ENTERIC VIRUS TRANSPORT FROM STREAMS TO MUNICIPAL WELLS IN ALLUVIAL AQUIFERS: CONTRASTING EXAMPLES FROM IOWA AND NEW YORK

We conducted studies in two locations in which there is evidence that pumping of municipal wells induces exfiltration of water from an adjacent stream. The aquifers in the two studies, one in Iowa and one in New York, were composed of coarse alluvial and glaciofluvial sand and gravel. The aquifer in the Iowa study site has both unconfined and confined parts while the aquifer in the New York study site is entirely unconfined. Both aquifers are used for municipal water supply but the aquifer in the Iowa study site serves a much larger population and the wells are pumped at a much higher rate. Evidence of exfiltration included hydraulic head measurements in the municipal wells as well as piezometers in and adjacent to the stream, stream discharge measurements, vertical streambed flux estimates from temperature data, and results of numerical modeling that explicitly simulates the stream-aquifer relationship.

We used human and animal enteric viruses as tracers of surface water into the aquifers. Sources of viruses include agricultural activities (dairy and concentrated swine farms), small wastewater treatment plants that discharge to the streams, and raw municipal sewage. Water samples from municipal wells and the streams were collected monthly or several times per month over a period of about one year by filtering large volumes of water through glass fiber filters (Iowa) or dialysis filters (New York). The samples were analyzed using RTq-PCR to identify and quantify virus material in units of genomic copies per liter (gc/L).

Results from the two sites differed markedly. At the Iowa site, eighteen samples from a municipal well 1250 m from the stream showed virus detections with a maximum concentration of 6.51 gene copies/L (gc/L) and a mean concentration of 0.26 gc/L. In contrast, the New York site showed only one virus detection (1.2 gc/L) in groundwater in municipal wells 75 m from the stream. Assuming that virus degradation did not occur during transport, factors that may control virus transport to the pumping wells include differences in pumping rate, in hydraulic conductivity, porosity, and hydraulic head gradient (contributors to average groundwater velocity and age) the groundwater flow path, and possibly the stream cross-sectional area over which the exfiltration occurs. In situations where virus transport from a stream seems likely, we suggest that site hydrogeology exerts a strong control on whether viruses arrive at the pumping wells.