Do Birds of a Feather Always Flock Together? Assessing Differences in Group and Individual Zoo Enclosure Usage by Comparing Commonly Available Methods

Round 1

Reviewer 1 Report

The study addresses the possible influence of the space use measurement technique (index) used on group versus individual level outcomes. This topic is particularly important in the context of animal welfare, where it is the individual experience that matters most. Understanding what most animals within a group may not always represent the needs of all, and may leave some animals to experience reduced welfare. Overall, the study is well designed and executed. The results are clearly reported, and potential confounding factors are identified and appropriately discussed. I have only a few minor comments for the authors to consider:

LN 29: the acronym “EI” is not defined prior to use; definition is provided in LN 68.

LN 60: consider changing “how” to “whether”

LN 101: a reference is needed to support this statement: “given the evidence that individual animals display different personalities…”

LN 210-217: It would be helpful to have the zone numbers identified (with reference to Figure 1), so that it is easier for the reader to understand which zones were combined.

LN 251: I found the description of the sign test difficult to follow. It suggests that a single one sample sign test was conducted “with the median group m/SPI value calculated for comparison against individual m/SPIs”. However, earlier it states that SPI values were also analyzed. Am I correct to assume that median group m/SPI was compared against individual m/SPI values, and median group SPI was compared against individual SPI values? Therefore, were 4 sign tests were conducted: one comparing m/SPI values and the other SPI values for each zoo separately? This interpretation seems to align with the results in LN 289-295.

LN 292-295: What are the values in the parentheses represent (e.g. (0.6068))?

LN 374: Change table heading to “Table 2”

LN 385: Change “tjese” to “these”

General comment for the author’s consideration: the graphs are not legible when printed in gray scale. You may wish to consider changing color saturation or using dashed lines, etc., to ensure that the article is more accessible in printed format.

Author Response

The study addresses the possible influence of the space use measurement technique (index) used on group versus individual level outcomes. This topic is particularly important in the context of animal welfare, where it is the individual experience that matters most. Understanding what most animals within a group may not always represent the needs of all, and may leave some animals to experience reduced welfare. Overall, the study is well designed and executed. The results are clearly reported, and potential confounding factors are identified and appropriately discussed. I have only a few minor comments for the authors to consider:

Thank you for the comments and helpful review of this paper. We are pleased that you find it a useful piece of work.

LN 29: the acronym “EI” is not defined prior to use; definition is provided in LN 68.

Edited.

LN 60: consider changing “how” to “whether”

Edited.

LN 101: a reference is needed to support this statement: “given the evidence that individual animals display different personalities…”

Reference included.

LN 210-217: It would be helpful to have the zone numbers identified (with reference to Figure 1), so that it is easier for the reader to understand which zones were combined.

Edited accordingly.

LN 251: I found the description of the sign test difficult to follow. It suggests that a single one sample sign test was conducted “with the median group m/SPI value calculated for comparison against individual m/SPIs”. However, earlier it states that SPI values were also analyzed. Am I correct to assume that median group m/SPI was compared against individual m/SPI values, and median group SPI was compared against individual SPI values? Therefore, were 4 sign tests were conducted: one comparing m/SPI values and the other SPI values for each zoo separately? This interpretation seems to align with the results in LN 289-295.

Yes, you are correct. Multiple sign tests were run. We have clarified this in the text.

LN 292-295: What are the values in the parentheses represent (e.g. (0.6068))?

Edited to explain that these are median SPI values.

LN 374: Change table heading to “Table 2”

I'm sorry but I don't understand the correction here. The table is already called Table 2? Please can you clarify?

LN 385: Change “tjese” to “these”

Edited.

General comment for the author’s consideration: the graphs are not legible when printed in gray scale. You may wish to consider changing color saturation or using dashed lines, etc., to ensure that the article is more accessible in printed format.

Thank you for the comment. I don't believe that these articles are very printed in greyscale as they are online only? So I think we are OK with colour graphs. But if the journal would like this changed, we would be happy to edit the colours.

Reviewer 2 Report

Group and individual usage of two enclosures by flamingos is monitored in this study. Three indexes were tested in a flock of Chilean flamingos (Phoenicopterus chilensis) and a flock of Greater flamingos (Phoenicopterus roseus). Quite small sample size (N=2 flocks) to understand enclosure use and inform future improvements, including multi-zoo comparison studies and post-occupancy evaluation of behaviour and space use across conditions (sic, lines 56-57). Flocks differences could be explained by the enclosure design and the species, but also by the randomness associated with a low sample size (N=2). The study also tests empirical methods for assessing an animal’s spatial use. This aim was fulfilled. The study provides helpful basic information to design the enclosure size and structure.

Minor comments

Line 29. The letters 'EI' should be defined.

Lines 38-39. This comment is minor and non-mandatory. Consider removing the words 'always', 'always', and 'same' in the first sentence of the Introduction. Wild animals evolved in completely different ecological conditions than those provided in zoos. Accordingly, zoo enclosures do not fully meet animal needs. A zoo aiming to meet wild animal natural requirements become a natural preserve at the very least. Declaring that zoo enclosures do not always meet the needs of an animal suggests that most of the time, zoos fulfil them, which is a good aim but not a good description of zoos.

Lines 124. Delete the word 'different' in the sentence "Two species of flamingo at two different zoological collections…". Otherwise, it must be included twice: "Two different species of flamingos at two different zoological collections…". 

