Review Reports - Local Segregation of Realised Niches in Lizards

Round 1

Reviewer 1 Report

I read with interest this paper on niche segregation in lizards. However, there are substantial issues with methodology and interpretation. I have detailed these below.

Line 44 - “[4] provided...” Do not use this method of citation when you are referring to an author specifically in a sentence. Use the authors name with the numbered citation after.

Line 59 – Do not use acronyms, spell out the terms.

Line 84 – I think the prediction would be the opposite, i.e., that the realized niche would differ in syntopy compared to allopatry. The classical pattern exhibited by Galapagos finches and others is differences in syntopy and similarity in allopatry.

Lines 84-91 – You should cite literature to justify these predictions, especially the links to relative abundance.

You should label locations as syntopic or allopatric, it is difficult for me to remember which are which.

A PCA to lower three variables to two for analysis seems like it obscures data rather than providing clarification.

You should provide a key to the colors used in the figures. I am not sure how to read these plots.

It is difficult to determine exactly what your results show. I have read and reread the methods and results sections and it looks like of the 24 (or more) statistical p-values reported, only one (maybe two) is significant indicating a statistical difference in niches (PC-PG Estrela). No more than we would expect by chance. However, in the text you insist that niches in allopatry are more similar than niches in syntopy and that the less abundant species is the one that changes. It looks like there is no statistical justification for your conclusion.

Line 413-417. So you used a test that is invalid in the first place?

I don’t understand how you have used a statistical analysis to support your conclusions. It seems that your analysis does not support your conclusions, but you go ahead and suggest that it does somehow. I may not understand what you have done, in which case you should better explain in your methods how you have set up your tests to test your hypotheses. Or, you should alter your conclusions to be consistent with the statistical output.

Author Response

Reviewer 1

I read with interest this paper on niche segregation in lizards. However, there are substantial issues with methodology and interpretation. I have detailed these below.

Line 44 - “[4] provided...” Do not use this method of citation when you are referring to an author specifically in a sentence. Use the authors name with the numbered citation after.

>>>> Corrected.

Line 59 – Do not use acronyms, spell out the terms.

>>>> The acronym ENMs is spelt out in the previous sentence. Repeating it throughout the text would make it wordy.

>>>> Thank you very much for pointing that possibility. However, the results presented confirmed our hypothesis: species in sympatry use the same realised niche while they differed in allopatry. With the changes applied in the ecospat analyses (see below), we suggest that the results are now easier to understand.

Lines 84-91 – You should cite literature to justify these predictions, especially the links to relative abundance.

>>>> We added two references about the link between abundance and character displacement.

You should label locations as syntopic or allopatric, it is difficult for me to remember which are which.

>>>> We indicated the study areas as allopatric and sympatric across the manuscript, namely in the captions of the Figures 3, 4, and 5, in the results, and in the discussion.

A PCA to lower three variables to two for analysis seems like it obscures data rather than providing clarification.

>>>> This procedure is constrained by the way ecospat works. It is not possible to perform ecospat analyses without the application of a PCA.

You should provide a key to the colors used in the figures. I am not sure how to read these plots.

>>>> We added the following sentence in Figures 2-5: ‘Stability is represented in blue, unfilling in green, and expansion in red. The red arrow is the distance between centroids’.

>>>> Ecospat analyses present two ways of calculating p-values: considering that the niches will be more similar than expected, or that the niches will be less similar than expected. We repeated the ecospat analyses this time choosing the alternative hypothesis of niches less similar than expected. The results were the same, but the p-values of the equivalency test changed and now they are easier to interpret: comparisons between sympatric areas continue to be non-significant, but the other comparisons are now all significant. The similarity test continues to provide non-significant results, as expected. We made clear in the methods that we are evaluating the hypothesis of niches being less similar than expected by random. All subsequent results and interpretations are made following this hypothesis.

Line 413-417. So you used a test that is invalid in the first place?

>>>> We did not. The similarity results only refined the results of the equivalency test. The results from the equivalency test continue to be valid, but the similarity test indicates the differences found because the species are located in different study areas.

>>>> Now that we changed the direction of the ecospat test to expect ‘niches less similar than random’, the results are clearer and the significant level coincide with the meaning of the results. Both ways of performing ecospat results are the same, but we admit that choosing the ‘niche more similar than random’ direction is harder to interpret.

