Articles | Volume 15, issue 7
Hydrol. Earth Syst. Sci., 15, 2229–2244, 2011
https://doi.org/10.5194/hess-15-2229-2011
Hydrol. Earth Syst. Sci., 15, 2229–2244, 2011
https://doi.org/10.5194/hess-15-2229-2011

Research article 18 Jul 2011

Research article | 18 Jul 2011

Modeling subsurface transport in extensive glaciofluvial and littoral sediments to remediate a municipal drinking water aquifer

M. Bergvall1,2, H. Grip1,*, J. Sjöström2, and H. Laudon1 M. Bergvall et al.
  • 1Department of Forest Ecology and Management, Swedish University of Agricultural Sciences, Umeå, Sweden
  • 2Swedish Defense Research Agency, Umeå, Sweden
  • *now at: Surbrunnsgatan 4, 11421 Stockholm, Sweden

Abstract. Few studies have been carried out that cover the entire transport process of pesticides, from application at the soil surface, through subsurface transport, to contamination of drinking water in esker aquifers. In formerly glaciated regions, such as Scandinavia, many of the most important groundwater resources are situated in glaciofluvial eskers. The purpose of the present study was to model and identify significant processes that govern subsurface transport of pesticides in extensive glaciofluvial and littoral sediments. To simulate the transport processes, we coupled a vadose zone model at soil profile scale to a regional groundwater flow model. The model was applied to a municipal drinking-water aquifer, contaminated with the pesticide-metabolite BAM (2,6-dichlorobenzoamide). At regional scale, with the combination of a ten-meter-deep vadose zone and coarse texture, the observed concentrations could be described by the model without assuming preferential flow. A sensitivity analysis revealed that hydraulic conductivity in the aquifer and infiltration rate accounted for almost half of the model uncertainty. The calibrated model was applied to optimize the location of extraction wells for remediation, which were used to validate the predictive modeling. Running a worst-case scenario, the model showed that the establishment of two remediation wells would clean the aquifer in four years, compared to nine years without them. Further development of the model would require additional field measurements in order to improve the description of macrodispersion in deep, sandy vadose zones. We also suggest that future research should focus on characterization of the variability of hydraulic conductivity and its effect on contaminant transport in eskers.

Download