The primary objectives of the Twin Lake experiments have been to study geologic heterogeneities and field-scale dispersion and to provide data for developing and evaluating groundwater flow and transport models. The experiment reported here was undertaken in 1987--1988 to expand the studies of heterogeneity and dispersion from a 40-m scale investigated in 1983 to a full 270-m scale from the point of tracer injection to the groundwater discharge zone. The hydrogeologic setting, the experimental and analytical methods, and the interpretation of experimental data are the topics of discussion in this paper. The 1983 and 1987--1988 tracer tests were used to evaluate the predictive abilities of various numerical groundwater flow and transport models. The modeling is done along the mean flow direction in the vertical plane. The two-dimensional finite-element modeling was successful in simulating the hydraulic head distribution but not the groundwater motion or the tracer transport. In addition, owing to the advection-dominated nature of the groundwater flow, the transport simulation suffered from numerical dispersion. Significant improvements in simulations were obtained using method-of-characteristics- and random-walk-based computer models. The influence of dimensionality on the transport simulation was evaluated by comparing the two- and three-dimensional random-walk simulations.

It was found that over the first 40 m along the mean flow path, the three-dimensional simulation does not reproduce the plume migration any better than the two-dimensional simulation does. Random-walk simulations of the transport observed in the 1983 and 1987--1988 tests are also done on the basis of the advection-dispersion equation and tracer test--derived velocities without invoking the concept of hydraulic conductivity and Darcy's law. These simulations were found to be superior to simulations based on the coupled flow and transport models. It was also found that the simulated hydraulic head distribution is insensitive to spatial variations in hydraulic conductivity and the simulated groundwater velocity distribution is inappropriate for transport modeling. It was concluded that the advection-dispersion models of the local-scale transport perform better in predicting the tracer migration at the Twin Lake aquifer using measured velocities rather than velocities simulated by a flow model. ¿ American Geophysical Union 1993