Enhancing success in grassland restoration by adding regionally propagated target species

•Transfer of hay and threshing material moderately increased the target species number.•Additional sowing led to a distinct higher establishment of target species.•Introduced species were able to colonize the adjacent untreated grassland.•We recommend the applied strip-approach for large-scale meadow restoration.

The majority of grasslands of European interest are currently assessed as having an unfavorable conservation status. In order to fulfil the targets of the EU Biodiversity Strategy to 2020 effort to restore the diversity of species-poor grasslands are required. Besides the improvement of site conditions by management agreements the active introduction of target species by the transfer of on-site harvested plant material is recommended. Starting in 2009, we tested different methods to optimize species-introduction on example of a lowland hay meadow located in Saxony-Anhalt, Germany.We set up an experiment to introduce target grassland species in prepared strips with four restoration variants: hay transfer, sowing of threshing material, and both methods combined with additional sowing of a regional seed mixture. We expected that (1) all applied methods lead to increasing number and cover of characteristic grassland species, (2) hay transfer is more successful than transfer of threshing material, (3) the highest number and cover of target species is reached after additional sowing of regional seed mixtures, and (4) species spread out from treated strips into the adjacent untreated grassland. We evaluated the success of the restoration measures yearly over a timeframe of six years.The transfer of hay and the use of threshing material moderately increased the target species number. However, there were only slight differences concerning the establishment of target species between both introduction methods. In contrast, additional seeding of the regional seed mixture had a significant positive effect on the number of established target species. Many of the target species were solely established by additional seeding or were more frequently found on additionally seeded plots.Species spread into the adjacent undisturbed grassland during a period of six years. The frequency of target species recorded in the adjacent undisturbed grassland was clearly related to the distance to treated strips, thus strips served as seed-source within the matrix of the existing low-diversity grassland. However, as high frequencies were mostly found only a few metres apart from strips, the spread of species across an existing low-diversity grassland have to be seen as long-term process and should be facilitated by appropriate management measures.Depending on the availability of suitable donor sites and regionally propagated seed material, restoration practitioners and farmers can choose the most advantageous method of species introduction as all applied methods led to an increase in number and cover of characteristic grassland species compared to the untreated control. We recommend additional seeding especially if the species inventory of the donor sites is not entirely consistent with the target community, when abundances of specific target species on available donor sites in the surrounding area are too low, target species are difficult to harvest on-site because of a very early, respectively late, seed-setting time, or the donor grasslands are simply lacking some specific target species.