Land use changes have seriously fragmented grasslands leading to extensive biodiversity loss worldwide. Habitat fragmentation affects grasslands at both local and landscape scales, adversely affecting the probability of species colonisation and extinction. In our study, we addressed the effects of fragment size and landscape-scale habitat connectivity on the vegetation composition in two grassland types, i.e., loess steppe fragments (situated on kurgans) and sand steppe fragments of the Pannonian forest steppe region.
We collected frequency data on vascular plant species in 12 1 m × 1 m quadrats in altogether 60 fragments along a connectivity gradient in sand steppes and loess steppe fragments. We analysed whether habitat specialists, generalists and exotic species were affected by the level of fragmentation based on species richness and traits related to local persistence (life span, clonality and soil seed bank type) and dispersal (dispersal strategy and seed mass) in the two grassland types using linear mixed-effects models.
Based on our results, both fragment size and landscape-scale connectivity were important in shaping the trait composition of the vegetation. We observed more fragmentation effects in generalist than in specialist species. We found that isolation resulted in fewer specialist species with autochorous dispersal strategy in loess steppe fragments, but, at the same time, also fewer exotics. Isolated loess steppe fragments harboured fewer generalist species with persistent seed bank. Large loess steppe fragments supported more wind-dispersed species than smaller ones. In isolated sand steppe fragments, generalists were more frequent with endozoochorous dispersal strategy and without clonal propagation. Life span, clonal propagation and seed mass did not depend on the level of fragmentation in the two grassland types.
Our results imply that both sand and loess steppe fragments can rely to some extent on the persistence of clonal perennial specialist and generalist species in small and isolated patches to mitigate fragmentation effects. In conclusion, these processes should be supported by the preservation of large fragments, increase of habitat connectivity combined with targeted management of exotic species.