All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Dear Dr. Szabolcs,
I am pleased to inform you that your manuscript has been accepted for publication in PeerJ! Congratulations!
Please note that one of the reviewers still suggested some very minor changes that can be performed during the proofreading stage of your manuscript. Please do not forget to change them, since, as indicated by the reviewer, they will significantly improve the reading of the indicated portions of your text.
Sincerely,
Daniel Silva
[# PeerJ Staff Note - this decision was reviewed and approved by Dezene Huber, a PeerJ Section Editor covering this Section #]
no comment
no comment
no comment
no comment
No comment
No comment
No comment
The authors have done a great job! The paper quality has improved a lot!
no comment
no comment
no comment
The following slight grammatical changes could improve the manuscript. Otherwise, the paper seems very well structured, understandable, & thorough; an excellent contribution to a highly understudied biological topic.
* line 58: end sentence with “mortality” & begin next with “However”…
* line 274: insert “of” between “because” and “the”
* line 373: comma after “view”
* line 377: comma after “level”
* line 384: comma after “mortality”
Dear Dr. Szabolcs,
After this new review round, three of the reviewers accepted your manuscript. Still, two other reviewers required minor reviews. Please note that although one of the reviewers indicated the acceptance of the manuscript, they also required minor changes regarding statistical analyses/descriptions and also in the discussion. Also, more than one reviewer indicated that text editing/prooferading is still required to improve the quality of the written English of your manuscript. Therefore, a further review by a native English speaker and/or a third-party company is necessary to guarantee that your text before it is accepted for publication in PeerJ.
Sincerely,
Daniel Silva
The manuscript has undergone significant improvements, that increased its readability and quality. However, I think the authors overlooked the major language revision suggested by several reviewers, which could have furhter enhanced the overall quality of this study.
I only have a couple of minor, detailed comments that are language related (see below).
no comment
no comment
Line 169: I would write as "snake movement might be affected as floods (...)"
Line 207: if you added the package above in the text, there is no need to do it again here.
Line 297: A few kms? Please specify.
Line 309: remove "that".
Line 388: Add "these" before "results".
Line 408: hibernacula.
The authors have done a great job in these revisions, and now the manuscript is much more fluid and concise, with a clearer explanation of what was done and the results obtained. I suggest a few more changes that are very important since the authors are working solely with observed data. Since no corrections for sampling errors have been made, which is already recognized in road ecology literature, in my point of view it is necessary to mention it explicitly in the manuscript.
No comment
As I mentioned in the first review, it is important to emphasize that the mortality estimates presented refer to the observed carcasses by the observer who conducted the survey. Since another reviewer suggested adding a mortality rate, I emphasize that this rate should be referred to as the observed rate, given that even the authors acknowledge factors that influence carcass detection, which means these presented mortality rates may not accurately represent the actual roadkill number of snakes. Therefore, I suggest referring to it as "road OBSERVED mortality rate" throughout the manuscript.
methods section:
- field methods: I reinforce that it should be mentioned that the authors are aware of sampling errors that influence detection (in a concise way), something like: "We estimated snake road mortality by counting the number of carcasses as an index of roadkill incidents, due to sampling errors not considered in the estimated rates..."
- analyses: emphasize that the calculated rates are observed rates.
Discussion section:
I didn't understand the authors' intention in mentioning in the discussion (L276-277) that they monitored a larger area. Out of curiosity, I looked at the results from Serbia and the authors present an estimated rate per day/km. Therefore, I suggest the authors also present the observed rate per km for better comparison. Considering the rate per km, the observed rate of Natrix natrix found by the authors is much lower than the estimated rate in Serbia. It seems somewhat contradictory the first result indicates a high snake mortality (L267-268) when in subsequent sentences, it is said to be a lower rate (L294-295). I suggest that the authors rewrite these sentences.
No comment.
No comment.
No comment.
scientific wording could be enhanced somewhat (see notes)
general structure of narrative could be changed from personal to more ambiguous (see notes)
n/a design seems sound
sound validity & stats, conclusions, relevant data
I think the authors have done a great reviewing work of this manuscript. References have been updated, the background looks sufficient, and the writing is clear.
The authors have included the estimation of roadkill rates, which is a great addition to this work, however, they did not include the units. In the reference you cite, roadkill rates are calculated as individuals/km/year. What are your mortality units? This is my only suggestion.
No comment
I just recommend authors to mention the units used for reporting roadkill rates. I think they forgot to include this.
Dear Dr. Szabolcs,
After this new review round, I believe your manuscript is almost ready for acceptance. Five reviewers have assessed the manuscripts. Four indicated the need for minor reviews. Still, the fifth reviewer indicated the need for major revisions. There are text portions that still need to be improved (a written English review is advisable), along with the preference for newer manuscripts than older ones.
