Persisting effects of seed bed preparation and early grazing on plant communities in grassland restoration

Semi-natural mountain grasslands are species-rich and provide a wide range of ecosystem services, such as biomass production for livestock grazing and carbon sequestration. Land abandonment but also habitat destruction are threatening these ecosystems highlighting the need for restoration when degraded. The transfer of plant material containing seed collected at non-degraded reference sites is increasingly used for restoration but the appropriate methods to prepare the transfer and to manage the grassland afterwards are still under debates. This study aims at testing the effects of seed bed preparation by harrowing and livestock grazing on the establishment of transferred targets and community trajectories towards the reference system. In a previous study, first year seedling counts showed a positive effect of seed bed preparation and a marginally significant negative effect of grazing on the seedling recruitment of the transferred species. We asked whether the results on first year seedling densities translate into significant effects on plant cover and whether these effects on plant cover change in the following three years. Both grazing and harrowing effects on species establishment were tested using a full factorial split-plot experimental design on five replicate sites. Data on species richness and vegetation cover for both the whole plant community and the transferred species were analysed during four years after seed transfer. The cover of transferred species increased whereas their species richness remained unchanged. We found a weak positive effect of harrowing on total cover whereas grazing had no effect. Seed bed preparation by harrowing had a clear positive effect on transferred target species cover. The negative effect of grazing on first-year seedling recruitment of transferred target species turned into a positive effect in the fourth year. However, this effect was only apparent in the harrowed plots resulting in a significant grazing-by-harrowing interaction. The similarity of the plant communities to the reference increased from the first to the fourth year and this increase was stronger in grazed and harrowed plots. In conclusion, the results confirmed the positive effect of seed bed preparation on restoration by seed transfer and demonstrated that initial grazing exclusion is not needed in mountain grasslands of our study area.