Assessing the effects of climate change on arthropod abundance in Azorean pastures: PASTURCLIM project's baseline monitoring data

Abstract Background The data we present are part of the project PASTURCLIM (Impact of climate change on pasture’s productivity and nutritional composition in the Azores). The project aims to assess the consequences of climate change (e.g. temperature increase) on the grass production and its quality for forage, as well as to assess changes in the arthropod communities associated with the Azorean intensive pastures. An in situ experiment was set up using Open Top Chambers (OTCs), in order to simulate an increasing of temperature (average of +1.2ºC) on pastures. In this contribution, we present the data relative to the arthropod sampling. New information We provide an inventory of all arthropods recorded inside OTCs and in control plots in three intensively managed pastures dominated by grasses in Terceira Island (Azores): two of them dominated by ryegrass, Loliummultiflorum Lam. (Poaceae), located respectively at 186 m and 301 m above sea level; and one field dominated by common velvetgrass, Holcuslanatus L. (Poaceae), located at an altitude of 385 m. A total of 41351 specimens were collected. Organisms collected belong to four classes, 15 orders, 60 families and 171 species/morphospecies (including 34 taxa identified only at order, family or genus level). Therefore, for only 137 taxa, we have a scientific name associated (n = 38918). A total of 75% of the species (n = 129 species) are considered introduced (including all the species with indeterminate colonisation status that are possibly also exotic species (n = 7622)), representing 71% of the total abundance (n = 29664 specimens). A total of 19% of the species (n = 33 species) are considered native non-endemic representing 28% of the total abundance (n = 11608 specimens). Only one endemic species was sampled, the wolf spider Pardosaacorensis Simon, 1883 (1% of the species), representing 0.2% of the total abundance (n = 79 specimens). Spiders (5056 specimens) and beetles (18310 specimens) were the dominant taxa representing, respectively, 20 and 78 morphospecies. Since the main aim of this study was to have a better knowledge on arthropod communities present in Azorean pastures under a simulated temperature increase, the principal novelty of this paper is the contribution with distribution and abundance data to a baseline knowledge on the future consequences of climate changes on arthropod communities in Azorean pastures.


Introduction
Climatic changes occurring on Earth imply mainly changes in temperature (Arnell et al. 2019, Pörtner et al. 2022) and in rainfall patterns (Ohba and Sugimoto 2019, Papalexiou and Montanari 2019), which affect ecosystems as well as their biodiversity (Sharma andDhillon 2018, Habibullah et al. 2022). Grasslands used as forage crops are affected at different levels by the increase of temperature: i) Increased growth rate, in which higher temperatures can stimulate the growth rate of forage crops. As a result, grasslands can produce more forage, which can be beneficial for livestock. However, changes in seasonal precipitation would reduce these benefits, particularly in areas with low summer rainfall (Hopkins and Del Prado 2007); ii) Drought stress, in which higher temperatures comes with the higher risk of drought, which can be detrimental to the growth of forage crops. Drought stress can reduce the yield of grasslands and result in poor-quality forage; iii) Changes in plant composition (Feeley et al. 2020), in which some species may become less abundant, while others may thrive, which can alter the nutritional value of the forage, decreasing protein and mineral nutrient concentrations, as well as altering lipid composition (DaMatta et al. 2010); iv) Changes in plant phenology, in which some grasses are affected as well as their functional traits and chemical composition (Lee et al. 2013, Piao et al. 2019, Ekholm et al. 2020, Melo et al. 2022. All these factors can lead to cascading effects on biodiversity and on ecosystem services (Selvaraj et al. 2013, Banerjee et al. 2018, Garcia et al. 2018, Mirás-Avalos and Baveye 2018, Moss and Evans 2022.
Adaptation to climate change for agriculture will be definitively a crucial point to overpass in order to avoid an economic crisis in the coming years (Aguiar et al. 2018, Rivera et al. 2018, Vizinho et al. 2021. Regarding wildlife, there has been great concern for many years concerning the decline of arthropods (Wilson 1987, Brantley and Ford 2012, Seibold et al. 2019, Halsch et al. 2021. In anthropised ecosystems such as for crops or pastures, they are responsible for many ecosystem services (e.g. pollination, decomposition of organic matter, pest control and predation), but can also be responsible for ecosystem disservices (e.g. pest, parasitism, herbivory, seed predation, crop damage) (Cardenas et al. 2022, Ferrante et al. 2022. Climate changes may also affect this balance of services and disservices by inducing shifts in species composition (Harter et al. 2015). Climate changes may influence species presence/absence, fluctuation of abundances and can even favour the dominance of some species in the ecosystem with the threat of creating a boom of pest species (Buchholz et al. 2013, Sohlström et al. 2022). The risk is higher on island ecosystems because of the limited area available and the usually lower altitudinal range. Therefore, climate changes represent a real threat for island biodiversity (Harter et al. 2015, Borges et al. 2019, Veron et al. 2019, Pörtner et al. 2022. Predictions for the Azores suggest a temperature increase between 1.6 and 2.72°C till the end of the century (respectively following the two scenarios from the PRAC: RCP4.5 and RCP 8.5). Changes in the rainfall pattern are also expected due to the increase in heavy rains and storms in the winter and prolonged droughts during the summer (Costa et al. 2017).
Nowadays, the main activity in the Azores is dairy and meat production. Thus, most of the land between the sea level and middle altitude (500 m) is used for agriculture (e.g. intensive pasture and forage crops) representing 56% of the territory (Costa et al. 2017). The impact of temperature increase on the arthropod communities of Azorean pastures is unknown.
Therefore, an in-situ experiment was established to collect baseline data in order to help understand how the increase of the temperature affects the arthropod communities associated with intensive pastures in the Azores.

