Deriving phytoplankton size classes from satellite data: Validation along a trophic gradient in the eastern Atlantic Ocean

In recent years, the global distribution of phytoplankton functional types (PFT) and phytoplankton size classes (PSC) has been determined by remote sensing. Many of these methods rely on interpretation of phytoplankton size or type from pigment data, but independent validation has been difficult due to lack of appropriate in situ data on cell size. This work uses in situ data (photosynthetic pigments concentration and cell abundances) from the north-east Atlantic, along a trophic. gradient, sampled from 2005 to 2010, as well as Atlantic Meridional Transect (AMT) data for the same region, to test a previously developed conceptual model, which calculates the fractional contributions of pico-, nano- and micro-plankton to total phytoplankton chlorophyll biomass (Brewin et al., 2010). The application of the model proved to be successful, as shown by low mean absolute error between data and model fit. However, regional values obtained for the model parameters had some effect on the relative distribution of size classes as a function of chlorophyll-a, compared with the results according to the original model. The regional parameterisation yielded a dominance of micro-plankton contribution for chlorophyll-a concentrations greater than 0.5 mg m(-3), rather than from 1.3 mg m(-3) in the original model. Intracellular chlorophyll-a (Chla) per cell, for each size class, was computed from the cell enumeration results (microscope counts and flow cytometry) and the chlorophyll-a concentration for that size class given by the model. The median intracellular chlorophyll-a values computed were 0.004, 0.224 and 26.78 pg Chla cell(-1) for pico-, nano-, and micro-plankton respectively. This is generally consistent with the literature, thereby providing an indirect validation of the method based on pigments to assign size classes. Using a satellite-derived composite image of chlorophyll-a for the study area, a map of cell abundance was generated based on the computed intracellular chlorophyll-a for each size-class, thus extending the remote-sensing method for mapping size classes of phytoplankton from chlorophyll-a concentration to mapping cell numbers in each class. The map reveals the ubiquitous presence of pico-plankton, and shows that all size classes are more abundant in more productive areas. (c) 2013 Elsevier Inc. All rights reserved.