About the Project
Honeybees (Apis mellifera) are a species introduced to North America for their pollination services in agro-industries. These species, while valuable economically, can have potential impacts on native pollinators and native plants. While studies have been done to show the impacts of honeybees on native bees and crop production, little attention has been given to other groups of pollinators, such as flies, butterflies, ants, wasps, and others.
This project aims to shed light on the species composition of pollinator groups in central and southern Alberta, and if the presence of honeybees impact them. To do this, we traveled between two agriculturally active sites ("ranches") in Alberta, parkland and mixed prairie, to catch pollinators as they landed on the reproductive organs of plants. Each ranch had twelve transects, with honeybee hives situated at varying distances as a treatment.
Non-metric multidimensional scaling (NMDS) ordination was completed for the statistically significant pollinator groups (ordination p-value ≤ 0.03) to find differences in pollinator composition within regions, treatments, and seasons. To compare pollinator composition between treatments and season, canonical discriminant analysis was then completed to assess for differences ordination may not have detected. Finally, to test the statistical strength of ordination groupings, Permutational multivariate analysis of variance (perMANOVA) was used. The ordination results indicated a treatment effect only in parkland, though r^2 values and perMANOVA results show some indication of treatment effect. Canonical discriminant analysis discovered some treatment effect in the late season. This information suggests that at a coarse scale, there is no effect of honeybees on pollinator group composition; the data, however, demands further taxonomic categorization and future study requires control transects be placed at further distances from hives.
Non-metric multidimensional scaling (NMDS) ordination was completed for the statistically significant pollinator groups (ordination p-value ≤ 0.03) to find differences in pollinator composition within regions, treatments, and seasons. To compare pollinator composition between treatments and season, canonical discriminant analysis was then completed to assess for differences ordination may not have detected. Finally, to test the statistical strength of ordination groupings, Permutational multivariate analysis of variance (perMANOVA) was used. The ordination results indicated a treatment effect only in parkland, though r^2 values and perMANOVA results show some indication of treatment effect. Canonical discriminant analysis discovered some treatment effect in the late season. This information suggests that at a coarse scale, there is no effect of honeybees on pollinator group composition; the data, however, demands further taxonomic categorization and future study requires control transects be placed at further distances from hives.
RESOURCES
ABOUT THE AUTHORMSc student in Conservation Biology at the U of A specializing in plant-pollinator network interactions in southern Alberta.
|
RANGELAND RESEARCH INSTITUTE"The RRI brings together expertise from within the faculty, the university, across Western Canada and beyond to conduct multi-disciplinary research and teaching in support of the long-term environmental and economic sustainability of rangelands."
|
ALBERTA NATIVE BEE COUNCIL"Our mission is to promote conservation of native pollinator communities through research and monitoring, advocacy, education, and collaboration with others."
|