On the use of the FluoroProbe®, a phytoplankton quantification method based on fluorescence excitation spectra for large-scale surveys of lakes and reservoirs.

Although microscope analysis is very useful for studying phytoplankton community composition, it does not allow for high frequency (spatial and/or temporal) data acquisition. In an attempt to overcome this issue, fluorescence-based approaches that use selective excitation of pigment antennae have spread rapidly. However, the ability of spectral fluorescence to provide accurate estimates of phytoplankton biomass and composition is still debated, and only a few datasets have been tested to date. In this study, we sampled of a wide range of water bodies (n=50) in the Ile-de-France region (North Central France). We used the resulting extensive dataset to assess the ability of the bbe-Moldaenke FluoroProbe II (FP) to estimate phytoplankton community composition in lakes and reservoirs. We demonstrated that FP data yields better estimates of total phytoplankton biovolume than do spectrophotometric chlorophyll a measures and that FP data can be further corrected using the average chlorophyll a to biovolume ratio among phytoplankton groups. Overall, group-specific relationships between FP and biovolume data were consistent. However, we identified a number of cases where caution is required. We found that Euglenophytes are expected to depart from the global FP vs. biovolume relationship of the 'green' group due to varying Fv/Fm and pigment content in response to environmental conditions (photoautotrophic vs. photoheterotrophic growth). Then, it appears necessary to consider the composition of the Chromophytes community in order to obtain a good agreement between both biomass estimation methods. Finally, we confirmed the misattribution toward the 'red' group of phycoerythrin-containing cyanobacteria and the occurrence of a strong scattering in the relationship between the FP vs. biovolume of the 'blue' group that can be partly attributed to the occurrence of large colony-forming cyanobacteria (e.g., Microcystis spp, Aphanizomenon flos-aquae). We propose correcting procedures to improve the quality of data obtained from spectral fluorescence tools in the context of large-scale sampling of lakes and reservoirs.