WATER-QUALITY CHANGES CAUSED BY RIVERBANK FILTRATION BETWEEN THE MISSOURI RIVER AND THREE PUMPING WELLS OF THE INDEPENDENCE, MISSOURI, WELL FIELD, 2003–2005

The city of Independence, Missouri, operates a well field in the Missouri River alluvial aquifer and follows source-water treatment rules for ground water that include maximum contaminant levels for disinfection byproducts, minimum residual disinfectant levels in finished water, and specific goals for removal or inactivation of bacteria, protozoa, and viruses. Results of a study by the U.S. Geological Survey and the city of Independence indicate riverbank filtration substantially improves Independence well field source-water quality. Samples analyzed from the Missouri River near the Independence well field, two pumping wells, and a collector well indicate riverbank filtration decreased turbidity, dissolved organic carbon and nutrient concentrations, total haloacetic acid formation, and total trihalomethane formation potential between the Missouri River and three wells of the Independence well field. Total coliform bacteria, Cryptosporidium, Giardia, and total culturable viruses were detected in water from the Missouri River, but undetected in all water samples from wells.

Minimum log removals calculated for samples separated by traveltime between the Missouri River and wells ranged from 0.8 to 3.5 for turbidity, from 1.5 to 2.1 for Giardia, from 0.4 to 2.6 for total culturable viruses and were infinite for total coliform bacteria. Concentrations for most constituents in Missouri River samples and well samples collected at times that approximated the traveltime from the Missouri River to wells indicate no clear relation between temporal changes in water quality in the Missouri River and in wells. The absence of any temporal trends in water quality between the Missouri River and the pumping wells indicates the strong influence of riverbank filtration. The rate of ground-water flow from the Missouri River to the wells ranged from 1.2 to 6.7 feet per day. These rates are less than rates typical of slow sand filters. Slower flow rates result in greater filtration and indicate riverbank filtration at this site may be at least as effective as a slow sand filter.