Estimating ecosystem metabolism from continuous multi-sensor measurements in the Seine River

Large rivers are important components of the global C cycle. While they are facing an overall degradation of their water quality, little remains known about the dynamics of their metabolism. In the present study, we used continuous multi-sensors measurements to assess the temporal variability of gross primary production (GPP) and ecosystem respiration (ER) rates of the anthropized Seine River over an annual cycle. Downstream from the Paris urban area, the Seine River is net heterotrophic at the annual scale (-226 gO(2) m(-2) year(-1) or -264 gC m(-2) year(-1)). Yet, it displays a net autotrophy at the daily and seasonal scales during phytoplankton blooms occurring from late winter to early summer. Multivariate analyses were performed to identify the drivers of river metabolism. Daily GPP is best predicted by chlorophyll a (Chla), water temperature (T), light, and rainfalls, and the coupling of daily GPP and Chla allows for the estimation of the productivity rates of the different phytoplankton communities. ER rates are mainly controlled by T and, to a lesser extent, by Chla. The increase of combined sewer overflows related to storm events during the second half of the year stimulates ER and the net heterotrophy of the river. River metabolism is, thus, controlled at different timescales by factors that are affected by human pressures. Continuous monitoring of river metabolism must, therefore, be pursued to deepen our understanding about the responses of ecosystem processes to changing human pressures and climate.