Reviewer 2 Report

Dear authors,

I have reviewed your manuscript “Local segregation of realised niches in lizards” for possible publication in the International Journal of Geo-Information. I found your manuscript very interesting with novel results and findings. I enjoyed reading it. I have, however, comments that require to be addressed before acceptance, and some suggestions to improve the quality and highlight the message of your study.

See comments below.

Abstract

L16. Chance "provide" to "are"

L27-28. You may want to close the abstract with a conclusion of your results, instead a general statement from the method.

Introduction

L16. Change "provide" for "are"

L34. Change to “forced”

L36. What other dimensions? Provide details for your readers

L41. Requires reference(s)

L46. Don’t you mean “migration”?

L55. Explain more “species ecological niche perspective”

L63. Change to “by two species in strict syntopy may…”

L66. Reference?

L79. Delete “of a group”

L90-91. Change to “the least abundant species within the community will shift its realised niche”.

L97-101. This paragraph should be moved before the objectives or into the Material and Methods (Study species)

Materials and Methods

L126. Use complete species name after period

L130. For this particular two species? Clarify

L154. How many individuals is “enough”?

L165. Describe these “favourable” conditions?

L170. How many people were involved in the sampling?

L172-173. Based on what? Previous published studies? Previous surveys?

L173. Change to “could not be”

L178. Add reference and version of R software used

L203. Reference?

Results

L235. Not really “very high”

L236. With exceptions. Also, you need to provide a reference for what you described as “very low”

Figures 2-5. Describe what different colours indicate

L365. “were not”, better provide a positive statement for a better understanding

L237. Provide as metric for the distance between centroids

L367. As above

Discussion

L374. And also, in other environmental variables, which were no measured here. This need to be discussed

L375. But your micro-niche only consists in three variables. This caveat needs to be discussed

L376. Consider adding information regarding the environmental variables, e.g. mean, standard deviation, max, min of temperature, humidity, and wind speed at each site.

L377. Shifts to different microhabitat?

L379-392. This is repetition of Results, there’s no discussion here (delete?)

L401. “in terms”

L423. Genera in italic

L425-428. This paragraph seems lost here as a random example. Discuss this with your findings

L430. Discuss how other variables (e.g. predation, presence of additional species at the other sites), although difficult to measure can have an effect on the species’ niches. Also, you need to justify the environmental variable you used in the context of their importance to determine the species’ occurrence, and not only because they are “easy” to measure

L434. Yes, but you can use other tools, such as hypervolume (see Blonder et al. 2014, Global Ecology and Biogeography)

L438. “on this issue”

L438. Similar to the abstract, this is the take-home-message, perhaps ending with a powerful statement on your findings, instead of mentioning what could be done if we had better data

Author Response

Reviewer 2

Abstract

L16. Chance "provide" to "are"

>>>> Done.

L27-28. You may want to close the abstract with a conclusion of your results, instead a general statement from the method.

>>>> We deleted the last sentence. Now, the last sentence of the abstract is the main conclusion of the study: ‘Our study demonstrates that sympatry may shift species’ realised niche’.

Introduction

L16. Change "provide" for "are"

>>>> Done.

L34. Change to “forced”

>>>> Done.

L36. What other dimensions? Provide details for your readers

>>>> We include some examples as resources and biotic interactions.

L41. Requires reference(s)

>>>> We included the reference [8].

L46. Don’t you mean “migration”?

>>>> No, the fundamental niche is defined in conditions where species immigrants are not necessary, where the system is closed. Pulliam (2000) introduced the source-sink theory in the definition of the ecological niche to show that it is possible to have viable populations in unsuitable populations (the sink) by the arrival of immigrants from a nearby healthy population (the source). This reference has not been included in the manuscript as it is beyond the objectives of our study.

L55. Explain more “species ecological niche perspective”

>>>> We added at the end of the sentence: ‘, i.e. without modelling the ecological niche’.

L63. Change to “by two species in strict syntopy may…”

>>>> Done.

L66. Reference?

>>>> We already have a reference [29] at the beginning of the sentence.

L79. Delete “of a group”

>>>> Done.

L90-91. Change to “the least abundant species within the community will shift its realised niche”.

