1 Introduction
European countries are spatially dominated by agricultural landscapes (Kleijn et al., 2012), yet the ongoing intensification of their management dramatically reduces biodiversity (Brooks et al., 2012; Cardoso et al., 2020). Even though species diversity and specific ecological traits are well known as key promoters of ecosystem functioning (Borer, Grace, Harpole, MacDougall, & Seabloom, 2017), ubiquitous ecosystem engineers, such as ants (Sanders & van Veen, 2011), are threatened by destruction (Hendrickx et al., 2007) and fragmentation of remaining semi-natural habitats interspersed between arable land (Ewers & Didham, 2006). In order to mitigate severe effects on the maintenance of ecosystem services provided by ants, such as biological pest control (Östman, Bengtsson, & Ekborn, 2001; Tscharntke et al., 2012), biodiversity restoration in modern cultivated landscapes holds great potential (Tscharntke, Thies, Kruess, Steffan-Dewenter, & Klein, 2005; Ekroos, Olsson, Rundlöf, Wätzold, & Smith, 2014). Extensively managed grassland ecosystems are among the most species-rich habitats in Northern and Central Europe (Krauss, Steffan-Dewenter, & Tscharntke, 2003) and paramount for the diversity of ants in temperate regions (Seifert, 2018). These important semi-natural environments are disappearing rapidly from European agricultural landscapes due to abandonment, afforestation and conversion to residential areas (Dengler, Janišová, Török, & Wellstein, 2014; Valkó et al., 2018). Set-aside land and other remnants of semi-natural habitats enhance the edge density in agricultural landscapes, which fosters the diversity and abundance of ground-dwelling predators like ants (Haaland, Naisbit, & Bersier, 2011; Martin et al., 2019) and correspondingly the success of biological pest control in adjacent farmland (Gardiner et al., 2009). However, yield-enhancing ecosystem services rely heavily on the ability of predator species to disperse into the agricultural matrix (Kohler, Verhulst, Van Klink, & Kleijn, 2008). The enhancement of perennial non-crop areas through newly established grassland strips, likely provides refuge habitat not only for common agrobiont species, but also for more specialized species, if they persist in the long-term (Dauber & Wolters, 2005).
Ants are eu-social insects and important consumers of herbivorous insects, which makes them a key taxon for ecosystem functioning of temperate grasslands (Wills & Landis, 2018). Many ant species can organize mass-recruitment if sufficient food sources are available (Seifert, 2018) and this spatial allocation of predatory workers enables ant colonies to respond effectively to a dynamic and heterogeneous density of prey in their environment (Way & Khoo, 1992). Yet, the importance of ants as consumers and ecosystem engineers is often underappreciated (Wills & Landis, 2018), even though they are the numerically dominant invertebrates in certain agricultural landscapes. Offenberg (2015) showed that the efficiency of ant-mediated biocontrol is comparable to chemical pesticides, which designates ants as a relevant target group towards the development of sustainable management practices of agroecosystems.
Similar to most other grassland taxa, ants are highly responsive to human impact, such as land use change (Dahms et al. 2005; Dauber et al. 2006). A recent study highlights that ant species richness, as well as functional diversity of ant communities, decreases with increasing land-use intensity in terms of mowing and grazing of grasslands (Heuss, Grevé, Schäfer, Busch, & Feldhaar, 2019). However, to maintain ant biodiversity and their role as biocontrol agents, not only the underlying mechanisms leading to the aforementioned decreases have to be elucidated, but also how habitat restoration, in terms of newly established grasslands, may affect ant communities. Along this line, it is essential to consider that colonies of all ant species in temperate regions require multiple years to establish, grow and reproduce (Dauber & Wolters, 2005; Seifert, 2018). Hence, in an agricultural landscape long-term set-aside area and permanent grassland interspersion between arable field crops is required to keep or bring ants back as ecosystem engineers.
Understanding the ecological function of a species in a particular habitat requires knowledge of species-specific traits, their dependence on environmental factors and ecological niches (Cadotte, Carscadden, & Mirotchnick, 2011; Gagic et al., 2015). Moreover, functional traits of ants, such as colony size (number of individuals), predation on pest insects (proportion of animal based resources in ant diet) and recruitment behaviour are closely linked to the biological control services they may provide (Perović et al., 2018). Moreover, ant functional traits are linked to species-specific responses of ants to habitat alteration and management intensity of agroecosystems (Ekroos, Rundlöf, & Smith, 2013).
This study aimed to document the development of ant community composition and functional diversity within newly established grassland strips of three years age and adjacent cereal crops. The results were compared to reference plots in traditionally used old grasslands, and control plots situated in the surrounding crops. Pest control potential was quantified in all habitats through predation experiments with sticky-cards using fruit flies as baits. Further, vegetation cover of all habitats was recorded, as a high and dense vegetation supports a higher supply of food resources (Kruess & Tscharntke, 2002; Siemann, 1998) and thus might affect the dependency of predatory arthropods such as ants to feed on experimentally exposed fruit flies.
At the start, newly established grasslands are assumed to provide habitats primarily for common agrobiont ant species (Dauber & Wolters, 2005) but they may also be colonized by habitat specialists if they persist in the long term and offer diverse ecological niches. Further, social insects such as ant colonies have high and continuous nutrient requirements and should therefore play a key role as biocontrol agents in agroecosystems. In account of that, the following research questions were addressed:
1) How do species richness and community composition of ants compare between newly established grasslands, old grasslands and surrounding cereal crops?
2) Are newly established grasslands able to develop the functional trait space covered by ant communities in old grasslands, and further the prevalence of traits related to biocontrol services?
3) How do the investigated habitat types and vegetation cover affect provided pest control services, and can predation efficacy be related to the aboveground activity of ants?