Mass balance evaluation of monitoring well purging: Part II. Field tests at a gasoline contamination site

Abstract The mass balance model simulations in Part I indicated that contaminant concentration data from typical ground water monitoring wells are inherently biased by mixing and averaging effects. This paper presents the results of field purging studies conducted at a site of subsurface gasoline contamination to test the mass balance purging model and to verify composite averaging effects. Samples collected from wells with different screen lengths, and from multi-level sampling pipes within both the aquifer and the sand pack of two wells provided data that confirmed mass balance inferences. The concentrations of aromatic contaminants and other water quality parameters in the monitoring wells depended on: (1) the well screen length; (2) the water levels achieved in the well during purging (as well as the volume of water purged); (3) when the well was sampled during water level recovery (as opposed to sampling method); (4) sand pack characteristics; and (5) vertical concentration variations in the ground water. The study also revealed that post-purging well concentrations underestimated ground water contamination by orders of magnitude due to composite averaging. These results indicate that typical monitoring wells provide only qualitative contaminant information, irrespective of how the wells are purged. Consequently, contaminant data collected from typical monitoring wells may only have limited value with respect to delineating, modeling and remediating contaminant plumes. As predicted by the mass balance model simulations in Part I and verified by this field study, water samples must be collected at discrete depths to quantitatively delineate ground water contamination.