Increased development in forestry, oil and gas, road infrastructure, and agriculture sectors across the Swan River Watershed (Alberta, Canada) has led to an increase in the impact they have on the riparian and aquatic ecosystems. These industries require the removal of vegetation for construction and operation, and some do not require buffer zones around waterbodies to protect aquatic habitat. These industries also use herbicides and fertilizers that may contain high levels of heavy metals, as well as glyphosate. In this thesis, I examined the changes to abundance of plant species chosen by Knowledge Keepers of the Swan River First Nation, as well as general plant categories (i.e., shrub, trees, and herbaceous), in relation to industrial development. I also examined how heavy metal and glyphosate content changed in these plant species with distance from industrial development. Sixty-seven sites were sampled across the Swan River Watershed, each with seven transects examining plant abundance. Tissue samples of chosen species were collected on the first, fourth, and seventh transects. I found significant ( = 0.05) changes to the abundance of aquatic and riparian plant species and categories in association with industrial activity. I also found significant ( = 0.05) changes to heavy metal concentrations in response to industrial presence and distance. There was a significant ( = 0.05) increase in the presence of glyphosate in plant species when forestry or agriculture were present at a site. These results were obtained after conducting a comparison of multiple Bayesian and frequentist regression analysis. There has been an increased interest in Bayesian analysis in ecology, however there is still some hesitation around its implementation due to hardware and software costs, time, and education. The Bayesian method resulted in smaller root-mean-squared-errors and increased precision. I also found that the time and costs were the same as the frequentist analysis, when using a dataset collected over one field season. The biggest barrier in the implementation of Bayesian analysis was the lack of accessible education through formal university courses.
