Water-quality and geohydrologic data were collected from 19 monitor wells and a stream in an agricultural area in southeastern Wisconsin. These sites were located along a 2,700-ft transect from a local ground-water high to the stream. The transect is approximately parallel to the horizontal direction of ground-water flow at the water table. Most of the wells were installed in unconsolidated deposits at five locations along the transect and include an upgradient well nest, a midgradient well nest, a downgradient well nest, wells in the lowland area near the stream, and wells installed in the stream bottom. The data collected from this study site were used to describe the water quality and geohydrology of the area and to explain and model the variations in water chemistry along selected ground-water flowpaths.

Water samples from most wells and the stream were analyzed for major ions, nutrients, pesticides, dissolved organic carbon, aluminum, tritium, CFCs, 15N, 18O, and dissolved gases. Measurements of temperature, pH, specific conductance, and dissolved oxygen were made in the field. Concentrations of all dissolved constituents were below Wisconsin ground-water quality enforcement standards. The concentrations of both nitrate and ammonium in precipitation concentrated by evapotranspiration are roughly equal to the concentrations of either in the shallow ground waters. The nitrogen and oxygen isotope data, however, indicate that soil ammonium, ammonium fertilizer, and animal waste are possible nitrate sources. Concentrated precipitation can also supply dissolved sulfate to the shallow ground waters and may be a principal source of pesticides to the ground water. However, some input of dissolved chloride to the ground water from mineral or anthropogenic sources is necessary.

X-ray diffraction analyses of samples from 2 cores show the most abundant mineral to be dolomite, with subordinate quartz, microclme, and plagioclase, and minor amounts of mica, hornblende, and chlorite. Hydraulic conductivities determined from slug tests at selected wells range from 0.006 to 55 feet per day, with most values between 0.4 and 12 feet per day.

A cross-sectional ground-water flow model, representing the water-table flow system, was developed for the site and was used to identify possible ground-water flowpaths for geocli^mical modeling. The model was calibrated against measured water levels and was most sensitive to variation in recharge and hydraulic conductivity. The calibrated model shows that downward flow from shallow to deeper wells within a nest may occur at the upgradient and midgradient well nests, but that flow from each well nest travels beneath downgradient nests to the stream. Pathline and travel-time analysis performed on the calibrated flow-model output yielded travel times to well screens that range from 5.8 to 59 years with a recharge of 4 inches per yr. Recharge dates based on tritium and CFC concentrations range from pre-1955 to 1986 and are consistent with flowpaths1 and travel times in the calibrated flow model.

Changes in water quality along ground-water flowpaths were evaluated using the geochemical model PHREEQC. Geochemical mole balance models of shallow ground-water formation show that the principal reaction, by an order of magnitude, is dissolution of dolomite with CO2 . Concentration factors in the mole-balance models range from 1 to 11, with most values between 5 and 10, which provides independent support for the concentration factor of 8 based on recharge estimates used in the flow model.

Ground water recharging at mid- and downgradient wells is oxic and contains dissolved nitrate, whereas the ground water discharging to the stream is anoxic and contains dissolved ammonium. Redox environments were defined at each well on the basis of relative concentrations of various dissolved redox-active species. Chemically permissible flowpaths inferred from the observed sequence of redox environments at well sites are consistent with flowpaths in the ground-water flow model. The transition from nitrate in recharging ground water to ammonium in ground water discharging to the stream suggests the possibility of nitrate reduction along the flowpath. None of the techniques employed in this study, however, were able to prove the occurrence of this reaction.