Characterization of the behavior of a sustainable urban drainage system treating road runoff through hydrologic modeling and soil contamination cartography

Decentralized stormwater management generally involves the use of infiltration systems, the comportment of which is not always correctly understood. Incorrect assumptions about the hydrologic behavior of such systems can undermine their roles in both the quantitative and qualitative management of runoff, especially in systems designed to treat pollution at the source. Given the technical difficulty of directly measuring water fluxes in already existing systems, this paper discusses the utility of two indirect methods, applied to a real site made up of a vegetative filter strip followed by a biofiltration swale: the use of a hydrologic model and the characterization of soil pollution. The hydrologic model involves uncertainties due to its simplifying hypotheses and the variability of soil hydraulic conductivity measurements. Still, it can provide a rough estimation of the relative frequency of infiltration in different parts of the system. The cartography of contaminants, which shows that the majority of pollution is accumulated close to the roadside, indirectly provides time-integrated information as to the repartition of infiltration fluxes and the processes of sedimentation and filtration at the study site. The high spatial resolution of soil contamination data also provides evidence of heterogeneous surface runoff and infiltration flows. Despite the limits of each approach, both methods lead to the conclusion that the majority of runoff volumes-and as a consequence, the majority of pollutant loads-is infiltrated in the part of the system nearest to the system inlet: the vegetative filter strip. Therefore, the biofiltration swale does not fulfill its initially intended role of depolluting runoff water, as most pollutants do not reach it but rather are retained in the surface soil close to the roadside.