Migratory Movements and Home Ranges of Geographically Distinct Wintering Populations of a Soaring Bird
, J. Brian Davis 1, D. Tommy King 2, L. Wes Burger 1, Bronson K. Strickland 1, Marsha A. Sovada 3, Guiming Wang 1 and Fred L. Cunningham 2Round 1
Reviewer 1 Report
This research article aimed at comparing the movement patterns of American White Pelicans sharing a breeding area but wintering in two geographically distinct regions. The authors used a consistent dataset of GPS-tracking data to separate the breeding population into two different wintering groups and to describe their hourly movements and their home ranges throughout the four biological seasons of the Pelicans. They presented interesting and well-supported conclusions stating that exogenous factors from the wintering area induced differences in the hourly movement patterns and the size of the home range during the winter, but these differences are not carried over to the Pelicans’ common breeding area.
General comments:
This research is very interesting as the results and the conclusions contribute to our understanding of the seasonal variability in migratory behaviors. Moreover, while many studies these days aimed at finding carry-over effects from one season to the other, this study highlights the fact that some behaviors are not necessarily carried over when they are induced by exogenous factors.
The study is well constructed, supported by sufficient and robust data, and the statistical analyses are performed adequately. To me, the introduction presents a great overview of the topic. The discussion is clear, not too lengthy, and adequately responds to the hypothesis with interesting interpretations and conclusions. Nevertheless, the result section was a little more arduous to read. I think small modifications to the figures would help to improve the understanding of the results (see my suggestions below).
Specific comments:
Abstract:
L.14-15. Move the sentence “We hypothesized that…” after “…Southern Gulf of Mexico (GOM)” (L.19).
Introduction:
L.76-77. “few studies have related individual migratory bird movement to human footprint index.”. Cite one or two of these few studies. Moreover, it could be interesting to add a few words about their conclusions.
L.92 + L.97 (and throughout the manuscript). You are talking about Northern and Southern GOM POPULATIONS. Do they really are distinct populations as they all meet in the same breeding area? Unless you have a priori information that distinguishes the individuals from these two wintering areas as different populations, it could be more appropriate to talk about “individuals wintering in different areas”.
L.101-103. “Relative abundance of pelicans […] populations (Figure S1).” Move to the previous paragraph when you present both populations. This will also avoid repetition in the next sentence.
Hypotheses 1 and 2: I suggest presenting hypothesis 2 first, and then hypothesis 1. This way, you start by assessing whether individuals from Southern GOM have shorter hourly movement distances and smaller home ranges. Then you assess if these differences, if any, are still present at the breeding site.
L.107. It is not clear in this sentence what “between seasons” refers to. Only wintering and breeding period? Or you also are considering migratory periods?
L.111-112. Delete this sentence as it only rewords the previous sentence (L.109-111). If you want to clarify your hypothesis, you can simply specify in the previous sentence that “wintering further south” means Southern GOM.
Methods:
L.118-121. I highly recommend adding a map with the boundaries of Northern and Southern GOM in the main text. This will help people not aware of your study system to figure it out easily. Also, you could add to this map the locations of your capture sites (L.133-138).
L.146. Refer to the map I recommended adding.
L.159. Delete “in this study”.
L.166-169. It is not clear how you considered or not the sedentary pelicans that remained at the non-breeding ground year-round. Since they do not migrate to the breeding ground, I suppose you have not considered them in your analyses. Otherwise, it may cause some problems since sedentary individuals could present a different pattern than migratory ones. Have you compared the analyses with and without these sedentary Pelicans? If you did not consider sedentary individuals in your analyses, add a short sentence that states that you ignored them for all further analyses.
L.169-171. Should be in the Statistical analyses section.
L.177-179. Refer to Figure S1.
L.231-232. You refer to models with “year” as a covariate (Table S1-S3), but not to those with “relative abundance index”. Either add a reference to Table 1-3, delete the reference to table S1-S3 here, or add models with both covariates in Table S1-S3. Otherwise, it is not clear why you refer to Table S1-S3 and what is the difference with tables in the main text.
Results:
For all your model selections, I suggest adding the AIC weights as it gives the reader a better insight into the quality of the models. Thus, you should add it to all three tables in the main text and the same for Supplementary Materials.
Moreover, to avoid losing information on your model selection, I recommend using model averaging instead of selecting the most parsimonious model. In many cases in your results, you discarded the best-fitted model while you have 2 or 3 models with delta AIC <2. At least, if you present the AIC weights it would comfort the reader with the selection of the most parsimonious model.
L.267-268. Refer to the map I recommended adding in the Methods.
L.299-300. Provide means and 95% CI for the 3 other seasons.
