TDK conference 2010

Twenty percent of insects require obligately nectar and pollen of flowers at least in a certain period of their life cycle. The pollination provided by them is an important ecosystem service, since sixty-five percent of angiosperms are pollinated by insects. Pollination is not only crucial in the maintenance of wild-growing plants, but it is irreplaceable in the agricultural production as well. The pollination of eighty-four percent of the cultivated plants, and the third of the agricultural production depends on animal pollinators. The decline of pollinators (primarily bees) led to serious 'pollination crisis' in some regions around the world in the end of the 20th Century. This is caused partly by landscape-level changes, therefore investigation of environmental factors influencing the pollination at landscape level has considerable importance. Beside in situ observations, insect-pollinated plants, grown under controlled circumstances (phytometer) are used to examine pollination efficiency. Exposing these phytometers at different agricultural areas, or different landscape context allows us to observe and sample the pollinator insects. The number and dry mass of fruits evolving from the flowers and seeds allows us to estimate the efficiency of pollination. During my research I examined the number of pollinators, and pollination success at different linear landscape elements. The study was conducted in the area surrounding Kakucs, in the summer 2010. I chose grassy margins and tree lines next to dirt roads adjacent to grasslands and arable fields, with 3-3 replications in each category, resulting in a total of 12 study sites. Along 100 m long transects I assessed the abundance of pollinators (bees, bumblebees, flies and butterflies) in 15 minutes two times in July and three times in August. The flowering plant species, the number of flowers and the major vegetation features (width of the linear component, height and density of vegetation, etc.) were also recorded. To measure the pollination success, I placed five flowering radishes (Raphanus sativus) as phytometer plants in each transect. I counted the pollinators on the flowering plants three times during 15 minutes (38 observation occasions in total). According to the general linear mixed models the presence of tree-lines and the density of tree-lines increased the abundance of pollinators. The presence of adjacent grasslands had no effect on the studied insect groups. Average radish fruitset was higher next to tree-lines than in grassy margins, however, this difference was not significant. My results suggest the importance of tree-lines in agricultural areas, which might provide sufficient habitat and ecological corridor for the insects playing an important role in pollination.