Contribution of picoplankton to phytoplankton dynamics and bio-optics of the eastern Caribbean Sea

The picoplankton contribution to the total biomass has not been considered when it is estimated using remote sensors. In this work, In situ chlorophyll-a values were compared to values obtained from; (a) Sea-viewing Wide Field-of-view Sensor (SeaWiFS) imagery, and (b) different algorithms' (calculated with data from a spectroradiometer) for the Caribbean Time Series Station (CaTS) from October 1997 to August 2002. Regression analysis suggests that the Moderate-resolution Imaging Spectroradiometer (MODIS) algorithm provided the best estimate of the in situ chlorophyll-a value (r 2 = 0.67089). The SeaWiFS OC-4_v4 algorithm overestimated chlorophyll-a values when the in situ value was <0.2 μg/L and underestimate it when the in situ value was >0.2 μg/L. This is due to sampling error resulting from the use of 0.7 μm GF/F filters. Picoplankton loss through 0.7 μm filters (Whatman GF/F) compared to 0.2 μm membrane filters (Millipore TCMF) was quantified for oceanic stations of the North Eastern Caribbean Basin and Mayagüez Bay. On average, a 20% loss of picoplankton in oceanic stations, and a 9% loss for coastal stations were observed. Size fractionated phytoplankton analysis revealed that picoplankton was the dominant size class in oceanic stations (accounting for 60–85% of the total phytoplankton biomass and 61–77% of the absorption of light by particulates), while larger phytoplankton (>2.0 μm) dominate coastal stations. Temporal and spatial variability was observed in the size distribution of the phytoplankton community in all the stations. Electrophoretic patterns of Restriction Fragment Length Polymorphism (RFLP) corroborated the variability. These results emphasize the importance of picoplankton variability when temporal and spatial scales are considered, and suggests that this group of photoautotrophs, rather than simply representing a “background noise”, constitutes an active and changing component of the microbial community in the open ocean and even in productive waters. To improve satellite estimates of phytoplankton biomass, future algorithms must take into account the contribution of the picoplankton to the phytoplankton population.