Recovery of native grass biodiversity by sowing on former croplands: Is weed suppression a feasible goal for grassland restoration?

Grassland restoration on former croplands offers good opportunity to mitigate the loss of grassland biodiversity. Weed suppression can be another benefit, which becomes increasingly important because of the high recent rate of abandonment of arable lands in Central and Eastern Europe. Our aim was to evaluate the usefulness of sowing two low-diversity seed mixtures followed by annual mowing, a frequently used restoration technique, in weed suppression. We found that rapidly forming cover of sown grasses effectively suppressed short-lived weeds and their germination except in the first year. The detected dense seed bank of short-lived weeds points out the possibility and threat of later weed infestation. In the short run perennial weeds cannot be suppressed easily by sowing and annual mowing. We found that the effectiveness of seed sowing followed by mowing in weed suppression can be different on sites with different history or seed mixture. Rapidly establishing perennial weeds, such as Agropyron species were only detected in former alfalfa fields; Cirsium arvense was found in former cereal and sunflower fields but not in former alfalfa fields. We found that the rate of weed suppression and success was influenced by the seed mixtures used. In several alkali restorations the high proportion of perennial weeds was detected in year 3. In loess restorations, much lower scores were typical. This was likely caused by the different seed mixture used. The loess seed mixture contained seeds of a clonally spreading tall-grass, Bromus inermis, which could compete more effectively with clonally spreading weeds, than could short grass species with or without tussock forming. Our findings indicate that post-restoration management require carefully designed actions that are fine-tuned addressing specific threats at the site level.