General description
Purpose: To provide baseline data on arthropod species richness and abundance from intensively managed pasture in Terceira Island (Azores) under natural and modified climatic conditions (e.g. increase in temperature via Open Top Chambers -OTCs). These data will allow us to assess the effects of climate change on arthropod's communities in Azorean pastures.
Additional information: Open Top Chambers (OTCs) are raised from the floor (around 5 cm) and allow free movement of all crawling arthropods around the pasture. Instead, for flying arthropods, OTCs represent an artificial barrier and data collected would present a bias due to this obstacle. Therefore, we focused on the collection of crawling arthropods using pitfall traps filled with ethylene glycol.

Project description
Title: PASTURCLIM -Impact of climate change on pasture's productivity and nutritional composition in the Azores

Study area description:
The study was conducted on the Archipelago of the Azores (North Atlantic), on Terceira Island (decimal coordinates 38.712925, -27.234912) which is the third largest island of the Archipelago with 400.2 km and a maximum altitude above sea level of 1021 m. The Azores are from volcanic origin and have a temperate oceanic climate, relatively wet with mild temperature at low altitude, all year long.

Design description:
The study areas were intensive pastures located at different elevations (Table 1). All pastures were dominated by grasses. The two fields at lower elevations (A and B) were covered by the annual ryegrass, Lolium multiflorum Lam. (Poaceae) and the field at higher elevation (C) was covered by the perennial common velvetgrass, Holcus lanatus L. (Poaceae).

Description:
The study was conducted in three intensive pastures on Terceira Island (Azores) (Fig. 1). In each field, 20 plots (1 x 1 m) were set up in an area of 100m where cattle were not allowed. Amongst those 20 plots, 10 were randomly chosen to be surrounded by an OTCs (in order to simulate an increase of +1.2ºC average), while the other 10 were considered as control plots. OTCs were built including a 1 m plot and a margin of 25 cm all around. The aim of this margin was to allow the same set-up of the pitfall traps as in the control plots (e.g. with one pitfall trap at each corner); it also allows free space for scientists to enter inside the OTCs without stepping on the plot. Temperature and relative humidity were recorded through data loggers (Easy Log: EL-USB-2) in control plots and inside OTCs.

