Systemic models of full-scale Surface Flow Treatment Wetlands: Determination by application of fluorescent tracers

• Tracer experiments were conducted over full-scale Surface Flow Treatment Wetlands.
• Vegetation cover influences the level of hydrodynamic dispersion.
• Three different systemic models were calibrated to simulate the observed RTD curves.
• Photolytic decay of uranine was successfully coupled to a systemic model.

In Rhin-Meuse hydrological basin (North-East of France), local authorities have encouraged the setup of Surface Flow Treatment Wetlands (SFTW) at the outlet of several small communities wastewater treatment plants. These systems are devoted to effluent polishing by providing potential pollutant mitigation effects. However, such systems are designed mainly empirically and resulting surfaces and shapes may not be optimal. In the present study, the hydrodynamic behavior of three full-scale SFTWs used for sewage tertiary treatment was assessed by means of multi-tracer experiments involving two fluorescent dyes: uranine and sulforhodamine B. Residence Time Distribution analysis shows that the three investigated wetlands displayed very different hydrodynamic properties. Mean residence times were lower in the ditches (1–3 h) than in the pond (mainly 20 h). The effective volume ratio was very low for all investigated wetlands. Sediment deposition as well as vegetation cover development may explain this result. Ditches behaved as Plug-Flow Reactors with dispersion whereas the pond underwent strong internal recirculation. The influence of vegetation cover on hydrodynamic dispersion was evidenced as it induced long tails in the tracer breakthrough curves. Three systemic model structures are proposed to describe wetlands hydrodynamics: combination of ideal reactors (Plug Flow and/or Continuous Stirred Tank Reactors) with varying degrees of complexity were able to reproduce accurately the experimental Residence Time Distributions. Combination with a first-order kinetic model allowed photochemical decay of uranine to be described. In the future, combination of the proposed hydrodynamic models with more complex kinetics models will constitute a valuable tool for process understanding and optimization.