Bienen, Wespen und ihre Gegenspieler in Kaffee-Anbausystemen auf Sulawesi: Bestäubungserfolg, Interaktionen, Habitatbewertung

In a comparison of 24 agroforestry systems, differing in plant diversity, light intensity, and forest distance, the effects of local and regional land use on bee- and wasp-communities and their ecological services are examined.The first part of the thesis shows clear evidence for pollinator limitation in the self-sterile lowland coffee and the self-fertile highland coffee. Pollinator diversity was affected by two habitat parameters indicating guild-specific nesting requirements: the diversity of solitary bees increased with increasing light intensity of the agroforestry systems, whereas the diversity of social bees decreased with increasing forest distance. These results reflect the pollination success on coffee with higher fruit set in agroforestry systems, which provide sunlight and are located closest to the forest. Both coffee species showed relations between pollinator diversity and fruit set. This result gives therefore an important contribution to the diversity-function discussion.The second part of the thesis analyses the diversity and abundance of trap-nesting bees, wasps, and their interactions with natural enemies (host-parasitoid), and the interactions with food resources, that are used for the larvae inside the trap nests (predator-prey). Both bee and wasp diversity were negatively affected by increasing forest distance but increased with light intensity of the agroforestry systems. The highest trophic level (diversity of parasitoids, and percentage parasitism) showed even a stronger relationship with forest distance than their host did. These results show that first, higher trophic levels are more affected from environmental changes such as isolation from natural habitats and second, agroecosystems close to natural forest should be more efficient in the natural control of pest insects. The quality of the 24 agroforestry systems was examined with respect to food availability by measuring the foraging trip durations of three trap-nesting species (a pollinator, a cacao caterpillar-hunting predator, and a spider-hunting intraguild predator). The foraging trip duration indicates that the selection of nesting sites is not always related to food availability and also depends on other factors. Therefore, some species may only survive in landscapes including different habitat types that include multiple required resources.The results lead to the following conclusions: Natural forest and forest fragments should be preserved in the vicinity of coffee agroforestry systems, so that forest-nesting social bees, pest predators, and parasitoids can bridge the distance to the agroforestry systems. Farmers should grow coffee beneath a diversity of shade tree species, but also provide sunlight to promote optimal microclimatic conditions, flowering herbs, and nesting sites for solitary bees and wasps.