Final Activity Report Summary - RSFLUX (Monitoring suspended matter fluxes in coastal waters using ocean colour remote sensing and transport modelling. An integrated approach) The general objective of the research was to develop an operational tool to quantify the fluxes of suspended sediments delivered by a river to the ocean in a complex and dynamic interface: a macrotidal estuary (the Gironde, south-west of France). Three complementary techniques were used to observe and reproduce the transport of suspended sediments supplied from the rivers to the ocean: (i) field measurements recorded on fixed autonomous stations, (ii) regular ocean colour satellite remote sensing and (iii) three-dimensional (3D) sediment transport modelling. The method was to combine field and satellite data in order to obtain relevant information about the tidal and seasonal transport of suspended sediments, then integrate the field and satellite data into the sediment transport model. A recently developed quantification algorithm was applied to daily ocean colour satellite to map the location and movements of the maximum turbidity zone (MTZ) in the estuary. The algorithm was validated using match-ups with turbidity measurements continuously recorded on fixed autonomous platforms equipped with optical sensors. Field measurements allowed observing the influence of tides and the trapping of sediments in the upper estuary during the low river flow period. Satellite observations (one full-year of data recorded daily by two sensors with a spatial resolution of 250 m) first showed the tidal transport of sediments along the whole estuary. Averaging satellite data on a monthly basis (i.e. two consecutive spring/neap tidal cycles) allowed minimising the tidal influence and thus observing the seasonal movements of sediments in the estuary. Combining field and satellite observations gave the opportunity to observe the complex dynamics of the MTZ which was trapped in the upper estuary during four months (low river flow period), moved rapidly in the centre of the estuary during the winter high river flow period then broke down in two parts and progressively extended on the central and lower parts of the estuary. The MTZ was then partly expelled to the ocean during spring peak floods while being refilled upstream by river inputs. Finally, it rapidly moved back to the upper estuary during the following low river flow period. The variations of the mean concentration of suspended sediments in the estuary (that can be observed from satellite data) closely followed this seasonal cycle. The observations obtained from the combined field and satellite data provided relevant information to calibrate a 3D sediment transport model. Results obtained with such models are quite consistent when reproducing the hydrodynamics (sea height variations, current velocities). But comparing the suspended sediment concentrations calculated and observed showed that the sediment transport simulated was not realistic. This was due to difficulties when reproducing complex flocculation processes occurring to suspended particles in estuaries and complex exchanges between the bottom and water column (deposition and resuspension processes). Different data integration techniques were tested to force the sediment transport model with field and satellite observations. As expected, satellite data associated with an optimal spatial coverage were proved to be more efficient than field data to force the model. The first method (called re-initialisation) was to impose to the model the suspended sediment concentrations retrieved from satellite data. Results proved to be satisfactory but limited to problems with sediment mass conservation. The second method was to fit several empirical parameters in the model (e.g. the particle settling velocity) during a simulation in order to minimize the differences between the concentrations calculated and observed. This technique proved to be the most efficient and satisfactory but is expensive in terms of calculation time. Currents efforts tend to simplify it to make it operational.
