690-700 Gw S-o 03

Seepage-face boundaries are common features of unconfined ground water flow. These boundaries exist at the intersection of an unconfined aquifer and a surface water body such as a lake, river, or wetland. The existence of a seepage-face boundary is necessary to provide a physical transition between the internal water-table boundary and the external equipotential boundary when an unconfined aquifer intersects a surface water body (Bear 1972). In the vicinity of a seepage-face boundary, the phreatic surface normally rises above the elevation of the external surface water level, thereby creating an interface at atmospheric pressure where the fluid exits the porous medium. The impact of seepage-face boundaries on the ground water flow and transport characteristics in an unconfined aquifer can be significant as the ground water velocities near the seepage face can be high (Vachaud and Vauclin 1975; Clement et al. 1996). This is because the seepage face is a sharp interface that separates the internal saturated zone where the fluid pressure is positive and the external seepage zone where the pressure is zero. Despite its importance, traditional solutions of unconfined flow problems routinely ignore the existence of seepage-face boundaries because of the difficulty of incorporating the mathematical description of the boundary in the analytical solution of the problem. In addition, several commonly used ground water models such as MODFLOW, MT3D, and RT3D do not explicitly consider the presence of seepage-face boundaries (Zheng and Bennett 2002; Clement 1997). Seepage faces are formed due to the dominance of three-dimensional flow patterns near an outflow boundary (Clement et al. 1996). Therefore, if we are to gain a better insight into the influence of seepage faces on the ground water flow and solute transport processes, we need to better understand the nature of the three-dimensional ground water velocity field near the outflow boundary. Laboratory sand tank models are useful research tools to visualize the velocity field under various outflow conditions. Several investigators have used sand tank experiments to better understand the processes of ground water flow and solute transport in porous media systems. For example, radial flow tanks have been used to demonstrate the ground water flow characteristics about a pumping well (Wyckoff et al. 1932; Boulton 1951; Hall 1955). Sand tanks have also been Abstract Laboratory and numerical modeling investigations were completed to study the unconfined ground water flow and transport processes near a seepage-face boundary. The laboratory observations were made in a radial sand tank and included measurements of the height of the seepage face, flow velocity near the seepage face, travel time distribution of multiple tracer slugs, and streamlines. All the observations were reliably reproduced with a three-dimensional, axi-symmetric, variably saturated ground water flow model. Physical data presented in this work demonstrate and quantify the importance of three-dimensional transport patterns within a seepage-face zone. The results imply that vertically averaged flow models that employ Dupuit approximations might introduce error in the analysis of localized solute transport near a seepage-face boundary. The experimental dataset reported in this work will also be of interest for those who are attempting to validate a numerical algorithm for solving ground water and contaminant discharge patterns near a surface-water boundary.