The Migration of the Eurasian Woodcock (Scolopax rusticola L.) in the Carpathian Basin at the Turn of the 19–20th Centuries
Michele Sorrenti
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper could have an historical value, but there are a lot of parts incomplete, in introduction and methods.
Line 101 to 104: It's not clear how the birds are detected as migratory or wintering in the area. It is not clear what controls were done in January.
Table 2: See above, it is not clear if and what observation were done in January.
Line 115 to 120: The choice to eliminate data before 10 February and after 30 April seem arbitrary. How could you separate possibile wintering birds in all February from migrating ones? That means that observation in January are present?
Line 284 to 300: it should be described precisely with ring and satellite telemetry data the origin of birds passing through Hungary.
Line 359 to 366: very generic description, that do not focus on the species, even it is well studied. A publication from Estonia is missing: Riho Marja et al. 2022. The Eurasian woodcock (Scolopax rusticola) is arriving in Estonia earlier than hundred years ago. Hirundo.
Author Response
Line 101 to 104: It's not clear how the birds are detected as migratory or wintering in the area.
Please see our comments to the Editor. We have added the necessary information to the text.
It is not clear what controls were done in January.Table 2: See above, it is not clear if and what observation were done in January.
We believe that these comments are not relevant in the context of our study because it is not about the number of observations in January, but the number of days since 1 January. This is methodological information. The comparison is based on the number of days since the base date (1 January), so the dates have been converted in this way, i.e. the number of days since 1 January.
Line 115 to 120: The choice to eliminate data before 10 February and after 30 April seem arbitrary. How could you separate possibile wintering birds in all February from migrating ones?
Please see our comments to the Editor. We have added the necessary information to the text. February is/was the migration period of the species in the study region.
That means that observation in January are present?
We deem this comment as irrelevant because data used are from 10 February, as clearly justified in the text, with reference to previous studies (Spina & Volponi 2008, Kullberg et al. 2015).
Line 284 to 300: it should be described precisely with ring and satellite telemetry data the origin of birds passing through Hungary.
All of these information are given in the Introduction.
Line 359 to 366: very generic description, that do not focus on the species, even it is well studied. A publication from Estonia is missing: Riho Marja et al. 2022. The Eurasian woodcock (Scolopax rusticola) is arriving in Estonia earlier than hundred years ago. Hirundo
Thank you for this valuable suggestion, we were unaware of this paper; hence, we have added more information to this paragraph.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe study provides an analysis of Woodcock occurences from 1894 to 1926 based on annual reports of the Hungarian Ornithological Centre. The authors try to attribute annual variations in arrival to different environmental conditions including any altitude/region effects.
Although i think that the part regarding altitudinal effects is solid and makes sense (migratory birds are seen earlier in areas at lower elevations), I cannot say the same for the climatic variable part. Much more details are needed in the methods to actually evaluate such effects and much caution is needed before attributing the results to large scale events as climate change. I would therefore suggest the authors to re-evaluate their methods and findings and provide further information to support the conclusions, according to the comments below:
L.18,27 etc. You examine the patterns of birds that migrate through Hungary, that is that they use the country as a stopover. So the decision a bird makes is whether stop or not. So in the abstract and elsewhere you cannot satte that migration "started". All these statements need to be rephrased.
L.39 species name should be in italics
L.42 no need for different paragraph
L.53 why is pairs in quotes?
L.54 I think the most proper terminology is "broad-front"
L.55 ...the migration route "to the wintering grounds" ....
L.213 and Figure 1. Why is migration phenology reported in 10-day intervals? I think a more classic approach would allow for comparisons with other regions. That is use daily numbers and calculate median date (peak passage) and interquartile interval of the migration passage.
L.214 arrival on 19 March? Maybe you mean February?
Figure 4 i would add an error bar with the sd for each point to help understand variation in each location. Also are the regions listed in some order? (i suppose altitudinal?)
L.235 VIF results are needed to see which variables are correlated and the strength (VIF values).
