Annual flower strips and honeybee hive supplementation differently affect arthropod guilds and ecosystem services in a mass-flowering crop

Intensively managed agricultural landscapes have degraded the provisioning of diverse and continuous forage and shelter habitats for arthropods and weakened the delivery of ecosystem services such as insect crop pollination and biological pest control. In response, farmers are incentivised to sow flower strips along field margins to counteract resource bottlenecks. Yet, it is poorly understood how effective this diversification practice is when combined with the supplementation of honeybee (Apis mellifera L.) hives, which is commonly used to boost insect pollination in flowering crops. Honeybees share floral resources with wild pollinators and natural enemies of pests, which could lead to competition for food resources. We sampled pollinators, natural enemies and their pests as well as estimated the benefit of insect pollination in 17 organic faba bean (Vicia faba minor L.) fields in southern Sweden either with or without sown annual flower strips and with or without added honeybee hives. In fields with flower strips, bumblebee (Bombus spp.) densities were redistributed from field edges to interiors but without affecting their overall densities. Flower strips enhanced silver Y moth (Autographa gamma L.) densities and carabid beetle Shannon diversity along the field edge, and overall spider activity density. The supplementation of honeybee hives enhanced honeybee densities, overall ladybird beetle densities, black bean aphid (Aphis fabae Scop.) densities along field edges, but deterred silver Y moths and pushed bumblebees towards the field interior. Bean mass per plant was higher in insect pollinated plants compared with bagged, self-pollinated plants. This insect pollination benefit was independent of honeybee hive supplementation and the flower strip treatment suggesting that faba bean fields were not deficient in pollinator visits. We conclude that flower strips did not provide sufficient floral resources to increase overall wild pollinator densities in faba bean fields. Yet, annual flower strips attracted and facilitated ground-dwelling predators, especially spiders, to faba bean fields, likely by providing beneficial shelter habitats. It is worth noting that 2018, in which we collected our data, was characterised by late frosts in spring followed by an unusually hot and dry summer. While these unforeseen weather conditions together with a relatively small sample size might limit the generalisation of our results, we argue that conducting experiments under such conditions provide insights into the effectiveness of agri-environmental schemes under climate change, especially considering that such weather conditions are becoming increasingly more frequent