>>>> Done.

L97-101. This paragraph should be moved before the objectives or into the Material and Methods (Study species)

>>>> We moved the paragraph to the end of the section Study species in Materials and Methods.

Materials and Methods

L126. Use complete species name after period

>>>> Corrected.

L130. For this particular two species? Clarify

>>>> Effectively, as indicated in the text, Espinho is the only syntopy point where both species are found together.

L154. How many individuals is “enough”?

>>>> At least 30 individuals. We added this information to the sentence.

L165. Describe these “favourable” conditions?

>>>> We specified warm, without raining and strong winds.

L170. How many people were involved in the sampling?

>>>> We added in L174-176 the following sentence: ‘Searching for marked wall lizards always involved two people: one person actively searched for lizards and recorded the local weather data; the second person recorded the wall lizard location, waiting always some metres behind to avoid disturbance until the moment of using the GPS receptor’.

L172-173. Based on what? Previous published studies? Previous surveys?

>>>> This is based on our own experience. A wall lizard is able to move 4-5 metres in few seconds, so after one hour is impossible to forecast the direction of its movement. We added ‘Based on our own experience’ at the beginning of the sentence.

L173. Change to “could not be”

>>>> Done.

L178. Add reference and version of R software used

>>>> We used ecospat version 3.1.

L203. Reference?

>>>> The reference [18] is indicated at the beginning of the paragraph.

Results

L235. Not really “very high”

>>>> The Referee is right: considering that the D index ranges from 0 to 1, values between 0.6 and 0.8 should be considered only as high.

L236. With exceptions. Also, you need to provide a reference for what you described as “very low”

>>>> All stability and unfilling values are lower than 0.1. We indicated this value of <0.1.

Figures 2-5. Describe what different colours indicate

>>>> We added the following sentence in Figures 2-5: ‘Stability is represented in blue, unfilling in green, and expansion in red. The red arrow is the distance between centroids’.

L365. “were not”, better provide a positive statement for a better understanding

>>>> We changed to ‘lower than’.

L237. Provide as metric for the distance between centroids

>>>> We make reference to the see the red arrows in Figures 2-5.

L367. As above

>>>> We make reference to the see the red arrows in Figures 2-5.

Discussion

L374. And also, in other environmental variables, which were no measured here. This need to be discussed

>>>> We discussed in L349-352 how using a different set of variables will produce different results. As Podarcis species are generalist, further research may clarify whether other variables might lead to niche segregation in Podarcis species.

L375. But your micro-niche only consists in three variables. This caveat needs to be discussed

>>>> We briefly discussed this issue in L349-352.

L376. Consider adding information regarding the environmental variables, e.g. mean, standard deviation, max, min of temperature, humidity, and wind speed at each site.

>>>> We created a table with the descriptive statistics of the environmental variables and a figure with boxplots for each species and study area. Both table and figure are placed as supplementary materials (Table S2 and Figure S1). We also analysed the differences for each variable by species and study areas. The results can be found in Table S3. We updated the methods and the results respectively in L212-214, and in L224-225 and L228-230.

L377. Shifts to different microhabitat?

>>>> We think the term ‘occupy’ is more precise as we are referring to micro-habitats and not to the realised niche. We prefer to reserve the term ‘shift’ when referring specifically to the species’ realised niche.

L379-392. This is repetition of Results, there’s no discussion here (delete?)

>>>> This is effectively a summary of the results, but we consider it necessary to show how the different comparison support the two main conclusions stated at the beginning of the discussion.

L401. “in terms”

>>>> Corrected.

L423. Genera in italic

>>>> Corrected.

L425-428. This paragraph seems lost here as a random example. Discuss this with your findings

>>>> We related our results to the previous studies making reference to the fact that we did not find niche segregation when comparing different species in the syntopic areas (Espinho, Estrela, Moledo). In these three areas, the two Podarcis species occurring there are using the same environmental variables recorded.

>>>> We justified in L348-349 the importance of these environmental variables in lizards. We also discussed the need of future research to analyse how other variables might promote niche segregation.

L434. Yes, but you can use other tools, such as hypervolume (see Blonder et al. 2014, Global Ecology and Biogeography)

>>>> Thank you very much for the suggestion. We added a comment about hypervolume in L401-403. We also added the suggested reference.