L.312-318. This sentence is quite long and unclear. Consider rewording it in two sentences.
L.373-378. Have you considered calculating the mean total distance traveled by individuals wintering in Northern vs Southern GOM for both migrations? This parameter is easy to estimate with at least one location estimated daily. Adding this would be more convincing than only saying their route is more linear.
L.382-383. Maybe using filled polygons instead of contours would be better or some sort of color gradient with darker colors showing areas most heavily used.
L.390. The value of delta AIC in the text is different from the one in Table 3 (1.11 vs 1.17).
Figures 1-5. Adding the season on the top-right corner of each graph could facilitate the reading of the figures without having to continuously refer to their caption to remember which graph is for which season.
Figures 2 & 4. Using dashed lines for non-significant relationships and solid lines for significant ones would help to relate the results presented in the main text to the figures. Again, this small modification of the graphs could considerably facilitate the reading of the figures and the comprehension of the results.
Discussion:
L411-413. Saying that migratory birds fly faster in spring than in autumn migration is not always true even though this is the pattern we generally observe in migratory birds. For example, Schmaljohann’s (2018) review showed that some species present the opposite pattern and more recently, Seyer et al. (2021) showed that spring migration was slower than autumn migration for an Arctic species. I suggest rewording for something like: “Migratory birds GENERALLY fly faster in spring migration…”.
L.418-421. How would you explain this? Do pelicans from Southern GOM fly more hours per day? Or they simply have a shorter route. Try explaining here.
L.431-432. Specify for which season(s) this applies. I understand that it is during the wintering season but make it more obvious.
L.457-459. Have you considered comparing the distance traveled by individuals from the Northern vs Southern GOM? Maybe hourly speed is similar for both groups, but Northern individuals are following a longer route than Southern GOM ones but with the same hourly speed.
Suggested references:
Schmaljohann, H. 2018. Proximate mechanisms affecting seasonal differences in migration speed of avian species. Scientific Reports 8:4106.
Seyer, Y., G. Gauthier, J. Bêty, J. Therrien, and N. Lecomte. 2021. Seasonal variations in migration strategy of a long-distance arctic-breeding seabird. Marine Ecology Progress Series 677:1–16.
Author Response
This research article aimed at comparing the movement patterns of American White Pelicans sharing a breeding area but wintering in two geographically distinct regions. The authors used a consistent dataset of GPS-tracking data to separate the breeding population into two different wintering groups and to describe their hourly movements and their home ranges throughout the four biological seasons of the Pelicans. They presented interesting and well-supported conclusions stating that exogenous factors from the wintering area induced differences in the hourly movement patterns and the size of the home range during the winter, but these differences are not carried over to the Pelicans’ common breeding area.
General comments:
This research is very interesting as the results and the conclusions contribute to our understanding of the seasonal variability in migratory behaviors. Moreover, while many studies these days aimed at finding carry-over effects from one season to the other, this study highlights the fact that some behaviors are not necessarily carried over when they are induced by exogenous factors.
The study is well constructed, supported by sufficient and robust data, and the statistical analyses are performed adequately. To me, the introduction presents a great overview of the topic. The discussion is clear, not too lengthy, and adequately responds to the hypothesis with interesting interpretations and conclusions. Nevertheless, the result section was a little more arduous to read. I think small modifications to the figures would help to improve the understanding of the results (see my suggestions below).
Response: Thank you for the positive words and constructive suggestions. We appreciate your interest in our study. Below are the comments and revised components we have added.
Specific comments:
Abstract:
L.14-15. Move the sentence “We hypothesized that…” after “…Southern Gulf of Mexico (GOM)” (L.19).
Response: We moved the sentence as was suggested (L. 17).
Introduction:
L.76-77. “few studies have related individual migratory bird movement to human footprint index.”. Cite one or two of these few studies. Moreover, it could be interesting to add a few words about their conclusions.
Response: We added one case study about effects of human disturbance on bird migration (Cabrera-Cruz et al. 2018; L76-79).
L.92 + L.97 (and throughout the manuscript). You are talking about Northern and Southern GOM POPULATIONS. Do they really are distinct populations as they all meet in the same breeding area? Unless you have a priori information that distinguishes the individuals from these two wintering areas as different populations, it could be more appropriate to talk about “individuals wintering in different areas”.
Response: We agreed that the terminology ‘population’ is not appropriate in this uncertain situation. We changed all the words into “individuals” (or “group” if it is more appropriate) throughout the manuscript
L.101-103. “Relative abundance of pelicans […] populations (Figure S1).” Move to the previous paragraph when you present both populations. This will also avoid repetition in the next sentence.
