Groundwater microbiological quality in Canadian drinking water municipal wells.

To verify previous conclusions on the use of bacterial indicators suggested in regulations and to investigate virological quality of groundwater, a 1-year study was undertaken on groundwater used as a source of drinking water in 3 provinces in Canada. Raw water from 25 municipal wells was sampled during a 1-year period for a total of 167 samples. Twenty-three sites were selected on the basis of their excellent historical bacteriological water quality data, and 2 sites with known bacteriological contamination were selected as positive controls. Water samples were analyzed for general water quality indicators (aerobic endospores, total coliforms), fecal indicators (Escherichia coli, enterococci, somatic and male-specific coliphages), total culturable human enteric viruses (determined by cell culture and immunoperoxidase), noroviruses (analyzed by reverse-transcriptase -- polymerase chain reaction (RT-PCR)), adenovirus types 40 and 41 (analyzed by integrated cell culture (ICC) - PCR), and enteroviruses and reoviruses types 1, 2, and 3 (analyzed by ICC-RT-PCR). General water quality indicators were found very occasionally at the clean sites but were frequently present at the 2 contaminated sites. Only one of 129 samples from the 23 clean sites was positive for enterococci. These results confirm the value of raw water quality historical data to detect source water contamination affecting wells that are vulnerable. Samples from the 2 contaminated sites confirmed the frequent presence of fecal indicators: E. coli was found in 20/38 samples and enterococci in 12/38 samples. Human enteric viruses were not detected by cell culture on MA-104 cells nor by immunoperoxidase detection in any sample from the clean sites but were found at one contaminated site. By ICC-RT-PCR and ICC-PCR, viruses were found by cytopathic effect in one sample from a clean site and they were found in 3 samples from contaminated sites. The viruses were not detected by the molecular methods but were confirmed as picornaviruses by electron microscopy. Noroviruses were not detected in any samples. The results obtained reinforce the value of frequent sampling of raw water using simple parameters: sampling for total coliforms and E. coli remains the best approach to detect contamination of source water by fecal pollutants and accompanying pathogens. The absence of total coliforms at a site appears to be a good indication of the absence of human enteric viruses.