L438. “on this issue”

>>>> Corrected.

L438. Similar to the abstract, this is the take-home-message, perhaps ending with a powerful statement on your findings, instead of mentioning what could be done if we had better data

>>>> We substitute the last sentence by ‘As our study demonstrates, biotic interactions among species may shift spatially the species’ realised niches’.

Reviewer 3 Report

line 34: "forged" should be changed to "forced"

line 115: what is a "wet crop"? From the areal photos in the figure it kind of looks like a mowed field of grass, with bales of hay or straw. "Wet crop" does not make much sense.

lines 122 and 123: what is a "woody walking passage"? Is it a path made out of planks of wood? If so, that would typically be called a "boardwalk." "Woody walking passage" does not make much sense. I found the exact area on Google Earth, and was able to zoom in, and it is definitely what would be called a "wooden boardwalk"

line 133: by "shore of a dam" you probably mean "shore of a reservoir." A reservoir would have shores, dams would not.

Figures 2-5: it would be helpful if the meaning of the coloration were made explicit. It can be inferred from the "small plots," but it would be helpful to include in the caption exactly what the blue/purple, green, and red signify.

Author Response

Reviewer 3

line 34: "forged" should be changed to "forced"

>>>> Corrected.

line 115: what is a "wet crop"? From the areal photos in the figure it kind of looks like a mowed field of grass, with bales of hay or straw. "Wet crop" does not make much sense.

>>>> We apologise for this error. We meant irrigated crops. In Coronado area, corn is cultivated; in Moledo area, potatoes are cultivated.

>>> Thank you for the term. We applied it accordingly.

line 133: by "shore of a dam" you probably mean "shore of a reservoir." A reservoir would have shores, dams would not.

>>>> Corrected.

>>>> We added the following sentence in Figures 2-5: ‘Stability is represented in blue, unfilling in green, and expansion in red. The red arrow is the distance between centroids’.

Reviewer 4 Report

This is a good work because you have done a very good field surveys across the entire territory and statistical analyses. Otherwise, I have a lot of recommendations to increase the quality of your paper. Be careful with the writing and mistakes.

The keyword “species’ realized niche” is repeated in the article title. In order to increase the visibility of your paper I recommend changing the keyword by “R package ecospat” which is more appropriate to hook more readers. If you change it by a different keyword, you will increase the probability that your paper could be found by future readers when they look for your paper in some databases like Scopus for example. If you repeat the same words in the article title and in keywords, less people could find your work. So, you must think about the visibility of your research.

Please, put the keywords in alphabetical order. The journal publishes the keywords in this way. Follow the rules of the journal.

In line 44, 47 and 48 there is a mistake. You must not put the references at the very beginning of the sentence. Please, put it at the end of the sentence. In this way you must start a sentence in capital letters. This is grammatically incorrect.

In the line 58 write in capitals the letters used for the acronym: so, you should write “Ecological Niche Models” just before “(ENMs)”, this is very useful for a fluent reading.

In Materials and Methods, you must write more explicit about the three environmental variables because is a very important part of the paper. As well you must write why you use them and how in this part of the article.

The first time you write a scientific name you must put the authors at the end of it, so, in line 104 and the following when writing “Podarcis bocagei” and the two lizards you must look for the authors and write them.

In line 137-139, in the Table 1 you must not put the text in the centre, you must justify it on the left and on the right, not centered. Make the same correction on tables 2 and 3.

From the line 156 to 172 you must use the passive voice. You must avoid the word “We”, please, use the indirect speech or the passive voice, but not the active in scientific texts. You must do the same correction at the beginning of Results from line 205 to 207. If you find the same mistake in the whole paper, please correct it.

In line 200 you must write “Stability” just before “(S)” to make the reading easier. Do the same for “Unfilling” and “Expansion” in the following lines.

From lines 279 to 360 the figure 3, 4 and 5 are wrong written. You must fix it because is very difficult to read the text of the figures.

In the references you must follow the rules of the journal.

You must write the year of the papers in bold letters.

You must write the volume of the journals in italics.

Author Response

Reviewer 4

>>>> Thank you for the recommendation. We changed the keyword. We also deleted the keyword ecospat.

