High-frequency monitoring of phytoplankton and cyanobacteria in a small urban lake
Résumé
Urban lakes are important spots of recreational activities. Excess nutrient input from their watershed associated with global warming increase the occurrence of toxic cyanobacteria blooms. In bathing areas, compliance with regulatory standards for cyanobacterial biomass and toxin concentrations requires a regular toxicity assessment to limit risks. Current methods for determining potentially toxic species are based on taxonomic microscope identification. They require very specialized skills and are time-consuming. Therefore, rapid, easy to implement methods must be developed. Field sensors of chlorophyll and/or cyanobacteria-specific pigments can achieve this goal. However, sensor data do not provide information on the species; their results can be very variable depending on the hardware characteristics and in some environmental conditions, they can be affected by high errors. In order to investigate the reliability of a monitoring system of cyanobacteria blooms, a survey was performed for three years (May 2016 to December 2018), in a small experimental lake (Lake Champs-sur-Marne, Great Paris, France). High-frequency measurements of physical-chemical variables, chlorophyll-a and phycocyanin fluorescence were collected at three points. In parallel, vertical profiles of the four main phytoplankton groups were measured at the same points, with a fluorometric probe, during regular field campaigns (n=75). Lab analysis of chlorophyll-a and phycocyanin were performed. A regulatory monitoring of the bathing area supplied lab chlorophyll-a data and a semi-quantitative assessment of dominant species. This dataset is analysed in order to figure out how cyanobacteria sensor data are consistent with the other data. The sensor reliability conditions for assessing the cyanobacteria biomass are defined.