Electronic Thesis and Dissertation Repository


Master of Science




Irena F. Creed


The repeated occurrence of Lake Erie’s harmful algal blooms suggests an inadequate phosphorus management system that results in excessive loads to the lake. In response, Canadian and United States’ governments have issued a new management objective, a 40% reduction in total and dissolved reactive phosphorus loads relative to 2008. To provide scientific evidence to guide managers toward achieving their management objective, we used the International Organization of Standardization (ISO) 31010 Bowtie Risk Analysis Tool to analyze the performance of the phosphorus management system. The effectiveness of agricultural best management practices (BMPs) and their adoption were combined into a Bayesian belief network model to predict watershed performance of each BMP. Then, the BMPs were analyzed for their probability of high risk phosphorus load reduction and achieving the management objective. Trade-offs were observed among the BMPs that will require decision makers to decide whether the management priority is to achieve the 40% load reduction objectives, or prevent further increase in the proportion of dissolved reactive phosphorus in the load, the identified culprit causing the repeated algal blooms.