SeaPRISM observations of three freshwater lakes: impacts of cyanobacteria and atmospheric correction on remote sensing

Tim Moore, Florida Atlantic University, Harbor Branch Oceanographic Institute, Ft. Pierce, United States, Hui Feng, University of New Hampshire, Durham, NH, United States, Steven A Ruberg, NOAA, Great Lakes Environmental Research Laboratory, Ann Arbor, MI, United States, Steven Greb, University of Wisconsin-Madison, Madison, United States, Nima Pahlevan, Science Systems and Applications, Inc., Lanham, MD, United States, Sarah Bartlett, University of Wisconsin Milwaukee, Milwaukee, WI, United States, Menghua Wang, NOAA/NESDIS/STAR, College Park, MD, United States, Malcolm McFarland, Florida Atlantic University, Harbor Branch Oceanographic Institute, Fort Pierce, FL, United States and Kyle Beadle, NOAA Ann Arbor, Great Lakes Environmental Research Laboratory, Ann Arbor, MI, United States
The remote sensing of freshwater cyanobacteria blooms has emerged as an important component for detection and prediction of early stage bloom initiation and ensuing water quality hazards. Recent advances in satellite capabilities, including higher spatial resolution and improved re-visit times, have opened up image access to virtually any aquatic system in the world. A variety of bio-optical algorithms have been developed for water quality products that can be applied to these remote sensing image sets. However, in many systems, the performances of these algorithms are unknown. In addition, the satellite image quality after atmospheric correction is also unknown for many systems. A necessary aspect of using satellite data for these purposes is the validation of atmospheric correction schemes and associated water-leaving radiance products, in addition to bio-optical algorithms and products. A network of field radiometers with standardized design and data processing protocols to support satellite validation activities, called AERONET-OC, has recently expanded from ocean systems into freshwater lakes. This component of the network has its own design and data processing protocol, tailored for monitoring optical conditions common to turbid/eutrophic systems. In this work, we compare the data sets from three AERONET-OC sites located in Lake Erie, Lake Okeechobee and Green Bay in Lake Michigan. These three sites are all shallow systems that experience cyanobacteria blooms routinely. The environmental and optical characteristics of these sites share similarities, but also differences in background conditions. This study examines the performance of remote sensing products at these sites from a time series of satellite matchups in the context of cyanobacteria and non-cyanobacteria conditions. The hourly measurements from the radiometers also are examined in the context of cyanobacteria bloom phases, including impacts of buoyant colonies on the light field on short-term time scales.