Please note that several reviewers provided attached PDF files to help you out during the review process.
Sincerely,
Daniel Silva
**Language Note:** The Academic Editor has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
no comment
no comment
no comment
The authors have done an excellent job of revision. The article now has a more fluid readability, and the objectives, methods, and results are more thoroughly explored throughout the paper.
I think I mentioned this before, and I might sound quite insistent, but I believe that Question 1 in the objective paragraph should not be treated as an objective of the study (Which species are killed on the roads?). I understand it might have been considered an objective at the beginning of your study, but it doesn't need to be included in the objectives anymore, especially considering that you did not find many snakes from other species.
From my perspective, it could be solely mentioned in the results section, where the paper already contains the following sentence: "We eventually excluded the latter three species from this study due to low sample sizes," so I don't see the question as a valid objective.
The figures still require attention to certain details.
Figure 3: The bars are displaced. There are 9 groups of bars when there should be 8. It seems the 9th bar belongs to the 1st group.
Figures 3 and 4, I believe that when we look at a figure, we should be able to fully understand it without needing to read the legend. For these graphs, it would be helpful to include an indicator for each species within the graph, so that it is not necessary to refer to the legend to understand which species each graph corresponds to. Also, I suggest changing the y-axis title to "Number of Carcasses" instead of "Number of Individuals."
After reading the response letter, I understood that the author removed the flood regime comparison. However, in the section 'Analyses with environmental drivers of land use, flood regime and traffic volume' it appears that this comparative element is still present. I think this deserves attention to confirm if the comparison was completely removed. I apologize if I misunderstood something.
You mentioned in the results, specifically in the temporal patterns section: 'By visual examination, we identified two peaks in the number of carcasses found in each year (Fig. 3A).' However, I could not discern the data for each year when looking at this figure. It appears that all the collected data is aggregated together.
The authors made significant improvements in the discussion section of the paper.
Minor comment: "bicycle" is misspelled in the second paragraph of the discussion.
This paper has been greatly improved by the work of the authors and the comments made by prior anonymous reviewers. I agree with all of the comments the two prior reviewers offered, and it is evident that the authors took these comments to heart and made the appropriate changes. For this reason, I think very few changes need to be made in order to publish this study.
This paper, in its edited form, is well-written in clear, unambiguous, professional English. Literature is well-reference and relevant and the study is well situated within the context of the field of road ecology. I have some comments (see below) for improving the captions of figures, which are themselves adequate. The overarching findings of spatial and temporal patterns in road mortality are clearly articulated in this revised version.
The use of latent variables (the authors refer to these as LV1-4) is interesting and novel to me in the context of road ecology. The use of GLMMs is an adequate statistical technique. The authors should consider adding a sentence or two in their methods (perhaps in the very short section “Number of road-killed snakes.”) that explains their reasoning for not transforming the data into a continuous field (e.g., kernel density estimates). A continuous measure would be more spatially robust than the current use of summed-morality in a road section, which spatially coarsens the mortality measurements to the extent of road sections, which are 1km in length. In other words; are finer spatial patterns in road mortality being obscured by a spatially coarse mortality measurement? To be clear, I don’t disagree with the authors’ decision to not transform the data, but more clearly stating the reasons behind this decision would benefit the reader.
The findings of this study contribute well to the growing body of road ecology literature that relates long-term mortality datasets to the influence of the surrounding ecology. This study is adequately novel, both in its geographic / species focus, and as well with the statistical techniques employed. The underlying data are robust and statistically sound, and the conclusions (particularly the “Conservation Implications”) are measured and relate directly to the research question posed at the outset.
A few brief comments for revision. words within the square-brackets [ ] are my suggested additions / changes.
L296. Instead of “road traffic influence…” consider: “indicating that risk of mortality was similar regardless of age category and sex on all sections.”
L426. “Alternatively, [the] higher number…”
L506. “… yet fortunately [mortality] was low…”
L511. Remove “fortunately”
L531. “… that these snakes are capable [of occupying] such newly…”
Figures
In the caption of Figure 2, I would suggest rewording “more strong colors” and “more light colors” to either “stronger / lighter” or “more saturated / less saturated”
As well in the caption of Figure 2, change the last sentence to: “The cells that are framed indicate significant correlations.”
For all of the panels in Figure 5, consider removing the grid pattern in the background. I personally find that it adds busy-ness to a figure already very full of information.
see attached pdf.
*
*
*
This is an interesting manuscript analysing mortality patterns of two snake species. Most analyses are commonly carried out at taxonomic class level due to small sample sizes. Here, authors were able to systematically collect enough roadkill data, allowing the analysis of spatial patterns of mortality.
