Dispersal and establishment limitation reduces the potential for successful restoration of semi-natural grassland communities on former arable fields.
Ex-arable fields have been suggested as potential sites for re-creation of semi-natural grasslands, but information is still limited on the temporal scales needed for the natural assembly of these communities and whether colonization is related to dispersal or establishment limitation. We investigated grazed ex-arable fields of different age and adjacent semi-natural grasslands in terms of species richness of plants, community similarity, colonization pattern and recruitment ability of 16 sown grassland species. The functional trait distribution of successful and unsuccessful colonizing species was compared using five traits related to dispersal and persistence: seed mass, seed bank persistence, specific leaf area, plant height and potential for lateral spread. The youngest ex-arable fields had the lowest species richness and contained communities with the lowest similarity to semi-natural grassland. Species richness and similarity to semi-natural grassland both increased with time since grazing started on ex-arable fields, but were still significantly lower than in semi-natural grasslands even after more than 50 years of grazing. Colonization was not related to any of the investigated functional traits. The rank order of the species in terms of abundance was correlated between young and old ex-arable fields suggesting that species performance remains the same with field age. Recruitment after sowing was generally lower in ex-arable fields than in semi-natural grasslands, although the basic recruitment ability varied between species. Recruitment did not change with field age, suggesting that dispersal limitation rather than establishment limitation caused the temporal pattern of colonization. However, establishment limitation may act as a filter for colonization of all ex-arable fields, regardless of their age. Synthesis and applications. This study demonstrates that the temporal scale for natural assembly of semi-natural grassland communities in ex-arable fields extends over 50 years, even when source pools are nearby. Results suggest that a field age-independent establishment limitation, combined with dispersal limitation, cause the delayed assembly in ex-arable fields. Management that aims to re-create semi-natural grassland communities in ex-arable fields should consider introducing seeds or improving germination conditions at the sites.