Scale-dependent effects of grazing on the species richness of alkaline and sand grasslands

Extensively managed pastures harbour rare and endangered species and have a decisive role in maintaining grassland biodiversity. Traditional herding of local robust cattle breeds is considered as a feasible tool for preserving these habitats. We studied the scale-dependent effects of grazing on the species richness and composition of three dry grassland types in the Great Hungarian Plain: Achilleo setaceae-Festucetum pseudovinae and Artemisio santonici-Festucetum pseudovinae alkaline grasslands, and Potentillo arenariae-Festucetum pseudovinae sand grassland. We asked the following questions: (1) Does extensive grazing have a scale-dependent effect on plant species richness of alkaline and sand grasslands? (2) How does grazing affect the proportion of specialists, generalists and weeds in the three grassland types? We sampled ten sites of each grassland type, including five extensively grazed and five non-grazed sites (altogether we had 30 sites). We used a series of nested plots each consisting of 10 plots from the size of 0.01 m² to 16 m². We revealed that grazing has contrasting effects in the three grassland types, and had a considerable effect on their species richness even at small scales. In both alkaline grassland types, total species richness was overall higher in grazed plots but it increased in a similar manner for both ungrazed and grazed habitats across plot sizes. Small-scale heterogeneity likely due to the uneven distribution of grazing, trampling and defecation together with mitigated rate of competition allowed more species to co-exist even at small scales in grazed alkaline grasslands. Grazing increased the richness of specialists, but likely due to the salt stress, establishment of weeds was hampered. Open gaps formed by trampling likely supported the establishment of several specialist species such as Plantago tenuiflora and Puccinellia limosa which are typical to open alkali grasslands. Contrary, in sand grasslands, we did not detect any effect of grazing on total species richness, likely due to the adverse effect of grazing on the species richness of specialists and weeds. In contrast with the former findings we detected significantly higher species richness in 0.01 m² and 0.0625 m² plots in the grazed sand grasslands, but found no differences at larger scales. Whilst species richness of specialists was significantly decreased, richness of weeds was increased by grazing. Decrease in the specialist species richness was likely due to the lack of their evolutionary adaptation to grazing. Degradation caused by grazing and trampling together with the propagule pressure from the neighbouring anthropo-genic habitats resulted in an increased richness of weeds in the grazed sites.