Response: We moved the sentence to L.90.
Hypotheses 1 and 2: I suggest presenting hypothesis 2 first, and then hypothesis 1. This way, you start by assessing whether individuals from Southern GOM have shorter hourly movement distances and smaller home ranges. Then you assess if these differences, if any, are still present at the breeding site.
Response: We switched the order of hypotheses 1 and 2 (L.102-108)
L.107. It is not clear in this sentence what “between seasons” refers to. Only wintering and breeding period? Or you also are considering migratory periods?
Response: We added the words “summer and winter” between “between seasons” (L. 109)
L.111-112. Delete this sentence as it only rewords the previous sentence (L.109-111). If you want to clarify your hypothesis, you can simply specify in the previous sentence that “wintering further south” means Southern GOM.
Response: W deleted the sentence.
Methods:
L.118-121. I highly recommend adding a map with the boundaries of Northern and Southern GOM in the main text. This will help people not aware of your study system to figure it out easily. Also, you could add to this map the locations of your capture sites (L.133-138).
Response: Thank you for the suggestion. We originally thought to add the figures of study sites in the main text. However, we already had 5 figures in the main text, and we thought adding more might cause longer pages for readers to read accordingly. However, we also agree that such maps are helpful for them. Therefore, we added a reference of Figure S1, which has the boundary line (L. 131-132) and added one more Supplemental Figure of capture site (Figure S2; L. 151).
L.146. Refer to the map I recommended adding.
Response: Other reviewers recommended to remove this paragraph because this is redundant to the description of study regions. So we removed the paragraph accordingly.
L.159. Delete “in this study”.
Response: We deleted the words.
L.166-169. It is not clear how you considered or not the sedentary pelicans that remained at the non-breeding ground year-round. Since they do not migrate to the breeding ground, I suppose you have not considered them in your analyses. Otherwise, it may cause some problems since sedentary individuals could present a different pattern than migratory ones. Have you compared the analyses with and without these sedentary Pelicans? If you did not consider sedentary individuals in your analyses, add a short sentence that states that you ignored them for all further analyses.
Response: We added the sentence to clarify our process in L.167-170 “Although we did not consider the behaviors of sedentary pelicans during summer seasons and migration periods in our analyses, we included sedentary pelicans during winter seasons in the analyses, assuming that breeding and migratory activities of pelicans would not influence their behaviors in the subsequent winter seasons.”
L.169-171. Should be in the Statistical analyses section.
Response: We moved the sentence into L. 250-252
L.177-179. Refer to Figure S1.
Response: We added the reference Figure S1 in L. 178
L.231-232. You refer to models with “year” as a covariate (Table S1-S3), but not to those with “relative abundance index”. Either add a reference to Table 1-3, delete the reference to table S1-S3 here, or add models with both covariates in Table S1-S3. Otherwise, it is not clear why you refer to Table S1-S3 and what is the difference with tables in the main text.
Response: We changed the reference into Tables 1-3 (L. 231)
Results:
For all your model selections, I suggest adding the AIC weights as it gives the reader a better insight into the quality of the models. Thus, you should add it to all three tables in the main text and the same for Supplementary Materials.
Moreover, to avoid losing information on your model selection, I recommend using model averaging instead of selecting the most parsimonious model. In many cases in your results, you discarded the best-fitted model while you have 2 or 3 models with delta AIC <2. At least, if you present the AIC weights it would comfort the reader with the selection of the most parsimonious model.
Response: Thank you for the suggestion about AIC weights for all tables. We also agree that the model averaging would be an alternative approach to correct the coefficients of models. However, when we would like to primarily emphasize the seasonal variation of movement speed and home range between the Northern and Southern GOMs (season * group interaction), choosing the simplest model was ideal for us to target the ultimate objectives. Nevertheless, we still agree that it is helpful for readers to check AIC weights in the tables, so we added the values in all tables (Tables 1-3 and Tables S1 and S2)
L.267-268. Refer to the map I recommended adding in the Methods.
Response: We added Figure S2 in L. 268
L.299-300. Provide means and 95% CI for the 3 other seasons.
Response: We added the means and 95% CIs for 3 seasons (L. 301-306)
L.312-318. This sentence is quite long and unclear. Consider rewording it in two sentences.
Response: We reduced the repetitive words which made long and unclear (L. 318-323).
L.373-378. Have you considered calculating the mean total distance traveled by individuals wintering in Northern vs Southern GOM for both migrations? This parameter is easy to estimate with at least one location estimated daily. Adding this would be more convincing than only saying their route is more linear.
