Theoretical, experimental and numerical investigation of flow and solute transport in saturated porous media subjected to violation to the continuum hypothesis

The partial differential equations typically used to solve flow and solute transport in porous media are based on the continuum approach for which the pores and soil grains are replaced by a continuum with averaged macroscopic parameters. However, in order to validate the use of the continuum approach over the domain of interest certain length scale constraints need to be satisfied. Three main criteria for length scale constraints were defined in the literature. Justification for the third length scale constraint was further defined in this work as a requirement for proper averaging rather than a consequence of certain mathematical manipulations.; In cases when these length scale constraints are not satisfied, the continuum approach may not be adapted and other sophisticated approaches need to be devised. This work attempts to experimentally shed light on the behavior of such systems and show how this violation impacts our ability to simulate such systems.; The motive for this work arose when modeling the pervious surround concept at uranium Tailings Management Facilities (TMF) in Northern Saskatchewan, Canada. The pervious surround concept provides a path of low resistance so that groundwater will move around rather than through the tailings. Typically, the pervious surround is a 1 m thick permeable layer constructed with crushed rock with grain sizes of up to 20 cm. Thus, the length scale constraints may not be satisfied in this region and hence the use of traditional groundwater flow and solute transport equations questioned.; A parametric study was conducted to highlight the influence of the pervious surround parameters on the performance of this system. This study demonstrated the impact of the permeability contrast on the effectiveness of the pervious surround and the impact of the sand layer between the surround and tailings on the downgradient concentrations and mass flux.; An experimental study was conducted to illustrate the impact of violating the length scale constraints on model simulations through a comparison of experimental and simulation data. The study demonstrated that there are indeed differences between measurements and simulations based on the assumption of the validity of the continuum hypothesis in this system. The condition of the interface between the different zones was shown to significantly impact the comparisons. Unfortunately, the condition of the interface is case dependent and hence no sharp conclusion may be drawn on its influence. However, the experimental work indicated that the downgradient peak concentrations were not significantly different from simulation, which will certainly help for design and licensing purposes.