Effects of spontaneous field margin vegetation and surrounding landscape on Brassica oleracea crop herbivory

Abstract

Semi-natural structures of agricultural landscapes are known to provide important habitat functions for natural enemies of crop herbivores. Although several functions such as resource and shelter are related to plant species, the role of the floristic composition is not well understood. Using a correlative approach, we analysed interactions between plant functional groups of spontaneous field margin vegetation and the regulation of crop herbivores. We further compared the influence of the local plant community with that of landscape structure at different scales (50–500m). We used Brassica oleracea crop plants as phytometers planted into field margins of 48 fields and observed correlations of explanatory variables (vegetation, landscape) and response variables (herbivore and natural enemy abundance, crop damage) for two consecutive years (24 fields×year−1).We observed significant positive and negative correlations between plant functional groups and variables related to crop herbivore regulation. In general, plant cover based-models provided a better fit than diversity-based models. Entomophilous flowering plants providing nectar showed a positive correlation to ladybird abundance and aphid parasitism and a negative relationship with major crop plant damage caused by leaf-chewing invertebrates. Cover and diversity of wild Brassicaceae species were positively correlated with flea beetle damage and abundance. Most correlations with plant functional groups remained significant after fitting variables of landscape structure. The density of semi-natural habitats showed predominantly positive correlations to crop herbivore regulation, whereas the density of crops was negatively correlated. The number of significant relationships decreased from natural enemy to herbivore abundance and to plant damage.Our results demonstrate the relative importance of field margin vegetation for crop herbivore regulation. The differential responses to plant functional groups support the idea that designing specific plant mixtures may improve biological control. Additional effects of landscape variables suggest that a combination of local and landscape-scale management is the most promising strategy to optimise this ecosystem service.