Please, put the keywords in alphabetical order. The journal publishes the keywords in this way. Follow the rules of the journal.

>>>> Corrected.

>>>> Following the advice by Referee 1, we put the name of the author before the number of the reference.

In the line 58 write in capitals the letters used for the acronym: so, you should write “Ecological Niche Models” just before “(ENMs)”, this is very useful for a fluent reading.

>>>> The acronym is spelled out in the previous sentence.

>>>> We explained we selected these variables as they are main drivers of lizards biogeography and activity patterns and provided two references [16,36]. We also indicated that temperature was always measured in the shade.

>>>> We added the authors of the species.

In line 137-139, in the Table 1 you must not put the text in the centre, you must justify it on the left and on the right, not centered. Make the same correction on tables 2 and 3.

>>>> Corrected.

>>>> We prefer the use of active voice because is a more direct style, easier to use than the passive voice, and it uses less words. We write all our manuscripts in active voice and this is the style we recommend in our lectures of scientific writing at the University of Porto.

In line 200 you must write “Stability” just before “(S)” to make the reading easier. Do the same for “Unfilling” and “Expansion” in the following lines.

>>>> We corrected the three names.

From lines 279 to 360 the figure 3, 4 and 5 are wrong written. You must fix it because is very difficult to read the text of the figures.

>>>> We have remade the figures to improve their reading.

In the references you must follow the rules of the journal.

>>>> We have used the style of the journal provided in Mendeley reference manager. We have checked the format once more and it looks the according to the journal guidelines.

You must write the year of the papers in bold letters.

>>>> Corrected. We apologise for the error. We were using the style defined by MDPI in Mendeley reference manager software.

You must write the volume of the journals in italics.

>>>> Corrected.

Reviewer 5 Report

The problem described by the paper is framed clearly and demonstrates the relevance of the research problem being addressed. The paper is well written and requires only minimal editorial fixes (lines 34 [forged instead of forced], 401 [jn instead of in] are two that I found).

The primary issues that I found relate to the presentation of the data and results - this could be done more clearly. Specifically:

Summary statistics for the measured climate variables should be included to demonstrate the range of variation of these critical measurements - both between and within the study areas.
Summary plots that specifically illustrate the relationships described in lines 375-392 would save the reader a lot of interpretive effort in pulling values out of Table 3 and seeking out the visual results in the figures.
The description of the ecospat components described in lines 200-203 doesn't make clear the critical linkage (if it exists) between more- and less-abundant species in the definitions of stability and expansion. This makes interpretation of the values in the last three columns of table three and tracking with the comparative claims made in lines 379-392 less straightforward to assess.
A formal test of spatial autocorrelation between observed locations of the same individuals would help support the assumption of spatial independence made in lines 173-174.

Some of the issues highlighted in the conclusions as confounding study area attributes should have been introduced earlier in the methods section as specific environmental attributes that are not considered in the analysis that it could have been anticipated would be influential - the types and characteristics of walls in the different study areas are an example of this.

The map scale bars included in Figure 1 should be re-generated to be legible when the integrated graphics are combined into the figure.

These issues are all resolvable and would produce a stronger and more clear paper from the perspective of demonstrating the linkage between the data, the analysis results, and the conclusions reached.

Author Response

Reviewer 5

The problem described by the paper is framed clearly and demonstrates the relevance of the research problem being addressed. The paper is well written and requires only minimal editorial fixes (lines 34 [forged instead of forced], 401 [jn instead of in] are two that I found).

>>>> Thank you for your kind comments. Both errors have been corrected.

The primary issues that I found relate to the presentation of the data and results - this could be done more clearly. Specifically:

Summary statistics for the measured climate variables should be included to demonstrate the range of variation of these critical measurements - both between and within the study areas.

>>>> We created a table with descriptive statistics and a figure with boxplots for each species and study area. Both table and figure are placed as supplementary materials (Table S2 and Figure S1). We also analysed the differences for each variable by species and study areas. The results can be found in Table S3. We wrote these results in L224-225 and L228-230. We also added a new paragraph in the methods section in L212-214.

Summary plots that specifically illustrate the relationships described in lines 375-392 would save the reader a lot of interpretive effort in pulling values out of Table 3 and seeking out the visual results in the figures.

