ABSTRACT
Continuous urban growth and agricultural production have caused a myriad of influxes of nutrients into water bodies, resulting in widespread eutrophication issues in lakes, rivers, reservoirs, and coastal bays. These environmental impacts have often triggered the formation of Harmful Algal Blooms (HABs), some of which are toxic due to the presence of Microcystis. Microcystis is a cyanobacteria which produces the hepatotoxin microcystin after the death of the algal cells and threatens human health and the ecosystem. This chapter demonstrates the satellite-based remote sensing monitoring capability used to generate the spatiotemporal distributions of microcystin in western Lake Erie. The Integrated Data Fusion and Machine learning (IDFM) algorithm is highlighted herein for monitoring the ecosystem status by taking advantage of the synergies of spatial, temporal, and spectral information from multiple sensors. The case study aims to:
create the capability for daily monitoring of microcystin that serves as an ecological indicator in Lake Erie for sustainable development within the Great Lake region,
exhibit the use of the unique hyperspectral band to generate a noticeable increase in prediction accuracy, especially in the range of low microcystin concentrations, and
compare the monitoring effectiveness of the traditional bio-optical models to that of IDFM-based models via various feature extraction techniques.
