Using Diel Patterns in Chlorophyll and Temperature to Predict the Emergence of Harmful Algal Blooms in the Lafayette River

Abstract: Harmful algal blooms (HABs) are a global problem that can have severe consequences for aquatic life and the industries that rely on it, such as aquaculture, fisheries, and tourism. In the Chesapeake Bay, one species of HAB that frequently causes blooms is Margalefidinium polykrikoides. However, the factors that contribute to these blooms are not well understood, making it difficult for state and federal agencies and stakeholders to predict emerging HAB events. Dr. Sophie Clayton, Ocean and Earth Sciences, and Lucia Tabacu, Mathematics and Statistics, propose to use a dataset collected over 11 years at a time series site in the Lafayette River to build a probabilistic model for detecting M. polykrikoides blooms using Functional Data Analysis (FDA). The dataset is variably abundant and available for analysis by the project PIs. By using FDA, the team plans to analyze the field data and leverage the fact that M. polykrikoidesmigrate vertically in the water column over the course of the day. The aim of this project is to develop an algorithm based on the 11 years of data that can provide early warning for HAB events in the Chesapeake Bay when applied to streaming chlorophyll data. Expanding efforts to preserve and improve the health of the Bay’s ecosystems can have numerous benefits for the many species that call it home, as well as the human communities that rely on it for their livelihoods. By taking steps to protect the Bay, we can ensure its protection so future generations can enjoy its resources