Partly taken from the materials and methods of https://doi.org/10.1016/j.baae.2022.12.003:To compare the activity densities of ground-dwelling predators between treatments with and without RAPs, carabids were sampled using pitfall traps, which were set up after each round of aphid counting (one per plot, twice per year; Brown & Matthews, 2016). The traps (with a volume of 400 ml and a width of 90 mm) were filled with a mixture of water and ethylene glycol (1:1; 120 ml) and dug at ground level into the middle of each plot. The traps were covered with a plastic roof and a metal grid (15 × 15 mm grid size) to avoid overflowing during rain and accidental rodent catches (Császár et al., 2018). The traps were activated for 7 days. Subsequently, all arthropods were transferred into 70% ethanol. Carabids were identified to species according to Hůrka (1996). Carabid feeding behavior was classified according to Homburg et al. (2014). To simplify the dataset, carabid feeding behavior was classified as predominantly granivorous (species mainly feed on seeds and fruits) or as carnivorous/omnivorous, because carnivorous and omnivorous species are potentially feeding on aphids and other non-plant material.