Grazing and plant recruitment in semi-natural grasslands - a tradeoff between competition and disturbance

In grazed habitats is a tradeoff between competition (decreases with the grazing intensity) and disturbance (increases). This may give tradeoffs between plant life stages. The seed production can be assumed to decrease at high disturbance intensities because of herbivory of reproductive organs, but also at high levels of competition (low grazing intensity) because of resource limitation. The establishment of new plants, in contrast, can be expected to be favoured by high disturbance intensities because of reduced light competition and litter accumulation. In pastures subject to Swedish subsidiaries for management of semi-natural grasslands, it is regularly controlled that the grazing is intense enough, and continuous grazing from May is usually advised. It is normally not controlled that grazing is weak enough to allow sufficient seed production. The subsidiary program is thus implicitly assuming establishment to be more important, or more of a bottleneck for vascular plant populations than seed production. Here, results from three studies exploring the tradeoff between disturbance and competition in semi-natural pastures are compiled. The first study focuses on fruit production and seedling density in three grazing regimes, representing different levels of competition and disturbance (continuous grazing MayOctober — low competition, high disturbance, thin litter layer; one-year interruptions of grazing — high competition, no disturbance, thick litter layer; and grazing mid-July to October — high competition, late disturbance, moderate litter layer. The second study compares species densities in two of the management regimes, continuous grazing and late grazing after six years of experimental grazing. The third study is a clipping and sowing experiment on one model species, and examines the general relationship between damage intensity, seed production, and establishment of seedlings. The first study shows that fruit production was increased 7-14 times by excluding grazing during one season, and 4-5 times by delaying the onset of grazing until late July. In spite of thicker litter layers, the net effect of reduced disturbance intensity and increased competition was an 3-4 times increased seedling density in both alternative grazing regimes, compared to continuous grazing. The late grazing treatment enhanced plant reproduction, which had after six years significantly increased species density. Many species showed significantly increased frequencies, whereas no species had been significantly less common in late grazing. The study of the relationship between disturbance and competition in the study species Gentianella campestris showed that disturbance was more influential than competition for the seed production. Seed production was a linearly decreasing function of the disturbance intensity. Seedling establishment, in contrast, was a threshold function of the disturbance intensity: the probability of establishment was not affected by the disturbance intensity (litter layer) until the disturbance was weak enough to produce a litter layer > 20–25 mm, corresponding to a vegetation height of c. 10–15 cm. In conclusion, a relaxation of the grazing intensity may in many Swedish semi-natural pastures have positive effects on plant diversity.