Abstract
Humans play a major role shaping the living conditions for not only ourselves, but also all other species on Earth. In fact, some species-rich habitat types require human management to uphold the biodiversity and related ecosystem services. One of the world’s most biodiverse habitats on small spatial scales, semi-natural grasslands, have been formed over the course of centuries through extensive grazing and mowing. However, due to political and economic reasons, up to 90% of the European semi-natural grasslands have been lost during the 20th century. To counteract these drastic losses, restoration actions are implemented in environmental policies across Europe. Yet, knowledge of the long-term restoration effects on biodiversity and ecosystem services is still limited. The vast need for future restoration also requires a better understanding of how different pre-conditions affect the restoration outcome, as well as how stakeholders perceive restoration, to be able to prioritise between sites and recognise the limitations of the restoration process. In this thesis, I examine restoration outcomes in Swedish semi-natural grasslands, in terms of plant diversity, associated ecosystem services and from the farmers’ and land-owners’ perspective. The outcome is also analysed in relation to environmental factors at the local and landscape scale. I found that the overall community composition recovered to resemble intact reference communities, but it took relatively long time (12-20 years). Moreover, the reference sites still had higher species richness both at large and small spatial scales, more grassland specialist species and a higher abundance of plant species important to the five tested ecosystem services (meat production, pollination, water retention, temperature regulation and cultural heritage). My results show that prioritising large, unfertilised, newly abandoned grasslands situated in landscapes containing a large grassland specialist species pool and high amounts of intact and remnant semi-natural grasslands, could speed up the plant recovery. However, prioritising fast results does not necessarily ensure long-term success at a larger spatial scale. Since restoration success can be interpreted differently depending on evaluation measure used, pre-defined, clear and realistic goals are essential. While the surveyed farmers and landowners overall perceived the restoration as successful, 40% were unsure whether the grasslands will be managed in the future. Low profitability still poses a threat to their maintenance and thus, also to the coupled biodiversity and ecosystem services. Policy changes are therefore urgently needed to facilitate incentives for sustained management of restored and intact European semi-natural grasslands in a long-term perspective.

Supervisors
Lindborg, Regina, Professor
Stockholm University, Faculty of Science, Department of Physical Geography.

Cousins, Sara A O, Professor
Stockholm University, Faculty of Science, Department of Physical Geography.

Öckinger, Erik, Docent
Swedish University of Agricultural Sciences, Faculty of Natural Resources and Agricultural Sciences, Department of Ecology.

Biogeography and Geomatics Unit