Comparison between model and field diesel mass recovery in dolomite karst

Abstract

Analytical subsurface mass removal models are used to evaluate the accuracy and efficiency of remediation approaches. Generally, in field settings, the initial mass of the contaminant source is unknown and thus these analytical approaches cannot be evaluated effectively for complex field conditions. This is more difficult in karst settings, which have preferential flow and perched water tables. At a site near Fittstown, OK, several passive and active remediation technologies were used and monitored after 2000 gal of diesel spilled at an isolated municipal supply well in dolomite. Seven analytical models to estimate hydrocarbon mass attenuation or recovery were tested to predict mass removal of diesel from the various mechanisms that occurred on the site. The analytical predictions of diesel mass removal were either underestimated or overestimated, depending on the remediation technology or attenuation process. Model uncertainties relative to field data mass estimates may be due to parameter sensitivity and variability, sampling methodology, or model assumptions for the analytical solutions. While the models have various uncertainties in mass estimations, they are demonstrated to be useful to provide remediation options and decisions based on these estimates prior to active remediation.