The 245 water samples collected from 56 wells at 45 sites in surficial sand-plain aquifers that underlie 600 square miles of Douglas, Kandiyohi, Pope, and Stearns Counties in west-central Minnesota contained wide ranges in concentrations of some constituents--sulfate 2 to 160 mg/L (milligrams per liter), chloride 1.6 to 64 mg/L, nitrite plus nitrate nitrogen < (less than) 0.1 to 72 mg/L, ammonia <0.01 to 3.3 mg/L , iron <0.003 to 8.6 mg/L, and managanese <0.001 to 1.5 mg/L . Substantial fluctuations in these constituents over time were observed. Nitrate concentrations exceeded the Minnesota drinking-water standard of 10 mg/L (as N) in 50 percent of the wells sampled.

Differences in concentrations of nitrate with depth below the water table were observed at 11 paired-well sites where one well was screened near the water table and another well was screened 10 or more feet below the water table. Similar differences in concentration with depth were observed in specific conductance, iron, chloride, and sulfate, but not as consistently. At eight sites, nitrate concentrations were higher at the water table (4.3 to 72 mg/L) than deeper in the aquifer (<0.1 to 27 mg/L). At the other three sites, nitrate concentrations were greater at depth than at the water table, although the contrast in concentrations was not as large. Increased concentration at depth is attributed to vertical mixing at the three sites, possibly caused by rapid infiltration of more dilute water from ephemeral ponds following storms. The more dilute water displaces the high-nitrate water to greater depth.

Nitrate concentrations fluctuated seasonally in response to the combined effects of precipitation and agricultural practices (irrigation and fertilization). Nitrite plus nitrate nitrogen concentrations in one Stearns County well fluctuated from 35 mg/L in August 1982, to 72 mg/L in May 1983, to 18 mg/L in May 1984. In 12 wells for which historical data were available, the mean nitrate concentration during this study (1982-84) was greater than the mean concentration during previous studies (1965-78), although the amount of increase in nitrate concentrations observed over the previous 10 or more years was not as great as the seasonal fluctuations observed during this study. Short-term seasonal fluctuations are greater than apparent historical increases in nitrate concentration.

Concentrations of sulfate, chloride, and, especially, nitrate were significantly higher in irrigated areas than in uncultivated areas. Mean nitrite plus nitrate nitrogen concentrations were 4.3 mg/L in uncultivated (natural) areas relatively unaffected by agriculture, 5.4 mg/L in nonirrigated cultivated areas, and 17 mg/L in irrigated cultivated areas. The mean nitrite plus nitrate nitrogen concentration in uncultivated areas was greater than the usually observed background levels of less than 1 mg/L (as nitrogen), which may indicate that areas thought to be natural actually are affected by agricultural land use. Several statistical tests indicate that nitrate concentrations are significantly higher in irrigated cultivated areas than in nonirrigated cultivated areas. The difference in nitrate concentrations between irrigated and nonirrigated areas suggests that nitrate from agricultural fertilizer is being leached during heavy rainfall or irrigation periods as a result of higher concentrations of nitrogen being maintained in the soil to enhance yields of corn and potatoes.

Four of eight wells sampled for herbicides in west-central Minnesota had detectable concentrations of the triazine herbicide atrazine that ranged from 0.2 to 0.6 micrograms per liter. These concentrations were well below the water-quality guidelines published by the Canadian Inland Waters Directorate, Water Quality Branch, which specify 100 micrograms per liter as the maximum permissible concentration for the group of triazine herbicides in a raw drinking-water supply.