Epsilon Open Archive

Abstract

A large proportion of the former area of species-rich semi-natural grasslands has been lost due to agricultural intensification and abandonment. Remaining small and isolated grassland fragments are often imbedded in more or less intensively managed landscapes. The loss and fragmentation of these grasslands have profound effects on overall biodiversity, and may affect processes and functions in the entire agricultural landscape. Therefore, semi-natural grasslands are key targets for conservation and restoration efforts. Until recently, evaluations have mainly assessed restoration outcomes in terms of species richness and abundance, often only of plants. Contrasting responses among interacting species to landscape and local factors can have consequences for the recovery of community composition, species interactions and ecosystem functions in restored habitat fragments. In this thesis, I examine the response of plant and pollinator (bees; Apoidea, and hoverflies; Syrphidae) communities to restoration of semi-natural grasslands in fragmented landscapes in south-central Sweden. Restoration effects were measured in terms of species richness, species composition, community trait composition, and composition of species and interactions within plant-pollinator networks. Following restoration, plant communities recovered both in terms of species richness and trait composition. These changes were reflected in a recovery of pollinator composition within plant-pollinator networks, and also a recovery in species interactions. The trait composition among pollinators was sustained both through abandonment and restoration, but responded to structural features of the vegetation. Plant communities had higher species richness in sites adjacent to intact grasslands than in those adjoining croplands. Hoverflies were not affected by grassland connectivity, but high connectivity facilitated the recovery of sedentary and specialized bee species. Time since restoration was an important driver in the recovery of plant communities, but not for pollinators or species interactions. This thesis shows that plant and pollinator communities as well as plant-pollinator interactions can be restored. The dispersal ability of both plants and pollinators, and the resource use of pollinators within and outside grassland fragments affected species responses to grassland connectivity. Connectivity to intact grasslands does not fully reflect the availability of resources and potential source populations within fragmented landscapes. Except for intact grasslands, abandoned and restored grasslands are also important for biodiversity and ecosystem functioning in the agricultural landscape.