Résumé
In the context of pesticide pollution in hydrosystem, limiting pesticide transfer from agricultural plot to surface waterbodies appears to be crucial. Two constructed wetlands were tested at both pilot scale of an experimental constructed wetland and field scale of the outfall of a subsurface drained watershed. Tracer experiment and pesticide mass balance studies allowed us to assess the dissipation potential of a subsurface flow constructed wetland. At field scale, climatic parameters water and pesticide's flows in and out have been measured and monitored. We also recorded the conditions of implementation since we were very closed to real conditions. For this purpose inquiries addressing the various actors were carried out by sociologists. The results of the performance regarding pesticide's dissipation are given mentioning that efficiency is strongly linked to pesticide properties and hydrological transfer period (from 20 to 90% of pesticide removal). Sociologic approaches and amenities assessments have revealed unsuspected relations of the farmers with the society and the environment, and vice versa. The implementations have resulted of a co-construction where each actor had personal involvement. Even if co-construction should be a driving line, solutions for appropriate incentives and land reallocation tools should be fought with politics and authorities in order to facilitate further realisations.