Optimizing a long‐term monitoring strategy to track emerging cyanobacterial blooms in a Great Lakes estuary

Abstract

Cyanobacterial harmful algal blooms (cHABs) are increasingly frequent and severe globally. Recently, the St. Louis River Estuary (SLRE) has experienced numerous cHABs, but the locations, timing, and environmental conditions associated with them are unknown. We first investigated environmental predictors of cyanobacteria through a multifaceted approach, highlighting key relationships to understand past cHABs, and then performed a detailed spatiotemporal redundancy analysis of sampling data to optimize site selection and sampling frequency for future monitoring. We employed an oversampling approach over 2 years at eight spatially heterogeneous stations. Random forest analysis paired with constrained ordination indicated that warm, late‐summer temperatures were an important predictor of cyanobacterial abundance (biovolume), and drought‐mediated stoichiometric nitrogen limitation may have facilitated dominance by diazotrophic cyanobacteria in 2023. Temporal redundancy analysis revealed no ideal sampling frequency, but late summer, weekly sampling, and as‐needed bloom response collections were deemed critical for tracking cHABs. A 30‐ to 40‐day lag effect between water temperature and cyanobacterial abundance suggested temperature as an early warning predictor of summer bloom intensity. Water quality was more spatially heterogeneous than phytoplankton communities, informing future monitoring site selection designed to comprehensively capture water quality variability and minimize phytoplankton community redundancy while prioritizing sites with past blooms. Our findings inform an optimized monitoring program for the SLRE and demonstrate a feasible monitoring optimization approach that could be developed and applied elsewhere to track risks from cHABs and other stressors.