Timothy R. Ellsworth, Stefan Mayer, Charles W. Boast, and Miguel Restrepo

A primary objective of this research is the development of a network model to characterize and predict water flow and chemical transport in soil. In particular, we are focusing on the potential for upscaling flow and transport processes. We have developed a computer model that simulates flow and transport in a three-dimensional heterogeneous flow network. This approach employs an efficient solution method, which is based on the assumption that iso-pressure surfaces are known a priori (for instance, steady-state, gravity-driven vertical flows in unsaturated soil with horizontal iso-pressure surfaces). The flow and transport heterogeneity in this system is thus assumed to be a result of conductance heterogeneity.

The flow and transport processes have been simulated for two spatial network conductance distributions: (1) an Uncorrelated Random Network conductance distribution (URN), and (2) an Anisotropic, Correlated Random Network conductance distribution (ACRN). These preliminary simulations of flow and transport employed a network consisting of 20 x 20 x 40 grid nodes. Particle tracking was used to characterize solute transport in the network. As expected, simulated breakthrough curves (BTC) for the URN rapidly converged to be consistent with a convective-dispersive (CDE) process. A 2-3 times greater vertical, relative to horizontal, correlation in conductance was present in the ACRN simulation. This resulted in a faster mean travel time, and also considerably greater tailing of solute relative to the URN transport simulations. Both distributions assumed a lognormal conductance distribution.