The assessment of the hydro-dispersive properties of a Nature-based Solutions porous media

Nature-based Solutions (NbS) are becoming very popular in literature as a promising strategy for adapting to climate change and manage stormwater in urban environment with numerous beneficial synergies. Although the benefits of using NbS, these systems are not as widespread as they should be because the water flow and solute transport dynamics are strongly dependent on the hydro dispersive properties of the medium which are not easy to determine.

In this way, this work presents several experimental investigations coupled with numerical analysis to define the hydro-dispersive properties of two soil substrates usually used in the drainage packages of Nbs.

Thus, to define the Soil Water Retention Curve (SWRC) and the Unsaturated Hydraulic Conductivity Curve (UHCC) the Hyprop device based on the modified evaporation method has been used. The traditional constrained van Genuchten-Mualem model has been used to fit the experimental points measured from the evaporation method.

Results from this experiment shown the goodness of the estimated hydraulic parameters values assessed with the traditional constrained van Genuchten-Mualem model, and this was confirmed by the low uncertainties of the individual parameters, indicated by the 95% confidence limits for the parameter values obtained, which were narrow as well as the curves goodness of fit described by the Root Mean Square Error (RMSE) which was very low.

Solute transport in the soil is governed by two main processes: advection and dispersion, both of which are essential for understanding the dynamics of contamination and solute mobility. In this way, to determine the longitudinal dispersivity (D_L) of the investigated porous media, which is a key factor in solute transport dynamics, two saturated soil columns were injected with a natural tracer (deuterium) to characterize non-reactive solute transport in the substrates. Results from these experiments show a complex interaction between a mobile and an immobile domain, as indicated by the breakthrough curves' notable tailing. Finally, the inverse parameter estimation of the HYDRUS-1D model was applied to the experimental data obtained from the soil column experiment to assess the D_L. The inverse optimization results have shown that the equilibrium models used in the optimization phase remained dependable in this instance, despite the breakthrough curves displaying notable tailing behaviour, indicating the presence of a complex interaction between the mobile and immobile flow domains.