Comparison of techniques used to count single-celled viable phytoplankton

Abstract

Four methods commonly used to count phytoplankton were evaluated based upon the precision of concentration estimates: Sedgewick Rafter and membrane filter direct counts, flow cytometry, and flow-based imaging cytometry (FlowCAM). Counting methods were all able to estimate the cell concentrations, categorize cells into size classes, and determine cell viability using fluorescent probes. These criteria are essential to determine whether discharged ballast water complies with international standards that limit the concentration of viable planktonic organisms based on size class. Samples containing unknown concentrations of live and UV-inactivated phytoflagellates (Tetraselmis impellucida) were formulated to have low concentrations (<100 mL⁻¹) of viable phytoplankton. All count methods used chlorophyll a fluorescence to detect cells and SYTOX fluorescence to detect nonviable cells. With the exception of one sample, the methods generated live and nonviable cell counts that were significantly different from each other, although estimates were generally within 100% of the ensemble mean of all subsamples from all methods. Overall, percent coefficient of variation (CV) among sample replicates was lowest in membrane filtration sample replicates, and CVs for all four counting methods were usually lower than 30% (although instances of ~60% were observed). Since all four methods were generally appropriate for monitoring discharged ballast water, ancillary considerations (e.g., ease of analysis, sample processing rate, sample size, etc.) become critical factors for choosing the optimal phytoplankton counting method.