A comparison of solute-transport solution techniques and their effect on sensitivity analysis and inverse modeling results

Five common numerical techniques for solving the advection-dlispersion equa tion (finite difference, predictor corrector, total variation diminishing, method of characteristics, and modified method of characteristics) were tes ted using simulations of a controlled conservative tracer-test experiment t hrough a heterogeneous, two-dimensional sand tank. The experimental facilit y was constructed using discrete, randomly distributed, homogeneous blocks of five sand types. This experimental model provides an opportunity to comp are the solution techniques: the heterogeneous hydraulic-conductivity distr ibution of known structure can be accurately represented by a numerical mod el, and detailed measurements can be compared with simulated concentrations and total flow through the tank. The present work uses this opportunity to investigate how three common types of results-simulated breakthrough curve s, sensitivity analysis, and calibrated parameter values-change in this het erogeneous situation given the different methods of simulating solute trans port. The breakthrough curves show that simulated peak concentrations, even at very fine grid spacings, varied between the techniques because of diffe rent amounts of numerical dispersion. Sensitivity-analysis results revealed : (1) a high correlation between hydraulic conductivity and porosity given the concentration and flow observations used, so that both could not be est imated; and (2) that the breakthrough curve data did not provide enough inf ormation to estimate individual, values of dispersivity for the five sands. This study demonstrates that the choice of assigned dispersivity and the a mount of numerical dispersion present in the solution technique influence e stimated hydraulic conductivity values to a surprising degree.