>>>> We made reference to the corresponding figures for these results: Fig. 3 for P. bocagei, Fig. 4 for P. carbonelli, and Fig. 5 for P. guadarramae. We think it is not necessary to create a new figure repeating the ecospat results for these study areas.

The description of the ecospat components described in lines 200-203 doesn't make clear the critical linkage (if it exists) between more- and less-abundant species in the definitions of stability and expansion. This makes interpretation of the values in the last three columns of table three and tracking with the comparative claims made in lines 379-392 less straightforward to assess.

>>>> We rephrased sentence in L208 as ‘Finally, ecospat quantifies in three components how the niches overlap in the environment space’ to make clearer the meaning of the stability, expansion, and unfilling measures. Then, in the discussion, we clarified to what measures we referred, that is, stability. We also changed the sentence in L365 to clarify the concept of spatial overlap: ‘Consequently, we found species highly overlapping (i.e. stability) their realised niches between study areas which are structurally different’.

A formal test of spatial autocorrelation between observed locations of the same individuals would help support the assumption of spatial independence made in lines 173-174.

>>>> We calculated the Moran’s I test for each study area and local weather measures. Results showed that all lizards points are independent as all Moran’s I estimates are close to 0. We explained this analyses in L180-182 and presented the results in L220-221. The values of the Moran’s I estimates are presented in Table S1.

>>>> The height of the walls is indicated in the Study area sections.

The map scale bars included in Figure 1 should be re-generated to be legible when the integrated graphics are combined into the figure.

>>>> We have re-generated the scale bars in Figure 1.

>>>> Thank you very much for helping us improving the manuscript.

Round 2

Reviewer 1 Report

I can see that the authors have made a good effort to answer my previous questions. The main issues turn out to be constraints of using the ecospat program to analyze the data. It is unfortunate that the ecospat program can only use a PCA as input and can only evaluate two populations at one time. However, the authors have provided a consistent evaluation of habitat use (based on three variables) among these three species.

I have only one suggestion. There is a fairly large and long literature on character displacement that has not been addressed in this paper. Habitat use can be considered a specie's trait, and results from this paper should be discussed in the context of pairwise comparisons for other traits. I think at least a paragraph in the discussion should address this idea in the literature.

Author Response

Dear Editors,

Please find enclosed a revised version of the manuscript entitled ‘Local segregation of realised niches in lizards’ by Neftalí Sillero, Elena Argaña, Cátia Matos, Marc Franch, Antigoni Kaliontzopoulou and Miguel A. Carretero. We have applied the suggestion made by the Referee. We explained how these changes were applied after each Referee’s comments. Our comments are indicated with >>>>.

We hope that these changes help to improve the quality of the manuscript for its consideration in ISPRS International Journal of Geo-Information.

The Authors

Reviewer 1

>>>> We appreciated all the help provided by the Referee to improve the manuscript. We have extended the discussion about the character displacement as requested in the penultimate paragraph and add three new references. Now, that paragraph stands as:

Many studies have identified shifts in the realised niches of invasive and introduced species when compared with their native ranges [30,31,62,63]. Although all these studies analysed whether there is niche conservatism, none showed shifts in the fundamental niche [4]. All niche shifts detected correspond to changes in the species’ realised niche [17–20]. Here, we provided evidence that lizards shifted their realised niches when in syntopy, depending on the conspecific with which they coexisted. Based on previous knowledge of the examined species, phylogenetic relationships [64], morphology [65,66] and structural microhabitat use [57], and supported by character displacement theory [29], we might have expected Podarcis bocagei and P. guadarramae to exhibit the less intense niche shifts across areas, as they are sister species [64]. However, these species have quite different structural habitat preferences [57,59], which is hypothesized to have driven their morphological divergence. As such, they might have segregated their niches adaptively across their entire (and extensively overlapping) distribution areas [64], but not entering in direct competition when they meet in sympatry. By contrast, P. bocagei and P. carbonelli are much more similar in both morphology and structural habitat use [59,67], a fact that could be expected to trigger more intense niche shifts across this species pair. However, these predictions are not verified in our system. Instead, the most feasible explanation in our study case regards relative local abundance (see above) [34], where asymmetric competition may explain the pattern of ecological character displacement observed [33].