my apologies.  I am writing these lines to warn about this problem that no one seems to notice: do we ignore the map, or should we check its accuracy? But... according to what criteria?  In the future, historians of science will write interesting reviews on how geopolitics have shaped the maps published in scientific journals... of biology!

Page 3, Drone noise. Perhaps where it says '(Fig. 1)' it should say '(Table 1)'

Page 15. Delete this page, as it is empty

Author Response

Please check the attachment. Thank you.

Author Response File: Author Response.pdf