Parameter-induced uncertainty in modeling vadose zone transport of VOCs

This paper evaluates varying results produced by parameter uncertainty and variability in a one-dimensional vadose zone transport model, VLEACH, predicting transport of volatile organic compounds (VOCs) from a contaminated near-surface vadose zone downward to the water table. The model is applied using field data for trichloroethylene (TCE) and tetrachloroethylene (PCE) at a Superfund site in southern California. The research first evaluates sensitivity of the model results to variations in input parameters taken one at a time. Results demonstrate that parameter variations within the ranges seen in field data or derived from the literature may produce variability in model results of more than an order of magnitude. Second, the research evaluates variability in the results generated by uncertainty in parameters when varying jointly, using Monte Carlo simulation techniques. Monte Carlo results demonstrate that model outcomes are strongly affected for the selected site by uncertainty in several parameters produced by a variety of factors, including held measurements of subsurface lithography, forecast mean annual precipitation, and chemical properties of VOCs.