Evaluation of cyanobacteria cell count detection derived from MERIS imagery across the eastern USA Ross S. Lunetta a , Blake A. Schaeffer b, ⁎, Richard P. Stumpf c , Darryl Keith d , Scott A. Jacobs e , Mark S. Murphy f a US EPA National Exposure Research Laboratory, 109 T. W. Alexander Drive, Mail Code: E243-05, Research Triangle Park, NC 27709, United States b US EPA National Health and Environmental Effects Research Laboratory, Gulf Ecology Division, 1 Sabine Island Drive, Gulf Breeze, FL 32561, United States c NOAA National Oceans Service, Center for Coastal Monitoring and Assessment, 1305 East West Highway, Silver Springs, MD 20910, United States d US EPA National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell Drive, Narragansett, RI 02882, United States e US EPA National Risk Management Research Laboratory, Land Remediation and Pollution Control Division, 26 Martin Luther King Drive, Cincinnati, OH 45268, United States f Innovatel, Incorporated. 2800 Meridian Parkway, Durham, NC 27713, United States abstract article info Article history: Received 27 August 2013 Received in revised form 30 May 2014 Accepted 9 June 2014 Available online 7 July 2014 Keywords: Cyanobacteria monitoring Harmful algal blooms Algorithm validation MERIS Sentinel-3 Water quality Inland waters across the United States (US) are at potential risk for increased outbreaks of toxic cyanobacteria blooms events resulting from elevated water temperatures and extreme hydrologic events attributable to climate change and increased nutrient loadings associated with intensive agricultural practices. Current monitoring ef- forts are limited in scope due to resource limitations, analytical complexity, and data integration efforts. The goals of this study were to validate an algorithm for satellite imagery that could potentially be used to monitor surface cyanobacteria events in near real-time to provide a compressive monitoring capability for freshwater lakes (N 100 ha). The algorithm incorporated narrow spectral bands specific to the European Space Agency's (ESA's) MEdium Resolution Imaging Spectrometer (MERIS) instrument that were optimally oriented at phyto- plankton pigment absorption features including phycocyanin at 620 nm. A validation of derived cyanobacteria cell counts was performed using available in situ data assembled from existing monitoring programs across eight states in the eastern US over a 39-month period (2009–2012). Results indicated that MERIS provided robust estimates for low (10,000–109,000 cells/mL) and very high (N 1,000,000 cells/mL) cell enumeration ranges (ap- proximately 90% and 83%, respectively). However, the results for two intermediate ranges (110,000–299,000 and 300,000–1,000,000 cells/mL) were substandard, at approximately 28% and 40%, respectively. The confusion asso- ciated with intermediate cyanobacteria cell count ranges was largely attributed to the lack of available taxonomic data and distinction of natural counting units for the in situ measurements that would have facilitated conver- sions between cell counts and cell volumes. The results of this study document the potential for using MERIS- derived cyanobacteria cell count estimates to monitor freshwater lakes (N 100 ha) across the eastern US. Published by Elsevier Inc. 1. Introduction Cyanobacteria and their cyanotoxins are unregulated contami- nants. However, cyanotoxins are included in the US Environmental Protection Agency (USEPA) Safe Drinking Water Act “Contaminant Candidate List” (USEPA, 2013a). Cyanobacteria blooms occur world- wide and are associated with human respiratory irritation, taste and odor of potable water, and human illness as a result of ingestion or skin exposure during recreational activities. The term bloom is de- fined here as anytime occurrence may result in negative environ- mental or health consequences (Smayda, 1997). Pets, domestic animals, and wildlife are also affected by exposure to cyanotoxins, with deaths reported annually (Backer, 2002). Cyanotoxins can be found in water bodies used for drinking, aquaculture, crop irrigation, and recreation (Stewart, Webb, Schluter, & Shaw, 2006). It has been hypothesized that cyanotoxins could even be transferred to crops designated for human consumption via spray irrigation (Hunter, Tyler, Carvalho, Codd, & Maberly, 2010). Cyanobacteria blooms can be aesthetically unappealing, which is enhanced by their tendency to concentrate along shorelines where they are encountered fre- quently by the public (Chorus, Salas, & Bartram, 2000). Increasing frequency, duration and magnitude of blooms within some systems has prompted management actions. Postulated causes of these blooms include excessive nutrient loads, introduction of invasive species (Budd, Drummer, Nalepa, & Fahnenstiel, 2001), and increas- ing temperatures from climate change and variability, where warm- er surface waters favor cyanobacteria growth (Paerl & Huisman, 2008). Alterations in land-cover (e.g., urbanization) and changes in land-use practices, such as intensive agricultural practices and biofu- el mandates (e.g., 2007 Energy Independence and Security Act) can Remote Sensing of Environment 157 (2015) 24–34 ⁎ Corresponding author. Tel.: +1 919 541 5571. E-mail address: schaeffer.blake@epa.gov (B.A. Schaeffer). http://dx.doi.org/10.1016/j.rse.2014.06.008 0034-4257/Published by Elsevier Inc. Contents lists available at ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse