Uncertainty in Predicting the Rate of Mass Removal Created by Soil Vapor Extraction Systems

Recently, several soil gas flow and vapor transport numerical models have been developed for use in designing soil vapor extraction (SVE) systems. This article examines how uncertainties in soil properties, specifically permeability, corresponds to uncertainties in the prediction of mass removal rates by numerical models. Scaling equations were first derived for both relevant geometric and nongeometric modeling parameters to enable the examination of the impact of uncertainties associated with spatial variations in soil properties on the prediction of mass removal rates in a somewhat general manner. Monte Carlo analyses of volatile organic compound removal from a hypothetical contaminated soil by SVE were then used to investigate the effect of system operation time and permeability variance on the uncertainty in mass removal rates as predicted by a numerical model. Results showed that uncertainty in the predicted mass removal rate increases as both mass removal increases and as the assumed permeability variance increases. These results indicate that the design of SVE system using deterministic modeling methods may not always correlate to an effective SVE system.