Assessing Temperature-Induced Changes in Arthropod Communities over One Year: A Comparative Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Major comments

The manuscript lacks clarity and transparency in the presentation of data and of the chosen statistics.

Data presentation

For the data, it is mandatory, that they are accessible. The arthropod data for 2020 are published in a separate paper and a link is given to its public repository. However, the metadata do not match:

The public metadata under

Monitoring grassland’s arthropods in a in situ climate change experimentation (Terceira, Azores, Portugal)

state: “Overall, we collected 41,351 specimens belonging to four classes, 15 orders, 60 families and 171 morphospecies (including 34 taxa identified only at order, family of genus level). Therefore, for only 137 taxa we have a scientific name associated (n = 38918).”
In contrast, the manuscript state: “A total of 7,951 adult arthropod specimens were collected from two fields over two summers (2020 & 2021), belonging to 4 classes, 17 orders, 49 families, and 106 morphospecies. A total of 89 of these morphospecies were identified to the biological species level (which will be referred to as ‘species’ in this paper), while 17 were identified at the order, family, or genus level. The 88 species-level identifications represent about 97% of the specimens (n =7,780).”

I urge the authors to make clear which data are used and explain the discrepancies. I would also suggest presenting the data as a supplement, 106 morphospecies are not that much and can easily presented in a table.

The data for the year 2021 are only available on request (see M&M lines 154-155). These data also need to be deposited in a public database and/or as supplement to the manuscript; otherwise the data-analysis cannot be reflected.

I also miss the vegetation data, the manuscript state in line 99-102 of M&M: “The two locations are classified as intensive pastures, with the mid-altitude field primarily covered by Italian ryegrass (Lolium multiflorum Lam., Poaceae) and high elevation field dominated by common velvet grass (Holcus lanatus L., Poaceae).” Have you done a vegetation assessment (species, densities)? Please provide these data.

For the temperature only a very general statement is given in M&M line 119-120: “Over the two years, the average temperature in the OTCs was 1.2°C (SD = 0.3°C) higher than in the control plots”. This is not appropriate: give the data file for the 80 plots and the two years as a supplement. Present an analysis: what is the variation interannually and between the two localities, how much of the overall variance is induced by the wind shielding panels (OTC), is this different for the mid-altitude and the high elevation field?

 

Statistics

Unfortunately, I had great problems to understand the statistic and was mainly unable to follow the decisions for inclusion or exclusion of factors. The design is nested and used in total 80 plots. In an one-to-one design 20 plots were placed in one field, where 10 plots used as control and 10 plots treated with panels acting as wind barriers and therefore increasing temperature. {Just as side note, it can be questioned whether these 20 plots are independent, reducing the sample site to 1 as no further fields were studied}. The same set up was used in a different field with altered altitude (mid-elevation vs. high elevation). The design was then repeated in the next year (2020 and 2021). It is clear from the design that the comparison is between the 10 control and the 10 treatment plots within one field. Influencing factors are year (2020, 2021) and altitude (mid versus high altitude) with the different plots as random factor. Later also biomass (dry mass of vegetation) was added.

The presented statistics deviate from this in all cases:

1. OTC influence on biomass (lines 239-246 + table 1): Only treatment as factor and year as random factor are included, altitude is excluded, and the random factor plots as well. Consequently, table 1 is a stump.
2. Effect of Year, Treatment, and Biomass (lines 252-287): It looks that the authors try to split the analysis into parts, to avoid testing the full model with all parameters included. Therefore, the table 2 summarizes the analyses for the year or year + treatment (neglecting altitude and plot) and table 3 using only treatment (without year, altitude or plot), and table 4 using biomass and sometimes add year (without altitude or plot).

I think that the statistics needs to be done on the full models, to test for all factors in parallel.

 

Minor comments:

Abstract
line 20-22: make it specific, at present it has no content
line 33-34: sentence is a repetition of the previous one
line 36-37: “other factors” Is this shown or only speculated? Make clear what you mean
line 37-40: again, make it specific, at present it has no content

 

Introduction
Lines 68-69: be precise, as you study ground dwelling arthropods, pollination is out of your scope

 

Material and Methods
The island has its peak at 1023 m, therefore I guess labeling the second field at 386 m with “high altitude” can be questioned.

Make clear that you used only one sampling event, 14 resp. 13 days long.

Lines 157-165: Why is the fresh grass biomass not presented? What was the rational for using the dry mass? Be concise with the terminology, you can weigh something, the result is a mass, see line 164 “the dry mass of the grass”.

 

Results
Figure 2a: delete

Figure 2b: show all eight columns in one row

Figure 3: Present field (altitude) also. Using identical colors across figures would be helpful, compare figure 2

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study makes an interesting contribution to the understanding of climate change by experimentally assessing the effects of microclimatic warming on arthropod communities in island ecosystems. In general, I would advise against referring to the methodological approach used in the title ('Open Top Chambers'), as this is not a paper focusing on methodology. I have a few other minor comments.

 

L33–34: Repeat the content of the previous sentence.
L49–52: This information provides a more specific framework. Please move these statements below, to the part where you refer to study islands as appropriate ecosystems that deserve to be studied.

L72–75: This information should have been presented earlier. Once you have defined islands as being vulnerable to climate change, I suggest referring to arthropods as a particular group of ectothermic animals on islands that may be particularly sensitive to rising temperatures.

L90. The Introduction should explain how arthropods are related to grass biomass. Herbivorous species should be more sensitive to changes in grass biomass. 

Figure 1: Please incorporate an image of an OTC.

L119. Confirm whether 1.9°C is consistent with the predicted warming in this area.

L126-128: This part needs to be explained separately for the control and treatment plots. In the OTS plots, the traps were positioned within a 25 cm margin around each plot. You previously mentioned that OTCs can act as barriers for flying insects, but it is not clear whether the microclimate in these locations is comparable with that inside the OTC plots.

 

L209. This is explained again below (L176).

Table 1: This table is not necessary; the information on a single model should be shown in the text.

Figures 2 and 3. Define the meaning of the treatments (C and OTC) in the legend of this figure.

Tables 2, 3, and 4. The Model Type is unnecessary as this is explained in the Methods section.

 

 

 

 

 

 

 

 

 

 

 

L76-77: This is the main objective of the study. For now, please do not mention the methodology used; simply refer to the assessment of the impact of climate change on arthropod communities in island ecosystems.

L91. The role of altitude as a factor buffering the impacts of high temperatures could be incorporated into these goals.

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

This  version is much improved in comparison to my previous review.