4.2 Predation experiments
Grasslands and cereal crops showed no difference in predation rates, but new grasslands increased ant activity compared to cereal crops. This suggests that new grasslands embedded in agricultural landscapes are able to provide a consistent amount of biological control services. Further, we could show that vegetation density influenced the outcome of sticky card experiments more than habitat type and that predation of experimentally exposed fruit flies as a proxy for pest control is generally lower on sites with higher vegetation density, such as newly established and old grasslands. These findings support the assumption that in new and old grassland transects predatory arthropods were less dependent to feed on experimentally exposed fruit flies, as the high vegetation density of these habitats likely provided a higher supply of food resources (Kruess & Tscharntke, 2002; Siemann, 1998) compared to cereal crops.
For the management of pest control it is important to consider that biocontrol services are not provided solely by ants, but by a diverse assemblage of ground-dwelling arthropod predators in the agricultural matrix (Meyer, Heuss, Feldhaar, Weisser, & Gossner, 2019). Carabids, spiders and wasps were also able to access fruit flies glued to sticky cards and their contribution, which was not assessed in this study, is most likely the reason for the lack of a correlation between predation rate and ant activity. Nevertheless, the results of the predation experiment are relevant because ants are among the most abundant predatory arthropods in certain agricultural landscapes and account for a significant part of the arthropod biomass (Wills & Landis, 2018).