Data for: Perennial flower strips increase pollinator and natural enemy abundance but show limited efficacy in pest control for adjacent crops

We sampled pollinators, natural enemies, and herbivores using visual observations, yellow sticky traps, pitfall traps and tiller counts, as well as estimated predation and decomposition rates using sentinel prey cards and bait lamina strips in ten pairs of pollinator attractive perennial flower strips and control field margins, and their adjacent cereal fields in Skåne, Sweden in 2021. Field margins (flower strip vs spontaneous vegetation control) were characterized by estimating the percentage of plant cover and the total floral area (for each species we calculated the number of floral units x average floral area) in eight 0.6 x 0.6 m quadrats evenly distributed along a 100 m transect. Data was collected twice during the main growing period of the flower mixture. Pollinators (hoverflies, honey bees, bumblebees, solitary bees and butterflies) visiting flowers were surveyed for 10 minutes along a 100 m long and 1 m wide transect in each field margin type. Pollinators were surveyed twice during the main period of the flower mixture on the same days as the margin characterization was done. Leaf-dwelling natural enemies and herbivores were sampled using yellow sticky traps (20 cm x 12.6 cm). Four traps of each type were placed along the 100 m transect in the field margins and another four in the adjacent crop area, at 10 m from the margins, for a total of 16 traps per site. Traps were spaced 20 m apart within each transect and remained in the field for seven days. Data was collected twice during the main period of the flower mixture. Ground-dwelling natural enemies were sampled using pitfall traps made from polypropylene beakers (12 cm diameter) filled with 200 mL of soapy water. Four traps of each type were placed along the 100 m transect in the field margins and another four in the adjacent crop area, at 10 m from the margins, for a total of 16 traps per site. Traps were spaced 20 m apart within each transect and remained in the field for seven days. Data was collected twice during the main period of the flower mixture. We counted and identified all arthropods found on four groups of five tillers located along each adjacent crop transect, spaced every 20 m, resulting in 80 crop tillers per site. Data was collected twice during the main period of the flower mixture. Sentinel aphid cards were set up in the field to estimate aphid predation rates. Four groups of two cards at ground level and two cards at vegetation level were set up along each adjacent crop transect, spaced every 20 m, resulting in 32 cards per site. Sentinel prey cards were exposed simultaneously during the first sampling interval of the tiller counts. After 24 hours of exposure, the sentinel prey cards were collected, and the remaining aphids were counted. Decomposition rates were estimated by setting up bait lamina strips filled with a standardized bait mixture. Four groups of five strips were placed along each 100 m transect, with groups spaced every 20 m, resulting in 80 strips per site. Within each group, strips were spaced 20 cm apart. The lamina strips were buried in the ground for 15 days, coinciding with the end of the surveys. After this exposure period, we recorded the number of pierced holes and calculated the decomposition rate by dividing the number of pierced holes by the total number of bait-filled holes. All data were aggregated across samples and survey rounds for each field margin habitat and the adjacent in-crop area. For further information, see methods in the manuscript Rodríguez-Gasol et al. ’Perennial flower strips increase pollinator and natural enemy abundance but show limited efficacy in pest control for adjacent crops’.