The scientific goal of the project is to develop methods for characterizing and predicting contaminant transport in aquifers. The sustainable management of groundwater has been a topic of greatest interest within the European strategy for improving the environment due to the growing risk of contamination of groundwater resources from a wide variety of stresses including point and diffuse sources of contamination, over-abstraction and saline intrusion. The fate of contaminants in the subsurface is controlled by the complex organization of groundwater flows, due to the existence of heterogeneous geological structures at different scales. It is also determined by (bio) chemical reactions that take place as contaminants travel in the subsurface. However, the effect of the flow heterogeneity is not currently taken into consideration in most existing models and theories, which represent a severe limitation of our predictive capabilities. However, recent theoretical, numerical and experimental developments suggest that significant progresses can be made to image the flow heterogeneity and integrate it into reactive transport models. Thus we propose an original methodology based on :(i) the development of inverse methods for imaging the flow organization in natural media from geophysical, flow and tracer measurements; (ii) the definition of reactive transport models that integrate this flow organization; (iii) the confrontation of models to relevant experimental data and field cases. Since the project results are expected to provide a long missing link between the flow organization and effective reactive transport, it will lead to more realistic, physically based models of transport in aquifers. The project will combine physical, geophysical and (bio) chemical methods. It will form the basis of a long term collaboration between the host group (Géosciences Rennes) and the previous IEF group (UPC Barcelona), that have developed different, yet highly complementary, approaches.