Boreal Peatland LIFE -project – The effect of peatland restoration on bryophyte and vascular plant species richness, community composition and community dispersion

The diminishing area of pristine nature combined with the degraded stage of human altered landscapes has led to international targets of restoring degraded areas to safeguard ecosystems and their services. Because from the viewpoint of the communities restoration is another disturbance the effect of restoration on community assembly processes might be similar to that of disturbance. We used boreal peatlands representing three peatland types (spruce mires, pine mires and fens) with two levels of productivity to test whether this is the case. Half of the sites are in pristine stage and half had been restored after being drained for forestry for several decades. We sampled bryophyte and vascular plant species before (n = 120), two years after restoration (n = 115) and five years after restoration (n = 52). ). In addition we analyzed a few alkaline fens in which we did not have different productivity levels. On the basis of our previous study of the changes in peatland community assembly processes after disturbance we predict that restoration i) has no influence on species richness, ii) has an effect on community composition, and iii) has no influence on community dispersion (i.e. variation in community composition among sites). We used linear mixed models, permutational multivariate analysis, and the test of homogeneity of multivariate to analyze changes in species richness, community composition and dispersion in two and five years after restoration. Restoration increased Sphagnum moss abundance in spruce mires and decreased forest moss abundance in spruce mires and in alkaline fens. In addition, vascular plant communities in pine mires were affected by restoration. Despite of these changes, the general effect of restoration on bryophyte and vascular plant species richness, community composition and dispersion are still very subtle and inconsistent after five years. Thus, a long-term monitoring in order to determine whether restoration affects community assembly processes is clearly required.