Response: We alternatively consider the used area size during migration as the surrogate of total migration distance. As we observed different individual pelicans had different numbers and the days of stopovers during migration, we think that the used area size during migration (Figure 5) would be also an alternative way to compare traveling distance (and we did not find statistical significance of used area difference between the Northern and Southern GOM groups for neither spring nor autumn migration). As this suggestion is related to the suggestions in Discussion, we note that we added a sentence of my interpretation in L. 427-430 and L. 467-470
L.382-383. Maybe using filled polygons instead of contours would be better or some sort of color gradient with darker colors showing areas most heavily used.
Response: Thank you for the suggestion. We thought half-transparent colors would be alternative. However, after the trial and error of changing filling and non-filling, we found non-filling is simpler and an option to show individual variation of home ranges and migration routes.
L.390. The value of delta AIC in the text is different from the one in Table 3 (1.11 vs 1.17).
Response: Thank you for the finding. We changed the text of 1.11 to 1.17 (L. 394).
Figures 1-5. Adding the season on the top-right corner of each graph could facilitate the reading of the figures without having to continuously refer to their caption to remember which graph is for which season.
Response: We added the text of each season in all figures.
Figures 2 & 4. Using dashed lines for non-significant relationships and solid lines for significant ones would help to relate the results presented in the main text to the figures. Again, this small modification of the graphs could considerably facilitate the reading of the figures and the comprehension of the results.
Response: We changed the solid lines to dashed lines where the slopes are not significant.
Discussion:
L411-413. Saying that migratory birds fly faster in spring than in autumn migration is not always true even though this is the pattern we generally observe in migratory birds. For example, Schmaljohann’s (2018) review showed that some species present the opposite pattern and more recently, Seyer et al. (2021) showed that spring migration was slower than autumn migration for an Arctic species. I suggest rewording for something like: “Migratory birds GENERALLY fly faster in spring migration…”.
Response: We added “generally” as is suggested (L. 418), and added two citations in L.419.
L.418-421. How would you explain this? Do pelicans from Southern GOM fly more hours per day? Or they simply have a shorter route. Try explaining here.
Response: We added a sentence of explanation “Because the size of migratory used area between two wintering groups did not differ during spring and autumn migrations, the total migration distances may not differ in the two wintering grounds between the Northern and Southern GOMs.” In L. 428-430.
L.431-432. Specify for which season(s) this applies. I understand that it is during the wintering season but make it more obvious.
Response: We added “during winter seasons” in L. 442
L.457-459. Have you considered comparing the distance traveled by individuals from the Northern vs Southern GOM? Maybe hourly speed is similar for both groups, but Northern individuals are following a longer route than Southern GOM ones but with the same hourly speed.
Response: Referred to my response about migratory used area as an alternative in the above suggestions (L. 467-470).
Suggested references:
Schmaljohann, H. 2018. Proximate mechanisms affecting seasonal differences in migration speed of avian species. Scientific Reports 8:4106.
Seyer, Y., G. Gauthier, J. Bêty, J. Therrien, and N. Lecomte. 2021. Seasonal variations in migration strategy of a long-distance arctic-breeding seabird. Marine Ecology Progress Series 677:1–16.
Author Response File: 
Author Response.docx
Reviewer 2 Report
I found the paper to be well written, and the topic is one that will be of interest to those who study animal movements.You state you hypotheses and methods clearly, and you analyses seem sound. You results are presented in a logical way that follows each hypothesis, and your conclusions seem sound. I only had a couple of minor changes: first, lines 145-147 are unnecessary; you already Northern and Southern GOM in your introduction. My other change was in your caption of Figure 2, where you use the term "migration phases". That is not an appropriate term, since 1) summer and winter are not migratory periods, and 2) phases would be more like events such as zugstimmung, zugdisposition, and zugenruhe. Something like "annual phase" or simply "season" might be better.
Author Response
I found the paper to be well written, and the topic is one that will be of interest to those who study animal movements.You state you hypotheses and methods clearly, and you analyses seem sound. You results are presented in a logical way that follows each hypothesis, and your conclusions seem sound. I only had a couple of minor changes: first, lines 145-147 are unnecessary; you already Northern and Southern GOM in your introduction. My other change was in your caption of Figure 2, where you use the term "migration phases". That is not an appropriate term, since 1) summer and winter are not migratory periods, and 2) phases would be more like events such as zugstimmung, zugdisposition, and zugenruhe. Something like "annual phase" or simply "season" might be better.
Response: Thank you for your interest in our study and constructive comments. We revised all the sentences where you suggested;
‘lines 145-147 are unnecessary’: We agree with your suggestion and removed the paragraph
‘Caption in Figure 2’: We changed the ‘migration phases’ into ‘seasons’ (L. 330 & 371).