Sampling description:
The focus of the study were the arthropods associated with pasture for foraging production. As OTCs represent a physical barrier for flying insects, our focus was made on crawling arthropods. OTCs were raised about 5 cm above the ground and allowed arthropod movement around the experimental area. Pitfall traps were then used for the sampling.
Grasses inside each plot were seasonally and manually harvested to evaluate the biomass. Therefore, pitfall traps were set up and collected before harvesting grasses.  Table 1.
Description of the locality, habitat, elevation and coordinates (in decimal degrees) of the three fields sampled in Terceira island, Azores.
Pitfalls were set up for 14 days, in each field, in the winter of 2020. During the summer of 2020, in the fields A and C, pitfall traps were set up for 14 days, while they were set up for 13 days in Field B.
Pitfall traps consisted in a 330 ml plastic cups, about 12 cm deep and 8 cm of diameter at the top (Fig. 2). Traps were filled with ethylene glycol. We used car's cooling liquid at 20% ethylene glycol and added few drops of soap to break the water tension. Specimens collected were then stored into ethanol (96%).
For each season (winter and summer), four pitfall traps were set up on each corner of each plot resulting in four traps per plot (Fig. 3). All traps were were active for 14 days, except during the summer, in field B, where the traps were active for 13 days.
In the winter (March 2020) and before sorting arthropods, the four traps of each plot were merged into one sample corresponding to the plot. For this reason, for the winter 2020 period, only the pitfall number 1 (PTF_1) appears in the column "eventID" that corresponds to four pitfall traps merged into one single sample. Then in the summer (September 2020), each pitfall trap was kept separately before sorting, resulting in four pitfalls for each plot (PTF_1; PTF_2; PTF_3; PTF_4).
In the Event table, the location ID name includes the following information: For example, the location ID "AC7_2020-09_PTF_3" corresponds to the "Field A Control Plot number 7_ collected in September 2020_ Pitfall trap _ Number 3" Quality control: After collection, specimens were stored in ethanol (96%) before sorting. Specimens, adults and juveniles, were identified in the laboratory by a trained parataxonomist (Sophie Wallon) and organised following a system of morphospecies (Oliver and Beattie 1996). Final identification was done by the senior author (Paulo A.V. Borges).

Figure 2.
A pitfall trap. The trap was then covered with a plastic dish raised from the ground to avoid overflow of the trap due to eventual rainfall (Photo credit: Sophie Wallon).
Assessing the effects of climate change on arthropod abundance in Azorean ...
For each species identified, a colonisation status (Endemic, Native (non-endemic), Introduced, Indeterminate) named as "establishmentMeans" in the Occurrence table, was attributed following .
Step description: Specimens were identified, based on the Azorean arthropods collection "Dalberto Teixeira Pombo Insect Collection (DTP), University of Azores" created and maintained by Professor Paulo A.V. Borges. A new collection reference was created, in the framework of the project PASTURCLIM, referencing each species occurring in the present dataset. If the specimen observed did not correspond to species/morphospecies recorded in any specimen already recorded in the Azorean arthropods collection or if its identification was not possible, then a new morphospecies number was attributed to that specimen (identificationRemarks in Occurrence

Column label Column description
Event ID An identifier for every single event and specific to the dataset. type The type of the related resource.
licence Information about rights held in and over the resource.
institutionID An identifier for the institution having custody of the object(s) or information referred to in the record.
collectionID An identifier for the collection or dataset from which the record was derived.
institutionCode The name in use by the institution having custody of the object(s) or information referred to in the record. collectionCode The acronym identifying the collection or dataset from which the record was derived.
datasetName The name identifying the dataset from which the record was derived. basisOfRecord The specific nature of the data record.
occurrenceID An identifier built as a "Globally Unique IDentifier". phylum Scientific name of the phylum in which the taxon is classified.
class Scientific name of the class in which the taxon is classified.
order Scientific name of the order in which the taxon is classified.
family Scientific name of the family in which the taxon is classified.
genus Scientific name of the genus in which the taxon is classified.
subgenus Scientific name of the sub genus in which the taxon is classified. specificEpithet The species epithet of the scientific name.
infraspecificEpithet Name of the lowest or terminal infraspecific epithet of the scientific name. taxonRank The taxonomic rank of the most specific name in the scientific name.
scientificNameAuthorship The authorship information related to the scientific name.

Additional information
We collected a total 41,351 specimens belonging to four classes, 15 orders, 60 families and 171 morphospecies (including 34 taxa identified only at order, family or genus level). Therefore, 137 taxa have a scientific name associated (n = 38918) (from now on "species")  Table 2.
Inventory of arthropods collected in three pastures (Fields A, B and C) in Terceira Island (Azores, Portugal) in control plots (C) and plots surrounded by an OTC (T).  Regarding the colonisation status, introduced species (also those with an "indeterminate" colonisation status that are most probably exotic species (n = 7622)) represented 71% (n = 29664 specimens) of the total abundance and 75% (129 species) of the total richness; 28% (n = 11608 specimens) of the total abundance and 19% (33 species) of the total richness were represented by native non-endemic species; finally, endemic species represented 0.2% (n = 79 specimens) of the total abundance and 1% (one species) of the total richness.

AC -Field
Spiders (Arachnida, Araneae) and beetles (Insecta, Coleoptera) were the two most diversified and abundant groups.
Our study is responding to the need to have baseline data to understand long-term insect decline patterns (Seibold et al. 2019). Setting monitoring programmes using arthropods is important for understanding and managing pest populations, detecting environmental changes, assessing the impact of management practices and identifying potential threats to biodiversity (Borges et al. 2019).