Model investigation of DNAPL distribution in the saturated zone for varying groundwater flow velocities and subsurface geometry

Chlorinated solvents like Trichloroethylene (TCE) and Tetrachloroethylene (PCE) are among the most widespread groundwater contaminants worldwide and are often released into the subsoil as Dense Non-Aqueous Phase Liquids (DNAPL). Due to common knowledge, the DNAPL moves vertically through the vadose and saturated zone. On capillary barriers e.g. the capillary fringe, it pools until the accumulated mass exceeds the entry pressure and continues its downward movement. Encountering impermeable layers, it follows gravity-driven the inclination of the barrier. At site investigations, DNAPLs are sometimes not only observed straight below of the original contamination source zone, but also in the downstream direction. Possible explanations are the existence of a second, undocumented point source or preferential flow paths in the subsurface due to high permeable geological layers. Apart from these aspects, another possible explanation is the conveyance of the DNAPL with high groundwater flow velocities, which can occur in gravel-sandy narrowing aquifers or in aquifers which are influenced by human activities, such as pumping. Although a lot of research is done in the area of DNAPL migration in the saturated zone with respect to influences of small and large scale heterogeneities (Bradford et al., 2003; Broholm et al., 2005; Fagerlund et al., 2006; Illangasekare et al., 1995; Jawitz et al., 2005; Kueper and Frind, 1988; Page et al., 2007; Rivett and Feenstra, 2005; Saenton et al., 2002), dissolution processes (Falta, 2003; Falta et al., 2005; Fure et al., 2006; Rivett and Feenstra, 2005; Zhu and Sykes, 2000) and mass flux (Christ et al., 2009; Fure et al., 2006; Illangasekare et al., 2006; Jellali et al., 2003; Page et al., 2007; Soga et al., 2004), there has as far as the authors know, not yet been a distinct focus on the influence of high groundwater flow velocities on the distribution, geometry and location of a DNAPL source zone. Moreover, studies of Sale and McWhorter (2001) and Miles et al. (2008) revealed, that one key parameter for the dissolution process and persistence of NAPLs within the saturated zone is the geometry and orientation of the NAPL source zone within the water flow. Therefore, the knowledge of the behaviour of a NAPL under different flow regimes as well as its position and orientation within the groundwater body becomes vital for site investigation and for the conceptual design of remediation plans and actions. In our previous studies we used a coupled approach of 2D small-scale laboratory experiments and multiphase modelling with the software TMVOC (Pruess and Battistelli, 2002) to define to what extent groundwater flow velocities affect the distribution of a TCE DNAPL body in the saturated zone under homogeneous conditions (Erning et al., 2010, submitted; Erning et al., 2009; Luciano et al., 2010). The results of these first investigations are briefly shown at the beginning of chapter 3. Summarized they revealed that the position of the DNAPL pool, its length and its maximum saturation are already influenced at groundwater flow velocities of 0.05 m/d in the homogeneous set-up. However, the changes were negligible in relation to the overall size of the DNAPL pool as long as flow velocities of 5.00 m/d were not exceeded. At higher flow velocities the position of the TCE pool was delayed to the downstream direction, the pool length was elongated and the maximum saturation, as a proportion of the dissolution processes, was decreased. In this study we compare the influence of high groundwater flow velocities (> 5.00 m/d) on the DNAPL pool distribution with effects due to a stratified medium by multiphase modelling. Different realizations of horizontal impermeable lenses were introduced into the model in order to differentiate the effects of preferential flow paths, pooling effects on capillary barriers and high groundwater flow velocities. The position and the dimensioning of the DNAPL pool at high flow velocities is analyzed in relation to the pool behaviour at no flow conditions with and without the respective geometrical set-up. The focal point of interest is to evaluate to what extent the layering of the geological stratum and the groundwater flow velocities have to be taken into account for general aspects of site investigation.