AME 81:109-124 (2018)  -  DOI: https://doi.org/10.3354/ame01865

ABSTRACT: Many persistent bioluminescent bays (biobays) worldwide are the result of dense accumulations of bioluminescent dinoflagellates. One such biobay, Mangrove Lagoon, is a man-made lagoon in Salt River Bay, St. Croix, US Virgin Islands, that exhibits year-round bioluminescence. The causes of variations in the abundances of dinoflagellates as well as other phytoplankton in the lagoon are unknown. The purpose of this research was to quantify the seasonal changes in phytoplankton community structure in Mangrove Lagoon, with emphasis on dinoflagellates. A secondary goal was to relate phytoplankton abundances to corresponding changes in nutrient availability, salinity, and water residence time. Weekly to bi-weekly water samples were collected in 2013 (February-December) and 2015-2016 (June-August). Phytoplankton community composition was determined using a combination of microscopy, high-performance liquid chromatography photopigment analysis, and ChemTax. Nutrient and salinity concentrations were also measured, while water residence times were calculated based on predicted tidal elevations. Dinoflagellates were a consistent and major component of the phytoplankton community. Blooms (defined as a significant increase in biomass as chl a) of diatoms, cryptophytes, and chlorophytes coincided with dinoflagellate blooms. Cyanobacterial blooms occurred mostly during the summer months under high salinity conditions. There were no correlations between phytoplankton blooms and nutrient concentrations, salinity, or nitrogen:phosphorus ratio (n = 78, p > 0.05). However, dinoflagellate blooms occasionally occurred during periodic, tidally driven long water residence times in the lagoon. Phytoplankton and dinoflagellate abundance dynamics reported in this study provide insights into the potential role of physical processes driving the variability in bioluminescence in Mangrove Lagoon.

KEY WORDS: ChemTax · Caribbean · Biobay · Mangrove · Dinoflagellate · Residence time · Spatiotemporal distribution