L.240 I don't get this. The most appropriate thing to do in cases of such high multicollinearity is to remove some of the variables..Or reshape them. For example a "trick" you could use is to substract daily max temp from daily min temp and name this variable daily temp range. This way you could avoid/lower multicollinearity.
L.269 How did you reach this conclusion? Multicollinearity can surely affect GAMs too!
L. 273 I would like to see the effect plots too.
L.276-277 I dont get what this means. You include year as a smooth term so what is this about? Also are you sure about this statement? To my knowledge edf statistic reflects the degree of non-linearity of a curve. So an edf > 2 implies a highly non-linear relationship. That is why effects plots are needed in order to inspect the strength of the relationships.
L.289 arrival not beginning, since birds pass from Hungary. See also first comment above.
L.292 there is nothing clear about this, you dont provide such evidence so this is rather speculative. Also what weather? Temerature, precipitation? You dont provide any plots to confirm this. Please draw a gam effect plot including both year and climating variable of interest to see what happens.
L315-318 but you don't analyse wind effects whatsoever.
L318 I cannot assess these effects based on the table only. Please provide effect plots
Author Response
The study provides an analysis of Woodcock occurences from 1894 to 1926 based on annual reports of the Hungarian Ornithological Centre. The authors try to attribute annual variations in arrival to different environmental conditions including any altitude/region effects.
Although i think that the part regarding altitudinal effects is solid and makes sense (migratory birds are seen earlier in areas at lower elevations), I cannot say the same for the climatic variable part. Much more details are needed in the methods to actually evaluate such effects and much caution is needed before attributing the results to large scale events as climate change. I would therefore suggest the authors to re-evaluate their methods and findings and provide further information to support the conclusions, according to the comments below:
L.18,27 etc. You examine the patterns of birds that migrate through Hungary, that is that they use the country as a stopover. So the decision a bird makes is whether stop or not. So in the abstract and elsewhere you cannot satte that migration "started". All these statements need to be rephrased.
Thank you for this suggestion. We have changed the wording throughout the text.
L.39 species name should be in italics
Corrected.
L.42 no need for different paragraph
Corrected.
L.53 why is pairs in quotes?
The species has a promiscuous reproductive biology, so it cannot nest in pairs, and this description is misleading in terms of the number of birds. It is now written (formulated/worded) in this way in some ornithological literature, but there are still some places where the size of the population is incorrectly given in pairs. This is why we used quotes.
L.54 I think the most proper terminology is "broad-front"
Corrected.
L.55 ...the migration route "to the wintering grounds" ....
We have the added the above-mentioned phrase to the text.
L.213 and Figure 1. Why is migration phenology reported in 10-day intervals? I think a more classic approach would allow for comparisons with other regions. That is use daily numbers and calculate median date (peak passage) and interquartile interval of the migration passage.
Peak passage (median date) was calculated by the daily data. Reporting such phenological data in 10-days intervals is common in ornithology. Most sources on Woodcock migration report only the first, second or third decade of the month. This is why we have choosen this way of reporting, making it easier to compare.
L.214 arrival on 19 March? Maybe you mean February?
It is not a typo. This is the calculated value for the median date of the migration.
Figure 4 i would add an error bar with the sd for each point to help understand variation in each location. Also are the regions listed in some order? (i suppose altitudinal?)
We have added to the caption about the order of the regions and changed the graph accordingly. In this figure, the median values are given. Because the median has no standard deviation, only the mean, therefore, we have refrained from plotting the median on the graph.
L.235 VIF results are needed to see which variables are correlated and the strength (VIF values).
VIF statistics have been reported in the manuscript. The predictors did not show multicollinearity after removing the variable “daily mean temperature”. The VIF results indicate that multicollinearity is not an issue any more among the numeric variables (Altitude above sea level: 1.10, Maximum temperature: 1.87, Minimum temperature: 1.86, Precipitation: 1.01, Year: 1.06).
