Towards the Development of a Prognostic tool for Harmful Algal Blooms
The Laurentian Great Lakes and many Canadian inland waters have experienced a resurgence of cyanobacteria-dominated harmful algal blooms (cHABs), which negatively impact recreational uses, aesthetics, taste and odor in drinking water. The presence of toxins can also have dire repercussions on aquatic wildlife and human health. The factors that influence the occurrence and magnitude of algal blooms and toxin production (e.g., nutrient enrichment, climate change) vary in space and time and are poorly understood. Thus, our ability to predict cHABs is currently limited and represents a major challenge for the management of our water resources. Founded upon cutting-edge machine-learning and Bayesian inference techniques, this research project aims (i) to identify the factors that regulate the occurrence of cHABs; (ii) to provide predictions of cHABs under different land-use and climate change scenarios; and (iii) to obtain a probabilistic mapping of areas around the Great Lakes that are characterized by an excessively high risk of cHAB formation. To showcase this modelling framework, the intern will use data collected from the Bay of Quinte (Lake Ontario, Ontario, Canada), a system that has been experiencing water quality issues, and where the elimination of cHABs represents one of the major challenges of eutrophication management.