High-frequency monitoring and modelling of cyanobacteria dynamics in urban lakes: Application of new approaches in Lake Enghien (France) and Lake Pampulha (Brazil)
Résumé
In order to better understand the cyanobacteria dynamics in freshwater bodies, a continuous, real-time monitoring system was developed by the PROLIPHYC project (funded by the French National Agency for Research ANR). It consists in a measurement buoy equipped with meteorological sensors and immersed probes to measure water temperature, fluorescence and dissolved oxygen at 15-min time step and send them to a database. This new monitoring system offers several advantages over historical techniques: survey of remote sites, reduction of overall cost of data collection and significant increase of quantity and quality of data. In 2009, such a monitoring system was implemented in Lake Enghien (France), an urban lake (mean depth 1.3 m, 41 ha) frequently affected by blooms of the cyanobacterium Planktothrix agardhii. The water temperature and the cyanobacteria biomass were simulated with the coupled model DYRESM-CAEDYM during summer and fall. The model was calibrated with the data collected for 15 days and validated over a 5-month period. The goodness of fit between hourly measurements and simulation results showed that DYRESM-CAEDYM is an appropriate model for simulating cyanobacteria dynamics at short and medium time scales in an urban lake. The next step of our research consists in investigating the link between watershed changes and cyanobacteria dynamics. The study site is a tropical urban lake located in Brazil. The inlet and outlet of Lake Pampulha (mean depth 5.0 m, 197 ha) will be equipped with sensors to measure the flow discharge and physical-chemical parameters (temperature, turbidity and conductivity). Automatic samplers will collect water for analysis of quality parameters. The phytoplankton biomass and the water temperature in the lake will be continuously monitored by a buoy equipped with fluorescence and CTD probes. This lake-watershed monitoring associated with the lake phytoplankton modelling will be used to assess the phytoplankton response to watershed management scenarii.