L.240 I don't get this. The most appropriate thing to do in cases of such high multicollinearity is to remove some of the variables..Or reshape them. For example a "trick" you could use is to substract daily max temp from daily min temp and name this variable daily temp range. This way you could avoid/lower multicollinearity.
VIF statistics have been reported in the manuscript. The model is now free from multicollinearity. Reviewer suggested a very good „trick”, however, we opted for testing minimum and maximum temperatures separately to see their effects separately. This way we can make claims about both the minimum and the maximum temperatures as predictors of migratory patterns.
L.269 How did you reach this conclusion? Multicollinearity can surely affect GAMs too!
We removed the variable “daily mean temperature” to remove multicollinearity. We have included the VIF results in the manuscript: „The predictors did not show multicollinearity after removing the variable “daily mean temperature”. After removing average temperature [], the VIF results indicate that multicollinearity is not a significant issue among the numeric variables (Altitude above sea level: 1.10, Maximum temperature: 1.87, Minimum temperature: 1.86, Precipitation: 1.01, Year: 1.06).”
- 273 I would like to see the effect plots too.
We are greatly indebted to Reviewer for this suggestion because without the effects plot the temporal and non-linear dynamics of the predictors cannot be visualized. The effects plots have been added in one single plot (Figure 5). Each and every predictor shows a non-linear dynamics. In addition, we have interpreted each predictor based on the effects plot.
L.276-277 I dont get what this means. You include year as a smooth term so what is this about? Also are you sure about this statement? To my knowledge edf statistic reflects the degree of non-linearity of a curve. So an edf > 2 implies a highly non-linear relationship. That is why effects plots are needed in order to inspect the strength of the relationships.
Reviewer is right in that she/he says that each and every predictor shows a non-linear dynamics. This is the case in our dataset, see effects plot (Figure 5). We are greatly indebted for Reviewer for pointing out the necessity of the effects plot.
L.289 arrival not beginning, since birds pass from Hungary. See also first comment above.
Corrected.
L.292 there is nothing clear about this, you dont provide such evidence so this is rather speculative. Also what weather? Temerature, precipitation? You dont provide any plots to confirm this. Please draw a gam effect plot including both year and climating variable of interest to see what happens.
Effects plot have been included as mentioned above. Reviewer is right in saying that the GAM-table does not show the non-linear dynamics and the relationship between the response/dependent variable („migration days”) and the predictor(s). Each and every predictor displays a non-linear dynamics, as mentioned earlier (see Figure 5).
L315-318 but you don't analyse wind effects whatsoever.
While this is indeed the case, it is worth noting here in the Discussion that other studies have found this factor to be the most important. Accordingly, we reagrd it as important to mention this factor in the Discussion, for which we did not have data.
L318 I cannot assess these effects based on the table only. Please provide effect plots
Effects plot have been included (Figure 5). Reviewer is right in saying that the GAM-table alone does not show the non-linear dynamics (or the directions and the shape of the curves) and the relationship between the response/dependent variable and the predictor(s).
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsReference to satellite radio tracking should be added and discussed.
Author Response
Reference to satellite radio tracking should be added and discussed.
We have added two references (Le Rest et al. 2019 and Schally et al. 2022) to the manuscript in relation to the recent tracking studies on the species’ migration.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors did great work addressing the comments and suggestions made. I believe that the ms is now much more complete and overall improved. I only have a couple of grammatic comments:
L.18 and L.29 change "the migration was" to "migration occurred"
L.53 still using quotes seems rather weird without proper explanation. I am not a woodcock expert, however, BirdLife International and any other source i checked reports population in pairs so I would remove the quotes.
Author Response
L.18 and L.29 change "the migration was" to "migration occurred"
We have changed „the migration was” to „migration occurred”.
L.53 still using quotes seems rather weird without proper explanation. I am not a woodcock expert, however, BirdLife International and any other source i checked reports population in pairs so I would remove the quotes.
We have deleted the quotes. However, it would be better to write “females” instead of pairs because of the reproductive biology of the species. EBBA2 give the European population size for example in “mature individuals”. It is left to the editor's decision.
