The role of geologic assumptions in solving complex contaminant transport problems

Abstract

Commonly used ground-water flow and con t transport computer models often assume a homogeneous, isotropic medium. Most hydrogeological systems are very complex and violate the basic assumptions underlying these models. However, even complex hydrogeological systems can be modeled if geologically defensible simplifying assumptions are made and if a large enough scale is applied. Aquifer parameters can be found which will reflect an average value for a representative volume of the aquifer system. In this work three computer models were used in succession to find selected aquifer parameters , characterize hydrologic conductivity fields and contaminant transport parameters. MODFLOW was used to characterize the ground-water flow. DK2D, an inverse solution to the con t transport equation was used to find contaminant transport parameters. The USGS Method of Characteristics was used to study the transport and fate of a DNAPL. A hydrogeologically complex site was chosen as a study site. The site is located on glacial fills overlying a confirmed aquifer in horizontally fractured dolomite and has been contaminated by several DNAPLs. Trichloroethene (TCE) was the contaminant chosen to be modeled. A systematic application of calibration and sensitivity analysis was made to check the validity of results from the models. It was found that ground-water could be head data could be matched using MODFLOW if certain geological assumptions were invoked, namely zones of hydraulic conductivities and the introduction of seeps. The TCE plume data could be matched if it was assumed that the TCE was undergoing first-order decay and linear isotherm adsorption. To use this technique to match observed data the investigator must understand both the models being used and the geology and hydrogeology of the site because the geologic setting is modified into a model compatible setting.