I also encourage authors to update references, which in many cases correspond to papers published more than 15 years ago.
Several parts of the text are not very well written. I recommend improving your writing and language so ideas are presented clearly.
Although I enjoyed reading this document, the way that authors present methods did not make me feel comfortable. Several parts of the methods are not clearly explained, which is a problem if other researchers wish to replicate this approach. This section is not presented with enough details. Then, my most basic suggestion is that authors pay special attention to this so all methods are clearly explained.
Please refer to the document I am attaching to have more details on this.
Dear Dr. Szabolcs,
After this new review round, the reviewers believe your manuscript is worthy of publication after an in-deep review of your study. Most issues of your text is related to the framing of the scientific findings you and your co-authors found. Therefore, please consider reviewing your manuscript and providing a rebuttal letter to inform the reviewers about the changes you implemented. Finally, please consider submitting your study to an in-deep editing and proofreading service in order to improve it.
Sincerely,
Daniel Silva
[# PeerJ Staff Note: The Academic Editor has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title). #]
Overall, this is an interesting paper reporting roadkill for two species of Natrix snakes in an area with two similar but distinctively managed landscapes. It is notable throughout the manuscript that the authors have made an impressive effort to include many variables, but I think that the paper, as it is, is rather confusing and hard to follow. It undoubtedly still needs some work to make it publishable.
I have some doubts about the validity of some variables used, namely traffic density.
Methods section in the manuscript is very extensive and hard to follow. Please find attached some suggestions in how to improve this section.
The findings are usefull to understand snake ecology and behaviour on roads. Also, to conservationists and road agencies to implement mitigation measures in the future.
Justification for the usage of some statistical methods are provided and well explained and conclusions make ecological sense and are based on literature.
Although I found the results very interesting (deserving publication), I struggled during the review process and I read the manuscript several times, especially to attempt to fill up the four sections. This is mainly due, to my humble opinion, to an initial problem of framing as explained in this review
The story is based on the comparison between “active versus protected floodplains”. The term “protected” is ambiguous in the context of this study. Human’s activities or installations are protected by dams or embankments, not the floodplain per se. More importantly, this study is not designed to assess the (negative) impact of constructions made to control flooding. However, the authors collected very nice data sets with considerable potential for conservation actions.
This can be illustrated with comments on the four main results that are presented in the first paragraph of the discussion.
1) “…road mortality was unexpectedly high and rather similar across years and months”
2) “…most importantly, mortality was higher in the flooded than in the flood-protected area…”
3) “…differences in mortality across road sections…”
4)”…these differences arose because road mortality of snakes was higher in sections surrounded by more natural habitat types and lower in sections surrounded by urban or agricultural areas”
Comments:
1) Why mortality was unexpectedly high? Natricine snakes are known to occur in high density in many places. Figures 3-5 (and associated statistics) reveal strong differences among years and months (perhaps the authors aimed to emphasize that the patterns were consistent across years? Even so this seems to not be the case). There is no need to state that mortality was strangely elevated and relatively invariable across seasons and years, fluctuations do not diminish the value of this work.
2) This is likely the main problem. Although the authors stated that habitats were not different between flooded than in the flood-protected area (with supporting statistics, but the input data were selected to please the expected outcome), the roads respectively surveyed in these two areas cannot be directly compared. For example, there are major differences in terms of distance between the roads and the riverbeds with enormous possible impact on road mortality for semi-aquatic species; further, a huge partly natural swamp area (between the Brodog Tisza Rivers) occurs only in the active flooded area, another strong potential source of contrasts in terms of population density or dispersal or seasonal migration.
3) This is a crucial result. Note that relative mortality risk cannot be estimated without information about the respective size of the populations and their spatial distribution. Grass snakes might be more abundant than dice snakes; or dice snakes may remain nearby rivers. However, certain sections of the roads clearly represent a threat for the snakes and these segments specifically deserve conservation actions.
4) Another crucial result that might be combined with the previous one.
Overall, I suggest to shift away from the “active versus protected floodplains” problematic to consider the road section one. By focusing on the road section issue, the article can be drastically shortened without loss of information.
Beside this major concern, several minor comments are proposed below.
The hypothesis linking mortality with the level of domestication of the floodplain is not clear. How can we tease apart the effect of road embankment from habitat artificialisation? Both can cause a decrease of road mortality, directly (embankment) or indirectly (habitat loss).
L 68: “ors” instead of “or”?
L74 : consider deleting « are highly mobile », indeed compared to birds, mammals and many other organisms snakes are not particularly mobile.
Figure 1 (map) very difficult to read for colorblind person.
The sections surveyed (36+22) should be clearly indicated in Figure 1 (at least a few crucial ones).
