SUMMARYTwo numerical methods for solving the problem of solute transport in unsteady flow in unconfined aquifers are studied. They are the method of characteristics (MOC) based on the finite difference method (FDM), and the finite element method (FEM). The FEM is further subdivided into four schemes: moving mesh, pseudo-Lagrangian (FEM1); stationary mesh, pseudo-Lagrangian (FEM2); pseudo saturated-unsaturated, Eulerian (FEM3); and non-stationary element, Eulerian (FEM4).Experiments on a one-dimensional flow case are performed to illustrate the schemes and to determine the effect of discretization on accuracy. In two-dimensional flow the above methods are compared with experimental results from a sand box model. Results indicate that for a similar degree of accuracy, the FEM requires less computational effort than the MOC. Among the four FEM schemes, FEM4 appears to be most attractive as it is the most efficient and most convenient to apply.

No abstract

Experiments in a sand-filled flume simulate contaminant movement in an unconfined aquifer from a long strip recharge basin. Sodium chloride is used as a tracer with its distribution being monitored via electrical conductivity measurements controlled and stored through a minicomputer. The effects of initial saturated depth and magnitude of the recharge rate on the pattern of contaminant movement are discussed. Comparisons with results from numerical models are provided for the speed of movement of the contaminant front and the thickness of the mixing zone as well as for the rise in the free surface level. INTRODUCTIONGroundwater pollution is usually less apparent than surface water pollution and is also more difficult and more costly to reverse. Unconfined aquifers are particularly susceptible to pollution from surface sources when little or no cleansing is afforded by overlying strata as the polluted water infiltrates down to the aquifer. The pollutant moves through the aquifer by convection and dispersion, and a knowledge of hydrodynamic dispersion in groundwater flow is required for accurate prediction of the extent of pollution at any specified time.Some early experimental investigations were concerned with fundamental characteristics of dispersion coefficients as exemplified by the papers of Ruiner [1962], Harleman and Rumer [1963], Banks and Ali [1964], and Hoopes and Harleman [1967]. Experimental studies were also extended to other conditions such as unsteady flow [Banks and Jerasate, 1962; Rumer, 1962] and two-dimensional dispersion [Bruch and Street, 1966]. Well flow situations have been dealt with by Hoopes and Harleman [1967], Kumar and Kimbler [1970], and Gelhar et al. [1972].The present investigation covers laboratory work on dispersion in groundwater domains with a moving free surface, a case which, in spite of its common occurrence, has received little attention.The study is confined to recharge of an idealized unconfined aquifer through an infinitely long spreading basin. The hydraulic conductivity, porosity, and average grain diameter are constant throughout the flow domain. The recharge water or the invading fluid is assumed to be nonreactive and of identical temperature, viscosity, and density to the host fluid. The two-dimensional flow in the aquifer is represented experimentally with a narrow sand-filled box.The experimental results are compared with theoretical predictions from numerical models developed for this problem. The experimental procedure and results are presented

SUMMARYTwo numerical methods for solving the problem of unsteady flow in unconfined aquifers are studied. They are an explicit finite difference method (FDM), and the finite element method (FEM). The FEM is further subdivided into three schemes: vertical displacement approach, explicit scheme (FEMl), normal velocity approach, explicit scheme (FEM2), and vertical displacement approach, implicit scheme (FEM3). Results from the above methods are compared with experimental results from a sand box model. Various factors affecting the accuracy and numerical stability are investigated. Results indicate that, for a similar degree of accuracy, the FEM requires less computational effort than the explicit FDM. Amongst the three FEM schemes, FEM3 appears to be most attractive as it is the most stable and economical of the three schemes compared.

No abstract