Elucidating the effects of river fluctuation on microbial removal during riverbank filtration

The transfer of microbial pathogens from surface or waste water can have adverse effects on groundwater quality at riverbank filtration sites. Previous studies on groundwater protection in sandy unconfined aquifers with the focus on virus transport and health based water quality targets, such as done in the Netherlands, revealed larger protection zones than zones limited by 60 days of groundwater travel time. The 60 days of travel time are the design criterion in Austria for drinking water protection. However, in gravel aquifers, microbial transport processes differ significantly to those in sandy aquifers. Preferential flow and aquifer heterogeneities dominate microbial transport in sandy gravels and gravel aquifers. Microbial mass transfer and dual domain transport models were used previously to reproduce these effects. Furthermore, microbial transport has mainly been studied in the field during steady state groundwater flow situations. Hence, previous microbial transport models have seldom accounted for transient groundwater flow conditions. These dynamic flow conditions could have immense effects on the fate of microorganisms because of the variations in flow velocities, which are dominating microbial transport. In the current study, we used a variably saturated, three-dimensional groundwater flow and transport model coupled to a hydrodynamic surface water model at a riverbank filtration site. With this model, we estimated the required groundwater protection zones based on 8 log10 viral reductions and compared them to the 60 days travel time zones. The 8 log10 removal steps were based on a preliminary microbial risk assessment scheme for enteroviruses at the riverbank infiltration sites. The groundwater protection zones were estimated for a set of well withdrawal rates, river fluctuation ranges and frequencies, river gradients and bank slopes. The river flow dynamics and the morphology of the riverbed and banks are potentially important factors affecting microbial transport processes during riverbank filtration, which were previously not accounted for. Acknowledgments We would like to thank the Austrian Science Funds FWF for financial support as part of the Doctoral program DK-plus W1219-N22 on Water Resource Systems and the Vienna Waterworks (MA31) as part of the GWRS-Vienna project. We would also like to thank the MA39 (IFUM) for helping at the preliminary risk assessment.