Kindler et al. 2017 cannot be used to infer that a single population occurs in the study zone. A lack of genetic barrier does not mean that homogeneous populations are equally distributed in a given sampling zone. A limited number of dispersers can maintain the flow of alleles across populations characterized by contrasted densities.
Age categorization are very crude but fine for the purpose of this study.
Y scale should be standardized in the figures 3, 4 & 5, even if some patterns will be flattened.
Consider combining Figures 3 & 4A& 4B. Each month and each year, numbers of roadkill of known ages and unknown age piled up in a single bar.
Were numbers corrected by sampling effort (number of surveys?).
NA, this is an empirical study. The leading author surveyed roads during three years and collected the data. Bicycle riding is very appropriate for this kind of survey.
The results are based on a huge amount of data, the patterns are clear and the suggestions to mitigate road mortality sound (some road sections must be fitted with culvert for example).
With limited efforts (mostly deleting useless sections and simplifying the manuscript) this study will provide fine scientific contribution with great conservation potential. I was frustrated to not read more discussion about age and sex effects. For example, in September October post-hatching dispersal can cause strong neonate mortality, but not in this study.
The article is interesting and has the potential to contribute to the knowledge about the impact of roadkill on the assessed areas. However, some terms, such as "roadkill," need to be consistently written throughout the text. The introduction of the article is good and effectively addresses the issue at hand, but the last paragraph needs some improvements.
The figures require attention to certain details and higher resolutions. Figures 1 and 2: I suggest using different colors to enhance the visibility of important elements. For the graph figures, I recommend providing titles for both axes. It would also be beneficial to establish a consistent pattern within the graphs (Figs. 3/5), such as presenting the same species in different environments and then the other species, or both species in the same environment and then the different environment.
Regarding the supplementary materials, I suggest reviewing them (or revising the text) as I noticed a discrepancy in the number of N. tessellata records compared to what was described in the results.
One of the main points to be revised in the study is that the analyses and the presented results need to be aligned with the objectives and specific questions of the study. Although I believe that all the questions have been answered, it is crucial to restructure the way the objectives, analyses, and results are presented, making it clearer and more concise how each question is being addressed and what the findings are. The authors conducted some analyses to understand aspects that were not explicitly stated in the objectives of the study, although they were related - or could even be implied. However, given the extensive data analysis performed, I believe that the study will be improved a lot by this restructuring, making it evident which question is intended to be answered with each analysis (or set of analyses).
I would like to know why these specific temporal and spatial analyses were chosen. There are temporal and spatial analyses commonly used in the literature that I would expect to find when it was mentioned that these patterns would be identified, such as the Rayleigh or Herman-Rasson tests for the temporal component (Landler et al., 2018) and aggregation analyses for the spatial component (Gunson and Teixeira, 2015; Coelho et al. 2014).
Landler, L., Ruxton, G.D. & Malkemper, E.P. Circular data in biology: advice for effectively implementing statistical procedures. Behav Ecol Sociobiol 72, 128 (2018). https://doi.org/10.1007/s00265-018-2538-y
Gunson K, Teixeira FZ. Road–wildlife mitigation planning can be improved by identifying the patterns and processes associated with wildlife‐vehicle collisions. Handbook of road ecology. 2015 Apr 1:101-9.
Coelho, A.V.P., I.P. Coelho, F.Z. Teixeira and A. Kindel. 2014. Siriema: Road Mortality Software. User’s Guide V.2.0. NERRF, UFRGS, Porto Alegre. Available at http://nerf-ufrgs.herokuapp.com/siriema.
One issue that I would like to draw attention to and that needs to be revised and discussed more by the authors is regarding the comparisons made on roadkill rates. Basing comparisons, such as one area having a higher or lower number of roadkill incidents, only on observed data is a critical point in this study. We already know that there are errors associated with any kind of sampling, and there are two important errors, well discussed in road ecology, that should be considered in the study (Barrientos et al., 2018). Hence, it is important to clearly state in the study that what is being presented is a comparison of observed data without considering any sampling errors and that the mentioned values represent observed carcass by the observers, but it does not imply that they encompass the real number of animals that died in the region during the study period. Therefore, caution must be exercised not to assert that one road/section has more roadkill incidents than another, but rather that researchers recorded more carcasses in those locations. Although the authors address one of the errors - carcass persistence - in the discussion, I suggest mentioning the error of observer efficiency as well, restructuring the way the results are presented and highlighting that they are observed results and not a more accurate estimation of reality.
Barrientos R, Martins RC, Ascensão F, D'Amico M, Moreira F, Borda-de-Água L. A review of searcher efficiency and carcass persistence in infrastructure-driven mortality assessment studies. Biological conservation. 2018 Jun 1;222:146-53.
I left some comments with doubts and suggestions in the attached pdf to help to improve the manuscript.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.