Assessing the Impact of Insect Decline in Islands: Exploring the Diversity and Community Patterns of Indigenous and Non-Indigenous Arthropods in the Azores Native Forest over 10 Years

Round 1

Reviewer 1 Report

Dear authors, please find my few observations in the text, the paper is well written and I will be glad to see it published.

Comments for author File: Comments.pdf

Author Response

Global response to reviewer 1:

Many thanks for your detailed revision of our manuscript.

All the typos you mentioned have been corrected.

On page 2, "climate" has been changed to "temperature, precipitation" as detailed in the cited literature [43].

On page 16, we have followed your suggestion and changed "we observed" to "we detected".

Author Response File: Author Response.docx

Reviewer 2 Report

This paper reports on the results of 10 years monitoring efforts of arthropods in native forest on one Azorean islands. This is for sure a most valuable data set and it allows first inferences on stability versus fluctuations among the fauna. In the present study, arthropods are partitioned according to the biogeographical status, and temporal trends are compared between indigenous and introduced species. As expected, the indigenous fraction of the observed arthropods showed rather weak trends and was more stable, whereas the introduced species were more erratic.

I found the paper well written and the data analysis up to date. However, I have two important suggestions. First, the species subsets that you are comparing are NOT 'communities' as they are usually defined in ecology. A community would embrace all organisms that co-occur (and potentially interact) in space and time. Better use a neutral descriptive term like 'subsets' or 'assemblages' THROUGHOUT the paper for the partitions you are analysing. For example: Fauth, J. et al. (1996), The American Naturalist, 147(2), 282-286.

Second, as you rightly state (Line 122) Malaise traps usually are flight-interception devices. However, all the non-insect arthropods and also some insects (Microcoryphia; worker caste specimens of ants) mentioned in your paper are flightless. That to me raises the question how representative Malaise trap catches are for these non-volant arthropods at all. Is there something special about the SLAM traps that makes them more suitable to monitor also flightless (i.e. ground-dwelling) arthropods, as compared to standard Malaise traps? Most ecologists would set up a monitoring programme for exactly the arthropod groups you address directly (L 130-134) using pit-fall traps. These are THE standard method for sampling ground-dwelling epigaeic arthropods. Please clearly explain why you opted for something different here. Moreover, would the analytical results be qualitatively different if you partition your data into volant vs. flightless arthropods? I am a bit concerned that non-volant arthropods somehow represent just a kind of by-catch of Malaise traps, and by-catches tend to be erratic in nature.

Specific comments

L 123: something appears to be missing here, no grammatically complete sentence.

L 133: typo, should read Microcoryphia

L 138: typo, should read ‘the checklist’

L 202: please explain briefly what the Gambin model is all about, since for sure many readers will not know about that. This is just another (statistical) approach to characterize the geometry of species-abundance-distributions by one single ‘curvature parameter’, conceptually not so much different from the classical Fisher’s alpha of the logseries distribution.

L 263: probably should read species per year, not individuals?

L 288: hypotheses can never be "verified" in natural sciences; you just did not test this particular hypothesis, which is all fine (to me). Yet, please adjust wording!

L 293: should read ‘confidence interval’. I did not get where I can find the information on statistical significance in the graph.

 

 

Style and grammar are fine to me (not being a native speaker of English language)

Author Response

Response to Reviewer 2 Comments

We want to thank the reviewer for his/her effort reading the manuscript and making suggestions and corrections, as it made the article clearer and more organized, without a doubt a better version of it.

Point 1: First, the species subsets that you are comparing are NOT 'communities' as they are usually defined in ecology. A community would embrace all organisms that co-occur (and potentially interact) in space and time. Better use a neutral descriptive term like 'subsets' or 'assemblages' THROUGHOUT the paper for the partitions you are analysing. For example: Fauth, J. et al. (1996), The American Naturalist, 147(2), 282-286.

 

Response 1:

We correct the term communities whenever it has been used to write about the subsets of the 'true' community (ecologically speaking). Therefore, we now only use this term when considering all arthropods of a given season and year. When partitioning according to biogeographical origin, we use the terms 'subset(s)' or 'assemblage(s)'.

 

 

Point 2: Second, as you rightly state (Line 122) Malaise traps usually are flight-interception devices. However, all the non-insect arthropods and also some insects (Microcoryphia; worker caste specimens of ants) mentioned in your paper are flightless. That to me raises the question how representative Malaise trap catches are for these non-volant arthropods at all. Is there something special about the SLAM traps that makes them more suitable to monitor also flightless (i.e. ground-dwelling) arthropods, as compared to standard Malaise traps? Most ecologists would set up a monitoring programme for exactly the arthropod groups you address directly (L 130-134) using pit-fall traps. These are THE standard method for sampling ground-dwelling epigaeic arthropods. Please clearly explain why you opted for something different here. Moreover, would the analytical results be qualitatively different if you partition your data into volant vs. flightless arthropods? I am a bit concerned that non-volant arthropods somehow represent just a kind of by-catch of Malaise traps, and by-catches tend to be erratic in nature.

 

Response 2:

The capture of non-volant arthropods is now discussed in the Materials and methods section. We take into account the extent of the trap's catchability as a consequence of its positioning and the structure of the environment. Indeed, the native forest of the Azores has a very complex structural complexity and SLAM traps can act as extensions of trees. Therefore, it is possible to trap common flightless arthropods while using this device to move within the forest. We include a picture as Figure 1 to give the reader a better idea of what the trap is and in what structural state it is placed.

We also take into account a long period of time during which the traps were not moved and remained in exactly the same place. Therefore, we acknowledge that they are not the best tool to sample the soil-dwelling arthropod, but their probability of being sampled is considered to be the same throughout the time period (as the traps stayed in the same conditions). In fact, even if we do not completely survey this arthropod assemblage, we can still monitor its variation over time, which was the main goal of our study.

 

Point 3: Specific comments

• L 123: something appears to be missing here, no grammatically complete sentence.
• L 133: typo, should read Microcoryphia
• L 138: typo, should read ‘the checklist’
• L 202: please explain briefly what the Gambin model is all about, since for sure many readers will not know about that. This is just another (statistical) approach to characterize the geometry of species-abundance-distributions by one single ‘curvature parameter’, conceptually not so much different from the classical Fisher’s alpha of the logseries distribution.
• L 263: probably should read species per year, not individuals?
• L 288: hypotheses can never be "verified" in natural sciences; you just did not test this particular hypothesis, which is all fine (to me). Yet, please adjust wording!
• L 293: should read ‘confidence interval’. I did not get where I can find the information on statistical significance in the graph.

 

Response 3:

• We have not found any problem with a sentence line 123.
• All the typos you mentioned have been corrected.
• We have added a small paragraph describing the GamBin model in the Materials and Methods section.
• L 288, we have adjusted our wording to use the verb "test" instead of "verify" in the sentence.
• In the graphs Figure 2 and Figure 10, significance is indicated by the presence of an asterisk next to the slope value. We have clarified the level of significance in the legend of these figures.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I have no further topics to mention with regard to this revised version. The clarification about the special methodology of Malaise trapping  applied in this case is very helpful for readers.