| Numerical models can be useful tools for engineering safe and sustainable solutions to address problems impacting fractured media such as long-term nuclear waste storage in the subsurface, the containment and remediation of hazardous dissolved contaminants. Many simulation approaches have been proposed for both single- and two-phase flow through single fractures; however, some of these models have yet to be validated to observed flow data. This thesis contains a quantitative evaluation of many of these models through comparisons between measured and simulated single- and two-phase flow behaviour and using detailed aperture measurements.; The results of the study illustrated that using a parallel-plate representation of a rough-walled fracture is inadequate. The effect of the fracture surface roughness must be accounted for when simulating both single- and two-phase flow rates using a roughness factor to modify the intrinsic permeability. Also demonstrated was the increase in tortuosity of the flow paths and suppression of the flow rate of one fluid caused by the presence of trapped residual of the other fluid during two-phase flow. It was demonstrated that the impact of the trapped residual could be simulated by modifying the Brooks-Corey relative permeability equations. Finally, the sensitivity of the flow rates to various modelling parameters was studied, with the results illustrating that nodal spacings less than the small-scale aperture correlation length are required for the pore-scale models, and the apertures should be calculated in a direction perpendicular to the local centreline of the fracture. |
