Long-term spatiotemporal variability and regime classification of Chlorophyll-a concentrations in Lake Erie using satellite products

Monitoring chlorophyll-a (Chl-a) concentrations in Lake Erie is crucial for managing harmful algal blooms (HABs) that pose risks to water quality and public health. While in-situ Chl-a monitoring continues, its ability to capture the overall Chl-a patterns across the lake is limited by local heterogeneity. Satellite products address this limitation by providing broader spatiotemporal coverage, overcoming the constraints of point-based measurements. Among the nine satellite products evaluated, MODIS-Terra was selected as the optimal choice due to its strong correlation with in-situ measurements (ρ = 0.71) and extensive long-term data availability (60.08 % pixel availability on valid dates). Using MODIS-Terra data, we analyzed the long-term spatiotemporal Chl-a dynamics in Lake Erie over a 24-year period (2000–2023). Composite patterns, derived by averaging 24 years of data for each location, revealed higher Chl-a concentrations in western Lake Erie and along the shorelines, with a decline towards the center of the lake. Chl-a concentration typically peaked in August or September during the HABs-prone period from June to October, with the timing varying by location. Furthermore, long-term trends at each location, assessed using the Mann-Kendall test, indicated an increasing trend in 34.43 % of the lake area and no trend in 65.51 %. The areas with increasing trends were primarily concentrated in the central and western regions of Lake Erie. Based on Chl-a levels and trends, we identified and classified nine Chl-a regimes across the lake. Each regime is characterized by distinct Chl-a levels and varied long-term trends, which differ from those observed in-situ measurements and from the average patterns seen across the entire lake. These findings provide a comprehensive understanding of the long-term spatiotemporal dynamics of Chl-a in Lake Erie, while emphasizing local Chl-a heterogeneity. Moreover, the long-term Chl-a analysis and regime classification can guide decision-making, supporting the prioritization of Chl-a management areas and the expansion of monitoring networks.