Modeling multicomponent NAPL transport in the unsaturated zone with the constituent averaging technique

Using composite constituents to simulate the average behavior and fate of groups of individual hydrocarbons in the subsurface may provide the means for dealing with mixtures of tens or hundreds of individual hydrocarbons while reducing the computational cost. There are two additional sources of uncertainty associated with this technique, namely, the time dependence of the effective properties of a composite constituent due to temporal composition changes, and the approximation in determining their effective property values. In this work, an approach to limit these sources of uncertainty is described and certified with published experimental data, that involved an artificially mixed kerosene. An inverse analysis is used to estimate the effective property values of composite constituents, and a methodology is introduced for optimizing component grouping, based on the minimization of an objective function. Aqueous solubility is found to be the optimal grouping criterion for the multicomponent organic mixture considered here. Under the assumption of equilibrium partitioning among phases, the effective property values of the composite constituents are expressed in terms of the property values of the member components using a step-wise inverse analysis. Simple arithmetic averages are found to provide poor approximations to the effective group properties and alternative schemes are proposed, based on experimental data.