Lines 151-152. Figure 1. Please explain what the X means. Is it the observer location? The drawings describe some areas beyond the reach of flamingos (for example, visitors viewings). What are white / empty areas? Readers may think neither people, flamingos, the observer, nor the reader must care about those empty areas. However, these intriguing empty spaces could include adjacent enclosures with predators or noisy services (child's playground, bar, restaurant, etc.). Please provide an exhaustive list of what was around the flamingo enclosures. An appendix may include this supplemental information.

Lines 155-162. Collecting days. The movements around the enclosures could change in each season stage. The readers may wonder if these birds were pre-breeding, breeding, post-breeding or wintering. Data could have been recorded in the flamingo's spring migration season. Migratory birds are quite restless those weeks. Please include a short comment about each flamingo species' breeding/non-breeding cycle. Some birds could be breeding (line 409: the nesting flamingos (…) were the reason for wider variation of Electivity Index). Perhaps the study results may not generalise to other species, enclosures, and zoos if the annual season was inappropriate.

Line 250. Any predictor with a VIF < 2 was removed from the model. Perhaps there is a typo, and predictors were removed when VIF was greater than two. For example, Line 333 declares that "All VIF values were below 2".

Lines 343-344. Electivity Index (EI) values range between -1.8 and +1.7 (Fig. 6, top), but they must range between -1 and 1 (line 237). Please, explain why the manuscript declares the Electivity Index range is [-1, +1] in line 237, but (<-1, >+1) in Figure 6. Perhaps the intervals drawn in Fig. 6 were incorrectly calculated. Were they 1 SD, 1 SE or the CI95%? Statistical intervals for bounded variables such as proportions, percentages, and SIs are always skewed except in the middle (0.5, 50%, 0).

Author Response

Group and individual usage of two enclosures by flamingos is monitored in this study. Three indexes were tested in a flock of Chilean flamingos (Phoenicopterus chilensis) and a flock of Greater flamingos (Phoenicopterus roseus). Quite small sample size (N=2 flocks) to understand enclosure use and inform future improvements, including multi-zoo comparison studies and post-occupancy evaluation of behaviour and space use across conditions (sic, lines 56-57). Flocks differences could be explained by the enclosure design and the species, but also by the randomness associated with a low sample size (N=2). The study also tests empirical methods for assessing an animal’s spatial use. This aim was fulfilled. The study provides helpful basic information to design the enclosure size and structure.

Thank you for the useful comment. We have included some further information in the discussion to explain any potential effects of the sample size on our findings.

Minor comments

Line 29. The letters 'EI' should be defined.

Edited.

Lines 38-39. This comment is minor and non-mandatory. Consider removing the words 'always', 'always', and 'same' in the first sentence of the Introduction. Wild animals evolved in completely different ecological conditions than those provided in zoos. Accordingly, zoo enclosures do not fully meet animal needs. A zoo aiming to meet wild animal natural requirements become a natural preserve at the very least. Declaring that zoo enclosures do not always meet the needs of an animal suggests that most of the time, zoos fulfil them, which is a good aim but not a good description of zoos.

Thank you for the comment. We have edited this but changed the first "always" to "fully".

Lines 124. Delete the word 'different' in the sentence "Two species of flamingo at two different zoological collections…". Otherwise, it must be included twice: "Two different species of flamingos at two different zoological collections…". 

Edited.

Lines 151-152. Figure 1. Please explain what the X means. Is it the observer location? The drawings describe some areas beyond the reach of flamingos (for example, visitors viewings). What are white / empty areas? Readers may think neither people, flamingos, the observer, nor the reader must care about those empty areas. However, these intriguing empty spaces could include adjacent enclosures with predators or noisy services (child's playground, bar, restaurant, etc.). Please provide an exhaustive list of what was around the flamingo enclosures. An appendix may include this supplemental information.

Thank you for the comment. We have expanded on nearby features to each enclosure and we have included further information in the caption to explain the figures. 

Lines 155-162. Collecting days. The movements around the enclosures could change in each season stage. The readers may wonder if these birds were pre-breeding, breeding, post-breeding or wintering. Data could have been recorded in the flamingo's spring migration season. Migratory birds are quite restless those weeks. Please include a short comment about each flamingo species' breeding/non-breeding cycle. Some birds could be breeding (line 409: the nesting flamingos (…) were the reason for wider variation of Electivity Index). Perhaps the study results may not generalise to other species, enclosures, and zoos if the annual season was inappropriate.

Thank you for the comment. We feel that this has already been addressed as we have included a large section in the discussion on breeding and nesting activities and differences between the two flocks. So we were already away of this potential seasonal affect between the two flocks. This evaluation is on page 16. Also, flamingos do not migrate. They are itinerant nomads, moving only when needed between feeding and breeding areas. We have noted the differences in activity due to the courtship display of the Chilean flamingo in April compared to the nesting activities of the greater flamingos in May.  We have added to our discussion here. 

Line 250. Any predictor with a VIF < 2 was removed from the model. Perhaps there is a typo, and predictors were removed when VIF was greater than two. For example, Line 333 declares that "All VIF values were below 2".

Thank you. This is indeed a typo! 

Lines 343-344. Electivity Index (EI) values range between -1.8 and +1.7 (Fig. 6, top), but they must range between -1 and 1 (line 237). Please, explain why the manuscript declares the Electivity Index range is [-1, +1] in line 237, but (<-1, >+1) in Figure 6. Perhaps the intervals drawn in Fig. 6 were incorrectly calculated. Were they 1 SD, 1 SE or the CI95%? Statistical intervals for bounded variables such as proportions, percentages, and SIs are always skewed except in the middle (0.5, 50%, 0).

Thank you for the comment. Each median is on the -1 to +1 scale, but as these are plotted with the standard deviation calculated from all data points for that time period that the median was calculated from. We have added to our explanation of this graph.