PARTITIONING INTER-WELL TRACER TEST TO ASSESS THE DISTRIBUTION OF A RESIDUAL DNAPL SOURCE IN THE BORDEN AQUIFER

During the past decade, partitioning inter-well tracer tests (PITTs) have evolved as a promising technique for the determination of the mass, distribution and morphology of 3-D heterogeneous dense non-aqueous phase liquid (DNAPL) source zones and hold considerable promise for assessing the effectiveness of remediation attempts.

To verify their applicability under field conditions, a PITT was conducted in the summer of 2004 in the unconsolidated Borden sand aquifer. Here, a well-characterized heterogeneous residual DNAPL source zone that resulted from the monitored infiltration of a 50L chlorinated solvent mixture in the spring of 1999 was subject to subsequent dissolution under natural groundwater flow conditions. From time and space integrated mass balances on detailed down-gradient transects, an estimated 5L (10%) of the initial DNAPL remained at the onset of the PITT.

During the PITT, 4000L of tracer mixture was injected during 3 days into four wells and four pumping wells were sampled during 60 days to assess the total mass, morphology and the spatial variability of the DNAPL source zone across the four sweep zones using partitioning and interfacial tracers. Overall, the PITT results are in keeping with independent indications on the source zone variability from ground penetrating radar reflection (GPR), continuous cores and the transect data. The calculated relative DNAPL-water interfacial areas were high as expected for this residual source zone. Also, the PITT derived pore-saturations (< 0.23%) were in the range of the core-derived, sweep zone-averaged saturations (< 0.39%) and the variation across the four sweep zones was in keeping with the considerable horizontal variability as shown by GPR reflection data. While the lower end of the range in PITT derived estimates of total remaining DNAPL (6–15L) were close to the independent estimate, PITT had the tendency to overestimate in DNAPL depleted zones. Lab experiments indicated that this mainly stems from an increased background sorption from sediments from which all initial DNAPL had dissolved but where residual Sudan IV was present, a color agent that was added to the initial DNAPL mixture. The influence of other hydrophobic, high-molecular weight components, commonly present in industrial DNAPLs, thus needs to be taken into account when using PITT for residual source zone characterization.